via electronic filing...summary of hdr’s assessment of anchor qea (2017) report.” in attachment...

Don Pedro Hydroelectric Project FERC Project No. 2299-082 – California

La Grange Hydroelectric Project

FERC Project No. 14581-002 – California August 15, 2019 VIA ELECTRONIC FILING Kimberly D. Bose, Secretary Federal Energy Regulatory Commission 888 First Street NE Washington DC 20426 Re: Response of Turlock Irrigation District and Modesto Irrigation District to Comments on the Draft Environmental Impact Statement Dear Secretary Bose: Turlock Irrigation District and Modesto Irrigation District (collectively, the “Districts”), owners and licensees of the Don Pedro Hydroelectric Project No. 2299 (“Don Pedro” or “Project”) and owners of the La Grange Hydroelectric Project No. 14581 (“La Grange”), hereby submit these comments in response to comments filed by various entities on the Draft Environmental Impact Statement (“DEIS”) for Don Pedro and La Grange, which was issued by the Federal Energy Regulatory Commission (“FERC” or “Commission”) on February 11, 2019. In these comments, the Districts: (1) support the positions of relicensing participants with respect to key non-flow measures proposed by the Districts in their October 11, 2017 Amended Final License Application (“AFLA”), including the Districts’ proposed infiltration galleries and a fish counting and barrier weir; (2) provide additional information regarding the development of a revised Bay-Delta Plan in California, which is a comprehensive plan on file with FERC that may be the basis for mandatory conditions included in a water quality certification filed by the California State Water Resources Control Board (“SWRCB”); and (3) clarify or correct select comments on the DEIS filed by relicensing participants.

August 15, 2019

In addition, the Districts’ comments include the following attachments:

Attachment A: “Districts’ Response to Resource Agency and NGO Comments on the Don Pedro/La Grange DEIS.” Attachment A provides the Districts’ point-by-point responses and clarifications in a matrix format to the comments on the DEIS filed by the U.S. Fish and Wildlife Service (“FWS”); U.S. Environmental Protection Agency; U.S. National Marine Fisheries Service (“NMFS”); California Department of Fish and Wildlife (“CDFW”); SWRCB; The Bay Institute; Tuolumne River Conservancy; Central Sierra Environmental Resource Center; and Conservation Groups.1

Attachment B: “Lower Tuolumne River Habitat Improvement Program.” In the DEIS, Commission staff did not recommend inclusion of the Lower Tuolumne River Habitat Improvement Program (“LTRHIP”), which was proposed jointly by the Districts and the FWS, because it was unclear to Commission staff: (1) how the LTRHIP projects would be funded; (2) where the projects would be located; (3) how the Districts would obtain access rights to property to implement and maintain the projects; (4) how compliance with other federal laws such as the Endangered Species Act and the National Historic Protection Act would be obtained; and (5) whether the scope of the projects would require a project site to be included within the Project boundary. Attachment B addresses each of these concerns by providing this information for the initial group of four habitat enhancement projects planned to be implemented under the LTRHIP within five years of license issuance. These projects also serve as examples of the types of collaborative habitat enhancement projects that may be recommended by the proposed Tuolumne Partnership Advisory Council (“TPAC”). The Districts note that FWS’ and NMFS’ comments on the DEIS expressly support the LTRHIP and the formation of the TPAC. The LTRHIP is a prime example of fish and wildlife enhancement projects that should be considered by FERC pursuant to Section 10(a) of the Federal Power Act (“FPA”), and should be recommended in the FEIS for inclusion in the Don Pedro license.

Attachment C: “Districts’ Response to NMFS Comments on the Draft Report for the Thermal Performance of Wild Juvenile O. mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature.” The Districts previously filed in Attachment C of the AFLA responses to comments made by NMFS on the subject Draft Report (see Farrell et al. 2017). The Districts are refiling their response to NMFS’ comments on the Draft Report because this information supports the Districts’ responses to NMFS’ comments on the DEIS in Attachment A hereto regarding water temperature considerations.

1 Conservation Groups include the following entities: California Sportfishing Protection Alliance, Tuolumne River Trust, Trout Unlimited, American Rivers, American Whitewater, Merced River Conservation Committee, Friends of the River, Golden West Women Flyfishers, Central Sierra Environmental Resource Center, Tuolumne River Conservancy, American River Touring Association, Inc., Sierra Mac River Trips, Inc., O.A.R.S. West, Inc., All-Outdoors California Whitewater Rafting, Inc.

August 15, 2019

Attachment D: “Watercourse Engineering Assessment of Temperature Variability in the Upper and Lower Tuolumne River.” Attachment D supports the Districts’ responses to NMFS in Attachment A hereto regarding diurnal temperature variability in the upper and lower Tuolumne River.

Attachment E: “Summary of HDR’s Assessment of Anchor QEA (2017) Report.” In Attachment B to their March 15, 2018 filing, the Districts commented on the inadequacy and limited utility of the report prepared by Anchor QEA, LLC (“Anchor”) regarding conceptual engineering plans for fish passage, which report was filed by NMFS with FERC on November 13, 2017. Attachment E hereto provides a summary of the Districts’ previous technical review of the Anchor report. Attachment F: “Additional Examples of Large-Scale and Experimental Predator Suppression Programs.” Attachment F includes a compilation of selected examples of large-scale management programs in North America to remove or suppress predatory fish to benefit native fish species and examples of experimental studies examining survival responses to non-native fish removal conducted in California. Attachment F refutes comments on the DEIS filed by Conservation Groups and NMFS, and supports the inclusion of the Districts’ proposed predator control and suppression plan as an integral non-flow measure to enhance native salmonids on the lower Tuolumne River.

I. The Infiltration Galleries and the Barrier Weir Should Be Recommended in the FEIS Because They Are Necessary and Appropriate Non-Flow Measures That Serve Project Purposes The Districts proposed in the AFLA to install and operate two in-river infiltration galleries on the lower Tuolumne River at approximately river mile (“RM”) 25.9. Each infiltration gallery would have a flow capacity of approximately 100 cubic feet per second (“cfs”) and would be connected via a steel pipe to a pump station located on the south bank of the river. The infiltration galleries would allow the Districts to release water supplies to the lower Tuolumne River that otherwise would be diverted for consumptive purposes. Thus, the purpose of the infiltration galleries is to add flexibility to the Districts’ use of the Project for water supply and generation purposes, while also enhancing habitat for native salmonids for 25 river miles between the Districts’ current water supply diversion location at approximate RM 52 and the infiltration galleries (RM 25.9). The Districts also proposed in the AFLA to install and operate a small barrier weir (approximately five feet of head at normal flows) near RM 25.5. The purpose of the barrier weir would be to block invasive, non-native striped bass and black bass from attempting to pass upstream to locations that may be used by rearing juvenile native salmonids. The weir would cause the invasive bass to congregate below the weir and would facilitate their removal or isolation in conjunction with the timing of the release of

August 15, 2019

outmigration pulse flows, as described in the Districts’ comprehensive predator suppression and control plan. The weir also would provide a year-round means of monitoring and accurately counting the upstream migration of native salmonids, which is not possible with the existing temporary fish-counting weir located at RM 24.5 because it cannot operate at flows greater than 1,500 cfs, thereby missing portions of the native salmonid upmigration. In addition, the weir would support efficient and cost-effective operation of the infiltration galleries by increasing the available net positive suction head. In response to Commission staff’s rejection in the DEIS of the Districts’ proposed infiltration galleries and the barrier weir, Conservation Groups, FWS, and NMFS filed comments supporting inclusion of the infiltration galleries in the Don Pedro license, and FWS filed comments supporting inclusion of the barrier weir. The Districts agree with these commenters because: (1) these measures are key to appropriately balancing the competing Project purposes of water supply, generation, and enhancement of fish and wildlife under Section 10(a) of the FPA; (2) these measures are integral components of the suite of flow and non-flow measures proposed by the Districts in the AFLA to balance competing Project purposes; and (3) these measures are necessary and appropriate for the operation of Don Pedro. Accordingly, Commission staff should reconsider these measures and recommend in the FEIS the inclusion of the infiltration galleries and the barrier weir in the Don Pedro license. Section 10(a) of the FPA requires the Commission to consider and balance measures that serve Project purposes, including measures for the “protection, mitigation, and enhancement of fish and wildlife,” measures to protect water supply, and measures for the “improvement and utilization of water-power development.” 16 U.S.C. § 803(a) (emphasis added). Inclusion of the infiltration galleries and the barrier weir in the Don Pedro license is within the Commission’s jurisdiction because they serve Project purposes authorized under Section 10(a), including protecting the water supply of the Districts and the City and County of San Francisco (“CCSF”), maximizing the availability of generation (i.e., the improvement of water-power development), and enhancing populations of and habitat for native salmonids. The Districts relied on the detailed studies and computer models conducted as part of the relicensing process to develop an integrated suite of river-specific flow and non-flow measures to balance the competing Project purposes of water supply, generation, and enhancing fish and wildlife. The individual measures that comprise the Districts’ integrated proposal are closely linked, and selectively eliminating measures reduces the overall effectiveness of the Districts’ proposal. For example, the effectiveness of the outmigration pulse flows proposed by the Districts’ is maximized with the barrier weir, which causes predators to congregate for easy removal before outmigration begins. While not necessary for the infiltration galleries to function, the barrier weir supports the operation of the infiltration galleries by improving their pumping efficiency and operability. The infiltration galleries allow for the enhancement of habitat for native salmonids for 25 river miles between the Districts’ current water supply diversion location at approximate RM 52 and the infiltration galleries (RM 25.9), while protecting water supply and maintaining flexibility to generate during the peak summer demand

August 15, 2019

period. See, e.g., Districts’ Response to AIR, Table 1 Generation Production Summary, located in each of Attachments C-J (filed May 14, 2018). Non-flow measures such as the infiltration galleries and the barrier weir are an effective means to enhance fisheries, and the elimination of these measures by Commission staff may require greater flows to attain similar results compared to those achieved by the combination of flow and non-flow measures proposed by the Districts. For example, the Districts’ modeling demonstrates that, when non-flow measures are incorporated, requiring flows greater than those proposed by the Districts in the AFLA does not measurably improve smolt productivity in the lower Tuolumne River. See DEIS Figures 3.3.2-26, 3.3.2-29, 3.3.2-32, and 3.3.2-25; Districts’ Response to AIR, Section 3.0 (filed May 14, 2018). Accordingly, under the Districts’ proposal, the infiltration galleries and the barrier weir can serve the FPA Project purpose of enhancing fish and wildlife without reducing water supply or opportunities for generation during key summer months. Commission staff’s rejection of these measures undermines the careful balancing of competing Project purposes by diminishing the potential benefits for over-summering salmonids (by reducing flows below that which the Districts proposed), significantly and adversely affecting the Districts’ and CCSF’s water supply needs (by not permitting flow recapture), and significantly and adversely affecting the Districts’ ability to generate during summer months (by reducing opportunities for peaking response). Thus, the infiltration galleries and the barrier weir are key to an appropriate balancing of competing Project purposes under Section 10(a), and would allow for the best adapted comprehensive use of the waterway to efficiently serve the Project’s purposes of water supply, generation, and the enhancement of fish and wildlife.2 In addition, because the infiltration galleries and the barrier weir would allow for the Districts to maintain flexibility to produce peaking power during the peak demand period (summer), these non-flow measures are integral to maintaining the generation capabilities of the Project, are “used and useful” in connection with the Project, and are “necessary and appropriate” in the maintenance and operation of the Project’s generating facilities. 16 U.S.C. § 796(11) (a “project” includes all miscellaneous structures “used and useful” in connection with a unit of development or “necessary or appropriate” in the maintenance and operation of the project). See, e.g., Public Utility District No. 1 of Pend Oreille County, Wash., 117 FERC ¶ 61,205, at ¶ 206 (2006) (finding a pumping station “used and useful” in connection with the operation of the project because it pumps water over a railroad dike, which ultimately allows the licensee to produce more power).

2 The Districts note that in the DEIS Commission staff did not have any problem considering whether increased flows would benefit fish; yet, Commission staff rejected non-flow measures that have similar or greater beneficial effects on fish and that more appropriately balance competing jurisdictional purposes.

August 15, 2019

II. Voluntary Agreement Process in California In addition to being supported by federal fish and wildlife agencies, the Districts’ proposed non-flow population and habitat enhancement measures, including the infiltration galleries, the barrier weir and related predation plan, and the projects to be developed as part of the LTRHIP, are currently under consideration by California agencies, including CDFW and the SWRCB, as part of a Voluntary Agreement that would constitute a Tuolumne River-specific alternative to the SWRCB’s update to the Water Quality Control Plan for the San Francisco Bay/Sacramento-San Joaquin Delta Estuary (“Bay-Delta Plan”). The Bay-Delta Plan is a comprehensive plan on file with FERC and is the basis for any conditions included in a water quality certification issued by the SWRCB pursuant to Section 401 of the Clean Water Act. The Bay-Delta Plan update may also affect any revised fish and wildlife recommendations filed by CDFW. The process to develop Voluntary Agreements as an alternative to the Bay-Delta Plan update is an ongoing process and is anticipated to conclude with the adoption of amendments to the Bay-Delta Plan by the SWRCB. Significant recent developments in this process include the presentation to the SWRCB by CDFW and the California Department of Water Resources (“CDWR”) on December 12, 2018, of a term sheet for a Voluntary Agreement specific to the Tuolumne River, and a resolution adopted by the SWRCB on the same date directing SWRCB staff to incorporate potential amendments to the Bay-Delta Plan update based on Voluntary Agreements related to the Tuolumne River. On March 1, 2019, CDFW and CDWR submitted to the SWRCB a “Planning Agreement” inclusive of a project description for environmental review of the Voluntary Agreements, a proposal for a process for the SWRCB to analyze the project description, and a proposal for a process for developing appropriate terms for and implementation of the Voluntary Agreements.3 Appendix A to the Planning Agreement includes project descriptions for the proposed Voluntary Agreements, including with respect to the Tuolumne River, flow measures based on the AFLA, with modifications,4 and examples of illustrative non-flow habitat measures, including gravel augmentation and maintenance, improvements to instream habitat morphology/complexity, an adult fish counting and barrier weir, predator control, a new downstream point of diversion (i.e., the infiltration galleries), and spill management. See Table 3 of Appendix A to the March 1, 2019 Planning Agreement. Appendix A6 and Appendix A10 (Section 1.10.7) of the Planning Agreement further detail the flow and non-flow measures under consideration by CDFW, CDWR, and

3https://www.waterboards.ca.gov/waterrights/water_issues/programs/bay_delta/docs/bay_delta/va_planning_agreement.pdf (last accessed August 15, 2019). 4 These modifications include the following minor changes to the AFLA-recommended flows: (1) increase the June 1-October 15 flows below the infiltration galleries from 75 to 125 cfs in Critical and Dry water year types; and (2) decrease the June 1-October 15 flows measured at La Grange from 350 cfs to 300 cfs in Wet, Above Normal and Below Normal year types. These modifications also include an additional floodplain pulse flow of 2750 cfs for 9 to 20 days, depending on the water year type, subject to a provision for relief in the event of a defined sequence of dryer water year types.

August 15, 2019

SWRCB as part of a Voluntary Agreement being negotiated with water users on the Tuolumne River, such as the Districts. The most recent schedule issued by the SWRCB regarding the process to review and analyze the terms of the Voluntary Agreements under negotiation contemplates the issuance of a report on the scientific basis of the Voluntary Agreements by October 30, 2019. This report would be a major milestone in the process leading to final adoption of the Voluntary Agreement. The Districts are currently in active negotiations regarding the terms of a Voluntary Agreement for the Tuolumne River. The terms under consideration by CDFW, CDWR, and the SWRCB, as presented by these agencies on March 1, 2019, are based on the Districts’ flow and non-flow measures proposed in the AFLA, with adjustments to the base and pulse flows proposed therein. Significantly, the terms contemplated for inclusion in the Voluntary Agreement include all of the non-flow measures proposed by the Districts in the AFLA, including a predation control weir and a downstream diversion point at the infiltration galleries, as well as a suite of illustrative habitat enhancement measures that are similar in type and scope to the projects contemplated by the Districts and the FWS as part of the LTRHIP. As described herein, the Commission should analyze these measures in the FEIS, in part, because conditions included in a water quality certification issued by the SWRCB and any revised 10(j) recommendations filed by CDFW are likely to be based on a Voluntary Agreement that includes these measures. III. Clarifications and Corrections to Select Comments on the DEIS

A. Predator Control and Suppression Plan As part of their integrated suite of flow and non-flow measures to serve and balance Project purposes, the Districts proposed in the AFLA a comprehensive predator control and suppression program that included non-flow measures such as isolating predatory fish at the barrier weir and other collection methods to control and suppress invasive, non-native bass. In the DEIS, Commission staff did not recommend the Districts proposed predator control and suppression plan “because [it is] not likely to be effective.” Commission staff concluded that “[s]imilar predator removal efforts by the [CDWR] did not noticeably reduce salmon mortality.” See DEIS at pg. 2-22. Instead of relying on the site-specific studies conducted by the Districts as part of the relicensing proceeding, Commission staff referenced a single example where a predator removal effort by the CDWR at the Clifton Court Forebay was purported to be unsuccessful. Id. at pg. 5-65. In their comments on the DEIS, Conservation Groups agree with Commission staff’s finding that the Districts’ proposed non-flow predator control and suppression efforts in the lower Tuolumne River would not benefit native salmonid populations. See Conservation Groups DEIS Comments at pgs. 28-29. Similarly, NMFS states that “there is no scientific consensus that non-native predator suppression programs such as what the Districts propose in the AFLA will measurably increase juvenile salmonid survival or increase adult salmonid abundance in the Tuolumne.” See NMFS DEIS Comments at

August 15, 2019

pg. 7. Notably, while FWS raised concerns with FERC staff’s rejection of the Districts’ non-flow measures because removing these measures may affect the assumptions incorporated into the Districts’ modeling, FWS’ comments on the DEIS did not object to the Districts predator control and suppression plan as a valid measure to enhance native salmonid populations in the lower Tuolumne River. For the reasons explained in the Districts’ April 12, 2019 comments on the DEIS, the Districts strongly disagree with Commission staff’s unsupported rejection of the predator control and suppression plan. As described therein, Commission staff references, but provides no detail regarding, a single purported example of a failed predator removal effort by CDWR at the Clifton Court Forebay. In contrast to this isolated reference, the site-specific studies conducted by the Districts as part of the relicensing proceeding conclude that predator control and removal efforts are essential to and have a high probability of increasing native salmonid survival in the lower Tuolumne River. See Districts’ DEIS Comments at pgs. 3-4. The Districts also pointed to other studies – Sabal et al. 2016, Cavallo et al. 2012, Davis et al. 2012 – that show reducing invasive bass numbers result in an increase in juvenile salmonid survival. Id. at Section 3.0, Table 1, Item No. 28. Since the Districts’ filing of their comments on the DEIS, the Districts have researched additional examples of large-scale management programs in North America to remove or suppress predatory fish to benefit native fish species. These include the following: (1) Northern Pikeminnow Predator Control Program on the Columbia and Snake Rivers for the purpose of protecting native salmonids that is estimated to have reduced predation of juvenile salmonids by 40% through sport fishing, site-specific gill nets, and angling at dams;5 (2) Upper Colorado River Endangered Fish Recovery Program, which is using measures such as electrofishing, sport fishing tournaments, net capture, and exclusion barriers;6 (3) Mechanical Removal of Non-Native Fishes in the Colorado River, which has reduced populations of non-native fish through electrofishing;7 (4) Great Lakes Restoration Initiative Action Plan, which has used pathway blocking, tributary barriers, and traps to create “[o]ne of the longest-running and most effective invasive control technology programs” for sea lamprey;8 and (5) Yellowstone’s Native Fish Conservation Plan, which has significantly reduced lake trout populations through entrapment nets, tagging, gill netting, and catch-and-kill programs.9 The Districts also have found the following examples of experimental studies examining survival responses to non-native fish removal in California: (1) Mokelumne River at Woodbridge Irrigation District Dam, which involved, in part, predator removal with boat-based electrofishing (Sabal et al. 2016); (2) North Fork River downstream of

5 Attachment F, Table 1. 6 Id. 7 Id. 8 Id. 9 Id.

August 15, 2019

the Delta Cross Channel (Cavallo et al. 2013), which involved predator removal by electrofishing; and (3) San Joaquin River between Port of Stockton and Lathrop, which involved relocating predator fish. A summary of the programs, methods, results, and conclusions are provided in Attachment F hereto. In addition, there is currently an ongoing collaborative project between FishBio and NMFS on the Stanislaus River to study the effect of predation on juvenile salmonid survival and migration to inform the development of a management plan.10 Based on this additional information to support the conclusions from the Districts’ studies and modeling, Commission staff should recommend in the FEIS inclusion of the Districts’ predator control and suppression plan in the Don Pedro license to benefit native salmonids in the lower Tuolumne River.

B. Conflict Between Downstream Fish Passage and the Tuolumne Wild and Scenic River

NMFS’ April 11, 2019 comments on the DEIS claim that its hybrid in-river collection facility for downstream fish passage, as outlined in the Anchor report, “is feasible and should be analyzed for constructability in the FEIS.” See NMFS DEIS Comments at pg. 7. NMFS further asserts that “the National Park Service has previously established that the construction of enhancement measures and specifically fish passage facilities at FERC licensed projects is consistent with the [Wild and Scenic Rivers] Act.” Id. The Districts strongly disagree with both statements and agree with Commission staff’s conclusion to the contrary, see DEIS at pgs. 3-169, 3-170, in part, because FERC likely would be barred by the Wild and Scenic Rivers Act (“WSRA”) from licensing NMFS’ proposal for downstream passage, as described in the Anchor report. The WSRA was enacted for the purpose of preserving selected river reaches in their free-flowing condition for the benefit and enjoyment of future generations, and includes a number of prohibitions and limitations specific to the Commission to further this purpose. Section 7(a) of the WSRA bars the Commission from licensing the construction of any facility “on or directly affecting” any designated river. 16 U.S.C. § 1278(a). In addition, Section 7(a) bars the Commission from licensing the construction of facilities above or below a designated segment of the wild and scenic rivers system if the facility “invades” the area or “unreasonably diminishes” the scenic, recreational, and fish and wildlife values present. Id. The Districts conducted a technical review of the Anchor report, which estimated that NMFS’ proposed hybrid collector system located near Wards Ferry Bridge would involve a dam with a height of approximately 75 feet. See Attachment B of Districts’ March 15, 2018 filing. A dam of this height in the canyon-like reach of the Tuolumne River between the Tuolumne Wild and Scenic River boundary and Wards Ferry Bridge (a distance of approximately 1.8 miles) likely would result in backwater effects on the designated Tuolumne Wild and Scenic River. Consequently, FERC would be barred 10 See http://www.savethestan.org/native-fish-recovery-efforts/ (last accessed August 15, 2019).

August 15, 2019

from licensing the construction of NMFS’ proposed dam because the backwater effects would “directly affect” and would “invade” the designated Tuolumne Wild and Scenic River. In addition, FERC may be barred from licensing construction of the dam because the reservoir environment created by the backwater from NMFS’ dam and the dam itself likely would “unreasonably diminish” the recreational value of the designated river segment. Recreational value would be diminished because the free-flowing condition of the designated Tuolumne Wild and Scenic River would be reduced and NMFS’ dam would be located in the same location where whitewater boaters completing a run of the designated reach typically exit the river. This expansion of the Don Pedro reservoir and the impairment of takeout activities likely would unreasonably diminish the recreational values of the whitewater boater experience on the Tuolumne Wild and Scenic River. Moreover, NMFS’ reference to the pleading filed by the U.S. National Park Service (“NPS”) in the Carmen-Smith (P-2242) relicensing proceeding is inapt. In the Carmen-Smith proceeding, NPS filed a comment letter in 2014 urging the Commission to allow construction of mitigation measures, including placement of gravel and large woody debris, and construction of recreation improvements within the McKenzie Wild and Scenic River corridor, which encompasses the entirety of the Carmen-Smith project. See Accession No. 20140103-5057. This NPS letter does not reflect the policy of the agency in charge of administering either the McKenzie or the Tuolumne Wild and Scenic Rivers, and does not reflect the current policy of the NPS. The McKenzie Wild and Scenic River is managed by the U.S. Forest Service (“USFS”). In the Carmen-Smith proceeding, the USFS was a party to a settlement agreement that resulted in an off-license agreement for implementation of five measures because the USFS agreed that the WSRA barred FERC from issuing a license authorizing construction of these measures. Thus, irrespective of the position of the NPS, which has no management role with respect to the McKenzie River (nor the lower designated segment of the Tuolumne River), the position of the USFS in the Carmen-Smith proceeding was that FERC was barred from licensing the construction of facilities under the WSRA because the measures would be “on or directly affecting” or would “invade” or “unreasonably diminish” the scenic, recreational, and fish and wildlife values of the McKenzie Wild and Scenic River. Further, the reference to NPS’ wayward filing in 2014 in the Carmen-Smith proceeding does not appear to reflect current NPS policy described in a filing on July 30, 2019, in Docket No. AD19-7, which policy recites the same prohibitions and limitations on FERC set forth in the WSRA described above. See Accession No. 20190730-5062.

C. Wards Ferry Take-Out Facilities The U.S. Bureau of Land Management (“BLM”), Conservation Groups, and the USFS fault Commission staff for not “recommending” inclusion of BLM Condition No. 13, which requires a Wards Ferry/Tuolumne River Take-Out Plan, as part of the “Staff Alternative” described in Section 2.3 of the DEIS. These commenters fail to

August 15, 2019

acknowledge that BLM Condition No. 13 is included in Section 2.4 of the DEIS titled “Staff Alternative with Mandatory Conditions.” Because Condition No. 13 was filed as a mandatory condition, FERC is required to include this condition in any license issued for Don Pedro. See Escondido Mutual Water Co. v. La Jolla Band of Mission Indians, 466 U.S. 765 (1984). While the Districts acknowledge that because Condition No. 13 is a mandatory condition it will be included as a condition in the Don Pedro license, the Districts strongly disagree with BLM’s assertion in its comment on the DEIS that Condition No. 13 (or any takeout activities at Wards Ferry Bridge generally) has a nexus to Project operations or Project effects. The record does not support commenters’ unsubstantiated assertions regarding the historical use of the Wards Ferry area as a takeout or alternative takeout locations. BLM alleges that prior to the construction of Don Pedro reservoir, “most whitewater boaters did not takeout at Wards Ferry Bridge” rather, “[t]he preferred takeout was at the former Highway 120/49 Bridge.” See BLM DEIS Comments at pg. 3. The very first takeout opportunity for boaters on the downstream reach of the Tuolumne Wild and Scenic River is at Wards Ferry Bridge, which is located 18 miles downstream from the nearest put-in location at Lumsden Campground. The next takeout opportunity is at Moccasin Marina, which is located an additional five miles downstream from Wards Ferry Bridge.11 While some boaters may have used locations in the vicinity of Moccasin Marina as a takeout prior to the construction of Don Pedro, including the former (but no longer existing) Highway 120/49 Bridge location, there is absolutely no evidence in the record to support the characterization that “most whitewater boaters did not takeout at Wards Ferry Bridge” or that the “preferred takeout” was at the former Highway 120/49 Bridge. The shortest possible whitewater trip on the Tuolumne River is 18 miles from Lumsden Campground to Wards Ferry Bridge, and this trip requires a full day to complete. As the river approaches the Don Pedro reservoir (but upstream of the influence of the reservoir), the gradient diminishes significantly and the frequency of flatwater increases. Thus, even prior to the existence of the Don Pedro reservoir (which is not the appropriate baseline from which to assess Project effects in a relicensing proceeding, see American Rivers v. FERC, 187 F.3d 1007, amended and reh’g denied, 201 F.3d 1186 (9th Cir. 1999)), adding an additional five miles of partial flatwater to an 18-mile-long river trip would likely have been a significant incentive to takeout at the very first opportunity at Wards Ferry Bridge. The Districts also dispute Conservation Groups’ characterization that “the filling of the reservoir has created conditions where shoreline access is steep and uneven” near Wards Ferry Bridge. See Conservation Groups DEIS Comments at pg. 30. On river right, commercial rafters pay a fee to the Districts to cross privately-owned property to reach Wards Ferry Road. While there are steep sections of the hillside on river right should one egress without using existing trails, there is also a switchback trail with a moderate grade

11 The former Highway 120/49 Bridge takeout location was near Moccasin Marina. However, this former highway bridge has been rerouted and the current iteration of the bridge could no longer be used as a takeout with nearby road access in the same manner as the former Highway 120/49 bridge may have been used.

August 15, 2019

that is easily navigated, even when carrying boats. On river left, the shore where boaters exit from the river is flat. As the trail progresses in elevation and onto BLM lands, the gradient increases significantly. However, the steep sections of this trail are not under the influence of the reservoir and thus are steep regardless of the elevation of the reservoir. Therefore, the reservoir did not “create[] conditions where shoreline access is steep and uneven;” rather, the Tuolumne River canyon in the vicinity of Wards Ferry Bridge is steep and uneven, and these conditions would exist even without the influence of the reservoir. Similarly, BLM laments the “increasing distance to carry boats and equipment up to the road as the reservoir lowers.” See BLM DEIS Comments at pg. 3. BLM seems to misunderstand that the natural state of the river is the lowest reservoir elevation. Under BLM’s logic, the Districts are benefitting boaters when the reservoir is higher by lessening the distance necessary to carry boats and equipment up to Wards Ferry Road, compared to the distance traveled from the river’s natural elevation. Raising the reservoir to shorten the walk to Wards Ferry Bridge for boaters is not an adverse Project effect. The Districts acknowledge that Section 10(a) of the FPA identifies recreation as a Project purpose and Section 4(e) requires equal consideration of recreational opportunities. The Districts’ AFLA proposal satisfies both of these standards and includes a significant amount of recreational facilities, recreational improvements, and recreational opportunities at and near the reservoir. In fact, Project-based recreational facilities and opportunities at and near Don Pedro are so abundant that it was necessary for the Districts to create a distinct agency – Don Pedro Recreation Agency – to manage Project-based recreation. There is no doubt that, even without the facilities described in Condition No. 13, recreation opportunities are given more than equal consideration in the Districts’ AFLA and recreation opportunities related to Project features and operations are maximized. Finally, the case cited by BLM for support that a designer boater takeout facility should be constructed by the Districts does not support BLM’s position. In Public Utility District No. 1 of Okanogan County, 144 FERC ¶ 62,018, at ¶ 102 (2013), the Commission required the licensee to add a boat launch/takeout to a new recreation area (parking, camping, picnic sites) required by the new license because the five-foot dam raise proposed by the licensee would inundate an informal takeout location 0.25 miles downstream from the new recreation area. The Commission’s primary reason for requiring construction of the new recreation area and the boat launch/takeout was because the existing licensed project did not have any recreation facilities – not a single one. This is in stark contrast to the availability of abundant Project-based recreational opportunities at Don Pedro. The Districts note that the recreation facility cited by BLM was never constructed because the license was terminated. Public Utility District No. 1 of Okanogan County, Wash., 168 FERC ¶ 62,084 (2019). In sum, while the designer Wards Ferry takeout facility will be required in a Commission license because it was included in a mandatory condition filed by BLM, the

August 15, 2019

Commission’s conclusion in the DEIS that such a facility lacks a nexus to the continued operation of the Project remains true and supported by the record in this proceeding. If you have questions regarding this filing, please contact the undersigned.

Respectfully submitted,

Steve Boyd Turlock Irrigation District P.O. Box 949 Turlock, CA 95381 (209) 883-8364 [email protected]

John B. Davids Modesto Irrigation District P.O. Box 4060 Modesto, CA 95352 (209) 526-7564 [email protected]

Enclosures cc: Don Pedro and La Grange Service Lists

AmericasActive:13841831.3

ATTACHMENT A

Districts’ Response to Resource Agency and NGO Comments

on the Don Pedro/La Grange DEIS

Don Pedro and La Grange Hydroelectric Projects Turlock and Modesto Irrigation Districts (TID/MID) FERC Nos. 2299 and 14581 1 Districts’ Response to Comments on the DEIS

ATTACHMENT A

Districts’ Response to Resource Agency and NGO Comments on the Don Pedro/La Grange DEIS

No. Entity

Comment Ltr Page Number / Location Issue Raised / Entity Position / Request to FERC Districts’ Response

1. USFWS Pg. 8 FERC staff did not recommend a streamflow gage be located downstream of the Districts’ proposed infiltration galleries at RM 24.5. FERC staff indicated they did not recommend a gage near RM 25 because it would not monitor compliance with potential license requirements or have a Project nexus. FERC staff recommended against inclusion of the Districts’ infiltration galleries at RM 25.9. Additional flows associated with the IGs provide recreational boating benefits upstream and downstream of RM 25. Therefore, a Project nexus exists with a need to monitor compliance.” “Recreational boaters and anglers increasingly rely on publically available stream flow data to plan their trips. Fox Grove Park and boat launch is located adjacent to the proposed infiltration galleries. During the DEIS public meeting that FERC staff held in Modesto on March 27, 2019, members of the public commented that this boat launch provides one of the few public access points for boaters on the Tuolumne River. Therefore, the USFWS recommends that the Districts be required to estimate stream flow near RM 25, and make that information publicly available for use by recreational boaters and anglers.”

The Districts agree that FERC has erred in not including the IGs as part of the future license for the Don Pedro Project. The IGs provide an enhancement to fish and wildlife of the lower Tuolumne River as specifically intended by and consistent with Section 10(a)(1) of the FPA. Flows associated with the IGs provide recreational boating opportunity upstream of the IGs, but no additional boating opportunity below the IGs beyond that summarized in Section 5.7.2 of the AFLA, Exhibit E, filed with FERC on October 11, 2017. The Districts disagree with USFWS’ recommendation that the Districts be required to install a streamflow gage near RM 25. This would have the effect of making the Districts responsible for mitigating for factors influencing flows in the reach between the La Grange gage at RM 52 and RM 25, such as riparian water withdrawals and outflows to groundwater under drought conditions. These factors are not Project effects, nor under the control of the Districts. The La Grange gage provides the appropriate measuring point for flow compliance. Compliance with flows required to be delivered below the La Grange gage, if FERC elects to include the IGs in the next license, would be measured by subtracting actual flows withdrawn by the IGs from the La Grange gage flows. Closed conduit


pipeline flows such as those in the pipeline exiting the IGs are routinely measured with a high degree of accuracy.

2. USFWS Pg. 9 “FERC staff recommended adoption of this measure [Spill Management Plan] but did not recommend adoption of the advisory committee that would be responsible for implementation. It is unclear how this plan will be implemented without the scientific and technical expertise that an advisory group would provide.

The Districts agree with USFWS that FERC should adopt the requirement for the Districts to form an advisory committee, referred to as the Tuolumne Partnership Advisory Committee, or TPAC, to assist with implementation of the Spill Management Plan. It is acknowledged that FERC cannot require agency participation, which would necessarily be on a voluntary basis. However, the Districts agree that forming the TPAC will provide the opportunity for such participation.

3. USFWS Pg. 9 “FERC staff did not recommend adoption of this measure [Lower Tuolumne River Habitat Improvement Program] saying that it was not a specific measure to protect fish and wildlife.”

FERC staff cited five specific reasons for not including the LTRHIP in the Staff alternative (see pgs 2-22 and 2-23 of the DEIS). The Districts are herewith submitting as Attachment B a full description of the first set of initial habitat improvement projects to be developed under the LTRHIP which fully addresses each of the five concerns raised by FERC staff in the DEIS. The Districts recommend that FERC now reconsider inclusion of the LTRHIP as part of the new license for the Don Pedro Project.

4. USFWS Pg. 10 Salmonid Monitoring - FERC staff did not recommend adoption of this measure in whole or part.

The USFWS recommends that the Districts develop a Salmonid Monitoring Plan for the lower Tuolumne River consisting of seven discrete measures. The Districts support the USFWS’ recommendations for paired rotary screw trap monitoring; operation of a seasonal adult upmigration counting weir near RM 24.5; snorkel surveys prior to LWM placements; and annual reporting of these activities. The Districts do not agree with any of the USFWS’ recommendations that involve annual carcass surveys (which are now carried out by CDFW) or any


related morphometric measurement of carcasses. These elements should continue to be the responsibility of CDFW.

5. USFWS Pg. 13 Water Temperature Monitoring Plan – USFWS recommended monitoring that included at a minimum, monitoring within four different areas of the lower Tuolumne River. FERC recommended adoption of the measure in part. FERC staff recognized that the EPA previously found that the lower Tuolumne is impaired for temperature, that Project operations currently contribute to the temperature impairment, and that the warmer water temperatures reduce habitat suitability for Chinook salmon and O. mykiss downstream of La Grange Diversion Dam, particularly for spawning, egg incubation, and smoltification. Yet, FERC staff did not recommend water temperature monitoring during normal operating conditions of the Project. Nor did FERC staff indicate whether Project operations with the FERC staff recommended minimum instream flows would result in the Project no longer contributing to the temperature impairment. Instead, FERC staff only recommended water temperature monitoring when Don Pedro Reservoir elevations will be lower than 600-feet elevation.

The Districts have consistently stated their strong disagreement with statements generally referring to “warmer water temperatures reduce habitat suitability for Chinook salmon and O. mykiss downstream of La Grange Diversion Dam.” These statements are unsupported in the current record before FERC. USFWS offers no reference or citation to “warmer water temperatures” so as to explain what specifically the current temperatures are warmer than; that is, what is the basis of comparison. USFWS presents no evidence or analysis showing actual “reduced habitat suitability” or any actual harm to fish populations in the Tuolumne River during spawning, egg incubation, or smoltification. USFWS’ claims amount to supposition. EPA’s assessment of temperature “suitability” was not based on any site-specific investigation of fish condition or habitat use, but solely based on guidance developed for Northwest fisheries. The EPA itself recently called into question the application of broad based temperature “criteria” citing the body of recent research which shows that fish species may become adjusted or adapted to local stream conditions. The Districts’ site-specific studies have shown that the temperatures in the Tuolumne River are supportive of all life stages of native salmonids, and have proposed enhanced flows which will continue to support these life stages. Moreover, the fact is that the USFWS routinely conducts condition assessments of fall-run Chinook juvenile


outmigrants and have consistently found these Tuolumne River fish to be in good condition. The Districts currently maintain a system of water temperature monitors in the lower Tuolumne River and plan to continue this monitoring. The Districts do not oppose maintaining this system and are open to adding stations, if shown to be useful.

6. USFWS Pg. 15 Revise the AFLA Woody Debris Management Plan to Include Rapid LWM and Woody Debris Removal

USFWS continues to advocate for “rapid LWM and Woody Debris removal” from the upper reaches of the Don Pedro Reservoir. As discussed at length in the Districts’ response to FERC’s DEIS, this is not feasible because of the remote location and lack of access to the preferred wood capture area, which protects both whitewater boaters’ egress at Wards Ferry Bridge and downstream recreation use of Don Pedro Reservoir. Current debris capture is performed by barge and boats which pen in the debris for later burning. There is no vehicle access to the wood capture area. Capturing and penning debris at Wards Ferry Bridge would result in a long upstream extension of debris which would significantly interfere with whitewater boaters’ safe and efficient egress from the Tuolumne River. There is no truck access to the river at Wards Ferry, and the debris would have to be removed by cranes and haul trucks working atop the bridge, which is not allowed under County road use regulations and would further interfere with boater, boat, and equipment egress. The Districts refer FERC staff and the USFWS to related comments submitted on the DEIS. The Districts also note that they have had numerous discussions with the USFWS’s sister agency – BLM – on this matter and have reached agreement with respect to debris handling.


However, regardless of the LWD collection method, the USFWS states that the primary concern is for the potential for debris rafts to affect CRLF, citing there is no CRLF survey data to prove otherwise. The Districts point out that FERC’s independent assessment of potential effects to CRLF due to Project operations resulted in a finding of “no adverse effect” based on (1) the lack of suitable habitat in the reservoir, (2) fish are the primary frog predators, and (3) no CRLF have ever been found within five miles of the Project. USFWS also is concerned that the LWD rafts encourage bullfrog predation on CRLF, which although unlikely is neither proven nor unproven. A full CRLF survey protocol is eight visits and very costly compared to the likelihood of an impact on CRLF. Personnel safety and feasibility of conducting the survey at the required times are also concerns that should be considered. Conducting surveys in high-flow years (e.g. 2017) would be risky to surveyors. Safety concerns also affects the season/timing of surveys and LWD work, because safety and feasibility may require doing the work after water levels have dropped. The physical constraints of the site, including constraints affecting personnel safety, may preclude the feasibility of conducting the full protocol level surveys.

7. USFWS Pg. 16 “Although Article 413 [Coarse Sediment Management Plan] does not specify the amount of material that should be augmented on a yearly basis, FERC staff did provide recommendations in other areas of the DEIS. FERC staff recommended augmentation of only 1,000 cubic yards per year (cu yds/yr). FERC staff based this

As the Districts have pointed out previously in their response to agency comments on the AFLA regarding the issue of sediment accumulation in the Don Pedro Reservoir:


volume on the lowest range of estimated amount of coarse bed material lost from storage in the lower Tuolumne River annually rather than the amount of material withheld by the Don Pedro Dam. The annual bedload sediment that the Districts’ dams capture was estimated to be an average of 18,800 cu yds/yr as detailed in the 2004 Coarse Sediment Management Plan for the Lower Tuolumne river (McBain and Trush, 2004). In a separate analysis by the Districts, they concluded that 23,560 cu yds/yr of coarse sediment is trapped by the Project (Stillwater Sciences 2013).”

(1) The McBain and Trush (2001; 2004) estimate was described as potential sediment delivery from the “unimpaired” watershed above Don Pedro as reported by a third party, and not intended as being an independent estimate attributable to McBain and Trush.

(2) The study plan describing the Districts’ bathymetry survey conducted in 2011 specifically stated the purpose of the work was to provide information pertaining to reservoir geometry to support development of the Districts’ proposed reservoir temperature model. It was not intended to be a study of the sediment yield of the watershed above Don Pedro Reservoir. Any such use of the bathymetry study would have had to take into account the limited reliability and usefulness of comparing a bathymetry survey conducted using 2011 technology to topographic maps of unspecified origin developed in the 1920s prior to closure of the original Don Pedro Dam. The differences in accuracy and precision of the two methods render any comparison as highly suspect. In fact, study W&AR-04 (the source cited by USFWS) specifically states that the estimate of storage loss due to sedimentation since the closure of the first Don Pedro Dam (1923) is less than 1 percent of the original estimated storage capacity of the current Don Pedro Reservoir, and more importantly, “[t]his percentage is within the uncertainty associated with the interpolated surfaces.” That is, the estimated change in storage


capacity should not be relied upon to establish sediment yields, quantities, or distribution.

Furthermore, there has been no study of bedload transport rates along the Tuolumne River. Don Pedro Reservoir has inundated known areas of extensive gravel deposition, including the Jacksonville area upstream of Railroad Canyon and various substantial gravel bars, including the original Don Pedro Bar, downstream of Railroad Canyon.

8. EPA Pg. 3 (pdf) The DEIS uses a temperature value of 22C for the lower Tuolumne River in its model to determine impacts to water quality and aquatic resources alternatives. This value is based on a study of peak performance, as measured by aerobic scope, for juvenile O. mykiss (Verhille et al., 2016). The DEIS consequently concluded that O. mykiss have likely adapted to the recent thermal shift of the lower Tuolumne River. We note that the duration of the aerobic scope test in Verhille was six hours, which is substantially shorter than the 96-hour duration recommended by EPA for testing acute effects on salmonids (“guidelines for Deriving Numerical National Water Quality Criteria for the Protection of Aquatic Organisms and their Uses, “ U.S. EPA, 1985). Also, the DEIS does not appear to examine population level impacts or impacts to life stages other than juvenile rearing (e.g., spawning, egg incubation, fry, emergence, smoltification). Similar to the DEIS, EPA acknowledges that there is an open scientific question about the thermal

The EPA makes the same error as NMFS and CDFW by mischaracterizing the tests conducted and reported by Verhille et al. (2016) as a test of “acute effects” on salmonids. The Districts refer EPA to Farrell et al. (2017) in the Districts’ AFLA, and for convenience are providing a copy as Attachment C with this filing. Regarding there being a “lack of consensus”, as explained in Verhille et al. (2016) and further enumerated in Farrell et al (2017), the results of the tests conducted on Tuolumne River juvenile O. mykiss demonstrate that the Tuolumne strain of O. mykiss “are locally adapted to higher water temperatures than the O. mykiss populations used to develop the EPA (2003) criteria.” EPA goes on to list all the various and sundry factors that may “influence the health of salmonids” and that should be incorporated into an assessment of appropriate temperature “criteria”, knowing full well that no such test that evaluates all these factors could actually be undertaken. In all of EPA’s comments, what is ignored is the actual health of the O. mykiss population in the Tuolumne River. This population has expanded rapidly since the ’96


adaptability of Central Valley salmonids; however, there is a lack of consensus at present regarding O. mykiss adaptation to the recent shift in the thermal regime of the lower Tuolumne River. EPA recommends that FERC consider a robust suite of data and endpoints that broadly address thermal physiological and ecological effects, both acute and chronic, when determining scientifically-sound temperature values. Include in the FEIS a description of additional studies and references available regarding protective thermal values of salmonids. We further recommend including in the FEIS a discussion of the broad range of physiological and ecological factors that can influence the health of salmonids, including potential negative effects of competition, disease, and predation under a warmer thermal regime. Further, EPA recommends that the FEIS consider additional temperature research regarding population level impacts and impacts to life stages other than juvenile rearing.

Settlement Agreement. Population figures provided in the DEIS demonstrate the continuing health of the Tuolumne O. mykiss population under the existing temperature regime.

9. EPA Pg. 4 (pdf) The three DEIS alternatives do not include 10(j) conditions from the NMFS or the CDFW. An adequate analysis of the third alternative (i.e., staff alternative with all mandatory conditions from other agencies) and potentially the second alternative, (the staff alternative including accepted conditions) if FERC decides to include a condition in its staff alternative will need to include these conditions. In the FEIS, include all submitted Section 4€ and 10(j) conditions in the staff alternative with mandatory

As the Districts explained in their May 14, 2018, submittal to FERC in response to FERC’s Additional Information Request, the generalized non-flow measures recommended by the agencies were not sufficiently detailed to enable mathematical modeling. Recommendations under Section 10(j) must be adequately defined so as to be able to understand exactly what is being recommended and the desired outcome. This was not the case in NMFS’ or CDFW’s Don Pedro 10(j) recommendations.


conditions and the staff alternative, if applicable. For any conditions that FERC determines to be invalid, provide supporting analysis.

10. NMFS Pg. 2 “NMFS recommends thermal criteria that are protective of ESA-listed populations, such as EPA (2003). The 18°C criterion from EPA (2003) is not an “upper aerobic performance limit” and it is inappropriate to compare results from an acute test of individuals (such as Farrell et al. 2017) to a threshold meant to protect against chronic negative effects to populations. In fact, the authors of Farrell et al. (2017) do not recommend new thermal criterion be based solely on their results. If new thermal criteria were to be developed for the Tuolumne River, many other factors would need to be considered. These include considerations of exposure time (chronic vs. acute), food availability, predator abundance and avoidance, susceptibility to disease, among other population-level effects which were considered during the development of the EPA (2003) criteria.

NMFS continues to misunderstand and mischaracterize the Districts’ swim tunnel studies, the findings reported in Verhille et al. (2016), and the further explanation of the tests and conclusions as provided in Farrell et al. (2017) and the application of the findings to Tuolumne River O. mykiss. As part of the La Grange licensing process, a systematic and collaborative effort was undertaken by licensing participants (LPs) to assess the potential for success of introducing anadromous fish to the upper Tuolumne River (UTR). One of the potential limiting factors of the UTR identified by LPs was temperature suitability for the two species under consideration – spring-run Chinook salmon and anadromous O. mykiss. A Temperature Technical Committee (TTC) was formed to collaboratively develop appropriate temperature metrics for the in-river life stages of these species for comparison to the existing temperature regime of the UTR. The TTC examined numerous data sources and relevant literature and developed a set of temperature suitability metrics for presentation to the larger plenary group of LPs. Recognizing that healthy fish populations may experience a range of in-situ water temperatures, the TTC developed life stage specific Upper Optimum and Upper Tolerable water temperature indices (UOWTI; UTWTI) for spring-run Chinook and anadromous O. mykiss. By definition, up to the maximum UOWTI, growth, reproduction, and fish


behavior are unimpaired. Between UOWTI and UTWTI, growth and reproduction may be reduced and above the maximum UTWTI, growth and reproduction cease. The maximum UTWTI temperature is not to be interpreted as either chronic or acute lethal. Based largely on the findings presented in Verhille et al. (2016), the TTC recommended a UOWTI of 18℃ and an UTWTI of 22℃ for juvenile O. mykiss. The Plenary Group, of which NMFS was a participant, concurred with these temperature metrics for the Tuolumne River with no party objecting. While the DEIS may have mislabeled the UTWTI as an “upper performance limit” instead of “upper tolerable”, NMFS, for its part, continues to describe Verhille et al. (2016) as an examination of O. mykiss juveniles’ “acute” response to temperature, despite the further extensive explanations provided in Farrell et al. (2017). As explained in considerable detail in the authors’ response to NMFS’ February 2017 comments on the Farrell et al. (2015) draft report (which was the basis for the Verhille et al., 2016 peer reviewed paper on the same subject), and as also addressed in response to similar comments submitted by CDFW in August 2016 (see W&AR-14 study documentation in the AFLA), NMFS’ characterization of Farrell et al. (2015; 2017) as an “acute” response experiment displays a fundamental and troubling misunderstanding of the study design, purpose, and results. The authors of Farrell et al. (2017) have gone to great lengths to carefully describe the biological significance of the study results and what absolute aerobic scope (AAS) measures, as well as how these data relate to the EPA (2003) 7DADM guidance for O. mykiss juveniles.


While Farrell et al. (2017) clearly states that their report is not intended to be the basis for changing the EPA Region 10 temperature guidance for Northwest salmonids suggested in EPA (2003), the study report, and the subsequent peer reviewed paper (Verhille et al. 2016), do demonstrate that the EPA (2003) “criterion” of 18°C lacks merit for the current juvenile O. mykiss population found in the lower Tuolumne River. At the very least, the report shows that 18°C would be a very conservative, and biologically indefensible, upper thermal limit for Tuolumne River O. mykiss juveniles. In NMFS’ DEIS comment letter, the EPA (2003) temperature guidance is also mischaracterized. First, NMFS mislabels EPA’s temperature guidance of 18℃ for juvenile O. mykiss as an EPA “criterion”, implying a regulatory standard, which it is not. EPA (2003) is strictly a guidance document. Next, NMFS describes the temperature of 18℃ as a “threshold meant to protect against chronic negative effects to populations”. The EPA (2003) reports make it abundantly clear that the temperature of 18℃ was judged to be the optimal temperature for peak salmonid growth. NMFS’ misunderstanding of the underlying biological intent of the EPA (2003) temperature guidance may contribute to the difficulty NMFS is having in understanding the Farrell et al. (2017) findings. The experiment reported in Farrell et al. (2017) measured the optimal temperature range of metabolic capacity for Tuolumne River O. mykiss, not acute CTmax temperatures. NMFS presents a quote from Farrell et al. (2017) which NMFS refers to as evidence that Farrell et al. (2017) did not intend to recommend a new “thermal criterion” for


Tuolumne River O. mykiss. NMFS presents only a portion of the statement attributed to Farrell et al. (2017). The complete quote is provided below for clarity.

“What our data should NOT be used for is to pick a new thermal criterion based solely on our aerobic scope curve. In fact, we do not suggest revising the 7DADM based solely on our AAS curve. We simply state that we believe our data are suggestive of local thermal adaptation in Central Valley fish and inconsistent with a blanket criterion for the population under consideration. Because the Tuolumne River O. mykiss fish outperform northerly populations at warm temperatures, the inference is that the current guidelines are overly conservative.”

Lastly, in the first part of the last paragraph on pg 2, NMFS begins with: “If new thermal criteria were to be developed for the Tuolumne River…” It’s worth pointing out again that there is no current numeric temperature criteria of 18℃, or any other value, for Tuolumne River O. mykiss. Temperature standards are set by the State of California for rivers under its jurisdiction, and the State has not adopted specific numeric temperature standard or “criteria” for the Tuolumne. NMFS’ recommendation as to what the State Water Board should consider in setting temperature criteria is helpful to understand.

11. NMFS Pg. 3 “It is erroneous to conflate “temperature tolerance” with protective “temperature criteria”. In order to meet EPA’s obligations under the Clean Water Act and the Endangered Species Act, EPA (2003) established temperature criteria to: “Recommend how States and Tribes can designate uses and establish temperature criteria that will protect salmonid

NMFS’ suggested use of a “dual temperature standard”-- one when considering “reintroduction”--and another when considering “protection” of existing fish--is not biologically defensible. NMFS proposes that for the same ESA-listed species in the same river two temperature standards should be applied. For Tuolumne River O. mykiss NMFS applies one thermal suitability metric for those O.


populations…..” In short, EPA (2003) is intended to support survival and growth of salmonids and protect against adverse, thermal impacts. Whereas, Boughton et al. (2018) distinguishes between “tolerable” and “intolerable” temperatures in order to establish the probability that a particular habitat would be occupied on a life stage specific basis. The criteria used in Boughton et al. (2018) and the accompanying technical analysis estimates the hydraulic, geomorphic and thermal components of capacity. This was done in order to predict maximum, life-stage specific, habitat capacity for Central Valley (CV) Spring-run Chinook salmon and CV steelhead in the upper Tuolumne River, under current conditions. In addition, EPA (2003) recognizes several cases where the general guidance provided in the document can be supplanted with site-specific criteria: “Another example may be where there is exceptionally high natural diurnal temperature variation and where the maximum weekly mean temperature is within the optimal temperature range but, because of the high diurnal variation, summer maximum temperatures exceed the State or Tribe’s numeric criteria. In this situation, a State or Tribe may choose to develop a site-specific numeric criterion based on a metric other that the 7DADM (e.g., a maximum weekly mean criterion plus a daily maximum criterion). There may be other situations as well when an alternative site-specific criterion would be appropriate.” (EPA 2003, p.34). The upper Tuolumne basin experiences high diurnal variation that make it a potential candidate for

mykiss already in the river and a different one for any “new” O. mykiss coming into the river. More specifically, any O. mykiss NMFS designates as being part of a “reintroduction effort”, even if the O. mykiss is already present, would have a different, less stringent, thermal suitability standard than existing O. mykiss not designated to be part of the “reintroduction”. The notion that an ESA-listed species used for colonization of new habitat or a “reintroduction” experiment does not need to be “protected” against adverse temperatures to the same degree as the existing population of the ESA-listed fish makes little biological sense. If anything, taking an endangered fish from an existing, “protective” temperature regimen and intentionally and knowingly exposing it to a harsher temperature regimen is itself a likely violation of the ESA mandate against harming an ESA-listed species. NMFS’ need to justify a “dual temperature standard” is exposed when NMFS attempts to explain why a NMFS’ study of the potential to introduce ESA-listed CV steelhead (anadromous O. mykiss) to the upper Tuolumne River adopted a much different temperature suitability metric than that applied by NMFS to the same ESA-listed O. mykiss in the lower Tuolumne River. NMFS’ rationale for applying dual temperature metrics is explained in its comments to the DEIS: “In short, EPA (2003) is intended to support survival and growth of salmonids and protect against adverse, thermal impacts. Whereas, Boughton et al. (2018) distinguishes between “tolerable” and “intolerable” temperatures in order to establish the probability that a particular habitat


a temperature metric that does not use the 7DADM metric.

would be occupied on a life stage specific basis. The criteria used in Boughton et al. (2018) and the accompanying technical analysis estimates the hydraulic, geomorphic and thermal components of capacity. This was done in order to predict maximum, life-stage specific, habitat capacity for Central Valley (CV) Spring-run Chinook salmon and CV steelhead in the upper Tuolumne River, under current conditions.” The EPA (2003) guidance for juvenile O. mykiss is a 7DADM temperature of 18℃. NMFS’ description, shown above, of the intent of the EPA (2003) recommended temperature of 18℃ is, at best, imprecise. The EPA (2003) documents make it very clear that the 18℃ temperature is intended as something more than “protective against adverse thermal impacts”. The 7DADM of 18℃ is recommended as the optimal temperature to maximize growth of juvenile O. mykiss [see EPA (2003), Issue Paper 5, Table 1; and pg 29 referring to Myrick and Cech (2000)]. Establishing thermal conditions intended to maximize the potential for peak growth of O. mykiss juveniles is a substantially different goal than “protecting” fish against adverse thermal effects. Conditions may be something less than optimal for growth, but still protective of viable fish populations. Boughton et al. (2018) states that rebuilding of CV steelhead stocks “involves reestablishing viable populations in river systems” blocked by dams. For juvenile O. mykiss in the upper Tuolumne River, Boughton et al. (2018) adopts temperature suitability metrics that are significantly higher than the recommended EPA (2003) temperatures. NMFS’ comments to the DEIS attempts to


justify the use of different, higher allowable temperatures in Boughton et al. (2018) by claiming that the Boughton et al. (2018) study can use “tolerable” instead of “protective” temperatures to estimate production capacity. This raises the question of why would you introduce ESA-listed fish into a reach where temperatures are not “protective” over the longer term; that is, where temperatures are not “protective”; that is, where temperatures do not “support survival and growth” using NMFS own definition of “protective”. Is not the purpose of the Boughton et al. (2018) study to examine if the upper Tuolumne might support a viable population of steelhead? Can the population grow and be viable if it must spend its fresh water life stage in a river where temperatures do not “support “survival and growth”? NMFS’ contortion of words on this matter is not biologically relevant. In a final attempt to justify the application of higher suitable temperatures to upper Tuolumne O. mykiss compared to lower Tuolumne O. mykiss, NMFS acknowledges that the EPA (2003) document provides “general guidance” and not strict “criterion” and identifies one such example where “site-specific criteria” might apply is where there exists high diurnal temperature variation. NMFS then declares that the upper Tuolumne River experiences high diurnal temperature variation as distinguished from “moderate” or “low” variation, and therefore may not be subject to the EPA (2003) 7DADM metric. NMFS does not provide temperature thresholds distinguishing “high” from “moderate” or “low” temperature variation. Comparing the lower Tuolumne River diurnal temperature variation to the upper Tuolumne, it is apparent that the lower Tuolumne River diurnal temperature variation is as large, or larger,


than the upper Tuolumne River diurnal temperature variation (see Attachment D to these comments). As shown in Attachment D, the diurnal temperature range experienced at approximately the mid-reach of the lower Tuolumne River O. mykiss habitat (RM 45) and near the lower end of that habitat (RM 40) is significantly greater than that experienced at the mid-and lower-reach in the upper Tuolumne River. Therefore, according to NMFS, the lower Tuolumne River qualifies for application of site-specific criteria and strict application of EPA (2003) would not be appropriate. Regarding what the EPA (2003) guidance document suggests about “site-specific criteria”, it is helpful to have a broader understanding of the purpose and findings of the EPA (2003) reports. As explained on page 1 of Issue Paper 5 of the EPA (2003) documents, Summary of Technical Literature Examining the Physiological Effects of Temperature on Salmonids, EPA’s purpose is to review “Information on the optimal or preferred ranges of temperatures that will be needed to promote long-term survival, growth, and reproductive success” of salmonid populations. Farther on page 1, EPA acknowledges the importance of understanding the metabolic capacity of different fish species because “[m]etabolic processes are directly related to temperature, and the metabolic rate increases as a function of temperature. Fish are metabolically efficient and most likely to thrive within the preferred range of temperatures”. On page 44 of Issue Paper 5, EPA (2003) reports: “The above research shows a wide range in the estimates of optimal temperature for rearing rainbow trout. This wide range may reflect the fact that the individual subspecies and specific stocks have


evolved differently to fit the characteristics of their home streams.” Farrell et al. (2017) is a study of the metabolic capacity of existing wild juvenile O. mykiss in the lower Tuolumne River. As directly acknowledged in EPA (2003), research shows a wide range in the estimates of optimal temperature for rearing rainbow trout, and based on the rigorous metabolic testing of juvenile O. mykiss in the lower Tuolumne River, this population exhibits average aerobic scope that remained within 5% of optimum from 17.8°C to 24.6°C.

12. NMFS Pg. 3 “The operations of the Project also affect seasonal flow patterns in the lower San Joaquin River and the Sacramento-San Joaquin Delta. As FERC determined in the scoping documents for the Project, and included in the DEIS, the geographic scope for water quality, water quantity and aquatic resources extends upstream to Hetch Hetchy Dam and downstream to San Francisco Bay.”

In this comment, NMFS is referring to the geographic scope of cumulative effects analysis identified in the FERC Scoping Document. The definition of “cumulative effects” under NEPA is the impact on the environment which results from the incremental impact of the action when added to other past, present, and reasonably foreseeable future actions regardless of what agency (Federal or non-federal) or person undertakes such other actions (40 CFR ~ 1508.7). The DEIS thoroughly evaluated cumulative effects on aquatic resources with an analysis extending from pg 3-213 to 3-229. The DEIS concluded that the San Joaquin River and the Delta are affected by numerous past and current cumulative actions that occurred and/or continue to occur in the San Joaquin watershed and the Delta.

13. NMFS Pg. 4 NMFS agrees with the Districts that the IGs should be incorporated into the license. However, NMFS states: “Because the infiltration galleries are intended to mitigate project impacts, FERC should include them as a part of its proposed action in the Final Environmental Impact Statement (FEIS).”

NMFS asserts that the Districts’ proposed infiltration galleries (IGs) “are intended to mitigate project effects”. This is incorrect. NMFS provides no evidence of the Don Pedro Project having an adverse effect on Tuolumne River O. mykiss. Under the current license flow requirements contained in the ’95 Settlement Agreement, O. mykiss


populations have significantly increased in numbers and geographic extent in the lower Tuolumne River. The purpose of the IGs is to further enhance temperature conditions for O. mykiss in the Tuolumne from La Grange Diversion Dam to RM 40, the dominant O. mykiss reach. Snorkel surveys conducted since 2008 have documented the increase in adult O. mykiss population in the lower Tuolumne River (see AFLA, Exhibit E, Table 3.5-19). The IGs would further and consistently expand the thermally suitable habitat for O. mykiss. This is a resource enhancement measure since current conditions support a healthy O. mykiss population.

14. NMFS Pg. 4 “It is unclear why FERC approved of and relies on a model that does not encompass the full geographic scope of the Projects’ effects.” “Because the Districts’ models do not extend beyond the Tuolumne River, they do not fully capture the benefits of increased flow on juvenile salmonid survival. Recent studies such as Buchannan et al. (2018) show that although flow has variable effects on salmonid survival in the interior delta, survival in the upstream, more riverine parts of the Delta/San Joaquin River increased during periods of high flows.”

Accurately modeling effects on fall-run Chinook and O. mykiss populations over a geographic region encompassing the Sacramento-San Joaquin River Delta and San Francisco Bay is beyond what is reasonably possible during the Don Pedro Project relicensing process, and if attempted, would likely produce little reliable or useful information about effects to these populations as they transit the lower San Joaquin River, the Delta, or the San Francisco Bay. FERC approved of and relied upon the Districts’ suite of models for the very fact that the models encompassed the geographic scope for which Project effects could be reliably determined. NMFS acknowledges studies completed by the Districts indicate a positive relationship between increased pulse flows and juvenile survival in the Tuolumne River. These relationships are incorporated directly into the Districts’ in-river fall-run Chinook population model. Potential benefits accruing to fish populations in the San Joaquin River and the Bay-Delta region due to the Project’s Preferred Plan of flow and non-flow measures would be speculative because


of the numerous interaction of factors affecting fish populations in these complex and more distant systems. As one example, precisely modeling the predation rate on salmonids by the various non-native predator species in the SJR, Delta, and Bay would be infeasible. As cited by NMFS, Buchanan and Skalski (2018) presented the results of their analysis of the relationship between flow and survival in the SJR and Delta and found the survival of fall-run Chinook smolts through this system to be consistently low under all flow conditions, likely due to the outsized impact resulting from predation. NMFS points out that survival through some parts of the system are higher than other parts. The Districts do not see the relevance of this observation since outmigrating fish have to make it through the entire Bay-Delta system to be part of future escapements. The Districts cannot be expected to cure the problems encountered by outmigrating salmonids in the SJR and Bay-Delta system; however, the Districts’ proposed plan includes measures which substantially increase outmigration survival in the Tuolumne River, the reach able to be reasonably benefitted by the Districts’ PM&E measures. NMFS also mischaracterizes the purpose and use of the Districts’ fish population models. NMFS asserts the models are intended to “make realistic predictions of anadromous populations.” This is not the case. The models are intended to be useful for comparing the effects of alternative flow and non-flow measures on the target fish populations. The Districts have been clear that the models were never intended to be precise predictors of future population levels.


The Districts have pointed out in their comments on the DEIS that the DEIS has not adequately considered the larger body of evidence available to it which demonstrates the effectiveness of predator control programs. This evidence is cited in the Districts’ comments. NMFS commits the same error as committed by FERC staff in the DEIS when it asserts that the DEIS rejects the Districts’ predator control program “because it is not known if it would have a measureable benefit…” It is certainly true that a measure cannot be “known” to have a benefit until it is actually implemented and demonstrated. Under this logic, FERC should not adopt any new flow or non-flow measure because it is impossible to “know” they are beneficial until after they are put into effect. FERC normally relies on well-reasoned and sound scientific analysis of proposed measures to make determinations of potential effects. The Districts demonstrated the significant positive benefits of predator control by the results of fish population models which have been approved and otherwise relied upon by FERC Staff in the DEIS.

15. NMFS Pg. 5 “...the Districts’ in-river production models incorporated the effects of the Districts’ proposed predator control and suppression program. In the proposed action described in the DEIS, FERC does not approve the predator suppression program because it is not known if it would have a measureable benefit to Chinook salmon or O. mykiss populations. Although we agree with FERC, it is not clear why FERC relied on a model that assigns disproportionate weight to the benefits of a predator suppression program after excluding it from the proposed action.”

The Districts disagree with characterization of predation related mortality as being assigned a “disproportionate weight”. In addition to multiple predation studies reviewed as part of the Districts’ Salmonid Information Synthesis Study, the Districts incorporated both smolt survival study results and RST passage information to quantify the relative magnitude of predation and other salmonid mortality sources. The Districts’ model development included multiple mortality sources including predation and were calibrated to RST passage data over 2010-2013 and validated against RST data from 1999-2009. The resulting relationships


reflect the best available information to assess the effects of predation and measures such as predator control recommended by the Districts. It is interesting to note that in the NMFS’ comment which starts on page 4, NMFS first affirmatively cites to the Districts’ studies as providing a positive flow:survival relationship on the Tuolumne River, then later in the same comment rejects the Districts’ population models which incorporate these same flow:survival relationships. Regarding the DEIS’ recommendation that the Districts’ predator control program not be made part of the new FERC license, this does not mean that the program would not be undertaken as an off-license measure. It is reasonably foreseeable that the Districts’ proposed predator control program will be undertaken. Both the USFWS and CDFW, via the CDFW presented Tuolumne River Voluntary Agreement alternative in the State’s Bay- Delta water quality control plan updates, have supported implementation of a predator control plan on the Tuolumne River.

16. NMFS Pg. 5-6 “FERC analyzes NMFS 10(a) and 10(j) Condition 5, Fish Passage Program Plan, relying solely on information provided by the Districts. NMFS has filed four relevant studies on the record: Anchor 2017, Pearse and Campbell 2017, Boughton et al. 2018, Speir et al. 2018, which analyze the biological, physical, and economic feasibility of reintroducing salmonids to historic habitat in the upper Tuolumne River. For instance, the Pearse and Campbell (2017) study found the mykiss trapped above the dams have both ancestry and adaptive genomic variation

The Districts conducted a review of the Anchor QEA (2017) study and identified a number of serious flaws in the study, including potential unevaluated fatal flaws. Due to the number of significant technical issues either left unresolved or not evaluated at all, the study scope and level of analysis does not satisfy what would be needed for even a pre-feasibility level of assessment, let alone a full feasibility level study (see Districts’ March 15, 2018, filing with FERC, Attachment B). The Districts have included a summary of HDR’s findings from the review of the Anchor QEA report as Attachment E to this filing.


consistent with being descendants of indigenous migratory populations. In the Final EIS, FERC should analyze NMFS’ Fish Passage Program Plan (10(j) and 10(a) Condition 5) using the full suite of relevant studies available, including the information NMFS’ has filed on the record.

Similar concerns apply to the Boughton et al. (2018) study primarily, but not solely, for its failure to even consider the hydraulic effects to spawning and rearing habitat due to the peaking hydropower operations of the CCSF Holm powerhouse on Cherry Creek at the head of the study reach. In contrast to Boughton et al. (2018), the Districts submitted a detailed study based on 2D modeling, supported by site-specific river-reach data, of spawning and rearing habitat for Chinook salmon and O. mykiss in the upper Tuolumne River. The Districts’ analysis evaluated the upper Tuolumne River reach, including its flow dynamics, in much greater detail and to a much greater level of accuracy than undertaken in Boughton et al. (2018) (see Districts’ March 15, 2018, filing with FERC, Attachment C, Appendix H). Both of these NMFS studies contain serious flaws and unsupported assumptions, and reflect an insufficient scope of analysis to support decision-making. Regarding the Pearse and Campbell (2017) report, the Districts’ undertook a similar assessment of O. mykiss genetic traits (see Districts’ March 15, 2018, filing with FERC, Attachment C, Appendix L). This independent assessment, completed by Cramer Fish Sciences (Cramer) and Genidaqs in January 2018, was based on collections of O. mykiss from the upper (above Don Pedro Reservoir) and lower (below La Grange Diversion Dam) Tuolumne River. The Cramer study found that the frequency of the anadromous variant (residing on the Omy5 genomic region) was at least as great in the lower Tuolumne fish as in the upper Tuolumne fish. Cramer concluded that “there are sufficient genetic resources within the O. mykiss


population downstream of La Grange Diversion Dam to result in the expression of anadromy. Consequently, there is no need to translocate fish from or facilitate connectivity to the upper basin to establish a steelhead run downstream of La Grange Diversion Dam, where there is ample genetic resources to support such a run.” NMFS’ Speir et al. (2018) study evaluated the potential effect to the electric rates of the two Districts of several alternative capital expenditures, intended to represent the cost of constructing and operating upstream and downstream fish passage facilities at the La Grange and Don Pedro projects. The Speir et al. (2018) study estimated, under the assumptions applied, that fish passage facilities with a capital cost of $170 million and an annual O&M cost of $0.4 million would require a rate increase to Turlock Irrigation District customers of 3.7% and to Modesto Irrigation Districts customers of 1.3%. In standard utility practice, rate increases are serious matters requiring the utmost prudence and care because of their significant economic and financial effect on individual, community, and business customers. As a first matter, any consideration of a rate increase due to a potential capital expenditure must be underpinned by a reliable and detailed cost estimate which, in turn, must be based on a realistic and constructible project. The flaws contained in the Speir et al. (2018) attempt at ratemaking are numerous. However, by far the most serious flaw is that no actual constructible, viable fish passage project on the Tuolumne River at a cost of $170 million has been identified, so there is no merit in projecting what the rate increase might be for a project that could never be built for


$170 million. The Speir et al. (2018) evaluation amounts to just playing with numbers. A reliable economic analysis cannot be performed until a technologically feasible project has been defined which could actually be built, function as intended, and achieve expected results, none of which were accomplished by the Anchor QEA 2017 study. The numerous serious flaws, including the potential for fatal flaws which went unevaluated, are described in detail in the Districts’ review of the Anchor QEA study which the Districts filed with FERC on March 15, 2018, prepared by HDR Engineering, Inc (HDR). As HDR shows, modifying the Anchor QEA concept as needed to meet minimum engineering requirements results in an estimated capital cost of $388 million (see Table 3 in Attachment B of the March 2018 filing). This is only the cost of the “hybrid concept” which provides for downstream passage of fish. Upstream passage consisting of a collect, handle, transport and release (CHTR) facility were estimated to cost an additional $44 million. Speir et al. (2018) added an “allowance” for annual O&M cost of $400,000 per year, an absurdly low amount. Operating only the CHTR, with its fish collection, fish separation, hauling and releasing of upstream migrants would be expected to take at least four FTEs working 40 hours per week for nine months (see Figure 2, page 15 of the Anchor QEA report), not including required maintenance during the non-migration season. Total O&M cost for both upstream and downstream passage would easily run $1 million per year, not counting any replacement cost and project life expectancy, which have been sources of major additional expenses at high-dam fish passage facilities. Therefore, at a minimum for


rate estimating purposes, a capital cost of $425 million and annual O&M of $1 million should be applied. At these estimated costs, rate increases would be much greater, and this would be the rate increase for only this single capital project, not counting other costs and investments likely to face the Districts (including other PM&E measures associated with the new FERC license). It is not a prudent approach to ratemaking to base a potential rate increase on the least possible cost of a proposed project. There are numerous other flaws in the Speir et al. (2018) report, including not accounting for the cost of all the usual bond covenants, such as sufficient debt coverage ratios and reserve funds. There can be no reliable estimates of potential rate impacts resulting from a project until there is an actual constructible project accompanied by reliable cost estimates.

17. NMFS Pg. 6-7 “Agencies have had numerous meetings discussing integrating the fish passage facility with rafting access and NMFS believes this coordination could result in a facility that meets the needs of both objectives. It’s not unusual for fish passage facilities to incorporate boat traffic (see FERC’s 12/21/2006 FEIS for p-2195-011). Similar to other fish passage facilities, the collection facility could be designed to eliminate the impact. Regarding consistency with the Wild and Scenic Rivers Act (Act), the National Park Service has previously established that the construction of enhancement measures and specifically fish passage facilities at FERC licensed projects is consistent with the Act (NPS, 2014).

The Districts’ filing with FERC dated March 15, 2018, included just such a constructability assessment, as well as an engineering, cost, and operational review. HDR’s critique also compared the hybrid collector to the Wells and Rocky Reach facilities on the Columbia River and found both to have very little in common with the Anchor QEA (2017) hybrid proposal (see Attachment B of the Districts’ March 15, 2018 filing). Regarding the NMFS reference to “numerous meetings” intended to coordinate fish passage and rafting access, we presume these were held with commercial and individual rafting interests on the Tuolumne. The Districts were unaware of, and not invited to, these meetings. This approach to planning and collaboration is in direct contrast to the open and collaborative process the Districts went


through with licensing participants over a two-year period to discuss and explore the question of fish reintroduction on the Tuolumne River. Contrary to NMFS’ statement, since there are no current head-of-reservoir downstream collectors, there can be no example that is “similar to other fish passage facilities” where the impact was “eliminated”. Regarding the NMFS interpretation that the Anchor QEA hybrid collector would be consistent with regulations governing use of Wild & Scenic River corridors, the key phrase used by NMFS is “fish passage facilities at FERC licensed projects”. There is evidence to suggest that the new dam planned to be built under the Anchor QEA concept would have backwater effects on the W&S reach extending beyond the “FERC licensed project” boundary, such action being prohibited under current federal law.

18. NMFS Pg. 7-8 “NMFS agrees with FERC that there is no scientific consensus that non-native predator suppression programs such as what the Districts propose in the AFLA will measurably increase juvenile salmonid survival or increase adult salmonid abundance in the Tuolumne.” “In analyzing various predator control methods, FERC did not analyze the increased flow and lower temperatures associated with a spring pulse that mimics the natural hydrograph, which has been shown to disrupt warm-water species spawning and dramatically decrease predator abundance.”

As the Districts pointed out in comments on the DEIS, the fall-run Chinook salmon population model provides substantial evidence that the Districts’ predator control program will increase juvenile survival in the Tuolumne River. Sound science is not a matter of consensus; it relies the existence of a rigorous biological analysis of data. In the case of the Tuolumne, it is also a matter of common sense that preventing striped bass from migrating into major rearing areas for fall-run Chinook juveniles will eliminate predation by striped bass in the reach above the barrier weir, allowing greater production of smolts in the upper 20 miles of the lower Tuolumne River. Although the DEIS does not fully analyze the hour-to-hour effects of pulse flow upon water temperature under the various flow proposals, the Districts disagree with the premise put forward by NMFS that managed pulse flows


can be an effective predator suppression strategy due to disruption of spawning by warm-water species. As documented in the 2013 Predation Study and monitoring data collected under Article 58, predatory striped bass populations have increased in the lower Tuolumne River. Given that striped bass reproduce at and tolerate a broad range of temperatures, migrate into the lower Tuolumne after being born elsewhere, and have been identified as a primary source of salmon mortality in the lower Tuolumne River, NMFS’ recommendations regarding spring pulse flows affecting striped bass “spawning” are unlikely to achieve similar direct benefits as predator control and removal strategies. It is evident from NMFS’ comments that there is general agreement as to the significant adverse effects predation is having on juvenile salmon in the Tuolumne. NMFS asserts that higher flows in the Tuolumne would reduce such predation. As evidence, NMFS cites Moyle (2002) and the Districts’ annual reports. NMFS does not list the title of the citation to Moyle (2002) and the Districts were not able to locate the quote attributable to Moyle or the paper itself. Marchetti and Moyle (2001) did report on a study of Putah Creek, but this study involved downstream displacement of non-native species after natural high flow events occurring two years in a row, and the native species involved in the study were not anadromous. The reference to the Districts’ annual studies and the quote provided by NMFS on page 8 of its comments tend to support the Districts’ view that flows are not an effective


long-term predator control measures, as the quote on pg 8 makes clear – the adult bass returned after the 1997 flood event, even though that event was the largest single flow that occurred in the last 50 years measured as over 60,000 cfs. Black bass temporarily flushed from the Tuolumne River will end up downstream in the larger San Joaquin River and would be likely to return to the Tuolumne and/or increase predation in the San Joaquin.

19. NMFS Pg. 8-9 “The Project blocks all large gravel and significantly reduces springtime flow in the lower Tuolumne River. These direct Project effects have prevented recovery from the anthropogenic effects FERC cited above. For instance, if the Project was not blocking all coarse gravel transport for over 100 years, the special run pools (SRPs) would have filled in overtime with naturally occurring bedload material. Furthermore, the Project reduces the duration, frequency and magnitude of peak flows that transport gravel, further preventing natural recovery from anthropogenic effects to the Tuolumne River such as gravel mining.”

NMFS’ statement referring to filling of special run pools amounts to pure speculation. While Don Pedro Reservoir bathymetry was obtained to support the development of the reservoir temperature model, there have been no studies of gravel transport or gravel migration rates under pre-project conditions in the reach currently inundated by the Don Pedro Reservoir. However, it is well documented that large areas of gravel deposition occurred near the historical mining areas near Jacksonville upstream of Railroad Canyon and along large depositional bars below Railroad Canyon, including the Don Pedro Bar.

20. NMFS Pg. 9 “NMFS provided FERC with a detailed plan for LWM augmentation which included size classes and density targets for the lower River. It is unclear why FERC determined that these target densities are not applicable to the Tuolumne. FERC did not provide analysis as to what factors may not be applicable to the Tuolumne. NMFS cited literature from watersheds that are directly comparable to the Tuolumne (e.g. Albertson et al. 2012, Ruediger and Ward 1996).”

While NMFS cited information about LWM from other watersheds, it has ignored the specific studies conducted on the Tuolumne River. In study W&AR-12, the Districts examined the LWM captured in the Don Pedro Reservoir, a better source of information for the Tuolumne than using sources not of the Tuolumne. These studies, covering different water year types and a five-year long study period, demonstrated that very little LWM captured by the Don Pedro Reservoir are of a suitable size to be long-lasting in the lower Tuolumne River, a class 6 river.


In any event, the Districts have agreed to a comprehensive Lower Tuolumne River Habitat Improvement Plan to be implemented in the Tuolumne River below La Grange Diversion Dam. The LTRHIP includes the potential for installation of appropriately sized LWM in the river. On page 10 of its comments, NMFS generally supports the LTRHIP.

21. NMFS Pg. 11-13 These “outmigration” pulse flows were designed to increase survival during critical periods of juvenile fall-run Chinook salmon outmigration when large numbers of juveniles are “smolt” size of 65mm or above. These flows were not designed to aid in juvenile steelhead outmigration, which outmigrate throughout the January–June timeframe, peaking in March, April, and May. NMFS supports the concept of increasing juvenile salmonid survival by increasing flows during outmigration windows. However, the FERC staff alternative spring pulse flows will not increase survival during the majority of the fall-run and steelhead outmigration period.

The FERC analysis includes the Districts’ proposed flow measures identified in the AFLA which provide for spring pulse flows that would provide benefits to juvenile salmonid habitat and promote increased survival of juvenile salmonids. The outmigration spring pulse flow timing is proposed to be adaptively managed following the methods provided in the AFLA. Nevertheless, review of historical O. mykiss trawl recoveries at Mossdale indicates the exceedingly rare recoveries of putative steelhead outmigrants from the San Joaquin River that are found during April and May with almost no recoveries found in other months. This suggests both low rates of anadromy as well as the majority of outmigration occurring during April and May, consistent with when Tuolumne River pulse flows occur.

22. NMFS Pg. 16 Fall pulse flows have been released the previous three years on the lower Tuolumne and were successful in initiating upstream migration from fall-run Chinook salmon (see figure 2 above). Wet-season initiation flows also provide other ecosystem benefits such as nutrient cycling and sediment mobilization and are a part of the natural hydrograph that native salmonids have adapted in. The FERC staff recommended

Longer term records of fall pulse flows do not demonstrate a positive benefit for adult migration into the Tuolumne River. Even where a flow-induced movement trend may appear to occur, it is unclear whether this trigger results in any greater escapement, or simply an earlier upriver movement. There is no data to suggest that earlier adult movements has a net beneficial effect on spawning success or future total returns.


alternative does not provide this important aspect of salmonid habitat.

23. NMFS Pg. 17 Coarse Sediment Management Plan - It is unclear where FERC staff derived 1,000 cubic yards per year, as there is no citation associated with this statement. Studies conducted on the Tuolumne River have determined that the Project captures approximately 18,800 cubic yards of coarse sediment per year (McBain and Trush 2004, p.23). A gravel program that is commensurate with the amount of coarse bed material lost to storage would add this amount to the lower Tuolumne, which is what NMFS gravel plan (10(j) and 10(a) Condition #2) would provide.

FERC clearly states on page 3-183 of the DEIS the origin of the figure of 1,000 cubic yards per year; that is, a site-specific study of the amount of gravel lost in the dominant spawning reaches over the time period examined in the study. The amount of spawning gravel lost in the lower Tuolumne below the Don Pedro and La Grange projects is the full effect of the project on spawning gravel (see study W&AR-04). The Districts’ proposed 10-year gravel augmentation quantities far exceed the estimated loss of spawning gravel in the dominant spawning reach of the lower Tuolumne River. NMFS’ proposed gravel quantities are not related to the Project’s effects to downstream gravel loss. NMFS uses gross estimates of reservoir capture under unimpaired watershed conditions, ignoring the potential effects of gravel capture by other watershed impoundments. Moreover, there are no studies of gravel transport in reaches above Don Pedro Dam to support NMFS’ contention.

24. NMFS Pg. 18 NMF’ recommended Fish Passage Program Plan is designed to be an iterative, adaptively managed process, where new information is learned and incorporated as the Project moves through time. There is great flexibility in how the Districts could design and implement a Fish Passage Program Plan. The Plan would utilize appropriate fish and habitat studies, reintroduction experiments, and analyses in an adaptive management approach that would ultimately result in fish passage facilities and operations that have the greatest chance of success. Because FERC has not analyzed all information relating to the feasibility of reintroduction in the Tuolumne (e.g. Anchor 2017,

The Districts address the issue of fish passage and fish reintroduction in Response Comment 16 above. Moreover, here once again is an example of NMFS applying a double-standard in its recommendations to FERC. With regard to fish passage (which NMFS desires), NMFS states that it is not necessary for its feasibility to be “reasonably foreseeable” in order for FERC to include it as a license condition. But when considering whether FERC should include the Districts’ predator control plan (which NMFS does not support) as a license condition, NMFS agrees with FERC that the feasibility of the predator control plan must be “known” in order to be included as a license


Pearse and Campbell 2017, Boughton et al. 2018) it is premature for FERC to determine that establishing viable populations of salmonids in the upper Tuolumne is not feasible. Furthermore, it is unclear why fish passage would have to be “reasonably certain” to occur for FERC to include it as a condition in the new license for the Projects.

condition (see page 5 of NMFS April 11, 2019, comments on the DEIS).

25. CDFW Page 2 of 32; para 3

“...the infiltration galleries would operate from June 1 through October 15. However, the water rights request for the infiltration galleries comprises a year-round water transfer. CDFW requests that the Districts explain the rationale behind this temporal discrepancy.”

The AFLA proposes a use of the IGs which corresponds to an amount of flow and a period of time when such use is projected to benefit oversummering O. mykiss. Other use of the IGs is not a matter for FERC to consider because such use relates solely to the consumptive use of water. Any water to be withdrawn by the IGs outside of the time period intended to benefit O. mykiss (July-September) would have to be over and above any required FERC-flow at the La Grange gage and then be limited to the withdrawal of that added flow by the IGs.


“CDFW finds that when we assume the same water withdrawals as the Districts assumed in their analysis with a scenario that includes infiltration galleries, the lower Tuolumne River shows negative values for instream flows below river mile 25 (Table 1).”

The CDFW assumption regarding withdrawals pertaining to the irrigation galleries as shown in Table 1 on Page 17 of the comment letter dated April 11, 2019 is incorrect. The Districts’ proposed interim flows as stated in the DEIS and analyzed by FERC are to be provided until both irrigation galleries are operational. Table 3.3.2-20 of the DEIS lists the Districts’ minimum flows with and without infiltration galleries as measured at the USGS gage below La Grange Diversion Dam (RM 51.7) and RM 25.9 along with the proposed interim flows. No negative instream flows would occur downstream of RM 25.


“CDFW recommends temperature monitoring at designated reservoir locations whenever reservoir elevations are projected to decrease below 700 ft.”

CDFW presents no evidence supporting its recommendation. Section 10(j) recommendations must be supported by substantial evidence, and not simply reflect the preferences of one or another agency.



“Since FERC staff does not recommend a predator removal plan...CDFW asks that the Districts’ model runs of their recommended flow scenarios be updated to reflect the lack of predator removal efforts and other staff recommendations and that all model assumptions be clearly articulated in this DEIS.”

CDFW and the Districts have worked together to develop an agreed-upon plan to present to the CA State Water Board (SWB) as an alternative to the SWB’s Bay-Delta Revised Water Quality Control Plan for the Tuolumne River. This Tuolumne River alternative includes most of the elements of the Districts’ predator control plan. The Districts would undertake this effort with CDFW support even if not included in the FERC license; therefore, the Districts’ predatory control plan is reasonably certain to be undertaken in the next license term and should continue to be included in FERC’s assessment of the different scenarios of alternative future conditions.


“...CDFW argues that fall pulse flows are necessary and have a major effect on the numbers of [fall-run Chinook] moving into the system.”

Again, CDFW is expressing an opinion, and does not provide substantial evidence of the need for fall pulse flows. While in some years fall-pulse flows may appear to influence movement of fall-run Chinook salmon, there is no data or analysis in the record, or any provided by CDFW, to conclude that fall-pulse flows have “a major effect”, or any effect at all, on the “numbers” of fall-run Chinook eventually returning to the Tuolumne (i.e., total escapement).


“FERC staff relies on the Districts’ commitment during spill years to ‘make reasonable efforts’ to shape the descending limb of the snowmelt runoff hydrograph to mimic natural conditions that would promote riparian recruitment...CDFW requests to go beyond the expectation of reasonable efforts and identify a specific recession rate that the Districts’ be required to meet on different water year types.”

The descending limb of a spring runoff event does not have to “mimic natural conditions” to promote riparian recruitment. No two water years are exactly alike and attempting to prescribe a specific recession rate for each water year type is not likely to result in mimicking natural conditions. Making reasonable efforts to shape the descending limb of the actual runoff event has a greater chance of promoting recruitment than establishing a blanket formula for how to manage the end of the runoff period.


“CDFW and the USFWS recommended the Districts develop a TRMP for the La Grange Project. CDFW

The La Grange Project is essentially a run-of-river project with total daily project inflow equal to total daily project


would like to understand why this recommendation was not accepted by the Districts.”

outflow. Run-of-river is, in essence, a form of management plan.


“...the current license includes ‘interpolation water’ which has been used for a fall pulse flow or to increase water volume in the Tuolumne River. CDFW is concerned that this license is unlikely to include ‘interpolation water...’”

As proposed by the Districts to simplify the determination of water year type and associated pulse and base flows, the April determination of water year type will designate the required instream flows based on the flows required to be released in each of the five potential water-year types. The Districts are not proposing to use a sliding scale extending from one “boundary” of one water year type to the next. The Districts’ proposed Spill Management Plan provides for the possibility of retaining water for a fall pulse flow under certain conditions of Spill, as generally described in the USFWS’ filing with FERC dated October 1, 2018.


“It would be beneficial for gravel augmentation to continue down to the end of the gravel-bedded reach, beyond RM 39, since CDFW data shows that fish [i.e., fall-run-Chinook] use those areas...”

The Districts’ proposed LTRHIP is intended to provide just such potential improvements. The Districts’ separate Coarse Sediment Management Plan is designed to improve spawning conditions in the dominant spawning reach which is located upstream of RM 39.


“...CDFW does not concur with the conclusion that providing 5 cfs to the La Grange plunge pool would have a substantial effect on the TID sluice gate channel because these locations are not connected...”

The Districts concur with the CDFW statement.

35. SWRCB Attachment A; page 2; para 2-3

“The 2018 Bay-Delta Plan is in effect and is posted on the State Water Board “Plans and Policies” webpage...The 2018 Bay-Delta Plan establishes narrative and numeric Lower San Joaquin River flow objectives for the protection of fish and wildlife beneficial uses. Of note, the numeric flow objectives require a percentage of unimpaired flow from February through June from each of the Stanislaus, Tuolumne, and Merced Rivers to protect fish and wildlife beneficial uses.

As the SWRCB notes further in its comment letter on the DEIS, negotiations on the 2018 Bay-Delta Plan are continuing between the Districts, CDFW, CNRA and SWRCB. The Districts, CCSF, CDFW and CNRA have jointly developed an alternative to the SWRCB’s plan described in the SWRCB’s Substitute Environmental Document. This joint submission is referred to as the Tuolumne Voluntary Agreement. It is anticipated that the Tuolumne Voluntary Agreement, in conjunction with voluntary agreements in other watersheds, will be


considered for adoption by the SWB in late 2020 or early 2021.

36. SWRCB Attach A; page 3; para 1

At the December 12, 2018 Bay-Delta Plan adoption meeting, the California Department of Water Resources and California Department of Fish and Wildlife presented information on a potential Delta watershed-wide agreement to protect fish and wildlife. Resolution No. 2018-0059 encourages stakeholders to continue to work together to reach voluntary agreements and present those agreements to the State Water Board for its review as soon as feasible. State Water Board staff continues to provide appropriate technical and regulatory information to assist in the completion of a Delta watershed-wide agreement, including potential flow and non-flow measures for the Tuolumne River, and associated analyses. Such an agreement could result in future amendments to the Bay-Delta Plan.

See response to SWRCB comment immediately above.


The 303(d) list designates impairments near the Projects for the following pollutants or stressors in the Lower Tuolumne River: chlorpyrifos, diazinon, Group A pesticides, mercury, and water temperature.

The DEIS adopts the EPA’s temperature guidance for the Pacific Northwest (EPA, 2003) as a “conservative” basis for evaluating thermal regimes. The Districts dispute the use of EPA (2003) as a basis for listing water temperature impairment. Article 58 monitoring data submitted to FERC annually shows continued spawning success and juvenile production across a range of water temperatures from fall through spring that are at times significantly in excess of EPA (2003) targets. In the FERC Staff analysis, FERC-adopted 7DADM targets are exceeded under both existing


conditions (Base Case) and all other flow proposals evaluated, some which call for instream flows as high as 50% of the unimpaired flow estimate. By this measure all proposals result in unsuitable habitat conditions at almost all locations evaluated in Table 3.3.2-24 and Table 3.3.2-25. The DEIS recognizes that EPA (2003) may not be an appropriate measure of habitat suitability and refers to several peer reviewed studies showing local temperature adaptations by various salmonid species, including O. mykiss. EPA (2011) concludes the issue of whether salmonid populations are adaptable to warmer conditions in California is an open and legitimate scientific question and encourages use of the most up-to-date research to evaluate the impact on fish populations. The Districts’ AFLA provides the best available science on water temperatures, fish conditions, thermal capacity of Tuolumne River and Central Valley salmonids, and temperature suitability applicable to the lower Tuolumne River.


“Although a formal adaptive management process was not included in State Water Board staff’s preliminary conditions, State Water Board staff will include a formal adaptive management process for implementation of the Lower San Joaquin River flow objectives identified in the 2018 Bay-Delta Plan.”

The Districts included a detailed adaptive management plan in the October 2017 AFLA. The same is included as part of the March 1, 2019, submittal made to the SWRCB. The Districts encourage the SWRCB and FERC to adopt this river-specific plan.


“FERC’s Staff Alternative does not provide adequate water quality monitoring to determine compliance with water quality objectives in the Lower Tuolumne River and Don Pedro Reservoir for the duration of their license. The Water Quality Monitoring Plan should include monitoring sites throughout affected river reaches throughout the Lower Tuolumne River

The DEIS recognizes the SWRCB section 401 mandatory conditions described in Section 2.2.5, including Condition 6: Develop a water quality monitoring plan and states “We anticipate that all valid section 401 conditions will be included in any new license issued for the project.”


and Don Pedro Reservoir, not just at locations near La Grange dam and La Grange Reservoir. The Water Quality Monitoring Plan should also include monitoring needed to inform assessment of biological goals, recreation related water quality and bioaccumulation monitoring components, in addition to dissolved oxygen and temperature.”

The Districts agree with the DEIS assessment on Page 3-85 that states “It is unclear how additional bioaccumulation data collected under Water Board preliminary 401 condition 6 would be used to guide project operation. Based on the above, there appears to be little basis for requiring the Districts to monitor recreation-related water quality or bioaccumulation in aquatic organisms.”


“Additional monitoring locations from river mile 26 to the confluence with the San Joaquin River are necessary to measure and track changes in water temperature throughout affected reaches.”

The Districts currently monitor water temperature at two locations downstream of RM 26 (RM 23.6 and 3.5) and would continue to collect water temperature at these locations. Although the Districts are willing to include additional monitoring locations as part of 401 certification, the Districts agree with the assessment by FERC on page 3-102 of the DEIS stating “Temperature effects of the Don Pedro Project diminish as water flows downstream where non-project diversions, irrigation returns, and tributaries have increasing influence on the river’s temperature; therefore, any temperature monitoring below the infiltration galleries, as recommended by California DFW and FWS, would not directly link to project operations.”


“According to California Data Exchange Center, the lowest elevation level Don Pedro Reservoir has reached since 2001 is about 670 feet...Don Pedro’s lowest reservoir elevation during the worst drought since record-keeping began is still 70 feet higher than the trigger elevation at which FERC’s Staff Alternative is requiring the Districts to begin temperature monitoring...Annual seasonal water temperature monitoring for the duration of a 30 – 50 year license at any reservoir level would help

There are no specific “temperature requirements” established for the Tuolumne River; therefore, monitoring reservoir temperatures to “meet temperature requirements” is not warranted.


determine the effects of the Projects’ operations on thermal conditions in Don Pedro Reservoir and the Lower Tuolumne River and offer solutions to adaptively manage Project operations to meet temperature requirements.”

42. SWRCB Attach A; page 5; para 2 – 4

(A) First paragraph states “The FERC Staff Alternative omits several non-flow PM&E measures proposed by the Districts, such as in-channel, riparian, and floodplain improvements in the Lower Tuolumne River.” (B) Second paragraph contains the statement [FERC staff conclude “implementing the Districts’ proposal would likely further benefit juvenile salmonids through the reestablishment of riparian vegetation and its associated increase in prey availability, which appears to be a major limiting factor in the lower Tuolumne River.”] (C) Third paragraph states “the validity of the Districts’ biological models is highly uncertain and remains challenged by outstanding agency comments that were not resolved in the final study reports for the juvenile fish production models. Agency criticisms of the Districts’ biological models include, but are not limited to, concerns that models do not recognize existing rearing and spawning habitat limitations or accurately represent temperature sensitivity, predation, and the effect of flow in establishing rearing and floodplain habitat benefits.

(A) The DEIS includes Additional Measures Recommended by Staff (Section 5.1.2) that includes both a LWM management plan to increase the amount of LWM downstream of the La Grange Diversion Dam and a coarse sediment management plan that includes gravel augmentation in the lower Tuolumne River between RM 39 and RM 52. Additionally, the FERC Staff Alternative includes development of a spill management plan to maximize the benefit of spill events for fall-run Chinook salmon floodplain rearing and increased minimum base flows to support riparian biota. B) The Districts do not dispute that implementation of flow and non-flow measures are relevant; however, the phrase regarding prey availability appears to be a “major limiting factor” in the lower Tuolumne River was disputed in the Districts’ comments on the DEIS as this statement is not only uncited and unsupported, the opposite is the case. The Districts’ various studies of benthic macroinvertebrates and other food sources indicate healthy prey communities in the lower Tuolumne River (see AFLA, Exhibit E, Table 3.5-24). (C) The Districts dispute statements regarding validity of biological models being “highly uncertain.” The models were developed in collaboration with agencies and no statements or other data were submitted across multiple workshops and other review opportunities that support


such characterizations of model validity in the comment. Further, no reports were submitted by any party suggesting habitat limitation for rearing salmonids. The SWRCB was present at model development workshops and did not request additional model calibration or validation results to inform statements regarding model uncertainty. The models represent the best available science and are used appropriately to compare relative salmonid productivity differences of various proposals.

43. The Bay Institute

Page 1 of 6; para 2

“The Project operators have made an assertion, which is also contained in the DEIS, that our proposal would result in 90% rationing in the SFPUC service area...However, that assertion does not accurately reflect our proposal nor how we modeled it.”

The Districts modeled The Bay Institute proposal as it was presented by The Bay Institute in its submittal to FERC. The Bay Institute does not explain how The Bay Institute modeled its own proposal.


Page 1 of 6; para 4

“Our diversion cutbacks were made in order to achieve the ecologically important threshold of providing 50-60% of unimpaired February-June flows necessary to restore and maintain salmonid populations and achieve state and federal salmon doubling mandates.”

Contrary to The Bay Institute’s assertion, the only data or information in the record before FERC on the effect of various flow proposals on fish populations is that provided by the Districts. Neither The Bay Institute nor any other party has provided to FERC any assessment of the effect of their proposal on fish populations. The Districts also made this point in comments on the SWRCB’s SED, and separately have filed these comments with FERC. By the SWRCB’s own estimate of the effects of 50% and 60% unimpaired flow from February through June, both of these percents of unimpaired flow result in fewer fall-run Chinook in the San Joaquin River than the 40% unimpaired flow, according to the SED. Unsupported assertions of effects on fish at the population level must be taken by FERC as just that – opinions not based on any substantial scientific analysis.


Page 2 of 6; para 3

“The agricultural diversions in the model (TID/MID) are where most of the water diverted from the

By this comment The Bay Institute lays bare its desired ultimate goal of reducing the amount of irrigated farmland


Tuolumne River is sent, therefore these diversions must bear the brunt of the cutbacks under any reasonable scenario...Implementing the December 2018 WQCP amendments will necessitate farmland that cannot be sustainably supported—the main question is how much farmland.”

in the Districts’ service territory. This bias colors the many assertions put forward by this party. Moreover, in Scoping Document II, FERC stated that alternatives designed to reduce the consumptive use of water “do not satisfy the NEPA purpose and need for the proposed action and are not reasonable alternatives for the NEPA analysis” prepared for the Don Pedro Project.


Page 4 of 6; para 5

“While we did not provide this supporting information previously, our proposed temperature criteria do have a biological basis as described in Appendices B & C of the Stanislaus SEP...” Note: the table provided by the Bay Institute identifies EPA (2003) temperature guidelines as being “fair,” which is defined as follows “Fair conditions represent those that result in limited temperature-related stress.” Optimal temperature conditions based on the SEP are identified, for Chinook and O. mykiss, as: (1) adult migration = 15.5 °C, (2) spawning/incubation = 12.5 °C, (3) juvenile rearing and migration = 16 °C.

The Bay Institute did not participate in the Don Pedro or La Grange licensing processes, made no comments on the FERC-approved study plans, no comments on any of the draft studies or suite of river-specific models developed as part of the licensing process. Citing to a process on the Stanislaus River provides little to no useful information for the Tuolumne River.


Page 5 of 6; para 2

“The DEIS presents temperature model results for our proposal...however due to the presence of errors in characterizing diversions associated with our proposal, we are assuming that the temperature modeling does not correctly characterize our proposal. The attached reservoir storage time series should allow improved temperature modeling to be done for our proposal.”

On page 14 of Scoping Document II dated July 25, 2011, FERC specifically identified that alternatives intended to reduce the consumptive use of water “do not satisfy the NEPA purpose and need for the proposed action and are not reasonable alternatives for the NEPA analysis” prepared for the Don Pedro Project.


Page 6 of 6; para 1

“While an unimpaired approach is the best way to restore instream flows in this system, the 40% starting point is clearly inadequate as we have demonstrated in our comments on the Phase 1 SED. We continue to believe that our proposal is the best example of a

There is no such demonstration provided by The Bay Institute of the effect of any percent of unimpaired flow on fall-run Chinook populations in the Tuolumne River, the San Joaquin River, or the Bay-Delta. Assertions presented without substantial scientific assessment at the


legally adequate flow regime for the Tuolumne River that is based on biological objectives.”

fish population level must be treated as unsupported opinions.

49. Tuolumne River Conservancy

Page 3 of 56; para 3

“Since both juvenile and adult protection should take precedence over fry, we recommend flows at a minimum 300 cfs level for the entire year.”

TRC provides no biological basis for the recommendation of emphasizing juvenile and adult habitat over fry. The Districts’ flow measures are life stage and seasonal specific.



“The water districts` proposed second infiltration gallery is essentially a new dam (weir) designed to raise the river water level to facilitate a new water diversion pump site.”

This is incorrect. The operation of the second IG is not dependent upon the barrier weir.



“As demonstrated in studies, maintaining cooler water temperatures that favor salmonids and dis-favor predatory nonnative fish is an effective tool to keep the bass downstream from the prime spawning and rearing habitats for salmonids.”

Striped bass have been shown to be prevalent in the Tuolumne River in all seasons throughout the entire reach of the lower Tuolumne River as observed during field studies during the Don Pedro and La Grange licensing processes. The Districts are not aware of what studies TRC is referencing, and no specific citations are provided.

52. CSERC 4 Issue: Water temperature monitoring Position: “We agree with the FERC that the project

operations are affecting water temperature in the Lower Tuolumne River and that temperatures in the lower Tuolumne River are negatively affecting salmonids. FERC’s conclusion regarding water temperatures reflects the need for data collection on water temperature as well as a basic level of monitoring of salmonids in the project-affected reach (the lower Tuolumne River below La Grange).”

Request to FERC: “As stated in USFWS Section 10(j) Condition 6, the FERC should require the Districts to develop and implement a Water Temperature Monitoring Plan within the first year of the new license that includes

The Districts strongly disagree with the assertion that temperatures in the lower Tuolumne River are “negatively affecting salmonids.” This is an assertion that has been so often repeated, parties assume it has biological or scientific support based on data from the Tuolumne River. This is not the case. There is no data or scientific analysis in the record before FERC supporting this assertion. In fact, the record before FERC supports the opposite finding.


the Project reservoirs, Project impoundments, and Project-affected reaches of the lower Tuolumne River.”

53. Conservation Groups (CGs)

9 Issue: DEIS analysis of baseline conditions of the projects and affected areas. Position:

“The DEIS does not accurately establish baseline conditions or identify appropriate mitigation measures based on its baseline conditions. It fails to provide “accurate information” for its description of baseline consumptive use and consumptive demand for Tuolumne River water and fails to provide “defensible reasoning” for its decision not to analyze the overappropriation of the Tuolumne River watershed.”

Request to FERC: “…the FEIS should supplement the analysis

of the proposed action’s baseline. The FEIS should provide an accurate evaluation of the appropriation of the Tuolumne River’s water resources, including the general condition of the overappropriation of the water resources of the Tuolumne River and adjacent groundwater basins”.

The DEIS describes baseline conditions to the extent necessary for FERC’s analysis. In the Affected Environment sections for each resource (Section 3) and elsewhere, the DEIS describes, references, and analyzes numerous empirical study results and modeling conducted by the Districts and other parties, which together provide a characterization of the baseline conditions of the projects and affected environment that is scientifically sound and thorough. FERC states in the DEIS that the No-Action alternative constitutes the baseline condition (i.e., “base case”) for purposes of the NEPA analysis. The action being considered by FERC is the continued generation of hydroelectric power at the Don Pedro Project and the La Grange Project. In Scoping Document II dated July 25, 2011, FERC clearly points out that alternatives addressing consumptive use do not satisfy the NEPA purpose and need for the proposed action. FERC recognizes water rights as being a state issue, and does not wade into matters under the jurisdiction of a state’s water rights law or regulations.

54. Conservation Groups

11-14 Issue: DEIS analysis of baseline conditions – groundwater and surface water resources Position: CGs state:

“The DEIS does not analyze the overappropriation of the Tuolumne River watershed, including groundwater resources, as a baseline condition. Therefore, it

See Districts’ comments provided in Response No. 53 immediately above.


improperly presents incremental improvements as substantive improvements”.

“Surface water resources are overappropriated”.

“Groundwater resources that are dependent on extensive recharge from flood irrigation are part of a baseline overappropriation of combined surface water and groundwater resources.

Request to FERC: “The FEIS should include a description of the

overappropriation of water resources in the area of the projects as a baseline condition. The FEIS should re-evaluate the balancing of resources in consideration of this more complete description of baseline conditions”.


14-15 Issue: DEIS alternatives analysis – range of alternatives Position:

“The DEIS fails to analyze a reasonable range of alternatives”.

“…the DEIS should consider a diverse range of operational alternatives that the State Water Board will likely to consider in making its CWA section 401 decision. This includes the Conservation Groups’ alternatives, which are reasonable means to meet the objectives of the State Board’s Bay-Delta Plan and the purpose and need of the Projects”.

Request to FERC: “…the FEIS must consider a reasonable range

of alternatives, including alternative flow scenarios that better protect fish and wildlife

FERC’s action in this proceeding does not include assessing the State of California’s Bay-Delta Plan.


resources. This obligation applies regardless of whether the dams serve multiple purposes; regardless of whether the Commission has exclusive authority to authorize the alternatives; to satisfy the requirement that cumulative effects be taken into account; and to satisfy the requirement that environmental consequences and mitigation measures be considered”.


16-17 Issue: DEIS alternatives analysis – completeness of Staff Alternative and Mandatory Conditions Position:

“The DEIS does not analyze the implementation of the State Water Board’s adopted Lower San Joaquin flow objectives as an alternative, or consider the ramifications of what would change if the San Joaquin flow objectives adopted by the State Water Board on December 12, 2018 are implemented”.

Request to FERC: “The FEIS must specify flows requirements

for the months of July-September and November-January in an alternative that analyzes the State Water Board’s preliminary § 401 conditions. It must also analyze how such flows would act with flows required by the State Water Board and with other measures to affect the environment and all beneficial uses”.

The DEIS did include an analysis of the SWRCB’s 40% unimpaired flow for the period of February through June. The DEIS showed that the SWRCB’s proposal to require minimum instream flows from February through June equal to 40% of the estimated unimpaired flow produces far fewer fall-run Chinook and O. mykiss juveniles than the Districts’ proposed plan while using much more water.


20-21 Issue: DEIS alternatives analysis – reduced exports alternative Position:

It would not be within FERC’s jurisdiction to require reduction of water exports, if indeed such an alternative was deemed superior. FERC must limit the alternatives it


“The DEIS does not analyze a reduced exports alternative”;

“Conservation Groups’ REA Comments, as in previous comments on scoping, recommended that the EIS evaluate a “Limited Delta Exports alternative” developed based on hypotheses from various specified sources”;

“Alternative mitigations to improve the successful passage of salmonids from the Tuolumne River through the Delta are squarely within the scope of equal consideration for “the purposes of … the protection, mitigation of damage to, and enhancement of, fish and wildlife” under § 4(e) of the FPA, as quoted and cited above. The fact that FERC does not have the authority to require Delta operations does not absolve FERC of the responsibility to evaluate feasible alternatives.”

Request to FERC: “The FEIS should include a Limited Delta

Exports alternative”.

considers to those that could be implemented within FERC’s jurisdiction.


21 Issue: DEIS alternatives analysis - Conservation Groups’ REA recommendations Position:

“…the FEIS should evaluate Conservation Groups’ REA recommendations as a complete alternative under NEPA”.

Request to FERC: “The FEIS should include the Conservation

Groups’ combined recommendations as an alternative under NEPA in the FEIS,

See Districts’ comments in Response No 57 above.


recommended in REA Comments as a “Conservation Groups’ Flow, Habitat and River Management Alternative.” It should include such alternative in its Developmental Analysis. Such inclusion would help to cure the present flaw that the alternatives in the DEIS are not sufficiently distinct from one another”.


23-25 Issue: DEIS analysis of cumulative effects and mitigation measures – floodplain inundation Position:

The analysis in the DEIS of project effects on floodplain inundation improperly excludes the water supply operation of project works from project effects and makes other assumptions that understate cumulative effects.”

In support of their position, the CGS discuss the following items in regards to the DEIS analysis:

o The analysis excludes project releases diverted into irrigation canals;

o The analysis excludes the months of February, May, and June, and therefore neglects the need for mitigation of project effects on floodplain habitat and various species benefits (salmonids, and riparian vegetation);

o The analysis “appears to discount the value of floodplain habitat in the river downstream of spawning reaches…”;

This is incorrect. FERC’s analysis of cumulative effects included an assessment of past, present, and potential future actions that may affect various resources. Water resource development was one of the actions considered.


o The use of averages to analyze the results within water-year types and across water-year types, which has the effect of “telescoping the years with the largest reductions into the average”.

Request to FERC: “The FEIS should evaluate all project effects

on floodplain habitats and species, including the effects of project releases to irrigation facilities, in at least the months of February through June”.


25-27 Issue: DEIS analysis of cumulative effects and mitigation measures – salmonid monitoring Position:

“The DEIS fails to require monitoring of salmonids based on the Districts’ representations about voluntary monitoring and the DEIS’s truncated approach to cumulative effects.”

In support of their position the CGs discuss the following:

o Monitoring activities need to be included as license conditions as relying on the 1995 settlement agreement or on “voluntary representation by the Districts that it shall conduct certain actions” without the appropriate “specificity on performance and compliance” is not sufficient for the DEIS analysis of cumulative impacts;

As noted in the DEIS, the Districts propose to continue salmonid monitoring, including the ongoing snorkeling surveys, RST monitoring, and weir monitoring, under any new licenses issued for the projects. The Staff Alternative also recommends and analyzes monitoring related to salmonid habitat conditions, including water temperature monitoring in Don Pedro Reservoir and the lower river that includes provisions for reporting monitoring results and identifying actions to address water temperatures that exceed the suitable range for salmonid survival; an operation compliance monitoring plan to document compliance with the flow and water level requirements included in the license; and monitoring at large woody material (LWM) enhancement sites in the lower river as part of the LWM management plan. In addition, the SWRCB’s mandatory Conditions 6 and 7 would require the Districts to develop water quality and water temperature monitoring plans, both of which were analyzed in the DEIS.


o FERC’s dismissal of additional monitoring activities and parameters on the basis they do not relate to project operations or would inform future project operations is “incomplete and short sighted”;

o “The Districts overwhelmingly control the operation of the lower Tuolumne River. It is reasonable and appropriate that they should monitor the fish in that reach of river.”

Request to FERC: “Prior to issuing the FEIS, staff should

consider what monitoring should be required in the licenses to ensure the effectiveness of proposed mitigation measures.”


31 Issue: Mitigation measures for effects on sedimentation and floodplain inundation Position:

“The DEIS fails to provide appropriate mitigation measures to remedy project effects on sedimentation and floodplain inundation.”

In support of their position the CGs discuss the following:

o The DEIS analysis and recommendation for the “quantity of required mitigation gravel is convoluted” and “arbitrarily and capriciously further reduces the required coarse sediment mitigation…”

o FERC’s rationale in the DEIS for limiting mitigation to salmon

As part of the Staff Alternative, FERC recommends that the Districts develop a plan to augment gravel annually for the term of any new license. FERC states that gravel augmentation is needed to mitigate the impacts of the projects on salmonid spawning habitat downstream of La Grange Diversion Dam, and recommends this mitigation for the entire license term instead of the 10-year coarse sediment augmentation program proposed by the Districts. Rather than attempting to fill the SRPs which are not related to the projects, FERC recommends that the coarse sediment management plan should focus on providing high quality spawning habitat for anadromous salmonids in the first 12.4 miles downstream of La Grange Diversion Dam. This reach, which includes the primary salmonid spawning reach, is slowly degrading in response to a reduction in coarse sediment supply and thus has the greatest potential for increased salmonid production in


spawning beds, discounts “the fact that gravel capture by project reservoirs is ongoing project effect in and of itself that requires mitigation in and of itself”;

o The DEIS description of actions by the Districts since 1995 that have reduced the effects of gravel capture is opaque;

o The DEIS “fails to account for the removal of large amounts of gravel from the lower Tuolumne River floodplain during the construction of Don Pedro Dam”.

Request to FERC: “The FEIS should re-analyze project effects

on the input and transport of both fine and coarse sediment and gravel. License conditions should specify amounts and timing of sediment replenishment, not simply leave those amounts to be determined. The FEIS should evaluate amounts that mitigate for past and ongoing, not just future, project effects.”

“In addition, the FEIS should evaluate license conditions to mitigate for the frequency and amount of floodplain inundation lost due to project operation. The FEIS should consider license conditions that include a calculated combination of instream flow and physical alteration of the streambed to restore the floodplain habitat lost caused by project operation. See Conservation Groups’ REA Comments, pp. 55-67.”

response to gravel augmentation. FERC’s recommendation would appropriately mitigate project effects because it would require an annual addition of gravel commensurate with the project’s ongoing level of impact to downstream spawning gravels as documented in the Districts’ study W&AR-04: Spawning Gravel in the Lower Tuolumne River. As stated previously in the Districts’ reply to the agencies’ and conservation groups’ requests, the Districts should not be required to mitigate any hypothetical project effects related to floodplain inundation because relicensing studies show that access to floodplain rearing habitat routinely occurs on a 2-4 year recurrence interval, depending upon duration. Nether in-channel rearing habitat nor salmonid access to floodplain habitat has been shown to limit salmonid productivity in the lower Tuolumne River (Stillwater Sciences 2017, HDR and Stillwater Sciences 2017).



43-44 Issue: Conclusions in the DEIS violate NEPA because they lack evidentiary basis and are therefore arbitrary and capricious – flow measures Position:

“The DEIS provides neither information nor analysis that shows that the Districts’ proposed flow measures have comparable benefits to aquatic habitat conditions as the flow measures proposed by agencies and Conservation Groups.”

Request to FERC: “In order to provide analysis of how “more

normative ecological processes” would benefit the aquatic environment, the Final EIS should start with evidence in the record such as the 2010 Delta Flow Criteria Report that describe and analyze the benefits of a more natural flow regime.200 For the same reasons, staff should also review and analyze the final Substitute Environmental Document for Phase I of the update of the Bay-Delta Water Quality Control Plan (2018)201 and the Water Quality Control Plan for the San Francisco Bay/Sacramento-San Joaquin Delta Estuary (2018).202 Staff should explicitly contrast the benefits that these documents describe from more natural flows in the Tuolumne River with the benefits staff ascribes to Districts’ flow recommendations.”

The statement in the DEIS that aquatic habitat conditions under the Districts’ proposal would be similar to those under the resource agencies/stakeholders’ recommendations is most logically attributed to FERC’s evaluation of WUA and water temperature modeling results that show increases in salmonid production similar or greater in magnitude to production increases resulting from the flow measures proposed by agencies and Conservation Groups. FERC points out that the Districts’ proposed flow regime would improve aquatic habitat conditions compared to the base case and, in contrast to the flow measures proposed by agencies and Conservation Groups, would continue to meet existing and projected water demands in the region. FERC correctly concludes that any incremental ecological benefits of the resource agencies/stakeholders’ flow regimes over those proposed by the Districts should be weighed against the cost of water used. FERC neither states nor implies in the DEIS that the Districts’ proposal would mimic the magnitude, duration, and timing of the unimpaired hydrograph or “create more normative ecological processes that would likely benefit native anadromous fish populations and their habitat.” The DEIS, therefore, need not provide supporting descriptions or analyses as requested by the Conservation Groups.


51 Issue: Conclusions in the DEIS violate NEPA because they lack evidentiary basis and are therefore arbitrary and capricious – June flows Position:

The DEIS’s recommendation of low June flows for O. mykiss fry is based on peer-reviewed habitat suitability indices documenting relationships between fry suitability and river discharge. These relationships are based on


“The DEIS’s recommendation of low June flows to manage for the fry lifestage of O. mykiss does not account for empirical evidence that demonstrates improved survival of O. mykiss with high June flows”.

Request to FERC: “In any finding based on the record, FERC

must identify the facts on which it relies, explain why these facts are reliable and relevant, and then demonstrate how the facts support its decision. The DEIS offers no reasoned explanation for why it favored the Districts’ habitat modeling over empirical evidence that shows that higher June flows, not lower flows, favor the survival and proliferation of juvenile O. mykiss in the lower Tuolumne River. The FEIS should correct this deficiency.”

empirical evidence derived from multiple datasets. In their assertion that snorkel survey results from 2008–2011 provide evidence of improved survival of O. mykiss with high June flows, the Conservation Groups rely on the fatally flawed argument that correlation (i.e., O. mykiss survival is higher in high flow years) implies causation (i.e., that high June flows are the cause of the high survival). The Conservation Groups’ conclusion is flawed because it ignores the fact that multiple factors influence survival of O. mykiss fry in the lower Tuolumne River and instead focuses only on flow.


51-52 Issue: Conclusions in the DEIS violate NEPA because they lack evidentiary basis and are therefore arbitrary and capricious – “interim” flows Position:

“Staff offers no evidence that incorporating Districts’ proposed “interim” flows for July through October 15 as the permanent license condition for that time period will provide suitable water temperatures for O. mykiss”.

Request to FERC: “The DEIS provides no evidence that the

action as proposed would provide suitable thermal conditions for O. mykiss in the lower Tuolumne River. The FEIS should change the proposed license condition to require the

Using water temperature modeling results presented in Table 3.3.2-24 of the DEIS, FERC compares the simulated exceedance of salmonid water temperature criteria that would occur under the environmental baseline conditions and each of the proposed and recommended flow regimes. The comparison, which includes the Districts’ proposed interim flows, includes criteria and time periods corresponding to each O. mykiss life stage. Although the modeled salmonid life stages do not specifically include the time period from July 1–October 15, this period is closely approximated by the Central Valley Steelhead, Juvenile Over-summer Rearing period of June–September, which uses a 7DADM water temperature criterion of 18°C. EPA (2003) recommends the 18°C 7DADM criterion to protect migrating juvenile salmonids


flows that Districts proposed as “with infiltration gallery” flows to provide such suitable conditions.”

“Absent such change, the FEIS should provide the output of temperature modeling for the actual proposed license condition. The FEIS should also qualify analysis of staff’s previous temperature modeling to reflect the that there is no existing or known proposed regulatory assurance that the Districts will operate to the modeled flows. In the event that staff does not change its recommended license condition, the FEIS should also explain how the actual condition protects O. mykiss in the lower Tuolumne River during the summer.”

and non-core (moderate to low density) juvenile rearing during the period of summer maximum temperatures.


53-54 Issue: Conclusions in the DEIS violate NEPA because they lack evidentiary basis and are therefore arbitrary and capricious – LTR summer flows Position:

“The DEIS provides no information or analysis to demonstrate that June 1 through October 15 flows of less than 200 cfs downstream of RM 25.5 will mitigate project effects on recreational opportunities and visual quality in the lower Tuolumne River in the Modesto urban corridor.”

Request to FERC: “The FEIS should re-evaluate flows in the

lower Tuolumne River downstream of RM 25.5 in consideration of values other than water supply.”

There is no evidence in the record that current recreational opportunities in the lower Tuolumne River will be adversely affected by the Districts’ proposed flow regime. To the contrary, based on the Districts’ study of boatable flows conducted on the lower Tuolumne River (see RR-03 in the AFLA), the availability of boatable flows will substantially increase under the Districts’ Preferred Plan.



54-57 Issue: Conclusions in the DEIS violate NEPA because they lack evidentiary basis and are therefore arbitrary and capricious – fall attraction flows Position:

“The DEIS’s rejection of fall attraction flows for salmon ignores conflicting evidence in the record based on more comparable data and analysis than the Klamath study cited by staff.”

Request to FERC: “The FEIS should re-evaluate the data

presented here and in the agencies’ and Conservation Groups’ REA comments, and find that fall pulse flows in the lower Tuolumne River are warranted. Fall pulse flows will help migrating adult Chinook salmon find the Tuolumne River.

The allegedly conflicting evidence in the record referred to in the Conservation Groups’ comment—namely the relatively low Chinook salmon escapement in the Tuolumne River when fall pulse flows were absent and relatively high escapement in the Stanislaus River when fall pulse flows occurred—ignores the fact that total Chinook salmon escapement in the Tuolumne River in the years cited by the Conservation Groups (2014 and 2015) was an order of magnitude lower than escapement in the Stanislaus River and had also been considerably lower in preceding years. While fall pulse flows have been shown in some cases to stimulate upstream migration by salmon, there is no evidence that they are correlated with total escapement. The Conservation Groups’ assertion that the greater Stanislaus River escapement was the result of fall pulse flows and the much lower Tuolumne River escapement is due to a lack of fall pulse flows is unsupported and false. Fall-run Chinook salmon escapement in both the Tuolumne and Stanislaus rivers is predominantly comprised of hatchery fish. The supposition that fall-run Chinook returning to their natal streams are aided by fall pulse flows is not supported by any data or analysis.


58-61 Issue: Conclusions in the DEIS violate NEPA because they lack evidentiary basis and are therefore arbitrary and capricious – UTR reintroduction Position:

“The finding in the DEIS that reintroduction of salmonids to the upper Tuolumne River is too speculative to warrant further study is not supported by substantial evidence and is inconsistent with the Commission’s

The Districts’ response to comments dealing with fish passage are provided in Response No 16 above and in Attachment E. Furthermore, all references to sustained runs of steelhead and/or spring Chinook in the Tuolumne River prior to construction of Wheaton Dam in the 1870s are based on anecdotal information. Studies of temperature conditions in the upper Tuolumne River below the occurrence of physical barriers show that over-summering temperatures


obligation to consider feasible mitigation of project effects”.

Request to FERC: “A systematic, problem-solving investigation

of how to address these obstacles is founded in fact and is reasonable. It would get directly to the issues that diverse parties have identified as critical path. Conservation Groups recommend that the FEIS evaluate a license condition that would order such an investigation. Absent such investigation, a project effect that has endured for 120 years will likely go unmitigated for another 40. Conservation Groups believe that the Commission has a responsibility to do more than walk away from it.”

“Staff should also consider staying determination on the remainder of NMFS’s 10(j) study recommendations relating to reintroduction of salmonids to the upper Tuolumne River pending the Commission’s evaluation of the outcome of this investigation.”

significantly exceed EPA (2003) temperature metrics, indicating, to the extent EPA (2003) can be relied on, that there was thermally suitable habitat for these species in the Tuolumne River except possibly during the wettest years (see AFLA 2017; Jayasundara et al. 2017). The problem-solving investigation proposed by the Conservation Groups was undertaken as part of the La Grange licensing proceeding in a collaborative process in which the Conservation Groups participated. The outcome of the various studies conducted showed that reintroduction had a very low probability of success and a very high cost; in other words, was infeasible.


68-71 Issue: Conclusions in the DEIS violate NEPA because they lack evidentiary basis and are therefore arbitrary and capricious – socioeconomic effects (agriculture) Position:

“The FEIS must re-evaluate the socioeconomic effects of water made unavailable to the Districts under different flow recommendations.”

Request to FERC:

See Districts’ comments in Response No 47 above.


“In sum, the DEIS accepts the premise that acre-feet left in the river mean revenue lost to the agricultural economy. The FEIS should consider and analyze feasible alternatives to achieve existing levels of agricultural production with less water.”


REFERENCES Boughton, D.A., S. John, C.J. Legleiter, R. Richardson, and L.R. Harrison. 2018. On the capacity

of Upper Tuolumne and Merced Rivers for reintroduction of steelhead and spring-run Chinook salmon. January 19, 2018. Filed January 24, 2018.

Cramer Fish Sciences–Genidaqs. 2018. Feasibility of successfully introducing anadromous fish into the upper Tuolumne River Basin: Tuolumne River – genetic analysis results of Oncorhynchus mykiss samples Technical Memorandum. Prepared for Turlock Irrigation District and Modesto Irrigation District. Prepared by Cramer Fish Sciences–Genidaqs, Sacramento California and M. Miller, University of California, Davis. January 2018.

EPA. 2003. EPA Region 10 guidance for Pacific Northwest state and tribal temperature water quality standards. April 2003. EPA 910-B-03-002. Available at: https://nepis.epa.gov/Exe/ZyPDF.cgi/P1004IUI.PDF?Dockey=P1004IUI.PDF.

_____. 2011. Letter from A. Strauss, Director, Water Division, United States Environmental Protection Agency, San Francisco, CA, to T. Howard, Executive Director, State Water Resources Control Board, Sacramento, CA. October 11, 2011.

Farrell, A., N. Fangue, and K English. 2017. Districts’ Response to NMFS Comments on the Draft Report for the Thermal Performance of Wild Juvenile Oncorhynchus Mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature. September 2017. Attachment to Amended Final License Application filed with FERC October 11, 2017.

FERC. 2019. Draft Environmental Impact Statement issued February 11, 2019. HDR and Stillwater Sciences. 2017. Lower Tuolumne River floodplain hydraulic assessment

study report. Don Pedro Project, FERC No. 2299. Prepared for Turlock Irrigation District, Turlock, CA, and Modesto Irrigation District, Modesto, CA. February 2017.

McBain & Trush (McBain & Trush, Inc.). 2004. Coarse sediment management plan for the lower Tuolumne River. Revised Final Report. Prepared for TRTAC; Turlock and Modesto Irrigation Districts; U.S. Fish and Wildlife Service, Anadromous Fish Restoration Program; and California Bay-Delta Authority. McBain and Trush, Inc., Arcata, CA.

Stillwater Sciences. 2017. Oncorhynchus mykiss habitat survey, amended study report. September. Don Pedro Project, FERC No. 2299. Prepared for Turlock Irrigation District, Turlock, CA, and Modesto Irrigation District, Modesto, CA. Filed October 11, 2017, with amended final license application.

Turlock Irrigation District and Modesto Irrigation District. 2012. Don Pedro Reservoir Bathymetric Study Report. Prepared by HDR Engineering, Inc. October 2012. Filed with FERC as part of the Amended Final License Application October 11, 2017 (W&AR-03).

_____. 2013. Lower Tuolumne River Lowest Boatable Flow Study (RR-03). Attachment to Don Pedro Hydroelectric Project Draft License Application. December 2013.

_____. 2013. Spawning Gravel in the Lower Tuolumne River Study Report (W&AR-04). Attachment to Don Pedro Hydroelectric Project Updated Study Report. December 2013.

_____. 2013. Salmonid Population Information Integration and Synthesis Study Report (W&AR-05). Attachment to Don Pedro Hydroelectric Project Draft License Application. December 2013.

_____. 2016. Fall Migration Monitoring at the Tuolumne River Weir 2015 Annual Report. Report 2015-6 in 2015 Report of Turlock Irrigation District and Modesto Irrigation District Pursuant to Article 58 of the License for the Don Pedro Project, No. 2299. Prepared by


FISHBIO, LLC Oakdale, CA. March _____. 2017. Oncorhynchus Mykiss Habitat Survey Amended Study Report (W&AR-12).

Prepared by Stillwater Sciences. Attachment to the Amended Final License Application. September 2017.

_____. 2017. Oncorhynchus mykiss habitat survey, amended study report (W&AR-12). September. Don Pedro Project, FERC No. 2299. Prepared for Turlock Irrigation District, Turlock, CA, and Modesto Irrigation District, Modesto, CA. Attachment to Amended Final License Application (accession no. 20171011-5067).

_____. 2017. Thermal Performance of Wild Juvenile Oncorhynchus Mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature. Final Report (W&AR-14). Attachment to Amended Final License Application. February 2017.

_____. 2017. Amended Final License Application for the Don Pedro Project. Filed with FERC October 11, 2017.

_____. 2017. Final License Application for the La Grange Project. Filed with FERC October 11, 2017.

_____. 2018. March 14, 2018 Reply Comments of TID/MID in response to comments, recommendations, and preliminary terms and conditions filed by agencies and stakeholders. Filed with FERC on March 15, 2018.

_____. 2018. May 14, 2018 Filing of Response to February 16, 2018 Request for Additional Information, Resource Agency Late Filing and other Related Information.

_____. 2018. July 30, 2018 Filing of Additional Information Requested by FERC on July 3, 2018. _____. 2019. Comments on DEIS for Don Pedro Project P-2299 and La Grange Hydroelectric

Project P-14581. www.FERC.gov E Library. Filed with FERC April 12, 2019. Verhille C.E., K.K. English, D.E. Cocherell, A.P. Farrell, and N.A. Fangue. 2016. High thermal

tolerance of a rainbow trout population near its southern range limit suggests local thermal adjustment. Conservation Physiology 4. DOI: 10.1093/conphys/cow057. Available at: https://www.researchgate.net/publication/311611607_High_thermal_tolerance_of_a_rainbow_trout_population_near_its_southern_range_limit_suggests_local_thermal_adjustment.

ATTACHMENT B

Lower Tuolumne River Habitat Improvement Program

Initial Group of Recommended Habitat Restoration Sites

LTRHIP DRAFT-2 Initial Group of Habitat Restoration Sites Page 1 June 11, 2019

Lower Tuolumne River Habitat Improvement Program Initial Group of Recommended Habitat Restoration Sites

1.0 Background and Purpose

Turlock Irrigation District and Modesto Irrigation District (collectively, the Districts) are co-licensees of the Don Pedro Hydroelectric Project (Project) located on the Tuolumne River. On October 11, 2017, the Districts filed an application to the Federal Energy Regulatory Commission (FERC or Commission) for a new license to continue operating the Project. On January 29, 2018, the US Fish and Wildlife Service (USFWS) and other parties filed comments and recommended terms and conditions with FERC. Following receipt of resource agency and other parties’ recommendations, FERC required the Districts to model the various proposals submitted, and on May 14, 2018, the Districts filed the resulting model runs. Shortly after the May 14, 2018, filing by the Districts, the USFWS and the Districts entered into discussions to explore areas of common ground. On October 1, 2018, the USFWS filed revised Section 10(j) recommendations with FERC, and on November 14, 2018, the Districts filed an Amendment to their Preferred Plan adopting the USFWS revised Section 10(j) recommendations.

The Lower Tuolumne River Habitat Improvement Plan (LTRHIP) is a key component of the USFWS revised Section 10(j) recommendations and the Districts’ amended Preferred Plan1. On February 11, 2019, FERC Staff issued a Draft Environmental Impact Statement (DEIS) for the Don Pedro Project. In the DEIS, FERC Staff did not recommend adoption of the LTRHIP into the new license. FERC Staff identified certain concerns regarding the proposed Lower Tuolumne River Habitat Improvement Program (LTRHIP) as detailed below:

“We do not recommend developing a plan to implement the Lower Tuolumne River Habitat Improvement Program because it is unclear: (1) precisely what habitat restoration projects would be funded, (2) where those projects would be located in the lower river, (3) how the Districts would obtain the rights needed to access a property for restoration and maintenance activities for each proposed improvement site, (4) how compliance with the ESA and National Historic Preservation Act (NHPA) would be obtained at each site, and (5) the details on the project design and scope of operation and maintenance activities that would occur at each habitat improvement site so that the Commission can determine whether the site should be included within the project boundary.”

In the present report, the Districts address each of the concerns FERC identified in the DEIS related to the LTRHIP. With this additional information, the Districts request that FERC include the LTRHIP, and related Tuolumne Partnership Advisory Council (TPAC), as a condition of the new Don Pedro license as provided for under Section 4 of the Federal Power Act’s mandate to give equal consideration to the purposes of enhancement of fish and wildlife, “including related spawning grounds and habitat” and Section 10(a)(1) in that any best adapted plan shall include “enhancement of fish and wildlife”.

1The LTRHIP is distinct from, and does not include, the Districts’ proposed Coarse Sediment Management Plan, designated as RPM-1, as contained in the AFLA, Exhibit E, Section 5.5.1. Under RPM-1, the Districts proposed to improve specific spawning riffles in the upper reaches of the lower Tuolumne River over a ten-year period using 75,000 tons of coarse sediment.

ATTACHMENT B


Described herein is an initial set of instream and floodplain habitat improvement and restoration projects that are recommended to be funded and implemented under the LTRHIP subject to consultation with the TPAC and receipt of permits. Included for each individual project are the following:

Detailed descriptions and preliminary drawings of the initial group of recommended projects.

Locations of each project. Property ownership and summary of the necessary rights for access and implementation

for each project. Details on project design and scope of monitoring and maintenance activities. Detailed description of how compliance with the Endangered Species Act (ESA) and

National Historic Preservation Act (NHPA) would be achieved. It is evident from both the USFWS’ October 1, 2018, filing and the Districts’ subsequent November 14, 2018, amendment to their Preferred Plan, that the LTRHIP is an important and extensive program with a total capital funding of $38 million and up to $1 million per year in annual expense funding for monitoring and maintenance. Individual project size would generally be prioritized to be less than $5 million in capital cost. It is likely that a dozen or more projects would receive capital funding. Prudent planning demands sufficient time be allotted to carefully select and vet individual projects and that lessons learned when implementing the first several projects be employed on subsequent efforts. However, owing to the extensive river habitat assessment work performed by the Districts and other parties dating back to the year 2000, leading up to the Coarse Sediment Management Plan (CSMP) report (McBain and Trush 2004), and the prior conduct of several successful habitat improvement projects (see AFLA, Exhibit E, pg 3-165), the Districts are in a position to be able to identify a first set of four (4) specific projects that address the concerns raised by FERC Staff in the DEIS. These projects are described in the sections below and in the attached documents. It is expected these projects would be fully implemented within the first five (5) years of the term of the new license, including all permitting, bidding, and construction. The next set of three (3) to five (5) projects would be submitted to FERC for approval in Year six (6) of the new license, consistent with the LTRHIP. 2.0 Goal of the Lower Tuolumne River Habitat Improvement Program The lower Tuolumne River downstream of La Grange Diversion Dam provides habitat for native salmonid species (fall-run Chinook salmon and O. mykiss). The channel and associated floodplains have been substantially modified over the last 150 years through in-channel gold and gravel mining activities, agricultural and water resource development, levee construction, and urban encroachment. The goal of the LTRHIP is to provide capital funding for design, permitting, and construction of specific non-flow measures involving instream habitat improvements and riparian and floodplain restoration, along with funding for subsequent monitoring and maintenance, all intended to benefit native salmonid species. Types of projects include those that provide improved connectivity between the river channel and adjacent floodplains, improved in-channel habitat conditions, slough development, and in-channel structural complexity (e.g., LWM installation). The first priority would be given to the upper 25-mile, gravel-bedded reach of the lower Tuolumne River.


There is a substantial history of study and investigation of the habitat conditions in and potential for improvements to the lower Tuolumne River. The Tuolumne River Restoration Plan (McBain and Trush, 2000) and the McBain and Trush July 2004 “Revised Final Coarse Sediment Management Plan for the Lower Tuolumne River” provide the foundation to assist with identification and planning of potential restoration actions that would meet the goals and objectives of the LTRHIP. Habitat studies conducted by the Districts during the ongoing relicensing process for the Don Pedro Project as reported in Exhibit E of the AFLA also serve to update and inform environmental enhancements for the Tuolumne River. This combined body of work is cited to enable FERC Staff to understand the amount of relevant background material available on the lower Tuolumne River. Having this extensive background information available begins to address FERC’s concerns about what projects would be undertaken and where they would be located. For the Tuolumne River, it is not a question of a lack of potential resource improvement projects, but which projects would have the greatest resource value consistent with the goals of the LTRHIP. Careful selection of projects and prudent design implementation drive the LTRHIP schedule and the role of the TPAC to ensure best use of the funds available. Based on the goals of the LTRHIP, the Districts have identified the first four (4) projects recommended to be implemented under the program. While additional consultation with all resource agencies is desired by the Districts and permitting of each site remains to be completed, each project has been sufficiently vetted and preliminary designs adequately detailed to enable reasonable certainty of moving the projects quickly to final design (following permitting) and construction once the new license is issued. A general discussion of the various permitting processes and the role of the TPAC members in these processes is provided below to address FERC’s general concern about efficient and proper handling of the various federal and state regulatory processes to be undertaken for each project. As reported below, resource agencies have developed a set of agreements related to the expedited consideration of permit requests for habitat improvement projects in the Central Valley region of California. Given the existence of the coordinated permitting process and the fact that each of these first four (4) recommended projects have existing preliminary designs and envision beneficial restoration actions previously implemented in other nearby locations, FERC Staff can be assured that each of the five areas of concern raised in the DEIS shall be adequately addressed in these projects. Furthermore, for those projects needing private landowner involvement, such coordination has already been underway. The sum of these previous efforts allow the Districts to conclude that it is highly likely this first set of four projects will be fully implemented and constructed within the first five years of the new license. In Year 5 of the new license, and every five years thereafter, the Districts will prepare a report covering the status of all work conducted and underway associated with the LTRHIP and, after review and comment by TPAC members, file a final report with FERC. The five-year status reports will also include information on the performance, monitoring, maintenance, and condition of each project undertaken as part of the LTRHIP.


3.0 LTRHIP Sites Recommended to be Implemented First A general location map of the four (4) projects is provided below in Figure 1; and Table 1 provides an overview of each of the projects. Appendix A contains a more detailed descriptions of each project, a presentation of the information already available for each project, the scope of work to be performed consistent with the goals of the LTRHIP, and a description of how each project addresses the five areas of concern raised by FERC Staff in the DEIS.


Figure 1. Location of the first group of LTRHIP improvement projects.


Table 1. Proposed first group of habitat restoration sites

RM Site Name and Goal Site Description Design Status

Estimated Cost (2020 $)

49.9- 50.1

Riffle 1C; Improve juvenile Chinook and O. mykiss Rearing

Upstream of New La Grange Bridge: Place approximately 40,100 cy of coarse sediment for:

Construction of alternating bars, creating a contemporary low flow channel width of 90 ft

Addition of two new spawning riffles Filling of a remnant mining pit on the south bank of the channel to an

elevation that is frequently inundation by contemporary flows

Initial 30% contour design - ready for review

$1.1 million

49.2- 49.6

Buck Flat Riffle 3A/3B Complex; Improve Chinook juvenile rearing

Downstream of New La Grange Bridge: Remove the concrete rubble, the old bridge sheet-piling abutments, and other prior construction materials. Place approximately 28,400 cy of coarse sediment for:

Construction of two new spawning riffles (inclusive of pool tails) Construction of alternating point bars on the north and south banks Enhancing an existing scour channel along the south bank (a result of

the 1997 flood) to provide off-channel rearing habitat and reduce potential salmonid stranding

Grading and lowering of adjacent surfaces and floodplain area to interact with contemporary flows

Initial 30% contour design - ready for review

$3.5 million

43.8- 44.6

Bobcat Flat Phase III (Bobcat East); Improve Chinook juvenile rearing and adult spawning

Bobcat Flat Phase III 30% Design Report – Final; McBain Assoc.; April 18, 2019:

Reconnect the river to its floodplains Establish more natural pool-riffle morphology Increase off-channel rearing habitat Total Cut 550,540 cy; Total Fill 55,600 cy Re-vegetate riparian zone as shown on plan Total Area of Improvements: 76.8 acres

30% design complete; Permitting underway

$6.0 million


RM Site Name and Goal Site Description Design Status

Estimated Cost (2020 $)

41.8 Riffle 24 (TLSRA); Improve in-channel salmonid habitat

Turlock Lake Area: Place 3,100 cy of coarse sediment to enhance in-channel salmonid

habitat at Riffle 24 and provide coarse sediment supply for downstream transport

Conceptual design complete; contour design ready for review

$750,000


4.0 Permitting of Potential LTRHIP Sites As a member of the TPAC, the USFWS will partner with the Districts to assist with permitting of restoration projects funded by the LTRHIP. The USFWS implements multiple aquatic and riparian habitat restoration projects each year throughout the Central Valley of California to benefit anadromous fish, including fall-run Chinook salmon and O. mykiss. These efforts include extensive environmental compliance and permitting activities. General project types include, but are not limited to, fish passage structures/facilities, screening of diversions, in-stream gravel augmentation, floodplain creation/reconnection, riparian vegetation establishment, and several combinations of those general elements within a single project footprint. As an active partner in aquatic and riparian habitat restoration projects throughout the Central Valley in general, and the successful implementation of LTRHIP specifically, USFWS has committed to provide leadership, support, and state agency coordination on all applicable environmental compliance activities for the specific projects developed under LTRHIP. Examples of recent projects similar to those expected to be developed and implemented as part of LTRHIP and which have been successfully permitted by USFWS are summarized in Table 2 below: Table 2. Example restoration projects successfully permitted by the USFWS. Project

Year Permitting Completed

Relevant Activities

Merced River Ranch (Merced River)

2010 Gravel augmentation, floodplain lowering and contouring, and on-site mine tailings processing

Bobcat Flat – Phase I/II (Tuolumne River)

2010-2011 Gravel augmentation, floodplain lowering and contouring, revegetation and on-site processing of armored bank material

Snelling/Henderson Park (Merced River)

2012 Gravel augmentation, floodplain lowering and contouring, and on-site mine tailings processing

Bobcat Flat – Duck Slough (Tuolumne River)

2015 Floodplain lowering, on-site processing of coarse sediment, conversion of dredger slough into low flow side channel habitat

Buttonbush Restoration Project (Stanislaus River)

2016 Gravel augmentation, floodplain lowering and contouring, and on-site processing of armored bank material

Mokelumne Spawning Habitat (Mokelumne River)

2009-Present Annual augmentation of gravel, floodplain lowering and contouring

Directly related to the Endangered Species Act (ESA) compliance concerns raised by FERC Staff, USFWS has worked extensively to enable restoration staff in their Fish and Aquatic Conservation Program to be able to complete the bulk of ESA permitting consultations for aquatic and riparian habitat restoration projects. This includes both internal ESA permitting consultations with USFWS Ecological Services (for non-salmonid fish and all other listed species) and external ESA permitting consultation with the National Marine Fisheries Service (NMFS; for listed salmonids).


These efforts have yielded three main outcomes. First, ESA permitting consultations for beneficial aquatic/riparian habitat restoration projects can now regularly be completed in under six (6) months. Second, USFWS is listed as an implementing agency under a relatively new statewide programmatic agreement with NMFS for beneficial aquatic and riparian habitat restoration projects. Third, USFWS staff will also be an implementing agency under a complementary programmatic agreement with USFWS Ecological Services expected to be completed by the end of 2019. Both of the programmatic agreements allow the types of beneficial aquatic and riparian habitat restoration projects proposed to be implemented as part of LTRHIP to receive ESA compliance within 1 to 2 months of final project design completion. The programmatic agreements assess potential impacts to relevant ESA-listed species for a suite of project types, including all of those that are expected to be pursued under LTRHIP and functionally complete the required ESA permitting consultations with a final review of a limited set of site-specific details for individual proposed projects. As an active partner in LTRHIP, USFWS has committed to utilizing these programmatic agreements to expedite ESA compliance for projects. Regarding compliance with Section 106 of the NHPA, the Districts’ consultant on NHPA for the Don Pedro and La Grange projects developed Historic Properties Management Plans (HPMP) for each project. The successful development of the two HPMPs followed a five year effort involving research into local and regional historic and cultural resources, extensive field investigation along the Tuolumne River, close coordination with local Indian tribes, and consultation with State Historic Preservation Office, all within the context of FERC relicensing and compliance. This wealth of experience enables HDR specialists to understand the local historic and cultural resources context, expedite field investigations and coordination with interested Tribes, and consult with State Office of Historic Preservation. The Don Pedro HPMP serves as a programmatic blueprint for planning and conducting related work along the Tuolumne River.

5.0 Conclusion In summary, each of the initial four projects planned to be implemented under the LTRHIP, subject to final TPAC coordination and successful permitting with the assistance of the USFWS, addresses the concerns raised by FERC Staff in the DEIS. The scope and preliminary design of each is completed or well underway (describing what will be built); the location of each project is precisely known (where it will be built); discussions with private property owners, where necessary for access or land use, are underway; an efficient permitting process is ensured through existing agency coordination agreements and HPMPs; and the scope of future maintenance is sufficiently understood to enable final access agreements to be put in place in a reasonable time frame. In Appendix A to this summary, each of the four projects is described in detail along with how each individually addresses the five concerns raised in the DEIS.


Appendix A Detailed Project Descriptions of Recommended Restoration Projects


RIFFLE 1C


Riffle 1C

New La Grange Bridge Area

The Riffle 1C Habitat Restoration Project addresses the concerns FERC identified in the DEIS related to the LTRHIP:

Construction, monitoring, and maintenance of in-channel and floodplain improvements to improve juvenile fall-run Chinook and O. mykiss rearing, consistent with the goals of the LTRHIP;

Located between RM 49.9 and RM 50.1; Rights of access necessary for construction, monitoring, and maintenance activities to be

secured by easement or contract from the primary property owner Stanislaus County for the term of the new license upon completion of final design and approval of permits for construction2; and,

Permits in compliance with all regulatory requirements, including ESA and NHPA, to be led by USFW under its agreements with state and federal agencies pertaining to review of stream and floodplain rehabilitation projects or related to NHPA, through direct consultation with SHPO and Tribes using protocols adopted for the Don Pedro and La Grange projects.

The project description below provides details of the conceptual project design, monitoring, and maintenance activities. Project Description Prior to 1997 Riffle 1C was located under New La Grange Bridge (RM 49.95). During the 1997 flood Riffle 1C was scoured and partially depleted of its coarse sediment. In addition, along the south bank directly upstream of La Grange Bridge at RM 50.0 a remnant mining pit is connected to the channel. In its current state, the Riffle 1C site is a run pool containing little channel complexity and salmonid habitat. Additionally, floodplain surfaces adjacent to Riffle 1C are largely disconnected from contemporary Tuolumne River flows. The restored project will provide for initiation of floodplain connectivity at river flows in the range of less than 1,000 cfs. Channel restoration of the Riffle 1C site includes placement of approximately 40,100 cy (including a design estimate of 32,100 cy plus an 8,000 cy [25%] contingency) of coarse sediment at this site. The following actions are anticipated to be taken (see Figures 1 and 2):

2 Easement from or contract with Stanislaus County provides sufficient right, title, and interest necessary for ensuring the LTRHIP project will perform as designed over the term of the new license. A final access and construction agreement with the County will be provided to FERC with the usual and customary request for FERC approval for construction. The site does not need to be included within the Project Boundary by virtue of a showing of the rights contained in the agreement with the County.


Construction of alternating bars between Station 2753+00 and 2767+00 (approximately RM 49.9 to 50.1), creating a contemporary low flow channel width of 90 ft;

Addition of two new spawning riffles between Station 2757+70 to 2759+70 and Station 2764+50 to 2767+00; and,

Filling of a remnant mining pit on the south bank of the channel (Station 2760+00 to Station 2764+00) to an elevation that is inundated under contemporary flows.

In addition to the proposed channel restoration action, approximately 30 acres of adjacent surfaces between New and Old La Grange bridges will be lowered to interact with contemporary flows for floodplain connectivity and off-channel habitat enhancement. Excavation of these adjacent surfaces will increase inundation frequency and duration, allow for planted and natural riparian establishment, provide a sediment source for in-channel restoration, and provide foraging and high flow refuge for rearing fall-run Chinook salmon. Figure 3 illustrates the current conceptual design for overall reach improvements between RM 49.0 to 50.5 (Riffles 1C and 3A/3B), including floodplain restoration. The proposed in-channel and floodplain actions will meet overall project goals as follows:

Gravel augmentation in the form of point bars will provide fry rearing habitat, as well as a future source of coarse sediment for downstream routing;

Construction of riffle and pool tails would provide area for salmonid spawning and improved invertebrate production;

Filling of mining pit will reduce predatory habitat, restore floodplain connectivity, and provide off-channel rearing habitat;

Project would preserve existing pool habitat upstream of New La Grange Bridge (RM 50.0) as well as Riffle 1B located upstream of the project; and,

Lowering of adjacent surfaces will increase inundation frequency and duration, providing juvenile salmonids foraging opportunities and refuge during higher flows; will improve success of riparian planting and natural regeneration; and provide a local source of coarse sediment for gravel augmentation projects.

Monitoring and Maintenance Monitoring of Riffle 1C Project will be conducted following construction of the improvements. The objectives of the Monitoring Plan will be to assess:

if the project was satisfactorily implemented as designed; which project goals were met; how project features persist and function through time and over a variety of flow

conditions; and, how lessons learned on this project can inform future restoration efforts.


Preliminary assessments, metrics, and protocols to be used for monitoring are summarized in the table below. The TPAC will advise what additional implementation and effectiveness monitoring may be warranted to verify and document that the project was built according to design and to document that project goals were met. The information collected will be used to inform future restoration projects undertaken as part of the LTRHIP. To the extent available and applicable, the LTRHIP and TPAC may utilize data obtained from other monitoring efforts being conducted on the lower Tuolumne River. Assessment Metric Protocol As-built survey Comparison of as-built survey

to 100% design GIS/CAD layer overlay to compare the two surfaces and calculate overall deviations

Riparian revegetation success Number of plants surviving after one year

Census of planted poles to determine the number deceased vs alive

Natural riparian recruitment Area of naturally established native riparian vegetation within the project footprint

Photographic documentation and estimate of area of naturally established native vegetation (square meters)

Side channel persistence Presence/absence of hydrologically connected side channel entrances and exits connecting the feature to the mainstem river

Photographic documentation over time

As-built surveys to assess new rearing habitat area and new floodplain area will be completed immediately after construction. Riparian revegetation success will be assessed one year after planting. Side channel persistence should be evaluated every three to five years. On-going maintenance activities over the term of the license may include minor gravel augmentation, re-grading, and/or revegetation to facilitate achievement of restoration goals. A report on the monitoring and maintenance activities, assessment of effectiveness of the restoration, and guidance for future maintenance will be prepared and submitted to FERC every five years. These activities will be supported by the LTRHIP annual maintenance fund. Property Rights Properties permanently affected by proposed floodplain and/or in-channel activities are summarized below:

Project Site Proposed Activity Number APN Name 1C Floodplain Work 94 008-025-003 Stanislaus County

1C In-Channel and Floodplain work 99 008-042-004 Stanislaus County

1C In-Channel and Floodplain work 103 008-042-013

Buck Flat Joint Venture et al (associated with Tuolumne River Conservancy)


Easements or contractual agreements will be obtained from Stanislaus County and the Buck Flat Joint Venture for rights to gain temporary and permanent access for construction, monitoring, and maintenance activities over the term of the FERC License. Final easements will be obtained upon completion of final design and approval of permits for construction and submitted to FERC as part of the request for approval of construction. While the Districts contend that the existence of the final easement or contract is sufficient evidence of the necessary right, title, and interest to secure site performance, the Districts are willing upon completion of construction of the Riffle 1C project, to prepare a boundary survey of the final post-construction project. If required, a project boundary can be provided to FERC in a geo-referenced electronic GIS file format consistent with 18 CFR 4.39 mapping specifications.


Figure 1. Tuolumne River, Riffle 1C 30% conceptual design subject to review by and consultation with the TPAC.


Figure 2. Tuolumne River, Riffle 1C 30% conceptual design cross-sections from Figure 1.


Figure 3. Overall restoration potential ready for review by and consultation with TPAC between RM 49.0 to 50.5, Riffles 1C and 3A/3B, including channel and floodplain restoration actions.


RIFFLE 3A-3B

Buck Flat


Buck Flat Riffle 3A/3B Complex

The Riffle 3A/3B Habitat Restoration Project addresses the concerns FERC identified in the DEIS related to the LTRHIP:



secured by easement or contract from the primary property owners Stanislaus County and the State of California for the term of the new license upon completion of final design and approval of permits for construction3; and,

Permits in compliance with all regulatory requirements, including ESA and NHPA, to be led by USFW under its agreements with state and federal agencies pertaining to review of stream and floodplain rehabilitation projects or related to NHPA, through direct consultation with SHPO and Tribes using protocols adopted for the Don Pedro and La Grange projects.

The project description below provides details of conceptual project design, monitoring, and maintenance activities. Project Description Riffle 3A/3B is located downstream of New La Grange Bridge between RM 49.2 and 49.6. The Riffle 3A/3B area contains in-channel rip-rap, grade control boulders, and remnant bridge abutments (concrete and metal debris) that were part of the gravel haul road and bridge used to transport reclaimed dredge tailings to the New Don Pedro Dam construction site in the late 1960s. These remnant bridge materials have affected the morphology of Riffle 3A/3B, reduced spawning habitat availability, and resulted in the loss of invertebrate production. As a result of the 1997 flood the left bank floodplain opposite the remnant haul road and a 300 ft long portion of the right bank at the downstream end of the site were heavily eroded. These eroded areas have created overly wide channels and adjacent scour channels. In general, bank erosion should be encouraged as a natural process, but without adequate coarse sediment to deposit and maintain low-flow confinement, the resulting channel has become a large pool with diminished spawning habitat. This site along with Riffle 1C are the only significant bedload impedance sites between the Old La Grange Bridge and Riffle 5B. Restoring sediment storage at this site would restore bedload transport continuity for 2.9 miles from Old La Grange Bridge nearly to Basso Bridge. 3 Easements from or contracts with Stanislaus County and State of California provide sufficient right, title, and interest necessary for ensuring the LTRHIP project will perform as designed over the term of the new license. Final access and construction agreements with the County and State will be provided to FERC with the usual and customary request for FERC approval for construction. The site does not need to be included within the Project Boundary by virtue of a showing of the rights contained in the agreements with the County and State.


Restoration of the Riffle 3A/3B site proposes removal of concrete rubble, bridge sheet-piling abutments, adjacent haul road fill, and other debris, and placement of approximately 28,400 cy (including a design estimate of 22,700 cy plus an 5,700 cy [25%] contingency) of coarse sediment at this site through the following actions (Figure 1 and Figure 2):

Placement of approximately 7,900 cy for construction of two new spawning riffles (inclusive of pool tails) between Station 2731+20 to 2735+10 and Station 2738+90 to 2741+50;

Cut and fill resulting in net placement of approximately 14,800 cy for construction of alternating bars between Station 2729+00 and 2745+00 (approximately RM 49.2 to 49.8), creating a contemporary low flow channel width of 90 ft; and enhancing an existing scour channel along the south bank (a result of the 1997 flood) to provide off-channel rearing habitat and reduce potential for salmonid stranding.

As part of the proposed channel restoration action, approximately 45 acres of adjacent surfaces between New and Old La Grange bridges will be lowered to interact with contemporary flows for floodplain connectivity and off-channel habitat enhancement. Excavation of these adjacent surfaces will increase inundation frequency and duration, allow for planted and natural riparian establishment, provide a sediment source for in-channel restoration, and provide foraging and high flow refuge for rearing fall-run Chinook salmon. The project will provide for initiation of floodplain connectivity at river flows in the range of 1,000 to 1,200 cfs. In addition, several isolated dredger swales, gravel ponds and wetlands fall within the TRC property to the north of the project. These features could be incorporated into the restoration actions at this site, increasing off-channel salmonid rearing habitat. Riparian vegetation on the south bank of the project site provides important cover for salmonids. Impacts to existing riparian vegetation will be minimized during construction of this project. Figure 3 below illustrates the current design for overall reach improvements between RM 49.0 to 50.5, (Riffles 1C and 3A/3B) including floodplain restoration. The proposed in-channel and floodplain actions would meet overall project goals as follows:

Gravel augmentation in the form of point bars will provide fry rearing habitat, as well as a future source of coarse sediment for downstream routing;

Construction of riffle and pool tails would provide area for salmonid spawning and improved invertebrate production;

Enhancement of adjacent dredge swales, ponds, and scour channels will reduce predatory habitat during periods of low flows, restore floodplain connectivity, provide off-channel rearing habitat, and reduce potential for salmonid stranding;

Project would preserve the existing pool habitat upstream between Riffle 3A and 3B (Station 2735+10 to 2738+90);

Areas of riparian vegetation along the south bank will be preserved to maintain valuable cover for salmonids; and,

Lowering of adjacent surfaces will increase inundation frequency and duration, providing juvenile salmonids foraging opportunities and refuge during higher flows, will improve success of riparian planting and natural regeneration, and provide a local source of coarse sediment for this and other gravel augmentation projects.


Monitoring and Maintenance Monitoring of Riffle 3A/3B Project will be conducted following construction of the improvements. The objectives of the Monitoring Plan will be to assess:


conditions; and, how lessons learned on this project can inform future restoration efforts.

Preliminary assessments, metrics, and protocols to be used for monitoring are summarized in the table below. The TPAC will advise what additional implementation and effectiveness monitoring may be warranted to verify and document that the project was built according to design and to document that project goals were met. The information collected will be used to inform future restoration projects undertaken as part of the LTRHIP. To the extent available and applicable, the LTRHIP and TPAC may utilize data obtained from other monitoring efforts being conducted on the lower Tuolumne River. Assessment Metric Protocol As-built survey Comparison of as-built

survey to 100% design GIS/CAD layer overlay to compare the two surfaces and calculate overall deviations







As-built surveys to assess new rearing habitat area and new floodplain area will be completed immediately after construction. Riparian revegetation success will be assessed one year after planting. Side channel persistence should be evaluated every three to five years. On-going maintenance activities over the term of the license may include minor gravel augmentation, grading, and re-vegetation to facilitate achievement of restoration goals. A report on the monitoring and maintenance activities conducted, assessment of effectiveness of the restoration, and guidance for future maintenance will be prepared and submitted to FERC every five years. These activities will be supported by the LTRHIP annual maintenance fund.


Property Rights The north and south banks of the Riffle 3A/3B site property are owned by the Buck Flat Joint Venture and Stanislaus County. Access to the site is from existing access roads off of J59 near the Hwy 132 intersection. Properties permanently affected by proposed floodplain and/or in-channel activities are summarized below: Project Site Proposed Activity Number APN Name 3A/3B Floodplain Work 37 008-016-011 Stanislaus County

3A/3B In-Channel 45 008-016-033 Marcia J & George M Ingals

3A/3B In-Channel and Floodplain work 46 008-016-034 State of California

3A/3B In-Channel and Floodplain work 99 008-042-004 Stanislaus County

3A/3B In-Channel and Floodplain work 103 008-042-013

Buck Flat Joint Venture et al (associated with Tuolumne River Conservancy)

Necessary easements will be obtained from the State of California, Stanislaus County, Buck Flat Joint Venture, and private property interests for rights to gain temporary and permanent access for construction, monitoring, and maintenance activities over the term of the FERC License. Final easements will be obtained upon completion of final design and approval of permits for construction and submitted to FERC as part of the request for approval of construction. While the Districts contend that the existence of the final easement or contract is sufficient evidence of the necessary right, title, and interest to secure site performance, the Districts are willing upon completion of construction of the Riffle 3A/3B Project, to prepare a boundary survey of the final post-construction project. If required, a project boundary can be provided to FERC in a geo-referenced electronic GIS file format consistent with 18 CFR 4.39 mapping specifications.


Figure 1. Tuolumne River, Riffle 3A/3B 30% conceptual design subject to review by and consultation with the TPAC.


design plan.

Figure 2. Tuolumne River, Riffle 3A/3B 30% conceptual design cross sections from Figure 1 above.


Figure 3. Habitat restoration and improvements between RM 49.0 to 50.5, Riffles 1C and 3A/3B ready for review by and consultation with TPAC, including channel and floodplain restoration actions.


BOBCAT EAST


Bobcat Flat Phase III (Bobcat Flat East)

The Bobcat Flat East Habitat Restoration Project addresses the concerns FERC identified in the DEIS related to the LTRHIP:



secured by easement or contract from the primary property owner Friends of the Tuolumne Inc. (Tuolumne River Conservancy) and other property owners for the term of the new license upon completion of final design and approval of permits for construction4; and

TRC is currently coordinating CEQA/NEPA and ACOE floodplain permit submittals. Permits in compliance with all regulatory requirements, including ESA and NHPA, to be led by USFW under its agreements with state and federal agencies pertaining to review of stream and floodplain rehabilitation projects or related to NHPA, through direct consultation with SHPO and Tribes using protocols adopted for the Don Pedro and La Grange projects.

The project description below provides details of conceptual project design, monitoring, and maintenance activities. Background Bobcat Flat Phase III (Bobcat East) is the final stage of a 334-acre 1.6 mile Tuolumne River restoration effort headed by the Tuolumne River Conservancy (TRC). Restoration of Bobcat Flat (see Figure 1) was initiated in 2003 as a multi-phase project to restore morphologic function and habitat for target species within the Dredger Tailing Reach of the Tuolumne River. To date, three restoration projects within the Bobcat Flat West property have been designed, permitted, and constructed. Combined, these three projects have restored approximately 18 acres of remnant scraped dredger tailing surfaces to functional riparian floodplains, restored geomorphic function and salmonid habitat by sorting and placing coarse sediment into 0.6 miles of the Tuolumne River mainstem channel, and constructed approximately 0.4 miles of high quality low-flow side channel salmonid rearing habitat by placing coarse sediment in an existing slough created by historical gold dredging. The Phase III project began in 2017 with a grant from U.S. Fish and Wildlife (USFWS) to collect existing site data, prepare two conceptual design alternatives, prepare a 30% design of the preferred alternative, and prepare supporting analysis and documentation. 4 Easement from or contract with Friends of the Tuolumne Inc. (Tuolumne River Conservancy) provides sufficient right, title, and interest necessary for ensuring the LTRHIP project will perform as designed over the term of the new license. A final access and construction agreement with the Conservancy will be provided to FERC with the usual and customary request for FERC approval for construction. The site does not need to be included within the Project Boundary by virtue of a showing of the rights contained in the agreement with the Conservancy.


Project Description Bobcat Flat Phase III encompasses approximately 190 acres and 0.8 miles of the Tuolumne River (RM 43.8 to 44.6) within the Bobcat Flat East property. Bobcat Flat Phase III is focused on creating a dynamic system where ecological and geomorphic aspects evolve together. The primary objectives of this project include:

Scale surfaces adjacent to the mainstem channel and reconnect the river to its floodplains so it can function under the contemporary regulated flow regime;

Reconstruct the lake–cascade channel morphology to be a more natural pool–riffle morphology by redistributing the elevation drop in the short steep riffles to create low-gradient riffles with a slope of less than 0.2%;

Reduce predatory habitat; Increase off-channel rearing habitat for fry and juvenile salmonids via construction of

low flow side-channels and annually inundated floodplain benches; and, Provide up to 80 acres for native plantings (wetland, riparian, and upland) and natural

recruitment.

As of April 18, 2019, McBain Associates, USFWS, and TRC had completed a 30% design of the preferred alternative. This preferred 30% design alternative meets the design goals and objectives presented above and are consistent with the goals of the LTRHIP. The 30% design includes four general types of rehabilitation feature (work areas):

FP: Floodplain bench IC: In-channel feature SC: Side channel W: Wetland

The project also includes riparian plantings. A general site overview of the improvements designed for Bobcat Flat East is provided as Figure 2, delineating extent and type of each feature to be constructed. Table 1 provides the design cut and fill volumes, area of each feature, and estimated usable coarse and fine sediment. A description of each of the features is also provided below.


Table 1. Design cut and fill volumes, area of each feature, and estimated usable coarse and fine sediment.

Floodplain Bench Features Floodplain benches will be constructed in thirteen areas by lowering adjacent surfaces to reconnect the floodplain bench to the contemporary flow regime, mainstem channel, constructed side channel, and existing groundwater elevation. The floodplain benches will achieve the following:


Lower the floodplain to an inundation target flow range specific for each area to:

o allow inundation at 500–1,000 cfs to provide frequent and long duration streamflow connection to the mainstem Tuolumne River to provide fry and juvenile salmonid velocity refuge and foraging opportunities.

o allow inundation at 1,000–5,400 cfs to provide a floodplain connection to the constructed side channel and juvenile refuge during flows above 1,000 cfs; and

o allow inundation at 7,000–9,000 cfs and provide additional habitat and velocity refugia for salmonids and other aquatic organisms during flood events;

Bring the floodplain closer to the groundwater and provide an opportunity to revegetate with native upland grasses, shrubs, and trees.

Excavating granular material from the dredger surface and placing clean gravel within the adjacent dredger pond;

Promote natural riparian regeneration, winter and spring salmonid fry and juvenile rearing habitat, and reduce stagnant water that provides habitat for invasive water hyacinth and predatory fish.

Remove armored layers and lowering the surface to facilitate soil development and natural riparian recruitment.

Material obtained from lowering of areas provide usable coarse sediment needed to construct other design features such as the in-channel bars, riffles, and coarse sediment recruitment piles for the mainstem channel.

Develop a healthy riparian zone along the constructed side channel to provide shading that will foster cooler water temperatures during the hot summer months.

In-Channel Features In-channel features will be constructed in ten areas with the following design objectives:

Area IC-1 will include placement of clean coarse sediment as a riffle enhancement area that the river will mobilize at flows above 3,000 cfs, provide a short-term source of coarse sediment intended to restore in-channel storage, and maintain existing and constructed alluvial features that may also be mobilized by flows greater than 3,000 cfs. The coarse sediment placed in the channel is scaled to the contemporary high flow regime, providing clean coarse sediment for salmonid spawning, invertebrate production, and juvenile and fry rearing.

Area IC-2 enhances the mainstem run and to provide a long-term, frequently replenished (after flows greater than 3,000 cfs) source of coarse sediment intended to maintain existing and constructed alluvial features that may also be mobilized by flows greater than 3,000 cfs.

Area IC-3 includes construction of a riffle placement of clean coarse sediment that can be mobilized by the contemporary flow regime and are suitable for adult salmon spawning and macroinvertebrate production. The riffle control elevation is designed to redistribute the low-flow (150–1,000 cfs) channel slope by backing water into the steeper upstream riffle.

Area IC-4 will increase coarse sediment storage, narrow the low-flow channel width, increase channel sinuosity and complexity, and improve habitat. Willow trenches at the upstream end will promote coarse and fine sediment deposition and reduce the risk of


immediate erosion of the point bar. IC-5 includes construction of a medial bar of clean gravel to immediately increase coarse

sediment storage, narrow the low-flow channel width, promote channel migration into the right bank, and increase in-stream channel habitat and complexity. Willow trenches proposed at the upstream end of the medial bar are intended reduce the risk of immediate erosion of the medial bar but are not intended to impede long-term mobility of the bar feature.

IC-6 includes expansion of the bankfull channel width by up to 100 feet while preserving the existing vegetation. Large riparian and upland trees will be salvaged for use as habitat wood in constructed side channels. Willows may be salvaged in the form of cuttings and used in the construction of willow trenches.

IC-7 includes construction of a riffle of clean coarse sediment that can be mobilized by the contemporary flow regime and are suitable for adult salmon and steelhead spawning and macroinvertebrate production.

IC-8 will increase coarse sediment storage, narrow the low-flow channel width, increase channel sinuosity, promote right bank channel migration, and improve habitat and channel complexity. Willow trenches constructed at the upstream end will promote coarse and fine sediment deposition and reduce the risk of immediate erosion of the point bar.

IC-9 includes construction of a medial bar to immediately increase coarse sediment storage, narrow the low-flow channel width, and increase in-stream channel habitat and complexity. Willow trenches proposed at the upstream end of the medial bar will reduce the risk of immediate erosion of the medial bar but should not impede long-term mobility of the bar feature.

IC-10 will augment the existing Riffle 18 by placing clean coarse sediment that can be mobilized by the contemporary high flow regime and are suitable for adult salmon and steelhead spawning and macroinvertebrate production. The IC-10 riffle control elevation is designed to redistribute the low-flow (150–1,000 cfs) channel slope by backing water into the downstream end of riffle IC-7.

Side Channel Features Side channel features will be constructed in three areas with the following design objectives:

The SC-1 will be constructed to be a low-flow side channel that flows at and above 150 cfs to provide year-round off-channel habitat and refugia for fry and juvenile salmonids. Adjacent features, such as benches, bars, and floodplain, are designed into the side channel and will inundate at flows ranging from 300 cfs to 7,000 cfs, providing contiguous velocity shelter for fry and juvenile salmonid rearing.

Areas SC-2 and SC-3 will be constructed as high-flow side channels that connect the SC-1 side channel to the mainstem river at flows of 1,000 cfs and greater. Mature riparian vegetation along the pond fringes, to the extent possible, will be preserved during construction. Any mature vegetation disturbed during construction may be used as large wood habitat features within the SC-1 side channel.


Wetland Features Each of the three wetland features will be constructed as a bench adjacent to a side channel: Area W-1 will be constructed bench adjacent to SC-1; and Areas W-2 and W-3 will be constructed adjacent to SC-2. The benches will allow flow from upstream to connect to the side channel and provide a suitable location for emergent wetland vegetation to establish; and provide immediate fry and juvenile salmonid rearing habitat, western pond turtle basking habitat, and habitat for wood ducks and other avian species. Riparian Planting Riparian planting designs for the Bobcat Flat Phase III project includes five vegetation zones: emergent wetland, riparian toe, riparian, riparian transitional, and upland. In addition, native sedges, rushes, and/or grasses will be hydro-seeded for erosion control and meadow enhancement. No planting will be done within the side channel or any existing scour channel or on its slopes in order to maintain or improve flood conveyance.


Figure 1. Bobcat Flat East Project Area

BOBCAT FLAT


Figure 2. General site overview of improvements planned for Bobcat Flat East Project subject to review by and consultation with TPAC.


Monitoring and Maintenance Monitoring of Bobcat Flat East Project will be conducted following construction of the improvements. The objectives of the Monitoring Plan will be to assess:


conditions; and, how lessons learned can inform future restoration efforts.

Preliminary assessments, metrics, and protocols to be used for monitoring are summarized in the table below. The TPAC will advise what additional implementation and effectiveness monitoring may be warranted to verify and document that the project was built according to design and to document that project goals were met. The information collected will be used to inform future restoration projects undertaken as part of the LTRHIP. To the extent available and applicable, the LTRHIP and TPAC may utilize data obtained from other monitoring efforts being conducted on the lower Tuolumne River. Assessment Metric Protocol As-built survey Comparison of as-built

survey to 100% design GIS/CAD layer overlay to compare the two surfaces and calculate overall deviations







As-built surveys to assess new rearing habitat area and new floodplain area will be completed immediately after construction. Riparian revegetation success will be assessed one year after planting. Side channel persistence should be evaluated every three to five years. On-going maintenance activities over the term of the license may include minor gravel augmentation, grading, and re-vegetation to facilitate achievement of restoration goals. A report on the monitoring and maintenance activities conducted, assessment of effectiveness of the restoration, and guidance for future maintenance will be prepared and submitted to FERC every five years. These activities will be supported by the LTRHIP annual maintenance fund.


Property Rights Properties permanently affected by proposed floodplain and/or in-channel activities are summarized below:

Project Site Proposed Activity Number APN Name Bobcat Flat East In-Channel work 66 008-020-016 W Joel Hall Et Al

Bobcat Flat East In-Channel work 68 008-020-023 Friends of the Tuolumne Inc (AKA Tuolumne River Conservancy)

Bobcat Flat East In-Channel and Floodplain work 74 008-021-011

Friends of the Tuolumne Inc (AKA Tuolumne River Conservancy)

Bobcat Flat East In-Channel work 77 008-021-025 W Joel Hall Et Al

Bobcat Flat East In-Channel and Floodplain work 78 008-021-026

Friends of the Tuolumne Inc (AKA Tuolumne River Conservancy)

Easements or agreements will be obtained from the Tuolumne River Conservancy and private property interests for rights to gain temporary and permanent access for construction, monitoring, and maintenance activities over the term of the FERC License. The easements will be obtained upon completion of final design and approval of permits for construction and submitted to FERC as part of the request for approval of construction. While the Districts contend that the existence of the final easement or contract is sufficient evidence of the necessary right, title, and interest to secure site performance, the Districts are willing upon completion of construction of the Bobcat Flat East Project, to prepare a boundary survey of the final post-construction project. If required, a project boundary can be provided to FERC in a geo-referenced electronic GIS file format consistent with 18 CFR 4.39 mapping specifications.


RIFFLE 24


Riffle 24 (TLSRA) The Riffle 24 Habitat Restoration Project addresses the concerns FERC identified in the DEIS related to the LTRHIP:

Construction, monitoring, and maintenance of in-channel improvements to improve juvenile fall-run Chinook and O. mykiss rearing, consistent with the goals of the LTRHIP;

Located at RM 41.8; Rights of access necessary for construction, monitoring, and maintenance activities to be

secured by easement or contract from the primary property owner State of California for the term of the new license upon completion of final design and approval of permits for construction5; and

Permits in compliance with all regulatory requirements, including ESA and NHPA, to be led by USFW under its agreements with state and federal agencies pertaining to review of stream rehabilitation projects or related to NHPA, through direct consultation with SHPO and Tribes using protocols adopted for the Don Pedro and La Grange projects.

The project description below provides details of conceptual project design, monitoring and maintenance activities.

Riffle 24 Project Description Riffle 24 was selected because it has public access within a relatively long reach (RM 39.4–43.0) that otherwise would not receive coarse sediment augmentation in the near future. Access is via Turlock Lake State Recreation Area. Coarse sediment augmentation at this site will not only increase spawning gravel availability and quality but will also route sediment downstream to the next augmentation site at Roberts Ferry Bridge. This site will also provide an interpretive site to allow public access to view spawning gravel restoration methods. The original concept developed in 2004 for the CSMP was recently enhanced to 30% design level to better meet objectives of the LTRHIP, subject to review and revision by the TPAC. The 30% design for the project includes placement of 2,500 cy of coarse sediment to enhance in-channel salmonid habitat at Riffle 24 and provide coarse sediment supply for downstream transport (see Figures 1 and 2):

Placement of approximately 1,700 cy of coarse sediment between Station 2337+10 and 2333+45;

Placement of approximately 700 cy of coarse sediment downstream of Station 2333+45; and,

Placement of approximately 1 foot of coarse sediment (about 100 cy) to the right back bar between Station 2336+75 and Station 2335+00.

5 Easement from or contract with State of California provides sufficient right, title, and interest necessary for ensuring the LTRHIP project will perform as designed over the term of the new license. A final access and construction agreement with the State will be provided to FERC with the usual and customary request for FERC approval for construction. The site does not need to be included within the Project Boundary by virtue of a showing of the rights contained in the agreement with the State.


A contingency of 25% for estimated coarse sediment volume is 600 cy for an estimated total volume of coarse sediment fill of 3,100 cy. Monitoring and Maintenance Monitoring of Riffle 24 Project will be conducted following construction of the improvements. The objectives of the Monitoring Plan will be to assess:


conditions; and, how lessons learned can inform future restoration efforts.

As-built surveys to assess new rearing habitat area will be completed immediately after construction. On-going maintenance activities over the term of the license may include minor gravel augmentation, grading, and re-vegetation to facilitate achievement of restoration goals. A report on the monitoring and maintenance activities conducted, assessment of effectiveness of the restoration, and guidance for future maintenance will be prepared and submitted to FERC every five years. These activities will be supported by the LTRHIP annual maintenance fund. The TPAC will advise what additional implementation and effectiveness monitoring may be warranted to verify and document that the project was built according to design and to document that project goals were met. The information collected will be used to inform future restoration projects undertaken as part of the LTRHIP. To the extent available and applicable, the LTRHIP and TPAC may utilize data obtained from other monitoring efforts being conducted on the lower Tuolumne River. Property Rights Properties permanently affected by proposed in-channel activities are summarized below: Project Site Proposed Activity Number APN Name Riffle 24 In-Channel 63 008-020-006 Turlock Lake State Park Riffle 24 In-Channel 64 008-020-009 Robert L & Patricia R. Cree

The south bank of the Riffle 24 site is owned by the Turlock Lake State Park. The site property on the north bank is owned by Robert and Patricia Cree. Access to the project site is from existing access roads off of Lake Road on the south side of the river, and Yosemite Boulevard on the north side. Easements will be obtained with the State of California and private property interests for rights to gain temporary and permanent access for construction, monitoring, and maintenance activities


over the term of the FERC License. The easements will be obtained upon completion of final design and approval of permits for construction and submitted to FERC as part of the request for approval of construction. While the Districts contend that the existence of the final easement or contract is sufficient evidence of the necessary right, title, and interest to secure site performance, the Districts are willing upon completion of construction of the Riffle 24 Project, to prepare a boundary survey of the final post-construction project. If required, a project boundary can be provided to FERC in a geo-referenced electronic GIS file format consistent with 18 CFR 4.39 mapping specifications.


Figure 1: Tuolumne River, Riffle 24 30% design plan subject to review by and consultation with TPAC.


Figure 2. Tuolumne River, Riffle 24 30% design cross sections.

ATTACHMENT C

Districts’ Response to NMFS Comments on the Draft Report

for the Thermal Performance of Wild Juvenile

ONCORHYNCHUS MYKISS in the Lower Tuolumne River: A

Case for Local Adjustment to High River Temperature

DISDR

OF

STRICTRAFT REWILD J

LOWAD

TS’ RESPEPORTJUVENI

WER TUODJUSTM

PONSE FOR TH

ILE ONCOLUMN

MENT TO

TurlocModest

TO NMHE THECORHYNE RIVEO HIGH

ck Irrigatito Irrigati

MFS COMERMAL

YNCHUSER: A C

H RIVER

DON

ion Districon Distric

An

MMENTL PERFOS MYKI

CASE FOR TEMP

N PEDROFER

ct – Turloct – Modes

nthony P. Nann A. FKarl K. E

Se

TS ON TORMANISS IN TOR LOCERATU

O PROJERC NO. 2

Prepared ck, Califosto, Califo

PreparedFarrell, PFangue, PEnglish, M

eptember 2

THE NCE THE CAL URE

ECT 2299

for: ornia ornia

d by: Ph.D. Ph.D. M.Sc.

2017

ATTACHMENT C

This Page Intentionally Left Blank.

W&AR-14 Page 1 Districts’ Response to NMFS Comments Thermal Performance of Juvenile O. mykiss Don Pedro Hydroelectric Project

1.0 BACKGROUND In January 2015, Turlock Irrigation District and Modesto Irrigation Districts (collectively, the Districts) released a draft report for the Thermal Performance of Wild Juvenile Oncorhynchus mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature (Farrell et al. 2015) to relicensing participants for review and comment. Comments on Farrell et al. 2015 were received from the Tuolumne River Trust, the California Sportfishing Protection Alliance, the State Water Resources Control Board, and the California Department of Fish and Wildlife. The report was subsequently finalized in February 2017 (Farrell et al. 2017). Farrell et al. 2017 includes copies of the abovementioned comment letters (as Appendix 5) and the Districts’ responses to these comments (as Appendix 6). In February 2017, the National Marine Fisheries Services (NMFS) provided additional comments on Farrell et al. 2015. The Districts herein provide a response to NMFS’ comments.


2.0 RESPONSE TO COMMENTS Thank you for the detailed, thorough and critical comments on the report (Farrell et al. 2015). It is unfortunate that, because of the timing of the review, the NMFS did not have the benefit of reviewing our revised report (Farrell et al., 2017), which includes our detailed responses to CDFW’s comments. We would encourage NMFS to review this document before moving on to our specific response below. Please also read the peer-reviewed published manuscript of this study (Verhille et al. 2016). NMFS’ February 2017 comment letter, Farrell et al. 2017, and Verhille et al. 2016 are attached to this document as Attachments A, B, and C, respectively, for your convenience. As we outline below, many of the issues seem to stem from a mismatch in expertise between the experimenters and the reviewers. We apologize for any lack of clarity on our part and for taking for granted that all reviewers would possess a similar baseline and core understanding of fundamental tenants of thermal biology and physiology of fishes because this communication barrier appears to have caused several misinterpretations. To be clear, our experiments are fundamental physiology. The theoretical and technical approach is based on well-accepted physiological and ecological principles established by Fred Fry and others over a 60-year period. What is different today is that technology and methodology are more reliable than 60 years ago, which makes our data all the more robust. Therefore, we feel that neither our data quality, nor our experimental approach should be challenged in the way it has been in the NMFS review. These direct challenges have necessitated us to defend our science, aggressively in places. Indeed, after a normal and thorough review by anonymous experts in the field, our work now appears in the peer-reviewed scientific literature (Verhille et al. 2016). Please understand that we view this exchange as positive and essential because we stand behind the numbers we have carefully generated. What troubled us most is the overreliance by the NMFS reviewers on the publication and opinions of Clark et al. (2013a). Rarely do scientific papers draw letters to the editor, but Clark et al. (2013a) drew two independent letters of concern that were subsequently published in the same journal, along with a response by Clark et al. (2013b). Clark et al. (2013a) felt very strongly that the OCLTT hypothesis cannot explain all of fish biology, and that other performance measures should be used in addition to aerobic scope. They did not suggest alternative metrics for aerobic scope, and replacements still have not emerged 4 years on. What Clark et al. (2013a) have never challenged is that aerobic scope (and related measurements) is a useful metric that tells us much about a fish. In fact, Dr. Clark continues to measure aerobic scope in many of his subsequent works, as do a wide spectrum of the fish community. Therefore, we are not alone in using this approach and perhaps a search of aerobic scope in the scientific literature will support this assertion. Perhaps the new review by Portner et al. (2017) will provide further assurances of the value and reliability of measuring aerobic scope (see Attachment D). Consequently, the issue at hand is not the utility of measuring aerobic scope or how accurately it was measured, but rather what can be inferred from such measurements. We have gone to great lengths in our responses to both CDFW, in the revised Farrell et al. 2017 report, in our published


manuscript, and now in our responses to NMFS’ comments below to further articulate what our data should and should not be used for in the context of managing lower Tuolumne fish and the 7DADM for O. mykiss. More generally, Farrell (2016) has dealt with some of these issues, noting that while we have ample resources to measure a fish’s capacity to obtain oxygen, we have an imperfect understanding of how and when a fish apportions this capacity to the various life supporting activities that require oxygen. Thus, an important knowledge gap exists between what we can measure and what we would like to know for management purposes. We do recognize the importance of our data for management purposes, and because of this, fundamental misunderstandings, misconceptions and misinterpretations must be addressed. It is paramount to us that we are as clear as possible on each of these points, necessitating the following dialog consistent with a rigorous scientific exchange. What we feel is the most important scientific fact to emerge, and of direct relevance to management, is that the measurements made by us on Tuolumne O. mykiss indicate that we must set this stock/strain of O. mykiss apart from other strains for which we have similar measurements. This suggests to us that local adaptation may have occurred and that the Tuolumne O. mykiss show a thermal physiology that is more like that of the subspecies of O. mykiss that has adapted to desert regions in the USA and that of O. mykiss selectively bred from an unknown Californian O. mykiss stock for a harsh, hot and arid new life in lakes of Western Australia. The idea of local adaptation within a species is not a new concept, but it is one that EPA 2003 left unaddressed when setting temperature guidance criteria some 15 years ago. Science has marched on and important advances have been made in our understanding of thermal tolerance of fishes. Indeed, we now have similar thermal performance data in hand for a California strain of Chinook salmon (Poletto et al. 2016). Specific Responses: In the hope of aiding NMFS to better understand our findings and our interpretations, rather than addressing the 28 pages of comments, point by point, and line by line, we group the comments into 4 general topics, which we address collectively: 1. Data quality; 2. Excessive critique beyond expertise; 3. Overemphasis on a single review (Clark et al 2013a); and 4. General statistics. We close by addressing the reviewer’s 10 specific questions (taken from pp. 9-10 of the review). 1. The quality of the data is frequently questioned by the reviewers Our data are of the highest quality. Our research team is composed of internationally recognized experts in the thermal physiology of fishes drawn from inside and outside California. We stand by the core data on absolute aerobic scope (AAS) of TR O. mykiss in our Report. Indeed, the work has already been internationally peer-reviewed to be published (Verhille et al. 2016). Therefore, the methodology and results have been taken through the rigors of the time-tested scientific evaluation process for acceptance by the wider scientific community. Moreover, the data appear in open-access literature for all to read. Consequently, any questions as to the quality of the data or the reliability of the experimental approach raised by the reviewers are now mute points. Indeed, until contradictory data are introduced to refute these data, they must stand.


Therefore, we have refrained from another detailed defense of the quality of the results despite the misleading comments and innuendo found in the NMFS review. If additional detail is needed for the purposes of clarity so that our Report can reach a broader readership and better serve scientists with a limited understanding of thermal performance in fishes, this can be done easily through the addition of supplementary information and explanations in footnotes. The data on AAS presented in the Report are now de facto scientifically sound. 2. Excessive, unreasonable, and overtly misleading commentary revealing a disconnect between the expertise of the NMFS reviewers with that needed to evaluate our data. The reviewers openly acknowledge that they do not understand either the study design and measurement principle, or the permitting restrictions imposed on the study that is reported. With respect to the first point, the reviewers state:

From NMFS: “We have not conducted aerobic scope experiments ourselves, and so the list below is to highlight areas where we did not fully understand the study design, methodological approach or data analysis.”

If the reviewers do not understand the basic methodology, how are they able to perform a reliable scientific review? Perhaps this explains why we found the review excessively long, containing much digression and excessive quotations. Of greater concern, however, are spurious claims that border on falsehoods when compared to what was said in our Report. Since spurious claims are obstructive to the goal of moving towards better management of TR O. mykiss, they must be noted. Unfortunately, in providing specific examples below, the reviewers force us to criticize the work of other scientists without them being able to defend themselves, which is not how science should proceed. a) The reviewers’ present unreasonable experimental demands given the permitting limitations for the experimental work, which are followed by comments that reveal insufficient understanding for proper data review and evaluation. For example, the NMFS review contains many suggestions of additional experimental work that should have been done on TR O. mykiss. To be clear once more, the study permit allowed us to use up to 50 fish, and no more. In addition, only 2 fish could be out of the river for a period of 24 h, and there could be no more than 3 fish mortalities, which greatly restricts the experimental design. Nevertheless, even with this limited number of fish, we generated scientifically sound and novel information of the thermal physiology of TR O. mykiss. These data led us to the conclusion that the TR population of O. mykiss shows features that differ from typical Northwest Pacific O. mykiss populations. Instead, the TR population of O. mykiss possesses features that closely resemble those O. mykiss populations known to have a superior thermal tolerance. Yet, the reviewer clearly misunderstood our constraints and misrepresents our experimental design with comments such as:


From the NMFS: “Two O. mykiss individuals were tested at each exposure temperature, rather than testing the same individual over the range of exposure temperatures”

What we actually did was test 2 fish each day using the same test temperature and varied the temperature on a daily basis over a wide range, but using new fish as the study permit required. Why cannot the reviewer simply accept that a solid experimental design was used to derive a complex curve by treating the dependent variable (i.e., temperature) as a continuous variable within the constraints of what the permitting allowed. Instead, the review goes on to suggest that we should conduct thermal acclimation and growth studies when such studies were impossible within our study permit. Another example is the section of the NMFS review that was introduced by:

From the NMFS: “We reviewed the Report in the light of the recommendations of Clark et al. (2013a), intended to ensure that swim-tunnel aerobic metabolism measurements are sound and comparable between studies.”

It is then followed by:

“Clark et al. (2013a) recommend conducting aerobic scope tests beyond what the fish experience in nature. We understand ESA collection permit restrictions prevented tests at temperatures higher than 25°C (p. 15), but these warmer conditions occur often in the lower Tuolumne River. How did this experimental constraint affect the test results? For example, the experiments yielded an average aerobic scope that remained within 5% of optimum from 17.8°C to 24.6°C. Since the upper range value of this estimate is near 25°C, are additional data points (from exposures at temperatures >25°C) necessary to produce a curve that more precisely estimates an optimum temperature or range for aerobic scope?”

Additional data points at sustained temperatures above 25°C would not be helpful in our efforts to define the optimal temperature range because it is possible that more fish would have died when test at temperatures above 25°C. We clearly show that O. mykiss began to die when exhaustively exercised at 25°C. As conservation-minded scientists, we did not want to kill any more fish, so why would we want to use higher temperatures when the writing was already on the wall: 25°C is too warm for TR O. mykiss to exhaustively exercise. Besides, only 3 fish deaths were possible for the entire set of experiments, as dictated by our study permit. b) We do not see the point of the reviewers recommending that other stocks beside TR O. mykiss be considered. We already have 40-yr old data on growth of Minnesotan O. mykiss (Hokanson et al. 1977), but current data for TR O. mykiss suggest they are inapplicable to warmer Californian rivers. This, quite frankly, is the point that is most important in all of this. The scientific literature tells us that the performance of Minnesotan fish is very unlikely to reflect TR O. mykiss. Thus, from a management perspective, what you need are data for TR O. mykiss, and this is what we have provided.


Critically, you must understand that what we have shown here is strong evidence that LTR O. mykiss can be fundamentally different, with respect to their thermal performance, compared with other O. mykiss stocks found around the world. The extension of this is important. It is simply not appropriate, and the literature no longer supports, managing specific populations as if they are all equal. This is inconsistent with our understanding of thermal performance in fishes, and further, is inconsistent with evolutionary principles and local thermal adaptation. In the absence of data, we should be conservative. In the presence of data, we should be continually working towards using the data to improve our practices, working towards population-specific thermal management. c) We agree with the NMFS that more than one performance metric should be incorporated for the determination of protective water temperature criteria.

From the NMFS, p. 7: “As explained below, the NMFS does not support the use of any single performance metric (aerobic scope or another measure) to determine protective water temperature criteria. We support using multiple lines of evidence gathered from studies of several physiological and ecological measures (EPA 2003; McCullough et al. 2009).”

However, the critical issue quickly becomes: which suite of performance criteria do you need, tested over what ecologically-relevant timescale, for which lifestage and/or population of fish etc.? We maintain that AS is a very useful metric, adding important information to the management tool box, and we present our expert interpretations for what these data tell us about TR O. mykiss relative to other O. mykiss populations. The lines of evidence used to set the EPA (2003) criteria that are applied to TR O. mykiss were not derived from studies using TR O. mykiss. Our physiological study along with other ecological observations for TR O. mykiss are building the multiple lines for evidence that support the conclusion that TR O. mykiss are locally adapted to higher water temperatures than the O. mykiss populations used to develop the EPA (2003) criteria. d) The NMFS comment on p. 4 is misleading.

From the NMFS: “We found the Report’s experimental approach and framework relied heavily on acceptance of the OCLTT hypothesis as an established fact or law.”

This is nonsense. A hypothesis is tested, not proven. This is the basis of the scientific method. What is a fact for fishes is that all activities require oxygen, either immediately to directly support that activity, or shortly after during the recovery phase. This is the only fact that we rely on in this regard. Furthermore, our data do not have to be framed by the OCLTT hypothesis. If we remove from our Report all reference to OCLTT, it would not affect any of our data. Therefore, trying to discredit the OCLTT framework is not only unproductive, it is flawed. The OCLTT hypothesis is a contemporary extension of the Fry paradigm, which has long been accepted by fish biologists (see Claireaux and Chabot 2016) and encompasses the Fry polygon.


Moreover, the reviewers openly admit a poor understanding of what Fred Fry measured in fish and then applied to their ecology, just like both Brett and Elliott did many years afterwards with the following comments:

From the NMFS: “In reviewing the Report, we found the need for much more explanation of how aerobic scope (the difference between resting and maximum metabolic rates) by itself can be assumed to define a fish’s capacity to carry out its essential life functions like swimming, defending territory, catching prey and feeding, digesting a meal, growing, and avoiding predators (p. 4), or of how this capacity or potential translates to vital rates (e.g., size-based fecundity, mortality rate, etc.) that affect population-level status.”

The reviewer then uses a sleight of hand to shift ownership of a well-established fact by stating:

From the NMFS: “The Report asserts that aerobic scope defines a fish’s capacity to perform its essential life functions”.

Our statement holds true because, as we explain below, the operable word is “capacity”, unlike the reviewers’ emphasis of the wrong word (“defines”). Indeed, to challenge this definition reveals a lack of understanding of simple bioenergetic principles. O2 is the currency of life for animals.

6O2 + C6H12O6 = 6CO2 + 6H2O.

Simply put, an animal generates CO2 and water using food and O2, which in the process of mitochondrial respiration generates ATP to fuel all essential life functions. Importantly, this defining principle says nothing about which of the life functions the O2 is used for at any given moment in time. What it does imply, however, is that if an animal cannot supply sufficient O2 for an activity, it cannot perform the activity as well or at all. This simple concept has been endlessly demonstrated by exposing animals to environmental hypoxia, which reduces O2 supply and at some level reduces what an animal can do. Thus, if the maximum CAPACITY to supply O2 (i.e., AAS) is a known entity, you then know what an animal can POTENTIALLY do. This very principle was adopted in our work and is one that was established almost a century ago for human studies and over 60 years ago for fish studies. What we can also do with the growing literature base in this area is make stronger conclusions and inferences about what the ecological significance of AS capacity means between species or among populations of the same species. Comparisons of relative AS values among species and populations is incredibly powerful. While AAS does not tell us directly how a fish apportions its capacity to use O2 on a given day, or among competing activities (Farrell, 2016), of all activities in fish, digestion and exercise are perhaps the best understood, O2-demanding functions which we try to interpret in our report.


3. Overemphasis on a single review article, Clark et al. 2013a The Review leans heavily on the words and recommendations of the Clark et al. (2013a & 2013b). Indeed, the critique of the experimental approach mostly relies on these contributions to the literature, which are cited to excess. It is important to clarify what these two works represent. First, Clark et al. (2013a) is a review (i.e., no original data are contained therein) that takes particular exception to the universality of the OCLTT hypothesis. Furthermore, these authors recommend the use of “multiple performance measures” without stating what these new “performance metrics” might be. Indeed, their review closes with:

“… it does provide a timely reminder that the OCLTT hypothesis is not universally applicable and there are other performance metrics that may work independently or synergistically with attributes such as aerobic scope to govern the thermal preferences and fitness of water-breathing organisms. A fruitful direction for future research is to investigate which performance metric(s) becomes progressively limited as ectothermic animals are shifted away from their preferred temperature (i.e. the temperature they select when given the opportunity).”

By making this statement (and from reading the entire review), the three authors of Clark et al. (2013a) could not provide a substitute “performance metric” for ASS at that time. Furthermore, an alternative has not emerged in the scientific literature to date. More importantly, some of the opinions contained in Clark et al. (2013a) were challenged by two independent letters to the journal’s Editor from two distinct research teams (Farrell 2013; Portner and Giomi, 2013). These two letters were published alongside another letter to the Editor (Clark et al. 2013b) in response to the two challenges. Thus, all three letters are opinions and are NOT a peer-reviewed source. All three letters address the validity of the opinions expressed in Clark et al. (2013a), which represents the perspective of just three scientists. Thus, Clark et al. (2013a) was not accepted carte blanche by the scientific community, in part because it offered no alternative to AAS. 4. Comments of the reviewers suggest a fundamental lack of understanding of statistics Our concern is clearly revealed with the following quote:

From the NMFS: “We noted the imprecise fit (r2 = 0.099) of the curve through the aerobic scope data points (Figure 4, p. 37). In contrast, the fit of the curve for the resting (routine) metabolic rate was good (r2 = 0.8), and better than for maximum metabolic rate (r2 = 0.49). Are these findings common in aerobic scope studies? How does the imprecise fit of the aerobic scope curve affect the reliability of the estimates of optimal temperature for aerobic scope, the comparability of the results with other studies, or its use in setting protective temperature criteria?”

Scientists understand that a regression shows the robustness of the relationship between two variables, X and Y. A horizontal line through data means that no such relationship exists between X and Y, and this would be reflected in a very low r2. Our point of presenting the statistic (r2 = 0.099) is that AAS does not statistically change appreciably over much of the


temperature range that we tested, i.e., TR O. mykiss maintain a near maximal capacity to supply oxygen over a surprisingly wide temperature range. Thus, by raising this as an issue, the reviewers, possibly inadvertently, are admitting that TR O. mykiss at 22C and 18C must have the same capacity to deliver oxygen to tissues, which is the primary conclusion of our Report. What then follows in our logic is that if there is no difference in aerobic capacity between these two temperatures, why are regulations using 18C as a cut off temperature? Instead, this thermal limit was derived primarily on a growth study performed 40 years ago and with O. mykiss that had been transplanted and hatchery-reared in Minnesota rather than a population from the TR that had been fully acclimatized to summer river temperatures, which was the case for our data. Here, we do not want to be overly disrespectful and critical of the historical data (Hokanson et al. 1977), but we must respectfully acknowledge that the scientific understanding of fish thermal physiology has moved on technically and intellectually since 1977. 5. Answers to specific questions from NMFS, p. 9 onwards. (1) Can you explain this study-design decision, and how it might have affected the resulting

estimates of the optimal temperature for aerobic scope? We believe oxygen is needed for all activities of a fish. Therefore, we measured the fish’s capacity to supply oxygen as a function of temperature and treating temperature as a continuous variable so that a complex curve could be fitted to the data for no more than 50 fish.

(2) Could the short-term (several hours of) temperature acclimation described above have affected the test results, or are longer acclimation intervals (e.g. weeks) required to affect aerobic scope responses? Aerobic scope curves can respond to thermal acclimation (e.g. over weeks at a new temperature), but the magnitude of the response is often species and/or population dependent. We do not feel that acclimation was a factor in the present experiments. Thus, the concern raised does not change the outcome of our results, but does introduce the possibility that this fish population could do even better at warmer temperature if they were allowed to first acclimate because it is well known that thermal acclimation is used by fishes to “improve performance” at the new acclimation temperature. It is a valid comment to query the ‘acclimatization’ state of wild caught fish. The typical pattern is that as fish are acclimated/acclimatized to higher temperatures, the performance curve (if it shifts at all) will shift in the direction that promotes better performance at warmer temperatures (i.e. the thermal optima will move to a higher temperature as acclimation/acclimatization temperatures increase). 75 percent of the fish tested in this study were sourced from locations with capture temperatures between 12.7 and 17.1°C. Therefore, if the TR O. mykiss used in the present study can be shown to acclimate to water temperatures warmer than they were experiencing at the time of the experiments, we have then provided a conservative estimate of temperature effects on the fish performance by looking only at the


effect of a rapid rise in water temperature from river temperature to which they were acclimated.

(3) How could the long-term temperature acclimation experienced by the individuals tested, and the inability to test the less tolerant individuals in the Tuolumne River population, have affected the test results? There are several ideas conflated in this question and the balance of the paragraph that follows. The reviewers’ terminology is imprecise. The idea that thermal acclimation/acclimatization could influence aerobic scope is addressed above. The suggestion by the reviewers that acclimation of individuals (which, importantly, is a process that happens to an individual within its lifetime) is somehow affecting our results in a way that would lead us to conclude that our data are not representative of other important individuals called the ‘less tolerant, non-surviving individuals” in the population doesn’t make sense. These ideas do not go together. The reviewer also mentions ‘long term exposures’ but the time course they are referring to is vague. Of course, we are only testing survivors, as would be the case for any experiment performed on a living organism. Our data do not directly address the ‘tolerance’ or ‘performance’ of the non-surviving members of a population, which in and of itself, doesn’t make sense. Assuming that some fish from a population don’t survive because they are ‘less tolerant’ is quite the speculation. There are many reasons why fish die. Anytime organisms are sampled from a population for study, a fundamental assumption is always that the fish sampled and the variation measured, is representative of the population.

(4) Since the upper range value of this estimate is near 25°C, are additional data points (from exposures at temperatures >25°C) necessary to produce a curve that more precisely estimates an optimum temperature or range for aerobic scope? Fish began dying at 25C. We know from the voluminous literature on the thermal limits of fishes under test conditions and time courses of exposure similar to those used here that more fish would die beyond 25C. It is possible that some individuals may have survived at warmer temperatures, but the purpose of the study design was not to kill fish. If you are interested in revealing the fundamental mechanisms underlying fish failure at high temperatures (which is largely the focus of the OCLTT debate and of Clark et al. 2013a), you must explore the limits of these curves. If you are interested in evaluating protective thermal management targets and you are constrained by permitting, testing fish approaching lethal limits is unethical. The range of temperatures over which fish demonstrate performance capacity is what is important because a fish in nature does not spend its entire life at an optimal temperature for aerobic scope.

(5) Are these findings common in aerobic scope studies? Our study findings are consistent with the results from other aerobic scope studies (e.g. Parsons 2011).


(6) How does the imprecise fit of the aerobic scope curve affect the reliability of the estimates

of optimal temperature for aerobic scope, the comparability of the results with other studies, or its use in setting protective temperature criteria? Dealt with above in the “Comments of the reviewers suggest a fundamental lack of understanding of statistics” section.

(7) How did the steps taken (repeated stimulation with high-velocity bursts to encourage swimming) affect the aerobic scope results, and the estimates of the optimal temperature for aerobic scope? This is a critique that we could have made mention of more clearly in the report. More than one method to elicit and measure MMR is regularly used in aerobic scope studies, appears as a suitable method in the peer-reviewed literature, and the validity of our application of it here is further evidenced by the peer-reviewed publication of our research (Verhille et al. 2016). Please see the published figures which include different symbols for the different methods to acquire MR. Visual inspection shows that there was no pattern between method and result.

(8) We noted the maximum metabolic rate was found to be lowest for fish exhibiting this behavior (p. 19); did this affect the precision of the maximum metabolic rate and aerobic scope results? Please see Verhille et al. 2016, Figure 1. Visual inspection shows that there was no pattern between method and result. Note, an underestimate of MMR would lead to a reduced AAS which would be an overly conservative result and could not lead to an overestimate of metabolic capacity.

(9) How does using an exponent from a fish other than O. mykiss affect the aerobic scope results, the estimates of optimal temperature for aerobic scope, or the comparability of the results with other studies? Scaling metabolic data has no internal effect on our study because the range of body mass of our fish was too small to use a scaling exponent or an ANCOVA with body mass. However, what scaling does allow for is a better comparison with existing data where body mass varies AMONG studies, as is the case when we compare our results with other data for our quality control purposes and interpretation. In fact, few data exist for active and resting mass exponents for a rainbow trout population at the size range investigated in this study. Wieser (1985; JEB 118:133-142) determined mass exponents of 0.08 to 7 g rainbow trout at much cooler temperatures (4°C to 12°C), and found similar mass exponents to what we applied. We applied a mass exponent of 0.95, which is the midpoint between the resting and active mass exponents reported in Lucas et al 2014, and, according to Wieser (1985), the midpoint between the resting and active mass exponents for fish at 12°C was 1.03. The decision to base the mass exponent on the more recent work by Lucas et al (2014), despite it being a different species, was based on methodologies. Active metabolic rate was


achieved by electrical shock in the Wieser (1985) work, whereas, Lucas et al. (2014) provoked activity through manual chases, which we considered to be a more similar approach to our critical swimming velocity tests.

(10) Did these fish undergo aerobic testing prior to death, and if so what were the exposure temperatures? We did not report any metabolic data for these fish because none were taken, see Appendix 4 of the report. Chlorine is part of our tunnel cleaning procedure and we made a terrible mistake in not neutralizing this fully before introducing fish W37 and W 38 into the tunnels for testing. They were dead before any measurements were made. The third fish died after the test, and again the data were not used in mathematically deriving the curve for aerobic scope. It has been known since the 1950’s that fish can exhaust themselves at high temperature and experience delayed mortality as a consequence.


3.0 REFERENCES Claireaux G. and Chabot D. 2016. Responses by fishes to environmental hypoxia: integration

through Fry’s concept of aerobic metabolic scope. Journal of Fish Biology (2016) 88, 232–251.

Clark, T.D., Sandblom, E. and F. Jutfelt. 2013a. Aerobic scope measurements of fishes in an era

of climate change: respirometry, relevance and recommendations. Journal of Experimental Biology 216:2771-2782.

Clark, T.D., Sandblom, E. and F. Jutfelt. 2013b. Response to Farrell and to Pörtner and Giomi.

Journal of Experimental Biology 216:4495-4497. Environmental Protection Agency (EPA). 2003. EPA Region 10 Guidance for Pacific

Northwest State and Tribal Temperature Water Quality Standards. April 2003. Farrell, A. P. 2016. Pragmatic Perspective on Aerobic Scope: Peaking, Plummeting, Pejus and

Apportioning. J Fish Biol 88: 322-343. Farrell, A. P., N. A. Fangue, C. E. Verhille, D. E. Cocherell, and K. K. English. 2015. Thermal

performance of wild juvenile Oncorhynchus mykiss in the Lower Tuolumne River: a case for local adjustment to high river temperature. Draft Report. Prepared by the Department of Wildlife, Fish, and Conservation Biology, University of California, Davis for Turlock Irrigation District and Modesto Irrigation District.

Farrell, A. P., N. A. Fangue, C. E. Verhille, D. E. Cocherell, and K. K. English. 2017. Thermal

performance of wild juvenile Oncorhynchus mykiss in the Lower Tuolumne River: a case for local adjustment to high river temperature. Final Report. Prepared by the Department of Wildlife, Fish, and Conservation Biology, University of California, Davis for Turlock Irrigation District and Modesto Irrigation District.

Hokanson, K. E. F., Kleiner, C. F., and Thorslund, T. W. 1977. Effects of constant temperatures

and diel temperature fluctuations on specific growth and mortality rates and yield of juvenile rainbow trout, Salmo gairdneri. J. Fish. Res. Bd. Can. 34, 639–648.

Lucas, J., Schouman, A., Plyphout, L., Cousin, X., and LeFrancois, C. (2014). Allometric

relationship between body mass and aerobic metabolism in zebrafish Danio rerio. J. Fish Bio. 84, 1171–1178. April 2014. doi:10.1111/jfb.12306

McCullough, D.A., Bartholow, J.M., Jager, H.I., Beschta, R.L., Cheslak, E F., Deas, M.L.,

Ebersole, J.L., Foott, J.S., Johnson, S.L., Marine, K.R., Mesa, M.G., Petersen, J.H., Souchon, Y., Tiffan, K.F. and W.A. Wurtsbaugh. 2009. Research in thermal biology: burning questions for coldwater stream fishes. Reviews in Fisheries Science 17(1):90–115.


Parsons, E. 2011. Cardiorespiratory physiology and temperature tolerance among populations of sockeye salmon (Oncorhynchus nerka). PhD Thesis. University of British Columbia, Canada.

Poletto, J. B, Dennis, C.E., Baird, S.E., Nguyen, T.X., Cabrera-Stagno, Valentina, Farrell, A.P.,

and Fangue, N.A. 2016. Unusual Aerobic Performance at High Temperatures in Juvenile Chinook Salmon, Oncorhynchus Tshawytscha. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5216678

Pörtner, H.-O. and Giomi, F. 2013. Nothing in experimental biology makes sense except in the

light of ecology and evolution. J. Exp. Biol. 216, 4494-4495. Portner, H.-O., Bock, C. and Mark, F.C. 2017. Oxygen- and capacity-limited thermal tolerance:

bridging ecology and physiology. Journal of Experimental Biology (2017) 220, 2685-2696 doi:10.1242/jeb.134585.

Verhille CE, English KK, Cocherell DE, Farrell AP, Fangue NA (2016) High thermal tolerance

of a rainbow trout population near its southern range limit suggests local thermal adjustment. Conserv Physiol 4(1): cow057; doi:10.1093/conphys/cow057.

Wieser W. 1985. Development and Metabolic Constraints of the Scope for Activity in Young

Rainbow Trout (Salmo Gairdneri). Biol. 118, 133-145 (1985).

DISTRICTS’ RESPONSE TO NMFS COMMENTS ON THE DRAFT REPORT FOR THE THERMAL PERFORMANCE OF WILD JUVENILE ONCORHYNCHUS MYKISS IN THE LOWER TUOLUMNE RIVER: A

CASE FOR LOCAL ADJUSTMENT TO HIGH RIVER TEMPERATURE

ATTACHMENT A

NMFS’ FEBRUARY 2017 COMMENT LETTER

February 6, 2017 In response refer to: WF:P-2299/P-14581.WCR:FERC

Kimberly D. Bose, Secretary Federal Energy Regulatory Commission 888 First Street, NE Washington, D.C. 20426 Re: National Marine Fisheries Service’s Comments on the Study, “Thermal Performance of Wild Juvenile Oncorhynchus mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature (Farrell et al. 2015), is filed to the Administrative Record of the Federal Energy Regulatory Commission’s Projects, New Don Pedro (P-2299) and LaGrange (P-14581), located on the Tuolumne River, California. Dear Secretary Bose: NOAA Fisheries Service (NMFS) provides our Review of the Study, Thermal Performance of Wild Juvenile Oncorhynchus mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature (Farrell et al. 2015). NMFS’ Review is hereby filed to the Administrative Records of the Federal Energy Regulatory Commission’s Projects, New Don Pedro (P-2299) and LaGrange (P-14581), located on the Tuolumne River, California. If you have any questions regarding the attached document, please contact William Foster (916-930-3617).

Sincerely,

Steve Edmondson FERC Branch Supervisor NMFS, West Coast Region

Document Attached CC: FERC Service Lists for P-2299 an P-14581

UNITED STATES DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration NOAA FISHERIES SERVICE WEST COAST REGION 650 Capitol Mall, Suite 5-100 Sacramento, California 95814-4706

20170207-5032 FERC PDF (Unofficial) 2/6/2017 10:41:38 PM

A REVIEW BY THE NATIONAL MARINE FISHERIES SERVICE

CALIFORNIA CENTRAL VALLEY OFFICE

INTRODUCTION The following is a review by the National Marine Fisheries Service (NMFS) of Thermal Performance of Wild Juvenile Oncorhynchus mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature (Report) (Farrell et al. 2015), prepared for the Turlock Irrigation District (TID) and the Modesto Irrigation District (MID). The investigations were conducted in 2014, for a water and aquatic resources (W&AR) study during the Federal Energy Regulatory Commission (FERC) re-licensing proceeding for the Don Pedro Hydroelectric Project (FERC No. 2299). In study W&AR-14 (Enclosure A), TID and MID plan to collect information about the site-specific temperature responses of steelhead or rainbow trout (Oncorhynchus mykiss), for use in a reassessment of the thermal guidelines (EPA 2003) applied in the lower Tuolumne River and elsewhere in California’s Central Valley. It is expected that TID and MID will file either the Report or a modified version in the administrative record for the Don Pedro FERC re-licensing proceeding. Both resident O. mykiss (i.e. “rainbow trout”) and the anadromous life-history form (i.e. “steelhead”) exist in the Tuolumne River downstream of the La Grange Dam (Zimmerman et al. 2009). The Tuolumne River steelhead belong to the California Central Valley (CCV) steelhead distinct population segment (DPS), listed as “threatened” under the federal Endangered Species Act (ESA) (Federal Register Notice, January 5, 2006). The juvenile offspring of steelhead also belong to the DPS, and are protected under the ESA even prior to their seaward migration. Due to the impassable La Grange Dam (height 131 feet), CCV steelhead have access to only the fifty-two miles of the “lower Tuolumne River” from the San Joaquin River upstream to the Dam; this segment is ESA-designated “critical habitat” for the CCV steelhead DPS (Federal Register Notice, September 2, 2005). For the CCV steelhead, the model of Lindley et al. (2006) predicts the historical habitat suitable for the species was limited by summertime conditions; due to the relatively lower elevations and warmer conditions in the lower Tuolumne River, the model predicts all 324 km (201 miles) of historically-suitable habitat was in the higher and colder upper Tuolumne River watershed (Lindley et al. 2006, pp. 9-10). Present-day dams that impound large reservoirs can maintain higher base flows and lower stream temperatures for some distance below the dam during summer than would be found in those reaches in the absence of the dam. In the Tuolumne River, flows from the drainage area above the La Grange Dam are impounded in four reservoirs: Hetch Hetchy, Lake Eleanor, Cherry Lake, and Don Pedro. The Don Pedro Reservoir is a short distance (2.6 miles) upstream of the La Grange Dam, and holds a large capacity of nearly 2 million acre-feet. Cold releases from Don Pedro quickly transit the small, riverine reservoir impounded by the


2

La Grange Dam, where in summer large volumes are diverted from the lower Tuolumne River into canals, appreciably lowering the flows downstream. The NMFS is concerned about flow management that could affect the adult and juvenile steelhead in the lower Tuolumne River, including the application or modification of water temperature criteria, standards, or basin-plan beneficial uses by anadromous fishes. The goal of the NMFS is to recover the CCV steelhead DPS, including the Tuolumne River population. Because water temperature affects the health of individual fish, it also affects entire populations and species assemblages (Richter and Kolmes 2005). In reviewing the Report, we sought to understand how the temperature-dependent aerobic scope responses of individual juvenile O. mykiss might inform actions to improve the Tuolumne River steelhead population, and thereby promote recovery. STUDY BACKGROUND Swimming tunnel respirometer experiments were conducted with juvenile O. mykiss in a streamside facility on the lower Tuolumne River, downstream of the La Grange Dam (river mile (RM) 52). The test fish (100-200 millimeter fork length, N=48) were captured in the following lower Tuolumne River locations (number): RM 50.7 (36); RM 51.6 (8); RM 50.4 (2); and RM 49.1 (2). Water pumped from the River was passed through the test facility, while its temperatures were variably controlled between 13 degrees Centigrade (°C) to 25°C. An individual fish was placed in the swim tunnel, and water velocities were adjusted to create the equivalent of an “aquatic treadmill” that required it to swim at various speeds. To measure the oxygen consumed by a fish during a test, the exchange of fresh (aerated) water with the tunnel was intermittently turned off; the rate of oxygen-level decline within the closed chamber provided an estimate of the fish’s aerobic metabolism (at the given activity level and water temperature). The experimental approach adopts the oxygen- and capacity-limited thermal tolerance (OCLTT) hypothesis, which proposes that the extremes of thermal tolerance are set by an animal’s inability to supply oxygen to its tissues above and beyond a basic routine need. “Aerobic scope” is the difference between the minimum and maximum oxygen consumption rates. Accordingly, the Tuolumne River tests measured how much oxygen was routinely consumed by a resting fish, its routine metabolic rate (RMR)1, followed by tests of how much oxygen could be maximally extracted from the water by a fast-swimming fish, termed its maximum metabolic rate (MMR). By subtracting RMR from MMR, the difference is the “absolute aerobic scope” (AAS) for a test 1 NMFS notes that the oxygen consumption rate for fish as measured in lab experiments is not equal to the metabolic

rate (Nelson 2016). It may be time to start considering MO2 as its own measurement and not a surrogate for metabolic rate. As the price and availability of better technology for direct calorimetry of aquatic organisms improves, more laboratories will want to use it and the term metabolic rate should be reserved for their results. As these direct calorimetric results on fish accumulate, more informed comparisons between the two techniques will be made and the understanding of energy flow through fishes improved. Energy, after all, is the currency of fish life and its accurate measurement has implications across the entirety of fish biology (Nelson 2016).


3

fish - a measure of the capacity of a fish to supply oxygen to its tissues above and beyond a basic routine need. The Report explains the biological significance of aerobic scope is that AAS defines a fish’s capacity to perform the activities essential to carry out its life functions, such as swimming, catching prey and feeding, digesting a meal, growing, avoiding predators, and defending territory (p. i; p. 4, pp. 21-22). Consistent with the OCLTT hypothesis, the Report provides Figure 1 (p. 33), a schematic plot depicting how theoretical aerobic scope values (y-axis) vary with water temperatures (x-axis); the point where the aerobic scope versus temperature curve reaches its peak is defined as the optimal temperature (Topt). The authors also compute factorial aerobic scope (FAS = MMR/RMR), a ratio described as another way of expressing a fish’s aerobic capacity across a range of water temperatures; the authors explain that FAS values of 2 or greater are biologically significant because they indicate an aerobic capacity sufficient for a fish to properly digest a full stomach (p. ii; pp. 4-5, p. 20, pp. 21-22), potentially grow faster at a higher temperature (p. 22), or maintain station in a flowing river (p. 21). The aerobic scope test results were used to determine an optimal temperature range for the Tuolumne River O. mykiss population, via plots of the swim tunnel metabolic measurements versus test temperatures, using best-fit mathematical models (lines or curves) along with their confidence intervals (Figure 4, p. 37). The experiments yielded a Topt for AAS of 21.2°C, an average AAS that remained within 5% of this peak from 17.8°C to 24.6°C, and FAS values > 2 for all fish up to 23°C (p. ii, pp. 23-24). The study results were compared with the 18°C criterion2 recommended by the U.S. Environmental Protection Agency (EPA 2003), and the Report concluded the test data “…provide sound evidence to establish alternative numeric criteria that would apply to the Tuolumne River O. mykiss population below La Grange Diversion Dam.” (p. 24). The Report recommends “…that a conservative upper performance limit of 22°C, instead of 18°C, be used to determine a 7DADM value for this population.” (p. 24). COMMENTS AND CONCERNS The Tuolumne River experiments do not test the OCLTT hypothesis; rather, this hypothesis is adopted as if it were widely accepted that aerobic scope alone defines a fish’s capacity to perform its essential life functions. The theoretical rationale for the OCLTT hypothesis holds that the biochemical and physiological capacities of fishes have evolved such that aerobic scope is maximized within a given temperature range (Topt for AAS) in order to optimize fitness-related performance, such as growth, reproduction and locomotion, while performance diminishes as aerobic scope decreases at higher and lower temperatures (Clark et al. 2013a). We found the Report’s experimental

2 The EPA recommends 18°C for the protection of Salmon/Trout Migration plus Non-Core Juvenile Rearing, where

“Trout” refers to Steelhead and coastal cutthroat (EPA 2003, p. 25). The numeric criterion value (18°C) is the recommended maximum of the 7-day average of the daily maximum temperature (7DADM).


4

approach and framework relied heavily on acceptance of the OCLTT hypothesis as an established fact or law:

“The experimental approach acknowledged the oxygen- and capacity-limited thermal tolerance (OCLTT) hypothesis, which proposes that the extremes of thermal tolerance are set by a fish’s inability to supply oxygen to its tissues above and beyond a basic routine need. (p. i). “This experimental approach is consistent with the oxygen- and capacity-limited thermal tolerance (OCLTT) hypothesis that has emerged as a conceptual model to assess thermal performance of aquatic animals and determine the fundamental thermal range for their distributions (Pörtner and Knust 2007; Pörtner and Farrell 2008). The OCLTT hypothesis proposes that the extremes of thermal tolerance will be set by a fish’s inability to supply oxygen to its tissues above a basic routine need.” (p. 4). “In the present study, we use 95% of the peak AAS value to set the optimal thermal range (Figure 1; the two temperatures that bracket Topt are termed a Pejus temperature, Tp). If, as predicted by the OCTTL hypothesis, a cardiorespiratory limitation exists for exercising salmonids during warming, AAS will decrease below 95% of peak AAS beyond the upper Tp, and often rapidly over just a few degrees before lethal temperatures are reached (Farrell 2009).” (p. 5).

The Report asserts that aerobic scope defines a fish’s capacity to perform its essential life functions:

“Thermal performance was assessed as the range of temperatures over which juvenile O. mykiss can increase aerobic metabolic rate (MR) beyond basic needs. This aerobic capacity could be used for any of the normal daily activities of O. mykiss in the Tuolumne River during its normal life history (swimming, catching prey and feeding, digesting a meal, growing, avoiding predators, defending territory, etc.). (p. 4).

“Therefore, our experimental approach directly measured MR under two states: routine metabolic rate (RMR), representing how much oxygen is needed by an individual O. mykiss to exist in the Tuolumne River and maximum metabolic rate (MMR), representing how much oxygen can be maximally extracted from the water for its tissues, typically when swimming. The capacity of the fish to supply oxygen to tissues above and beyond a basic routine need is then calculated by subtracting RMR from MMR, which is termed the absolute aerobic scope (AAS = MMR - RMR). Therefore, AAS defines a fish’s capacity to perform the activities essential to carry out its life functions.” (p. 4). [Underline emphasis added].

While the Report adopts the OCLTT hypothesis, we observe that all experts do not agree with a wide application of the theory, and a healthy debate is ongoing about the ecological relevance of


5

aerobic scope to overall fitness (Clark et al. 2013a; Farrell 2013; Pörtner and Giomi 2013; Clark et al. 2013b). Clark et al. (2013a) warn that the OCLTT hypothesis should not be assumed to be a law, but should undergo more testing (like all hypotheses). Clark et al. (2013b) observe that the temperature where aerobic scope is maximal (Topt for AAS) cannot be assumed to be the same as the optimal temperature for a fish (Topt); they propose that the “onset of loss of performance” of an animal can occur where aerobic scope is maximal because other metrics of performance can be markedly reduced at the Topt for AAS. What seems a reasonable alternative to the OCLTT hypothesis is that the optimal temperature of the animal will be context dependent, governed by interacting optimal temperatures of different physiological and ecological performance metrics (Clark et al. 2013b). The studies of Hokanson et al. (1977) illustrate how different thermal optima can exist for different endpoints. In this research, 50-day laboratory tests were conducted with O. mykiss fed excess rations and exposed to seven constant temperatures (between 8 and 22°C) and six diel temperature fluctuations (±3.8°C about mean temperatures from 12 to 22°C). For constant temperature treatments, the maximum specific growth rate occurred at 17.2°C, with an average specific mortality rate of 0.35% per day at this optimum temperature and lower. Specific growth rate at a fluctuating temperature of 22.2±3.8°C was zero, and mortality rate was 42.8% per day during the first 7 days. In experiments under a fluctuating regime of 15.5 to 17.3°C, the average specific mortality was 0.36% per day. Combining data on specific growth and mortality rates, the authors were able to predict the yield for a hypothetical population under the temperature regimes; they determined a population would exhibit zero increase over 40 days at a constant 23°C and a fluctuating temperature of 21±3.8°C because under these temperature conditions, specific growth rate balances specific mortality rate. Considering these laboratory results and corroborating field information, they recommended a mean weekly temperature of 17±2 °C for O. mykiss, so that maximum yield is not appreciably reduced for populations living in the fluctuating temperature regimes normally found in the natural environment. In their results interpretation, Hokanson et al. (1977) considered both growth and survival endpoint responses (and their interaction), recognizing that population production involves a balance between individual growth and mortality rate of the population.

The temperatures affecting O. mykiss smoltification must also be considered to assess steelhead production. Smoltification involves the physiological, morphological, biochemical, and behavioral changes before and during seaward migration and ocean entry. Myrick and Cech (2005) studied food consumption and growth endpoints versus temperature in O. mykiss (from the Nimbus State Fish Hatchery, American River, California); while they found temperatures approaching 19°C would increase growth rates, they referred to multiple studies suggesting that if juvenile steelhead are exposed to temperatures much above 11°C during the presmolt and smolt periods, their ability to osmoregulate and survive in salt water would be significantly impaired. They recommended study of how long juvenile O. mykiss can be exposed to higher temperatures before returning to temperatures that are better suited for smolting (p. 328). The EPA (2003) recommends a maximum 7DADM of 14°C for the protection of waters where and when the early stages of steelhead smoltification occurs or may occur (p. 31). The Report found


6

aerobic scope maximal at 21.2°C, with an average that remained within 5% of this peak from 17.8°C to 24.6°C; however, when O. mykiss smoltification is included as an endpoint these temperatures are much higher than those deemed protective. The CCV steelhead simulation model of Satterthwaite et al. (2010) incorporates survival-based as well as growth-based variables. Their model suggests the greatest management concern with respect to preserving anadromy is reduced survival of emigrating smolts during their downstream migration (freshwater, estuarine, and ocean). They emphasize the potential importance to O. mykiss anadromy of large changes in both freshwater survival or growth rates; both factors are emphasized, and both are affected by temperature. Several other variables are also at play in the model, including a temperature-dependent measure of basal metabolic rate (Table 1, p. 4). In reviewing the Report, we found the need for much more explanation of how aerobic scope (the difference between resting and maximum metabolic rates) by itself can be assumed to define a fish’s capacity to carry out its essential life functions like swimming, defending territory, catching prey and feeding, digesting a meal, growing, and avoiding predators (p. 4), or of how this capacity or potential translates to vital rates (e.g., size-based fecundity, mortality rate, etc.) that affect population-level status. As stated earlier, the goal of the NMFS is to increase the Tuolumne River steelhead population. If aerobic scope data can promote this goal, we need to better understand how it is linked to the field-based, population-level data used to determine steelhead viability (Lindley et al. 2007). In Enclosure B, we provide a reproduced table from McCullough et al. 2009 (p. 92) that arranges research topics concerning the thermal biology of fishes within their various levels of biological organization. We used the table to place the Tuolumne River aerobic scope experiments in the context of other research that investigates physiological or ecological endpoints. Assessed in this broader context, the Tuolumne River experiments appear to evaluate O. mykiss response at the physiological level (aerobic scope performance) and extrapolate the biological significance of the test results across levels #2, #3, and #4 (consistent with the acceptance of OCLTT theory); the Report concludes its test results warrant a new water quality criterion for the lower Tuolumne River (#5 policy-level implications). The Report is not clear about the linkages; how is the aerobic scope performance of individual Tuolumne River O. mykiss a sufficient measure of population-level fitness under the thermal regime? How is the singular aerobic scope measure sufficient to establish a protective water temperature criterion? Clark et al. (2013a) observe that aerobic scope is likely to be an important mechanism contributing to the ecology of fishes because it should set the limit for the magnitude of oxygen-demanding processes that can be performed simultaneously. However, they caution that aerobic scope may continue to increase until temperature approaches lethal levels and then decline rapidly as death ensues, thereby providing little insight into the preferred temperature or performance of aquatic ectotherms (p. 2771). MMR has often been reported to first increase with temperature, then plateau or even decrease due to detrimental effects of high temperature on convective oxygen transport (Fry & Hart, 1948; Farrell et al., 2009; Portner, 2010, as cited in Norin and Clark 2016), causing a bell-shaped relationship between aerobic scope and temperature. It is becoming clearer, however, that MMR, AAS, or both does not always plateau


7

or decrease at high temperatures in fishes (Norin and Clark 2016). Many studies show that in general, fishes appear to exhibit a continuous increase in MMR, AAS, or both until the point where temperature is at the upper end of the natural range and close to or at lethal levels (Norin and Clark 2016), including the study on the Central Valley Chinook salmon juveniles (Poletto et al. 2016). Recent evidence suggests that there is considerable individual variation in the thermal sensitivity of MMR, AAS, or both (Poletto et al. 2016), with high-MMR individuals responding less to heating compared with low-MMR individuals (Norin and Clark 2016). In fact, the aerobic scope for the Central Valley juvenile Chinook salmon continues to increase at water temperatures up to 25 or 26°C, where substantial mortality occurred (Poletto et al. 2016). We are concerned that negative effects also could occur at temperatures much lower than near-lethal levels, where aerobic scope is continuing to increase with temperature. As explained below, the NMFS does not support the use of any single performance metric (aerobic scope or another measure) to determine protective water temperature criteria. We support using multiple lines of evidence gathered from studies of several physiological and ecological measures (EPA 2003; McCullough et al. 2009). The Report makes numerous comparisons between its Tuolumne River aerobic scope results and the 18°C summer criterion recommended by the EPA (2003), but the EPA’s criterion was derived from several types of studies and not aerobic scope experiments.

The Report establishes how metabolic rate measurements and related computations (e.g., AAS, FAS, etc.) were used to derive an optimal (peak) temperature and optimal temperature range for aerobic scope. Then, in numerous instances these numerical results are compared and contrasted with the EPA’s 18°C criterion. These comparisons appear to assume the EPA’s criterion was based on aerobic scope studies, but our understanding is the EPA (2003) did not use aerobic scope information to establish its recommended summer 18°C criterion. The Report does not establish how EPA used aerobic scope information to develop their criterion, or if they used aerobic scope information at all. It also does not adequately acknowledge or discuss the other types of thermal suitability information the EPA (2003) used to derive its 18°C criterion. For these reasons, we found several of the Report’s comparisons to be invalid, and list examples below:

“Interestingly, by setting the 7DADM criterion for salmon and trout migration as 20°C, rather than 18°C, EPA (2003) acknowledged that juvenile Pacific Northwest O. mykiss have sufficient aerobic scope for the energetic demands of river migration even at a temperature 2°C above the 7DADM for juvenile growth.” (p. 2). “This information should help define more accurate criteria for thermal performance of juvenile O. mykiss rearing in the lower Tuolumne River. Specifically, the temperature indices and the shape of the aerobic scope curve derived in the present study can also be compared with those of other O. mykiss populations and with the EPA (2003) recommendations.” (p. 5).


8

“Based on the EPA (2003) 7DADM criteria alone, one would predict that wild O. mykiss captured from the Tuolumne River for the present tests would show the following… Maximum metabolic rate (MMR) will increase with test temperature and reach a peak around 18°C according to the EPA criterion…Absolute aerobic scope (AAS) has a Topt around 18°C according to the EPA criteria…AAS will decline at a temperature just above 18°C…Factorial aerobic scope (FAS) will decline with increasing temperature, reaching a value < 2 (i.e., MMR is less than twice RMR) at a temperature just above 18°C.” (p. 6).

“These results for MMR are inconsistent with our prediction #2 derived from EPA (2003) criteria where MMR was expected to peak near to 18°C.” (p. 15).

“Thus, given the good agreement with existing literature for MR measurements combined with the fact that the shape of the response curves will be independent of the methodological concerns noted above, we are confident in using these response curves to test the predictions based on EPA (2003) and our alternative predictions.” (p. 20). “Collectively, the results show clear deviations from our predictions based on EPA (2003), and consistency with the alternative predictions, which suggests the likelihood that the Tuolumne River O. mykiss population is locally adjusted to warm thermal conditions. In particular, the Topt for AAS was 21.2°C, markedly higher than 18°C. Furthermore, AAS at 18°C was numerically the same as that at 24.5°C.” (p. 20). “These data on the RMR, MMR, AAS, and FAS were consistent with higher thermal performance in Tuolumne River O. mykiss compared to that used to generate the 7DADM value of 18°C using Pacific northwest O. mykiss (EPA 2003).” (p. 24).

In these examples, the Report suggests the EPA (2003) conducted or reviewed aerobic scope experiments using O. mykiss from the Pacific Northwest, and these test data were used in the derivation of its 18°C criterion. If this is so, the Report should describe or cite the experiments the EPA conducted or considered, what the results were, and how the EPA used the aerobic scope data to derive its 18°C criterion. The EPA also considered the results of several other types of studies to establish its thermal criteria (EPA 2003, Table 1, p. 16), and so the Report should at least discuss them when comparing recommendations. We reviewed the Report in the light of the recommendations of Clark et al. (2013a), intended to ensure that swim-tunnel aerobic metabolism measurements are sound and comparable between studies. We have not conducted aerobic scope experiments ourselves, and so the list below is to highlight areas where we did not fully understand the study design, methodological approach or data analysis.


9

Two O. mykiss individuals were tested at each exposure temperature, rather than testing the same individual over the range of exposure temperatures. Clark et al. (2013a) note the latter design allows repeated-measures analyses. Can you explain this study-design decision, and how it might have affected the resulting estimates of the optimal temperature for aerobic scope?

The Tuolumne River temperatures where test fish were captured ranged between 12.7

and 17.1°C, and the fish were held in tanks (prior to testing) in Tuolumne River water between 12.5 and 13.6°C (p. 9). The time from fish capture in the river to placement into holding tanks ranged up to 2 hours, and fish remained in holding tanks for up to another 3 hours before transfer to a swim tunnel respirometer. In the swim tunnel, fish were held for another 1 hour, underwent a 1-hour training swim, and then recovered for 1 hour, all at 13±0.3°C. Then, water temperatures were increased to the test temperature for each pair of fish (ranging from 13 to 25°C), at least 8 hours before swimming tests began. (p. 10). Experts have noted the physiological mechanisms of thermal acclimation in modulating aerobic scope are not well understood (Clark et al. 2013b, p. 2778). Could the short-term (several hours of) temperature acclimation described above have affected the test results, or are longer acclimation intervals (e.g. weeks) required to affect aerobic scope responses?

Experts also note there appear to be few studies of the potential thermal acclimation of metabolism during long-term exposures (Clark et al. 2013b, p. 2778). The test fish were sampled from the small population of age-1 and older survivors of exposures to the warm thermal conditions in the lower Tuolumne River, over one or more summers. It seems the aerobic scope responses of the less-tolerant individuals in a population are relevant if this response information is to be used to establish thermal criteria protective at the population level. Obviously the individuals that do not survive (over one or more summers) could not be tested with the field-based study design used. How could the long-term temperature acclimation experienced by the individuals tested, and the inability to test the less tolerant individuals in the Tuolumne River population, have affected the test results?

Clark et al. (2013a) recommend conducting aerobic scope tests beyond what the fish experience in nature. We understand ESA collection permit restrictions prevented tests at temperatures higher than 25°C (p. 15), but these warmer conditions occur often in the lower Tuolumne River. How did this experimental constraint affect the test results? For example, the experiments yielded an average aerobic scope that remained within 5% of optimum from 17.8°C to 24.6°C. Since the upper range value of this estimate is near 25°C, are additional data points (from exposures at temperatures >25°C) necessary to produce a curve that more precisely estimates an optimum temperature or range for aerobic scope?


10

We noted the imprecise fit (r2 = 0.099) of the curve through the aerobic scope data points (Figure 4, p. 37). In contrast, the fit of the curve for the resting (routine) metabolic rate was good (r2 = 0.8), and better than for maximum metabolic rate (r2 = 0.49). Are these findings common in aerobic scope studies? How does the imprecise fit of the aerobic scope curve affect the reliability of the estimates of optimal temperature for aerobic scope, the comparability of the results with other studies, or its use in setting protective temperature criteria?

During tests at faster swimming speeds (to determine maximum metabolic rate), roughly 50% of the test fish used their caudal fin to prop themselves on the downstream screen to avoid swimming (p. 12). How did the steps taken (repeated stimulation with high-velocity bursts to encourage swimming) affect the aerobic scope results, and the estimates of the optimal temperature for aerobic scope? We noted the maximum metabolic rate was found to be lowest for fish exhibiting this behavior (p. 19); did this affect the precision of the maximum metabolic rate and aerobic scope results?

To account for individual variation in body mass, metabolic rate was corrected for fish

mass using the exponent 0.95, a value halfway between the life-stage-independent exponent determined for resting and active zebrafish (p. 11). We noted Cech et al. (1990) use the exponent 0.67 in their tests with multiple species, including O. mykiss (p. 97). How does using an exponent from a fish other than O. mykiss affect the aerobic scope results, the estimates of optimal temperature for aerobic scope, or the comparability of the results with other studies?

Appendix 2 (p. 47) lists two fish mortalities (W37 and W38) that were attributed to chloride residue in the swim tunnels, but it is not clear how this cause of death was determined. Did these fish undergo aerobic testing prior to death, and if so what were the exposure temperatures?

The Report’s conclusions are unclear about how the aerobic scope results are to be used. As mentioned in the “Introduction,” the Tuolumne River investigations were conducted as part of a study (W&AR-14) for the FERC re-licensing proceeding, one that intends to provide information to reassess the temperature guidelines (EPA 2003) for protection of the O. mykiss in the lower Tuolumne River. In this context, we found the wording of the Report’s “Conclusion” (p. 24) could be interpreted to recommend either:

1) A maximum 7DADM value of 22°C should be applied as the new summer criterion to protect the lower Tuolumne River O. mykiss population, replacing the EPA’s recommended maximum 7DADM of 18°C.


11

2) The aerobic scope study results (including the upper performance limit of 22°C) should be used in a new criterion-setting process, to determine an alternative to the 18°C criterion recommended by the EPA.

Below, we consider both possible interpretations of the Report’s conclusions, and provide comments and suggestions for each. If the Report’s conclusion meant to recommend replacement of the EPA’s current summer criterion (a maximum 7DADM of 18°C) with one based on the Tuolumne River aerobic scope studies (a maximum 7DADM of 22°C), we disagree. The Report’s “primary purpose” statement establishes that the swim tunnel test data (metabolic rate measurements and related computations, e.g., AAS, FAS, etc.) were the single line of evidence for determining the optimal temperature range for juvenile O. mykiss in the Tuolumne River:

“The primary purpose of this study is to determine the thermal performance of the subadult (100-200 mm fork length; FL) O. mykiss population inhabiting the lower Tuolumne River (LTR) to assess any local adjustment in thermal performance. Thermal performance was assessed as the range of temperatures over which juvenile O. mykiss can increase aerobic metabolic rate (MR) beyond basic needs. This aerobic capacity could be used for any of the normal daily activities of O. mykiss in the Tuolumne River during its normal life history (swimming, catching prey and feeding, digesting a meal, growing, avoiding predators, defending territory, etc.). Thus, MR measurements were used to determine the optimal temperature range for Tuolumne River O. mykiss.” (p. 4). [Underline and bold emphasis added]. “In the present study, we use 95% of the peak AAS value to set the optimal thermal range (Figure 1; the two temperatures that bracket Topt are termed a Pejus temperature, Tp).” (p. 5). [Underline and bold emphasis added].

The Report’s sole reliance on aerobic scope data is justified by an assertion that AAS defines a fish’s capacity to perform essential life function activities (p. 4). We do not find adequate justification in the Report (or referenced therein) to agree that aerobic scope information alone defines a fish’s capacity to carry out its life functions. As discussed above, more explanation is warranted about how a measure of the aerobic capacity to perform a life-history function (in short-term tests, under a given temperature regime) is adequate to assess how well a temperature regime promotes population-level fitness in the natural environment. The EPA’s (2003) criterion-setting approach for juvenile salmon and O. mykiss considered multiple lines of evidence gathered from studies of several physiological and ecological metrics. These included:

The water temperatures (over a one-week exposure) causing death;


12

The optimal water temperatures for growth, under limited and unlimited food supplies; The thermal preferences of rearing juveniles, in both laboratory and field studies; The water temperatures causing impairments to juvenile smoltification; and The water temperatures leading to minimal, elevated, and high disease risks.

(EPA 2003, Table 1, p. 16). The EPA noted the importance of ecological as well as physiological considerations when establishing numeric water temperature criteria (EPA 2003, p. 18). Accordingly, they considered not only “physiological optimum temperatures” (where functions like growth, swimming, heart performance, etc. are optimized), but also “ecological optimum temperatures” which they described as those where fish do best in the natural environment - considering food availability, competition, predation, and fluctuating temperatures. The EPA summarized their considerations

for salmon and trout juvenile rearing:

“EPA looked at both laboratory and field data and considered both physiological and ecological aspects. Optimal growth under limited food rations in laboratory experiments, preference temperatures in laboratory experiments where fish select between a gradient of temperatures, and field studies on where rearing predominately occurs are three independent lines of evidence indicating the optimal temperature range for rearing in the natural environment. As highlighted in Tables 1 and 2 (and shown in detail in the technical issue papers3) these three lines of evidence show very consistent results, with the optimal range between 8 - 12°C for bull trout juvenile rearing and between 10 - 16°C for salmon and trout juvenile rearing.” (EPA 2003, p. 19). [Footnote added].

The EPA recommended a maximum 16°C 7DADM for salmon and trout “core” juvenile rearing, in a river basin’s mid-to-upper reaches with moderate to high juvenile densities. The 16°C criterion is to: (1) safely protect juvenile salmon and trout from lethal temperatures; (2) provide upper optimal conditions for juvenile growth under limited food during the period of summer maximum temperatures and optimal temperatures for other times of the growth season; (3) avoid temperatures where juvenile salmon and trout are at a competitive disadvantage with other fish; (4) protect against temperature-induced elevated disease rates; and (5) provide temperatures that studies show juvenile salmon and trout prefer and are found in high densities (EPA 2003, p. 27). As mentioned earlier, the historical habitat suitable for summer-rearing O. mykiss was in the higher and colder upper Tuolumne River watershed (Lindley et al. 2006). In the Tuolumne River, the La Grange Dam (at RM 52) currently blocks the O. mykiss in its lower elevations from reaching the upper watershed. The EPA recommended a maximum 18°C 7DADM criterion for salmon and trout migration plus “non-core” juvenile rearing, recognizing that juvenile salmonids will experience waters that have a higher temperature than their optimal thermal range. This criterion is to: (1) safely protect 3 The 3-year interagency cooperation that produced EPA Region 10 Guidance for Pacific Northwest State and Tribal

Temperature Water Quality Standards (EPA 2003) also produced 6 scientific issue papers, 2 scientific peer review panel reports, and the integration of public comments.


13

against lethal conditions for both juveniles and adults; (2) prevent migration blockage conditions for migrating adults; (3) provide optimal or near optimal juvenile growth conditions (under limited food conditions) for much of the summer, except during the summer maximum conditions, which would be warmer than optimal; and (4) prevent adults and juveniles from high disease risk and minimize the exposure time to temperatures that can lead to elevated disease rates (EPA 2003, p. 28). This criterion is applicable in a river basin’s mid-to-lower reaches, with moderate to low juvenile densities (downstream of the core juvenile rearing reaches). In the Tuolumne River, all of the habitat currently accessible to rearing juvenile O. mykiss is “non-core” (due to the impassable La Grange Dam). This point is highly relevant to the management of the thermal regime in the lower Tuolumne River because its steelhead population is now confined to the lower reaches, where flow and temperature management must accommodate both its fall-run Chinook salmon and its steelhead. While individual O. mykiss that express the steelhead life history smolt and migrate to saltwater, they do so only after remaining in freshwater (and over-summering) for 1 to 3 years. In contrast, fall-run Chinook juveniles are predominantly “ocean-rearing” fish that reside a shorter time in freshwater and do not over-summer in-river. Myrick and Cech (2004) recognize how the extended, over-summer residence time before smolting for O. mykiss (compared to Chinook salmon) make them more vulnerable to alterations of the natural thermal regime, and to the management of thermal regimes designed to protect juvenile fall-run Chinook salmon (p. 115). As mentioned above, the EPA also recommend a maximum 7DADM of 14°C for O. mykiss smoltification. The interval for O. mykiss smoltification in the lower Tuolumne River could extend from spring to early summer. Dams and diversions in the watershed cause large-scale alterations in the magnitude, frequency, timing, duration, and rate-of-change of streamflows in the lower Tuolumne River (McBain and Trush 2000; Anderson 2009), and these flow alterations affect the temperature regime during the O. mykiss smoltification and out-migration intervals (Gordus 2009). Anderson (2009, p. 14) explains how snowmelt floods in the Tuolumne River once provided large-magnitude, sustained flows peaking from late April to early June, then receding in July and August. With the onset of impoundment and diversion, snowmelt floods have been largely eliminated from the annual hydrograph and replaced with FERC license-required spring “pulse flows” intended to stimulate smolt emigration. The snowmelt recession limb of the annual Tuolumne River hydrograph has also been altered; the duration and rate of the snowmelt recession plays an important role in the juvenile rearing and smolt outmigration life stages. Before the large dams and diversions were in place, these recession flows connected the snowmelt peak flows to the summer baseflows, and generally extended into July during dry years and into August in wetter years. The receding limb of the spring pulse flows that now occur in the lower Tuolumne River (under the FERC license) is often early, short, and steep (Anderson 2009, p. 14), resulting in diminished flood plain inundation duration (Gard 2009, Table 1, p. 9) as well as warmer water temperatures in the late spring and early summer months (Gordus 2009).


14

In their temperature criteria-setting considerations, the EPA (2003) recognized that dams can significantly reduce the river flow rate, thereby causing juvenile migrants to be exposed to high temperatures for a much longer time than they would under a natural flow regime (p. 7). If the Report’s conclusions are meant to recommend the future development of a site-specific alternative to the EPA’s recommended criterion (18°C maximum 7DADM), we do not find adequate explanation of how the Tuolumne River aerobic scope results should be used in that process. The California State Water Quality Control Board (State Board) has applied the EPA criteria (2003) to evaluate long-term, field-based assessments of temperature suitability for juvenile O. mykiss (and other salmonids) in the lower Tuolumne River4; the temperature data and analysis were submitted to the EPA, who found the information supports a listing of the lower Tuolumne River on the Federal Clean Water Act § 303(d) list of impaired water bodies (due to unsuitably high water temperatures for the life stages of anadromous fishes). Consequently, the EPA is requiring the State Board to develop a total maximum daily load (TMDL) for temperature in the lower Tuolumne River, by the year 2021. The EPA (2003) recognized that situations exist where their general guidance can be supplanted with site-specific criteria (p. 34). The NMFS supports options within EPA’s guidance for the development of alternative criteria, where the general numeric criteria are unattainable or inappropriate (Enclosure C). If the Report is recommending the use of its Tuolumne River aerobic scope results in the development of a new temperature criterion, it is not clear how. Neither the Report, nor the Turlock and Modesto Irrigation Districts’ Study W&AR-14 (which plans to assess the existing lower Tuolumne River temperature criteria) describe plans or a process to coordinate with the State Board or the EPA to develop alternative criteria. We do not hold that aerobic metabolism studies alone can define the singular optimum temperature for a fish, or establish protective temperature criteria for O. mykiss. Below, we list and discuss topics for consideration in any process to set new temperature criteria or develop a temperature-related TMDL for the lower Tuolumne River.

1. Review the long-term dataset of water temperatures measured in the lower Tuolumne River during the summer.

The California Department of Fish and Wildlife collected water temperature data from 1998 through 2006, from several locations in the lower Tuolumne River (Gordus 2009). The data clearly indicate summer temperatures in the lower Tuolumne River exceed the EPA’s summer recommendations (2003) for juvenile O. mykiss rearing, over several miles of the lower

4 The California Department of Fish and Wildlife collected data from 1998 through 2006, using in situ devices

(probes) deployed at various locations in the lower Tuolumne River, and that continuously measure and record water temperatures. The analysis of the data applied the EPA criteria (2003) as the benchmarks to determine the impairments of the anadromous fish beneficial uses established by the State Board.


15

Tuolumne River, and do so across multiple water-year types and summer weather conditions. It is likely additional data are available from years earlier than 1996, and new data have been collected since 2006. This long-term dataset could be used to extend the thermal suitability analysis, for use in any process for setting new temperature criteria. For example, it could be used with flow gage and air temperature data to assess how lower Tuolumne River water temperatures change under various water-year types, discharge regimes (from the La Grange Dam), and atmospheric conditions, including the conditions during the recent California drought.

2. Review information about the general trends observed in O. mykiss distribution and abundance in thermal “zones” of rivers that grade from colder to warmer, including in California.

The Report discusses the redband rainbow trout (O. mykiss gairdneri) found in eastern Oregon and Idaho (p. ii., pp. 2-3) where summer temperatures can reach 26ºC or greater (Zoellick 1999). While the presence of this cold-water fish is documented at these high temperatures, studies of its distribution and abundance suggest strong negative correlations between trout densities and stream temperatures. In the John Day River basin (Oregon), Li et al. (1994) found negative correlations between redband trout densities of all age-classes and maximum stream temperatures (Table 3, p. 634), and a negative relationship (r = –0.71, P < 0.05) between trout biomass:invertebrate biomass ratios and stream temperatures (Figure 7, p. 637). Ebersole et al. (2001) found redband trout in twelve studied northeast Oregon stream reaches, with some individuals present where mean daily maximum temperatures were 25°C; however, the authors reported a negative correlation (r = –0.70, P = 0.01) between mean trout density and mean daily maximum water temperatures (Figure 3, p. 6). Zoellick (2004) studied the redband trout in two southwestern Idaho creeks, and found trout density negatively correlated (r = –0.76, P = 0.03) with increases in water temperature in both creeks (Figure 4, p. 24); density and biomass were greater in the stream with lower daily maximum water temperatures (range 18° to 22°C) than in the stream with higher temperatures (range 20.2° to 26°C) (Figure 3, p. 23). The Report also notes the presence of O. mykiss in southern California streams where ambient summer water temperatures routinely exceed 25°C. However, the O. mykiss population sizes and densities in these warmer streams and stream reaches are low, and negatively correlated with temperature. Douglas (1995) studied the O. mykiss in the Santa Ynez River watershed (Santa Barbara County, California), where extensive runoff occurs during winter rains but the higher flows of winter/spring rapidly diminish by summer. As a result, much of the mainstem Santa Ynez and most of the smaller streams dry up by late summer or fall each year, including areas where winter and spring O. mykiss spawning occurs. Surface water at those times is limited to large, isolated pools or groundwater seepage flows in the upper reaches of some streams (p. 9). The author found the juvenile O. mykiss density negatively correlated (r = –0.61, P < 0.05) with the temperature difference between shallow (warmer) and deep (cooler) habitats within the same stratified pools (Table 2, p. 27). Various aspects of temperature (range and maxima for all habitats, and the difference in mean temperatures of shallow and deep pool sub-habitats) showed strong negative correlations (range -0.50 to -0.68) with the density of adult trout (Table 3, p. 28).


16

This study of O. mykiss densities was limited to the upper reaches of the Santa Ynez watershed because mean water temperatures were 21°C a short distance (~2 kilometers) below Bradbury Dam, and then rose to over 25°C farther downstream and reached maxima near 29°C around Lompoc, California (p. 56-58). Matthews and Berg (1997) examined summer (August) O. mykiss distribution in two adjacent pools in Sespe Creek (Ventura County, California) where water temperature reached 28.9°C. In “Pool 1” water temperatures ranged from 21.5°C at the bottom to 28.9°C at the surface. After 5 August, trout were no longer found in this pool, suggesting they had moved out of the high temperature water, or died. In the adjacent “Pool 2” surface water temperatures were as high as 27.9°C, but temperatures on the bottom remained cooler (17.5 to 21.0°C) than in Pool 1, and O. mykiss aggregations were consistently observed in this lower stratum throughout the study period despite lower dissolved oxygen values. Regression analysis showed mean O. mykiss numbers per pool region were inversely related to water temperature and dissolved oxygen: the mean number of trout per region decreased as dissolved oxygen and temperature increased (significant regression coefficients, P < 0.05) (Table II, p. 67; Figure 8, p. 68). Eaton et al. (1995) assembled a large field-based database, the “Fish and Temperature Database Matching System” (FTDMS), to match spatial and temporal stream temperature records with fish sampling events; this information was used to estimate the yearly temperature regimes used by 30 freshwater fishes common across the U.S. (including O. mykiss). An impressive total of 207,846 weekly mean fish/temperature datasets from 29 states were generated and analyzed. The results showed good agreement between assignments of species to coldwater, coolwater, or warm water “guilds” based on earlier laboratory mortality test data. The coldwater “trout-salmon-whitefish” species (including O. mykiss) were found at distinctly lower temperatures than the others, comprising the 9 lowest FTDMS data points on a plot including all 30 fishes (Figure 2, p. 13), with an obvious separation (near 23°C) between the coldwater guild and the others. In a study of 130 sites in California’s San Joaquin River watershed (including the Tuolumne River), Moyle and Nichols (1973) used correlative analyses (between environmental variables and species abundances) to identify four distinct fish species “associations.” They found the O. mykiss (rainbow trout) association overlapped the least with the others; strong correlative relationships were observed between the distribution and abundance of O. mykiss and the variables temperature and elevation (which were highly negatively correlated with one another). In general, the “Rainbow Trout Association” was found in the colder, higher-elevation permanent streams; the “California Roach Association” was found in the small, warm intermittent tributaries to the larger streams; the “Native Cyprinid-Catostomid Association” was found in the larger low elevation streams; and the “Introduced Fishes Association” was found in the low-elevation, intermittent streams that had been highly modified. From 1993 to 1995, Brown (2000) sampled twenty sites in the lower San Joaquin River drainage (including four sites in the lower Tuolumne River) to characterize fish communities and their associations with measures of water quality and habitat quality. His results indicate the native


17

fish community can persist in the human-modified stream reaches below the major foothill dams, but the downstream range of the community appeared to be limited, particularly in the Merced and Tuolumne rivers. Also, although the native community is still present, introduced species may be present at the same sites in low to moderate percentage abundances (p. 263). We acknowledge that correlative relationships do not establish or confirm causation, and so we sought out studies that examined the linkages between aerobic metabolism and the distribution or abundance of O. mykiss. Cech et al. (1990) studied the metabolic responses and tolerance limits of several California fishes to temperature and dissolved oxygen. Their experiments were conducted in flow-through respirometers, with the following test fishes: O. mykiss, Sacramento sucker, Sacramento squawfish (common name changed to Sacramento pikeminnow), hardhead, California roach, riffle sculpin, and tule perch; all are native Californian species that comprise recognized species associations. Resting metabolic responses were measured because they were thought to be a basic property of each species, and could be used to compare species under similar conditions of temperature and hypoxia; the authors explain that active metabolic rates would not be comparable because the rate of energy usage during activity depends on the type of activity (thermodynamic path) more than some basic property of individual species (p. 97). When the test results were compared with relevant field observations of occurrence and distributional limits (e.g. Moyle and Nichols 1973), the authors found good correspondence, suggesting a major role for temperature and dissolved oxygen as ecological determinants of California stream fish distributions. Based on their data and other physiological studies, they concluded the effect of temperature on O. mykiss metabolism prevents them from inhabiting the squawfish-sucker-hardhead and California roach zones during the warmer seasons (except that O. mykiss could be found in these lower-elevation zones if cold outflows existed downstream of dams) (p. 101). The general trends observed in O. mykiss distribution and abundance in the warmer thermal “zones” of rivers and streams, including in California, suggest the species may exist there in reduced numbers, but will not thrive in the warmer conditions. During any process for setting new temperature criteria for the lower Tuolumne River, consideration should be given to the already-limited size of the colder-water zone directly downstream of the La Grange Dam (Gordus 2009), and the potential adverse consequences to O. mykiss of further reducing the downstream extent of this zone if warmer criteria are adopted.

3. Review the long-term snorkeling data for O. mykiss to understand the site-specific distribution of the population in the lower Tuolumne River.

A preferred temperature range is that which the fish most frequently inhabits when allowed to freely select temperatures in a thermal gradient (McCullough 1999, p. 10). Among the independent lines of evidence considered by the EPA during its criteria-setting process (2003) were preference temperatures indicated by laboratory experiments (where fish select between a gradient of temperatures), and field studies indicating the stream locations where juvenile salmonid rearing predominately occurs. The temperatures selected by fish in the field may be


18

important companion studies to conduct alongside aerobic scope experiments because one cannot assume that a fish will occupy or seek out temperatures where aerobic scope is maximal. For example, Norin et al. (2014) found that while the Topt for aerobic scope was 38°C for a tropical fish, when given the opportunity the fish selected a median temperature of 31.7±0.5°C, and spent only 10±3% of the time at temperatures >36°C. In the Tuolumne River, the respirometer experiments yielded a Topt for aerobic scope of 21.2°C and an average aerobic scope that remained within 5% of this peak from 17.8°C to 24.6°C (p. ii, pp. 23-24). However, the O. mykiss in the lower Tuolumne River do not appear to seek out temperatures that maximize their aerobic scope. Snorkeling surveys have been conducted annually (sometimes multiple times per year) in the lower Tuolumne River over the past 31 years (1982-2014) (TID and MID 2015). These long-term snorkel survey data (TID and MID 2015, Figure 3; reproduced in Enclosure D) clearly show that O. mykiss over-summering in the lower Tuolumne River are found in the colder waters (Gordus 2009) near their upstream migration limit (the La Grange Dam). This is consistent with the behavior of the redband trout of Eastern Oregon. Anderson et al. (2011) studied their seasonal migrations in the Donner and Blitzen River (Oregon), using radio telemetry and passive integrated transponder (PIT) tags. During both summers of the study, temperatures in the lower river exceeded the 24.3°C ultimate upper incipient lethal temperature of the redband trout (Bear et al. 2007), and approached its 29.4°C critical thermal maximum (Rodnick et al. 2004). While the authors expected to observe adult (mature) trout moving upstream in spring (to find suitable spawning habitats), they did not expect to also observe the long-distance upstream migrations (at the same time of year) by sub-adult (immature) trout (p. 10); they interpreted these movements as actions to find favorable thermal conditions and good foraging opportunities. Gamperl et al. (2002) found that redband trout from the Little Blitzen River and nearby Bridge Creek both exhibited a preferred temperature of approximately 13°C. A visual overview of Figure 3 (Enclosure D) suggests the lower Tuolumne River O. mykiss in summer are consistently clustered in the uppermost ~5 miles (from RM 46.9 upstream to RM 51.6). By performing computations with the survey data from 1997-2014, we found 79% of the O. mykiss snorkel observations were in this uppermost 4.7 miles of the lower Tuolumne River. When 1997-2014 snorkel data from only the warmest months (July and August) are considered, 86% of the O. mykiss observations were in the upper 4.7 miles. The last two years of available snorkeling data (2013-2014) were recorded in very warm and dry (drought) conditions; in these two years, 96% of the O. mykiss observations were in the upper 4.7 miles. The Report refers to fishery survey data gathered since 1997 to define the extent of “primary rearing habitat” for O. mykiss, which the authors consider to be the 12.4 miles from RM 39.6 extending upstream to near the La Grange Dam at RM 52 (p. 1). We adopted the Report’s temporal (1997) and spatial (RM 39.6) divisions to further review the snorkel data, and found that surveys from 1997-2014 reported 99% of the O. mykiss were observed in the uppermost 12.4 miles of the lower Tuolumne River. This result calls in to question the Report’s use of RM 39.6


19

to delineate the downstream limit of the primary O. mykiss habitat; this location appears to define the downstream extent of the total habitat because very few (only 1%) of individuals were observed in the sites downstream of this location. We suggest a more realistic downstream demarcation for the primary O. mykiss summer habitat is well above RM 12.4, and closer to RM 5. Also, the Report noted maximum water temperatures at RM 12.4 have exceeded 27°C during the summer months (p. 1), a near-lethal temperature for O. mykiss; such high temperatures are a plausible explanation for the upstream, clustered distribution of O. mykiss consistently observed in the snorkeling data. The model produced by Lindley et al. (2006) predicts that all 201 miles of stream habitat in the Tuolumne River watershed suitable for summer occupancy by O. mykiss was historically found upstream of the La Grange Dam (which now blocks O. mykiss in the lower Tuolumne River from reaching the upper watershed). If RM 12.4 is considered the present downstream extent of suitable summer habitat for O. mykiss in the Tuolumne River, then the existing habitat accessible to CCV steelhead is a low 6% of the historical amount. If RM 5 is considered the present downstream extent of suitable summer habitat, then the existing accessible steelhead habitat is an even lower 2.5% of the historical amount. Also notable is that access for juvenile salmonids to rearing habitats in permanently-flowing tributaries has been entirely blocked by the La Grange Dam. Any process of setting new temperature criteria for the lower Tuolumne River should consider the potential negative effects of reducing even further the size of the already-limited colder water habitat below the La Grange Dam (Gordus 2009).

4. Review the lower Tuolumne River snorkeling data to understand the abundance of O. mykiss, to assess the site-specific population status and trends.

Our review of the summarized Tuolumne River snorkel surveys suggests the numbers of O. mykiss have been, and continue to be, very low under the flow and thermal regimes (TID and MID 2015, Table 2; reproduced in Enclosure E). Over the 31 years monitored, the numbers of O. mykiss observed (survey data in all sites combined) averaged 182 per year. We noted the remarkably low numbers from 1987-95, when in 7 of the 9 years surveyed zero individual O. mykiss were observed in any of the snorkeled locations, which spanned ~25 miles of the River. During the best two years of this interval, a single O. mykiss was observed in 1992, and 3 individuals were observed in 1995 (Table 2). The more recent data (from 1997 to the present) suggest some (modest) improvement, with the observed number averaging 306 per year (all sites combined). However, the numbers are still very low in most years, and the data are uneven; as recently as 2014, only 53 O. mykiss were observed in 12 survey sites that spanned ~20 miles of the lower Tuolumne River. We understand observations from snorkel surveys will underestimate the actual numbers of over-summering O. mykiss present. However, the numbers observed are so low they undeniably


20

suggest a lower Tuolumne River generally depauperate in O. mykiss, especially when one considers the long-term duration (30+ years) of the surveys and how they have been conducted over several miles of the lower Tuolumne River. During any process of setting new temperature criteria for the lower Tuolumne River, the snorkel, water temperature, and flow-gage datasets could be used to further explore relationships between summer flows/temperatures and the abundances and distributions of O. mykiss. Fish densities could be calculated (by dividing observed or estimated fish numbers by the river miles surveyed), allowing comparisons with other California and western U.S. streams (Platts and McHenry 1988); the existing snorkel numbers suggest very low salmonid densities in the lower Tuolumne River. Continued monitoring of the responses could occur to evaluate responses as flow release adjustments occur.

5. Consider that unless the very low size of the total O. mykiss population surviving over the summer in the lower Tuolumne River is substantially increased, a viable CCV steelhead population cannot be achieved.

A common misconception is that declines in anadromous fish populations are caused either by factors in their freshwater habitat or by changes in their marine environment – but the effects are not mutually exclusive (Lindley 2009). Regarding the freshwater effects (i.e., in the lower Tuolumne River), CCV steelhead juveniles must remain in their natal rivers (and over-summer) for 1 to 3 years before they smolt and migrate seaward (Myrick and Cech 2004). Therefore, the in-river conditions during summer (including water temperature) are key factors limiting survival to the age of smoltification. The lower Tuolumne River snorkeling data indicate the over-summer survival of the total O. mykiss population is very low, much too low to achieve a viable steelhead population (measured by adult returns, see Lindley et al. 2007). This is because only a fraction of the (already very low) total juvenile O. mykiss population can be expected to smolt; to that fraction, the low natural rates of smolt-to-returning-adult survival must be considered. Mortality during the smolt migration and early ocean occupancy may exceed 95% (Kendall et al 2015). Unless several thousand more sub-adult O. mykiss are produced, and survive in the lower Tuolumne River at least one summer until they smolt, only a small number of adult steelhead can be expected to return. The available data about adult steelhead returning to the Tuolumne River reflect this situation. Since 1940, very low numbers of adults have been observed passing upstream in the lower Tuolumne River. Older accounts report 66 steelhead passed upstream of the former Dennett Dam (RM 16.2) between October 1 and November 30, 1940, and 5 steelhead passed in late October of 1942 (CDFG 1993, in Stillwater Sciences 2013, p. 5-35 to 5-36). More recent counts at the lower Tuolumne River weir (installed in 2009, at RM 24.5) indicate very few steelhead enter the Tuolumne River:


21

2009: One O. mykiss was documented passing upstream on November 7, 2009 (Stillwater Sciences 2013, p. 4-1).

2010: During monitoring from September 9, 2010 to December 1, 2010, no O. mykiss were recorded passing upstream through the weir (FISHBIO 2011, p. 11).

2011: Four O. mykiss were recorded passing through the weir between September 16, 2011 and December 31, 2011 (FISHBIO 2012, Table 3, p. 11). One individual was recorded as an ad-clipped fish. We note these fish may not have been adult steelhead because of their small sizes, ranging from length 14.2 to 16.5 inches.

2012: Three O. mykiss (no adipose fin clips) were recorded passing through the weir between September 24, 2012 and December 31, 2012 (FISHBIO 2013, Table 3, p. 13); however, we note these fish may not have been adult steelhead because of their small sizes, ranging from length 14.5 to 16 inches.

2013: No O. mykiss were recorded passing the weir between September 24, 2013 and December 31, 2013 (FISHBIO 2014, p. 11).

2014: No O. mykiss were recorded passing the weir between September 29, 2014 and December 31, 2014, (FISHBIO 2015, p. 12); during a second monitoring interval from January 1, 2014 to May 7, 2014, no O. mykiss were recorded passing the weir (p. 16).

Collectively, this information suggests that very low numbers of adult steelhead return to the lower Tuolumne River. Although the weir monitoring has mostly focused on the study of fall-run Chinook, the monitoring period of late September through December since 2011 covers the peak immigration of adult steelhead to Central Valley rivers, including the lower Tuolumne River. Within a weir monitoring season, there may be smaller temporal gaps for many reasons, including high flow conditions that may affect weir operations or prevent them altogether. Also, the weir location may be upstream of some existing steelhead spawning habitat (FISHBIO 2015, p. 2). This information suggests that the actual numbers of immigrating adult steelhead may be higher than reported. In addition, rotary screw trap monitoring data indicate that a few to no steelhead smolts leave the lower Tuolumne River (Stillwater Sciences 2013, p. 3-1; p. 4-1), while we acknowledge that screw traps may not be efficient at capturing steelhead smolts. It is not completely understood why some O. mykiss migrate to the ocean while others do not. Life history expression is thought to be partly under genetic control (Pearse et al. 2014), as well as influenced by environmental factors. Sloat and Reeves (2014) investigated (in laboratory stream mesocosms) the influences of individual condition, standard metabolic rate, and rearing temperature on anadromous versus resident life history expression. Given the high relevancy of this study, we provide below an overview of the experimental methods. Juvenile O. mykiss spawned from anadromous parents (Clackamas River Hatchery, Oregon) were separated into “dominant” and “subordinate” behavioral groups, based on competitive outcomes. Fish


22

considered “dominant” were the individuals that established and defended feeding territories, while those considered “subordinate” dispersed downstream within the artificial stream mesocosm. Within each separated group, standard metabolic rate was measured in a randomly selected subset of individuals, using intermittent-flow respirometry (in tests similar to the Report’s “swim tunnel” metabolic experiments). Fish belonging to the dominant group were found to exhibit higher standard metabolic rates than fish from the subordinate group. The high and low standard metabolic rate groups were then reared under alternative “warm” and “cold” thermal regimes to complete the experiments. Rearing occurred until fish exhibited phenotypic traits associated with either anadromous (smolting) or resident (sexual maturation) life histories. The warm regime consisted of seasonally adjusted temperatures between 6 and 18°C, and the cold regime ranged between 6 and 13°C. Individual body size was tracked throughout the experiment, to determine how standard metabolic rate, rearing temperature, and individual growth trajectories influenced life history expression. Whole-body lipid content was determined in a random selection of 30 males and 30 females per life history type, per temperature treatment. This lipid data was collected because temperature-driven changes in energy storage may have important consequences for salmonid life histories; juveniles exposed to lower stream temperatures tend to allocate more energy towards storage, at a cost to somatic growth, and energy reserves (in the form of lipids) are needed to initiate the sexual maturation process. The study results suggest that warmer temperature regimes that maximize growth at the expense of energy storage will reduce the probability of freshwater maturation (resident life history expression). However, the authors cautioned against the interpretation that warmer thermal regimes will result in a higher abundance of anadromous individuals (steelhead) because the numerical abundance of salmonids commonly decreases with increasing temperature. They warn that an increase in the proportion of smolts within O. mykiss populations at higher temperatures may be offset by decreases in total population size (p. 499). It is also noteworthy that the maximum exposure temperature in the “warm” regime did not exceed 18°C. The low total O. mykiss population size in the lower Tuolumne River, along with its low steelhead population, appear consistent with the scenario observed by Sloat and Reeves (2014). So, while improvements in smolt survival (in-river, through the Delta and Bay, and in the ocean) will be necessary to improve the Tuolumne River steelhead population (and for DPS recovery), these efforts will be insufficient unless the in-river total O. mykiss population is increased substantially. Accordingly, any process for setting new temperature criteria should consider the adverse effects upon the Tuolumne River steelhead population of the very low production and over-summer survival of O. mykiss, which has been consistently observed in the lower Tuolumne River (TID and MID 2015, Table 2).


23

6. Review information about thermal stress in O. mykiss, and consider the linkages between biomarker studies and population-level effects.

Research into the underlying molecular and physiological processes that result in different degrees of thermal tolerance might lead to improved models of thermal tolerance (McCullough et al. 2009, p. 93). We sought out clues to understanding why the total population size of O. mykiss in the lower Tuolumne River is so low, and why its O. mykiss seek out summer water temperatures below the reported optimum temperature for aerobic scope. As explained above, O. mykiss are also observed at low densities in the warmer zones of other streams, and have been observed moving (seasonal migrations and diel movements) to colder habitats when conditions warm to near the upper range of their optimal temperature for aerobic scope. If these observations are not explained by aerobic scope performance, perhaps other metrics or endpoints could be considered in a new criterion-setting process. The Report includes “important indications” in the Tuolumne River experiments that some of the tested O. mykiss were stressed at temperatures of 23-25°C; for example, 4 out of 13 (31%) individuals tested at 23-25°C had a FAS < 2, two fish regurgitated their stomach contents at 25°C, and the only fish mortality occurred during recovery after testing at 25°C (p. 20). Field and lab research using physiological indicators of thermal stress in individual fish appear promising, and could complement studies such as water temperature and population-size monitoring.

Feldhaus (2006) studied the link between summer temperatures and physiological performance indices for juvenile redband rainbow trout in the John Day River watershed (Eastern Oregon); the goal was to define summer water temperatures as physiologically “suitable,” “marginal,” or “unsuitable.” The studies suggest that while these fish can tolerate summer daily maximum temperatures in excess of 22ºC, they are physiologically compromised as indicated by the induction of heat-shock proteins, a biomarker for thermal stress (p. 50-51; Iwama et al. 1998). Kammerer and Heppell (2012) conducted studies of the wild redband trout in the John Day River watershed, in two systems with contrasting thermal regimes (colder and warmer) to test the hypothesis that trout are physiologically healthy below a certain temperature threshold, and above this threshold performance declines. They discuss (p. 2) how fish exposed to high temperatures undergo a cellular stress response, in which individual cells mount a host of responses to cope with the stress; heat-shock proteins are described as “cellular guardians” produced in the liver that reduce damage to other cellular components when an animal is exposed to high temperatures. The authors found relative heat-shock protein levels in liver and fin tissue increased with temperatures (Figure 4, p. 8), and were correlated with reduced fish size and changes in body fat, particularly above a 23 °C thermal threshold (p. 11). In a follow-up laboratory study, Kammerer and Heppell (2013) assessed the long-term effects of elevated summer stream temperatures on growth, the cellular stress response, and whole-body lipids in juvenile summer steelhead (Skamania–Columbia basin hatchery stock). Replicated groups of fish were exposed to three different temperatures, 15, 23, and 25ºC, for 25 consecutive


24

days followed by a 55-day recovery period at 15ºC. At temperatures ≥ 25ºC, steelhead consumed significantly less food per day and had elevated feed conversion rates, but experienced slower growth, reduced body size, lower body fat, and elevated heat-shock protein levels relative to fish exposed to 23ºC and 15ºC. Also, growth decreased 24.4% and 27.1% for length and mass, respectively, between 15ºC and 23ºC, and an additional 60% and 56.5%, respectively, between 23ºC and 25ºC. Mortalities occurred only in the 25ºC exposures, beginning 10 days after the start of the experiments. While growth increments and lipid levels recovered to control levels after water temperature was reduced, body size of the fish exposed to 25ºC lagged throughout the experiment. Heat-shock protein levels were detectable up to 25 days after the thermal stress. The authors concluded the results show that thermal stress affects performance, and they suggested a thermal threshold of 23ºC, after which exposed steelhead incur a “physiological debt.” Werner et al. (2005) measured expression of heat-shock protein 72 in white muscle tissue of juvenile steelhead parr (O. mykiss) in the Navarro River watershed (California), and concluded the fish were experiencing cellular stress when maximum daily water temperatures reached 20-22.5ºC. In a related laboratory study, Viant et al. (2003) performed long-term (10-week) exposures of juvenile O. mykiss to 15 and 20ºC, and measured metabolic condition, stress response and growth rate endpoints. In addition to heat-shock proteins, they analyzed multiple “traditional” biomarkers of metabolic condition (using nuclear magnetic resonance-based metabolite profiling) in muscle and liver. The authors found their laboratory results support the conclusion from earlier Navarro River field study, that heat-shock protein 72 is an indicator of thermal stress in juvenile steelhead parr; in fish that experienced a temperature increase from 15 to 20ºC, heat-shock protein levels remained elevated in 15% of these fish for 10 weeks. The authors also found good correlations between increased heat-shock protein synthesis and reduced metabolic condition. While they did not observe reduced steelhead growth rates, they noted the highest exposure temperature was 20ºC and the sufficient food provided likely supported the increased metabolic demand at that temperature or lower. In their studies, Kammerer and Heppell (2012) found good correlations between liver- and fin-tissue heat shock protein 70 in O. mykiss, which suggests a non-lethal sampling method (fin clips) could be used to measure thermal stress. This is relevant because concerns exist over even small “take” of the rare, ESA-listed O. mykiss in the lower Tuolumne River, which could occur during scientific collection. Before setting new temperature criteria to protect the lower Tuolumne River O. mykiss, heat shock proteins or other biomarkers could be examined in fish living in the existing (baseline) summer temperature conditions, to understand if they are experiencing thermal stresses.

7. Consider that at higher temperatures in the lower Tuolumne River, food supply may limit the energy available for growth and survival of O. mykiss, and exert population-level effects.

Bioenergetics models rely on the balanced energy equation of Winberg (1956), as cited in Hartman and Brandt (1995): G = C – (R + SDA) – F – U, where all units are in energy and: G is


25

growth; C is consumption; R is metabolism; SDA is specific dynamic action (or the heat increment associated with biochemical transformation of ingested food into forms that can be metabolized); F is egestion (the removal of indigestible food); and U is excretion (the elimination of waste products resulting from metabolic processes). First, it is evident from this energy equation that metabolism is not the only temperature-dependent variable influencing growth. Second, it would be incorrect to assume growth can be maintained as temperature increases, even if R (or aerobic scope, MMR – RMR) is unchanged or higher at the warmer temperature, unless energy (food) consumption is increased. Weber et al. (2014) applied a bioenergetics mass balance equation similar to the one above, in their study of O. mykiss habitat growth potential in the John Day River basin (Oregon); their focus was on measurements of (invertebrate) food abundance and stream temperature. They observed nearly five-fold differences in the abundances of benthic and drifting invertebrate assemblages among the reaches studied. Their modeling demonstrated that relationships between temperature and salmonid physiological rates are insufficient when attempting to estimate the growth potential of a stream habitat – they found consideration of local invertebrate food availability can greatly increase the accuracy of bioenergetics-based predictions of salmonid growth (p. 1165). Kendall et al. (2014) review the processes and patterns of anadromy and residency in steelhead and rainbow trout, including the interactions among genetics, individual condition, and environmental influences; what is striking is the broad role of stream temperature acting within and between these influences. The genetic basis for smolt transformation is discussed; as explained earlier, the processes of smolt transformation are also influenced by temperature. The expression of genes associated with physiological traits such as metabolism and food conversion efficiency is discussed; both metabolism and food conversion efficiency are temperature-dependent. In reviewing the role of individual fish condition on O. mykiss anadromy and residency, growth, size, and lipid content are discussed; temperature influences all of these factors. Stream temperature itself is discussed as an environmental variable, along with others that are themselves not independent of temperature, such as stream flow, habitat, food supply, competitor density, and freshwater migration challenges (Figure 4, p. 329). The reviewers recommend future investigation of the effects of temperature on O. mykiss anadromy and residency; they especially single out the need to better understand the role of energy allocation (p. 337). The Report relies solely on data from its lower Tuolumne River aerobic scope experiments to assess the scope for growth for O. mykiss under a thermal regime. The experiments yielded FAS (factorial aerobic scope = MMR/RMR) values > 2 for all fish up to 23°C (p. ii, pp. 23-24). The Report explains that FAS values of 2 or greater are biologically significant because they indicate an aerobic capacity sufficient for a fish to properly digest a full stomach (p. ii; pp. 4-5, p. 20, pp. 21-22), and to potentially grow faster at a higher temperature (p. 22). However, demonstration of aerobic scope sufficient for digestion up to 23°C does not guarantee O. mykiss growth or survival in the lower Tuolumne River at this temperature. As discussed above, juvenile O. mykiss


26

experiencing longer-term thermal stress (indicated by heat-shock protein synthesis) at higher temperatures (23-25ºC) may consume less food, experience slower growth and reduced body size, and have lower lipid content (Kammerer and Heppell 2013). At lower, but sub-optimal exposure temperatures (~20ºC) heat-shock protein synthesis and metabolic indicators also signal stress in O. mykiss, but reduced growth may not occur if sufficient food is available (Werner et al. 2005). Any process to develop new Tuolumne River temperature criteria should include bioenergetics considerations, which suggest O. mykiss growth would likely decline if food availability and consumption were kept constant while lower Tuolumne River temperatures increased. Also, the bioenergetics equation discussed above does not consider additional temperature-influenced factors such as competition for food and space between cold-water O. mykiss and fishes that are more warm-adapted. For example, temperature-dependent competition between O. mykiss and Sacramento pikeminnow (Ptychocheilus grandis) has been shown to reduce the condition and abundance of juvenile O. mykiss at higher temperatures (Reese and Harvey 2002). This raises the possibility of negative competitive effects with other fishes (native and introduced) sympatric with the O. mykiss in the lower Tuolumne River.

8. Review information about the abundances and distributions of native and non-

native fishes in the lower Tuolumne River, consider the potential negative effects to O. mykiss of competition and predation, and how these effects might worsen under a warmer thermal regime.

As discussed above, the development of existing water temperature criteria for juvenile O. mykiss considered not only “physiological optimum temperatures” (where functions like growth, swimming, heart performance, etc. are optimized), but also “ecological optimum temperatures” where fish do best in the natural environment -- considering food availability, competition, predation, and fluctuating temperatures (EPA 2003, p. 18). While the distribution and abundance of fish species is partially dependent upon physiological responses, this pattern is also controlled by the interactions among competing species and predator-prey relationships (McCullough et al. 2009, p. 102). We previously discussed research documenting negative correlations between O. mykiss abundances and maximum temperatures in small creeks in the high deserts of Oregon and Idaho, and in southern California. In these systems, the negative effects of interspecific competition or predation are likely small because few fishes are present, and O. mykiss are themselves the only or top predator. In contrast, the Tuolumne River is a large river with a drainage area upstream of the La Grange Dam of ~1,550 square miles. Brown and Ford (2002) describe how these California rivers (including the Tuolumne River) with large reservoirs can maintain native species for varying distances below the dams, with nonnative warm-water species dominating their lower reaches (p. 332).


27

Surveys in the lower Tuolumne River by seining, snorkeling, and electrofishing found 33 fish taxa, 12 native to California and 21 introduced species (Ford and Brown 2001, Table 9, p. 279). The native O. mykiss, hitch, prickly sculpin, Sacramento blackfish, Sacramento splittail, and tule perch were relatively rare, never exceeding 1% of the total catch with any of the sampling methods (Table 10, p. 282); introduced species were present throughout the lower Tuolumne River, and were dominant downstream of Waterford (RM 32). Predatory fishes such as bass and catfish were present. The authors discuss their lower Tuolumne River fish survey results in the context of the potential negative thermal effects on juvenile salmonids that occur below lethal levels; they include the physiological changes that affect growth, disease resistance, predator avoidance, and smolting, as well as ecological effects such as increased predator activity or increased food requirement (without an increase in food supply). Since temperatures in the lower Tuolumne River often vary within the range at which these effects seem to occur, these effects (and their combinations) may be important influences on the survival of young salmonids (p. 311). Yearly snorkel surveys since 1982 have also consistently documented the presence of native cool water and introduced warm water fishes in the lower Tuolumne River (TID and MID 2015). After 1996, these species have been observed in lower numbers in upstream sites, which has been attributed to the higher minimum summer flow releases (p. 4). However, as recently as 2014, the numbers of Sacramento pikeminnow observed were far greater than the numbers of O. mykiss (Table 1, p. 16), and warm water fishes observed were bluegill sunfish (Lepomis macrochirus), redear sunfish (L. microlophus), green sunfish (L. cyanellus), largemouth bass (Micropterus salmoides), smallmouth bass (M. dolomieu), and spotted bass (M. punctulatus). In the Report, we found inadequate discussion of how the aerobic metabolism responses of lower Tuolumne River O. mykiss might inform their competitive or predator/prey interactions with the high number of cool- and warm-water fishes also present. Cech et al. (1990) illustrate how the routine metabolic responses of multiple fishes (measured in swim tunnel respirometers at various temperatures) might be used to evaluate their spatial overlap in rivers and streams. Any process to develop new temperature criteria should consider that colder summer conditions in the lower Tuolumne River could reduce the spatial overlap of juvenile O. mykiss with their potential competitors or predators, while a warmer thermal regime is likely to have the opposite result.

9. Consider that recent genetic studies of O. mykiss in California indicate the test fish used in the Tuolumne River aerobic performance experiments are introgressed by fish with coastal steelhead ancestry.

Pearse and Garza (2015) collected genotype data from >1,900 O. mykiss sampled at one or more locations in 15 tributary sub-basins of the Sacramento and San Joaquin rivers, including locations above and below barriers to anadromy. In the Tuolumne River, samples were collected in both the upper watershed (above the impassable La Grange and Don Pedro dams) and the lower Tuolumne River. They analyzed the genetic diversity and population structure differentiation between populations above and below dams, and the relationship of Central


28

Valley O. mykiss populations to coastal California steelhead. The introgression by both hatchery rainbow trout strains (which have primarily native Central Valley ancestry) and imported coastal steelhead stocks was also evaluated. Central Valley O. mykiss above and below dams within the same tributary were not found to be each other’s closest relatives, and the study found no relationship between genetic and geographic distance among below-barrier populations. The results suggest introgression by hatchery rainbow trout strains does not appear to be widespread among above-barrier populations. However, the clustering of the lower Tuolumne River (below-barrier) population with Nimbus Fish Hatchery and American River samples indicates the lower Tuolumne River O. mykiss have been introgressed by fish with coastal steelhead ancestry (p. 13). The authors advise that future conservation, restoration, and mitigation efforts should take these findings into account when working to meet recovery planning goals. Any process to develop new temperature criteria should be done in the light of this new genetic information, and consider that the aerobic scope experiments were performed with O. mykiss captured in the lower Tuolumne River. In the future, O. mykiss different from the existing lower Tuolumne River stock may be used in steelhead recovery efforts in the watershed.

10. Consider that water temperature is not the only factor negatively affecting the lower Tuolumne River O. mykiss and steelhead populations, but it is a very important one.

In closing, although water temperature is the factor predominantly discussed in this review, we recognize it is not the only factor capable of negatively affecting the lower Tuolumne River O. mykiss and steelhead populations. Degradation or loss of habitat unrelated to temperature (e.g., due to impassable dams, channel modifications, gravel supply interruptions, large wood supply interruptions, etc.), as well as hatchery operations and other factors have played a role. However, a protective temperature regime will be necessary to achieve the ESA recovery goals for the lower Tuolumne River steelhead population. We also recognize that policy decisions will be population-focused, and based on a weight-of-the-evidence rather than on a case for local adjustment to high river temperature that rests on a single study or finding.


29

LITERATURE CITED Anderson, C.R. 2009. Direct Testimony of Craig R. Anderson, M.S. on behalf of the National Marine Fisheries Service. Before the U.S. Federal Energy Regulatory Commission. Exhibit No. NMF-4, filed in FERC project dockets P-2299-065 and P-2299-053, September 11, 2009. Anderson M., Giannico, G., and S. Jacobs. 2011. Seasonal migrations of adult and sub-adult redband trout in a high desert basin of Eastern Oregon, USA. Ecology of Freshwater Fish 2011. Bear, E.A., McMahon, T.E., and A.V. Zale. 2007. Comparative thermal requirements of westslope cutthroat trout and rainbow trout: implications for species interactions and development of thermal protection standards. Transactions of the American Fisheries Society 136:1113–1121. Brown, L.R. 2000. Fish communities and their associations with environmental variables, lower San Joaquin River drainage, California. Environmental Biology of Fishes 57:251–269. Brown, L.R. and T. Ford. 2002. Effects of flow on the fish communities of a regulated California river: implications for managing native fishes. River Res. Applic. 18:331–342. California Department of Fish and Game (CDFG). 1993. Steelhead documentation for the Tuolumne River 1940 and 1942. Memorandum from William Loudermilk to George Neillands and Steve Baumgartner, California Dept. of Fish and Game, Region 4. Fresno, CA. February. Cech, Jr. J.J., Mitchell, S. J., Castleberry, D.T., and M. McEnroe. 1990. Distribution of California stream fishes: influence of environmental temperature and hypoxia. Environmental Biology of Fishes 29: 95-105. Clark, T.D., Sandblom, E. and F. Jutfelt. 2013a. Aerobic scope measurements of fishes in an era of climate change: respirometry, relevance and recommendations. Journal of Experimental Biology 216:2771-2782. Clark, T.D., Sandblom, E. and F. Jutfelt. 2013b. Response to Farrell and to Pörtner and Giomi. Journal of Experimental Biology 216:4495-4497. Douglas, P.L. 1995. Habitat relationships of oversummering rainbow trout (Oncorhynchus mykiss) in the Santa Ynez drainage. Master’s Thesis, University of California at Santa Barbara. Eaton, J.G., McCormick, J.H., Goodno, B.E., O'Brien, D.G., Stefan, H.G., Hondzo, M. and R. M. Scheller. 1995. A field information-based system for estimating fish temperature tolerances. Fisheries 20(4):10-18.


30

Ebersole, J.L., Liss, W.J., and C.A. Frissell. 2001. Relationship between stream temperature, thermal refugia and rainbow trout Oncorhynchus mykiss abundance in arid-land streams in the northwestern United States. Ecology of Freshwater Fish 10:1–10. Farrell, A.P., Fangue, N. A., Verhille, C.E., Dennis E. Cocherell, D.E., and K.K. English. 2015. Thermal performance of wild juvenile Oncorhynchus mykiss in the lower Tuolumne River: a case for local adjustment to high river temperature. Prepared for the Turlock Irrigation District, Turlock, California, and the Modesto Irrigation District, Modesto, California. 58 pp. Farrell, A.P. 2013. Aerobic scope and its optimum temperature: clarifying their usefulness and limitations. Journal of Experimental Biology 216:4493-4494. Federal Register Notice, 70 Fed. Reg. 52488, September 2, 2005. Endangered and Threatened Species: Designation of Critical Habitat for Seven Evolutionarily Significant Units of Pacific Salmon and Steelhead in California. Final Rule. Federal Register Notice, 71 Fed. Reg. 834, January 5, 2006. Endangered and Threatened Species: Final Listing Determinations for 10 Distinct Population Segments of West Coast Steelhead. Final Rule. Feldhaus, J.W. 2006. Heat shock proteins and physiological stress in redband rainbow trout (Oncorhynchus mykiss gairdneri) in the South Fork John Day River, Oregon. Thesis submitted to Oregon State University in partial fulfillment of the requirements for the degree of Master of Science, April 27, 2006. 127 pp. FISHBIO. 2015. Fall Migration Monitoring at the Tuolumne River Weir, 2014 Annual Report. Submitted to Turlock Irrigation District and Modesto Irrigation District. Prepared by: Chris Becker, Jason Guignard, and Andrea Fuller. FISHBIO Environmental, LLC, Oakdale, California. March 2015. FISHBIO. 2014. Fall Migration Monitoring at the Tuolumne River Weir, 2013 Annual Report. Submitted to Turlock Irrigation District and Modesto Irrigation District. Prepared by: Chris Becker, Jason Guignard, and Andrea Fuller. FISHBIO Environmental, LLC, Oakdale, California. March 2014. FISHBIO. 2013. Fall Migration Monitoring at the Tuolumne River Weir, 2012 Annual Report. Submitted to Turlock Irrigation District and Modesto Irrigation District. Prepared by: Todd Wright, Jason Guignard, and Andrea Fuller. FISHBIO Environmental, LLC, Oakdale, California. March 2013.


31

FISHBIO 2012. Fall/Winter Migration Monitoring at the Tuolumne River Weir, 2011 Annual Report. Submitted to Turlock Irrigation District and Modesto Irrigation District. Prepared by: Ryan Cuthbert, Chris Becker, and Andrea Fuller. FISHBIO Environmental, LLC, Oakdale, California. March 2012. FISHBIO 2011. Fall/Winter Migration Monitoring at the Tuolumne River Weir, 2010 Annual Report. Submitted to Turlock Irrigation District and Modesto Irrigation District. Prepared by: Chris Becker, Ryan Cuthbert, and Andrea Fuller. FISHBIO Environmental, LLC, Oakdale, California. March 2011. Ford, T. and L.R. Brown. 2001. Distribution and abundance of Chinook salmon and resident fishes of the lower Tuolumne River, California. Pages 253-303 in Contributions to the biology of Central Valley salmonids. California Department of Fish and Game, Fish Bulletin 179, Volume 2. R.L. Brown, ed. Fry, F.E.J. and J.S. Hart, 1948. The relation of temperature to oxygen consumption in the goldfish. Biol. Bull. 94:66-67. Gamperl, A.K., Rodnick, K.J., Faust, H.A., Venn, E.C., Bennett, M.T., Crawshaw, L.I., Keeley, E.R., Powell, M.S., and H.W. Li. 2002. Metabolism, swimming performance, and tissue biochemistry of high desert redband trout (Oncorhynchus mykiss ssp.): evidence for phenotypic differences in physiological function. Physiological and Biochemical Zoology 75(5):413-431. Gard, M.F. 2009. Direct Testimony of Mark F. Gard, Ph.D. on behalf of the U.S. Fish and Wildlife Service. Before the U.S. Federal Energy Regulatory Commission. Exhibit No. FWS-6, filed in FERC project dockets P-2299-065 and P-2299-053, September 11, 2009. Gordus, A.G. 2009. Direct Testimony of Andrew G. Gordus, Ph.D. on behalf of California Department of Fish and Game. Before the U.S. Federal Energy Regulatory Commission. Exhibits No. DFG-4 and DFG-7, filed in FERC project dockets P-2299-065 and P-2299-053, September 13, 2009. HDR. 2013. Initial Study Report. Don Pedro Project, FERC No. 2299. Prepared for Turlock Irrigation District and Modesto Irrigation District. Prepared by HDR Engineering, Inc., January 2013. Hokanson, K.E.F., Kleiner, C.F., and T.W. Thorslund. 1977. Effects of constant temperatures and diel temperature fluctuations on specific growth and mortality rates and yield of juvenile rainbow trout, Salmo gairdneri. J. Fish. Res. Board Can. 34:639-648. Hartman, K.J. and S.B. Brandt. 1995. Comparative energetics and the development of bioenergetics models for sympatric estuarine piscivores. Can. J. Fish. Aquat. Sci. 52:1647-1666.


32

Iwama, G.K., Thomas, P.T., Forsyth, R.B., and Vijayan, M.M. 1998. Heat shock protein expression in fish. eviews in Fish Biology and Fisheries 8:35-56. Kammerer, B.D. and S.A. Heppell. 2012. Individual condition indicators of thermal habitat quality in field populations of redband trout (Oncorhynchus mykiss gairdneri). Environ Biol Fish DOI 10.1007/s10641-012-0078-2. Kammerer, B.D. and S.A. Heppell. 2013. The effects of semichronic thermal stress on physiological indicators in steelhead. Transactions of the American Fisheries Society, 142:5, 1299-1307, DOI:10.1080/00028487.2013.806349. Kendall, N.W., McMillan, J.R., Sloat, M.R., Buehrens, T.W., Quinn, T.P., Pess, G.R., Kuzishchin, K.V., McClure, M.M., and R.W. Zabel. 2015. Anadromy and residency in steelhead and rainbow trout (Oncorhynchus mykiss): a review of the processes and patterns. Can. J. Fish. Aquat. Sci. 72:319–342. Li, H.W., Lamberti, G.A., Pearsons, T.N., Tait, C.K., Li, J.L., and J.C. Buckhouse. 1994. Cumulative effects of riparian disturbances along high desert trout streams of the John Day Basin, Oregon. Transactions of the American Fisheries Society 123(4):627-640. Lindley, S.T., Schick, R.S., Agrawal, A., Goslin, M., Pearson, T., Mora, E., Anderson, J.J., May, B., Greene, S., Hanson, C., Low, A., McEwan, D., MacFarlane, R. B., Swanson, C., and J. G. Williams. 2006. Historical population structure of Central Valley steelhead and its alteration by dams. San Francisco Estuary and Watershed Science Volume 4, Issue 1, Article 2. Lindley, S.T., R.S. Schick, E. Mora, P.B. Adams, J.J. Anderson, S. Greene, C. Hanson, B.P. May, D.R. McEwan, R.B. MacFarlane, C. Swanson, and J.G. Williams. 2007. Framework for assessing viability of threatened and endangered salmon and steelhead in the Sacramento-San Joaquin Basin. San Francisco Estuary and Watershed Science Volume 5, Issue 1 [February 2007], article 4. Lindley, S.T. 2009. Direct Testimony of Steven T. Lindley, Ph.D. on behalf of the National Marine Fisheries Service. Before the U.S. Federal Energy Regulatory Commission. Exhibit No. NMF-6, filed in FERC project dockets P-2299-065 and P-2299-053, September 11, 2009. Matthews, K.R. and N.H. Berg. 1997. Rainbow trout responses to water temperature and dissolved oxygen stress in two southern California stream pools. Journal of Fish Biology 50:50-67. McBain & Trush. 2000. Habitat restoration plan for the lower Tuolumne River corridor. Final Report. Prepared for: The Tuolumne River Technical Advisory Committee, with assistance from the U.S. Fish and Wildlife Service Anadromous Fish Restoration Program. Prepared by: McBain & Trush, Arcata, California. March 2000. 217 pp.


33

McCullough, D. A. 1999. A review and synthesis of effects of alterations to the water temperature regime on freshwater life stages of salmonids, with special reference to Chinook salmon. Prepared for the U.S. Environmental Protection Agency Region 10, Seattle, Washington. EPA 910-R-99-010. 291 pp.

McCullough, D.A., Bartholow, J.M., Jager, H.I., Beschta, R.L., Cheslak, E F., Deas, M.L., Ebersole, J.L., Foott, J.S., Johnson, S.L., Marine, K.R., Mesa, M.G., Petersen, J.H., Souchon, Y., Tiffan, K.F. and W.A. Wurtsbaugh. 2009. Research in thermal biology: burning questions for coldwater stream fishes. Reviews in Fisheries Science 17(1):90–115. Metcalfe, N. B., T. E. Van Leeuwen, and S. S. Killen. 2016. Does individual variation in metabolic phenotype predict fish behaviour and performance? Journal of Fish Biology 88:298-321. Moyle, P.B. and R.D. Nichols. 1973. Ecology of some native and introduced fishes of the Sierra Nevada foothills in central California. Copeia: 478-490. Myrick, C.A. and J.J. Cech, Jr. 2004. Temperature effects on juvenile anadromous salmonids in California’s Central Valley: what don’t we know? Reviews in Fish Biology and Fisheries 14: 113–123. Myrick, C.A. and J.J. Cech, Jr. 2005. Effects of temperature on the growth, food consumption, and thermal tolerance of age-0 Nimbus-strain steelhead. North American Journal of Aquaculture 67:324–330. Nelson, J. A. 2016. Oxygen consumption rate v. rate of energy utilization of fishes: a comparison and brief history of the two measurements. Journal of Fish Biology 88:10-25. Norin, T., Malte, H. and T.D. Clark. 2014. Aerobic scope does not predict the performance of a tropical eurythermal fish at elevated temperatures. Journal of Experimental Biology 217: 244-251. Norin, T. and T. D. Clark. 2016. Measurement and relevance of maximum metabolic rate in fishes. Journal of Fish Biology 88:122-151. Norin, T., H. Malte, and T. D. Clark. 2016. Differential plasticity of metabolic rate phenotypes in a tropical fish facing environmental change. Functional Ecology 30:369-378. Pearse, D.E., Miller, M.R., Abadia-Cardoso, A., and J.C. Garza. 2014. Rapid parallel evolution of standing variation in a single, complex, genomic region is associated with life history in steelhead/Rainbow Trout. Proceedings of the Royal Society B Biological Sciences 281, 20140012.


34

Pearse, D.E. and J.C. Garza. 2015. You can't unscramble an egg: population genetic structure of Oncorhynchus mykiss in the California Central Valley inferred from combined microsatellite and single nucleotide polymorphism data. San Francisco Estuary and Watershed Science 13(4) Article 3:1-17. Platts, W.S. and M.L. McHenry. 1988. Density and biomass of trout and char in western streams. Gen. Tech. Rep. INT-241. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station. 17 p. Poletto, J.B., Cocherell, D.E., and N.A. Fangue. 2016. Thermal performance in juvenile hatchery Oncorhynchus tshawytscha: aerobic scope tests over a range of environmental temperatures. Technical Report: The United States Environmental Protection Agency Region 9 – Pacific Southwest Region, March 15. Pörtner, H.O. and F. Giomi. 2013. Nothing in experimental biology makes sense except in the light of ecology and evolution. Journal of Experimental Biology 216:4494-4495. Reese, C.D. and B.C. Harvey. 2002. Temperature-dependent interactions between juvenile steelhead and Sacramento pikeminnow in laboratory streams. Transactions of the American Fisheries Society 131:599–606. Richter, A. and S.A. Kolmes. 2005. Maximum temperature limits for chinook, coho, and chum salmon, and steelhead trout in the Pacific Northwest. Reviews in Fisheries Science 13(1): 23-49. Rodnick, K.J., Gamperl, A.K., Lizars, K.R., Bennett, M.T., Rausch, R.N. and E.R. Keeley. 2004. Thermal tolerance and metabolic physiology among redband trout populations in south-eastern Oregon. Journal of Fish Biology 64: 310–335. Satterthwaite, W.H., Beakes, M.P., Collins, E.M., Swank, D.R., Merz, J.E., Titus, R.G., Sogard, S.M., and M. Mangel. 2010. State-dependent life history models in a changing (and regulated) environment: steelhead in the California Central Valley. Evolutionary Applications 3:221-243. Sloat, M.R. and G.H. Reeves. 2014. Individual condition, standard metabolic rate, and rearing temperature influence steelhead and rainbow trout (Oncorhynchus mykiss) life histories. Can. J. Fish. Aquat. Sci. 71:491–501. Stillwater Sciences. 2013a. Oncorhynchus mykiss population study report, W&AR-10. Don Pedro Project, FERC No. 2299. Prepared for Turlock Irrigation District and Modesto Irrigation District. Prepared by Stillwater Sciences, Inc., December 2013. Stillwater Sciences. 2013b. Salmomid population information integration and synthesis study report, W&AR-05. Don Pedro Project, FERC No. 2299. Prepared for Turlock Irrigation District and Modesto Irrigation District. Prepared by Stillwater Sciences, Inc., January 2013.


35

Turlock Irrigation District and Modesto Irrigation District (TID and MID). 2015. 2014 Snorkel Report and Summary Update, Lower Tuolumne River Annual Report 2014-5. Prepared for Turlock and Modesto Irrigation Districts by Stillwater Sciences, Berkeley, California. March 2015. U.S. Environmental Protection Agency (EPA). 2003. EPA Region 10 Guidance for Pacific Northwest State and Tribal Temperature Water Quality Standards. EPA 910-B-03-002. Region 10 Office of Water, Seattle, Washington. 57 pp. Viant, M.R., Werner, I., Rosenblum, E.S., Gantner, A.S., Tjeerdema, R.S., and M.L. Johnson. 2003. Correlation between heat-shock protein induction and reduced metabolic condition in juvenile steelhead trout (Oncorhynchus mykiss) chronically exposed to elevated temperature. Fish Physiology and Biochemistry 29:159-171. Werner, I., Smith, T.B., Feliciano, J. and M.L. Johnson. 2005. Heat shock proteins in juvenile steelhead reflect thermal conditions in the Navarro River watershed, California. Transactions of the American Fisheries Society 134(2):399-410. Winberg, G.G. 1956. Rate of metabolism and food requirements of fishes. Belorussian University, Minsk. (Transl. Fish. Res. Board Can., Transl. Ser. No. 194, 1960). Zimmerman, C.E., Edwards, G.W. and K. Perry. 2009. Maternal origin and migratory history of steelhead and rainbow trout captured in rivers of the Central Valley, California. Transactions of the American Fisheries Society 138:280–291. Zoellick, B.W. 1999. Stream temperatures and the elevational distribution of redband trout in southwestern Idaho. Great Basin Naturalist 59(2):136-143. Zoellick, B.W. 2004. Density and biomass of redband trout relative to stream shading and temperature in southwestern Idaho. Western North American Naturalist 64(1):18-26.




ATTACHMENT B

THERMAL PERFORMANCE OF WILD JUVENIILE ONCORHYNCHUS MYKISS IN THE LOWER TUOLUMNE RIVER:

A CASE FOR LOCAL ADJUSTMENT TO HIGH RIVER TEMPERATURE

THERMAL PERFORMANCE OF WILD JUVENILE ONCORHYNCHUS MYKISS IN THE

LOWER TUOLUMNE RIVER: A CASE FOR LOCAL ADJUSTMENT TO HIGH

RIVER TEMPERATURE

FINAL REPORT DON PEDRO PROJECT

Prepared for: Turlock Irrigation District – Turlock, California

Modesto Irrigation District – Modesto, California

February 2017

Thermal Performance of Wild Juvenile Oncorhynchus mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature1

February 2017

Prepared for: Turlock Irrigation District – Turlock, California Modesto Irrigation District – Modesto, California Prepared by: Anthony P. Farrell, Ph.D. Canada Research Chair in Fish Physiology, Culture and Conservation University of British Columbia Vancouver, BC, Canada Nann A. Fangue, Ph.D. Associate Professor of Fish Physiological Ecology University of California, Davis Department of Wildlife, Fish, and Conservation Biology Davis, CA Christine E. Verhille, Ph.D. Postdoctoral Research Fellow University of California, Davis Department of Wildlife, Fish, and Conservation Biology Davis, CA Dennis E. Cocherell Senior Research Associate University of California, Davis Department of Wildlife, Fish, and Conservation Biology Davis, CA Karl K. English, M.Sc. Senior Vice-President LGL Limited Sidney, BC, Canada

1 This work has been published in the peer reviewed literature as: Verhille CE, English KK, Cocherell DE, Farrell AP, Fangue NA (2016) High thermal tolerance of a rainbow trout population near its southern range limit suggests local thermal adjustment. Conserv Physiol 4(1): cow057; doi:10.1093/conphys/cow057.

Foreword

FOREWORD In July 2011, as part of the Don Pedro Hydroelectric Project (No. 2299) Federal Energy Regulatory Commission (FERC) relicensing proceeding, Turlock Irrigation Districts (TID) and Modesto Irrigation District (MID) (collectively, the Districts) proposed to study the influence of temperature on juvenile Tuolumne River Oncorhynchus mykiss, as part of a suite of investigations described in the Temperature Criteria Assessment (Chinook and Oncorhynchus mykiss) (W&AR-14) Study Plan, as provided in the Districts’ Proposed Study Plan. In its December 2011 Study Plan Determination, FERC determined that the Districts were not required to complete the Temperature Criteria Assessment (Chinook and Oncorhynchus mykiss), but indicated that empirical data collected on the thermal capability of Tuolumne River fish would be considered in the Don Pedro Project relicensing proceeding. The Districts elected to complete an investigation of the thermal performance of juvenile O. mykiss in the lower Tuolumne River, given the importance that empirical evidence on this subject would have in the relicensing proceeding. In June 2014, the Districts finalized the Local Adaptation of Temperature Tolerance of O. mykiss Juveniles in the Lower Tuolumne River Study Plan and posted the document to the Don Pedro Project relicensing website. On June 30, 2014, the Districts invited relicensing participants to attend, prior to the start of fieldwork, a site visit to observe the onsite laboratory set-up and a demonstration of the study approach and the equipment to be used. The demonstration, held on July 10, 2014, was attended by a representative from the California Department of Fish and Wildlife and members of the public. Fieldwork for the study began later that month and was completed in August. In January 2015, the Districts sent a draft study report to relicensing participants for 30-day review and comment. Comments on the draft study report were received from the Tuolumne River Trust, the California Sportfishing Protection Alliance, the State Water Resources Control Board, and the California Department of Fish and Wildlife (Appendix 5). The Districts provide responses to these comments in Appendix 6 of this report. In November 2016, this study was published in the peer reviewed journal Conservation Physiology. The journal article is appended to this report as Appendix 7.

Executive Summary

W&AR-14 ii Project Report Thermal Performance of Juvenile O. mykiss Don Pedro Project

EXECUTIVE SUMMARY The purpose of this study was to investigate the thermal performance of juvenile Oncorhynchus mykiss that populate the lower Tuolumne River in the Central Valley region of California with respect to the seasonal maximal water temperatures they experience during the summer months. The study tested the hypothesis that the Tuolumne River O. mykiss population below La Grange Diversion Dam is locally adjusted to the relatively warm thermal conditions that exist in the river during the summer. The basis for this hypothesis is peer-reviewed scientific literature that indicates that salmonid species, including O. mykiss, can adjust to local thermal conditions. In the current study, O. mykiss were locally caught and tested, and then returned safely within ~ 1 day of capture to the Tuolumne River. The experimental approach acknowledged the oxygen- and capacity-limited thermal tolerance (OCLTT) hypothesis, which proposes that the extremes of thermal tolerance are set by a fish’s inability to supply oxygen to its tissues above and beyond a basic routine need. The experimental approach also acknowledged that every activity of a fish in a river (swimming, catching prey and feeding, digesting a meal, avoiding predators, defending territory, etc.) requires oxygen above and beyond a basic routine need and that salmonids have evolved to maximize their oxygen supply when they fuel muscles during exhaustive swimming. Consequently, the tests performed here directly measured how much oxygen can be maximally extracted from the water by a fish (its maximum metabolic rate; MMR) and how much oxygen is routinely needed by that fish to exist (its routine metabolic rate; RMR). These measurements were performed in a swim tunnel respirometer (the equivalent of an aquatic treadmill) at different test temperatures ranging from 13°C to 25°C. Then, by subtracting RMR from MMR, we determined over this temperature range the capacity of O. mykiss to supply oxygen to tissues above and beyond a basic routine need, which is termed the absolute aerobic scope (AAS = MMR - RMR) and defines the fish’s capacity to perform the activities essential to complete its life history. Factorial aerobic scope (FAS = MMR/RMR) was also calculated as another way of expressing a fish’s aerobic capacity. These measurements were performed over a wide range of test temperatures (13°C to 25°C), which allowed us to determine the dependence of aerobic capacity on water temperature. These short-term direct measurements of temperature effects on fish metabolism did not assess the potentially beneficial physiological and biochemical changes that would be associated with thermal acclimation during longer-term growth studies (i.e., weeks). As expected, the routine need for oxygen of these fish (RMR) increased exponentially with test temperature from 13°C to 25°C (36 different fish each tested at a single temperature). For these same fish, MMR also increased over the same range of test temperatures, but to a lesser degree. As a result, the absolute capacity to supply oxygen to tissues above routine needs (AAS) reached a peak at 21.2°C (as modeled for all fish by a mathematical equation). Moreover, there was a wide temperature range around this optimum where AAS changed very little. For example, the statistical 95% confidence limit for peak AAS extended between 16.4°C and 25°C. Likewise, 95% of the numerical peak for AAS (i.e., 5.84 mg O2 kg-0.95 min-1) could be maintained between 17.8°C and 24.6°C. By being able to maintain peak AAS across a range of test temperatures that clearly spans the 7-Day Average of the Daily Maximum (7DADM) criterion of 18°C set out by

Executive Summary

W&AR-14 iii Project Report Thermal Performance of Juvenile O. mykiss Don Pedro Project

EPA (2003) for Pacific Northwest O. mykiss, Tuolumne River O. mykiss population has a broader range of thermal performance than previously thought. Thus, the physiological measurements presented in this report supports the hypothesis that the O. mykiss population found in the Tuolumne River downstream of La Grange Diversion Dam is locally adjusted to the relatively warm thermal conditions that typify the summer months. Indeed, all fish that were tested from 13°C to 24°C recovered quickly from an exhaustive swim test and then were successfully returned to the river. Some of these test fish were inadvertently recaptured up to 11 days later in their original river habitat and appeared to be in excellent condition when visually inspected. Also, three of the four fish tested at 25°C were successfully returned to the river after their arduous experimental tests. The upper thermal performance limit (i.e., the temperature where AAS is zero) for Tuolumne River O. mykiss could not be determined with the present experiments due to conditions set forth by the National Marine Fisheries Service (NMFS), but the present data suggest that it must lie above 25°C. The conclusion of the study is that the thermal range over which the Tuolumne River O. mykiss population can maintain 95% of their peak aerobic capacity is 17.8°C to 24.6°C. Moreover, up to a temperature of 23°C, all test fish could at least double their routine oxygen uptake (a FAS value >2.0), which we suggest is sufficient aerobic capacity for the fish to properly digest a meal. Finally, based on a video analysis of the swimming activity of O. mykiss in the Tuolumne River, fish at ambient water temperatures were predicted to have an excess aerobic capacity well beyond that needed to swim and maintain station against the river current in their usual habitat. These results support the hypothesis that the thermal performance of wild O. mykiss from the Tuolumne River represents an exception to that expected based on the 7DADM criterion set out by EPA (2003) for Pacific Northwest O. mykiss. Moreover, given that the average AAS remained within 5% of peak performance up to a temperature of 24.6°C and that all Tuolumne River O. mykiss maintained a FAS value >2.0 up to 23°C, we recommend that a conservative upper aerobic performance limit of 22°C, instead of 18°C, be considered in re-determining a 7DADM for this population. This wide range of thermal performance for O. mykiss from the Tuolumne River is consistent with that found for O. mykiss populations already known to be high-temperature tolerant, such as the redband strain of rainbow trout (O. mykiss gairdneri) in the high deserts of Eastern Oregon and Idaho, steelhead trout from the south coast of California, and selected and hatchery-maintained strains of O. mykiss in Western Australia and Japan. Whether the high thermal performance that was demonstrated for the O. mykiss of the Tuolumne River downstream of La Grange Diversion Dam arose through genetic selection or physiological acclimatization was beyond the purpose and scope of the present study.

Table of Contents

W&AR-14 iv Project Report Thermal Performance of Juvenile O. mykiss Don Pedro Project

TABLE OF CONTENTS Section No. Description Page No. FOREWORD.................................................................................................................................. i EXECUTIVE SUMMARY .......................................................................................................... ii INTRODUCTION......................................................................................................................... 1

Thermal Tolerance and Thermal Performance ......................................................................... 1

7-day Average of the Daily Maxima (7DADM) .......................................................................... 2

Current Evidence for Local Physiological Acclimatization and Genetic Selection ................ 3

Justification and Purpose of the Study ....................................................................................... 5

Predictions and Alternate Predictions ........................................................................................ 7

Predictions Derived From EPA (2003) ....................................................................................... 8

Alternative Predictions for a Thermally Adjusted Population ................................................ 8

METHODS .................................................................................................................................... 9

Permitting Restrictions that Influenced the Experimental Design .......................................... 9

Fish Collection, Transport, and Holding .................................................................................... 9

Swim Tunnel Respirometry ....................................................................................................... 10

Measuring Metabolic Rates ....................................................................................................... 11

Experimental Protocol ................................................................................................................ 12

Data Quality Control, Model Selection and Analyses ............................................................. 14

RESULTS .................................................................................................................................... 16

DISCUSSION .............................................................................................................................. 19

Data Quality ................................................................................................................................ 19

Evidence for Local Thermal Adjustment ................................................................................. 21

Ecological Relevance of the Present Findings .......................................................................... 22

CONCLUSION ........................................................................................................................... 25

ACKNOWLEDGEMENTS ....................................................................................................... 26

BIBLIOGRAPHY ....................................................................................................................... 27

FIGURES ..................................................................................................................................... 33

TABLES ....................................................................................................................................... 40

APPENDICES ............................................................................................................................. 44

Table of Contents

W&AR-14 v Project Report Thermal Performance of Juvenile O. mykiss Don Pedro Project

LIST OF FIGURES Figure 1. Schematic representation of the resting metabolic rate (= routine; RMR) and

maximum metabolic rate (MMR) and aerobic scope (AS = MMR-RMR). ........... 34 Figure 2. Representative photographic images (a-d) of the four capture locations for

Tuolumne River O. mykiss, by river mile. ............................................................. 36 Figure 3. Map of capture and recapture locations of all Tuolumne River O. mykiss test

fish .......................................................................................................................... 37 Figure 4. The relationships between test temperature and the routine (RMR) and

maximum metabolic rate (MMR) of Tuolumne River O. mykiss. Absolute aerobic scope (AAS) and factorial aerobic scope (FAS) were derived from the MR measurements .................................................................................................. 38

Figure 5. The relationship between tail beat frequency (TBF; Hz) and metabolic rate (MR; mg O2 kg-0.95 min-1) ....................................................................................... 39

LIST OF TABLES Table 1. Literature values of critical thermal maximum (CTmax) for O. mykiss

populations. ............................................................................................................ 41 Table 2. Literature values for routine metabolic rate (RMcR), maximum metabolic rate

(MMR), absolute aerobic scope (AAS) and factorial aerobic scope (FAS) of juvenile salmonid fishes. ........................................................................................ 42

LIST OF APPENDICES Appendix 1. Tuolumne River 7-Day Average of Maximum Daily Temperatures Appendix 2. Capture Release Table Appendix 3. PIT Code and Recapture Table Appendix 4. Experimental Data Table Appendix 5. Comments Received on the Draft Study Report Appendix 6. Response to Comments on the Draft Study Report Appendix 7. High Thermal Tolerance of a Rainbow Trout Population near its Southern Range

Limit Suggests Local Thermal Adjustment

List of Abbreviations

W&AR-14 vi Project Report Thermal Performance of Juvenile O. mykiss Don Pedro Project

LIST OF ABBREVIATIONS 7DADM 7-Day Average of the Daily Maximum 95% CI 95% Confidence Limits AAS Absolute Aerobic Scope (MMR-RMR) AS Aerobic Scope BP Barometric pressure CTmax Critical Thermal maximum EPA U.S. Environmental Protection Agency ESA Endangered Species Act FAS Factorial Aerobic Scope (MMR/RMR) FL Fork Length of the fish ILT Incipient Lethal Temperature M Mass of fish MMR Maximum Metabolic Rate MR Metabolic Rate O2 Oxygen O2(A) Tunnel water oxygen concentration at beginning of seal O2(B) Tunnel water oxygen concentration at end of seal OCLTT Oxygen- and Capacity-Limited Thermal Tolerance PIT Passive Integrated Transponder RM River Mile RMR Routine Metabolic Rate T Time TBF Tail Beat Frequency Tcrit Critical Temperature when performance (e.g., aerobic scope) reaches zero Topt Optimal Temperature when performance (e.g., aerobic scope) reaches a

peak Tp Pejus Temperature when performance (e.g., aerobic scope) decreases from

its peak. In the present study, Tp is set when absolute aerobic scope decreases to 95% of the peak capacity at Topt

V Tunnel volume α(O2) Solubility of oxygen in water % O2Sat Percent saturation of oxygen in water

Introduction

W&AR-14 1 Project Report Thermal Performance of Juvenile O. mykiss Don Pedro Project

INTRODUCTION The Tuolumne River has been significantly affected by human activity since the mid-1800s, including in-channel and overbank mining of gold and gravel, urban and agricultural encroachment, and water resource development. Summertime water diversions from the Tuolumne River near La Grange, CA have been occurring for over 120 years. These changes have contributed to a unique river habitat for the O. mykiss population that lives in the Tuolumne River downstream of the La Grange Diversion Dam located at River Mile 52 (RM 52). Year round, the Don Pedro Dam located near RM 54 releases cool water to the river (10-13°C) even during the hottest periods in summer. As this water flows downstream it can gain or lose thermal energy depending on its surrounding environment. In summer months, the average river temperature increases appreciably with distance downstream of the dam (see Appendix 1). At RM 49, for example, river temperature peaked at 20.2°C in July 2014. However, cooler river temperatures are associated with cloud cover and over night, and deeper ponds in the river do show some thermal stratification. In 2013, a detailed study of summertime temperatures in the Tuolumne River was performed between ca. RMs 3-37 (HDR 2014). Based on observations from monitoring surveys conducted since 1997 (Ford and Kirihara 2010; Stillwater Sciences 2012), O. mykiss rearing habitat extends from RM 52 to ca. RM 30, with spawning habitat in 2013 documented from RM 50 to about RM 39 (FISHBIO 2013). Review of this information suggests that primary rearing habitat for O. mykiss since 1997 has been concentrated upstream of RM 39.6, where peak water temperatures have occasionally exceeded 27°C during the summer months. Therefore, the realized habitat of O. mykiss during summer presently covers a distance of ca. 12.4 river miles, where water temperature varies within the range of 11°C to 28°C. Any difference between where a fish actually lives (the realized habitat) and its fundamental habitat is determined by behavior (Matthews and Berg 1997). Thermal Tolerance and Thermal Performance Fundamental habitat of any animal is determined in part by its thermal tolerance limits to warm and cold. Even humans, who normally regulate body temperature at 37°C (98.4°F), quickly succumb if body temperature cannot be maintained below 45°C in extreme heat. However, the body temperature of a fish such as O. mykiss in the Tuolumne River is not regulated in the same way as that of humans. Instead, it is always the same as the surrounding river temperature, except for brief (seconds to minutes), non-steady states whenever a fish moves rapidly between regions of thermal stratification. Nevertheless, a fish warmed or cooled beyond its thermal limits will rapidly succumb, just like a human. Scientists commonly measure the thermal tolerance limit of a fish using either incipient lethal temperature (ILT) or critical thermal maximum (CTmax) tests. An upper ILT test acutely exposes fish to a suite of elevated temperatures and reports the temperature at which 50% of the test fish succumb. In contrast, an upper CTmax test warms (ca. 0.3°C per min) a fish until it can no longer maintain its upright orientation and reports the temperature when 50% of the fish roll over. Fish can rarely live for more than a few minutes at its CTmax.

Introduction


While CTmax values have been widely used to distinguish thermal tolerance differences among fish species, CTmax does not always discriminate more subtle physiological adjustments in thermal tolerance expected within a fish species in response to season and/or genetic differences. For example, a CTmax value of 29°C is reported for trout acclimated to temperatures ranging from 12 to 20°C (Table 1). While CTmax values for O. mykiss can certainly be similar over a wide range of thermal acclimation temperatures and populations, there are exceptions because CTmax can increase in some studies of thermal acclimation of O. mykiss (Table 1), as it does when killifish are thermally acclimated (Fangue et al. 2006). The sub-species redband trout has the highest CTmax for the genus O. mykiss and red-band trout live in dessert environment. Any insensitivity of the CTmax measurement likely stems from relatively short test exposure times (min) and the rapid but sometimes variable warming rates that are employed when measuring CTmax. Regardless, CTmax is always higher than the temperature that a fish can tolerate for hours to days and certainly higher than the temperature at which a fish can no longer swim aerobically. Consequently, despite its relative ease of measurement, CTmax, which is a measure of thermal tolerance, is increasingly being replacing by fish biologists with metrics that measure thermal performance. Metrics such as growth are preferred because they have some ecological relevance but have the disadvantage of requiring 30 or more days for a fish to achieve sufficient growth to determine its optimal temperature (or range of temperatures) for growth. Also, growth studies indirectly assess the effects of temperature on fish energetics and usually require rearing fish in controlled conditions that do not account for the full range of bioenergetic functions necessary for survival in nature (e.g. foraging, migration, competition, predation avoidance). An alternative metric for performance acknowledges that all activities of a fish ultimately require oxygen (O2). Therefore, it is possible to directly assess a fish’s need for and capacity to deliver oxygen and use these measures as an ecologically relevant metric of fish performance. Furthermore, by making such measurements over a range of temperature, as first done some 60 years ago (e.g., Fry 1947), it is possible to accurately characterize the thermal effects on a fish’s ability to deliver oxygen to its tissues, which is a direct measurement of energetic capacity to support the bioenergetic functions necessary for survival in nature. Unlike growth studies that require wild fish to be removed from their natural environment into a controlled artificial environment for months, studies of oxygen uptake can be performed in days. While methods to characterize fish thermal performance using oxygen uptake have an extremely long history, watershed managers have only started to embrace these thermal performance metrics over the past decade. As a result, existing regulatory criteria tend not to have considered these metrics, which can be measured at a local scale. 7-day Average of the Daily Maxima (7DADM) One of the thermal criteria used by EPA to protect fish is the 7-day average of the daily water temperature maximum (7DADM). The explicit recommendation in EPA (2003) for juvenile O. mykiss in summer rearing habitats is a 7DADM <18°C. A key study that influenced the current 7DADM criterion for O. mykiss from the Pacific Northwest is the growth study of Hokanson et al. (1977), which was reviewed in Issue Paper 5 (EPA 2001). Growth is considered

Introduction


as a very powerful integrator of environmental, behavioral and physiological influences of a fish’s fitness. Hokanson et al. (1977) measured growth of juvenile O. mykiss from the Great Lakes in Minnesota using constant and fluctuating (a daily temperature oscillation of ± 3.8°C) thermal regimes. O. mykiss grew maximally at 16-18°C, termed the optimum temperature (Topt) for growth. However, there are a number of concerns with applying these results to O. mykiss from the Tuolumne River. Foremost, O. mykiss are not native to Minnesota; they are an introduced species. Second, the thermal and other environmental conditions in Minnesota are far from similar to those encountered by O. mykiss in the Tuolumne River (below we show clear scientific support for local thermal adaptation of fishes, including O. mykiss). Moreover, the work of Hokanson et al. (1977) pre-dated the routine statistical packages that can place a statistical 95% confidence interval (CI) around data such as growth and oxygen uptake. This is an important data gap because EPA (2003) states that: “Each salmonid life stage has an optimal temperature range (our emphasis). Physiological optimum temperatures are those where physiological functions (e.g., growth, swimming, heart performance) are optimized. These temperatures are generally determined in laboratory experiments.” Therefore, this key study established a temperature optimum for growth rather than a thermal range for peak growth performance. EPA (2003) recommends 20°C as the 7DADM criterion for salmon and trout migration. Curiously, this criterion acknowledges that Pacific Northwest O. mykiss have sufficient aerobic scope for the energetic demands of river migration at a temperature that is 2°C higher than the 7DADM for growth in juveniles (18°C). River migration can be the most energetically challenging activity a salmonid can undertake and certainly requires more energy allocation than is used for feeding and growth. A juvenile salmonid in a river or stream will hold station and use darting behavior to opportunistically capture food drifting downstream. Thus they need energy for periodic sprint and burst activities, plus the cost of digesting and assimilating the captured food (specific dynamic action or heat increment of digestion). Furthermore, Hokanson et al. (1977) discovered that “At temperatures in excess of the growth optimum, mortality rates were significantly higher during the first 20 days of this experiment than the last 30 days.” The implication of this observation is that a proportion of the test fish were either initially better suited for high temperature or became better suited after living for 20 days at a supra-optimal temperature when compared to the fish that died during the initial 20-day period. In view of this uncertainty surrounding the applicability of the 7DADM for O. mykiss to O. mykiss in the Tuolumne River, we now review some of the literature that supports the possibility for local physiological acclimation or genetic adaptation to warm temperature within the O. mykiss genus. Current Evidence for Local Physiological Acclimatization and Genetic Selection Thermal acclimation is a physiological process whereby an ectothermic animal, such as a fish, can potentially perform better after being placed in a new environment. Thermal acclimation involves a suite of physiological and biochemical changes that occur over a period of several weeks. Thus, if a fish living in say 14oC water is transferred to 20oC, its performance would progressively improve as it acclimates to the new temperature. This processes is referred to as

Introduction


thermal plasticity within a species. The extensive knowledge on thermal acclimation among fish species dates back well into the 1940s. Thermal plasticity, however, has limits that vary from species to species, which is a result of thermal adaptation within a species. As early as the late 1960s, Bidgood and Berst (1969) used upper ILT data to conclusively demonstrate that juvenile O. mykiss from four anadromous Great Lakes populations could thermally acclimate, i.e., warm acclimation increased their upper ILT. Likewise in California (CA) there is wide variation in the thermal performance curves for hatching success among different strains of O. mykiss (Myrick and Cech 2001). While this variability includes the Eagle Lake and Mt. Shasta strains, these two strains had been shown earlier to have a similar CTmax (Myrick and Cech 2000). Thus, in the early 2000s, evidence for thermal acclimation was extensive within the species O. mykiss. Evidence for thermal adaptation within the species O. mykiss was limited at the time of Issue Paper 5 (EPA 2001). Nevertheless, the work did acknowledge the possibility of genetic adaptation by asking is there enough evidence for genetic variation within a species to warrant geographically-specific or stock-specific water temperature standards. The conclusion was “The literature on genetic variation in thermal effects indicates occasionally significant but very small differences among stocks and increasing differences among subspecies, species, and families of fishes. Many differences that had been attributed in the literature to stock differences are now considered to be statistical problems in analysis, fish behavioral responses under test conditions, or allowing insufficient time for fish to shift from field conditions to test conditions”. In fact, Issue Paper 5 (EPA 2001) cited (see its Table 1) Sonski (1983), who identified the Topt for growth of redband trout (O. mykiss gairdineri) as 20°C, which is the highest value for the genus O. mykiss. Therefore, evidence did exist in the literature prior to 2001 that the genus O. mykiss can perhaps be genetically adapted to local environmental conditions. Since 2001, the peer-reviewed scientific literature has provided ample and convincing support for thermal adaptation at the population level and among a wide variety of fish species (e.g., killifish populations on the Atlantic coast, Fangue et al., 2006; stickleback populations in the Pacific Northwest, Barrett et al., 2011). Importantly, included are salmon and trout species belonging to the Oncorhynchus genus. For example, Eliason et al. (2011) showed that populations of adult sockeye salmon (O. nerka) in British Columbia’s Fraser River watershed are adjusted to perform best at the local temperature conditions that they experience during their spawning river migration. Indeed, their maximum aerobic swimming capacity is also well matched with the range of hydraulic challenges that the different populations face migrating upstream to their spawning area (Eliason et al. 2013). In addition, wild populations of redband trout, a sub-species of O. mykiss, inhabit natural desert environments in Oregon and Idaho where summer stream water temperatures can exceed 30°C. New thermal performance studies provide evidence for local thermal adaptation of redband trout (Rodnick et al. 2004) and the redband trout’s ability to genetically adapt when acclimated to a common set of experimental conditions has found support (Narum et al. 2010, 2013, Narum and Campbell 2015).

Introduction


O. mykiss is an introduced fish species on every continent except Antarctica. Moreover, selective breeding of O. mykiss has been effective in selecting for high temperature tolerance. For example, Hartman and Porto (2014) found evidence for temperature-dependent growth and differences in feeding performance among three O. mykiss strains. Also, severe thermal exposures in a hatchery program in Western Australia have produced in just over 20 generations a line of O. mykiss that is thermally tolerant (Morrissy 1973; Molony 2001; Molony et al. 2004; Chen et al. 2015). During summer extremes, the juvenile O. mykiss continue to swim and feed even when water temperature reaches 26°C (Michael Snow, Department of Fisheries, Government of Western Australia, pers. comm.). The founder O. mykiss population for this thermally tolerant line was transplanted during the last century from CA with the intention of setting up a recreational fishery for O. mykiss in Western Australia. Japanese researchers have similarly selected a strain of rainbow trout that show high thermal tolerance (Ineno et al., 2005). Therefore, clear and compelling scientific knowledge exists for local adjustments and genetic selection of high thermal performance of O. mykiss. This new knowledge has been largely added to the scientific literature subsequent to the 18°C 7DADM being identified for O. mykiss in the Pacific Northwest by the EPA (2003). EPA (2003) did acknowledge that local adjustment was possible and that well-designed studies could be used to identify site-specific thermal adjustments. The present study aims to provide such evidence for the O. mykiss population inhabiting the lower Tuolumne River. Justification and Purpose of the Study The primary purpose of this study is to determine the thermal performance of the sub-adult (100-200 mm fork length; FL) O. mykiss population that inhabits the lower Tuolumne River (LTR) to assess any local adjustment in thermal performance. Thermal performance was assessed as the range of temperatures over which juvenile O. mykiss can increase aerobic metabolic rate (MR) beyond basic needs. This aerobic capacity could be used for any of the normal daily activities of O. mykiss in the Tuolumne River during its normal life history (swimming, catching prey and feeding, digesting a meal, growing, avoiding predators, defending territory, etc.). Thus, MR measurements were used to determine the optimal temperature range for Tuolumne River O. mykiss. This experimental approach is consistent with the oxygen- and capacity-limited thermal tolerance (OCLTT) hypothesis that has emerged as a conceptual model to assess thermal performance of aquatic animals and determine the fundamental thermal range for their distributions (Pörtner and Knust 2007; Pörtner and Farrell 2008). The OCLTT hypothesis proposes that the extremes of thermal tolerance will be set by a fish’s inability to supply oxygen to its tissues above a basic routine need. The ecological relevance of the OCLTT hypothesis is exemplified through performance measurements in eelpout (Zoarcidae viviparous) and spawning Pacific salmon. The temperature at which oxygen supply to the tissues of eelpout becomes limiting closely corresponds with the temperatures where growth performance and abundance of eelpout decrease in the German Wadden Sea (Pörtner and Knust 2007). In spawning Pacific salmon, temperature ranges for upstream migration success correspond with the temperature range across which absolute aerobic scope is maximal (Eliason et al. 2011). More recently, Chen

Introduction


et al. (2015) demonstrated a broad thermal range for absolute aerobic scope in the thermally tolerant O. mykiss strain from Western Australia. Salmonids are examples of fish that have evolved to maximize oxygen supply to exhaustive swimming muscles. Therefore, our experimental approach directly measured MR under two states: routine metabolic rate (RMR), representing how much oxygen is needed by an individual O. mykiss to exist in the Tuolumne River and maximum metabolic rate (MMR), representing how much oxygen can be maximally extracted from the water for its tissues, typically when swimming. The capacity of the fish to supply oxygen to tissues above and beyond a basic routine need is then calculated by subtracting RMR from MMR, which is termed the absolute aerobic scope (AAS = MMR - RMR). Therefore, AAS defines a fish’s capacity to perform the activities essential to carry out its life functions. Factorial aerobic scope (FAS = MMR/RMR) is another way of expressing aerobic capacity by characterizing how much a fish can increase oxygen uptake beyond routine needs (RMR). A key activity for survival in nature, namely feeding and digestion, is expected to require up to a doubling of a fish’s RMR for a large meal (Jobling 1981; Alsop and Wood 1997; Fu et al. 2005; Luo and Xie 2008), i.e., an FAS value of 2 allows for proper digestion of a large meal. Measurements of fish MR were obtained using the equivalent of an aquatic treadmill (a swimming tunnel respirometer) and at different test temperatures (from 13°C to 25°C). By mathematically modeling these data, the optimal temperature (Topt) for the peak AAS could be established for juvenile O. mykiss. The Topt window (or thermal range) is defined by Parsons (2011) as “the range in temperatures where maximum aerobic scope is maintained". In the present study, we use 95% of the peak AAS value to set the optimal thermal range (Figure 1; the two temperatures that bracket Topt are termed a Pejus temperature, Tp). If, as predicted by the OCTTL hypothesis, a cardiorespiratory limitation exists for exercising salmonids during warming, AAS will decrease below 95% of peak AAS beyond the upper Tp, and often rapidly over just a few degrees before lethal temperatures are reached (Farrell 2009). The critical temperature (Tcrit) is the temperature when there is no aerobic scope and therefore aerobic activities beyond basic needs, including swimming, are impossible. Thus, whenever a fish is warmed beyond its Tp, maximum oxygen delivery progressively fails to quantitatively keep up with the need for increased oxygen delivery just to maintain the resting state (Farrell 2009). As a result, the factorial aerobic scope (AMR/RMR) decreases with temperature. Thus, an important index when considering FAS is the temperature when FAS decreases below a value of 2 because it would not be possible to double RMR for the digestion of a large meal (Jobling 1981; Alsop and Wood 1997; Fu et al. 2005; Luo and Xie 2008). Thus, the primary study goal is to determine if there is evidence for local temperature ‘adjustment’ in Tuolumne River O. mykiss by establishing the temperatures that set the thermal range for Topt (at 95% of peak) and determining how rapidly AAS declines between the upper Tp and Tcrit for Tuolumne River O. mykiss. This information should help define more accurate criteria for thermal performance of juvenile O. mykiss rearing in the lower Tuolumne River. Specifically, the temperature indices and the shape of the aerobic scope curve derived in the present study can also be compared with those of other O. mykiss populations and with the EPA (2003) recommendations.

Introduction


While the curve relating AAS with temperature has been coined a Fry aerobic scope curve (Fry 1947), curves that describe the effect of temperature on a measure of organismal performance (e.g., RMR, MMR, AAS, growth) are more generally called thermal reaction norms (Huey and Kingsolver 1979; Schulte et al. 2011). Reaction norms typically have a shape in which the performance index increases with increasing temperature, reaches a peak at some intermediate temperature, and declines with a further temperature increase. Importantly, the specific shape and position of these performance curves can vary among species and in response to thermal variation in a fish’s environment. The magnitude and timescale of environmental temperature exposure are both critical and persistent differences in local thermal conditions over evolutionary time scales may result in compensatory adaptive changes in local populations (Hochachka and Somero 2002). On a shorter time scale, and if temperature varies on a daily or seasonal basis at a given locality, fish may compensate for the temperature difference over weeks to months - termed thermal acclimatization for natural settings or simply thermal acclimation when only temperature is manipulated under controlled laboratory conditions. Fish can also respond immediately (seconds to hours) to acute thermal challenges using either behavioral (e.g., attraction and avoidance), or physiological and biochemical responses (e.g., changes in heart rate and heat shock proteins). Although the theoretical basis for how patterns of thermal performance can be shaped by local thermal regimes is now well understood and this theory provides the framework for the present study, our study was not designed to distinguish between the mechanisms of local thermal adaptation (which implies a proven genetic change) and acclimatization. Consequently, rather than using the term ‘adaptive’, we say that the fish are acclimatized to the local conditions and will use the general term that fish are ‘well adjusted’ to local environmental conditions, if we find that to be the case. However, fish were sampled from the coldest section of their habitat and their response to acute warming was examined. Therefore, our short-term, direct measurements of temperature effects on fish oxygen uptake could not assess the likely beneficial effects of thermal acclimation due to conditions for fish removal set forth by the National Marine Fisheries Service (NMFS). EPA (2003) also states that: “Ecological optimum temperatures are those where fish do best in the natural environment considering food availability, competition, predation, and fluctuating temperatures. Both (sic lab-based and field based measurements) are important considerations when establishing numeric criteria.” Importantly, Issue Paper 5 (EPA 2001) comments that “Field testing of fish survival under high temperatures is not usually done. If such methods were feasible, the improved realism would be helpful.” Therefore, the present experiments established a field laboratory beside the Tuolumne River so that the thermal performance of wild O. mykiss acclimatized to field conditions could be tested without prolonged transport and holding of fish. Predictions and Alternate Predictions Given the EPA (2003) 7DADM and the current scientific literature, it is possible to make two types of contrasting predictions for the upper thermal performance of wild O. mykiss captured from the Tuolumne River: a) predictions based on the EPA (2003) 7DADM criterion, and b) alternative predictions based on contemporary literature for local thermal adjustment.

Introduction


Predictions Derived From EPA (2003) Based on the EPA (2003) 7DADM criteria alone, one would predict that wild O. mykiss captured from the Tuolumne River for the present tests would show the following:

1. Routine metabolic rate (RMR) will increase exponentially until the test temperature approaches the upper thermal limit for O. mykiss (i.e., CTmax), which depending on the O. mykiss strain and acclimation temperature, is 26°C to 32°C (see Table 1).

2. Maximum metabolic rate (MMR) will increase with test temperature and reach a peak around 18°C according to the EPA criterion.

3. Absolute aerobic scope (AAS) has a Topt around 18°C according to the EPA criteria. 4. AAS will rapidly decline at a temperature just above 18°C. 5. Factorial aerobic scope (FAS) will decline with increasing temperature, reaching a value

< 2 (i.e., MMR is less than twice RMR) at a temperature just above 18°C. Alternative Predictions for a Thermally Adjusted Population Based on recent peer-reviewed studies, the present study tested the hypothesis that the Tuolumne River O. mykiss population below La Grange Diversion Dam is locally adjusted to the relatively warm thermal conditions that exist in the river during the summer. One would then predict that the results of the present study would show the following:

1. RMR will increase exponentially until the test temperature approaches the upper thermal limit for O. mykiss (i.e., CTmax), which is ca. 26°C to 32°C depending on the study.

2. MMR will increase with test temperature and reach a peak that is above 18°C. 3. AAS will have a Topt that is above 18°C. 4. AAS will decline at a temperature above 18°C. 5. FAS will decline with increasing temperature, but maintain a value > 2 at temperatures

above 18°C.

Methods


METHODS Permitting Restrictions that Influenced the Experimental Design Wild Tuolumne River O. mykiss were collected under National Marine Fisheries Service Section 10 permit # 17913 and California Fish and Wildlife Scientific Collecting Permit Amendments. No distinction was made between resident (rainbow trout) and anadromous (steelhead) life history forms, and both are referred to as O. mykiss throughout this document. For permitting purposes, these fish are considered as “ESA-listed California Central Valley steelhead, O. mykiss”. Fish collection (to a maximum of 50 fish) was allowed between RM 52.2 and RM 39.5, and between June 1 and September 30, 2014. Fish collections were not allowed at river water temperatures that exceeded 70°F (21.1°C). Incidental fish recaptures were authorized in addition to the initial take limit (n=50), with these reported as ‘additional take’ under the NMFS permit reporting conditions. Because indirect fish mortality was limited to 3 fish, no more than 2 fish were captured per day as a precautionary measure to limit indirect mortalities. Also, temperatures were not tested randomly and most of the highest temperatures were tested last to preclude premature termination of the work should there be high-temperature related mortality. Preliminary experiments were performed with hatchery reared O. mykiss to ensure that all the equipment was fully functional and properly calibrated prior to testing wild fish. All experimental procedures were approved by the University of California Davis’ Institutional Animal Care and Use Committee (Protocol # 18196). All fish capture and handling activities were conducted by experienced FISHBIO personnel. Fish Collection, Transport, and Holding Fish capture was conducted via seine net (0.32 cm nylon mesh, 1.8 m high, 9 m long). Several precautions were used during capture activities in order to minimize handling of non-target fish. These included 1-2 snorkelers in the water identifying O. mykiss of the target size range (100-200 mm) prior to seine sweeps, as well as the use of a mesh size allowing fish smaller than the target fork length to avoid capture. Captured fish within the target range were transferred to a partially submerged transport tank via a large scoop net to minimize handling and avoid air exposure during transfer. Each captured fish was scanned for presence of a PIT tag to ensure that the fish had not been tested previously. Upon capture, a water temperature logger (Onset Computer Corporation) was placed in the transport tank with the fish recording temperature at 15-min intervals through the duration of the fish holding/testing period. These loggers remained in the water with the fish throughout all transport, experimental protocols and handling until fish were returned to the river. In total, 48 O. mykiss were captured between July 11 and August 13, 2014 (Appendix 2). Each fish was given a unique identification (‘W’ for wild, followed by a number between 01 and 48). Two fish were captured and tested daily using four capture locations (Figure 2). The fish ID, capture location (River Mile, RM), and any recaptures are shown in Figure 3 and summarized in Appendix 3. Most of the test fish (36) were captured from a single site (RM 50.7), 8 fish were

Methods


captured at RM 51.6, 2 at RM 50.4 and 2 at RM 49.1 (Figure 3). Instantaneous water temperature and dissolved oxygen (DO) levels were recorded at the time of capture, and varied between 12.7 and 17.1°C. Temperature loggers were placed at RM 40, 42, 44, 46, and 48-50 from early June to late September, 2014. From the logged temperature data, 7DADM at each RM location was calculated and plotted in Appendix 1. Additional information about release locations, water temperatures, time of day, and general comments are summarized in (Appendix 2). Fish were placed individually into 13-l plastic transport tanks, modified with numerous 0.8 cm diameter holes drilled at least 2.0 cm from the bottom to ensure sufficient water movement through the transport container. The fish, inside its transport tank, was placed into an individual insulated Yeti cooler filled with 25 l fresh river water and driven to the experimental field site (< 20-min journey). Water temperature and DO were re-measured in the transport tanks on arrival and fish were transferred from the coolers to outdoor holding tanks (300 l) filled with flow-through Tuolumne River water between 12.5 and 13.6°C. This water-to-water fish transfer minimized handling stress and eliminated air exposure. The holding tanks received river water passed through a coarse foam filter then a 18-l gas equilibration column for aeration. This water was split between the holding tanks and the sump tank supplying the swim tunnels. Oxygen content in all vessels remained above 80% air saturation at all times. Time from fish capture in the river to placement into holding tanks ranged from 60 to 120 min. Fish remained in holding tanks for 60 to 180 min before being transfer to a swim tunnel respirometer. Swim Tunnel Respirometry Individual fish were tested in one of two, 5-l automated swim tunnel respirometers (Loligo, Denmark). As with the holding tanks, swim tunnels were supplied with Tuolumne River water but via a fine pressurized 20-µm pleated filter; then a 180-l temperature-controlled sump, which operated as a partial recirculating system; and an 18-l gas equilibration column. The sump was continuously refreshed with air-equilibrated river water, turning over the entire system every 80-90 min. Additionally, an aquarium grade air pump supplied air stones in each tunnel bath for aeration. For temperature control, water from the sump was circulated through a 9500 BTU Heat Pump (Aqua Logic Delta Star. Model DSHP-7), and returned to the sump through a high volume pump (model SHE1.7, Sweetwater®, USA), where two proportional temperature controllers (model 72, YSI, Ohio) were mated to one 800 W titanium heater each (model TH-0800, Finnex, USA), resulting in temperature control precision of ± 0.5°C across a temperature range of 12 to 26°C. To prevent buildup of ammonia waste in the water, ammonia-absorbing zeolite was kept in the system’s sump and replaced weekly. Swim tunnel water baths were refreshed with the aerated sump water approximately every 20 min. Water oxygen saturation was monitored using dipping probe mini oxygen sensors, one per tunnel, connected to AutoResp software through a 4-channel Witrox oxygen meter (Loligo). Water temperature in the swim tunnel was monitored with a temperature probe connected through the Witrox system and temperature loggers (see Fish collection, transport, and holding).

Methods


To limit disturbance of fish, swim tunnels were enclosed with black shade cloth. Above each tunnel, video cameras with infrared lighting (Q-See, QSC1352W, China) were mounted to continuously monitor and record (Panasonic HDMI DVD-R, DMR-EA18K, Japan) fish during swims and overnight routine metabolism measurements. Measuring Metabolic Rates All routine and swimming metabolic rates were measured using intermittent respirometry, a well-established technique that is the gold standard in fish biology (reviewed in Steffensen 1989; Cech 1990). A flush pump connected each tunnel chamber with an aerated external bath to allow control of tunnel sealing (during oxygen measurements) and flushing with fresh, aerated water. The pump was controlled automatically through AutoResp software and a DAQ-PAC-WF4 automated respirometry system (Loligo). When the flush pump for the swim tunnel was off, no gas or water exchange occurred within the tunnel and so the oxygen level in the tunnel water declined due to fish respiration. Therefore, the rate at which oxygen declined in the tunnel was an estimate of aerobic metabolism. Oxygen drop (in mg O2) was calculated for a minimum 2-min period when the tunnel was sealed. To restore oxygen levels in the swim tunnel, a flush pump connected to the external water bath refreshed tunnel water for periods of 2 to 5 min. Oxygen levels were never allowed to fall below 80% saturation. Swim tunnels were bleached and rinsed weekly to prevent accumulation of bacteria. At the beginning and end of the 2-month experiment, background oxygen consumption measures of both tunnels without fish were performed. No oxygen consumption for these controls was detected, even at the highest test temperature (25°C). Two-point temperature-paired calibrations at 100% and 0% oxygen saturation were performed weekly on the oxygen probes. The 100% calibration was performed in aerated distilled water. The 0% calibration was performed in 150 ml distilled water with 3 g of sodium sulfite (Na2SO) dissolved. Percent oxygen saturation was converted to oxygen concentration ([O2], mg O2 l-1) using the formula:

[O2] = % O2Sat/100 x α(O2) x BP. Where %O2Sat is the percent oxygen saturation of the water read by the oxygen probes; α(O2) is the solubility coefficient of oxygen in water at the water temperature (mg O2 l-1 mmHg-1); BP is barometric pressure in mmHg. Metabolic rate (MR in mg O2 kg-0.95 min-1) for resting and swimming fish was calculated according to the formula:

MR = {[(O2(A) – O2(B)) x V] x M-0.95} x T-1

Where O2(A) is the oxygen concentration in the tunnel at the beginning of the seal (mg O2 l-1); O2(B) is the oxygen concentration in the tunnel at the end of the seal (mg O2 l-1); V is the volume of the tunnel (l); M is the mass of the fish (kg); T is the duration of the seal (min).

Methods


To account for individual variation in body mass, MR was allometrically corrected for fish mass using the exponent 0.95. This value is halfway between the life-stage-independent exponent determined for resting (0.97) and active (0.93) zebrafish (Lucas et al. 2014). Experimental Protocol Fish were placed individually into the swim tunnels between 1300 h and 1600 h on the day of capture. Water temperature in the swim tunnels was set to 13 ± 0.3°C (i.e., close to the habitat water temperature) and fish were given a 60-min adjustment period to this temperature prior to a 60-min training swim. Each tunnel was equipped with a variable frequency drive motor designed to generate a laminar water flow through the swimming section of the tunnel (calibrated to water velocity using a digital anemometer with a 30-mm vane wheel flow probe; Hönzsch, Germany). During the training swim, water flow velocity was gradually increased until the fish moved off of the tunnel floor and began to swim (usually at ca. 30 cm s-1). Once the fish began swimming, water velocity was further increased to 5-10 cm s-1 above the initial swimming speed and held for 50 min. To complete the training swim, water velocity was increased to a maximum of 50 cm s-1 for the last 10 min, which was the expected maximum swimming velocity of 150 mm fish at 13°C (Alsop and Wood 1997). Previous studies have shown that training swim protocols result in better swimming performance in critical swimming velocity tests performed the next day (Jain et al. 1997). Fish then recovered for 60 min at 13±0.3°C before water temperature was increased to the test temperature for each pair of fish (ranging from 13 to 25°C). Water temperature was increased in increments of 1°C 30 min-1 and the time that the test temperature was reached was noted, which for the highest test temperature (25°C) took ca. 24 h. Thus, all fish in the study reached their test temperature at least 8 h before swimming tests began the following morning. Measurements of MR began 30 min after the fish reached the test temperature and continued until 0700 h. The lowest four MR measurements collected during this overnight period were averaged to estimate RMR. Critical swimming velocity tests at the test temperature began between 0800 h and 0900 h for each fish. MMR was measured in two phases: a critical swimming velocity test followed by a burst swimming test. For the critical swimming velocity test, water velocity was again gradually increased until the fish moved off of the chamber floor and began to swim. Once a fish was swimming consistently, water velocity was gradually increased to 30 cm s-1 over a 10-min period and then held at 30 cm s-1 for 20 min. If a higher initial swimming velocity was required to elicit continual swimming, the fish was held at this initial velocity for 20 min as its first test velocity. Water velocity was then increased in increments of 3 to 6 cm s-1 every 20 min until the fish failed to swim continuously. The velocity increment was set to ~10% of the previous test velocity, i.e., if the previous test velocity was between 20 to 39 cm s-1, the velocity increment was 3 cm s-1; when the previous test velocity was between 40 to 49 cm s-1, the velocity increment was 4 cm s-1. Active metabolic rate was monitored at each test velocity by closing the tunnel for either two 7-min or one 17-min measurement periods after the first 3 min of being flushed with fresh water. Water in the tunnel never dropped below 80% air saturation, which is an oxygen level expected to be considered normoxic. At the end of a measurement period, the next test velocity began with a 3-min flush period. Whenever a fish fell back in the swimming chamber

Methods


and made full body contact with the downstream screen in the tunnel, water velocity was lowered to 13 to 17 cm s-1 for 1 min, and the 20-min timer stopped. After a 1-min recovery, the test velocity was gradually restored over a 2-min period and then the 20-min timer was restarted. Failure velocity was defined when the fish fell back to the downstream screen a second time during the same test velocity. The time of this failure velocity was noted. For each test velocity, video recordings were observed for quantification of tail beat frequency (TBF measured in Hz). Three 10-s sections of video, where the fish was continuously holding station without contact with the downstream screen, bottom or side of the tunnel were identified. If three replicates were not possible throughout the entire 20-min interval, two replicates were used. If only one replicate was possible, that interval was not quantified. For each of the three (or two) sections, video was slowed to 1/4 to 1/8 of real time speed, and the number of tail beats were counted over 10 s of real time. The 2 or 3 replicates were then averaged. The same methodology was applied to video recordings taken of fish swimming in the river at temperatures of 14°C and 20°C during the study period. Approximately 50% of the wild fish did not respond as expected to the critical swimming velocity protocol, but instead used their caudal fin to prop themselves on the downstream screen to avoid swimming. This behavior was regularly observed at test velocities well above the measured maximum swimming velocity for other fish. Consequently, to estimate MMR for these fish, swimming activity was evoked by rapidly increasing water velocity to a transient velocity stimulus of 70 to 100 cm s-1 (increase over 10 s and hold for 30 s or less), then decreasing the velocity back to the test velocity. Fish tended to briefly burst swim off of the downstream screen when velocities exceeded 70 cm s-1. After the transient velocity increase, the fish was allowed to swim without interference (at the test velocity) as long as it continued to swim. For some fish, it was necessary to apply the transient velocity stimulus several times to keep the fish swimming. These fish were otherwise swum identically to fish that swam continuously; i.e., with 20-min test velocity periods and with metabolic rate measurements taken during each test velocity period. Failure for these fish was considered to occur when the fish did not swim upstream to prevent contact with the downstream screen, despite the water velocity being increased to 100 cm s-1 and returning to test velocity three times. After a critical swimming velocity trial was terminated, all fish were allowed to recover at velocities of 13-17 cm s-1 for 20 min. The subsequent burst swimming test entailed a series of metabolic rate measurements taken at higher, short-duration (30-s) water velocities. To begin the burst swimming test, the water velocity was reset to the initial critical swimming velocity test increment specific to the individual fish—i.e., the first velocity increment at which the fish swam continuously for 20 min. The burst swimming protocol involved swimming a fish at its initial critical swimming velocity test increment for up to 10 min before the water velocity was rapidly increased over ca. 10 s to the maximum speed the fish could swim without contacting the downstream screen and held for ca. 30 s (or less if the fish fell back on to the downstream screen). After the 30-s burst, the velocity was decreased back to the initial critical test velocity for ca. 30 s. This protocol was repeated multiple times for at least 5 min and up to 10 min. Metabolic rate was measured for these fish by flushing the tunnel for the first 3 min of the 10-min continuous swim, then sealing the tunnel for the remaining time. Similarly, the tunnel was flushed for no more than the first

Methods


3 min of the 10-min burst swim, and sealed for the remaining time. After completion of the burst swim protocol, fish were allowed at least 60 min of recovery at the test temperature. Following the 1-h recovery period after the swim tests, water temperature in the tunnels was lowered to ca. 13-15°C over a 30-min period. Fish were then transferred into the individual transport tanks and placed in the flow through holding tanks before measurement and tagging procedures. Fish were anaesthetized for < 5 min with CO2 (produced by dissolving 2 Alka-Seltzer tablets in 3 l river water) and without losing gill ventilatory movements. The fork length (FL, mm) and mass (g) for each fish was measured, and half duplex PIT (Oregon RFID) tags were placed into the abdominal cavity of the fish through a 1-mm incision through the body wall, just off center of the linea alba. All equipment was sterilized with NOLVASAN S prior to tagging, and wounds were sealed with 3M VetBond. Fish were returned to the transport coolers filled with 13-15°C river water to revive (observed to swim and maintain equilibrium) before being transported to the river capture site for release. At the release site, river water was gradually added to the transport cooler to equilibrate the fish to river water temperature at a rate of 1-2°C h-1 before release. Once the acclimated to the river temperature, fish were allowed to swim away volitionally. To summarize, prior to release back to the river, all fish were subjected to:

• a 1-h adjustment period in the swim tunnel at 13°C; • a 1-h training swim at 13°C that began at ca. 1600 h; • a 1-h recovery period at 13°C before the water temperature was warmed to the test

temperatures; • holding at the test temperature for at least 8 h before testing for MMR; • swimming at various activity levels for minimally 2 h and maximally 6 h until they

reached exhaustion; • a 1-h recovery period at test temperature; • decrease from test temperature to 13-15°C over 30 min; and • morphometric measurement and tagging.

Data Quality Control, Model Selection and Analyses Routine metabolic rate quality control (QC) was performed by visually inspecting over night video recordings for fish activity. Data from any fish showing consistent activity over night was discarded. Data from three fish (W7, W8, and W17) were discarded based on this criterion. RMR was calculated by averaging the lowest 4 metabolic rate measurements from 30 min after the fish reached the test temperature to 0700 the next morning. There were two methods of establishing MMR: 1) Swimming (critical swimming velocity and burst performance), and 2) Agitated behavior (i.e., random movements and struggling) in the tunnel. QC criteria for MMR involved assessment of fish behavior in the tunnel via the video, and MR response to incremental increases in tunnel speed. MMR was reported as the single highest MR measurement. The highest MRs observed in this study were concurrent with fish exhibiting intense agitation. For fish not exhibiting intense agitation, the swimming MMR was used as overall MMR. Four of these ‘non-agitated’ fish (W2, W13, W14, and W15) were

Methods


discarded due to failure of MR to increase incrementally despite continuous station-holding swimming with tunnel velocity increases of more than 15 cm s-1. Four different relationships were examined: 1) RMR versus test temperature, 2) MMR versus test temperature, 3) AAS versus test temperature, and 4) FAS versus test temperature. Model fitting was performed in R (http://cran.r-project.org) using the ‘lm’ function. Four different models were tested: linear, quadratic, antilog base 2, and log base 2 model. To select the model that best described each data set, the r2 and residuals of each model type were compared. The model with the highest r2 was chosen, except, when the r2 of different models were identical, the model with the lowest residual SE was chosen. Confidence intervals and predicted values based on the best-fit model were calculated using the ‘predict’ function, also in R.

Results


RESULTS The experimental data table, including raw RMR, MMR, AAS, and FAS data for individual fish are presented in Appendix 4.

1. Routine metabolic rate (RMR) increased exponentially over the range of test temperatures from 13°C to 25°C. This thermal response was fitted with a statistically significant (P=5.83x10-13) relationship (Figure 4A), where: RMR (mg O2 kg-0.95 min-1) = 5.9513 - 0.5787x + 0.02x2

x = temperature (°C). Thus, RMR at 13°C averaged 2.18 ± 0.45 (95% CI) mg O2 kg-0.95 min-1 and reached 5.37 ± 0.41 (95% CI) mg O2 kg-0.95 min-1 at 25°C. Consequently, the fish’s oxygen demand (cost of basic living) increased by 2.5-fold over the 12°C range for test temperature. These results for RMR are consistent with our prediction #1 derived from EPA (2003) criteria and the identical alternative prediction #1. They state that RMR should increase exponentially until the test temperature approaches the upper thermal tolerance limit for O. mykiss, which according to published CTmax values is 26°C to 32°C (see Table 1). This prediction could not be fully tested because permitting restrictions prevented test temperatures higher than 25°C, a temperature that is clearly lower than the CTmax because fish survived and even swam for several hours at 25°C.

2. Maximum metabolic rate (MMR) increased linearly with test temperature up to the maximum test temperature of 25°C. This thermal response was fitted with a statistically significant (P=8.94x10-7) relationship (Figure 4B), where: MMR (mg O2 kg-0.95 min-1) = 1.6359 + 0.3835x x = temperature (°C) Thus, MMR at 13°C averaged 6.62 ± 1.03 (95% CI) mg O2 kg-0.95 min-1 and increased up to 11.22 ± 0.86 (95% CI) mg O2 kg-0.95 min-1 at the highest test temperature (25°C). Consequently, the maximum oxygen delivery at 25°C was 1.7-times greater than that at 13°C. These results for MMR are inconsistent with our prediction #2 derived from EPA (2003) criteria where MMR was expected to peak near to 18°C. Instead, these MMR results are consistent with our alternative prediction #2 that the Tuolumne River population of O. mykiss is locally adjusted to warmer temperature, as demonstrated by peak MMR occurring at least 7°C higher than 18°C.

3. Absolute aerobic scope (AAS) was largely independent of test temperature over the range 13-25°C. Indeed, it was only at the two extremes of test temperature that any change in

Results


AAS was statistically discernable. Because of the weak dependence of AAS on test temperature, the best statistical model for these AAS data only approached statistical significance (P=0.06; Figure 4C) where: AAS (mg O2 kg-0.95 min-1) = -5.7993+1.1263x -0.0265x2 x = temperature (°C). This mathematical relationship generated a Topt at 21.2°C with a peak AAS of 6.15 ± 0.71(95% CI) mg O2 kg-0.95 min-1. These results for AAS are inconsistent with our prediction #3 based on EPA (2003) criteria, but are consistent with our alternative prediction #3 that the Tuolumne River population of O. mykiss is locally adjusted by having Topt for AAS that is greater than 18°C, i.e., 21.2°C.

4. Contrary to our prediction #4 and our alternative prediction #4, AAS did not significantly decline above the optimal temperature. In fact, the numerical change in average AAS was surprisingly small over the entire test temperature range. Thus, rather than having a well-defined peak to the AAS curve, as expected for fish with a narrow thermal range and as schematically depicted in Figure 1, the results revealed a rather flat curve more similar to one typical of a temperature generalist. Simply, O. mykiss in the lower Tuolumne River were able to maintain peak AAS over a wide range of test temperatures well above 18°C. This fact can be best illustrated by two metrics, the thermal range for the statistical 95% CI of AAS and the Topt window for 95% of the peak AAS (i.e., 5.84 mg O2 kg-0.95 min-1). The statistical 95% confidence limits for peak AAS extend from 16.4°C to 25°C. Consequently, the numerical decrease in average AAS from 6.15 ± 0.71(95% CI) mg O2 kg-0.95 min-1 at Topt to 5.78 ± 1.09(95% CI) mg O2 kg-0.95 min-1 at 25°C was only 6% and did not reach statistical significance. Indeed, the AAS measured at 24.5°C (5.89 ± 1.05 (95% CI) mg O2 kg-0.95 min-1) was numerically identical to that measured at 18°C (5.89 ± 0.80 (95% CI) mg O2 kg-0.95 min-1). But when measured at 13°C, AAS was 4.36 ± 1.21(95% CI), which was below the 95% CI for the peak AAS value. The numerical 95% peak AAS could be maintained from 17.8°C to 24.6°C, which is a more conservative thermal range for Topt.

5. Although individual variability in FAS was considerable, on average the Tuolumne River population of O. mykiss could at least double their RMR across the entire test temperature range from 13 to 25°C. On an individual fish basis, a FAS value exceeding 3.5 was achieved in individual fish tested at 13, 16, and 22°C. Factorial aerobic scope (FAS) declined with temperature. This thermal response was fitted with a statistically significant (P=2.92x10-4) relationship (Figure 4D) where FAS = 2.1438 + 0.1744x -0.0070x2

Results


x = temperature (°C). Consequently, the average FAS at 13°C was 3.32 ± 0.41 (95% CI) and decreased to 2.13 ± 0.33 (95% CI) at 25°C. This result is inconsistent with our prediction #5 derived from EPA (2003) criteria, but consistent with our alternative prediction #5 that FAS will remain above a value of 2 at temperatures well above 18°C. Indeed, all individual fish tested up to 23°C had a FAS value >2, with only 4 out of 14 fish tested at 23°C, 24°C and 25°C having a FAS value <2.

6. During swim tests at test temperatures of 14°C and 20°C, a statistically significant linear

relationship (P=2.05 x10-5 for 14°C and 0.009 for 20°C) was determined between MR and Tail Beat Frequency (TBF) (Figure 5). For fish tested at 14°C, this relationship was:

MR (mg O2 kg-0.95 min-1) = 0.75 (TBF) + 1.05

For fish tested at 20°C, this relationship was: MR (mg O2 kg-0.95 min-1) = 1.04 (TBF) + 1.89

Video analysis of fish in the lower Tuolumne River at 14°C and 20°C revealed that a fish holding station against a river current required a TBF of 2.94 and 3.40 Hz, respectively. From these TBF values, it was possible using Figure 5 to interpolate a MR associated with O. mykiss holding station in normal habitat against the Tuolumne River current. These values were 3.26 mg O2 kg-0.95 min-1 at 14°C and 5.43 mg O2 kg-0.95 min-1 and 20oC. These estimates indicate that the cost of holding station increased MR by 1.50- and 2.04-fold, respectively, and used up about half of the available FAS (67% and 49%, respectively) at these two temperatures. This meant that the remaining FAS was 2.0 at 14°C and 1.7 at 20°C.

7. After exhaustive exercise, fish quickly recovered their RMR without any visible consequences when they were inspected before being returned to the river. After a 60-min recovery period, MR either had returned to RMR, or was no more than 20% higher than RMR. There were only two exceptions to this generality. Two fish tested at 25°C regurgitated rather large meals of aquatic invertebrates during the recovery from the swim test, and one of these fish died abruptly during the recovery period. No other fish mortality occurred as a result of testing the fish. Further evidence of post-release recovery was provided by the six fish that were inadvertently recaptured 1 to 11 days after they had been tested and returned to the river (Figure 3, Appendix 3). All these fish were recaptured in their same habitat unit and within 20 m of the original capture location. All recaptured fish were visually in good condition. Three of these recaptured fish had been tested at one of the highest test temperatures, 23°C.

Discussion


DISCUSSION Data Quality This report contains the first metabolic rate data for the Tuolumne River O. mykiss population, which were used to characterize their capacity for aerobic performance over a wide test temperature range, one that extended above 18°C. The absolute values for RMR and MMR can be compared with the scientific literature even though caution is needed whenever differences exist in body mass, acclimation temperature, populations and species among the studies. As a generality, a doubling or tripling of RMR is considered a normal biological response for an acute 10°C temperature change (Schmidt-Nielsen 1994). For the Tuolumne River O. mykiss population, RMR increased by 2.5-times for a 12°C change, from 2.18 mg O2 kg-0.95 min-1 at 13°C to 5.37 mg O2 kg-0.95 min-1 at 25°C. By comparison, a study of thermally acclimated and smaller sized (5-7 g) Mount Shasta and Eagle Lake O. mykiss found that RMR was similar (2.3-2.8 mg O2 kg-0.95 min-1) at 14°C, but lower (2.9-3.1 mg O2 kg-0.95 min-1) at 25°C (Myrick and Cech 2000, Table 2). Similar RMR values are reported in a wide range of studies for juvenile salmonids (Table 2). Also, when compared with other field-based measurements, but on wild adult salmon (coho, pink and sockeye) at temperatures of 10-16°C (2.9 – 4.3 mg O2 kg min-1; Farrell et al. 2003), the RMR measured in this study for O. mykiss was again lower at these temperatures. The main methodological challenge with accurately measuring RMR in fish is eliminating spontaneous locomotory activity, which can potentially elevate MR in salmonids more so than any other activity. (Note: An overestimate of RMR reduces the AAS estimate). Therefore, considerable effort was used to select the minimum MR rate measurements to estimate RMR and to use video analysis to confirm that the fish were inactive during the MR measurement, an additional quality control measure that was introduced by Cech (1990). As a result, the variance for RMR of Tuolumne River O. mykiss was small despite the fact that the measurements were field-based. The variance was much less than that reported for a field study with adult sockeye salmon (individual RMR values varied by about 2-times) where the experimental protocol was limited to only one RMR measurement (Lee et al. 2003). As a result of this low variance, the statistical model explained 80% of the variance in RMR. Therefore, we are confident in the RMR measurements generated for this report. Normally, RMR is measured in a post-absorptive state (i.e., following a period of starvation for usually 24 h) because the digestive process is an activity that requires an increase in RMR (Jobling 1981). In the present study, however, the digestive state of the wild fish could not be controlled because the fish would take a day or longer to fully digest a meal and return to a post-absorptive state (Jobling 1981). In fact, feces were regularly found in the swim tunnels after the overnight acclimation period, which indicated that fish in the river were feeding and that the digestive process had continued for at least part of the overnight period. Therefore, although the present measurement of RMR could have been elevated by a variable contribution for digestion, our RMR values still agree with, or fall below, comparable literature values, suggesting that digestion was not a major contributor to the RMR values measured here. Nevertheless, we cannot be certain that we measured standard metabolic rate (SMR), which is more typically used

Discussion


in traditional laboratory experiments to assess AAS. SMR would be lower than RMR, which would result in an underestimate of AAS and FAS when compared with literature that used SMR for these estimates. The methodological challenge with accurately measuring MMR in wild fish is that fish vary in their willingness to participate in forced activity because they are naive to the holding conditions and to the actual swim challenge. Thus, while it is impossible to overestimate MMR and AAS, MMR and AAS can be underestimated if a fish chooses not to swim maximally. While it is possible that MMR, and therefore AAS, were underestimated in this field study, we gave the wild fish a training swim and then used four different testing protocols to generate a MMR measurement to minimize this complication. Indeed, because some of the wild Tuolumne River O. mykiss were reluctant to perform a Ucrit protocol, a burst swimming protocol was used to generate MMR. The four protocols were:

1. continuous swimming with incremental increases in velocity; 2. a combination of continuous swimming and short velocity bursts to push fish off of the

downstream screen; 3. a 10-min burst protocol of alternating 30 s of a very high velocity burst with 30 s of low

velocity burst (aimed at maintaining moderate swimming); and 4. spontaneous intense activity during RMR measurements (rarely used, but sometimes MR

was greater than the for other 3 protocols). For Tuolumne River O. mykiss, the linear regression of MMR versus temperature estimated that MMR at 13°C was 6.62 mg O2 kg-0.95 min-1 and increased to 11.22 mg O2 kg-0.95 min-1 at 25°C. The statistical model for MMR explained 50% of the individual variance for the O. mykiss tested. We are unaware of any data in the literature assessing the response of MMR to warming in juvenile O. mykiss, other than the recent study on thermally tolerant O. mykiss (~30 g) tested in Western Australia. These fish had a peak AAS of ~10 mg O2 kg-1 min-1 that was similar when tested at both 18oC and 20oC, but decreased when measured at 25oC, the only other test temperature examined above 20oC (Chen et al. 2015). These authors report that 90% of peak AAS was maintained between 13oC and 20oC. Also, the average MMR value 7.4 mg O2 kg-0.95 min-1 here at 15°C is at the high end of the range (2.9 to 8.3 mg O2 kg-0.95 min-1) reported in the literature for smaller (2-13 g) O. mykiss (Table 2). Also at 15°C, we found an average AAS of 5.1 mg O2 kg-0.95 min-1 and FAS of 3.2, both of which were on the high end of the range of reported values in the literature (1.8-5.8 mg O2 kg-0.95 min-1 and 2.2-5.8, respectively, Table 2). When compared with similar field measurements on wild adult salmon (coho, pink and sockeye) at temperatures of 10-16°C (8.6-12.6 mg O2 kg min-1; Farrell et al. 2003), the MMRs measured here overlap with the lower end of this range. The individual variation for MMR was greater than that for RMR in Tuolumne River O. mykiss, but less than the individual variation reported for MMR values in a field study of adult sockeye salmon (Lee et al. 2003). It is interesting that the variation in MMR correlated with behavior, such that the fish that displayed frequent spontaneous activity during RMR and Ucrit tests had the highest MMR within a temperature group. Fish that swam continuously throughout a Ucrit test without many extra stimuli to encourage swimming invariably had the next highest MMR. The lowest MMR was for fish that propped themselves with their caudal fin to avoid swimming despite repeated stimuli with short velocity bursts and this behavior may have resulted in an underestimate of MMR.

Discussion


Reaction norms, defined by the shape of the response curves in Figure 4, allow for proper mathematical and statistical consideration of the thermal range of performance, a concept that is fully endorsed by EPA (i.e., the 7DADM designation “recognizes the fact that salmon and trout juveniles will use waters that have a higher temperature than their optimal thermal range.”). Indeed, given the rather flat reaction norm centered around a Topt of 21.2°C shown here for the Tuolumne River O. mykiss, it is certainly more appropriate to talk about a thermal range of performance. Thus, given the good agreement with existing literature for MR measurements combined with the fact that the shape of the response curves will be independent of the methodological concerns noted above, we are confident in using these response curves to test the predictions based on EPA (2003) and our alternative predictions. Evidence for Local Thermal Adjustment Our predictions based on EPA (2003), as listed above, assumed that the Tuolumne River O. mykiss population would perform similarly to Pacific Northwest O. mykiss populations used to set the 7DADM by EPA (2003). Our alternative predictions, however, allow for the possibility of local thermal adjustment to a warmer river habitat. Collectively, the results show clear deviations from our predictions based on EPA (2003), and consistency with the alternative predictions, which suggests the likelihood that the Tuolumne River O. mykiss population is locally adjusted to warm thermal conditions. In particular, the Topt for AAS was 21.2°C, markedly higher than 18°C. Furthermore, AAS at 18°C was numerically the same as that at 24.5°C. Therefore, we discovered that the Tuolumne River O. mykiss population has a wide thermal range for optimal performance. Indeed, one fish was inadvertently recaptured in good visual condition from its original habitat location in the Tuolumne River 11 days after being tested at 23°C for 14 h and performing demanding swim tests. All the same, given that the CTmax could not be determined in the present work and that MMR increased up to the highest test temperature (25°C), it was impossible to determine the upper thermal limit when MMR collapses, which means that alternate metrics must be used to set the upper thermal limit for the Tuolumne River O. mykiss population. The present work provides three useful metrics of the optimal temperature range for the Tuolumne River O. mykiss population. Using the Topt of 21.2°C for the mathematical peak of AAS, the least conservative metric is the thermal range that is encompassed by the 95% CI for peak AAS, which set the thermal optimum range between 16.4°C and 25°C. The next metric, which was nearly as conservative as the first, is the thermal range where AAS remained numerically within 5% of the peak AAS at 21.2°C, which set the thermal optimum range between 17.8°C and 24.6°C. The small difference between these two temperature ranges is more a result of the individual variation in the data. The third and most conservative metric defines the temperature range where the FAS value for every fish tested was >2, which would set the thermal optimum range between 13 and 22°C, although the average FAS value was 2.13 at 25°C. Thus, the performance of the Tuolumne River O. mykiss population remained sufficiently elevated well beyond 18°C, which is compelling evidence of local adjustment to warm conditions. Yet, there were important indications that the thermal testing and intensive swim imposed on them outside of their normal habitat over a 24-h period taxed a small percentage of individuals at

Discussion


temperatures of 23-25°C. In the present study, the telltale signs were that 4 out of 13 individuals tested at 23-25°C had a FAS < 2. Similar to the present study, Chen et al. (2015) report that FAS was 1.4-1.8 at 25oC for the thermally tolerant O. mykiss in Western Australia. In the next section, we suggest that fish need a FAS value of about 2 for proper digestion of a meal. Interestingly, two fish regurgitated their stomach contents at 25°C, a symptom common during extreme athletic exertion in humans when metabolic rate over-taxes oxygen supply. Such individual variability in upper thermal performance is not unexpected. Indeed, Hokanson et al. (1977) reported heightened mortality only during the initial 20 days of a growth trial for O. mykiss at supra-optimal temperatures. Lastly, the only fish mortality occurred in the recovery period (a phenomenon known as ‘delayed mortality’) after one fish was tested at 25°C. Ecological Relevance of the Present Findings Establishing ecological relevance of physiological data, such as those collected for the present report, has always been a challenge because of the multiple factors that influence fish distributions, behaviors and performance in the wild. Here, we measured the aerobic capacity of the Tuolumne River O. mykiss population in a field setting to improve the ecological relevance by minimizing fish transport and handling. After a rapid recovery from our exhaustive swim and thermal tests (as seen the 60-min recovery of MR after the swimming test), test fish appeared to reestablish their original habitat in the Tuolumne River because a portion of them were inadvertently recaptured in the river within 20 m of their original capture site. This excellent recovery behavior from intense testing seemed to be independent of the test temperature because fish were recaptured after a wide range of test temperatures (16-23°C; see Appendix 3) To provide ecological relevance to physiological findings some 60 years ago, Fry (1947) introduced the concepts of a fish being metabolically loaded and metabolically limited to explain environmental effects on fishes. Simply put, a metabolic load from an environmental factor increases the oxygen cost of living (e.g., it costs energy to detoxify a poison, or, as in the present study case, a thermal increase in RMR). Conversely, a metabolic limit from an environmental factor decreases the MMR, leaving less oxygen available for activities. More broadly, the allocation of energy and tradeoffs is now a fundamental tenant of ecological physiology, especially in fishes (see review by Sokolova et al. 2012). Like all other temperature studies with fish, we found that RMR increased between 13 and 25°C, but there was nothing untoward in the magnitude of this thermal response (a 2.5-times increase in RMR over this temperature range). MMR increased with temperature from 13 to 25°C, which would mean that as fish encounter higher temperatures, they have the capacity to perform an activity at a higher absolute rate, i.e., swim faster to capture food or avoid predators, digest meals faster, detoxify chemicals faster, etc. They certainly swam harder with temperature in the present study. Thus, the Tuolumne River O. mykiss population can perform to a higher capacity level at 25°C compared with either 13°C or 18°C. The temperature that the Tuolumne River O. mykiss population is predicted to have its highest absolute capacity for aerobic activity, the Topt for AAS, was 21.2°C. FAS, which measures the capacity for a proportional increase in RMR, typically decreases with temperature in fishes (Clark et al. 2011), as was the case here. Thus, the present finding for FAS was not unexpected. Moreover, being able to maintain FAS above 2 (i.e., being able to at least

Discussion


double its RMR; FAS = 2) may relevance for two important ecological activities for fish: digesting a full stomach and maintaining station in a flowing river. Many laboratory studies with fish have examined the metabolic cost of digesting a full stomach (i.e., ad libitum feeding in a laboratory). The peak oxygen cost of digesting a meal increases with temperature and meal size, but peak MR does not increase by more than 2-fold at the temperatures used here and with a typical meal size (2% of body mass per feeding) for a salmonid in culture (e.g., Jobling 1981; Alsop and Wood 1997; Fu et al. 2005; Luo and Xie 2008). Therefore, a FAS value of 2 can be used as an index that a fish has the aerobic capacity to digest a full meal, and all individual fish achieved this performance up to 23°C. As a result of high temperature, a fish would digest the same meal with a similar overall oxygen cost but at a faster rate. This means that the fish could eat more frequently and potentially grow faster at a higher temperature with a FAS >2. Thus, the important ecological consideration is whether or not there is sufficient food in the Tuolumne River to support the highest MR associated with high temperature. All available studies suggest that the Tuolumne River population is not food limited, including direct studies of Tuolumne River Chinook salmon diet (TID/MID 1992, Appendix 16), long-term benthic macro-invertebrate sampling data collected from 1988–2008 (e.g., TID/MID 1997, Report 1996-4; TID/MID 2009, Report 2008-7), as well as the relatively high length-at-age for O. mykiss sampled in 2012 (Stillwater Sciences 2013). Indeed, the O. mykiss sampled for the current study were apparently feeding well in the river during summer months given the high condition factors (see Appendix 2), feces being regularly found in the swim tunnel and two test fish regurgitating rather large meals post-exhaustion. We do not know, however, whether a wild fish would eat meals as large as 2% of body mass, as in laboratory studies. Here, we took advantage of the video analysis of the swimming behaviors of individual O. mykiss in the Tuolumne River habitat to provide a second evaluation of the ecological relevance of MR data. This analysis revealed a common set of swimming behaviors that O. mykiss used to maintain station in the water current, as well as darting behaviors used either to protect their territory or to grab food floating down the river. Because maintaining station against a water current requires a sustained swimming activity that is functionally analogous to steady swimming at one of the velocity increments in the swim tunnel, it was possible to estimate the tail beat frequency (TBF) while performing this normal river activity. Then, using Figure 5, the TBF for station holding was compared with the TBF used while swimming in the swim tunnel to determine a MR. Thus, the estimated oxygen cost of maintaining station in the Tuolumne River by O. mykiss at 14°C was found to increase metabolism to 1.5-times RMR, leaving fish with a FAS of 2, and therefore plenty of aerobic scope for additional activities besides maintaining station. Similarly at 20°C, maintaining station increased metabolic rate to twice RMR, and the remaining FAS was 1.7. Therefore, by combining laboratory and field observations, we can conclude that the Tuolumne River O. mykiss population at 20°C have an aerobic capacity to easily maintain station in their normal river habitat and additionally nearly double their RMR for other activities, or relocating to a lower water flow area to perform other activities. According to Issue Paper 5 (EPA 2001) “Acclimation is different from adaptation. Adaptation is the evolutionary process leading to genetic changes that produce modifications in morphology,

Discussion


physiology, and so on. Acclimation is a short-term change in physiological readiness to confront daily shifts in environmental conditions. The extent of the ability to tolerate environmental conditions (e.g., water temperature extremes) is limited by evolutionary adaptations, and within these constraints is further modified by acclimation.” Here we could not evaluate the possibility that the Tuolumne River O. mykiss population can thermally acclimate to warmer river temperatures as the summer progresses, due to the restrictions on the number of fish removed from the river (a maximum of 50 individuals) and their habitat temperature. Since the instantaneous temperature in the habitat where the test fish were captured was between 12.7 and 17.1°C (see Appendix 1), the upper thermal performance determined here may have underestimated thermal performance if the Tuolumne River O. mykiss can acclimate to temperatures higher than the river temperature they were captured in. In this regard, the thermal acclimation study of Mount Shasta and Eagle Lake O. mykiss (Myrick and Cech 2000) is particularly informative. Growth rate of the Mount Shasta strain was fastest at acclimation temperatures of 19 and 22°C, temperatures that bracket the Topt for AAS determined here for Tuolumne River O. mykiss. However, growth of the Mount Shasta strain stopped at 25°C, which is consistent with our result that FAS approached a value of 2 at 25°C. In contrast, growth rate for the Eagle Lake strain was fastest at 19°C and decreased at 22°C. The Eagle Lake strain actually lost weight at 25°C, which indicated that food intake was not keeping pace with the energy requirements to sustain the RMR at this temperature, perhaps because of a limitation on AAS. Thus, the Mount Shasta strain of O. mykiss was better able to thermally acclimate to temperatures above 20°C than the Eagle Lake strain. With clear evidence that California strains of O. mykiss grow optimally at acclimation temperatures >18°C and that local differences among strains amount to as much as a 3°C shift in the optimum temperature for growth, there already existed a precedent that the thermal range for optimal performance can reach 22°C for local populations of O. mykiss. Indeed, the new data presented here adds to this evidence of local adjustments of O. mykiss to warm river habitats, because while Topt for AAS was 21.2°C, AAS remained within 5% of the peak AAS up to 24.6°C and all fish maintained a FAS value >2 up to 23°C.

Conclusion


CONCLUSION High quality field data were generated on the physiological performance of Tuolumne River O. mykiss acutely exposed to a temperature range of 13 to 25°C. These data on the RMR, MMR, AAS, and FAS were consistent with higher thermal performance in Tuolumne River O. mykiss compared to those data used to generate the 7DADM value of 18°C using Pacific Northwest O. mykiss (EPA 2003). These new data are consistent with recent peer-reviewed literature that points to local thermal adjustments among salmonid populations. Therefore, these new data provide sound evidence to establish alternative numeric criteria that would apply to the Tuolumne River O. mykiss population below La Grange Diversion Dam. Given a measured Topt for AAS of 21.2°C, and that the average AAS remained within 5% of this peak performance up to 24.6°C, and all fish maintained a FAS value >2 up to 23°C, we recommend that a conservative upper performance limit of 22°C, instead of 18°C, be used to re-determine a 7DADM value for this population.

Acknowledgements


ACKNOWLEDGEMENTS We acknowledge that fish were collected consistent with permitting of Stillwater Sciences and FISHBIO. We thank the Turlock Irrigation District for the use of facilities and the provision of laboratory accommodation, and especially Steve Boyd for his assistance with setting up the field lab and accommodations at the La Grange Diversion Dam site. We are grateful for the assistance and guidance provided throughout this study by all the individuals who participated in the weekly conference calls, especially John Devine and Jenna Borovansky who organized and directed project meetings, coordinated the weekly calls and compiled reviewer’s comments on the draft report. Trinh Nguyen and Sarah Baird, working for the Fangue Lab at UC Davis, provided invaluable help with data analysis and fieldwork support. Also thanks to Anne Todgham for sending home cooked comfort food into the field. We are especially thankful to the extremely positive, professional and capable FISHBIO team, including Jason Guignard, Andrea Fuller, Mike Kersten, Jeremy Pombo, Shane Tate, Patrick Cuthbert and Mike Phillips, for their important contributions to the field work, such as assistance in research facility set up and fish capture, tagging and release. Finally, thank you to Ron Yoshiyama for proof reading and editing advice on this report and Dawn Keller for her assistance with the final formatting of this report. Funding for this project was provided by Turlock Irrigation District, Modesto Irrigation District, and the City and County of San Francisco.

Bibliography


BIBLIOGRAPHY Alsop, D. and Wood, C. (1997). The interactive effects of feeding and exercise on oxygen

consumption, swimming performance and protein usage in juvenile rainbow trout (Oncorhynchus mykiss). J. Exp. Biol. 100, 2337–2346.

Barrett, R. D. H., Paccard, A., Healy, T. M., Bergek, S., Schulte, P. M., Schluter, D., and

Rogers, S. M. (2011). Rapid evolution of cold tolerance in stickleback. Proceedings. Biol. Sci. / R. Soc. 278, 233–238.

Becker, C. D. and Wolford, M. G. (1980). Thermal resistance of juvenile salmonids sublethally

exposed to nickel, determined by the critical thermal maximum method. Environ. Pollut. 21, 181–189.

Bidgood, B. F. and Berst, A. H. (1969). Lethal temperatures for Great Lakes rainbow trout.

J. Fish. Res. Bd. Can. 26, 456–459. Brett, J. R. (1964). The respiratory metabolism and swimming performance of young sockeye

salmon. J. Fish. Res. Bd. Can., 21(5), 1183–1226. Brett, J. R. and Glass, N. R. (1973). Metabolic rates and critical swimming speeds of sockeye

salmon (Oncorhynchus nerka) in relation to size and temperature. J. Fish. Res. Bd. Can., 30, 379–387.

Carline, R. F. and Machung, J. F. (2001). Critical thermal maxima of wild and domestic strains

of trout. Trans. Am. Fish. Soc. 130, 1211–1216. Cech, J. J. Jr. (1990). Respirometry. In: Schreck, C.B., Moyle, P.B. (eds). Methods of fish

biology. American Fisheries Society, Bethesda, Maryland. pp. 335–362. Chen, J. Q., Snow, M., Lawrence, C. S., Church, A. R., Narum, S. R., Devlin, R. H. and Farrell,

A.P.(2015). Selection for upper thermal tolerance in rainbow trout (Oncorhynchus mykiss Walbaum). J. Exp. Biol. 218, 803-812.

Clark, T. D., Jeffries, K. M., Hinch, S. G., & Farrell, A. P. (2011). Exceptional aerobic scope and

cardiovascular performance of pink salmon (Oncorhynchus gorbuscha) may underlie resilience in a warming climate. The Journal of Experimental Biology, 214(Pt 18), 3074–81. doi:10.1242/jeb.060517

Currie, R. J., Bennett, W. A., and Beitinger, T. L. (1998). Critical thermal minima and maxima

of three freshwater game-fish species acclimated to constant temperatures. Environ. Biol. Fishes. 51, 187–200.

Eliason, E. J., Clark, T. D., Hague, M. J., Hanson, L. M., Gallagher, Z. S., Jeffries, K. M.,

Gale, M. K., Patterson, D. A., Hinch, S. G., and Farrell, A. P. (2011). Differences in thermal tolerance among sockeye salmon populations. Science. 332, 109–112.

Bibliography


Eliason, E. J., Clark, T. D., Hinch, S. G., and Farrell, A. P. (2013). 521 Cardiorespiratory

collapse at high temperature in swimming adult sockeye salmon. Conserv. Physiol. 1, 1−19.

Fangue, N. A., Hofmeister, M., and Schulte, P. M. (2006). Intraspecific variation in thermal

tolerance and heat shock protein gene expression in common killifish, Fundulus heteroclitus. J. Exp. Biol. 209, 2859–2872.

Farrell, A. P. (2009). Environment, antecedents and climate change: lessons from the study of

temperature physiology and river migration of salmonids. J. Exp. Biol. 212, 3771–80. Farrell, A.P., Lee, C.G., Tierney, K., Hodaly, A., Clutterham, S., Healey, M., Hinch, S. & Lotto,

A. (2003). Field-based measurements of oxygen uptake and swimming performance with adult Pacific salmon using a mobile respirometer swim tunnel. Journal of Fish Biology. 62, 64-84. doi:10.1046/j.0022-1112.2003.00010.x

FISHBIO (2013). Salmonid Redd Mapping Study Report. Prepared for the Turlock Irrigation

District and the Modesto Irrigation District by FISHBIO, Oakdale, CA Ford, T. and Kirihara, S. (2010). Tuolumne River Oncorhynchus mykiss monitoring report.

Prepared by Turlock Irrigation District/Modesto Irrigation District, California and Stillwater Sciences, Berkeley, California for Federal Energy Regulatory Commission, Washington, D.C. January.

Fry, F. E. J. (1947). Effects of the environment on animal activity. Publ. Ontario Fish. Res. Lab.

55, 1–62. Fu, S. J., Xie, X. J., and Cao, Z. D. (2005). Effect of meal size on postprandial metabolic

response in southern catfish (Silurus meridionalis). Comp. Biochem. Physiol., A. 140, 445–451.

Galbreath, P. F., Adams, N. D., Sherrill, L. W., and Martin, T. H. (2006). Thermal tolerance of

diploid versus triploid rainbow trout and brook trout assessed by time to chronic lethal maximum. Environ. Biol. Fishes. 75, 183–193.

Gamperl, A. K., Rodnick, K. J., Faust, H. A., Venn, E. C., Bennett, M. T., Crawshaw, L. I., …

and Li, H. W. (2002). Metabolism, swimming performance , and tissue biochemistry of high desert redband trout (Oncorhynchus mykiss ssp.): Evidence for phenotypic differences in physiological function. Physiol. Biochem. Zoo. 75(5), 413–431.

Hartman, K. J. and Porto, M. A. (2014). Thermal performance of three rainbow trout strains at

above-optimal temperatures. Trans. Am. Fish. Soc. 143(6), 1445-1454. HDR Engineering, Inc. (2014). In-River Diurnal Temperature Variation. 81 pp.

Bibliography


Hochachka, P. W. and Somero, G. N. (2002). Biochemical adaptation: Mechanism and process in physiological evolution. Oxford University Press. 480 pp.

Hokanson, K. E. F., Kleiner, C. F., and Thorslund, T. W. (1977). Effects of constant

temperatures and diel temperature fluctuations on specific growth and mortality rates and yield of juvenile rainbow trout, Salmo gairdneri. J. Fish. Res. Bd. Can. 34, 639–648.

Huey, R. B. and Kingsolver, J. G. (1979). Integrating thermal physiology and ecology of

ectotherms: A discussion of approaches. Am. Zool. 19, 357–366. Ineno, T., Tsuchida, S., Kanda, M., and Watabe, S. (2005). Thermal tolerance of a rainbow trout

Oncorhynchus mykiss strain selected by high-temperature breeding. Fish. Sci. 71, 767−775.

Jain, K. E., Hamilton, J. C., and Farrell, A. P. (1997). Use of a ramp velocity test to measure

critical swimming speed in rainbow trout (Oncorhynchus mykiss). Comp. Biochem. Physiol. 117A, 441–444.

Jobling, M. (1981). The influences of feeding on the metabolic rate of fishes: a short review.

J. Fish Biol. 18, 385–400. LeBlanc, S., Middleton, S., Gilmour, K. M., and Currie, S. (2011). Chronic social stress impairs

thermal tolerance in the rainbow trout (Oncorhynchus mykiss). J. Exp. Biol. 214, 1721−31.

Lee, R. M. and Rinne, J. N. (1980). Critical thermal maxima of five trout species in the

southwestern United States. Trans. Am. Fish. Soc. 109, 632–635. Lee, C. G., Farrell, A. P., Lotto, A., Hinch, S. G., Healey, M. C., & MacNutt, M. J. (2003).

Excess post-exercise oxygen consumption in adult sockeye (Oncorhynchus nerka) and coho (O. kisutch) salmon following critical speed swimming. The Journal of Experimental Biology, 206(18), 3239–3251.

Lucas, J., Schouman, A., Plyphout, L., Cousin, X., and LeFrancois, C. (2014).Allometric

relationship between body mass and aerobic metabolism in zebrafish Danio rerio.J. Fish Bio. 84, 1171–1178. April 2014. doi:10.1111/jfb.12306

Luo, Y. P. and Xie, X. J. (2008) Effects of temperature on the specific dynamic action of the

southern catfish, Silurus meridionalis. Comp. Biochem. Physiol. A. 149, 150–156 Matthews, K. R. and Berg, N. H. (1997). Rainbow trout responses to water temperature and

dissolved oxygen stress in two southern California stream pools. J. Fish Biol. 50, 50–67. McGeer, J. C., Szebedinsky, C., McDonald, D. G., and Wood, C. M. (2000). Effects of chronic

sublethal exposure to waterborne Cu, Cd or Zn in rainbow trout. I: Iono-regulatory

Bibliography


disturbance and metabolic costs. Aquatic Tox. 50(3), 231–243. doi:10.1016/S0166-445X(99)00105-8

Molony, B. (2001). Environmental requirements and tolerances of rainbow trout (Oncorhynchus

mykiss) and brown trout (Salmo trutta) with special reference to Western Australia: A review. Fisheries Research Reports No. 130. Western Australia. 28 pp.

Molony, B. W., Church, A. R., and Maguire, G. B. (2004). A comparison of the heat tolerance

and growth of a selected and non-selected line of rainbow trout, Oncorhynchus mykiss, in Western Australia. Aquaculture 241, 655–665.

Morrissy, N. (1973). Comparison of strains of Salmo gairdneri Richardson from New South

Wales, Victoria and Western Australia. Aust. Soc. Limnol. Bull. 5, 11–20. Myrick, C. A. and Cech, J. J. Jr. (2000). Temperature influences on California rainbow trout

physiological performance. Fish Physiol. Biochem. 22, 245–254. Myrick, C. A. and Cech, J. J. Jr. (2001). Temperature Effects on Chinook Salmon and Steelhead:

a Review Focusing on California’s Central Valley Populations. Bay-Delta Modeling Forum, Technical Publication 01-1.

Myrick, C. A. and Cech, J. J. Jr. (2005). Effects of temperature on the growth, food consumption,

and thermal tolerance of age-0 Nimbus-strain steelhead. North Am. J. Aquaculture 67, 324–330.

Narum, S. R., Campbell, N. R., Kozfkay, C. C., and Meyer, K. A. (2010). Adaptation of redband

trout in desert and montane environments. Mol. Ecol. 19, 4622–4637. Narum, S. R., Campbell, N. R., Meyer, K. A., Miller, M. R., and Hardy, R. W. (2013). Thermal

adaptation and acclimation of ectotherms from differing aquatic climates. Mol. Ecol. 22, 3090–3097.

Narum, S. R. and Campbell, N. R. (2015). Transcriptomic response to heat stress among

ecologically divergent populations of redband trout. BMC Genomics. 16(103). DOI 10.1186/s12864-015-1246-5

Parsons, E. 2011. Cardiorespiratory physiology and temperature tolerance among populations of

sockeye salmon (Oncorhynchus nerka). PhD Thesis. University of British Columbia, Canada.

Pörtner, H. O. and Knust, R. (2007). Climate change affects marine fishes through the oxygen

limitation of thermal tolerance. Science 315, 95–97. Pörtner, H. O. and Farrell, A. P. (2008). Physiology and climate change. Science 322, 690–692.

Bibliography


Rasmussen, J. B., Robinson, M. D., Hontela, A., and Heath, D. D. (2012). Metabolic traits of westslope cutthroat trout, introduced rainbow trout and their hybrids in an ecotonal hybrid zone along an elevation gradient. Biol. J. Linn. Soc., 105, 56–72. doi:10.5061/dryad.rr388.functional

Recsetar, M. S., Zeigler, M. P., Ward, D. L., Bonar, S. A., and Caldwell, C. A. (2012).

Relationship between fish size and upper thermal tolerance. Trans. Am. Fish. Soc. 141, 1433–1438.

Rodnick, K. J., Gamperl, A. K., Lizars, K. R., Bennett, M. T., Rausch, R. N., and Keeley, E. R.

(2004). Thermal tolerance and metabolic physiology among redband trout populations in south-eastern Oregon. J. Fish Biol. 64, 310–335.

Scarabello, M., Heigenhauser, G. J. F., & Wood, C. M. (1991). The oxygen debt hypothesis in

juvenile rainbow trout after exhaustive exercise. Respiration Physiology, 84(2), 245–259. Scarabello, M., Heigenhauser, G. J. F., and Wood, C. M. (1992). Gas exchange, metabolite status

and excess postexercise oxygen consumption after repetitive bouts of exhaustive exercise in juvenile rainbow trout. J. Exp. Biol. 167, 155–169.

Scott, M. A. (2012). Performance of wild and domestic strains of diploid and triploid rainbow

trout (Oncorhynchus mykiss ) in response to environmental challenges. Master thesis. University of British Columbia. 69 pp.

Schulte, P.M., Healy, T.M., and N.A. Fangue (2011). Thermal performance curves, phenotypic

plasticity, and the time scales of temperature exposure. Integ. Comp. Bio. 51, 691−702. Schmidt-Nielsen, K. (1994). Animal physiology: adaptation and environment. 4th ed. Cambridge

University Press. Sloat, M. R. and Reeves, G. H. (2014). Individual condition, standard metabolic rate, and rearing

temperature influence steelhead and rainbow trout (Oncorhynchus mykiss) life histories. Can. J. Fish. Aqu. Sci., 71, 491–501.

Sokolova, I. M., Frederich, M., Bagwe, R., Lannig, G., and Sukhotin, A. A. (2012). Energy

homeostatsis as an integrative tool for assessing limits of environmental stress tolerance in aquatic invertebrates. Mar. Env. Res. 79, 1–15.

Sonski, A.J. 1983. Heat tolerance of redband trout. Annu. Proc. Tex . Chap. Am. Fish. Soc.

5, 66−76. Steffensen, J. 1989. Some errors in respirometry of aquatic breathers: How to avoid and correct

for them. Fish. Physiol. Biochem. 6, 49-59.

Bibliography


Stillwater Sciences (2012). Tuolumne River 2011 Oncorhynchus mykiss monitoring summary report. Prepared by Stillwater Sciences, Berkeley, California for the Turlock Irrigation District and Modesto Irrigation District. January.

Stillwater Sciences (2013). Oncorhynchus mykiss Scale Collection and Age Determination Study

Report. Prepared by Stillwater Sciences, Berkeley, California for the Turlock Irrigation District and Modesto Irrigation District. January.

Strange, R. J., Petrie, R. B., and Cech, J. J. Jr. (1993). Slight stress does not lower critical

thermal maximums in hatchery-reared rainbow-trout. FOLIA Zool. 42, 251–256. Turlock Irrigation District and Modesto Irrigation District (TID/MID). (1992). Report of Turlock

Irrigation District and Modesto Irrigation District Pursuant to Article 39 of the License for the Don Pedro Project. Turlock, California. 8 Volumes. April.

TID/MID. (1997). 1996 Report of Turlock Irrigation District and Modesto Irrigation District

Pursuant to Article 58 of the License for the Don Pedro Project, No. 2299. Turlock, California. 6 Volumes. March.

TID/MID. (2009). 2008 Report of Turlock Irrigation District and Modesto Irrigation District

Pursuant to Article 58 of the License for the Don Pedro Project, No. 2299. Turlock, California. March.

U.S. Environmental Protection Agency (EPA). 2001. Issue Paper 5, Summary of Technical

Literature Examining the Physiological Effects of Temperature on Salmonids. Available online at: http://yosemite.epa.gov/R10/WATER.NSF/6cb1a1df2c49e4968825688200712cb7/5eb9e547ee9e111f88256a03005bd665/$FILE/Paper%205-Literature%20Temp.pdf

U.S. Environmental Protection Agency (EPA). 2003. EPA Region 10 Guidance for Pacific

Northwest State and Tribal Temperature Water Quality Standards. Available online at: http://www.epa.gov/region10/pdf/water/final_temperature_guidance_2003.pdf

Van Leeuwen, T. E., Rosenfeld, J. S., and Richards, J. G. (2011). Adaptive trade-offs in juvenile

salmonid metabolism associated with habitat partitioning between coho salmon and steelhead trout in coastal streams. J. Anim. Ecol. 80(5), 1012–23. doi:10.1111/j.1365-2656.2011.01841.x

Van Leeuwen, T. E., Rosenfeld, J. S., and Richards, J. G. (2012). Effects of food ration on SMR:

influence of food consumption on individual variation in metabolic rate in juvenile coho salmon (Onchorhynchus kisutch). J. Anim. Ecol. 81(2), 395–402. doi:10.1111/j.1365-2656.2011.01924.x

FIGURES

Figures


Figure 1. Schematic representation of the resting metabolic rate (= routine; RMR) and

maximum metabolic rate (MMR) and aerobic scope (AS = MMR-RMR) for a temperature specialist. See text for details. Topt = optimum temperature, Tp = pejus temperatures which set the thermal window or range in which 95% of the peak value for AS can be maintained; Tcrit = critical temperatures where there is no aerobic scope. (Parsons 2011).

Figures


(a) RM 51.6

(b) RM 50.7

Figures


(c) RM 50.4

(d) RM 49.1 Figure 2. Representative photographic images (a-d) of the four capture locations for Tuolumne

River O. mykiss, by river mile.

Figures


Figure 3. Map of capture and recapture locations of all Tuolumne River O. mykiss test fish.

Figures


Figure 4. The relationships between test temperature and the routine (RMR) and maximum metabolic rate (MMR) of Tuolumne River O. mykiss. Absolute aerobic scope (AAS) and factorial aerobic scope (FAS) were derived from the MR measurements. Each data point represents an individual fish tested at one temperature. These data were given a best-fit mathematical model (solid line or curve) and the 95% confidence intervals for this line are indicated by the broken lines.

Figures


Figure 5. The relationship between tail beat frequency (TBF; Hz) and metabolic rate (MR; mg O2 kg-0.95 min-1) measured when

Tuolumne River O. mykiss were swimming continuously in a swim tunnel at 14 °C or 20°C. The solid black line represents the linear regression based on the data for N=7 fish at 14°C and N=5 fish at 20oC). The blue dashed lines represent the estimated TBF (2.94 Hz at 14°C and 3.40 Hz at 20°C (bottom graph) taken from videos of O. mykiss maintaining station in a water current in their normal Tuolumne River habitat.

TABLES

Tables


Table 1. Literature values of critical thermal maximum (CTmax) for O. mykiss populations.

Acclimation CTmax Heating rate Mass Length Strain Reference Temperature

(°C) (°C) (°C min-1) (g) (cm) Source

8 26.9 ± 0.12 0.1 11 – 18 Washington Becker and Wolford 1980

9.8 27.9 ± 0.05 0.3 15,3 ± 0.25 Pennsylvania Carline and Machung 2001

10 28.5 ± 0.28 0.02 Lee and Rinne 1980

10 28.0 ± 0.12 0.3 ~15 ~10 Missouri Currie et al. 1998

10 27.7 ±0.08 0.3 12.9 ± 0.6 California Myrick and Cech 2000

11 27.5 0.3 8.0 ± 1.6 California Myrick and Cech 2005

11 * 29.0 ± 0.05 0.3 2.4 ± 0.5 British Columbia Scott 2012

13 27.9 ±0.14 0.33 21.8±0.4 Ontario Leblanc et al. 2011

14 28.5±0.11 0.3 13.8 ± 0.8 California Myrick and Cech 2000

14 29.4 ±0.1 0.03% 41 - 140 Oregon Rodnick et al. 2004

15 29.4 ± 0.08 0.3 Strange et al. 1993

15 29.1 ± 0.09 0.3 ~15 ~10 Missouri Currie et al. 1998

15 27.7 ±0.03 0.0014 # 89.9 ± 5.4 11.9 – 0.3 North Carolina Galbreath et al. 2006


15 ~29.65 0.083 & Miyazaki, Japan Ineno et al. 2005

15 29.0 ± 0.02 0.3/0.1 30.2 ± 0.3 13.0 ± 0.4 Western Australia Present study

18 31.2 0.3 4.1 – 20 Arizona Recsetar et al. 2012


19 29.9 ± 0.17 0.3 11.8 ± 0.7 California Myrick and Cech 2000

20 29.35 ± 0.19 0.02 Lee and Rinne 1980

20 29.8 ±0.12 0.3 ~2 ~4 Missouri Currie et al. 1998

20 ~30.4 0.083 & Miyazaki, Japan Ineno et al. 2005

20 31.14±0.03 0.3 10.8±0.1 Hatchery (British Columbia) Hartman and Porto 2014

20 31.20±0.03 0.3 11.9±0.1 Hatchery (Virginia) Hartman and Porto 2014

20 31.29±0.02 0.3 9.5±0.1 Hatchery (Maryland) Hartman and Porto 2014

22 30.9 ± 0.13 0.3 9.29 - 0.99 California Myrick and Cech 2000

25 31.75 ± 0.1 0.3 6.1 - 0.63 California Myrick and Cech 2000

*fish held at 10 ~12°C. & temperature was increased at 5°C h-1. % temperature was increased at 2°C h-1. # temperature was increased at 2°C day-1.

Tables


Table 2. Literature values for routine metabolic rate (RMcR), maximum metabolic rate (MMR), absolute aerobic scope (AAS) and factorial aerobic scope (FAS) of juvenile salmonid fishes.

Species Source1 (test location)

Mass (g) Temperature (°C) Metabolic rates (mg O2 kg-0.95 min-1)

FAS Reference

Acclimate Test RMR MMR AAS

Rainbow trout Hatchery (L) 13 15 15 0.5 2.9 2.4 5.8 Alsop and Wood 1997 Hatchery (L) 6 15 15 1 2.8 1.8 2.8 Alsop and Wood 1997 Hatchery (L) 2-3 15 15 3.9 8.7 4.8 2.2 Scarabello et al. 1991 Hatchery (L) 6 15 15 2.5 8.3 5.8 3.3 Scarabello et al. 1992 Hatchery (L) 18 17 17 3.9 7 3.1 1.8 McGeer et al. 2000 Eagle Lake Wild2 (L) 6.9 10 10 2.6 Myrick and Cech 2000 Eagle Lake Wild2 (L) 7.2 14 14 2.8 Myrick and Cech 2000 Eagle Lake Wild2 (L) 14.1 19 19 2.6 Myrick and Cech 2000 Eagle Lake Wild2 (L) 13.4 22 22 2.9 Myrick and Cech 2000 Eagle Lake Wild2 (L) 5 25 25 3.1 Myrick and Cech 2000 Mt. Shasta Wild2 (L) 10 10 10 2 Myrick and Cech 2000 Mt. Shasta Wild2 (L) 7.5 14 14 2.3 Myrick and Cech 2000 Mt. Shasta Wild2 (L) 24.5 19 19 2.2 Myrick and Cech 2000 Mt. Shasta Wild2 (L) 15 22 22 2.4 Myrick and Cech 2000 Mt. Shasta Wild2 (L) 5.4 25 25 2.9 Myrick and Cech 2000 Steelhead trout Wild (H/F) 1.7 8.3 8.3 1.8-3.4 5.7-9.1 Van Leeuwen et al. 2011 Hatchery (H/F) 3.3 12.3 12.3 1.9-3.6 5.5-9.7 Van Leeuwen et al. 2011 Rainbow trout

Wild2 (L) (territorial) 60-80 13 0.6-1.9 Sloat and Reeves 2014 Wild2 (L) (dispersing) 60-80 13 0.6-1.5 Sloat and Reeves 2014

Rainbow cutthroat hybrid

Hatchery (F) 20-70 9.5-11 9.5-11 2.3 Rasmussen et al. 2012

Cutthroat trout West slope Wild (F) 20-100 9.5-11 9.5-11 2.6 Rasmussen et al. 2012 Redband trout Wild Bridge Creek (F) 92 (150-200 mm) 12-24* 13 1.8 8.5 6.7 4.7 Gamperl et al. 2002 Wild Bridge Creek (F) 108 (150-200 mm) 12-24* 24 4.5 14 9.5 3.1 Gamperl et al. 2002 Wild Little Blitzen River (F) 58 12-18* 13 2.4 12 9.6 5.0 Gamperl et al. 2002 Wild Little Blitzen River (F) 71 12-18* 24 5.6 14 8.4 2.5 Gamperl et al. 2002

Tables


Species Source1 (test location)

Mass (g) Temperature (°C) Metabolic rates (mg O2 kg-0.95 min-1)

FAS Reference

Acclimate Test RMR MMR AAS

Wild 12 Mile Creek (F) 56 19-30 (23.4)* 24 4.7 18.3 13.6 3.9 Rodnick et al. 2004 Wild Rock Creek (F) 50 12-27 (18.7)* 24 4.7 18 13.3 3.8 Rodnick et al. 2004

Wild Bridge Creek (F) 63 13-21 (17)* 24 4.6 15.6 11 3.4 Rodnick et al. 2004

Sockeye salmon Wild (L) 37 (170 mm) 5 5 0.9 7.6 6.7 8.4 Brett 1964 Wild (L) 33(160 mm) 10 10 1.4 8.7 7.3 6.2 Brett 1964 Wild (L) 55 (190 mm) 15 15 1.7 14.2 12.5 8.4 Brett 1964 Wild (L) 63 (190 mm) 20 20 2.1 13.1 11 6.2 Brett 1964 Wild (L) 52 (180 mm) 24 24 0.8 12.7 11.9 15.9 Brett 1964 Wild (L) 20-60 5.3 5.3 0.5 6.9 6.4 13.8 Brett and Glass 1973 Wild (L) 19-60 15 15 0.9 9.9 9 11.0 Brett and Glass 1973 Wild (L) 20-50 20 20 1.7 12.5 10.8 7.4 Brett and Glass 1973 Coho salmon Wild (H/F) 3.9 8.3 8.3 1.5-3.1 3.6-6.2 Van Leeuwen et al. 2011 Hatchery (H/F) 5.4 12.3 12.3 1.1-2.3 3.8-6.5 Van Leeuwen et al. 2011 Wild2 (F) 40-100 9.5-11 9.5-11 3.2 Rasmussen et al. 2012 Wild (L) 4.3 14 14 1.6 Van Leeuwen et al. 2012

Redband trout Wild 12 Mile Creek (F) 94 19-30 (23)* 14 1.6 Rodnick et al. 2004 Wild 12 Mile Creek (F) 94 19-30 (23)* 24 2.3 Rodnick et al. 2004 Wild 12 Mile Creek (F) 94 19-30 (23)* 26 4.8 Rodnick et al. 2004 Wild 12 Mile Creek (F) 94 19-30 (23)* 28 5.6 Rodnick et al. 2004 Wild Rock Creek (F) 54 12-27 (19)* 14 1.8 Rodnick et al. 2004 Wild Rock Creek (F) 54 12-27 (19)* 24 3.7 Rodnick et al. 2004 Wild Rock Creek (F) 54 12-27 (19)* 26 5.7 Rodnick et al. 2004 Wild Rock Creek (F) 54 12-27 (19)* 28 6.1 Rodnick et al. 2004 Wild Bridge Creek (F) 79 13-21 (17)* 14 2.3 Rodnick et al. 2004 Wild Bridge Creek (F) 79 13-21 (17)* 24 4.2 Rodnick et al. 2004 Wild Bridge Creek (F) 79 13-21 (17)* 26 5.6 Rodnick et al. 2004 Wild Bridge Creek (F) 79 13-21 (17)* 28 6.7 Rodnick et al. 2004

1 L = laboratory; H = hatchery; F=Field. 2 Spawned in a hatchery. *Acclimations to cycled temperature regime of range indicated, and average in brackets if reported.

APPENDICES

THERMAL PERFORMANCE OF WILD JUVENILE ONCORHYNCHUS MYKISS IN THE LOWER TUOLUMNE RIVER: A CASE FOR LOCAL

ADJUSTMENT TO HIGH RIVER TEMPERATURE

APPENDIX 1

TUOLUMNE RIVER 7-DAY AVERAGE OF MAXIMUM DAILY TEMPERATURES

Appendices


Appendix 1. Tuolumne River 7-day average of maximum daily temperatures (7DADM) from June 1 to September 30, 2014. Thermograph data provided by TID (Patrick Maloney).

10

12

14

16

18

20

22

24

26

28

6/1/14 7/1/14 8/1/14 9/1/14RM 52 RM 51 RM 50 RM 49 RM 48

RM 46 RM 44 RM 42 RM 40



APPENDIX 2

CAPTURE RELEASE TABLE

Appendices


Appendix 2. Capture release table. Fish capture and release locations and physical conditions.

Fish ID

Capture Release Habitat Unit ID

(Stillwater 2010) Est. RM Comments

Coordinates Date Time Temp (°C) Coordinates Date

Time Temp (°C)

W01 N - 37.66574 W - 120.44421 7/13 9:45 12.9 N - 37.66574

W - 120.44421 7/14 15:35 14.4 4 FW Riffle (side channel #3) 51.5

W02 N - 37.66574 W - 120.44421 7/13 11:24 13.2 N - 37.66574


W03 N - 37.66532 W - 120.44482 7/14 11:04 13.5 N - 37.66518


W04 N - 37.66538 W - 120.44470 7/14 11:08 13.5 N - 37.66518


W05 N - 37.66524 W - 120.44424 7/15 9:50 12.8 N - 37.66544


W06 N - 37.66536 W - 120.44474 7/15 10:53 12.9 N - 37.66544

W - 120.44449 7/16 12:00 13.4 4 FW Riffle (side channel #3) 51.5 Fish not measured or PIT tagged to

limit handling

W07 N - 37.66544 W - 120.44449 7/16 9:52 12.9 N - 37.66510

W - 120.44515 7/17 13:16 14 4 FW Riffle (side channel #3) 51.5

W08 N - 37.66544 W - 120.44449 7/16 10:10 12.7 N - 37.66510

W - 120.44515 7/17 13:16 14 4 FW Riffle (side channel #3) 51.5

W09 N - 37.66586 W - 120.45826 7/17 9:10 13.5 N - 37.66581

W - 120.45829 7/18 14:36 16 11 FW Riffle 50.7

W10 N - 37.66586 W - 120.45826 7/17 9:24 13.5 N - 37.66581

W - 120.45829 7/18 14:36 16 11 FW Riffle 50.7

W11 N - 37.66581 W - 120.45829 7/18 8:40 13.7 N - 37.66581

W - 120.45829 7/19 14:49 15.5 11 FW Riffle 50.7

W12 N - 37.66581 W - 120.45829 7/18 8:40 13.7 N - 37.66581

W - 120.45829 7/19 14:49 15.5 11 FW Riffle 50.7

W13 N - 37.66579 W - 120.45832 7/20 8:48 13.4 N - 37.66585

W - 120.45823 7/21 13:59 15.3 11 FW Riffle 50.7

W14 N - 37.66579 W - 120.45832 7/20 8:48 13.4 N - 37.66585

W - 120.45823 7/21 13:59 15.3 11 FW Riffle 50.7

W15 N - 37.66585 W - 120.45823 7/21 8:35 13.3 N - 37.66579

W - 120.45834 7/22 13:47 15.0 11 FW Riffle 50.7 7/21- recaptured a PIT tagged fish #114779, 114769, or 114734

Appendices


Fish ID




Time Temp (°C)

W16 N - 37.66585 W - 120.45823 7/21 8:35 13.3 N - 37.66579

W - 120.45834 7/22 13:47 15.0 11 FW Riffle 50.7

W17 N - 37.66579 W - 120.45834 7/22 10:23 13.6 N - 37.66579

W - 120.45839 7/23 14:13 15.4 11 FW Riffle 50.7

W18 N - 37.66579 W - 120.45834 7/22 10:28 13.6 N - 37.66579

W - 120.45839 7/23 14:13 15.4 11 FW Riffle 50.7

W19 N - 37.66579 W - 120.45834 7/23 10:10 13.5 N - 37.66574

W - 120.45786 7/24 14:29 15.3 11 FW Riffle 50.7

W20 N - 37.66579 W - 120.45834 7/23 10:27 13.5 N - 37.66574

W - 120.45786 7/24 14:29 15.3 11 FW Riffle 50.7

W21 N - 37.66579 W - 120.45828 7/24 9:00 13.5 N - 37.66571

W - 120.45794 7/25 14:00 15.4 11 FW Riffle 50.7 7/24- recaptured PIT tag #114752

W22 N - 37.66579 W - 120.45828 7/24 9:00 13.5 N - 37.66571

W - 120.45794 7/25 14:00 15.4 11 FW Riffle 50.7

W23 N - 37.66582 W - 120.45830 7/25 9:05 13.5 N - 37.66582

W - 120.45830 7/26 13:33 15.1 11 FW Riffle 50.7

W24 N - 37.66582 W - 120.45830 7/25 9:05 13.6 N - 37.66582

W - 120.45830 7/26 13:33 15.1 11 FW Riffle 50.7

W25 N - 37.66565 W - 120.45826 7/27 8:15 13.6 N - 37.66565

W - 120.45826 7/28 14:15 14.5 11 FW Riffle 50.7

W26 N - 37.66565 W - 120.45826 7/27 8:15 13.6 N - 37.66565

W - 120.45826 7/28 14:15 14.5 11 FW Riffle 50.7

W27 N - 37.66565 W - 120.45826 7/28 9:15 13 N - 37.66565

W - 120.45826 7/29 14:15 14.9 11 FW Riffle 50.7

W28 N - 37.66565 W - 120.45826 7/28 9:15 13 N - 37.66565

W - 120.45826 7/29 14:15 14.9 11 FW Riffle 50.7

W29 N - 37.66565 W - 120.45826 7/29 9:30 13.3 N - 37.66574


W30 N - 37.66565 W - 120.45826 7/29 9:18 13.3 N - 37.66574

W - 120.45788 7/30 16:30 14.7 11 FW Riffle 50.7

W31 N - 37.66565 W - 120.45826 7/30 9:00 13.3 N - 37.66565


W32 N - 37.66565 W - 120.45826 7/30 9:07 13.3 N - 37.66565

W - 120.45826 7/31 13:38 15.1 11 FW Riffle 50.7

Appendices


Fish ID




Time Temp (°C)

W33 N - 37.66565 W - 120.45826 7/31 9:05 13.1 N - 37.66565

W - 120.45826 8/1 13:42 15.0 11 FW Riffle 50.7

W34 N - 37.66565 W - 120.45826 7/31 9:05 13.1 N - 37.66565

W - 120.45826 8/1 13:42 15.0 11 FW Riffle 50.7

W35 N - 37.66565 W - 120.45826 8/1 9:02 13.2 N - 37.66565

W - 120.45826 8/2 15:40 15.8 11 FW Riffle 50.7

W36 N - 37.66565 W - 120.45826 8/1 9:30 13.2 N - 37.66565

W - 120.45826 8/2 15:40 15.8 11 FW Riffle 50.7

W37 N - 37.66565 W - 120.45826 8/6 9:18 13.4 -- -- -- -- 11 FW Riffle 50.7 Mortality- due to chloride residue in

tunnel

W38 N - 37.66565 W - 120.45826 8/6 9:28 13.4 -- -- -- -- 11 FW Riffle 50.7 Mortality- due to chloride residue in

tunnel

W39 N - 37.66565 W - 120.45826 8/7 9:08 13.5 N - 37.66668

W - 120.46420 8/8 17:31 15.8 11 FW Riffle 50.7

W40 N - 37.66565 W - 120.45826 8/7 9:30 13.5 N - 37.66668

W - 120.46420 8/8 17:31 15.8 11 FW Riffle 50.7

W41 N - 37.66643 W - 120.46432 8/8 11:18 15.5 N - 37.66643

W - 120.46432 8/9 16:00 16.7 14 BC Riffle 50.4

W42 N - 37.66643 W - 120.46432 8/8 11:35 14.6 N - 37.66643

W - 120.46432 8/9 16:00 16.7 14 BC Riffle 50.4

W43 N - 37.66426 W - 120.48132 8/9 11:40 17.1 N - 37.66308

W - 120.48160 8/10 15:13 18.0 25 BC Riffle 49.1 Mortality- post- swim test transport

W44 N - 37.66426 W - 120.48132 8/9 11:40 17.1 N - 37.66308

W - 120.48160 8/10 15:13 18.0 25 BC Riffle 49.1

W45 N - 37.66565 W - 120.45826 8/13 10:25 14.4 N - 37.66565

W - 120.45826 8/14 14:10 15.2 11 FW Riffle 50.7 Fish not PIT tagged to limit handling

after study termination per NMFS Section 10 permit conditions

W46 N - 37.66565 W - 120.45826 8/13 10:59 13.9 N - 37.66565

W - 120.45826 8/14 14:10 15.2 11 FW Riffle 50.7 Fish not PIT tagged to limit handling after study termination per NMFS

Section 10 permit conditions

W47 N - 37.66565 W - 120.45826 8/14 9:08 13.6 N - 37.66565

W - 120.45826 8/14 14:10 15.2 11 FW Riffle 50.7 Fish released w/o testing per NMFS Section 10 permit conditions

W48 N - 37.66565 W - 120.45826 8/14 9:15 13.6 N - 37.66565

W - 120.45826 8/14 14:10 15.2 11 FW Riffle 50.7 Fish released w/o testing per NMFS Section 10 permit conditions

Appendices



APPENDIX 3

PIT CODE AND RECAPTURE TABLE

Appendices


Appendix 3. PIT code and recapture table. Only five out of seven recapture fish are included in this table because PIT IDs were not recorded for two of the recaptured fish. See Figure 3 for details on the two unidentified recaptured fish, and recapture location for all recaptured fish. Days post-release is the number of days after release the PIT was recaptured.

Fish ID PIT

Test Temp (˚C)

PIT recap freq

Days post

release

W01 114756 13

W02 114745 13

W03 114743 13

W04 114720 13

W05 114764 15

W06 -- 15

W07 114755 19

W08 114807 19

W09 114779 21

W10 114773 21

W11 114769 23

W12 114734 23 1 11

W13 114750 17

W14 114759 17

W15 114741 14

W16 114766 14

Fish ID PIT

Test Temp (˚C)

PIT recap freq

Days post

release W17 114752 16 1 1

W18 114808 16

W19 114803 20

W20 114723 20

W21 114786 22

W22 114730 22

W23 114809 18 1 3

W24 114714 18

W25 114787 23

W26 114725 23

W27 526260 17

W28 526292 17

W29 526299 24

W30 526275 24

W31 526297 19

W32 526212 19

Fish ID PIT

Test Temp (˚C)

PIT recap freq

Days post

release W33 526226 13

W34 526211 13

W35 526285 25

W36 526263 25

W37 -- --

W38 -- --

W39 526255 23 1 5

W40 526298 23 1 5

W41 526227 24

W42 526235 24

W43 526284 25

W44 526252 25

W45 -- 19

W46 -- 19

W47 -- --

W48 -- --

Appendices



APPENDIX 4

EXPERIMENTAL DATA TABLE

Appendices


Appendix 4. Experimental data table. RMR: routine metabolic rate; MMR: maximum metabolic rate; AAS: absolute aerobic scope; FAS: factorial aerobic scope; K: condition factor (mass x 105 / FL3).

Fish ID

Test Temp (°C)

RMR (mg O2

kg-0.95min-1)

MMR (mg O2

kg-0.95min-1)

AAS (mg O2

kg-0.95min-1) FAS FL

(mm) Mass

(g) K Body Depth (mm)

Body Width (mm)

Quality Control

W01 13 1.97 7.46 5.49 3.78 112 15.7 1.12 21 9

W02 13 2.25 110 13.3 1.00 19 9 DISCARD; tunnel leak confirmed

W03 13 1.85 6.40 4.55 3.46 102 10.9 1.03 19.5 11

W04 13 1.75 7.12 5.37 4.08 102 10.6 1.00 19.5 14

W05 15 2.80 7.05 4.24 2.51 113 13.4 0.93 22.0 11

W06 15 2.37 5.98 3.61 2.52 ~160 ~29.2 0.87

W07 19 9.79 126 21.4 1.05 22.5 12 DISCARD; activity during RMR


W09 21 3.96 11.19 7.23 2.82 125 20.2 1.03 24.0 12

W10 21 2.86 8.66 5.80 3.03 197 79.6 1.04 36.0 20

W11 23 3.94 10.99 7.05 2.79 132 24.3 1.06 21.0 12

W12 23 3.88 8.73 4.85 2.25 131 25.1 1.12 24.0 13

W13 17 1.89 141 29.4 1.05 26.0 14 DISCARD; no MR increase with velocity 33 to 53 cms-1



W16 14 2.53 5.61 3.08 2.22 137 28.4 1.10 24.0 12


W18 16 2.31 8.26 5.95 3.58 133 25.9 1.10 25.0 10

W19 20 3.75 9.95 6.19 2.65 147 38.4 1.21 28.0 11

W20 20 3.66 10.83 7.16 2.96 134 28.1 1.17 25.0 11

W21 22 3.09 11.15 8.06 3.61 124 21.7 1.14 21.0 10

Appendices


Fish ID

Test Temp (°C)

RMR (mg O2

kg-0.95min-1)

MMR (mg O2

kg-0.95min-1)

AAS (mg O2


(mm) Mass


Body Width (mm)

Quality Control

W22 22 2.89 9.73 6.84 3.37 115 15.8 1.04 19.0 8

W23 18 2.73 9.35 6.62 3.42 164 47.1 1.07 30.0 18

W24 18 2.81 6.97 4.16 2.48 133 22.6 0.96 21.0 13

W25 23 4.11 11.23 7.12 2.73 121 18.7 1.06 20.0 11

W26 23 3.90 9.43 5.53 2.42 129 23.4 1.09 23.0 12

W27 17 2.76 9.57 6.81 3.47 134 24.9 1.03 21.0 13

W28 17 2.87 8.69 5.81 3.02 122 19.9 1.10 24.0 12

W29 24 5.31 13.41 8.10 2.52 104 13.0 1.16 18.0 10

W30 24 5.26 9.17 3.91 1.74 115 16.5 1.08 19.0 12

W31 19 2.81 8.07 5.26 2.87 138 29.0 1.10 24.0 10

W32 19 3.21 6.71 3.51 2.09 140 27.2 0.99 28.0 11

W33 13 2.17 6.97 4.80 3.21 117 16.4 1.02 19.0 8

W34 13 2.02 6.40 4.38 3.17 105 12.2 1.05 19.0 7

W35 25 4.87 10.09 5.21 2.07 130 27.4 1.25 26.0 10

W36 25 7.01 13.12 6.11 1.87 111 12.4 0.91 17.0 7

W37 Mortality- due to chloride residue in tunnel

W38 Mortality- due to chloride residue in tunnel

W39 23 3.76 7.11 3.36 1.89 101 12 1.02 17.0 6

W40 23 4.76 14.41 9.65 3.03 122 18.5 1.16 20.0 10

W41 24 4.87 10.04 5.17 2.06 131 23.1 1.03 22.0 12

W42 24 3.94 10.04 6.10 2.55 138 25.5 0.97 22.0 12

W43 25 5.54 9.03 3.49 1.63 107 14.5 1.18 19.0 8

W44 25 6.13 12.61 6.48 2.06 113 14.9 1.03 19.0 8

W45 19 3.49 11.76 8.27 3.37 ~101 ~11.5 1.12 ~16 ~10

W46 19 3.51 7.59 4.08 2.16 ~108 ~13.1 1.04 ~17 ~10

Appendices


Fish ID

Test Temp (°C)

RMR (mg O2

kg-0.95min-1)

MMR (mg O2

kg-0.95min-1)

AAS (mg O2


(mm) Mass


Body Width (mm)

Quality Control

W47 Fish released w/o testing per NMFS Section 10 permit conditions

W48 Fish released w/o testing per NMFS Section 10 permit conditions

Appendices



APPENDIX 5

COMMENTS RECEIVED ON THE DRAFT STUDY REPORT

1

March 2, 2015 Ms. Rose Staples HDR, Inc. [email protected] Re: Comments on January 31, 2015 draft of Thermal Performance of Wild Juvenile Oncorhynchus Mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature. Dear Ms. Staples,

The California Sportfishing Protection Alliance (CSPA) and the Tuolumne River Trust (TRT) submit the following comments on the January 31, 2015 draft of Thermal Performance of Wild Juvenile Oncorhynchus Mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature (“Study”).

Overview

Based on our review of the Study and some of the background material cited in the

Study, including the EPA (2003) Region 10 Guidance For Pacific Northwest State and Tribal Temperature Water Quality Standard that the Study in significant part seeks to address, it appears to us that the Study proposes to recommend to regulators a change in the established EPA (2003) temperature benchmark for a 7DADM value for the population of O. mykiss in the lower Tuolumne River based on site-specific evidence.

The EPA (2003) guidelines recognize that site-specific thermal criteria for salmonids

may be developed that are more appropriate for specific locations and populations than are the general criteria promulgated in the guidelines. Evaluation of physiological response in a target population is an appropriate approach to development of site-specific conditions. We accept the premise of the Study that site-specific physiological study of the response of fish to water temperature may demonstrate that such response in a specific population is different than broader, more general and geographically unspecific studies of the response of fish to water temperature have shown.

Neither CSPA nor the Tuolumne River Trust has fisheries physiologists on staff, and

neither has the resources to hire a consulting fisheries physiologist at this time. We therefore

2

have no comment at this time on the experimental approach adopted within the Study, the value of the metrics adopted, or the execution of the Study. We may bring in an outside consultant at a later point in the ILP process to evaluate these and other technical aspects of the Study.

Instead, we confine our comments to the implicit and explicit argument that Study results

can “be used to determine a 7DADM value for this population.” (Study Conclusion, p. 24). The Study does not evaluate the physiological response of the population of O. mykiss in the lower Tuolumne River over time.

There are limitations to the Study that the Study does not acknowledge. Chief among

these limitations is that the Study does not evaluate physiological response of the population of O. mykiss in the lower Tuolumne River over time. On the contrary, 75% of the test fish were sourced from a location one mile downstream of La Grange Powerhouse, where temperatures at capture ranged from 12.7° C to 17.1° C. While the Study is critical of Hokanson (1977) for an issue concerning confidence intervals, the Study does not address Hokanson’s use of a 40-day period to evaluate physiological response. Other studies (e.g. Brett 1956; Bidgood 1969) similarly address long-term exposure to less-than-optimal thermal conditions. The Study does not acknowledge this limitation. It is akin to trying to determine the best overall athletic performance in a decathlon based on performance in the sprint alone.

Thermal conditions in the summer in most of the lower Tuolumne River are much more

comparable to a marathon than a sprint. In the absence of adequate flow, grinding ambient temperatures with daily highs greater than 90° F for four months, and greater than 100° F on multiple days, create long-term water temperatures that are stressful to juvenile and adult O. mykiss. A City of San Francisco biologist has acknowledged on the record in this proceeding that O. mykiss populations in the lower Tuolumne River are substantially smaller than populations downstream of rim dams in the Sacramento River drainage, where water temperatures are generally much lower than temperatures in the lower Tuolumne River.1 A change in the 7DADM value for the population of O. mykiss in the lower Tuolumne River is not warranted based on the evidence presented. The document should therefore be re-cast as a study, rather than walking what appears to us to be a gray line between a study and a position paper that advocates a departure from established guidance.

Before any adjustment to the established (EPA 2003) temperature benchmark for a

7DADM value for the population of O. mykiss in the lower Tuolumne River is considered based on site-specific conditions and response, further investigation and evaluation would be required. The Study should explicitly state this, and should describe additional evidence needed before any change in the 7DADM value for the population O. mykiss in the lower Tuolumne River might appropriately be evaluated.

1 See Dr. Ronald Yoshiyama, “Commentary on Evaluating the Temperature-Related Flow Requirements of Steelhead-Rainbow Trout (Oncorhynchus Mykiss) in the Lower Tuolumne River: A Literature Review and Synthesis,” eLibrary no. 20120807-5082 (July 5, 2012), p. 2: “The actual numbers of adult and juvenile trout in the lower Tuolumne River were not accurately known until recently. Routine fish monitoring by the Districts indicates relatively low numbers of trout have been present over the past 1-2 decades--i.e., far below the numbers occurring in the Sacramento River mainstem and tributaries.”

3

We discuss additional limitations of the Study and additional evidentiary needs below.

The Study results alone do not warrant site-specific summer water temperature criteria for O. mykiss in the lower Tuolumne River.

The Study is careful in its language not to state outright that its results alone can be used to develop alternative summer temperature criteria for the lower Tuolumne River. The Executive Summary states:

Moreover, given that the average AAS remained within 5% of peak performance up to a temperature of 24.6°C and that all Tuolumne River O. mykiss maintained a FAS value >2.0 up to 23°C, we recommend that a conservative upper performance limit of 22°C, instead of 18°C, be used to determine a 7-Day Average of the Daily Maximum (7DADM) value. (Study, p. ii, emphasis added).

The Conclusion states in greater context: High quality field data were generated on the physiological performance of Tuolumne River O. mykiss acutely exposed to a temperature range of 13 to 25°C. These data on the RMR, MMR, AAS, and FAS were consistent with higher thermal performance in Tuolumne River O. mykiss compared to that used to generate the 7DADM value of 18°C using Pacific northwest O. mykiss (EPA 2003). These new data are consistent with recent peer-reviewed literature that points to local thermal adjustments among salmonid populations. Therefore, these data provide sound evidence to establish alternative numeric criteria that would apply to the Tuolumne River O. mykiss population below La Grange Diversion Dam. Given a measured Topt for AAS of 21.2°C, and that the average AAS remained within 5% of this peak performance up to 24.6°C, and all fish maintained a FAS value >2 up to 23°C, we recommend that a conservative upper performance limit of 22°C, instead of 18°C, be used to determine a 7DADM value for this population. (Study, p. 24, emphasis added) The use of the passive voice (“be used to determine”) is at once imprecise as to the nature

and context of such use and imprecise as to who will or should use it. In our view, the appropriate use of the Study results would be to 1) evaluate their limitations; 2) develop additional investigations that might be necessary to scientifically justify consideration of adjusting thermal criteria for the population of O. mykiss in the lower Tuolumne River, 3) enumerate and evaluate regulatory and policy issues that might be involved in adjusting these criteria; and 4) assemble these necessary components and, based on this ensemble, develop a process for considering and evaluating site-specific water temperature criteria.

However, the Study provides no such context and proposes no such process. While the

Study does not explicitly say that its results alone can be used to develop alternative summer temperature criteria for the lower Tuolumne River, the Districts have already used the results of the Study to advocate that temperatures greater than those of the EPA (2003) criteria be considered appropriate to determine amount of usable habitat in the lower Tuolumne. The draft

4

Lower Tuolumne River Instream Flow Study—Evaluation of effective usable habitat area for over-summering O. mykiss distributed by the Districts’ consultants to relicensing participants on February 27, 2015 adopts a higher range of suitable temperatures for over-summering O. mykiss based on the present Thermal Performance Study:

Although the majority of historical (1996–2009) snorkel survey observations of O. mykiss in the lower Tuolumne River have occurred at temperatures of 20°C (68°F) or below (Ford and Kirihara 2010), O. mykiss have been routinely observed occupying Tuolumne River habitats at temperatures ranging from 11–25°C (52–77°C). Using wild juvenile O. mykiss collected from the Tuolumne River in the summer of 2014, a recently completed thermal performance study (Farrell et al. 2014) found a peak in the absolute aerobic scope (AAS) vs. temperature curve at 21.2°C (70°F), higher than the 19°C (66°F) growth rate optimum identified by Myrick and Cech (2001). Because Farrell et al. (2014) also found that the AAS of the wild O. mykiss test fish remained within 5% of the peak AAS between 17.8°C (64°F) to 24.6°C (76°F), these site-specific empirical data with broader temperature thresholds were selected for evaluation of thermal suitability for O. mykiss. In the current study, the temperatures of 18°C (66.4°F), 20°C (68°F), 22°C (71.6°F), and 24°C (75.2°F) were evaluated over each of the summer months (June through September) when these temperatures can be exceeded in the lower Tuolumne River.2 In skipping from study to study, any caveats and limitations that might be present or

implied disappear. In order to avoid such misuse, the authors of the current Study should be more explicit in its caveats and should describe the limitations of its conclusions.

The Study may be limited because it analyzes a single lifestage.

The Study examines only the juvenile lifestage of O. mykiss in the lower Tuolumne River. The Clean Water Act requires that the most sensitive resources be protected. It is not clear whether the adult lifestage, which is also present during the summer time period, is more, equally or less sensitive to high water temperatures. Before adjustments of summer temperature criteria for O. mykiss in the lower Tuolumne River could be considered, an evaluation of the physiological response of adult O. mykiss in the lower Tuolumne River would need to conducted, in addition to completing the evaluation of the physiological response of juveniles. The Study makes comparisons between O. mykiss in the lower Tuolumne River and populations that are more permanent and defined and that have more common characteristics.

The Study draws comparisons with other populations of rainbow trout that have demonstrated higher temperature tolerances than the figures given for juvenile rearing in the EPA (2003) Criteria. Several of these are cited in the EPA document, including redband trout in Eastern Oregon, southern California coastal steelhead, and trout introduced in Australia.

2 Stillwater Sciences, 2015, Lower Tuolumne River Instream Flow Study—Evaluation of effective usable habitat area for over-summering O. mykiss. Draft Report. Prepared by Stillwater Sciences, Davis, California for Turlock Irrigation District, Turlock California and Modesto Irrigation District, Modesto, California. Distributed to relicensing participants via e-mail by Ms. Rose Staples on February 27, 2015, pp. 2-3.

5

Certainly at least the redband and southern California steelhead are more likely to share common ancestry and even genetics than the fish in the lower Tuolumne River, where the population was extremely small due to low project flows until 1995. The current Tuolumne population is likely a combination of residual lower river fish, wild or hatchery fish washed down from La Grange (themselves possibly the result of production in La Grange Reservoir or originating in Don Pedro Reservoir), and some number of anadromous individuals of unknown origin and their progeny. It is further likely that the population is being replenished from these sources on an ongoing basis, and that some portion of the fish that are there in several years will have little directly in common with the current population. This is particularly likely under dry or drought conditions, when a greater proportion of the existing population may be expected to perish. Managing a changing population based on ascribed thermal tolerances of an existing population is questionable both scientifically and as policy.

It is likely that the present population in the lower Tuolumne is temperature tolerant

because it has had to be in order to survive, and that improved thermal conditions would create a larger population. Improved thermal conditions would certainly increase the volume of suitable habitat by pushing thermal limitations further downstream. It is a policy as well as a scientific question whether to manage to the highest suitable temperature (whatever that may be) or to manage to what is likely to produce a stronger population. On a policy and recreational basis, it is hard to justify a small population managed for small fish. If the population were more robust, the argument for managing to a higher temperature would be more credible. There is no bioenergetics study of O. mykiss in the lower Tuolumne River that would support management for water temperatures higher than those recommended in EPA (2003) guidance.

The Districts declined in 2011 to conduct a bioenergetics study of O. mykiss in the lower Tuolumne River as recommended by the Department of Fish and Wildlife.3 The Commission did not order this study. The current Study recognizes: “the important ecological consideration is whether or not there is sufficient food in the Tuolumne River to support the highest MR associated with high temperature.” (Study, p. 22). The Study supports the hypothesis that sufficient food is present only with anecdotal data:

All available studies suggest that the Tuolumne River population is not food limited, including direct studies of Tuolumne River Chinook salmon diet (TID/MID 1992, Appendix 16), long-term benthic macro-invertebrate sampling data collected from 1988–2008 (e.g., TID/MID 1997, Report 1996-4; TID/MID 2009, Report 2008-7), as well as the relatively high length-at-age for O. mykiss sampled in 2012 (Stillwater Sciences 2013). Indeed, the O. mykiss sampled for the current study were apparently feeding well in the river during summer months given the high condition factors (see Appendix 2), feces being regularly found in the swim tunnel and two test fish regurgitating rather large meals post-exhaustion. (ibid).

3 See California Department of Fish and Wildlife, Comments on Proposed Study Plan, eLibrary 20111024-5118, p. 55 ff., proposed Bioenergetics Study.

6

It is one thing to say that there is apparently sufficient food in the lower Tuolumne for the small population of O. mykiss located in a relatively small section of the river. It is quite another to argue in the absence of a targeted study that food production is great enough to support a larger population at the highest metabolic rate associated with high water temperatures. There is no evidence to support such a finding. If food is indeed unusually abundant, why is the O. mykiss population in the lower Tuolumne River neither greatly abundant nor characterized by large numbers of large fish? Conclusion and recommendations

The summer water temperature criteria that are apparently recommended in the Study, and that are more definitively recommended based on the present Study in the just-released draft study entitled Lower Tuolumne River Instream Flow Study—Evaluation of effective usable habitat area for over-summering O. mykiss, are not warranted by the evidence the Study has collected. If the Districts wish to persist in seeking to define site-specific summer water temperature criteria for the lower Tuolumne River, they should affirmatively address the scientific and policy issues we have described above. In brief, these are

1. Follow-up site specific physiological studies must address elevated water

temperatures over an extended period of time, ideally over an entire summer. 2. Follow-up site specific physiological studies must be conducted on adult as well

as juvenile O. mykiss. 3. Follow-up site specific physiological studies must address the likely multiple

sources and ongoing replenishment of the O. mykiss population of the lower Tuolumne River.

4. The Districts should perform a bioenergetics study for juvenile and adult O. mykiss in the lower Tuolumne River.

In addition, the Study should be edited so that the Executive Summary and the

Conclusion place the value of the findings in the appropriate context of how they might inform a comprehensive review of site-specific summer thermal conditions in the lower Tuolumne River.

Please contact Chris Shutes if you have any questions. Thank you for the opportunity to

comment on the draft of the Study entitled Thermal Performance of Wild Juvenile Oncorhynchus Mykiss in the Lower Tuolumne River: A Case for Local Adjustment to High River Temperature.

Respectfully submitted,

Patrick Koepele Chris Shutes Executive Director FERC Projects Director Tuolumne River Trust California Sportfishing Protection Alliance [email protected] [email protected]

THERMAL PERFORMANCE OF WILD JUVENILE ONCORHYNCHUS MYKISS IN THE LOWER TUOLUMNE RIVER:

A CASE FOR LOCAL ADJUSTMENT TO HIGH RIVER TEMPERATURE

APPENDIX 6

RESPONSE TO COMMENTS ON THE DRAFT STUDY REPORT

Appendices


Overarching Reply Comments To CDFW’s Review of the Current Study

On August 31, 2016, California Fish & Wildlife (CDFW) provided comments on the draft report entitled “Thermal Performance of Wild Juvenile O. mykiss of the Lower Tuolumne River” issued in January 2015. It is evident from the comments received from the reviewers that the study team has not been clear enough in describing:

a) the quality of the experimental work and the scientific rigor that was applied;

b) the applicability of the data generated relative to the larger question regarding the conservation of O. mykiss in the Tuolumne River; and

c) what types of data could provide further insight into the thermal ecology of Tuolumne

River O. mykiss Therefore, in addition to our detailed reply comments provided in this Appendix 6, we offer the following discussion to deal with certain issues that were raised in the CDFW review comments, issues that lie both within and outside of the primary purpose of our report. Hopefully, along with our detailed response document to the comments, this will better explain why we took the particular study approach that we did, namely using a temperature-dependent metabolic performance measure (i.e. aerobic scope), the ecological value of which is supported by a large volume of scientific literature. Further, the researchers conducting the study applied state-of-the-art methods and measurement techniques. Therefore, the information generated is applicable to the management of Tuolumne River O. mykiss.

1) What does absolute aerobic scope (AAS) tell us? AAS is a capacity measure or index that

has comparative value. We measure this ‘capacity’ or ‘potential’, if you like. It is clear that the present experiments were not intended to directly address how such capacity would or could be used by the fish. Indeed, very few fish studies have even attempted to study capacity allocation given the inherent difficulties of such an effort. Nevertheless, the most important guiding principle is that if a fish has no aerobic capacity, no activities can be performed other than those dealing with basic survival (basal metabolism in human terms). Conversely, if aerobic capacity is evident, as we discovered across a wide range of test temperatures for Tuolumne River O. mykiss, then that capacity is available for use for activity across the temperature range. AAS is a well-grounded scientific measurement that has only improved with time since its first inception by Fred Fry 60 years ago. We now have better measurement equipment available, as was used in this study, that gives us more reliable, more accurate and more frequent recordings, plus we have video to monitor the fish. Furthermore, we have a much greater appreciation of where errors can be introduced, how large they might be and how they can be avoided. Indeed, an entire special issue of an International journal (Journal of Fish Biology) was devoted to this topic in 2016, which attests to the rigor of the experimental approach we adopted. In the conduct of this study, we followed published

Appendices


principles and guidelines, i.e., our study was state-of-the-art. Few studies, including those used by EPA 2003, have tested wild fish. We tested wild Tuolumne River O. mykiss to ensure direct relevance of the data. AAS simply characterizes what capacity is available; further experiments would be informative to characterize how Tuolumne River O. mykiss allocate this capacity, including the potentially interactive effects of thermal acclimation, growth or reproduction. Thus, while comments and criticisms along these lines may potentially be relevant to the broader management of Tuolumne River O. mykiss, they indicate a misunderstanding of the purpose and use of our study. The present study demonstrates the fact that Tuolumne River O. mykiss have the capacity for the performance of ecologically relevant traits across the wide range of relatively higher temperatures experienced in the lower Tuolumne River.

2) One thing that is clear from our work and of critical importance is that the study populations included in the EPA criteria documents are ‘northerly’ populations. This should not be in dispute. The only work on southern populations comes from Dr. Joseph Cech’s lab and post-dates the EPA document which was used to set the 7DADM. Also clear is that the O. mykiss benchmark temperatures were established over a dozen years ago and considerable new science has amassed on thermal effects on fishes. Indeed, it may be the most intensely studied topic within fish biology over the past 10 years.

3) Since the early 2000’s or so, population-specific thermal sensitivity research, especially

for fishes, has expanded greatly, including further methodological and interpretive advancements. It is now widely accepted that local populations of fish of the same species can differ in thermal sensitivity, and it has been consistently demonstrated that their sensitivity is usually matched to their native or local thermal regimes whenever this has been properly tested. These observations are consistent with what is termed local thermal adaptation. Therefore, a logical link should be that thermal regulatory criteria should acknowledge the local population’s thermal sensitivity. The main point here is that any regulatory guideline should properly reflect the fish species and location to which they are intended to apply. Indeed, the EPA 2003 supporting document directly acknowledges this, but also notes there was insufficient data at that time to provide an informed opinion. The database on local adaptation within a species has now changed enormously. Thus, whenever evidence for local adaptation of a particular population of fish emerges, it is entirely reasonable to challenge the applicability of a more general guideline. This study tested wild Tuolumne River O. mykiss to ensure direct relevance of the data.

Population-specific performance is seen in many traits: growth, lethal limits, swimming performance, metabolic performance and aerobic scope, and each of these traits can be shaped by temperature at a variety of interacting timescales [i.e. acute (seconds to minutes), acclimatory (days to weeks; perhaps ‘chronic’ using the CDFW reviewer’s terminology), and adaptive]. Indeed, these are complex traits, and ecologists agree that these traits have implications at the level of the population. We acknowledge that there is debate about specific ‘implications’, but we try to be clear and precise, as well as conservative, as to what our data on Tuolumne River O. mykiss have revealed for the first time.

Appendices


As a general rule and an example, positive growth rates occur under conditions that are conducive to survival. Exactly how growth and survival translates to population dynamics requires considerably more detailed study beyond measuring growth rate, and perhaps modeling, which is never perfect without reliable input variables. Natural selection directly operates at the level of the individual, and effects become manifest at the population level. Therefore, understanding effects on individuals and knowing the physiological mechanisms that operate within individuals are key pieces of knowledge to obtain before attempts to extrapolate to the population level can be made with confidence. Consequently, we performed experiments that targeted individual, wild juvenile fish and probed mechanisms of the thermal tolerance that are well established in the mainstream fish literature. To reiterate, we performed our experiments on wild Tuolumne River fish (not hatchery fish as used for EPA 2003), captured from their native habitat and tested streamside. This experimental design is particularly powerful in estimating innate, real-time AAS capacity for this specific population.

4) AAS allows us to make comparisons. For example, we can safely conclude that the lower Tuolumne O. mykiss population does comparatively better at warmer temperatures than northern O. mykiss populations because we have shown aerobic performance across a temperature range that includes temperatures higher than those tolerated in northern populations. Consequently, our data only addresses the ‘blanket’ 7DADM guideline for all O. mykiss populations across the US, in one specific manner: we no longer have confidence in the growth studies used by EPA in 2003 to set guidelines for lower Tuolumne River O. mykiss because our AAS data for lower Tuolumne rainbow trout clearly show that this population is unlike and definitively different from more northern populations. Consequently, it is a confidence issue. Of course, any new guidelines should only be considered in close consultation with the EPA, and using the best available science and its modern interpretation. We did not suggest otherwise in the report and continue to hold this viewpoint. This is what the data are telling us, nothing more and certainly nothing less.

5) What our data should NOT be used for is to pick a new thermal criterion based solely on our aerobic scope curve. In fact, we do not suggest revising the 7DADM based solely on our AAS curve. We simply state that we believe our data are suggestive of local thermal adaptation in Central Valley fish and inconsistent with a blanket criterion for the population under consideration. Because the Tuolumne River O. mykiss fish outperform northerly populations at warm temperatures, the inference is that the current guidelines are overly conservative.

6) We also assume, perhaps incorrectly, that all of the scientists working on thermal

requirements of fishes would appreciate, without repeated statement, the fact that this study addresses physiological mechanisms related to temperature and temperature alone, and to juvenile fish alone. Nowhere did we extrapolate our findings to other life stages because we are aware of, and therefore sensitive to, some species of fish showing stage-specific thermal sensitivity. We also know that multiple stressors can interact (e.g. temperature sensitive metabolism x food) in additive or synergistic ways, so nowhere do we suggest

Appendices


that our data are the sole requirement to determine a 7DADM. However, the value of population-specific, site-specific data should not be underestimated.

7) Additional data that may be helpful to managing Tuolumne River O. mykiss might include:

comparative thermal sensitivity literature from other studies and other populations; knowledge of food resources available to the fish in question; and life-stage sensitivities that could reveal a ‘weak link’ in life history. Of course, this is not exhaustive, but it does acknowledge possible additional information that would be useful. We understand that at least some of this data is already available.

8) While we never suggest that a new 7DADM value be extracted solely from our data, we do suggest that the current value is conservative for Tuolumne River juvenile O. mykiss. Also, we know as a fact that a higher thermal tolerance than that reflected by the EPA 2003 7DADM exists within the O. mykiss genome because publications on local thermal selection (e.g. Australian and Japanese rainbow trout) conclusively illustrate that these populations feed and grow at temperatures well in excess of 20°C.

It is true that we do not know the growth capabilities of Tuolumne River O. mykiss but given our new understanding of FAS values, the Tuolumne River O. mykiss have sufficient aerobic capacity to eat a large meal, they had food in their stomachs when captured, have an abundance of food in the Tuolumne River, and have been videoed swimming to capture food passing by at temperature well above the EPA recommended 7DADM. This all provides additional evidence that juvenile O. mykiss captured from the lower Tuolumne River are feeding and growing in the current thermal regime. Growth studies would be useful to confirm rates of growth, but the present study supports the Tuolumne River O. mykiss’ significant capacity for growth. If there is any doubt that a higher thermal tolerance than that reflected by the 7DADM exists within the O. mykiss genome, we simply have to turn to the established physiological literature on a variety of O. mykiss that through natural selection live in the deserts of Idaho and Oregon, namely the redband trout, Oncorhynchus mykiss gairdneri. This variety deals with, as well as swims and feeds in, summer temperatures that can reach 26oC.

It is our hope that these remarks clarify some of the apparent misunderstanding of the design and purpose of the study. Below we respond to individual comments received on the draft report.

Appendices


A Note about ‘Acute’ and ‘Chronic’ Temperature Response There is often much discussion and debate about ‘acute’ and ‘chronic’ temperature response in fish. For the purposes of this discussion, we view ‘acute’ as relevant over the timescale of seconds to hours to days. Over longer timescales, weeks to months, temperature is considered ‘chronic’. However, some scientists reserve the term chronic to a certain portion of the lifecycle of a test animal, e.g., mammalian toxicology. Binning the effects of temperature on fishes into categories like ‘acute’ and ‘chronic’ is not a straightforward task and to attempt to do so is a dramatic oversimplification of both experimental methodologies and organismal biological, physiological and behavioral responses. This is in part because fish can acclimate to a new temperature and this acclimation can follow different time courses depending on the process being studied, and because a fish is rarely exposed to a single, static temperature for many weeks. Thus, while there are very good experimental reasons to control the acclimation temperature for groups of fishes before testing (as you would do in laboratory acclimation studies of thermal tolerance or growth), these tests are artificial and eliminate naturally-occurring thermal oscillations as well as fish behavioral selection of particular thermal habitat. We argue that it is much more insightful to understand the biologically-relevant oscillations in environmental temperatures of a particular system, which likely include daily fluctuations, fluctuations occurring over seasons, and/or variation in spatial temperature distributions. It is also critical to understand how these temperature profiles interact with the response variable that you are measuring (e.g. molecular responses as compared to organismal growth – each of which will have a distinct response pattern and response time). For example: heat shock proteins show an acclimation response in a matter of hours, whereas whole animal physiology can take weeks to acclimate. Lastly, one of the more challenging tasks for scientists is to understand how fish behaviorally utilize their thermal habitat as a reflection of their physiological capacities and limits. Consequently, how fish respond, physiologically and behaviorally, to environmental temperature change is a function of previous thermal history (e.g. seasonal acclimation), the magnitude and timescale of the thermal change (e.g. how high did the temperature rise, how quickly), and the duration of the exposure (e.g. how long was the new thermal exposure). Regulations should incorporate data that speak to each of these aspects. We point out that the 7DADM is neither an “acute” or “chronic” regulation, but it is in fact designed to incorporate temperature oscillations. Incorporating thermal heterogeneity into fish habitat, when done properly, is certainly more appropriate than managing to a static (chronic) thermal target (which we all should be able to agree is completely artificial to fishes that evolved in habitats with thermal variability). Importantly, no single study exists, or can be designed, that completely incorporates the complexities of thermal exposures and measured endpoints to ‘spit out’ the perfect thermal regulatory criteria for a particular species. Thus, regulations are based on a collection of data/experiments spanning so called ‘chronic’ and ‘acute’, biologically relevant thermal exposures and incorporating a variety of well-studied and understood endpoints. Or, in some

Appendices


cases, when data are not available for strong support, regulations should be reasonably protective. With specific respect to the study conducted on the Lower Tuolumne River, the fish were seasonally acclimated to the prevailing summer river conditions. We knew the temperature at which they were captured, but not the temperatures that they had experienced or for how long they had experienced them. We minimized the potential effects of thermal acclimation of processes that take many hours or weeks (fish were tested immediately, i.e. within hours, following capture from the river). Lastly, metabolic performance capacity was measured as a function of an incremental warming protocol that lasted no longer than 6 hours of exposure to a test temperature between 13 and 25°C, depending on the individual.

Appendices


Comment # (page #) Comment Districts’ Response

TRT/CSPA-1 (p. 2)

The Study does not evaluate the physiological response of the population of O. mykiss in the lower Tuolumne River over time. There are limitations to the Study that the Study does not acknowledge. Chief among these limitations is that the Study does not evaluate physiological response of the population of O. mykiss in the lower Tuolumne River over time. On the contrary, 75% of the test fish were sourced from a location one mile downstream of La Grange Powerhouse, where temperatures at capture ranged from 12.7°C to 17.1°C. While the Study is critical of Hokanson (1977) for an issue concerning confidence intervals, the Study does not address Hokanson’s use of a 40-day period to evaluate physiological response. Other studies (e.g. Brett 1956; Bidgood 1969) similarly address long-term exposure to less-than-optimal thermal conditions. The Study does not acknowledge this limitation.

We could have been clearer about stating the design and intent of the study. The Report has been amended accordingly. However, the study plan prepared for the study and reviewed previously by the commenter spelled out the specific design and purpose of the study. This never changed. We do not understand the commenters’ concern regarding lack of evaluating responses over time. This was not the objective of the Report as clearly explained in the original study plan. Indeed, the permits issued by the resource agencies for fish removal would NOT permit more than 2 fish to be studied at a time and over time – this was the maximum number of fish that could be removed from the river. To reliably measure growth rate, at least 40 days would be needed to detect responses and rates because the fish have to change their mass by a reliably detectable amount. This was never intended, as explained in the study plan. Nor could we have done this within the limitations of the permits issued. Instead, we measured oxygen uptake, which uses a different time scale, and it can be reliably measured over periods of minutes. Also, we went to great lengths to follow and analyze oxygen uptake over a nearly 24-hour period to examine its variability and ensure our estimates of RMR were as accurate as possible for a field study. Also, we carefully measured maximum oxygen uptake in the manner used by both Fry and Brett (who was Fry’s student), but using modern technology with greater accuracy and precision. Therefore, we can state with confidence what the fish’s

Appendices



capacity was in terms of delivering oxygen to tissues over a broad temperature range. If, however, the comment concerning “over time” is that it might be beneficial to study fish that were acclimated to different temperatures (14°C was the coolest temperatures found in the lower Tuolumne River during 2014 study period), this is a valid comment. Nevertheless, it is well known that thermal acclimation is used by fishes to “improve performance” at the new acclimation temperature. Therefore, if the lower Tuolumne River O. mykiss used in the present study can be shown to acclimate to water temperatures warmer than they were experiencing at the time of the experiments, we have then provided a conservative estimate of temperature effects on the fish performance by looking only at the effect of a rapid rise in water temperature from river temperature to which they were acclimated. The concern raised does not change the outcome of our results, but does introduce the possibility that this fish population could do even better at warmer temperature if they were allowed to first acclimate. The comment of the reviewer goes on to be critical of our critique of the general application of Hokanson’s data on rainbow trout that were studied in the American midwest to a population of rainbow trout in the Central Valley, CA. The fact is, as proven by the study, Tuolumne O. mykiss juveniles displayed a physiology and thermal tolerance quite different from more northern populations of rainbow trout. In fact, we point out that they are more similar to O. mykiss populations that have adapted to desert streams!

Appendices



We agree that the experimental approach used in this study differs fundamentally from the approach used by EPA to formulate its temperature recommendations for the Pacific Northwest. Three points of clarification are below.

1. Our criticism of Hokanson (1977) is two-fold. Foremost, today’s knowledge of local adaptation of a wide range of species from sticklebacks, through killifish to salmonids tells us that it may be inappropriate to apply studies that are geographically separated by large distances and differing climates. For example, use of data from studies of trout from central USA to the same species locally adapted to California river systems may be inappropriate. Indeed, this was the primary driver for the present study. Our commentary on the work of Hokanson (1977) is valid, as it does not criticize the quality of the data per se, rather the application of the results. Our concern about confidence limits in Hokanson (1977) is a minor one, driven in part because against our a priori predictions of fish performance, we found that the Tuolumne River fish were unexpectedly tolerant of acute changes in temperature and performed similarly over a wide range of temperatures. This introduces the statistical issue of when does warm temperature create an unfavorable fish performance. We think confidence limits are needed. Without confidence limits for a study such as Hokanson (1977), which is nearly 40 years old and did not have access to the statistical tools now

Appendices



commonly available, we cannot retrospectively interrogate these older data.

2. Lastly, how these fish exploit the local thermal gradients in the lower Tuolumne River was not part of the objectives of the current study. Nevertheless, it is possible to speculate. Perhaps they behave similar to the sockeye salmon that Brett studied in the 1970’s, by diurnally moving to warm reaches to feed and returning to cooler reaches to digest their food. This type of behavior would take advantage of warm habitats. In any event, whether such behavior occurs or does not occur has no effect on the conclusions of the present study.

TRT/CSPA-2 (p. 2)

A City of San Francisco biologist has acknowledged on the record in this proceeding that O. mykiss populations in the lower Tuolumne River are substantially smaller than populations downstream of rim dams in the Sacramento river drainage, where water temperatures are generally much lower than temperatures in the lower Tuolumne River.

Comparisons of differing O. mykiss population sizes from spatially distinct river systems citing a statement by a “City of San Francisco biologist” lacks scientific rigor and should be disregarded. Just as one example, the two rivers under comparison would have substantially different geomorphological histories and structures which may be a more important factor affecting population sizes. Other factors may also play key roles in abundance such as total spawning area, differing food sources, predation pressures, fishing activities, etc. These issues have nothing to do with temperature in this system.

TRT/CSPA-3 (p. 2)

Before any adjustment to the established (EPA 2003) temperature benchmark for a 7DADM value for the population of O. mykiss in the lower Tuolumne River is considered based on site-specific conditions and response, further investigation and evaluation would be required. The Study should explicitly state this, and

It should be noted that EPA (2003) does not provide specific temperature recommendations to the lower Tuolumne River or any California river system. It is a general recommendation that applies to all populations of rainbow trout. The report’s discussion is not intended as the basis for changing the EPA (2003) 7DADM recommendations.

Appendices



should describe additional evidence needed before any change in the 7DADM value for the population O. mykiss in the lower Tuolumne River might appropriately be evaluated.

Instead, our work simply suggests that the current value of 18°C lacks merit for the current O. mykiss population found in the lower Tuolumne River. Minimally, 18°C as the 7DADM value is a very conservative upper thermal limit based on the results of the current study. This report is not intended to preempt consultation with EPA. We strongly believe that the new data collected here are a firm basis for opening such a dialogue about site-specific temperature criteria in general as well as for the Tuolumne River O. mykiss. We suspect that EPA would welcome this dialogue being opened as their 2003 report acknowledged the possibility of local adaptation. EPA 2003 simply cited that the scientific evidence at that time was weak. The scientific evidence is now much stronger.

TRT/CSPA-4 (p. 3)

The Study results alone do not warrant site-specific summer water temperature criteria for O. mykiss in the lower Tuolumne River.

The Districts assert that the site-specific empirical evidence that exists for Tuolumne River O. mykiss warrants considerable weight when compared to data from completely different regions of the country. Also, see response to TRT/CSPA-3

TRT/CSPA-5 (p. 3)

In our view, the appropriate use of the Study results would be to 1) evaluate their limitations; 2) develop additional investigations that might be necessary to scientifically justify consideration of adjusting thermal criteria for the population of O. mykiss in the lower Tuolumne River, 3) enumerate and evaluate regulatory and policy issues that might be involved in adjusting these criteria; and 4) assemble these necessary components and, based on this ensemble, develop a process for considering and evaluating site-specific water temperature criteria.

While evaluating natural background provisions and use attainability exceptions to the EPA (2003) 18°C 7DADM recommendations are beyond the scope of the current study, the Districts (TID/MID 2014, Attachment A) previously demonstrated that potential re-operation of the Don Pedro Project to meet EPA (2003) temperature recommendations was infeasible under a range of potential scenarios evaluated, including “without dams” scenarios. Given the infeasibility of meeting the EPA 18°C 7DADM benchmark and that the results of the current study

Appendices



demonstrated near-optimum physiological performance and active feeding at temperatures well above 18°C, consideration of site-specific exceptions to this recommendations are warranted. We are pleased to read that the reviewer appears to agree with us that further steps are warranted. Therefore, the real issue is not what is contained in the report, but rather what should follow from it.

TRT/CSPA-6 (p. 4)

The authors of the current Study should be more explicit in its caveats and should describe the limitations of its conclusions.

As noted above in the response to TRT/CSPA-1 the data generated here would in our opinion represent the most conservative estimate of temperature effects by only looking at the effect of a rapid change from acclimation temperatures. To speculate on how well O. mykiss from the lower Tuolumne River might perform if they were allowed to acclimate to even higher water temperatures is beyond the scope of the present study. We agree that we could have been clearer about stating the limitations of the study, but it is clear that these limitations are more pertinent to the future actions and not the conclusions that are based on our data.

TRT/CSPA-7 (p. 4)

The Study examines only the juvenile lifestage of O. mykiss in the lower Tuolumne River. The Clean Water Act requires that the most sensitive resources be protected. It is not clear whether the adult lifestage, which is also present during the summer time period, is more, equally or less sensitive to high water temperatures. Before adjustments of summer temperature criteria for O. mykiss in the lower Tuolumne River could be considered, an evaluation of the physiological response of adult O. mykiss in the lower Tuolumne River would need to conducted, in addition to completing the evaluation of the physiological response of juveniles.

At no time in the report do we state that our results for juvenile fish are directly applicable to other life stages of this species. Although some studies have examined the relative thermal tolerance of juvenile and adult salmonid life stages, evaluation of the thermal performance of adult O. mykiss was outside the scope of the study plan. The decision to use juvenile vs adult-sized fish was made on the basis of higher relative abundance and the ability to capture them with beach seines vs angling that may result in reduced swimming performance and necessitate longer recovery times for adult fish that were captured by that method.

Appendices



Again, future steps might be to study other life stages, but this possibility does not challenge the present results. We would also like to note that in order to advance to the adult life stage, fish must survive the juvenile life stage. And juveniles are evidence of a successful life stage in this river system. It would be an odd, and unsustainable, biological adjustment to have juvenile fish be well acclimated to local conditions only to prove fatal when it reaches the adult life stage.

TRT/CSPA-8 (p. 4)

The Study makes comparisons between O. mykiss in the lower Tuolumne River and populations that are more permanent and defined and that have more common characteristics.

The current Tuolumne population is likely a combination of residual lower river fish, wild or hatchery fish washed down from La Grange (themselves possibly the result of production in La Grange Reservoir or originating in Don Pedro Reservoir), and some number of anadromous individuals of unknown origin and their progeny. It is further likely that the population is being replenished from these sources on an ongoing basis, and that some portion of the fish that are there in several years will have little directly in common with the current population. This is particularly likely under dry or drought conditions, when a greater proportion of the existing population may be expected to perish. Managing a changing population based on ascribed thermal tolerances of an existing population is questionable both scientifically and as policy.

This comment is highly speculative. Moreover, we do not fully understand what the reviewer means by “populations that are more permanent and defined and that have more common characteristics”. The population that we have studied and that is protected is a resident of the river system, one that has a barrier upstream in the form of a dam and a potential thermal barrier downstream. How they arrived there and how they adapted is not a concern of this Report. This Report focuses on the thermal capacity of the fish that currently reside in the river and are protected by current regulations.

Appendices



Thus, while the supposition that future lower Tuolumne River O. mykiss populations may have little directly in common with the current population is interesting, it does not affect the conclusions of our study since we are limited to testing the current population. Furthermore, if the intent is to do future stocking of this river system with fish from either a hatchery or another wild source, then similar experiments could be performed on those populations. The use of ‘wild’ or ‘local’ fish for the Tuolumne River would then have to be redefined.

TRT/CSPA-9 (p. 5)

There is no bioenergetics study of O. mykiss in the lower Tuolumne river that would support management for water temperatures higher than those recommended in EPA guidance. The Districts declined in 2011 to conduct a bioenergetics study of O. mykiss in the lower Tuolumne River as recommended by the Department of Fish and Wildlife. The Commission did not order this study.

The EPA guidance was not based on any bioenergetics studies. While the Districts were not required to undertake a direct bioenergetics study in FERC’s May 21, 2013 study determination, it should be understood that all bioenergetics (activities) require oxygen. The current study characterized the maximal capacity to deliver oxygen for any and all activities. Indeed, for any energetic model the currency can be oxygen or Joules. Regardless, the sum of all the bioenergetics inputs cannot in the long term exceed feeding input (the fish would be starving) or the maximum aerobic capacity (which is exactly what we measured). In addition, the multitude of factors that go into a bioenergetics study would require a large number of individuals to be removed from the river, well in excess of the authorized fish take, and many of these fish would have to be sacrificed for such a study. The current study design allowed direct examination of physiological performance without the need for either high levels of fish take or sacrificing fish. Indeed, we successfully returned all but three of the study fish to the river. Further, the O. mykiss

Appendices



population studies and resulting in-river population model does include a bioenergetics component. To be clear, the study was implicitly designed to minimize the impact of fish removal from the river for experimentation.

TRT/CSPA-10 (p. 6)

Follow-up site specific physiological studies must address elevated water temperatures over an extended period of time, ideally over an entire summer.

We do not understand the commenters’ concern regarding lack of evaluating responses over time. This was not the objective of the Report, as clearly pointed out in the study plan. Indeed, the permits issued for fish removal would NOT permit more than 2 fish to be studied at a time and over time – this was the maximum number of fish that could be removed from the river. Perhaps the following clarifies matters. To reliably measure growth rate, at least 40 days would be needed to detect a response because the fish have to change their mass by a reliably detectable amount. We clearly could not do this with the permits issued, nor did the study plan (previously reviewed by CDFW) ever suggest this was the intent of the study. Instead, as detailed in the study plan, we measured oxygen uptake, which uses a different time scale, and it can be reliably measured over periods of minutes. Also, we went to great lengths to follow and analyze oxygen uptake over a nearly 24-hour period to examine its variability and ensure our estimates of RMR were as accurate as possible for a field study. Also, we carefully measured maximum oxygen uptake in the manner used by both Fry and Brett, but using modern technology with greater accuracy and precision. Therefore, we can state with confidence what the fish’s capacity was in terms of delivering oxygen to tissues over a broad temperature range.

Appendices



If, however, the comment concerning “over time” is that we need to study fish that were acclimated to different temperatures (14°C was the coolest temperatures found in the lower Tuolumne River during the 2014 study), this is a valid comment. Nevertheless, it is well known that thermal acclimation is used by fishes to “improve performance” at the new acclimation temperature. Therefore, if the lower Tuolumne River O. mykiss used in the present study can be shown to acclimate to water temperatures warmer than they were experiencing at the time of the experiments, we have then provided a conservative estimate of temperature effects on the fish performance by looking only at the effect of a rapid change in water temperature from river temperature to which they were acclimated. This concern does not change the outcome of our results, but does introduce the possibility that this fish population could do even better at warmer temperature if they were allowed to first acclimate.

TRT/CSPA-11 (p. 6)

Follow-up site specific physiological studies must be conducted on adult as well as juvenile O. mykiss.

The decision to use juvenile vs adult-sized fish was made on the basis of higher relative abundance and the ability to capture them with beach seines vs angling that may result in reduced swimming performance and necessitate longer recovery times for adult fish that were captured by that method. However, we would note that in order to advance to the adult life stage, fish must survive the juvenile life stage. And juveniles are evidence of a successful life stage in this river system. It would be an odd, and unsustainable, biological adjustment to have juvenile fish be well acclimated to local conditions only to prove fatal when it reaches the adult life stage.

Appendices



The suggestion to study Tuolumne River adult O. mykiss would seem to indicate agreement with the Districts’ assertion that the EPA 2003 recommended temperatures are invalid because the EPA research involved no site-specific studies nor even results of research on CA O. mykiss, neither juvenile nor adult. Note also that Hokanson (1977) on which EPA 2003 recommendation is based, did not consider multiple life stages.

TRT/CSPA-12 (p. 6)

Follow-up site specific physiological studies must address the likely multiple sources and ongoing replenishment of the O. mykiss population of the lower Tuolumne River.

See response to TRT/CSPA-8 above. However, the commenter is suggesting that O. mykiss from different locations in the lower Tuolumne (even within a mile of each other) would have differing thermal capacities, while the EPA 2003 paper proposes that all O. mykiss populations in the entire Pacific NW and CA should be considered to have the same thermal capability. We agree with the commenter that site-specific empirical information is a much better measure of performance. Nonetheless, our Report concerns one specific population.

TRT/CSPA-13 (p. 6)

The Districts should perform a bioenergetics study for juvenile and adult O. mykiss in the lower Tuolumne River.

See response to TRT/CSPA-9 and -12.

TRT/CSPA-14 (p. 6)

The Study should be edited so that the Executive Summary and the Conclusion place the value of the findings in the appropriate context of how they might inform a comprehensive review of site-specific summer thermal conditions in the lower Tuolumne River.

The Executive Summary has been amended to address this comment. Also, see response to TRT/CSPA-3. Again, the present study suggests that applying the EPA 2003 recommendation to the Tuolumne River O. mykiss population is overly conservative.

SWRCB-1 (p. 1)

Study Plan 14 was not required by FERC in its Final Study Plan Determination and is not supported by the State Water Board, California Department of Fish and

In its December 2011 SPD, FERC stated that it would consider additional empirical evidence from the Tuolumne River. The development, evaluation and application of

Appendices



Wildlife (CDFW), United States Fish and Wildlife Service (USFWS), or the National Marine Fisheries Service (NMFS).”

empirical evidence that would reduce uncertainties regarding temperature-related effects on Tuolumne River salmonids is the primary purpose of this study. FERC’s emphasis on empirical evidence has further encouraged the Districts to identify and consider new evaluations that could contribute to more focused understanding of potential influences of temperature on LTR salmonids, which led to the development of this study approach and report.

SWRCB-2 (p. 2)

The report does not explicitly state that its results alone demand a change in the 7DADM temperature outlined in the 2003 USEPA Guidance. Rather the report states that this information should be used to determine a 7DADM value specific to Tuolumne River O. mykiss. However, the report does not outline a process to be used to determine a scientifically acceptable and defensible 7DADM specific to the Tuolumne River O. mykiss.

The recommendation lies well beyond the objective of the present report. See response to TRT/CSPA- 3 and -4.

SWRCB-3 (p. 2)

“State Water Board staff recommends that any process to develop temperature criteria specific to the Tuolumne River follow a similar process as the EPA Guidance. Two additional examples of the recommended process include: The Final Staff Report for the Klamath River Total Maximum Daily Loads (TMDLs) Addressing Temperature, Dissolved Oxygen, Nutrient, and Microcystin Impairments in California (NCRWQCB 2010), and The Effects of Temperature on Steelhead Trout, Coho Salmon, and Chinook Salmon Biology and Function by Life Stage; Implications for Klamath Basin TMDLs (NCRWQCB 2005).”

The references provided by the SWRCB deal with TMDL development, a different process than that required for amending the present temperature guidance. In any event, the Districts look forward to working with the SWRCB on temperature issues on the lower Tuolumne River.

Appendices



SWRCB-4 (p. 2)

“It is important to point out that the Report focuses on increased water temperature effects on only one parameter (oxygen consumption) and one life stage (juvenile) for O. mykiss. Study Plan 14 and the Report do not evaluate long term effects of increased water temperature as well as the other life stages of O. mykiss. Questions that might be evaluated as part of a more comprehensive study include, but are not limited to: 1. What is/are the effect(s) of increased temperature conditions on other life stages of O. mykiss or the long-term effects of this short-term exposure on O. mykiss? 2. How does temperature influence other factors which may affect salmonids, such as food availability and disease?

We thank the commenter for explicitly stating what our Report achieved. It would seem from this that the reviewer has no difficulty with accepting our data. Again it seems that the reviewer is making suggestions for future steps, which we have commented on above: See response to TRT/CSPA-1 & TRT/CSPA-9. Regardless, it is important to remember that the present study significantly expands the knowledge base regarding O. mykiss on the lower Tuolumne River. There were no studies for this population prior to the present study, which was a rigorous and comprehensive examination of thermal performance on juvenile wild fish. Indeed, the reviewer does not challenge the quality of the data in hand. We would argue that empirical data are a better indicator of thermal performance than largely unrelated information the applicability of which is difficult to measure. Related to temperature’s influence on other factors, prior studies of food sources on the Tuolumne under the existing temperature and flow regime have indicated healthy BMI populations and that prior studies have not found any significant disease issues with Tuolumne River salmonids.

SWRCB-5 (p. 2)

Study Plan 14 and the Report only consider increased temperature effects on fish persisting in the Tuolumne River under current conditions. Study Plan 14 and the report fail to examine the effects of increased river temperatures on the recovery of O. mykiss populations in the Tuolumne River.

The current study was able to examine the effect of increased temperature on juvenile O. mykiss persisting in the Tuolumne River under current conditions. While examination of questions related to conditions affecting future O. mykiss populations in the lower Tuolumne River are beyond the scope of the current study, the present study indicates an ability of the local population to adjust to local conditions.

Appendices



The Districts are uncertain as to what is meant by “recovery of O. mykiss” under “increased river temperatures”.

CDFW-1 (p. 2)

The study design identifies acute exposure to stressful warmer water temperatures at the individual level; therefore, the study cannot inform development and/or revision of population level chronic water temperature criteria. In their report, the authors compare their acute water temperature results to the United States Environmental Protection Agency's chronic population criteria (USEPA 2003) which is inappropriate.

The reviewer has a fundamental misunderstanding of the study design and purpose. The reviewer thinks we were conducting an acute survival study not testing metabolic capacity (through eliciting maximum metabolic rates using swim tests) at chronic water temperatures. We ask that the reviewer please read the overarching statement where we clearly explain what AAS measures tell us, and how these data relate to the 7DADM. Also, see our response to TRT/CSPA-3: The report discussion is not intended as the basis for changing the EPA (2003) 7DADM recommendations. Instead, our work simply suggests that the current value of 18°C lacks merit for the current O. mykiss population found in the lower Tuolumne River. Minimally, 18°C as the 7DADM value is a very conservative upper thermal limit based on the results of the current study. Please note that the casual use of ‘stress’ and ‘stressful’ should be avoided. Defining ‘stress’ in fish requires rigorous experiments at a population-specific level, careful endpoint selection and interpretation, and well-described exposure conditions. Assuming that our test conditions were ‘stressful’ and assuming that warm temperatures are necessarily ‘stressful’ to Tuolumne River O. mykiss is an unsupported statement. Our data, in fact, suggest that at temperatures much higher than 18C, the tested fish maintain maximum AS. We do not see how a fish that is “stressed”

Appendices



would be able to maintain its AAS given that “stress” is a metabolic load (see Fry, Beamish and Brett reviews of this in the last century) that necessarily limits aerobic performance.

CDFW-2 (p. 2)

Anadromous salmonids populations throughout the Pacific Northwest (including California) are declining primarily because of poor reproductive success and recruitment back into the population (Yoshiyama et al 2001).

It must be noted that this is not what the referenced paper said or concluded.

CDFW-3 (p. 2)

The intent of the USEPA (2003) analysis was to reverse that trend by presenting chronic population water temperature criteria.

This is the reviewer’s interpretation of the intent of EPA (2003) but this was not the stated rationale for the document. In fact, the EPA (2003) report did acknowledge that “local adjustment was possible and that well-designed studies could be used to identify site-specific thermal adjustments”. This was one of the reasons for conducting the present study, which we believe was well-designed and well-executed. It produced definitive and reliable data. Importantly, the 7DADM criteria incorporates information to estimate thermal optima and performance breadth using a diversity of thermal performance metrics (e.g. lethal limits, longer term thermal experiments related to growth, and many others) that operate several timescales of thermal exposure. The focus by the reviewer in casting the EPA 7DADM criteria as exclusively chronic is misleading and incorrect. Please see our response to TRT/CSPA-3, as well as the overarching response document at the front of this response to comments.

Appendices



CDFW-4 (p. 2)

Chronic criteria and population criteria are always lower than acute and individual criteria. The authors presented higher acute/individual water temperature criteria based on a single study, but failed to extrapolate the results to a lower chronic population criteria that would be protective for reproductive success and recruitment to maintain a sustainable (i.e. viable) population. Survival rates are based on amount of time exposed, as well as temperature exposure, and are extremely well described in the scientific literature.

As stated above, the reviewer’s fundamental misunderstanding appears to be that we were conducting a survival study not a test of metabolic capacity. There is no need to extrapolate our study results to a lower chronic temperature criteria since we were making direct measurements of the optimal temperature range for Tuolumne River O. mykiss based on their metabolic capacity. Indeed, we would argue that there is no reliable methodology to extrapolate from acute to chronic studies. However, if a fish cannot perform after an acute temperature change, it is unlikely to perform well with a chronic change unless it can thermally acclimate. We show that the fish do well with an acute thermal change, and these data do not appear to be in dispute. Please see the overarching response document at the front of these responses to comments where we explain what should and should not be gleaned from our data as well as specific remarks on how our data ‘scale’ up to population level functions. Also see our response to CDFW-3 regarding chronic criteria.

CDFW-5 (p. 3)

Executive Summary, page i, second paragraph. The authors stated, "The study tested the hypothesis that the Tuolumne River O. mykiss population below La Grange Diversion Dam is locally adjusted to the relatively warm thermal conditions that exist in the river during the summer". What is the authors' definition of "locally adjusted"?

"Locally adjusted" was defined on page 6 as a hypothesis that can be tested by confirming or refuting by evaluating the predictions on page 7 of our report. In short, the hypothesis is: Tuolumne River O. mykiss are “locally adjusted” if they have higher metabolic capacity (absolute aerobic scope) at temperatures above 18o C. A finding like this would be in contrast to the earlier data based solely on northern fish data. Locally adjusted is a term that includes, and does not distinguish between, local adaptation and local

Appendices



acclimation/acclimatization. The assignment of ‘adaptive’ happens at the level of the population, not for individuals. This is in contrast to acclimation/acclimatization, which are traits/responses that change at the individual level, over the course of a particular individual’s lifetime. A simple example would be how season influences fur thickness in bears. This is a trait that varies by individual, by season. Please see the last sentence of the executive summary where we explicitly state that our study does not distinguish between these two explanations as this was not our objective. Also, please see the entire section in the Introduction (Current Evidence for Local Physiological Acclimatization and Genetic Selection) articulating how acclimation/acclimatization/adaptation is defined. Thus, we used a term that encompassed both mechanistic explanations.

CDFW-6 (p. 3)

Executive summary, page i, third paragraph, last sentence. The authors state, "Therefore, the experimental approach also acknowledges that every activity of a fish in a river (swimming, catching prey and feeding, digesting a meal, avoiding predators, defending territory, etc.) requires oxygen consumption above a basic routine need and that salmonids have evolved to maximize their oxygen supply when they fuel muscles during exhaustive swimming". This statement leads to three questions; 1) This test appears to study basic survival, but does the study address reproductive success and recruitment? 2) Does this experimental design measure activities related to spawning, immune function and general overall stress? and 3) Isn't this the case for all vertebrates, that an animal's physiological function evolved to fuel their

As described in the study plan prepared for this study and submitted for review prior to conducting the study, we did not aim to study basic survival. Therefore the reviewer misunderstands the study objectives. We measure aerobic capacity. Survival would require a minimum of SMR, but this study shows that these fish had the capacity to more than double their metabolic rate at specific temperatures. This information can be used to assess whether there is the capacity for activities well beyond survival, such as growth, immune function, reproduction, predator avoidance, etc. We cannot comment on how the fish use this capacity. These are behavioral decisions made by this fish not by us. However, it we know that the energetic cost of an activity is maximally a doubling of metabolic rate, and this fish has this capacity, it is reasonable to then conclude that the fish has the capacity to undertake this activity.

Appendices



muscle under non-resting (exercise) or stress conditions?

Regarding question (3), the issue is not whether an animal’s physiological function has evolved to fuel their muscles during exercise, the issue is their capacity to fuel their muscles at different temperatures.

CDFW-7 (pp. 3 and 4)

Executive Summary, page i, fourth paragraph. The authors state "As expected for a fish, RMR [Routine Metabolic Rate] increased exponentially with increasing test temperature from 13°C to 25°C (36 different fish, each at a single test temperature)". Basically the RMR is a fish in a resting state, thus if their RMR increased with temperature in a resting state, this indicates the fish are becoming stressed in the warmer temperatures without exertion. They analyzed their results using a mathematical model. What would the results look like if the results were analyzed using standard statistical analysis for each temperature group? Further they presented temperature ranges from 16.4 °C to 25°C and 17.8°C to 24.6°C, suggesting the higher temperatures are protective for basic survival. This leads to the question, do the authors agree that the 16.4°C and 17.8°C temperature levels (i.e. lower end of range) to be a more protective temperature at a chronic population exposure level to provide optimal reproductive success and recruitment rather than the higher temperature's the author are advocating? It’s vitally important to remember that just because a fish or a fish population survives at a certain temperature; it does not automatically mean that the fish or the fish population thrives at the same temperature range. The ability to “thrive”, carries with it the ability to

The commenter’s statement “thus if their RMR increased with temperature in a resting state, this indicates the fish are becoming stressed in the warmer temperatures without exertion” is simply and fundamentally incorrect. We know of no theoretical reasoning or literature to support such a claim. Arrhenius in the 1920’s showed that all rate functions, including many biological ones, increase with an exponent of 2-3. To suggest otherwise reveals a fundamental lack of understanding about how temperature affects ectotherms. Why does RMR go up with increasing temperature? It has to do with simple laws of thermodynamics – fish are ectotherms, their body reflects water temperature. As fish/molecules warm up, they collide more frequently. Biochemical rates, such as ATP turnover, increase. Thus RMR increases. The ‘amount’ or how temperature sensitive this process is varies across species and reflects variation in biochemistry (I could go on a long tangent here about protein evolution etc.) and we express this temperature sensitivity with calculated temperature quotients, or Q10s. This is an expression for how much a rate (like a metabolic rate) changes with every 10C change in temperature. Ecologically relevant Q10s are usually between 1.5 and 3 in fishes. The lower the Q10, the less temperature sensitive a species is and the less MR changes as temperature changes.

Appendices



successfully grow and reproduce at sufficient levels that keep the both the individual fish, and the fish population, in good condition (i.e. adequate reproductive viability). The authors further state, "Thus, the maintenance of AAS [Absolute Aerobic Scope] across nearly the entire test temperature range clearly shows that the Tuolumne River O. mykiss population has a broad range of thermal performance". Isn't this case for all vertebrates? The authors further state, "Indeed, the AAS of the Tuolumne River O. mykiss population was atypical when compared with cold-adjusted, O. mykiss from the Pacific Northwest, whose thermal performance optimum is reported as 18°C" (USEPA 2003). What exactly is meant by "atypical"? What is meant by "cold-adjusted" fish from the Pacific Northwest when all salmonids are cold water fish that evolved in cold waters that originated from snow melt and ground water seepage into the river systems? The reference to the USEPA (2003) 18°C as a thermal performance optimum is incorrect. The USEPA (2003) report did not discuss thermal performance, but rather concentrated developing sub-lethal chronic population criteria to improve reproductive success and recruitment to reverse a declining population trend. It is inappropriate, and therefore not scientifically valid, to compare acute individual results to chronic population criteria. The last sentence suggesting the upper thermal performance is above 25°C is pure speculation on part of the authors and should be deleted.

Another incorrect interpretation of increasing RMR with increasing temperature is that when you see an increase in MR, it indicates stress. If a fish were stressed with acute warming the increase in oxygen uptake would have an even higher exponent. However, because MMR does not similarly increase with stress, AAS must then decrease with acute warming due to the following equation; AAS = [MMR – (RMR +stress)]. Therefore the fact that AAS was maintained across temperature despite an increase in RMR with temperature argues AGAINST the very claim the reviewer is making. Interestingly, fish in a variety of situations can behaviorally and deliberately seek out warm temperatures in order to avoid the ‘dampening’ effects of cool temperatures on activities such as growth. Thus seeking warm water is not ‘stressful’. For example, if you want to get big, and if you have access to lots of food, you might select warm temperatures to process food at a faster rate, smoltify sooner etc. The perspective that cold is always ‘less stressful’ than warm is pervasive throughout the CDFW comments, and is not ecologically relevant and has no biological basis. Fish have evolved with a physiology suited to historic thermal conditions - not the ones imposed by CDFW, EPA, and other regulators. This is called Natural Selection. While Darwin and others advanced this idea centuries ago, biologist are only now beginning to appreciate natural selection at a mechanistic level. To suggest otherwise does not

Appendices



acknowledge the seasonal/temporal fluctuations in temperature that are part of a fish's life history! Issues regarding the casual use of ‘stress’, the 7DADM as a chronic criteria, and misunderstandings about what is meant by local adjustment have been dealt with in the clarifier statement at the beginning of this attachment and in responses above. The question put forward after “18°C” (USEPA 2003)” is a rhetorical question since the reviewer provided the results of his “standard statistical analysis” later in his review. The statistical questions are dealt with below in the methods/results. Related to this comment, it is incorrect and purposely misleading to suggest we are advocating for higher temperatures, we are reporting on the results of our study which show that these fish have the capacity to conduct various energetically demanding tasks at temperatures above 18°C. The report discussion is not intended as the basis for changing the EPA (2003) 7DADM recommendations. Instead, our work simply suggests that the current value of 18°C lacks merit for the current O. mykiss population found in the lower Tuolumne River. As a minimum, 18°C as the 7DADM value is a very conservative upper thermal limit based on the results of the current study. Related to the question about thermal response of “all vertebrates” the short answer is “no”. Some vertebrates can have a much narrower range of temperatures where thermal performance, as indexed by AAS, is much narrower (e.g.

Appendices



Fraser River sockeye salmon). This is another point where fundamental misunderstanding of the thermal performance literature for fishes has apparently led this reviewer down a confused path. What the study and the data demonstrate is a relatively flat curve, which is consistent with capacity being temperature INSENSITIVE and acclimation/adaptation to a rather wide thermal range. Our use of “atypical” means different from the O. mykiss which are the basis for EPA’s 18°C criteria. The reviewer clearly believes that all salmonids are genetically programmed for cold water and that there has not been sufficient time for any to become locally adjusted (warm or cold). Scientific evidence indicates that salmonids from different locations within the Pacific Northwest and Canada have different optimal temperatures for day to day existence, migration and feeding in freshwater environments (Parsons 2011; plus other references).

CDFW-8 (p. 4)

Executive Summary, page ii, first paragraph. What do the authors mean when indicating that the fish are locally adjusted? The fish are blocked by a series of dams, preventing them to migrate upstream to cooling temperatures, so they have no choice but to live in a warmer environmental regime. The authors also stated they lost 1 of 4 fish acutely exposed to 25°C. Do the authors agree that 25% fish exposed to 25°C would die, especially if they are chronically exposed to this and higher temperatures?

The explanation of what was meant by “locally adjusted” is provided above. No, the authors do not agree that “that 25% of fish exposed to 25°C would die, especially if they are chronically exposed to this and higher temperatures”. Our research did not attempt to answer this question, and it would be inappropriate to try to do so. A different type of study with more fish tested at higher temperatures and more sensitive mortality endpoints would be required to answer this question. What we did show instead was that if fish were at 25°C and swum to exhaustion, 25% died. This observation is very different from the assertion regarding survival, which was not

Appendices



measured. Of course, “survival” is time-dependent. A CTmax measurement is used by some to measure “survival, but this is a matter of a few minutes. Even the fish that died in our experiments at 25°C lasted longer than a few minutes! Thus, the debate around “survival” and temperature has a long and unresolved history, and thermal resistance and tolerance are better terms at a mechanistic level. This is part of the reason why modern day physiologists measure AAS to assess thermal performance. The study was never intended to define specific chronic thermal exposure limits. We remind the reviewer that our work simply suggests that the current value of 18°C lacks merit for the current O. mykiss population found in the lower Tuolumne River. At a minimum, 18°C as the 7DADM value is a very conservative upper thermal limit based on the results of the current study. It is unclear to us what is unclear about this very specific articulation of the study goal and main finding.


Executive Summary, page ii, second paragraph. The authors state, "The conclusion of the study is that the thermal range over which the Tuolumne River O. mykiss population can maintain a 95% of peak aerobic activity from 17.8°C to 26.6°C". How long can these fish withstand this activity? In the last sentence they state that "Finally, based on a video analysis of the swimming activity of O. mykiss in the Tuolumne River, fish at ambient water temperatures were predicted to have excess aerobic capacity well beyond that needed to swim and maintain station against the river current in their usual habitat". However, don't all vertebrates have

Some of the fish could maintain this level of activity for hours in the swim tunnel but in the wild most of their lives occur at much lower activity levels and peak activity only occurs for a few seconds to feed or avoid predators. Also, please see clarifying document at the beginning of this attachment where the issue of energy allocation is explained and the value of understanding AAS capacity as a comparative metric is restated. However, no animal ever lives for prolonged periods near its maximum AAS. Therefore, we agree with the contention. Jared Diamond for example suggested that maximum

Appendices



excess aerobic capacity to survive and meet the basic needs of survival; how are these trout any different from any other living creature? Just because a fish can survive a short duration elevated temperature exposure event (i.e. minutes) does not mean that it can withstand the same elevated temperature for a long exposure event (i.e. days, weeks, and/or months). A human analogy helps us understand key physiological concepts and keep them separate. For example, an Olympic marathon runner can run 26.2 miles in approximately two hours; however, this same runner cannot maintain the same pace for days, weeks, and months. The point here is that the Olympic runner is training for an acute event but in so doing he/she is not enabling him/herself to maintain an acute pace over a chronic period of time (days, weeks, and months). The ability of fish to survive an acute event is not indicative of a fish's ability to survive a chronic event. As was stated above, acute tolerance is always higher than chronic tolerance. USEPA set chronic criteria while the authors of this report conducted an acute study. At best, this study's results may be used to inform development of acute level criteria (i.e. temperature tolerance over short duration) but is does not translate to predicting a chronic level criterion (i.e. temperature tolerance over long durations).

sustained performance in lactating mammals was limited by food movement across the gut; high endurance athletes and lumberjacks appear to have similar problems. Biologists who more broadly measure daily energy expenditures in wild animals rarely find that metabolic rate is on average 2X basal rates. Thus, the finding that Tuolumne River O. mykiss have a FAS of >2 for much of the thermal range we studied must have impressed this reviewer. Yes, each vertebrate species will have excess aerobic capacity to perform and survive at some range of temperatures. The issue we specifically address in this Report is “what is this range of temperatures”. There are obviously lots of differences between Tuolumne River trout and other living creatures, but the only difference relevant in this study is the optimum temperature range for Tuolumne River O. mykiss compared to that for other populations of O. mykiss. Tuolumne River O. mykiss have been observed living and feeding in a river which has higher water temperatures than most other O. mykiss populations. See our discussion of comparative context for help in understanding this point. With regard to the human analogy paragraph: This comment confuses several important concepts that we have explained above and in the clarifying document at the beginning of the attachment. The first part is about how metabolic energy is allocated and we’ve responded to this already. The next bit introduces acute and chronic exposures, which we’ve addressed above as well.

Appendices



CDFW-10 (p. 5)

Executive Summary, page ii, last paragraph. The USEPA (2003) criterion is not an upper performance level for fish. The authors are comparing acute results to a chronic value, an individual result to a population criteria, and survival to reproductive success and recruitment, which are all inappropriate comparisons. The authors need to conduct the same test in other rainbow trout stocks throughout the Pacific Northwest to make a similar comparison to this study before rendering a conclusion that the Tuolumne River rainbow trout have evolved higher population acute water temperature tolerance. The authors recommend " ... we recommend that a conservative upper performance limit of 22°C, instead of 18°C, be used to determine a 7-Day Average of the Daily Maximum (7DADM) value". However, for cold water fish, such as trout, it would be more appropriate, conservatively speaking, to use the lower water temperatures values (17.8°C) the authors presented in their study. Their comparison to the redband trout is also inappropriate because the redband trout evolved under a totally different set of environmental conditions compared to coastal rainbow trout/steelhead. Coastal rainbow trout evolved across thousands of years in river systems that originate in high mountain elevations and connect to the Pacific Ocean. Today's rainbow trout have been exposed to river systems, blocked by dams for less than 100 years, which is insufficient on the evolutionary scale to adapt to today's river water conditions.

We agree that USEPA 2003 criterion was not the upper performance level, but it was the optimal temperature for peak growth. See explanation above, regarding that fact that our experiment was measuring the optimal temperature range for Tuolumne River O. mykiss, not acute CTmax temperatures for these fish. If there is no peak for ASS we must then talk about a thermal range for peak AAS, which is what we do. The reviewer repeatedly returns to the idea that trout are a cold-water species that cannot adapt to warm conditions. We agree that most trout populations are post-glacial invaders, but there is a groundswell of evidence that indicates exceptions to the rule. Red band trout are a documented exception. So are the hatchery-selected rainbow trout in Western Australia and Japan. We believe Tuolumne River O. mykiss are another exception based on the data presented in our Report. Please see previous response to acute/chronic criteria. The reviewer appears to be very certain that 100 years of exposure to higher water temperatures in not sufficient for rainbow trout to become locally adjusted to higher temperatures than other rainbow trout populations. In the report, we thoroughly review the published literature that addresses thermal adaptation among rainbow trout populations and that demonstrates supports for local thermal adaptation. The reviewer cannot be correct with their assertion, as it has taken far shorter for this to occur. Moreover, geneticists are increasingly of the belief that Gene

Appendices



X Environment effects can take over in about 7 generations and be evident in as little as 2 generations. It is also not clear why the reviewer is focused on potential adaptation happening only over the last 100 years in California in response to dams. There are many natural systems in California (pre-dam) where fish would have encountered warm temperatures that would be comparatively warmer than northern latitudes. In drought years, trout can be trapped in shrinking ponds that get quite warm. Some survive and the survivors add resilience to the population. Please look into some of the portfolio effect literature. The opinion presented here by the reviewer is only one perspective. To think that we have been imposing artificial high temperature selection on California fish over the last 100 years is incomplete and quite likely incorrect. An argument could be made that constant, year-round cold-water access for fish immediately below dam is ‘unnatural’ selection and could contribute to the loss of high temperature resilience by dampening selective high temperature signals that would have, historically, occurred naturally.

CDFW-11 (p. 5)

Introduction, page 1, first paragraph. The authors' state, "However, cooler river temperatures are associated with cloud cover and over night [sic], and deeper ponds in the river do show some thermal stratification". Did the authors document the daily temperature difference during the hot summers, and identify and document any cool refugia or deep pools locations and measure water temperatures?

It is well known from the literature, human behavior and animal behavior that air temperature cools with cloud cover or at night. Groundwater seeps into rivers also provide cool refugia. To argue otherwise is folly. Indeed, the whole idea behind a 7DADM is that temperature fluctuates overnight and from day to day!!! Extensive studies were performed of water temperatures of the lower Tuolumne River and the reviewer is referred to

Appendices



these studies that have been previously provided to CDFW as part of the relicensing process.


Introduction, Page 1, second paragraph. The location in river miles was discussed as to where rainbow trout are commonly found with temperatures ranging from 11°C to 28°C. This is true; however, these fish have no other choice but to live under these environmental conditions because their natural migratory route to cooler high elevation waters is blocked by dams. If a fish can survive under a set of environmental (i.e. acute and chronic) conditions, including "thriving" (i.e. reproductive success over many generations etc.), then this fish has demonstrated that it has the capacity to withstand higher temperatures. However, not knowing the environmental conditions which other fish populations are actually exposed to and not knowing their population viability, the justification for changing temperature criteria based upon other fish stocks is scientifically invalid.

From these comments it seems that the reviewer agrees with our contention that this fish population has a limited and constrained habitat. Also they must be surviving, growing and reproducing in this environment. Therefore, they are likely adapted to the local conditions through natural selection. We are simply showing that this population has an excess aerobic capacity to perform over much of this thermal range. This is an important advance of knowledge, especially since it is not widely shared among other more cold-adapted rainbow trout populations, including those introduced to the midwest of the USA and were used by Hokanson (1977). Please also note: The report discussion is not intended as the basis for changing the EPA (2003) 7DADM recommendations. Instead, our work simply suggests that the current value of 18°C lacks merit for the current O. mykiss population found in the lower Tuolumne River. Minimally, 18°C as the 7DADM value is a very conservative upper thermal limit based on the results of the current study.

CDFW-13 (p. 6)

The entire [Thermal Tolerance and Thermal Performance] section discusses acute thermal tolerance in relation to survival, but does not present any information about chronic exposures in relation to reproductive success and recruitment to maintain a sustainable population. On page 2, paragraph 1, last sentence, the authors state "Regardless, CT max is always higher than the temperature that a fish can

See explanation above, regarding that fact that our experiment was measuring the optimal temperature range for Tuolumne River O. mykiss, not acute CTmax temperatures or survival for these fish. CTmax measures thermal tolerance; AAS measures capacity. The fact that when a fish is about to die at CTmax can be

Appendices



tolerate for hours to days and certainly higher that the temperature at which a fish can no longer swim aerobically". The CTmax is a lethal temperature, at which point a fish can no longer swim aerobically. The tunnel test conducted by the authors accomplished the same end point where the fish were pushed to exhaustion and could no longer swim aerobically. So how does the tunnel test as presented by the authors differ from CTmax as stated in this paragraph?

higher than when FAS is 2 seems to be a reasonable statement. Please also see previous comments regarding acute and chronic metrics, and how our study relates to population metrics like reproduction and survival.

CDFW-14 (p. 6)

7-day Average of the Daily Maxima (7DADM), page 2, second paragraph, last sentence. The authors state, "Interestingly, by setting the 7DADM criterion for salmon and trout migration as 20°C, rather than 18°C, USEPA (2003) acknowledged that juvenile Pacific Northwest O. mykiss have sufficient aerobic scope for the energetic demands of river migration even at a temperature 2°C above the 7DADM for juvenile growth". However, the authors failed to mention the 20°C migration criteria is conditioned with a provision to restore or provide the natural thermal regime; or to provide or restore cold water refugia. Examples of cold water refugia or natural cool regime would include the confluence of cold tributaries at the main stem river or where groundwater exchanges with the river flow (hyporheic flow) that would provide cold water refugia for fish to escape maximum temperatures. Waters in tributaries for large rivers in the Central Valley have been diverted, eliminating cold water refugia at the confluence of these tributaries and groundwater pumping in the valley has lowered groundwater levels, thus removing natural cool ground water seeps into the

We have removed this statement from the Report to avoid potentially misleading any reader. This is not an important issue to us, but was meant to illustrate that even the blanket 7DADM had exceptions. We note that the reference to the Corbett report is misleading as it did not draw this conclusion.

Appendices



valley's rivers. (Corbett, F., T. Harter, and M. Sneed. 2011. Subsidence due to excessive groundwater withdrawal in the San Joaquin Valley, California. American Geophysical Union. Fall Meeting Abstract #H23H-1397.)

CDFW-15 (p. 7)

Justification and Purpose of the Study, page 4, first paragraph, last sentence. The authors state, "Thus, MR [Metabolic Rate] measurements were used to determine the optimal temperature range for Tuolumne River O. mykiss". Can the authors provide a definition for what they consider "optimal temperature range" and differentiate an acute and chronic optima range? Do the authors consider the hottest temperature as optimal or would a cold water fish be in excellent condition at a lower temperature from a chronic exposure perspective?

Please carefully review Figure 1, including the legend. The requested information is stated clearly there. We clearly do not measure chronic temperatures so the distinction is meaningless for this Report. We clearly do not consider the hottest test temperature as optimal and that is also clear from Figure 1.

CDFW-16 (p. 7)

Justification and Purpose of the Study, page 5, the first paragraph describes the "aquatic treadmill" similar to Parsons (2011) and Figure 1 that is presented on page 33 in this Study report. The peak Topt in Figure 1 appears to be the maximum acute temperature (Tmax) at the peak of maximum oxygen consumption and not necessarily an optimal temperature. From the peak temperature to higher temperatures, oxygen consumption decreases, suggesting the fish is exhausted and no longer capable of absorbing oxygen similar to what occurs in hyperventilation with humans. It is vitally important to remember that water at higher temperatures have lower oxygen concentrations, which is noteworthy because oxygen crosses the cellular membrane via a concentration gradient. Thus, lower oxygen

This appears to be one of the key sources of the reviewer’s confusion on the purpose of our study. The maximum acute temperature (Tcrit in Figure 1) is not at the peak of maximum oxygen consumption. To be clear on the definitions that we have used, the temperature at which peak ASS occurs is DEFINED as Topt. This is the accepted definition. Tcrit would be when ASS fell to 0, but we never saw this with the present experiments. Furthermore, the statistics argue that there is no specific peak AAS as such, only a large thermal range over which there is no statistically significant change in AAS. Thus, there is not

Appendices



concentrations in the water decrease the concentration gradient forcing the fish to use more energy to pull oxygen across their gill membrane, similar to hyperventilation of a human at the 8,000-foot elevation where the oxygen concentration is lower than that which occurs at lower elevations. Clark et al. (2013) Figure 1 B (page 2772) demonstrates that Topt is midway up the aerobic scope and not at the peak of the slope. They further state "ToptAS provides little insight into the preferred temperature or performance of aquatic ectotherms, but rather aerobic scope continues to increase until temperatures approaches lethal levels, beyond which aerobic scope declines rapidly as death ensues." We agree with Clark et al. (2013) that the curves peak should be considered a Tmax, not a Topt.

a Topt as such, only a range over which a peak AAS is maintained. We are not therefore dealing with a mountain peak, but instead a prairie plateau! In this regard, the broad thermal performance of AAS more closely resembles the eurythermal killifish and goldfish. As a former postdoctoral supervisor of Tim Clark and co-author, Dr. Farrell is very familiar with his research and publications. We agree that AAS only tells us what capacity exists. Thermal preference, as pointed out by Clark et al. 2013, is a completely separate issue and should not be confused with Topt. However, if a fish wants to maximize the capacity to perform activities then it should choose Topt. The fish may or may not choose or prefer this temperature, but then there can be situations when they cannot – e.g. overwinter in the Tuolumne River, when the fish are likely to acclimate to cooler seasonal temperatures. Clearly a steelhead trout in the Pacific Ocean could not prefer a Topt of say 18°C for growth (as dictated by the 7DADM simply because such conditions do not exist within their known habitat range at sea. They are found at much lower temperature.


Justification and Purpose of the Study, page 5, second paragraph, last sentence. The authors state, "Specifically, the temperature indices and the shape of the aerobic scope curve derived in the present study can also be compared with those of other O. mykiss populations and with the EPA (2003) recommendations". It is inappropriate to compare results from an acute stress test conducted for basic

See explanation above, regarding that fact that our experiment was measuring the optimal temperature range for Tuolumne River O. mykiss, not acute CTmax temperatures for these fish. We fundamentally disagree with the reviewer on this point and again refer to how we are interpreting our work and how it relates to the EPA criteria. The report is not intended as

Appendices



survival needs and then make inferences to a population needing protection at the chronic criterion level. Again, acute level does not equate to chronic level when it comes to conducting tests and/or developing protective criteria. The USEPA criteria are chronic not acute; therefore, any reference to USEPA criteria in this report for purposes of changing chronic criteria is unfounded and is therefore not scientifically valid.

the basis for changing the EPA (2003) 7DADM recommendations. Instead, our work simply suggests that the current value of 18°C lacks merit for the current O. mykiss population found in the lower Tuolumne River. Minimally, 18°C as the 7DADM value is a very conservative upper thermal limit based on the results of the current study. Given the wealth of published literature on the ecological relevance of AAS measures, we think it rather inappropriate for a comment like ‘not scientifically valid’ to appear in this forum.


Justification and Purpose of the Study, page 5, last paragraph. This paragraph summarizes Fry (1947) as presented in Parsons 2011. The "tunnel" experiment is an acute test that measures acclimation rather than adaptation. Central Valley salmonids evolved across thousands of generations to adapt to their living environment before the construction of dams. Fish that exist today have not evolved under today's environmental conditions because the time period has been too short for adaptation. Yes, O. mykiss can acclimate on an acute basis, but cannot adapt on a chronic basis in the less than 140 years since the construction of dams which blocked their historic spawning grounds. Similar to Parson (2011) description, resistance or adaptation is a result of the evolutionary process that takes generations to develop and cause a genetic change across those generations in a population (Guthrie 1980). Tolerance or acclimation is a result of an individual, or

We never make any claim that this fish population is adapted. We simply say that the evidence is in support of local adaption. We do not want to sound like a broken record, as almost all of these issues are variously and repeatedly dealt with in the responses above and in the clarifying document. There are a couple of points we must emphasize. Our experiment was measuring the optimal temperature range for Tuolumne River O. mykiss, not acute CTmax temperatures for these fish. If O. mykiss can’t acclimate on a chronic basis to warmer water, we should not find them living and feeding in these warm water locations where they are observed in the Tuolumne River. Lastly, populations are made up of individuals. If the individual does not have the AAS to perform, the population

Appendices



a group of individuals, repeated exposure across the life of the individual that causes a physiological change. Individual based temperature exposure tolerance does not expand to all individuals in the population, but population based exposure adaptation transfers to all individuals within the population. Population thresholds are designed to protect a population; whereas, an individual threshold is designed to protect an individual or small group of individuals. A population threshold will have minimal health effects for all the individuals, including the most weak, in that population (USEPA 2989; Air RISK). Therefore, the population exposure threshold tends to be lower in value (i.e. more restrictive) than the individual exposure threshold. In summary, population thresholds are always less than an individual threshold and chronic thresholds are always less than acute thresholds. Thus, the reported fish water temperature experiment address individual level, but have limited usefulness as a basis for a full understanding of resistance or adaptation at the population level. As such, the tunnel stress test provides great information about tolerance and acclimation at the individual level, but is inappropriate to extrapolate the results to adaptation for chronic population exposure criteria.

will cease to exist. Furthermore, natural selection acts on individuals and the results are reflected in populations. Therefore, if we do not understand the effects of temperature at the level of individuals, we have no hope of properly understanding the population effects.

CDFW-19 (p. 9)

Predictions Derived from EPA (2003), page 6. The authors proposed predictions based the USEPA (2003) criteria are irrelevant because the USEPA (2003) criteria were not based on an acute stress test. Is data presented in Table 1 based on acute or chronic tests?

We agree with the reviewer on what the USEPA 2003 report does and does not contain. This does not mean that we cannot make predictions, which is all we do. The report discussion is not intended as the basis for

Appendices



The USEPA (2003) 18°C criterion is not based on maximum metabolic rate (MMR) acute test as presented in Figure 1, but is a chronic criterion which is lower than acute criterion. The USEPA (2003) never stated an AAS Topt metric, nor discussed this study design, to develop a chronic population criterion.

changing the EPA (2003) 7DADM recommendations. Instead, our work simply suggests that the current value of 18°C lacks merit for the current O. mykiss population found in the lower Tuolumne River. Minimally, 18°C as the 7DADM value is a very conservative upper thermal limit based on the results of the current study. This report is not intended to preempt consultation with EPA. We strongly believe that the new data collected here are a firm basis for opening such a dialogue about site-specific temperature criteria in general as well as for the Tuolumne River O. mykiss.

CDFW-20 (p. 9)

Alternative Predictions of Thermal Adjustment, page 6. On what are the predictions based? Again, this study design is an acute stress test. It is well known that O. mykiss can survive in temperature above 18°C, but the study design does not answer the questions as to what is the chronic population threshold for reproductive success and recruitment to maintain sustainable populations across many future generations. The study design also does not address how well the O. mykiss immune system functions to ward off disease or how well a cold water fish can escape a warm water predator, especially when the water temperature are in the optimal range for the warm water predator. This study design can measure individual cold water fish short sprint energy to avoid a predator, but does not indicate how long a cold water fish can escape in a predatory warm water fish optimal temperature zone.

We agree with the reviewer on what the USEPA 2003 report does and does not contain. This does not mean that we cannot make alternative predictions, which is all we do. The reviewer continues to blindly refer to this population of rainbow trout as a cold-water species, when they clearly live in river temperatures reaching 24°C. The report discussion is not intended as the basis for changing the EPA (2003) 7DADM recommendations. Instead, our work simply suggests that the current value of 18°C lacks merit for the current O. mykiss population found in the lower Tuolumne River. Minimally, 18°C as the 7DADM value is a very conservative upper thermal limit based on the results of the current study.

CDFW-21 (p. 9)

Fish Collection, Transport, and Handling, pages 8 to 9. Most of the study fish were caught in the upper coolest

The choice to collect fish from cool reaches had nothing to do with the distribution of fish in the river with respect to

Appendices



reaches of the river. However, if these fish are adjusted to warm temperatures, why were they present in the coolest waters of the river? The fact that most of the fish were found and captured in the coolest waters of the river is indicative that, at the population level, O. mykiss in the lower Tuolumne River are seeking cooler water to reside in even though warmer water is available to them.

temperature. Jumping to this conclusion is incorrect, unsupported, and misleading. The distribution of O. mykiss within the Tuolumne River is affected by many factors, only one of which is temperature, e.g there may be more predators downstream. Prior studies on the Tuolumne River have documented O. mykiss in warmer water. All of these studies have been provided to CDFW and we refer the reviewer to the many submittals on this subject. The decision to collect fish from relatively cold reaches was twofold. First, by collecting fish from cooler reaches, they were more likely to have a relatively cool thermal history (acclimatization) as compared to fish from warmer reaches. Because thermal history has a positive relationship with performance (i.e. if fish are acclimated to warmer temperatures, performance at warmer temperatures improves), testing cold-acclimatized fish should lead us to the most conservative AAS curve, with respect to temperature, that we could obtain. Certainly, when data are to be used for thermal criterion discussion, conservatism is desired for fish protection. Also it makes for easier comparison with existing data on rainbow trout. Secondly, collecting fish from cooler areas minimized our chances of capture-related mortalities (due to rapid temperature change during capture, release, or transport) during the peak of summer. This was an unnecessary risk that we thought best to avoid so that our study would not be shut down early. Our experiment showed that there is a wide range of temperatures where Tuolumne River O. mykiss have the

Appendices



aerobic scope to live and thrive. O. mykiss are found in the Tuolumne River at the range of water temperatures tested in our experiment. The distribution of O. mykiss within the Tuolumne River is affected by numerous factors only one of which is temperature.

CDFW-22 (p. 10)

Experimental Protocols, page 11, last paragraph. The authors state, "Water velocity was then increased in increments of 3 to 6 cm s-1 every 20 min until the fish failed to swim continuously". Is this an acceptable fisheries technique to allow an animal to work to the point of complete exhaustion? Would it be better to do a timed test by stopping the test before the fish is completely exhausted?

This experimental design has been used in numerous studies and approved by university research protocols. Please refer to the clarifying document and the cited special issue on methodology. Importantly, this comment reveals a fundamental misunderstanding related to why we are using swimming tests and exhaustion. To properly measure AAS, we need a method to estimate MMR and swimming to exhaustion happens to be one way to do this. Swimming fish for an extended period of time and a submaximal swimming velocity would not elicit maximum metabolic rates. A human example might help. Would a jogger out for a 20 min casual, timed run reach MMR? No. Would a runner chased to exhaustion exhibit MMR near the endpoint, yes. Even the cited and supported work by Clark et al. 2013 uses this methodology and Clark and Norin, 2016 used the approach used by us. It seems that the reviewer is adopting double standards.

CDFW-23 (p. 10)

Experimental Protocols, page 12, third paragraph. The authors state, "Approximately 50% of the wild fish did not respond to the critical swimming velocity protocol but instead used their caudal fin to prop themselves on the downstream screen to avoid swimming". Is this a sign the fish were already stressed before the

It is unclear what the reviewer considers stress. See previous comments regarding the pitfalls of using this term casually. This was not a sign of ‘stress’ and there was no discernable pattern between tail propping with test temperature. Making the speculation that this behavior had something to do with

Appendices



experimentation started and possibly a result of too warm temperatures to begin with?

elevated test temperatures is incorrect. It is common for some fish to be ‘non participants’ in tests like these. We see this in all species that I’m familiar with and the literature provides many examples. These fish are wild and a swim tunnel setting is novel. Some individuals require different motivations (stepwise velocity increases, bursting velocities etc.) to help to orient them to the current and elicit sustained swimming. In salmonids, tail propping is often seen at low velocities to save energy rather than swimming, but as velocities continue to increase, salmon will then begin to swim. Once they start swimming, they almost always perform to exhaustion. This fish are built for continuous swimming.

CDFW-24 (p. 10)

Data Quality Control, Model Selection and Analyses, page 13, last paragraph. The authors state, "Routine metabolic rate quality control (QC) was performed by visually inspecting over night [sic] video recordings for fish activity" and that "data from any fish showing consistent activity over night [sic] was discarded". Why were the data discarded? Was the fish activity a sign of stress before the experiments started? In addition the authors state, "For fish exhibiting intense agitation, the swimming MMR was used as overall MMR." Four of these 'non-agitated' fish (W2, W13, W14, and W15) were disgarded due to failure of MR to increase incrementally; despite continuous station-holding swimming with tunnel velocity increases of more than 15 cm s-1". Were these fish already stressed? How does inclusion of these data influence study results? It is important that data not be "selected" in order to bias study results. Scientific integrity requires that data not

It is unclear what the reviewer considers stress. See previous comments regarding the pitfalls of using this term casually. Please review our definition of RMR and reread comments on the effect of stress on RMR and AAS in the previous comments. It is clear that metabolic measure of active fish is NOT the physiological state representative of RMR. Active fish have metabolic rates somewhere between RMR and MMR and are not we were measuring in order to calculate AAS. We are keenly aware of the fact that rigorous science includes all valid data points and rigorous scientists do not ‘throw out’ valid data. All of our methods are consistent with published methods, see the previously referenced special issue of Journal of Fish Biology. Our rationale is clear and justified and supported by the scientific (published) literature, which we’ve cited. The insinuation that there is even a hint of a

Appendices



be thrown out for invalid reasons, including if the results cannot be explained or if they are different than expected.

scientific integrity issue of the co-authors of this study is completely out of line, inappropriate, and disappointing. Certainly, there is a less combative way to query the inclusion or exclusion of specific data points.

CDFW-25 (pp. 10 and 11)

Results, Page 15, Number 1, third paragraph, second sentence. The authors state, "They state that Routine Metabolic Rate (RMR) should increase exponentially until the test temperature approaches the upper thermal tolerance limit for O. mykiss, which according to published CTmas values is 26°C to 32°C (see Table 1 )". Who is "they"? If "they" is the USEPA, this is an incorrect statement because the USEPA did not include RMR studies in their review. Myrick and Cech's Table 1 had significant less food consumption and decreased growth rates and increased mortality in their 25°C test fish compared to their 10°C, 14°C, and 19°C exposed fish. In their Table 2 results, fish consumed significantly less oxygen at 25°C compared to fish exposed to 10°C, 14 °C, and 19°C. O. mykiss can survive in acute warm temperatures as demonstrated by the authors, but cold water fish still need cold water refugia sometime during the day. According to Myrick and Cech (2000) there is very little thermal difference between fish stocks per their comparison of other research studies (see Table 5) under similar experimental conditions.

“They” is referring to the predictions not to any group or report. The sentence should have started with “These predictions” not “They”.

CDFW-26 (p. 11)

Results, page 15, Number 2. The authors state, "These results for MMR are inconsistent with our prediction #2 derived from EPA (2003) criteria where MMR was

Chronic/acute criteria are dealt with in CDFW-1 and in the clarifying document.

Appendices



expected to peak near to 18°C". This statement is irrelevant because the authors are comparing chronic population criteria to acute individual results. Again, chronic and population thresholds are always less than acute and individual thresholds.

CDFW-27 (pp. 11 and 12)

Results, page 16, Number 3, third paragraph. The authors state, "These results for AAS are inconsistent with our prediction #3 based on EPA (2003) criteria, but are consistent with our alternative prediction #3 that the Tuolumne River population of O. mykiss is locally adjusted by having Topt for AAS that is greater than 18°C i.e., 21.2°C." This statement is irrelevant because the authors are comparing a chronic population criterion to acute individual results. Again, chronic and population thresholds are always less than acute and individual thresholds.


CDFW-28 (p. 12)

Results, page 16, Number 4, last sentence. The authors state, "The numerical 95% peak AAS could be maintained from 17.8°C to 24.6°C, which is a more conservative thermal range for Topt". However, based on the authors results, and because rainbow trout are a cold water fish, a true conservative thermal range would be from 16.4°C to 17.8°C.

How does this reviewer know what is “true” for Tuolumne River O. mykiss? The 16.4-17.8 °C range is more conservative but not necessarily the “true conservative thermal range”. Why not pick, 17.8 to 17.9 to be more conservative still. Such selection is arbitrary and has no statistical basis. Statistically AAS does not change over the range of temperatures stated. Therefore we adopt a scientific rigor that the reviewer does not appear to appreciate. Furthermore, the reviewer has repeatedly ignored the results of our tests of Tuolumne River O. mykiss and based his whole review on the assumption that our study fish are no different from other O. mykiss populations.

Appendices



Again the reviewer ignores the data by calling this population a cold-water species when data show that they perform equally well at temperatures in excess of 20°C. It’s also important to mention that ‘coldwater fish’ is not at all a precise term. What is cold, what is warm? Modern thermal ecology literature suggests that some Om populations are simply not as ‘cold water’ as previously thought.

CDFW-29 (p. 12)

Results, pages 16 to 17, Number 5. Same comment as for comparing an acute stress test results to a chronic population criterion. The author state, "Indeed, all individual fish tested up to 23°C has a FAS [Factorial Aerobic Scope] value >2, with only 4 out of 14 fish tested at 23°C, 24°C, and 25°C having a FAS value <2." A chronic population threshold is formulated to protect the weakest individuals in a population, so by using a lower criterion these 4 weaker fish should have better physiological function and survival.


CDFW-30 (p. 12)

Results, page 17, Number 7. The authors state, "Two fish tested at 25°C regurgitated rather large meals of aquatic invertebrates during the recovery from the swim test, and one of these fish died abruptly during the recovery period". Since, fish were exposed to an exhaustive state, this causes us to question whether or not this an appropriate testing technique where the test has to force an animal to complete exhaustion, especially for a group of fish that may be already stressed due to having to live in environmental conditions of altered flows and habitats that they did not evolve with.

Same comment on casual use of ‘stress’ and ‘stressed out’. What, scientifically, do these qualitative judgment statements mean? They have no value. We have already explained above why we use exhaustion to elicit MMR. We either measure it or we do not. We are dealing with wild fish and cannot hold them outside of the river to ensure a full post-prandial state. We simply cannot believe that the reviewer holds the following premise “a group of fish that may be already stressed due to having to live in environmental conditions of altered flows and habitats that they did not evolve with.” How on earth would the reviewer reach such a speculative

Appendices



conclusion without data to back it up? The amount of supposition and lack of scientific data to back such assertions is a total shock.

CDFW-31 (p. 12)

Discussion, Data Quality, page 18, first paragraph. This section provides a brief summary of the results and comparison to other aerobic studies. The Department completed an analysis of variance as presented in the following Table 1 using the Study's data presented in Appendix 4. Temperatures at and below 18°C were significantly different for RMR, MMR, and FAS compared to the highest temperatures at and above 22°C. For RMR, which is a fish at rest, is this an indication the fish at the warmer temperatures were already stressed before the experiment started?

It is simply not possible to respond to a comment about a re-analysis of the data when no statistical details are given. The statistical test, how the treatment groups were defined, n values, degrees of freedom, alpha level, and p-values are all missing. Note: imprecise use of stress and lack of understanding of the simple effects of temperature on biological rates are repeated once more.

CDFW-32 (p. 13)

There is an inverse relationship between water temperature and oxygen concentration. As temperature increases, oxygen decreases. As such, at the warmer temperature with less oxygen, are the fish stressed to the point they are hyperventilating, thus increasing their metabolism trying to pull in as much oxygen as possible from a low oxygen environment.

The movement of oxygen from water into the blood of a fish is governed by the partial pressure of oxygen in the water and not by oxygen concentration. Therefore the reviewer does not understand the basic principles of oxygen movement into fish by suggesting that the decrease in oxygen concentration with temperature triggers hyperventilation. Fish ventilation responds to oxygen partial pressure not concentration in the water. If the reviewer thinks otherwise they have been misled. Furthermore, oxygen concentration decreased by only 10% per 10°C. Therefore the decrease is around 15% for the entire range of test temperatures. Despite this minor change, the fish increase MMR and maintain AAS, so this is a non-issue. The idea that these fish are stressed has been repeatedly dealt with above.

Appendices



CDFW-33 (pp. 13 and 14)

The Department also graphed the mean results for each temperature as presented in Appendix 4. Note at 22°C for RMR, AAS, and FAS and at 21°C for MMR there is a sudden change in the slope of the graph. Does this change in slope indicate there is a sudden change in the physiological function of the fish and a clinical sign that the fish are highly stressed? A highly stressed animal is considered to be in poor condition.

Our response to CDFW-31 applies here too. It is simply not possible to respond to a graph that we do not have. The equation, the R2 and p value are needed. We are not sure of what clinical sign that the reviewer refers to.

CDFW-34 (p. 14)

Discussion, Data Quality, page 19, Protocol Number 2. The authors state, "2 a combination of continuous swimming and short velocity bursts to push fish off of the downstream screen". Was this an indication the fish was already tired and stressed at the beginning of the experiment?

Of course a fish is tired when it is exhausted. That is why it has reach MMR and we stop the experiment. The fish nevertheless were observed to recover quickly.

CDFW-35 (p. 15)

Evidence for Local Thermal Adjustment, page 20, first paragraph. The authors state, "Our predictions based on EPA (2003), as listed above, assumed that the Tuolumne River O. mykiss population would perform similarly to Pacific Northwest O. mykiss populations used to set the 7DADM by USEPA (2003)". The predictions based on USEPA are irrelevant because the USEPA did not perform tunnel stress techniques or use such data to develop their chronic population criteria recommendations. The authors recommend using 21.2°C rather than 18°C, but they are comparing an acute/individual result to a chronic/population recommendation. Have the authors considered other techniques to determine what cold water fish, such as O. mykiss can chronically sustain normal/optimal physiological function, including immune function, reproductive success and recruitment, at their

Please see the clarifying document at the beginning of this attachment where we reiterate what our study results reveal and should be used for. That the reviewer suggests “The authors mention these test fish have a wide optimal thermal performance range, but this is true for all living organisms;” suggest a complete lack of understanding of thermal biology. Some Antarctic ice fish die at about 4°C.

Appendices



recommended temperature of 21.2°C? It is well understood that cold water fish can simply survive at warmer temperatures to a point, but what about their entire life cycle needs at the individual and population levels? The authors mention these test fish have a wide optimal thermal performance range, but this is true for all living organisms; what do the authors consider "optimal"?

CDFW-36 (p. 15)

Evidence for Local Thermal Adjustment, page 20, first paragraph, last sentence. The authors also state, "However, given that the CTmax could not be determined in the present work and that MMR increased up to the highest test temperature (25°C), it was impossible to determine the upper thermal limit when MMR collapses, which means that alternate metrics must be used to set the upper thermal limit for the Tuolumne River O. mykiss population". Since the "upper thermal limit" is survival based, can the author's present reproductive success and recruitment base criteria with this type of testing?

Reproductive success and recruitment were not part of this study’s purpose.

CDFW-37 (p. 15)

Evidence for Local Thermal Adjustment, page 20, second paragraph. The authors state, "The present work provides three useful metrics of the optimal temperature range". What is meant by "optimal temperature range"? Topt appears to be more of a temperature maximum (Tmax) than a Topt. A temperature maximum does not necessarily mean it is an optimal temperature. Fry (1947) page 56, Figure 27, does not state the peak of activity as optimal, but refers to the "potential range of activity" and the "scope for activity". Fry further reduces the area of the activity curve by discussing

Please see response to CDFW-15. The thermal performance literature has evolved from the fundamental work of Fry to appreciate that there are many forms of thermal performance curves (not a static performance curve, incorporating all metrics for a particular species) like CDFW reviewer is presenting. These curves are shaped by many factors, including the 5 (or 6 depending on the citation) classifications of factors that Fry delineated in his work. They are also shaped by timescale (acute, acclimatory, adaptive), the fish’s life history, the fish’s

Appendices



"controlling factors". The USEPA (2001) as presented below provides a number of "controlling factors". Did the authors for this study consider controlling factors as described by Fry to adjust their activity curve?

evolutionary thermal history, etc. We do not disagree with the reviewer on the importance of these aspects and nowhere do we discount any of these details. It is interesting that Fry 1947, while an absolutely critical piece of early work, is also not augmented by the reviewer with more recent literature on thermal performance curves, their utility, and their interpretations. We cite several. Please revisit the clarifying document where we clearly articulate how our results articulate with the 7DADM.

CDFW-38 (p. 16)

Evidence for Local Thermal Adjustment, page 20, last paragraph, first sentence. The authors state, "Yet, there were important indications that a small percentage of individuals were taxed at 23-25°C by the thermal testing and intensive swim imposed on them outside of their normal habitat over a 24-h period." In the fourth sentence they further state, "In the present study, the telltale signs were that 4 of 13 individuals [31 %] tested at 23-25°C had a FAS <2." This supports the concept that a chronic population base threshold is to protect the weakest individuals in a population and cannot be formulated by using just one simple acute stress test.

See explanation above, regarding that fact that our experiment was measuring the optimal temperature range for Tuolumne River O. mykiss, not acute CTmax temperatures for these fish. It is incorrect for the reviewer to repeatedly call our test an ‘acute stress test’ for many reasons. As stated above, the optimal temperature range for Tuolumne River O. mykiss suggested by our swim tunnel tests is lower than the temperatures at which 4 fish had FAS values less than 2.

CDFW-39 (p. 16)

Evidence for Local Thermal Adjustment, pages 21 to 22, top line. In the same paragraph, the authors state, "Lastly the only fish mortality occurred in the recovery period (a phenomenon known as 'delayed mortality') after one fish was tested at 25°C". What is the point of mentioning 'delayed mortality'? The end result is one of four fish (25%) died at the highest temperature when forced to swim until completely exhausted.

“Delayed mortality” is just a term used to describe this category of mortality. Certainly, given the repeated interest in the duration of thermal exposure by the reviewer, the importance of expressing whether the death was immediate in response to an acute 25°C exposure or if the response was the result of a more prolonged high temperature exposure can be appreciated.

Appendices



CDFW-40 (pp. 16 and 17)

Ecological Relevance of the Present Findings, page 21, third paragraph. The authors state, "MMR increased with temperature from 13 to 25°C, which would mean that as fish encounter higher temperatures, they have the capacity to perform an activity at a higher absolute rate, i.e., swim faster to capture food or avoid predators, digest meals faster, detoxify chemicals faster, etc.". Are the authors saying rainbow trout are better off at 25°C instead of <19°C? Their interpretation does not make any sense. We agree a fish will have burst of energy no matter what the temperature, however, the question remains how long can they maintain this energy consumption under chronic warm temperatures at 21.2°C? It takes energy to reproduce, how does exposure to chronic warm temperature impact reproduction success and recruitment into the population? Clark et al. (2013), page 2779, stated that there is a range of optimal temperatures for different processes and life histories and these optimal temperatures are different from ToptAS· They used an example for adult pink salmon where a ToptAS is at 21°C, but if reproduction occurred at 21°C would fail because the optimal temperature for spawning is <14°C. They further stated on page 2780, that fish have different physiological functions at different optimal temperatures as presented in their Figure 7B.

No, we are saying that O. mykiss have greater aerobic scope at 21.2°C than at the other temperatures tested. Therefore, they would be better off from a physiological energy perspective at temperatures near 21.2°C than the other temperatures tested. Some of the fish swam for multiple hours at a high rate at 21.2°C and higher temperatures. Please review the clarifying document for our perspectives about how our data do and do not address the 7DADM EPA criteria and how our data relate to energy allocation. Please also note: The report discussion is not intended as the basis for changing the EPA (2003) 7DADM recommendations. Instead, our work simply suggests that the current value of 18°C lacks merit for the current O. mykiss population found in the lower Tuolumne River. Minimally, 18°C as the 7DADM value is a very conservative upper thermal limit based on the results of the current study. The reviewer indicates that he does not think our interpretation makes sense. However, it is equally possible that our interpretations do make sense and we have either not communicated them simply and clearly enough or the reviewer is not understanding the study goals and interpretations. We have tried very hard to restate how are data are or are not relevant to the comments from CDFW.

CDFW-41 (p. 17)

Ecological Relevance of the Present Findings, page 22, first paragraph, third sentence. The authors state "As a result of high temperature, a fish would digest the same

The commenter’s statement regarding the relationship between digestion rate and temperature is based on basic physiology of fishes and reveals a fundamental lack of

Appendices



meal with a similar overall oxygen cost but at a faster rate". This study did not measure how fast fish can digest their food at increasing water temperatures, therefore this statement stating that a fish would digest their food at a faster rate at higher temperatures is an assumption based on speculation. As the authors discussed, this study design measured oxygen demand to demonstrate fish have extra burst energy from a resting state to seek and catch food, but does not include measuring the rate of digestion. All animals digest their food during the resting state, otherwise their digestive tract would cramp-up during high activity.

understanding of the effects of temperature on another physiological rate process in fishes – digestion. (Please read the works of Jobling in the 1980-90’s and Fu in 2000’s. Indeed ecologists have postulated that some fishes actively feed in very warm areas so they can digest more!) We did not study nor use the terminology ‘extra burst energy’ as suggested by the reviewer. These are the reviewer’s words and should not be mistaken for ours. We also note that the last sentence is a bit misapplied in using anthropomorphic words like ‘cramp up’ in an attempt to describe some sort of physiological relationship or process. Regarding the comment that “[a]ll animals digest their food during the resting state”, we assume the reviewer is familiar with the behaviors of pelagic fish. Many swim continuously and eat and swim at the same time. Indeed, filter feeders must swim to feed. We note that the reviewer did not have any comment on the middle paragraph on page 22 of our report, where we provided clear evidence of O. mykiss maintaining their station in the Tuolumne River at 20°C where their metabolic rate (derived from tail beat frequency) was twice their RMR but substantially below their MMR at that temperature. The commenter is disregarding data that we provide while preferring to offer spurious speculation.

CDFW-42 (p. 17)

Ecological Relevance of the Present Findings, page 22, last paragraph. The authors state "Here we did not evaluate the possibility that the Tuolumne River O. mykiss population can thermally acclimate to warmer river temperatures as the summer progresses, due to the

This comment reveals fundamentally incorrect ideas and misunderstandings about what acute versus acclimatory processes are and what time scales are relevant. We kindly ask the reviewer to review the various report documents noting the definitions of these terms. Please also review the

Appendices



available sample of a maximum of 50 individuals and their habitat temperature." Actually the authors did evaluate if Tuolumne River trout can acclimate, because this was an acute stress test designed for that purpose. Up to a limit, animals can acclimate to an acute environmental change, but how do these animals reproduce successfully under chronic environmental changes such as migratory routes being blocked and under different water flow regimes that they did not evolve with?

study design. To evaluate thermal acclimation experimentally, you must control and/or measure the fish’s thermal acclimation history so that it is known. With those data in hand you can then attribute results to the effect of acclimation. Our fish were wild-caught and acclimatized to the river where thermal history was not measured. Thus, we did not test acclimation. The reviewer comments here regarding ‘acute’ are not relevant and the later part of the last sentence regarding migration routes and flows are also not relevant. If permitting had allowed us to keep each fish out of the river for 4 weeks, we would have measured acclimation. So the experiments could be performed with our mobile physiology lab, but was not tested because of the available permit.

CDFW-43 (p. 17)

Conclusions, page 24. As previously stated, the USEPA did not use acute tunnel stress test to evaluate a chronic population criterion. They included a number of factors as part of their evaluation. It is inappropriate to compare results from an acute individual test to a chronic population threshold. Since O. mykiss are a cold water fish, it would be more appropriate and conservative to use their lower range results (16.4 and 17.8°C) to protect this fish, particularly where reproduction success appears to be low because the population has been declining for decades since the dams were constructed (Yoshiyama et al., 2001).

See explanation above, regarding that fact that our experiment was measuring the optimal temperature range for Tuolumne River O. mykiss, not acute CTmax temperatures for these fish. See response to CDFW-1 specific to the repetitive comment regarding chronic/acute criteria. It is nice to see the reviewer finally acknowledge here that the 7DADM incorporates several types of thermal performance data in setting the criteria. One of the strengths of our study is that we have used a state-of-the-art approach (i.e. AAS) to contribute to this understanding. We have clarified how our data relate to the 7DADM in the clarifying document.

Appendices



Note that we do not think it to be appropriate to casually call O. mykiss cold water fish without using more precise and descriptive terminology. The literature has expanded over the last 10-20 years to show that at least some populations of O. mykiss, including those in this study, may not all be equally ‘cold-water’ as previously grouped.

CDFW-44 (p. 18)

Figures, page 33, Figure 1. The Topt appears to be an acute maximum temperature at the peak of maximum oxygen consumption and not necessarily an optimal temperature. From the peak temperature and higher, oxygen consumption decreases, suggesting the fish is exhausted and no longer capable of absorbing oxygen similarly to hyperventilation. See comment above for Page 20, second paragraph.

Addressed above.

CDFW-45 (p. 18)

Figures, page 37, Figure 4. See comment above for Page 20 second paragraph. Per Fry (1947) page 56, Figure 27, did the authors for this study consider controlling factors to adjust their activity curve? For the Factorial Aerobic Scope curve, the peak is approximately 13°C and decreases as temperatures increases. Clark et al. (2013) Figure 6, page 2778, presents a similar Factorial Aerobic Scope curve where the ToptAS is at the peak of the curve representing the lowest temperature at 11°C. Using Clark et al. (2013) concept, the authors Figure 4 peak at 13°C should be considered the Topt for Tuolumne River rainbow trout, not the maximum temperatures.

Addressed above.

CDFW-46 (p. 18)

Figures, page 44, Appendix 1. Were O. mykiss observed, or attempts made to capture fish, between River Mile 39.5 (permit limit location) and River Mile 49? River water temperatures were above 18°C, so it

Not during the 2014 study but O. mykiss were observed in the Tuolumne River below river mile 49 in 2015 and in other years.

Appendices



would be worthwhile information to know if a healthy number of rainbow trout occupied this area. River Mile 48 appears to be below the 21.1°C permit requirement.

CDFW-47 (p. 18)

Figures, page 47, Appendix 2. Fish W43 died. Did this fish die from delayed Capture Myopathy as a result of handling and exposure to high temperatures? Capture Myopathy results in the death of a captured wild animal during or after the animal has been captured and released.

The fish that died after the swim tunnel test was one of two fish that regurgitated their stomach contents after being tested at 25°C. We don’t know the reason for this mortality but it was likely associated with the excess metabolic demand resulting from digesting a full stomach of food combined with the test used to determine MMR using the swim tunnel. The other fish that regurgitated their stomach contents after the swim tunnel test did not die.

CDFW-48 (p. 18)

Figures, page 49, Appendix 4. All the data should be included for peer review, particularly for the fish that were discarded. What would the analysis look like if the discarded fish data was included? According to the Quality Control column the discarded fish were removed because of activity during RMR or no MR increase. Does this indicate the fish were already stressed? Which fish died?

Addressed above. We followed state-of-the-art peer reviewed methods, and will not comment further on the reviewer’s arbitrary assignment of the descriptor ‘stressed’ to any of our study animals without any supporting data.

CDFW-49 (p. 18)

Figures, page 50. Four of 14 fish tested at 23°C, 24°C and 25°C had a FAS< 2. These results of less than 2 at the highest test temperatures indicate these fish were highly stressed at these temperatures.

Addressed above.



APPENDIX 7

HIGH THERMAL TOLERANCE OF A RAINBOW TROUT POPULATION NEAR ITS SOUTHERN RANGE LIMIT SUGGESTS LOCAL

THERMAL ADJUSTMENT

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Volume 4 • 2016 10.1093/conphys/cow057

Research article

High thermal tolerance of a rainbow troutpopulation near its southern range limitsuggests local thermal adjustmentChristine E. Verhille1, Karl K. English2, Dennis E. Cocherell1, Anthony P. Farrell3 and Nann A. Fangue1,*

1Department of Wildlife, Fish and Conservation Biology, University of California Davis, Davis, CA 95616, USA2LGL Limited, Sidney, British Columbia, Canada V8L 3Y83Department of Zoology and Faculty of Land and Food Systems, University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z4

*Corresponding author: Department of Wildlife, Fish and Conservation Biology, University of California Davis, Davis, CA 95616, USA.Tel: +1-530-752-4997. Email: [email protected]

Transformation of earth’s ecosystems by anthropogenic climate change is predicted for the 21st century. In many regions,the associated increase in environmental temperatures and reduced precipitation will have direct effects on the physio-logical performance of terrestrial and aquatic ectotherms and have already threatened fish biodiversity and important fish-eries. The threat of elevated environmental temperatures is particularly salient for members of the Oncorhynchus genusliving in California, which is the southern limit of their range. Here, we report the first assessments of the aerobic capacityof a Californian population of wild Oncorhynchus mykiss Walbaum in relationship to water temperature. Our field measure-ments revealed that wild O. mykiss from the lower Tuolumne River, California maintained 95% of their peak aerobic scopeacross an impressive temperature range (17.8–24.6°C). The thermal range for peak performance corresponds to local highriver temperatures, but represents an unusually high temperature tolerance compared with conspecifics and congenericspecies from northern latitudes. This high thermal tolerance suggests that O. mykiss at the southern limit of their indigen-ous distribution may be locally adjusted relative to more northern populations. From fisheries management and conserva-tion perspectives, these findings challenge the use of a single thermal criterion to regulate the habitat of the O. mykissspecies along the entirety of its distribution range.

Key words: aerobic scope, fish, metabolic rate, Oncorhynchus mykiss, swimming, temperature

Editor: Steven Cooke

Received 10 October 2016; Revised 24 October 2016; Editorial Decision 29 October 2016; accepted 1 November 2016

Cite as: Verhille CE, English KK, Cocherell DE, Farrell AP, Fangue NA (2016) High thermal tolerance of a rainbow trout population near itssouthern range limit suggests local thermal adjustment. Conserv Physiol 4(1): cow057; doi:10.1093/conphys/cow057.

IntroductionRainbow trout (Oncorhynchus mykiss Walbaum 1792) isregarded as a cold-water fish species with an indigenousrange stretching across an immense temperature gradient,from the subarctic climate region of the Bering Sea to theMediterranean climate region of Northern Baja California(Reyes, 2008). Despite this large temperature gradient and

distribution range, the optimal temperature range for wildO. mykiss aerobic performance capacity has been deter-mined only for indigenous populations inhabiting temperateclimates. Local adaptation of thermal performance existswithin the teleosts (Angilletta, 2009), but has never beenshown for wild O. mykiss populations across their nativerange. Without knowledge of the variation in thermal per-formance among populations of O. mykiss, fish conservation

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1© The Author 2016. Published by Oxford University Press and the Society for Experimental Biology.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

by guest on Decem

ber 13, 2016http://conphys.oxfordjournals.org/

Dow

nloaded from

managers apply regulatory water temperature criteriaderived for higher latitude populations of O. mykiss for pro-tection of lower latitude populations.

The present study considered the thermal performance of apopulation of O. mykiss located in a river near the southernlimits of its native range and was prompted by a number ofrecent events. Foremost, global indicators show that 2014 and2015 were the warmest years on record for the earth’s climate(Blunden and Arndt, 2015; NOAA National Centers forEnvironmental Information, 2016). Animal populations, suchas Californian O. mykiss, which exist at the latitudinal ext-remes of their biogeographical range, are expected to experi-ence the most profound negative effects of such climatechanges (Lassalle and Rochard, 2009). Second, for a fish thattends to favour pristine, cold water in most of its native habi-tat, native O. mykiss populations inhabiting the extremelywarm summer temperatures of Californian rivers are evidenceof considerable phenotypic plasticity (or genetic variability)within the species, allowing acclimation (or adaptation) tomuch warmer environmental temperature regimes. Indeed,severe thermal exposures in southern Western Australia haveproduced a line of introduced, hatchery-reared O. mykiss(Morrissy, 1973; Molony, 2001; Molony et al., 2004) thatswim and feed at 26°C (Michael Snow, Department ofFisheries, Government of Western Australia, personal commu-nication) and retain 50% of their peak aerobic capacity at25°C (Chen et al., 2015). Interestingly, the founder populationfor this thermally tolerant hatchery strain was transplantedfrom California during the last century for recreational fisher-ies. Thus, with climate change continuing to shift baseline riv-er water quality and availability (Sousa et al., 2011; Swainet al., 2014), especially in central California, where the intensi-fication of weather extremes is triggering water crises andextreme droughts (Dettinger and Cayan, 2014), knowledge ofthe local thermal requirements of vulnerable key fish speciesbecomes ever more pressing (Moyle et al., 2011).

Fish can adjust to warmer habitat temperatures byrelocating to a cooler refuge (if available), thermally accli-mating or thermally adapting (Farrell and Franklin, 2016);responses that all operate at different time scales. Indeed, thesuggestion that fish might tailor their metabolic rate to habi-tat temperature has a long and strong history across a widerange of aquatic habitats and species (Fry, 1947, 1971; Brettand Groves, 1979; Elliott, 1982; Jobling, 1994; Hochachkaand Somero, 2002; Donelson et al., 2012). In fact, local ther-mal adaptation has been thoroughly characterized for otherfish species, such as stickleback populations (Barrett et al.,2011), temperate killifish (Fangue et al., 2006) and tropicalkillifish (McKenzie et al., 2013). Even within the genusOncorhynchus, Fraser River watershed populations of sock-eye salmon (O. nerka Walbaum 1792) have apparently tunedtheir thermal performance to meet the energetic needs of theironce-in-a-lifetime upstream migration (Farrell et al., 2008;Eliason et al., 2011, 2013). The ability of O. mykiss to accli-mate thermally is well documented (Myrick and Cech, 2000),

and there appears to be the genetic potential for thermaladaptation given the successful selective breeding of O. mykisslines that perform well at high temperatures (Australian lines,Molony et al., 2004; Japanese lines, Ineno et al., 2005).Nevertheless, assessments of the aerobic capacity in relation towater temperature of wild O. mykiss at the southern extent oftheir range in California are lacking. What is known for twoCalifornian strains of O. mykiss (Eagle Lake and MountShasta; Myrick and Cech, 2000) is that the thermal perform-ance curves for hatching success differ (Myrick and Cech,2001) despite similar upper thermal tolerance values (CTmax).In addition, the Eagle Lake and Mount Shasta strains of O.mykiss grew fastest at different acclimation temperatures (19and 22°C, respectively), but growth ceased at 25°C in bothstrains (Myrick and Cech, 2000).

The accumulating evidence for variation in thermal per-formance within and among Pacific salmon and rainbowtrout populations seems incongruous with the criteria used bythe US Environmental Protection Agency (EPA) to regulatewater temperatures. The EPA uses a regulatory 7 day averageof the daily water temperature maximum (7DADM) of 18°Cfor all juvenile O. mykiss over their entire native US rangefrom southern California into Alaska (US EnvironmentalProtection Agency, 2003). One way to bring greater insightinto population-specific thermal tolerance and to take localadaptation and acclimation into consideration for regulatorypurposes is to use a well-established non-lethal approach tostudy the thermal physiology of O. mykiss populations inhabit-ing unusually warm habitats. Therefore, we examined O. mykissthat inhabit the Tuolumne River below La Grange DiversionDam, which is the most downstream habitat for O. mykiss ina watershed that drains ~2500 km2 of the Western SierraNevada mountain range. This river reach is characterized bya longitudinal thermal gradient, which increases from 12°Cto occasionally as high as 26°C during summer warming overa ~25 km stretch of river. By measuring metabolic scope foractivity (Fry, 1947), we tested the hypothesis that O. mykissresiding below the La Grange Diversion Dam on theTuolumne River may be locally adapted to the summer habi-tat temperatures that can reach 26°C. Mechanistically, ourexperimental approach builds on a fish’s ultimate requirementto have the capacity to supply oxygen for all activities (e.g.for foraging, digestion, growth, migration, predator avoid-ance and reproduction). The capacity to provide oxygenbeyond basic needs is termed absolute aerobic scope (AAS),which, in field situations (e.g. Pörtner and Knust, 2007;Nilsson et al., 2009; Gardiner et al., 2010; Eliason et al., 2011;Rummer et al., 2014), can be estimated from the differencebetween routine metabolic rate (RMR) and maximal meta-bolic rate (MMR). Thus, by measuring RMR and MMR overa wide range of water temperatures, the portion of the tem-perature range where AAS (i.e. the capacity for aerobic activ-ity) is maximized can be defined. Such information is lackingfor wild O. mykiss in central California. For the presentstudy, a temporary respirometry laboratory was built besidethe Tuolumne River. This laboratory allowed wild juvenile

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research article Conservation Physiology • Volume 4 2016

by guest on Decem


Dow

nloaded from

O. mykiss to be tested at temperatures between 13 and 25°Cbefore they were returned to their original habitat within24 h, as required by the experimental permits.

This study has implications beyond the thermal needs forresident aquatic species because this segment of theTuolumne River is part of a watershed that provides munici-pal water to >2.4 million residents of the San Francisco BayArea and agricultural irrigation water to the Central Valley(Turlock Irrigation District and Modesto Irrigation District,2011). The recent drought in central California has left reser-voirs at historic lows (California Department of WaterResources, 2015) and has challenged the capacity to balancethe environmental water flow needs of aquatic biota with thehuman requirements from this watershed for domestic, agri-cultural and recreational use. Juvenile O. mykiss livingbelow the La Grange Diversion Dam have been observedexploiting summer Tuolumne River temperatures from 12 to26°C over 25 river km (HDR Engineering, Inc., 2014).There are no additional cool-water inputs (except for raresummer rains), resulting in progressive warming of the waterreleased from the Dam as it flows downstream. Establishingthe optimal temperature range for aerobic performance ofwild Californian O. mykiss will provide fish conservationmanagers with scientific support for temperature criteria thatallow for optimization of this balance between human andfish requirements.

Materials and methodsPermitting restrictions that influenced theexperimental designWild Tuolumne River O. mykiss were collected underNational Marine Fisheries Service Section 10 permit no.17913 and California Fish and Wildlife Scientific CollectingPermit Amendments. No distinction was made between resi-dent (rainbow trout) and anadromous (steelhead) life-historyforms. For permitting purposes, these fish are considered asESA-listed California Central Valley steelhead, O. mykiss.Fish collection (up to a maximum of 50 individuals) wasallowed only between river kilometer (RK) 84.0 and RK 63.6,and capture temperatures could not exceed 21.1°C. This per-mit allowed only two fish to be captured and tested each day,and all fish had to be returned to their original river habitat.Given that indirect fish mortality was limited to three fish, aprecautionary measure included testing fish at the highesttemperatures last (i.e. not randomly assigning test tempera-ture). Additionally, the permit restricted test temperatures to≤25°C. All experimental procedures were approved by theInstitutional Animal Care and Use Committee (protocol no.18196; the University of California Davis).

Fish collection, transport and holdingTwo wild O. mykiss were collected daily [a total of 44 fish;22.4 g (SEM = 1.78, range 10.5–79.6 g) and 125.7mm

(SEM = 2.88)] from four primary locations on the TuolumneRiver (Supplementary material, Fig. S1). The two fish wereimmediately scanned for a passive integrated transponder(PIT) tag to preclude re-testing a fish. The fish were transferreddirectly to a 13 litre container partly submerged in the riverbefore being driven to the streamside field laboratory(<20min) in insulated coolers filled with 25 litres of fresh riverwater. A water temperature logger (recording every 15min;Onset Computer Corporation, USA) remained with the fishuntil testing was completed and the fish was returned to theriver. At the field laboratory, located immediately downstreamfrom the La Grange Diversion Dam, fish were transferred toholding tanks (300 litres) filled with flow-through TuolumneRiver water (directly from the dam) that had passed through acoarse foam filter and then an 18 litre gas-equilibration col-umn for aeration (12.5–13.6°C, >80% air saturation). Thus,field-acclimatized fish were placed into the holding tankswithin 60–120min of capture and remained there for 60–180min before being transferred to one of two 5 litre auto-mated swim tunnel respirometers (Loligo, Denmark). Routineand maximal metabolic rates were then measured at tempera-tures between 13 and 25°C (1°C increments).

Swim tunnel respirometersThe swim tunnel respirometers received aerated TuolumneRiver water from an 80 litre temperature-controlled sump thatwas refreshed every 80–90min. Water temperature was regu-lated within ±0.5°C of the test temperature by passing sumpwater through a 9500 BTU Heat Pump (Model DSHP-7, AquaLogic Delta Star, USA) with a high-volume pump (modelSHE1.7, Sweetwater®, USA). Additionally, two proportionaltemperature controllers (model 72, YSI, USA) each ran an800W titanium heater (model TH-0800, Finnex, USA) locatedin the sump. The water temperature in the swim tunnels wasmonitored with a temperature probe connected through afour-channel Witrox oxygen meter (Loligo). All temperature-measuring devices were calibrated bi-weekly to ±0.1°C of aNational Institute of Standards and Technology certified glassthermometer. Ammonia build-up was prevented by zeolite inthe sump, which was replaced weekly. Water oxygen saturationin each swim tunnel was monitored continuously using a dip-ping probe mini oxygen sensor connected to AutoResp software(Loligo) through the Witrox system (Loligo). Video cameraswith infrared lighting (Q-See, QSC1352W, China) continuouslyrecorded (Panasonic HDMI DVD-R, DMR-EA18K, Japan) fishbehaviour in the swim tunnels, which were shaded by blackcloth to limit fish disturbance. A variable frequency drive motorgenerated laminar water flow through the swimming section(calibrated using a digital anemometer with a 30mm vanewheel flow probe; Hönzsch, Germany) in each swim tunnel.

Metabolic rate measurementRoutine and active metabolic rates of fish in the swim tunnelrespirometers were measured using intermittent respirometry

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Conservation Physiology • Volume 4 2016 Research article

by guest on Decem


Dow

nloaded from

(Steffensen, 1989; Cech, 1990; Chabot et al., 2016; Svendsenet al., 2016). The swim tunnel was automatically sealed duringmeasurements and flushed with fresh, aerated sump waterbetween measurements (AutoResp software and a DAQ-PAC-WF4 automated respirometry system, Loligo). Oxygenremoval from the water by the fish (in milligrams of oxygen)was measured for a minimal period of 2 min when the swimtunnel was sealed, without oxygen levels falling below 80%air saturation. No background oxygen consumption wasdetected without fish (performed at the end of each day withboth swim tunnels; Rodgers et al., 2016) even at the highest testtemperature (25°C). Each oxygen probe was calibrated weeklyat the test temperatures using 100% (aerated distilled water)and 0% (150 ml distilled water with 3 g dissolved Na2SO3) air-saturated water.

Oxygen uptake was calculated according to the followingformula:

⎡⎣ ⎤⎦{ }( )

( )= ( ) − ( ) × × ×

− −

− −t t V M T

Oxygen uptake in mg O kg min

O O ,

20.95 1

2 1 2 20.95 1

where O2(t1) is the oxygen concentration in the swim tunnel atthe beginning of the seal (in milligrams of oxygen per litre);O2(t2) is the oxygen concentration in the tunnel at the end ofthe seal (in milligrams of oxygen per litre); V is the volume ofthe swim tunnel (in litres); M is the mass of the fish (in killo-grams); and T is the duration of the measurement (in minutes).Allometric correction for variable body mass used the expo-nent 0.95, which is halfway between the life-stage-independentexponent determined for resting (0.97) and active (0.93) zebra-fish (Lucas et al., 2014).

Experimental protocolFish were introduced between 13.00 and 16.00 h each dayinto a swim tunnel at 13 ± 0.3°C, which was close to the riv-er temperature at which most fish were caught, and left for60min before a 60min training swim (Jain et al., 1997), dur-ing which water flow velocity was gradually increased to5–10 cm s−1 higher than when swimming started (typicallyat 30 cm s−1) and held for 50min before a 10min swim at50 cm s−1 (the anticipated maximal prolonged swimmingvelocity for a 150 mm fish at 13°C; Alsop and Wood, 1997).Recovery for 60min preceded the incremental increases inwater temperature (1°C per 30min) up to the test tempera-ture. Oxygen uptake (10–30min, depending on the test tem-perature, and followed by a 5–10min flush period) wascontinuously measured throughout the night until 07.00 h.Estimates of RMR for each of the 44 tested fish were calcu-lated by averaging the lowest four oxygen uptake measure-ments at the test temperature for the minimum 8 h overnightperiod (Chabot et al., 2016). Visual inspection of the videorecordings confirmed that fish were quiescent during thesemeasurements with the exception of three fish that were

discarded owing to consistent activity throughout the night(Crocker and Cech, 1997), which reduced the RMR mea-surements to 41 fish.

Critical swimming velocity and burst swimming protocols(Reidy et al., 1995; Killen et al., 2007; Clark et al., 2013;Norin and Clark, 2016) were used to determine MMR. Theybegan between 08.00 and 09.00 h and lasted 2–6 h. For thecritical swimming velocity test, water velocity was graduallyincreased until the fish continuously swam at 30 cm s−1 for 20min. Water velocity was incrementally increased every 20 minby 10% of the previous test velocity (3–6 cm s−1) until the fishwas no longer able to swim continuously and fell back tomake full body contact with the downstream screen of theswimming chamber. The fish recovered for 1 min at 13–17 cms−1, the lowest velocity setting of the swim tunnel, beforerestoring the final water velocity over a 2 min period andrestarting the 20 min timer. Fatigue was defined as when thefish made full body contact with the downstream screen of theswim tunnel a second time at the same test velocity or failed toresume swimming. Active metabolic rate was measured at eachtest velocity using a 3 min flush period and a 7–17 min meas-urement period. All fish swam for 20 min at one water vel-ocity, but almost 50% of the wild fish used their caudal fin toprop themselves on the downstream screen of the swim tunnelto avoid swimming faster, which required a secondary meas-urement of maximal metabolic rate using a burst swimmingprotocol. For the burst swimming protocol, tunnel velocitywas set to and held for 10 min at the highest critical swimmingvelocity test increment where that fish had continuously swum.Afterwards, water velocity was rapidly (over 10 s) increased to70–100 cm s−1, which invariably elicited burst swimmingactivity for 30 s or less, when water velocity exceeded 70 cms−1. This protocol was repeated multiple times for 5–10 min,while oxygen uptake was measured continuously. The MMRwas assigned to the highest active metabolic rate measuredwith the active respirometry methods. Occasionally, fish exhib-ited intense struggling behaviours with an even higher oxygenuptake, which was assigned MMR. The MMR was not esti-mated for four fish, which failed to swim and raise their meta-bolic rate appreciably with any of the methods, resulting in atotal of 37 fish with RMR and MMR measurements. Absoluteaerobic scope (AAS = MMR − RMR) and factorial aerobicscope (FAS = MMR/RMR) were calculated.

All fish recovered in the swim tunnel at a water velocityof 13–17 cm s−1 and at the test temperature for 1 h whilemeasuring oxygen uptake. Water temperature was thendecreased to ~13°C over a 30 min period before the fish wasremoved, measured, PIT tagged and put into a holding tankbefore release at the capture site. Fish were individuallyanaesthetized for <5 min with CO2 (2 Alka-Seltzer tabletsdissolved in 3 litres of river water) for morphometric mea-surements [fork length (FL), in millimetres; and body mass,in grams], condition factor calculation (CF = bodymass × 105/FL3), and PIT tagging. Half duplex PIT (OregonRFID) tags were placed into the abdominal cavity via a

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4


by guest on Decem


Dow

nloaded from

1 mm incision through the body wall, just off-centre of thelinea alba. All equipment was sterilized with NOLVASAN Sprior to tagging, and incisions were sealed with 3MVetBond. Revived fish were immediately transported to thecoolers filled with 13–15°C river water. At the release site,river water was gradually added to the cooler to equilibratethe fish to river water temperature at a rate of 1–2°C h−1

before fish were allowed to swim away voluntarily.

Measurements of tail beat frequencyThe tail beat frequency (TBF; number of tail beats per 10 s,reported in Hz) of fish swimming continuously and holdingstation without contacting the downstream screen of the res-pirometer was measured using the average of two or three10 s sections of video recordings played back at either one-quarter or one-eighth of real time. The TBF was then relatedto swimming speed and temperature. Tail beat frequencies ofundisturbed fish holding station in the Tuolumne River weremeasured from footage from underwater video camerasanchored within 1m of O. mykiss schools and left to recordfor up to 4 h (GoPro Hero 4). The TBFs were determinedusing the same methodology applied to respirometer videorecordings (n = 15 at 14°C and n = 1 at 20°C).

Data analysisA statistical model was fitted to individual data [performedin R (R Core Development Team, 2013) using the ‘lm’ func-tion] to determine the best relationships between the testtemperature and RMR, MMR, AAS and FAS. The statisticalmodel (linear, quadratic, antilogarithic base 2 and logarith-mic base 2 were tested) with the highest r2 and lowestresidual SE being reported. Confidence intervals and pre-dicted values based on the best-fit model were calculated inR using the ‘predict’ function. Variances around metabolicrate measurements are reported as 95% confidence intervals(CIs).

ResultsAs anticipated, basic oxygen needs (RMR) increased expo-nentially by 2.5-fold from 13 to 25°C (from 2.18 ± 0.45(95% CI) to 5.37 ± 0.41 mg O2 kg

−0.95 min−1). This thermalresponse was modelled by: RMR (in mg O2 kg−0.95

min−1) = 5.9513 – 0.5787 (temperature, in °C) + 0.0200(temperature, in °C)2 (P < 0.001, r2 = 0.798; Fig. 1A). TheMMR increased linearly by 1.7 times (from 6.62 ± 1.03 to11.22 ± 0.86 mg O2 kg−0.95 min−1) from 13 to 25°C. Thisthermal response was modelled by: MMR (in mg O2 kg−0.95

min−1) = 1.6359 + 0.3835 (temperature, in °C) (P < 0.001,r2 = 0.489; Fig. 1B). Given that MMR almost kept pace withthe thermal effect on RMR, AAS had a rather flat reactionnorm that was largely independent of the test temperaturerange. This thermal response was modelled by: AAS (in mgO2 kg−0.95 min−1) = −5.7993 + 1.1263 (temperature, in °C)− 0.0265 (temperature, in °C)2 (P = 0.060, r2 = 0.098;Fig. 1C). Using this model, peak AAS (6.15 ± 0.71mg O2

kg−0.95 min−1) was centred at 21.2°C. Nevertheless, theunexpected flat reaction norm meant that 95% of peak AASwas maintained from 17.8 to 24.6°C, which is a broad ther-mal window for peak AAS that extends well beyond the7DADM value of 18°C for O. mykiss.

Factorial aerobic scope is a useful metric of whetheror not a fish might have the required aerobic capacity toperform a specific activity, e.g. a doubling of RMR (i.e.FAS = 2) might be needed to digest a full meal properly(Jobling, 1981; Alsop and Wood, 1997; Fu et al., 2005; Luoand Xie, 2008). As expected, FAS decreased with tempera-ture (Clark et al., 2013), a thermal response modelled by:FAS = 2.1438 + 0.1744 (temperature, in °C) − 0.0070 (tem-perature, in °C)2 (P < 0.001, r2 = 0.344; Fig. 1D).

In addition, given the need to integrate AAS or FAS withinan ecological framework (see Overgaard et al., 2012; Clarket al., 2013; Farrell, 2013, 2016; Pörtner and Giomi, 2013;Ern et al., 2014; Norin et al., 2014), we used measurements ofTBF to estimate the oxygen cost required by a wild O. mykissto maintain station in the river currents of typical habitats inthe Tuolumne River. A steady TBF used for this activity atambient temperatures of 14 and 20°C was 2.94 and 3.40 Hz,respectively (see supplemental video, available online). Usingrespirometer swimming data to relate TBF to oxygen uptakeat 14 and 20°C (P < 0.001, r2 = 0.35; and P = 0.009,r2 = 0.33, respectively; Fig. 2), in-river TBF values of 2.94 and3.40 Hz corresponded to 3.26 and 5.43 mg O2 kg

−0.95 min−1,respectively. Therefore, we suggest that wild fish observedholding station in the Tuolumne River increased RMR by1.5 times at 14°C and by 2.0 times at 20°C, an activity thatwould use 49 and 67%, respectively, of the available FASmeasured at these two temperatures.

Fish recovery after exhaustive swimming tests was quickand without any visible consequences. The RMR at the endof the 60 min recovery period was either elevated by nomore than 20% or fully restored; an observation consistentwith previous laboratory studies of O. mykiss recovery (Jainet al., 1997; Jain and Farrell, 2003). Two fish tested at 25°Cwere the only exceptions. These two fish regurgitated theirgut contents during recovery and one then died abruptly.Inadvertent fish recapture provided some information on fishsurvival after being returned to the river. Six PIT-tagged fishwere recaptured at 1–11 days post-testing within 20 m oftheir original capture location; all were visually in good con-dition, and three of these fish had been tested at 23°C.

DiscussionThe present study is the first to consider the thermal responsefor an O. mykiss population so close to the southerly bound-ary of the natural distribution range for indigenousO. mykiss. We clearly show that 95% of peak AAS wasmaintained over an unexpectedly broad thermal window(17.8–24.6°C) and that all fish tested could maintain an

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5


by guest on Decem


Dow

nloaded from

FAS >2.0 up to 23°C. Moreover, we place these findingsinto an ecological context by suggesting that the level of FASat temperatures at least as high as 20°C may be more thanadequate to maintain station in the local water current of theTuolumne River and probably to digest a meal properly andoptimize growth, which is a very powerful integrator ofenvironmental, behavioural and physiological influences ona fish’s fitness. Moreover, fish were tested on site andreturned afterwards to the river, making the work locallyrelevant for the O. mykiss population, sensitive to conserva-tion needs and globally relevant by addressing the followingbroad question: are fish at the extreme edges of their biogeo-graphical range more physiologically tolerant because of thethermal extremes they experienced there?

The present results show good quantitative agreement withvarious previous studies with O. mykiss that have measured

some of the variables measured in the present study. Forexample, the 2.5-fold exponential increase in RMR from 13 to25°C (from 2.18 ± 0.45 (95% CI) to 5.37 ± 0.41 mg O2

kg−0.95 min−1) compares well with laboratory studies of RMRreported at 14°C (2.3–2.8 mg O2 kg−0.95 min−1; Myrick andCech, 2000) for 7 g Mount Shasta and Eagle Lake O. mykiss,and at 25°C (~6.5 mg O2 per kg

−0.95 min−1; Chen et al., 2015)for 30 g Western Australian O. mykiss. Therefore, concernsabout handling stress and specific dynamic action were min-imal. Likewise, MMR increased linearly by 1.7 times (from6.62 ± 1.03 to 11.22 ± 0.86 mg O2 kg

−0.95 min−1) from 13 to25°C, comparing well with previous laboratory measurementsof MMR reported at 15°C (2.8–8.7 mg O2 kg−0.95 min−1) for2–13 g O. mykiss (Scarabello et al., 1992, Alsop and Wood,1997) and with the peak MMR at 20°C (~11.13 mg O2 perkg−0.95 min−1) for Australian O. mykiss (Chen et al., 2015). Asa consequence of MMR nearly keeping pace with the thermal

Figure 1: The relationships between test temperature and the routine (RMR; A) and maximal metabolic rate (MMR; B) of Tuolumne RiverOncorhynchus mykiss. The three methods used to measure MMR (see Materials and methods section) are distinguished by different symbols.Absolute aerobic scope (AAS; C) and factorial aerobic scope (FAS; D) were derived from the metabolic rate measurements. Each data pointrepresents an individual fish tested at one temperature. These data were given a best-fit mathematical model (continuous line or curve), andthe 95% confidence intervals for each line are indicated by the shaded area. The RMR and FAS were smoothed to a polynomial fit of the formy = x + I(x2), where y is RMR or FAS, x is temperature, and I is a constant. The MMR and AAS were smoothed to a linear fit of the form y = x + c,where c is a constant. For RMR, degrees of freedom (d.f.) = 34, P < 0.001, residual standard error (RSE) = 0.561 and r2 = 0.798. For MMR,d.f. = 35, P < 0.001, RSE = 1.580 and r2 = 0.489. For AAS, d.f. = 35, P = 0.060, RSE = 1.490 and r2 = 0.098. For FAS, d.f. = 34, P < 0.001,RSE = 0.506 and r2 = 0.344. The asterisk indicates the one fish that died abruptly after the swimming test.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6


by guest on Decem


Dow

nloaded from

effect on RMR, AAS was largely independent of test tempera-ture. Directly comparing our AAS values with other studiesrevealed that our result for AAS at 15°C (5.10 mg O2 kg−0.95

min−1) was at the high end of previous laboratory measure-ments of AAS (1.8–5.8 mg O2 kg

−0.95 min−1) for O. mykiss at15°C (Scarabello et al., 1992, Alsop and Wood, 1997;McGeer et al., 2000), but lower than peak AAS (~7.3 mg O2

per kg−0.95 min−1) at 20°C in Australian O. mykiss (Chenet al., 2015). Likewise, our FAS values were bracketed byvalues obtained in previous laboratory studies. At 24°C, FAS(2.13 ± 0.33) was greater than that reported at 25°C (1.8) forWestern Australian O. mykiss (Chen et al., 2015), but com-pared with FAS values for juvenile rainbow trout (1.8–5.8) at13°C (Scarabello et al., 1992, Alsop and Wood, 1997,McGeer et al., 2000), our FAS at 13°C (3.32 ± 0.41) was inthe middle of the range.

To place the present data for Californian O. mykiss intoperspective, we have compared (Fig. 3) their reaction normwith those published for juveniles of northern O. mykiss (datafrom Fry, 1948) and Australian hatchery-selected O. mykiss

(data from Chen et al., 2015), as well as adult northern popula-tions of selected Pacific salmon populations (data from Leeet al., 2003a and Eliason et al., 2011). Among the native O.mykiss populations, the Lower Tuolumne River juvenileCalifornian O. mykiss are likely to experience the highest tem-peratures during summer (up to 26°C), although the introducedAustralian O. mykiss population had experienced selectiontemperatures ≥25°C (Chen et al., 2015). Notably, AAS at 24°C for Tuolumne River O. mykiss is greater than other O.mykiss populations and only bettered by the Chilko sockeyesalmon population, one of several sockeye salmon populationsthat are known to have a peak AAS at the modal temperaturefor their upstream spawning migration (Eliason et al., 2011,2013). Thus, the present data are in line with evidence of intra-specific matching of metabolic rate to local water temperatures

Figure 2: The relationship between tail beat frequency (TBF; in hertz)and metabolic rate (MR; in mg O2 kg

−0.95 min−1) measured whenTuolumne River Oncorhynchus mykiss were swimming continuouslyin a swim tunnel at 14 (A) or 20°C (B). The continuous black linerepresents the linear regression based on the data for n = 7 fish at14°C and n = 5 fish at 20°C. The vertical dashed lines represent theestimated TBF (2.94 Hz at 14°C and 3.40 Hz at 20°C) taken fromvideos of O. mykiss maintaining station in a water current in theirnormal Tuolumne River habitat. At 14°C, the relationship betweenTBF and MR followed the equation MR = 0.75TBF + 1.05, withdegrees of freedom (d.f.) = 41, P < 0.001, residual standard error(RSE) = 1.27 and r2 = 0.35. According to this formula, the MR for theTBF measured in the river (2.943 Hz) at 14°C was estimated to be3.26 mg O2 kg

−0.95 min−1. At 20°C, the relationship between TBF andMR followed the equation MR = 1.04TBF + 1.89, with d.f. = 15,P = 0.009, RSE = 1.29 and r2 = 0.33. According to this formula, the MRfor the TBF measured in the river at 20°C (3.402 Hz) was estimated tobe 5.43 mg O2 kg

−0.95 min−1.

Figure 3: Absolute aerobic scope (AAS) for three strains ofOncorhynchus mykiss, i.e. a northern strain (Fry, 1948), an Australianstrain (Chen et al., 2015) and the California strain reported in thismanuscript, compared with AAS measurements of Chehalis Cohosalmon (Oncorhynchus kisutch Walbaum 1792) and Gates Creek,Weaver Creek (Oncorhynchus nerka Walbaum 1792; Lee et al., 2003a)and Chilko Creek sockeye salmon (Eliason et al., 2011). The best-fitline of the relationship between AAS and temperature of the speciesand populations from other publications was predicted using asecond-order polynomial linear regression performed on the raw data(Lee et al., 2003a; Chen et al., 2015) or data extracted from plots (Fry,1948; Eliason et al., 2011) from the original publications. Coefficientestimates from the linear regression analysis were then used todetermine the peak aerobic scope and the temperaturescorresponding to the peak and 95% of peak AAS.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7


by guest on Decem


Dow

nloaded from

within the Oncorhynchus genus. Although the peak AAS of theAustralian O. mykiss population was 50% greater than for theother two O. mykiss populations, Tuolumne River O. mykisshad the broadest and highest thermal window (17.8–24.6°C)among the O. mykiss populations (20.5–22.4°C from Fry,1948; and 12.8– 18.6°C from Chen et al. 2015).

Whether the matching of Tuolumne River O. mykiss meta-bolic performance to local habitat temperatures is a result ofthermal acclimation or local adaption, as in the WesternAustralian O. mykiss, will need study well beyond the presentwork. Thermal acclimation usually results in fish performingbetter at the new temperature. For example, thermal acclima-tion offsets the effect of acute warming on RMR in 5–7 gMount Shasta and Eagle Lake O. mykiss (2.3–2.8 mg O2

kg−0.95 min−1 at 14°C and 2.9–3.1 mg O2 kg−0.95 min−1 at25°C; Myrick and Cech, 2000), which would normally doubleRMR over this temperature range, as observed here. Warmacclimation can also increase upper thermal tolerance limits,as it did for four anadromous Great Lakes populations ofjuvenile O. mykiss (Bidgood and Berst, 1969). Given that ourfish were captured at and, presumably, thermally acclimatizedto between 14 and 17°C, it would be of interest to test wildfish with a warmer thermal acclimation history. But evenwithout thermal acclimation, the present data suggest thatTuolumne River O. mykiss and those for northern O. mykiss(Fry, 1948; see Fig. 3) have the aerobic capacity temporarily,if not regularly, to exploit temperatures well above 18°C,which is the upper thermal limit suggested by EPA guidancedocuments (US Environmental Protection Agency, 2003).

Nevertheless, we caution that such local tailoring may notbe evident in all salmonid species. For example, the thermalphysiology of Atlantic salmon (Salmo salar Linnaeus 1758)from northern and southern extremes of their European rangedid not show any major difference (Anttila et al., 2014). Allthe same, a sub-species of redband trout (O. mykiss gairdneri),which are apparently adapted to high summer temperatures ofNorth American desert streams (Narum et al., 2010; Narumand Campbell, 2015), are likewise capable of high levels ofswimming performance up to 24°C (Rodnick et al., 2004) andhigher swimming performance for a warm vs. a cool creekpopulation (Gamperl et al., 2002). Redband trout have beenobserved actively feeding at 27–28°C (Sonski, 1984; Behnke,2010), but thermal selection of wild redband trout is centredbetween 13 (Gamperl et al., 2002) and 17°C (Dauwalter et al.,2015). How O. mykiss behaviourally exploit the steep sum-mer thermal gradient in the Tuolumne River below the LaGrange Diversion Dam (from 12 to 26°C over 25 km; HDREngineering, Inc., 2014) is another unknown. Even withoutthese important details, Tuolumne River O. mykiss appearphysiologically to be tolerant of the thermal extremes theyexperience.

The capacity of a fish to deliver oxygen to support activ-ities in water of varying quality is a concept originally intro-duced for fishes >60 years ago (Fry, 1947). The oxygen- and

capacity-limited thermal tolerance hypothesis broadens thisconcept and provides a mechanistic explanation (Pörtner,2001; Pörtner and Farrell, 2008; Deutsch et al., 2015), but iscurrently under debate (Overgaard et al., 2012; Clark et al.,2013; Farrell, 2013; Pörtner and Giomi, 2013; Ern et al.,2014; Norin et al., 2014). An accepted fact is that a metabolicload from an environmental factor (e.g. temperature) canincrease the oxygen cost for living (i.e. RMR). Consequently,like all other temperature studies with fish, the magnitude ofthe 2.5-fold increase in RMR observed here over a 12°C tem-perature range (between 13 and 25°C) was expected.However, temperature did not limit MMR, which increasedlinearly with acute warming, and the peak MMR was notresolved. The statistical models, which were based on individ-ual responses and 1°C temperature increments from 13 to25°C, predicted a peak AAS at 21.2°C for Tuolumne RiverO. mykiss and a FAS >2.0 up to 23°C. As the allocation ofenergy and trade-offs are recognized and fundamental tenantsof ecological physiology, especially in fishes (Sokolova et al.,2012), we suggest that being able to at least double RMR hasecological relevance for two behaviours that are likely to influ-ence survival of O. mykiss, maintaining station in a flowingriver and processing a large meal.

Snorkeling in the Tuolumne River provided visual obser-vations of O. mykiss maintaining station in the river currentfor prolonged periods that were punctuated by hiding underthe river bank and by darting behaviours to capture preyand to protect their position. Maintaining station required asteady TBF similar to the situation in the swim tunnel respir-ometer, which allowed us to estimate a metabolic cost ofmaintaining station in typical Tuolumne River habitats at 14and 20°C (a 1.5- to 2-fold increase in RMR) and the aerobicscope available for additional activities (1.7–2 times RMR).Although darting behaviours are likely to be fuelled anaer-obically, O. mykiss must (and were clearly able to) repay thepost-exercise excess oxygen debt (Lee et al., 2003b) whilemaintaining station in the river current. The rapid recoveryof RMR after exhaustive exercise in the swim tunnel suggeststhat O. mykiss had the capacity to repay post-exercise excessoxygen debt rapidly at temperatures as high as 24°C.Although digestion of a meal at high temperatures proceedsmore rapidly and with a higher peak metabolic rate, the totaloxygen cost of the meal remains similar. Thus, fish can the-oretically eat more frequently and potentially grow faster ata higher temperature provided there is a sufficient FAS fordigestion within the overall AAS. Given that peak metabolicrate during digestion of a typical meal for a salmonid doesnot necessarily double RMR at the temperatures used here(e.g. Jobling, 1981; Alsop and Wood, 1997; Fu et al., 2005;Luo and Xie, 2008), an FAS value of 2 should be a reason-able index, and all O. mykiss tested had this capacity up to23°C. Indeed, the fish were apparently feeding well in the riv-er, given a high condition factor (1.1 SEM = 0.01), the faecaldeposits found in the swim tunnel and two fish regurgitatingmeals when tested at 25°C. Meal regurgitation would beconsistent with an oxygen limitation, given that aquatic

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8


by guest on Decem


Dow

nloaded from

hypoxia impairs digestion in O. mykiss (Eliason and Farrell,2014). Indeed, feeding and growth are suppressed at supra-optimal temperatures (Hokanson et al., 1977; Brett and Groves,1979; Elliott, 1982; Myrick and Cech, 2000, 2001). Takentogether, these data suggest that Tuolumne River O. mykisswere doing well in their habitat and had the aerobic capacityto do so.

Our metabolic measurements, which show good quantita-tive agreement with controlled laboratory O. mykiss studies,represent a major challenge to the use of a single thermal cri-terion to regulate O. mykiss habitat when determining conser-vation criteria along the entire Pacific coast and perhapselsewhere. The 7DADM of 18°C for O. mykiss draws heavilyon a growth study performed in Minnesota (Hokanson et al.,1977). Therefore, it will be important to examine whether thepeak AAS at 21.2°C for Tuolumne River O. mykiss is asso-ciated with a peak growth rate. In this regard, the peakgrowth rate of another Californian O. mykiss population (theMount Shasta strain) occurred at acclimation temperatures(19–22°C; Myrick and Cech, 2000) above the 7DADM andwithin the thermal window for 95% peak AAS for TuolumneRiver O. mykiss. The Mount Shasta O. mykiss strain alsostopped growing at 25°C, the same temperature at which FASfor Tuolumne River O. mykiss approached 2. In contrast, theCalifornian Eagle Lake O. mykiss strain grew fastest at 19°Cand lost weight at 25°C (Myrick and Cech, 2000). Thus, theMount Shasta and Tuolumne River O. mykiss populationsare better able to acclimate thermally to temperatures >20°Cthan the Eagle Lake strain. With clear evidence that aCalifornia strain of O. mykiss can grow faster at acclimationtemperatures >18°C and that strains may differ in their theoptimal temperature for growth by as much as 3°C, there is aprecedent that local populations of O. mykiss can performwell above 18°C. Our findings also highlight the need forfuture experiments that consider replicate populations fromthroughout the species range to assess how widespread intra-specific variation in aerobic scope in O. mykiss might be.Continual development and refinement of the metrics used tobest inform regulatory criteria should be an ongoing pursuit,particularly if regional standards are to be implemented and ifthe criteria move away from what may now be consideredconservative. Probabilistic modelling approaches associatedwith a diversity of water temperature standards should bedeveloped in order for managers to understand the balancebetween standards that are conservative compared with thosethat are more risky.

The capacity to balance the essential environmentalrequirements of aquatic biota with human requirements isbecoming increasingly challenging across the globe because ofrecent increases in severe drought and record high temperatureoccurrences, a trend that climate change models project willcontinue. We suggest that broadly applying regulatory criteria,such as the 18°C 7DADM criterion for Pacific Northwest O.mykiss populations, to all North American O. mykiss is nolonger realistic and, in the present case, overly conservative.

The high degree of thermal plasticity discovered here for theTuolumne River O. mykiss population, which corresponds tolocal thermal conditions, adds to the accumulating evidence ofthe capacity for local adaptation among populations withinthe Oncorhynchus genus, including O. mykiss. Importantly,this work clearly illustrates that, owing to thermal plasticity,broad application of a single temperature criterion for fishprotection and conservation is not scientifically supported,especially for fish populations at the extreme limits of the spe-cies’ indigenous range.

Supplementary materialSupplementary material is available at ConservationPhysiology online.

AcknowledgementsWe thank Stillwater Sciences and FISHBIO for fish collectionassistance, and the Turlock Irrigation District for the use offacilities. We also thank Steve Boyd, John Devine, JennaBorovansky, Jason Guignard, Andrea Fuller and RonYoshiyama. Fieldwork was supported by Trinh Nguyen,Sarah Baird, James Raybould, Mike Kersten, Jeremy Pombo,Shane Tate, Patrick Cuthbert and Mike Phillips. All experi-ments followed ethical guidelines approved by the Universityof California Davis’ Institutional Animal Care and UseCommittee (protocol no. 18196). Tuolumne River O. mykisswere collected under National Marine Fisheries ServiceSection 10 permit no. 17913 and California Fish and WildlifeScientific Collecting Permit Amendments. We thank Ken Zilligand Dr Jamilynn Poletto for valuable manuscript feedback.

FundingThis work was supported by the Turlock Irrigation District,the Modesto Irrigation District, the City and County of SanFrancisco and UC Davis Agricultural Experiment Station(grant 2098-H to N.A.F.). A.P.F. holds a Canada ResearchChair and is funded by National Sciences and Energy ResearchCouncil of Canada.

ReferencesAlsop D, Wood C (1997) The interactive effects of feeding and exercise

on oxygen consumption, swimming performance and proteinusage in juvenile rainbow trout (Oncorhynchus mykiss). J Exp Biol10: 2337–2346.

Angilletta MJ (2009) Thermal Adaptation: a Theoretical and EmpiricalSynthesis. Oxford University Press, Oxford, UK.

Anttila K, Couturier CS, Overli O, Johnsen A, Marthinsen G, Nilsson GE,Farrell AP (2014) Atlantic salmon show capability for cardiac accli-mation to warm temperatures. Nat Commun 5: 4252.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9


by guest on Decem


Dow

nloaded from

Barrett RDH, Paccard A, Healy TM, Bergek S, Schulte PM, Schluter D,Rogers SW (2011) Rapid evolution of cold tolerance in stickleback.Proc Biol Sci 278: 233–238.

Behnke R (2010) Trout and Salmon of North America. The Free Press,New York, NY.

Bidgood BF, Berst AH (1969) Lethal temperatures for Great Lakes rain-bow trout. J Fish Res Board Can 26: 456–459.

Blunden J, Arndt DS (2015) State of the climate in 2014. Bul AmMeteor Soc 96: S1–S267.

Brett JR, Groves TDD (1979) Physiological Energetics. In Hoar WS,Randall DJ, Brett JR, eds, Fish Physiology Volume VIII Bioenergeticsand Growth, Academic Press, New York, NY, pp 279–352.

California Department of Water Resources (2015) Reservoir water stor-age. http://cdec.water.ca.gov/cgi-progs/reservoirs/STORAGEW

Cech JJ Jr (1990) Respirometry. In CB Schreck, PB Moyle, eds. Methods forFish Biology. American Fisheries Society, Bethesda, MD, pp 335–362.

Chabot D, Steffensen JF, Farrell AP (2016) The determination of stand-ard metabolic rate in fishes. J Fish Biol 88: 81–12.

Chen JQ, Snow M, Lawrence CS, Church AR, Narum SR, Devlin RH,Farrell AP (2015) Selection for upper thermal tolerance in rainbowtrout (Oncorhynchus mykiss Walbaum). J Exp Biol 218: 803–812.

Clark TD, Sandblom E, Jutfelt F (2013) Aerobic scope measurements offishes in an era of climate change: respirometry, relevance andrecommendations. J Exp Biol 216: 2771–2782.

Crocker CE, Cech JJ Jr (1997) Effects of environmental hypoxia on oxy-gen consumption rate and swimming activity in juvenile whitesturgeon, Acipenser transmontanus, in relation to temperature andlife intervals. Environ Biol Fish 50: 759–769.

Dauwalter DC, Fesenmyer KA, Bjork R (2015) Using aerial imagery tocharacterize redband trout habitat in a remote desert landscape.Trans Am Fish Soc 8487: 1322–1339.

Dettinger M, Cayan DR (2014) Drought and the California Delta—amatter of extremes. San Francisco Estuary and Watershed Sci 12:1–6.

Deutsch C, Ferrel A, Seibel B, Pörtner HO, Huey RB (2015)Ecophysiology. Climate change tightens a metabolic constraint onmarine habitats. Science 348: 1132–1135.

Donelson JM, Munday PL, McCormick MI, Pitcher CR (2012) Rapidtransgenerational acclimation of a tropical reef fish to climatechange. Nat Clim Change 2: 30–32.

Eliason E, Farrell AP (2014) Effect of hypoxia on specific dynamicaction and postprandial cardiovascular physiology in rainbow trout(Oncorhynchus mykiss). Comp Biochem Physiol A Mol Integr Physiol171: 44–50.

Eliason EJ, Clark TD, Hague MJ, Hanson LM, Gallagher ZS, Jeffries KM,Gale MK, Patterson DA, Hinch SG, Farell AP (2011) Differences inthermal tolerance among sockeye salmon populations. Science332: 109–112.

Eliason EJ, Wilson SM, Farrell AP, Cooke SJ, Hinch SG (2013) Low car-diac and aerobic scope in a costal population of sockeye salmon,Oncorhynchus nerka, with a short upriver migration. J Fish Biol 82:2104–2112.

Elliott JM (1982) The effects of temperature and ration size on thegrowth and energetics of salmonids in captivity. Comp BiochemPhysiol B Comp Biochem 73: 81–91.

Ern R, Huong DTT, Phuong NT, Wang T, Bayley M (2014) Oxygen deliv-ery does not limit thermal tolerance in a tropical eurythermalcrustacean. J Exp Biol 217: 809–814.

Fangue NA, Hofmeister M, Schulte PM (2006) Intraspecific variation inthermal tolerance and heat shock protein gene expression in com-mon killifish, Fundulus heteroclitus. J Exp Biol 209: 2859–2872.

Farrell AP (2013) Aerobic scope and its optimum temperature: clarify-ing their usefulness and limitations – correspondence on J. Exp.Biol. 216, 2771–2772. J Exp Biol 216: 4493–4494.

Farrell AP (2016) Pragmatic perspective on aerobic scope: peaking,plummeting, pejus and apportioning. J Fish Biol 88: 322–343.

Farrell AP, Franklin CE (2016) Recognizing thermal plasticity in fish.Science 351: 132–133.

Farrell AP, Hinch SG, Cooke SJ, Patterson DA, Crossin GT, Lapointe M,Mathes MT (2008) Pacific salmon in hot water: applying aerobicscope models and biotelemetry to predict the success of spawn-ing migrations. Physiol Biochem Zool 81: 697–708.

Fry FEJ (1947) Effects of the environment on animal activity.Publications of the Ontario Fisheries Research Laboratory 55: 1–62.

Fry FEJ (1948) Temperature relations of salmonids. Proc Can CommFreshwater Fish Res 1st Meeting. App. D. 1–6.

Fry FEJ (1971) The effect of environment factors on the physiology offish. Fish Physiol 6: 1–98.

Fu SJ, Xie XJ, Cao ZD (2005) Effect of meal size on postprandial meta-bolic response in southern catfish (Silurus meridionalis). CompBiochem Physiol A Mol Integr Physiol 140: 445–451.

Gamperl AK, Rodnick KJ, Faust HA, Venn EC, Bennett MT, CrawshawLI, Keeley ER, Powell MS, Li HW (2002) Metabolism, swimmingperformance, and tissue biochemistry of high desert redbandtrout (Oncorhynchus mykiss ssp.): evidence for phenotypic differ-ences in physiological function. Physiol Biochem Zool 75:413–431.

Gardiner NM, Munday PL, Nilsson GE (2010) Counter-gradient vari-ation in respiratory performance of coral reef fishes at elevatedtemperatures. PLoS One 5: e13299.

HDR Engineering, Inc. (2014) In-river diurnal temperature variation.Don Pedro Project FERC No. 2299.

Hochachka PW, Somero GN (2002) Biochemical Adaptation. OxfordUniversity Press, Oxford, UK.

Hokanson KEF, Kleiner CF, Thorslund TW (1977) Effects of constanttemperatures and diel temperature fluctuations on specific growth

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10


by guest on Decem


Dow

nloaded from

and mortality rates and yield of juvenile rainbow trout, Salmogairdneri. J Fish Res Board Can 34: 639–648.

Ineno T, Tsuchida S, Kanda M, Watabe S (2005) Thermal tolerance of arainbow trout Oncorhynchus mykiss strain selected by high-temperature breeding. Fish Sci 71: 767–775.

Jain KE, Farrell AP (2003) Influence of seasonal temperature on therepeat swimming performance of rainbow trout Oncorhynchusmykiss. J Exp Biol 206: 3569–3579.

Jain KE, Hamilton JC, Farrell AP (1997) Use of a ramp velocity test tomeasure critical swimming speed in rainbow trout (Oncorhynchusmykiss). Comp Biochem Physiol A Mol Integr Physiol 117: 441–444.

Jobling M (1981) The influences of feeding on the metabolic rate offishes: a short review. J Fish Biol 18: 385–400.

Jobling M, Baardvik BM (1994) The influence of environmental manipula-tions on inter–and intra–individual variation in food acquisition andgrowth performance of Arctic charr, Salvelinus alpinus. J Fish Biol 44:1069–1087.

Killen SS, Costa I, Brown JA, Gamperl AK (2007) Little left in the tank:metabolic scaling in marine teleosts and its implications for aer-obic scope. Proc Biol Sci 274: 431–438.

Lassalle G, Rochard E (2009) Impact of twenty-first century climatechange on diadromous fish spread over Europe, North Africa andthe Middle East. Glob Change Biol 15: 1072–1089.

Lee CG, Farrell AP, Lotto A, MacNutt MJ, Hinch SG, Healey MC (2003a)The effect of temperature on swimming performance and oxygenconsumption in adult sockeye (Oncorhynchus nerka) and coho(O. kisutch) salmon stocks. J Exp Biol 206: 3239–3251.

Lee CG, Farrell AP, Lotto A, Hinch SG, Healey MC (2003b) Excess post-exercise oxygen consumption in adult sockeye (Oncorhynchus ner-ka) and coho (O. kisutch) salmon following critical speed swim-ming. J Exp Biol 206: 3253–3260.

Lucas J, Schouman A, Plyphout L, Cousin X, LeFrancois C (2014)Allometric relationship between body mass and aerobic metabol-ism in zebrafish Danio rerio. J Fish Biol 84: 1171–1178.

Luo XP, Xie XJ (2008) Effects of temperature on the specific dynamicaction of the southern catfish, Silurus meridionalis. Comp BiochemPhysiol A Mol Integr Physiol 149: 150–156.

McGeer JC, Szebedinsky C, McDonald DG, Wood CM (2000) Effects ofchronic sublethal exposure to waterborne Cu, Cd or Zn in rainbowtrout. I: Iono-regulatory disturbance and metabolic costs. AquatToxicol 50: 231–243.

McKenzie DJ, Estivales G, Svendsen JC, Steffensen JF, Agnèse JF(2013) Local adaptation to altitude underlies divergent thermalphysiology in tropical killifishes of the genus Aphyosemion. PLoSOne 8: e54345.

Molony BW (2001). Environmental requirements and tolerances ofrainbow trout (Oncorhynchus mykiss) and brown trout (Salmo trut-ta) with special reference to Western Australia: a review. FisheriesResearch Reports No. 130. Western Australia, p 28.

Molony BW, Church AR, Maguire GB (2004) A comparison of the heattolerance and growth of a selected and non-selected line of rain-bow trout, Oncorhynchus mykiss, in Western Australia. Aquaculture241: 655–665.

Morrissy NM (1973) Comparison of strains of Salmo gairdneriRichardson from New South Wales, Victoria and Western Australia.Aust Soc Limn Bull 5: 11–20.

Moyle PB, Katz JVE, Quinones RM (2011) Rapid decline of California’snative inland fishes: a status assessment. Biol Conserv 144: 2414–2423.

Myrick CA, Cech JJ Jr (2000) Temperature influences on California rainbowtrout physiological performance. Fish Physiol Biochem 22: 245–254.

Myrick CA, Cech JJ Jr (2001) Temperature Effects on Chinook Salmonand Steelhead: a Review Focusing on California’s Central ValleyPopulations. Bay-Delta Modeling Forum, Technical Publication 01-1.

Narum SR, Campbell NR (2015) Transcriptomic response to heat stressamong ecologically divergent populations of redband trout. BMCGenomics 16: 103.

Narum SR, Campbell NR, Kozfkay CC, Meyer KA (2010) Adaptation ofredband trout in desert and montane environments. Mol Ecol 19:4622–4637.

Nilsson GE, Crawley N, Lunde IG, Munday PI (2009) Elevated tempera-ture reduces the respiratory scope of coral reef fishes. GlobChange Biol 15: 1405–1412.

NOAA National Centers for Environmental Information (2016) State ofthe Climate: Global Analysis for December 2015. http://www.ncdc.noaa.gov/sotc/global/201512.

Norin T, Clark TD (2016) Measurement and relevance of maximummetabolic rate in fishes. J Fish Biol 88: 122–151.

Norin T, Malte H, Clark TD (2014) Aerobic scope does not predict theperformance of a tropical eurythermal fish at elevated tempera-tures. J Exp Biol 217: 244–251.

Overgaard J, Andersen JL, Findsen A, Pedersen PBM, Hansen K,Ozolina K, Wang T (2012) Aerobic scope and cardiovascular oxy-gen transport is not compromised at high temperatures in thetoad Rhinella marina. J Exp Biol 215: 3519–3526.

Pörtner HO (2001) Climate change and temperature-dependent bio-geography: oxygen limitation of thermal tolerance in animals.Naturwissenschaften 88: 137–146.

Pörtner HO, Farrell AP (2008) Physiology and climate change. Science322: 690–692.

Portner HO, Giomi F (2013) Nothing in experimental biology makessense except in the light of ecology and evolution – correspond-ence on J Exp Biol 216: 2771–2782. J Exp Biol 216: 4494–4495.

Pörtner HO, Knust R (2007) Climate change affects marine fishes throughthe oxygen limitation of thermal tolerance. Science 315: 95–97.

R Core Development Team (2013) R: a Language and Environment forStatistical Computing. R Foundation for Statistical Computing,Vienna. http://www.R-project.org

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11


by guest on Decem


Dow

nloaded from

Reidy SP, Nelson JA, Tang Y, Kerr SR (1995) Post-exercise metabolicrate in Atlantic cod and its dependence upon the method ofexhaustion. J Fish Biol 47: 377–386.

Reyes KK (2008) Reviewed distribution maps for Oncorhynchus mykiss(Rainbow trout), with modelled year 2100 native range map basedon IPCC A2 emissions scenario. www.aquamaps.org, version ofAugust 2013.

Rodgers GG, Tenzing P, Clark TD (2016) Experimental methods inaquatic respirometry: the importance of mixing devices andaccounting for background respirometry. J Fish Biol 88: 65–80.

Rodnick KJ, Gamperl AK, Lizars KR, Bennett MT, Rausch RN, Keeley ER(2004) Thermal tolerance and metabolic physiology among red-band trout populations in south-eastern Oregon. J Fish Biol 64:310–335.

Rummer JL, Couturier CS, Stecyk JAW, Gardiner NM, Kinch JP, NilssonGE, Munday PI (2014) Life on the edge: thermal optima for aerobicscope of equatorial reef fishes are close to current day tempera-tures. Glob Change Biol 20: 1055–1066.

Scarabello M, Heigenhauser GJF, Wood CM (1992) Gas exchange,metabolite status and excess postexercise oxygen consumptionafter repetitive bouts of exhaustive exercise in juvenile rainbowtrout. J Exp Biol 167: 155–169.

Sokolova IM, Frederich M, Bagwe R, Lannig G, Sukhotin AA (2012)Energy homeostasis as an integrative tool for assessing limits of

environmental stress tolerance in aquatic invertebrates. MarEnviron Res 79: 1–15.

Sonski AJ (1984) Comparison of heat tolerances of redband trout,Firehole River rainbow trout and Wytheville rainbow trout. TexasParks and Wildlife Department, Fisheries Research Station.

Sousa PM, Trigo RM, Aizpurua P, Nieto R, Gimeno L, Garcia-Herrera R(2011) Trends and extremes of drought indices throughout the 20thcentury in the Mediterranean. Nat Hazard Earth Sys Sci 11: 33–51.

Steffensen J (1989) Some errors in respirometry of aquatic breathers:how to avoid and correct for them. Fish Physiol Biochem 6: 49–59.

Svendsen MBS, Bushnell PG, Steffensen JF (2016) Design and setup ofintermittent-flow respirometry system for aquatic organisms J FishBiol 88: 26–50.

Swain DL, Tsiang M, Haugen M, Singh D, Charland A, Rajaratnam B,Diffenbaugh NS (2014) The extraordinary California drought of2013/2014: character, context, and the role of climate change. BullAm Meteorol Soc 95: S3.

Turlock Irrigation District and Modesto Irrigation District (2011) Pre-application document, No. 2299. Turlock, CA.

US Environmental Protection Agency (2003) EPA Region 10 Guidance forPacific Northwest State and Tribal Temperature Water QualityStandards. http://www.epa.gov/region10/pdf/water/final_temperature_guidance_2003.pdf

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12


by guest on Decem


Dow

nloaded from



ATTACHMENT C

HIGH THERMAL TOLERANCE OF A RAINBOW TROUT POPULATION NEAR ITS SOUTHERN RANGE LIMIT SUGGESTS

LOCAL THERMAL ADJUSTMENT

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Volume 4 • 2016 10.1093/conphys/cow057

Research article

High thermal tolerance of a rainbow troutpopulation near its southern range limitsuggests local thermal adjustmentChristine E. Verhille1, Karl K. English2, Dennis E. Cocherell1, Anthony P. Farrell3 and Nann A. Fangue1,*

1Department of Wildlife, Fish and Conservation Biology, University of California Davis, Davis, CA 95616, USA2LGL Limited, Sidney, British Columbia, Canada V8L 3Y83Department of Zoology and Faculty of Land and Food Systems, University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z4

*Corresponding author: Department of Wildlife, Fish and Conservation Biology, University of California Davis, Davis, CA 95616, USA.Tel: +1-530-752-4997. Email: [email protected]

Transformation of earth’s ecosystems by anthropogenic climate change is predicted for the 21st century. In many regions,the associated increase in environmental temperatures and reduced precipitation will have direct effects on the physio-logical performance of terrestrial and aquatic ectotherms and have already threatened fish biodiversity and important fish-eries. The threat of elevated environmental temperatures is particularly salient for members of the Oncorhynchus genusliving in California, which is the southern limit of their range. Here, we report the first assessments of the aerobic capacityof a Californian population of wild Oncorhynchus mykiss Walbaum in relationship to water temperature. Our field measure-ments revealed that wild O. mykiss from the lower Tuolumne River, California maintained 95% of their peak aerobic scopeacross an impressive temperature range (17.8–24.6°C). The thermal range for peak performance corresponds to local highriver temperatures, but represents an unusually high temperature tolerance compared with conspecifics and congenericspecies from northern latitudes. This high thermal tolerance suggests that O. mykiss at the southern limit of their indigen-ous distribution may be locally adjusted relative to more northern populations. From fisheries management and conserva-tion perspectives, these findings challenge the use of a single thermal criterion to regulate the habitat of the O. mykissspecies along the entirety of its distribution range.

Key words: aerobic scope, fish, metabolic rate, Oncorhynchus mykiss, swimming, temperature

Editor: Steven Cooke

Received 10 October 2016; Revised 24 October 2016; Editorial Decision 29 October 2016; accepted 1 November 2016

Cite as: Verhille CE, English KK, Cocherell DE, Farrell AP, Fangue NA (2016) High thermal tolerance of a rainbow trout population near itssouthern range limit suggests local thermal adjustment. Conserv Physiol 4(1): cow057; doi:10.1093/conphys/cow057.

IntroductionRainbow trout (Oncorhynchus mykiss Walbaum 1792) isregarded as a cold-water fish species with an indigenousrange stretching across an immense temperature gradient,from the subarctic climate region of the Bering Sea to theMediterranean climate region of Northern Baja California(Reyes, 2008). Despite this large temperature gradient and

distribution range, the optimal temperature range for wildO. mykiss aerobic performance capacity has been deter-mined only for indigenous populations inhabiting temperateclimates. Local adaptation of thermal performance existswithin the teleosts (Angilletta, 2009), but has never beenshown for wild O. mykiss populations across their nativerange. Without knowledge of the variation in thermal per-formance among populations of O. mykiss, fish conservation

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1© The Author 2016. Published by Oxford University Press and the Society for Experimental Biology.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

managers apply regulatory water temperature criteriaderived for higher latitude populations of O. mykiss for pro-tection of lower latitude populations.

The present study considered the thermal performance of apopulation of O. mykiss located in a river near the southernlimits of its native range and was prompted by a number ofrecent events. Foremost, global indicators show that 2014 and2015 were the warmest years on record for the earth’s climate(Blunden and Arndt, 2015; NOAA National Centers forEnvironmental Information, 2016). Animal populations, suchas Californian O. mykiss, which exist at the latitudinal ext-remes of their biogeographical range, are expected to experi-ence the most profound negative effects of such climatechanges (Lassalle and Rochard, 2009). Second, for a fish thattends to favour pristine, cold water in most of its native habi-tat, native O. mykiss populations inhabiting the extremelywarm summer temperatures of Californian rivers are evidenceof considerable phenotypic plasticity (or genetic variability)within the species, allowing acclimation (or adaptation) tomuch warmer environmental temperature regimes. Indeed,severe thermal exposures in southern Western Australia haveproduced a line of introduced, hatchery-reared O. mykiss(Morrissy, 1973; Molony, 2001; Molony et al., 2004) thatswim and feed at 26°C (Michael Snow, Department ofFisheries, Government of Western Australia, personal commu-nication) and retain 50% of their peak aerobic capacity at25°C (Chen et al., 2015). Interestingly, the founder populationfor this thermally tolerant hatchery strain was transplantedfrom California during the last century for recreational fisher-ies. Thus, with climate change continuing to shift baseline riv-er water quality and availability (Sousa et al., 2011; Swainet al., 2014), especially in central California, where the intensi-fication of weather extremes is triggering water crises andextreme droughts (Dettinger and Cayan, 2014), knowledge ofthe local thermal requirements of vulnerable key fish speciesbecomes ever more pressing (Moyle et al., 2011).

Fish can adjust to warmer habitat temperatures byrelocating to a cooler refuge (if available), thermally accli-mating or thermally adapting (Farrell and Franklin, 2016);responses that all operate at different time scales. Indeed, thesuggestion that fish might tailor their metabolic rate to habi-tat temperature has a long and strong history across a widerange of aquatic habitats and species (Fry, 1947, 1971; Brettand Groves, 1979; Elliott, 1982; Jobling, 1994; Hochachkaand Somero, 2002; Donelson et al., 2012). In fact, local ther-mal adaptation has been thoroughly characterized for otherfish species, such as stickleback populations (Barrett et al.,2011), temperate killifish (Fangue et al., 2006) and tropicalkillifish (McKenzie et al., 2013). Even within the genusOncorhynchus, Fraser River watershed populations of sock-eye salmon (O. nerka Walbaum 1792) have apparently tunedtheir thermal performance to meet the energetic needs of theironce-in-a-lifetime upstream migration (Farrell et al., 2008;Eliason et al., 2011, 2013). The ability of O. mykiss to accli-mate thermally is well documented (Myrick and Cech, 2000),

and there appears to be the genetic potential for thermaladaptation given the successful selective breeding of O. mykisslines that perform well at high temperatures (Australian lines,Molony et al., 2004; Japanese lines, Ineno et al., 2005).Nevertheless, assessments of the aerobic capacity in relation towater temperature of wild O. mykiss at the southern extent oftheir range in California are lacking. What is known for twoCalifornian strains of O. mykiss (Eagle Lake and MountShasta; Myrick and Cech, 2000) is that the thermal perform-ance curves for hatching success differ (Myrick and Cech,2001) despite similar upper thermal tolerance values (CTmax).In addition, the Eagle Lake and Mount Shasta strains of O.mykiss grew fastest at different acclimation temperatures (19and 22°C, respectively), but growth ceased at 25°C in bothstrains (Myrick and Cech, 2000).

The accumulating evidence for variation in thermal per-formance within and among Pacific salmon and rainbowtrout populations seems incongruous with the criteria used bythe US Environmental Protection Agency (EPA) to regulatewater temperatures. The EPA uses a regulatory 7 day averageof the daily water temperature maximum (7DADM) of 18°Cfor all juvenile O. mykiss over their entire native US rangefrom southern California into Alaska (US EnvironmentalProtection Agency, 2003). One way to bring greater insightinto population-specific thermal tolerance and to take localadaptation and acclimation into consideration for regulatorypurposes is to use a well-established non-lethal approach tostudy the thermal physiology of O. mykiss populations inhabit-ing unusually warm habitats. Therefore, we examined O. mykissthat inhabit the Tuolumne River below La Grange DiversionDam, which is the most downstream habitat for O. mykiss ina watershed that drains ~2500 km2 of the Western SierraNevada mountain range. This river reach is characterized bya longitudinal thermal gradient, which increases from 12°Cto occasionally as high as 26°C during summer warming overa ~25 km stretch of river. By measuring metabolic scope foractivity (Fry, 1947), we tested the hypothesis that O. mykissresiding below the La Grange Diversion Dam on theTuolumne River may be locally adapted to the summer habi-tat temperatures that can reach 26°C. Mechanistically, ourexperimental approach builds on a fish’s ultimate requirementto have the capacity to supply oxygen for all activities (e.g.for foraging, digestion, growth, migration, predator avoid-ance and reproduction). The capacity to provide oxygenbeyond basic needs is termed absolute aerobic scope (AAS),which, in field situations (e.g. Pörtner and Knust, 2007;Nilsson et al., 2009; Gardiner et al., 2010; Eliason et al., 2011;Rummer et al., 2014), can be estimated from the differencebetween routine metabolic rate (RMR) and maximal meta-bolic rate (MMR). Thus, by measuring RMR and MMR overa wide range of water temperatures, the portion of the tem-perature range where AAS (i.e. the capacity for aerobic activ-ity) is maximized can be defined. Such information is lackingfor wild O. mykiss in central California. For the presentstudy, a temporary respirometry laboratory was built besidethe Tuolumne River. This laboratory allowed wild juvenile

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2


O. mykiss to be tested at temperatures between 13 and 25°Cbefore they were returned to their original habitat within24 h, as required by the experimental permits.

This study has implications beyond the thermal needs forresident aquatic species because this segment of theTuolumne River is part of a watershed that provides munici-pal water to >2.4 million residents of the San Francisco BayArea and agricultural irrigation water to the Central Valley(Turlock Irrigation District and Modesto Irrigation District,2011). The recent drought in central California has left reser-voirs at historic lows (California Department of WaterResources, 2015) and has challenged the capacity to balancethe environmental water flow needs of aquatic biota with thehuman requirements from this watershed for domestic, agri-cultural and recreational use. Juvenile O. mykiss livingbelow the La Grange Diversion Dam have been observedexploiting summer Tuolumne River temperatures from 12 to26°C over 25 river km (HDR Engineering, Inc., 2014).There are no additional cool-water inputs (except for raresummer rains), resulting in progressive warming of the waterreleased from the Dam as it flows downstream. Establishingthe optimal temperature range for aerobic performance ofwild Californian O. mykiss will provide fish conservationmanagers with scientific support for temperature criteria thatallow for optimization of this balance between human andfish requirements.

Materials and methodsPermitting restrictions that influenced theexperimental designWild Tuolumne River O. mykiss were collected underNational Marine Fisheries Service Section 10 permit no.17913 and California Fish and Wildlife Scientific CollectingPermit Amendments. No distinction was made between resi-dent (rainbow trout) and anadromous (steelhead) life-historyforms. For permitting purposes, these fish are considered asESA-listed California Central Valley steelhead, O. mykiss.Fish collection (up to a maximum of 50 individuals) wasallowed only between river kilometer (RK) 84.0 and RK 63.6,and capture temperatures could not exceed 21.1°C. This per-mit allowed only two fish to be captured and tested each day,and all fish had to be returned to their original river habitat.Given that indirect fish mortality was limited to three fish, aprecautionary measure included testing fish at the highesttemperatures last (i.e. not randomly assigning test tempera-ture). Additionally, the permit restricted test temperatures to≤25°C. All experimental procedures were approved by theInstitutional Animal Care and Use Committee (protocol no.18196; the University of California Davis).

Fish collection, transport and holdingTwo wild O. mykiss were collected daily [a total of 44 fish;22.4 g (SEM = 1.78, range 10.5–79.6 g) and 125.7mm

(SEM = 2.88)] from four primary locations on the TuolumneRiver (Supplementary material, Fig. S1). The two fish wereimmediately scanned for a passive integrated transponder(PIT) tag to preclude re-testing a fish. The fish were transferreddirectly to a 13 litre container partly submerged in the riverbefore being driven to the streamside field laboratory(<20min) in insulated coolers filled with 25 litres of fresh riverwater. A water temperature logger (recording every 15min;Onset Computer Corporation, USA) remained with the fishuntil testing was completed and the fish was returned to theriver. At the field laboratory, located immediately downstreamfrom the La Grange Diversion Dam, fish were transferred toholding tanks (300 litres) filled with flow-through TuolumneRiver water (directly from the dam) that had passed through acoarse foam filter and then an 18 litre gas-equilibration col-umn for aeration (12.5–13.6°C, >80% air saturation). Thus,field-acclimatized fish were placed into the holding tankswithin 60–120min of capture and remained there for 60–180min before being transferred to one of two 5 litre auto-mated swim tunnel respirometers (Loligo, Denmark). Routineand maximal metabolic rates were then measured at tempera-tures between 13 and 25°C (1°C increments).

Swim tunnel respirometersThe swim tunnel respirometers received aerated TuolumneRiver water from an 80 litre temperature-controlled sump thatwas refreshed every 80–90min. Water temperature was regu-lated within ±0.5°C of the test temperature by passing sumpwater through a 9500 BTU Heat Pump (Model DSHP-7, AquaLogic Delta Star, USA) with a high-volume pump (modelSHE1.7, Sweetwater®, USA). Additionally, two proportionaltemperature controllers (model 72, YSI, USA) each ran an800W titanium heater (model TH-0800, Finnex, USA) locatedin the sump. The water temperature in the swim tunnels wasmonitored with a temperature probe connected through afour-channel Witrox oxygen meter (Loligo). All temperature-measuring devices were calibrated bi-weekly to ±0.1°C of aNational Institute of Standards and Technology certified glassthermometer. Ammonia build-up was prevented by zeolite inthe sump, which was replaced weekly. Water oxygen saturationin each swim tunnel was monitored continuously using a dip-ping probe mini oxygen sensor connected to AutoResp software(Loligo) through the Witrox system (Loligo). Video cameraswith infrared lighting (Q-See, QSC1352W, China) continuouslyrecorded (Panasonic HDMI DVD-R, DMR-EA18K, Japan) fishbehaviour in the swim tunnels, which were shaded by blackcloth to limit fish disturbance. A variable frequency drive motorgenerated laminar water flow through the swimming section(calibrated using a digital anemometer with a 30mm vanewheel flow probe; Hönzsch, Germany) in each swim tunnel.

Metabolic rate measurementRoutine and active metabolic rates of fish in the swim tunnelrespirometers were measured using intermittent respirometry

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3


(Steffensen, 1989; Cech, 1990; Chabot et al., 2016; Svendsenet al., 2016). The swim tunnel was automatically sealed duringmeasurements and flushed with fresh, aerated sump waterbetween measurements (AutoResp software and a DAQ-PAC-WF4 automated respirometry system, Loligo). Oxygenremoval from the water by the fish (in milligrams of oxygen)was measured for a minimal period of 2 min when the swimtunnel was sealed, without oxygen levels falling below 80%air saturation. No background oxygen consumption wasdetected without fish (performed at the end of each day withboth swim tunnels; Rodgers et al., 2016) even at the highest testtemperature (25°C). Each oxygen probe was calibrated weeklyat the test temperatures using 100% (aerated distilled water)and 0% (150 ml distilled water with 3 g dissolved Na2SO3) air-saturated water.

Oxygen uptake was calculated according to the followingformula:

⎡⎣ ⎤⎦{ }( )

( )= ( ) − ( ) × × ×

− −

− −t t V M T

Oxygen uptake in mg O kg min

O O ,

20.95 1

2 1 2 20.95 1

where O2(t1) is the oxygen concentration in the swim tunnel atthe beginning of the seal (in milligrams of oxygen per litre);O2(t2) is the oxygen concentration in the tunnel at the end ofthe seal (in milligrams of oxygen per litre); V is the volume ofthe swim tunnel (in litres); M is the mass of the fish (in killo-grams); and T is the duration of the measurement (in minutes).Allometric correction for variable body mass used the expo-nent 0.95, which is halfway between the life-stage-independentexponent determined for resting (0.97) and active (0.93) zebra-fish (Lucas et al., 2014).

Experimental protocolFish were introduced between 13.00 and 16.00 h each dayinto a swim tunnel at 13 ± 0.3°C, which was close to the riv-er temperature at which most fish were caught, and left for60min before a 60min training swim (Jain et al., 1997), dur-ing which water flow velocity was gradually increased to5–10 cm s−1 higher than when swimming started (typicallyat 30 cm s−1) and held for 50min before a 10min swim at50 cm s−1 (the anticipated maximal prolonged swimmingvelocity for a 150 mm fish at 13°C; Alsop and Wood, 1997).Recovery for 60min preceded the incremental increases inwater temperature (1°C per 30min) up to the test tempera-ture. Oxygen uptake (10–30min, depending on the test tem-perature, and followed by a 5–10min flush period) wascontinuously measured throughout the night until 07.00 h.Estimates of RMR for each of the 44 tested fish were calcu-lated by averaging the lowest four oxygen uptake measure-ments at the test temperature for the minimum 8 h overnightperiod (Chabot et al., 2016). Visual inspection of the videorecordings confirmed that fish were quiescent during thesemeasurements with the exception of three fish that were

discarded owing to consistent activity throughout the night(Crocker and Cech, 1997), which reduced the RMR mea-surements to 41 fish.

Critical swimming velocity and burst swimming protocols(Reidy et al., 1995; Killen et al., 2007; Clark et al., 2013;Norin and Clark, 2016) were used to determine MMR. Theybegan between 08.00 and 09.00 h and lasted 2–6 h. For thecritical swimming velocity test, water velocity was graduallyincreased until the fish continuously swam at 30 cm s−1 for 20min. Water velocity was incrementally increased every 20 minby 10% of the previous test velocity (3–6 cm s−1) until the fishwas no longer able to swim continuously and fell back tomake full body contact with the downstream screen of theswimming chamber. The fish recovered for 1 min at 13–17 cms−1, the lowest velocity setting of the swim tunnel, beforerestoring the final water velocity over a 2 min period andrestarting the 20 min timer. Fatigue was defined as when thefish made full body contact with the downstream screen of theswim tunnel a second time at the same test velocity or failed toresume swimming. Active metabolic rate was measured at eachtest velocity using a 3 min flush period and a 7–17 min meas-urement period. All fish swam for 20 min at one water vel-ocity, but almost 50% of the wild fish used their caudal fin toprop themselves on the downstream screen of the swim tunnelto avoid swimming faster, which required a secondary meas-urement of maximal metabolic rate using a burst swimmingprotocol. For the burst swimming protocol, tunnel velocitywas set to and held for 10 min at the highest critical swimmingvelocity test increment where that fish had continuously swum.Afterwards, water velocity was rapidly (over 10 s) increased to70–100 cm s−1, which invariably elicited burst swimmingactivity for 30 s or less, when water velocity exceeded 70 cms−1. This protocol was repeated multiple times for 5–10 min,while oxygen uptake was measured continuously. The MMRwas assigned to the highest active metabolic rate measuredwith the active respirometry methods. Occasionally, fish exhib-ited intense struggling behaviours with an even higher oxygenuptake, which was assigned MMR. The MMR was not esti-mated for four fish, which failed to swim and raise their meta-bolic rate appreciably with any of the methods, resulting in atotal of 37 fish with RMR and MMR measurements. Absoluteaerobic scope (AAS = MMR − RMR) and factorial aerobicscope (FAS = MMR/RMR) were calculated.

All fish recovered in the swim tunnel at a water velocityof 13–17 cm s−1 and at the test temperature for 1 h whilemeasuring oxygen uptake. Water temperature was thendecreased to ~13°C over a 30 min period before the fish wasremoved, measured, PIT tagged and put into a holding tankbefore release at the capture site. Fish were individuallyanaesthetized for <5 min with CO2 (2 Alka-Seltzer tabletsdissolved in 3 litres of river water) for morphometric mea-surements [fork length (FL), in millimetres; and body mass,in grams], condition factor calculation (CF = bodymass × 105/FL3), and PIT tagging. Half duplex PIT (OregonRFID) tags were placed into the abdominal cavity via a

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4


1 mm incision through the body wall, just off-centre of thelinea alba. All equipment was sterilized with NOLVASAN Sprior to tagging, and incisions were sealed with 3MVetBond. Revived fish were immediately transported to thecoolers filled with 13–15°C river water. At the release site,river water was gradually added to the cooler to equilibratethe fish to river water temperature at a rate of 1–2°C h−1

before fish were allowed to swim away voluntarily.

Measurements of tail beat frequencyThe tail beat frequency (TBF; number of tail beats per 10 s,reported in Hz) of fish swimming continuously and holdingstation without contacting the downstream screen of the res-pirometer was measured using the average of two or three10 s sections of video recordings played back at either one-quarter or one-eighth of real time. The TBF was then relatedto swimming speed and temperature. Tail beat frequencies ofundisturbed fish holding station in the Tuolumne River weremeasured from footage from underwater video camerasanchored within 1m of O. mykiss schools and left to recordfor up to 4 h (GoPro Hero 4). The TBFs were determinedusing the same methodology applied to respirometer videorecordings (n = 15 at 14°C and n = 1 at 20°C).

Data analysisA statistical model was fitted to individual data [performedin R (R Core Development Team, 2013) using the ‘lm’ func-tion] to determine the best relationships between the testtemperature and RMR, MMR, AAS and FAS. The statisticalmodel (linear, quadratic, antilogarithic base 2 and logarith-mic base 2 were tested) with the highest r2 and lowestresidual SE being reported. Confidence intervals and pre-dicted values based on the best-fit model were calculated inR using the ‘predict’ function. Variances around metabolicrate measurements are reported as 95% confidence intervals(CIs).

ResultsAs anticipated, basic oxygen needs (RMR) increased expo-nentially by 2.5-fold from 13 to 25°C (from 2.18 ± 0.45(95% CI) to 5.37 ± 0.41 mg O2 kg

−0.95 min−1). This thermalresponse was modelled by: RMR (in mg O2 kg−0.95

min−1) = 5.9513 – 0.5787 (temperature, in °C) + 0.0200(temperature, in °C)2 (P < 0.001, r2 = 0.798; Fig. 1A). TheMMR increased linearly by 1.7 times (from 6.62 ± 1.03 to11.22 ± 0.86 mg O2 kg−0.95 min−1) from 13 to 25°C. Thisthermal response was modelled by: MMR (in mg O2 kg−0.95

min−1) = 1.6359 + 0.3835 (temperature, in °C) (P < 0.001,r2 = 0.489; Fig. 1B). Given that MMR almost kept pace withthe thermal effect on RMR, AAS had a rather flat reactionnorm that was largely independent of the test temperaturerange. This thermal response was modelled by: AAS (in mgO2 kg−0.95 min−1) = −5.7993 + 1.1263 (temperature, in °C)− 0.0265 (temperature, in °C)2 (P = 0.060, r2 = 0.098;Fig. 1C). Using this model, peak AAS (6.15 ± 0.71mg O2

kg−0.95 min−1) was centred at 21.2°C. Nevertheless, theunexpected flat reaction norm meant that 95% of peak AASwas maintained from 17.8 to 24.6°C, which is a broad ther-mal window for peak AAS that extends well beyond the7DADM value of 18°C for O. mykiss.

Factorial aerobic scope is a useful metric of whetheror not a fish might have the required aerobic capacity toperform a specific activity, e.g. a doubling of RMR (i.e.FAS = 2) might be needed to digest a full meal properly(Jobling, 1981; Alsop and Wood, 1997; Fu et al., 2005; Luoand Xie, 2008). As expected, FAS decreased with tempera-ture (Clark et al., 2013), a thermal response modelled by:FAS = 2.1438 + 0.1744 (temperature, in °C) − 0.0070 (tem-perature, in °C)2 (P < 0.001, r2 = 0.344; Fig. 1D).

In addition, given the need to integrate AAS or FAS withinan ecological framework (see Overgaard et al., 2012; Clarket al., 2013; Farrell, 2013, 2016; Pörtner and Giomi, 2013;Ern et al., 2014; Norin et al., 2014), we used measurements ofTBF to estimate the oxygen cost required by a wild O. mykissto maintain station in the river currents of typical habitats inthe Tuolumne River. A steady TBF used for this activity atambient temperatures of 14 and 20°C was 2.94 and 3.40 Hz,respectively (see supplemental video, available online). Usingrespirometer swimming data to relate TBF to oxygen uptakeat 14 and 20°C (P < 0.001, r2 = 0.35; and P = 0.009,r2 = 0.33, respectively; Fig. 2), in-river TBF values of 2.94 and3.40 Hz corresponded to 3.26 and 5.43 mg O2 kg

−0.95 min−1,respectively. Therefore, we suggest that wild fish observedholding station in the Tuolumne River increased RMR by1.5 times at 14°C and by 2.0 times at 20°C, an activity thatwould use 49 and 67%, respectively, of the available FASmeasured at these two temperatures.

Fish recovery after exhaustive swimming tests was quickand without any visible consequences. The RMR at the endof the 60 min recovery period was either elevated by nomore than 20% or fully restored; an observation consistentwith previous laboratory studies of O. mykiss recovery (Jainet al., 1997; Jain and Farrell, 2003). Two fish tested at 25°Cwere the only exceptions. These two fish regurgitated theirgut contents during recovery and one then died abruptly.Inadvertent fish recapture provided some information on fishsurvival after being returned to the river. Six PIT-tagged fishwere recaptured at 1–11 days post-testing within 20 m oftheir original capture location; all were visually in good con-dition, and three of these fish had been tested at 23°C.

DiscussionThe present study is the first to consider the thermal responsefor an O. mykiss population so close to the southerly bound-ary of the natural distribution range for indigenousO. mykiss. We clearly show that 95% of peak AAS wasmaintained over an unexpectedly broad thermal window(17.8–24.6°C) and that all fish tested could maintain an

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5


FAS >2.0 up to 23°C. Moreover, we place these findingsinto an ecological context by suggesting that the level of FASat temperatures at least as high as 20°C may be more thanadequate to maintain station in the local water current of theTuolumne River and probably to digest a meal properly andoptimize growth, which is a very powerful integrator ofenvironmental, behavioural and physiological influences ona fish’s fitness. Moreover, fish were tested on site andreturned afterwards to the river, making the work locallyrelevant for the O. mykiss population, sensitive to conserva-tion needs and globally relevant by addressing the followingbroad question: are fish at the extreme edges of their biogeo-graphical range more physiologically tolerant because of thethermal extremes they experienced there?

The present results show good quantitative agreement withvarious previous studies with O. mykiss that have measured

some of the variables measured in the present study. Forexample, the 2.5-fold exponential increase in RMR from 13 to25°C (from 2.18 ± 0.45 (95% CI) to 5.37 ± 0.41 mg O2

kg−0.95 min−1) compares well with laboratory studies of RMRreported at 14°C (2.3–2.8 mg O2 kg−0.95 min−1; Myrick andCech, 2000) for 7 g Mount Shasta and Eagle Lake O. mykiss,and at 25°C (~6.5 mg O2 per kg

−0.95 min−1; Chen et al., 2015)for 30 g Western Australian O. mykiss. Therefore, concernsabout handling stress and specific dynamic action were min-imal. Likewise, MMR increased linearly by 1.7 times (from6.62 ± 1.03 to 11.22 ± 0.86 mg O2 kg

−0.95 min−1) from 13 to25°C, comparing well with previous laboratory measurementsof MMR reported at 15°C (2.8–8.7 mg O2 kg−0.95 min−1) for2–13 g O. mykiss (Scarabello et al., 1992, Alsop and Wood,1997) and with the peak MMR at 20°C (~11.13 mg O2 perkg−0.95 min−1) for Australian O. mykiss (Chen et al., 2015). Asa consequence of MMR nearly keeping pace with the thermal

Figure 1: The relationships between test temperature and the routine (RMR; A) and maximal metabolic rate (MMR; B) of Tuolumne RiverOncorhynchus mykiss. The three methods used to measure MMR (see Materials and methods section) are distinguished by different symbols.Absolute aerobic scope (AAS; C) and factorial aerobic scope (FAS; D) were derived from the metabolic rate measurements. Each data pointrepresents an individual fish tested at one temperature. These data were given a best-fit mathematical model (continuous line or curve), andthe 95% confidence intervals for each line are indicated by the shaded area. The RMR and FAS were smoothed to a polynomial fit of the formy = x + I(x2), where y is RMR or FAS, x is temperature, and I is a constant. The MMR and AAS were smoothed to a linear fit of the form y = x + c,where c is a constant. For RMR, degrees of freedom (d.f.) = 34, P < 0.001, residual standard error (RSE) = 0.561 and r2 = 0.798. For MMR,d.f. = 35, P < 0.001, RSE = 1.580 and r2 = 0.489. For AAS, d.f. = 35, P = 0.060, RSE = 1.490 and r2 = 0.098. For FAS, d.f. = 34, P < 0.001,RSE = 0.506 and r2 = 0.344. The asterisk indicates the one fish that died abruptly after the swimming test.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6


effect on RMR, AAS was largely independent of test tempera-ture. Directly comparing our AAS values with other studiesrevealed that our result for AAS at 15°C (5.10 mg O2 kg−0.95

min−1) was at the high end of previous laboratory measure-ments of AAS (1.8–5.8 mg O2 kg

−0.95 min−1) for O. mykiss at15°C (Scarabello et al., 1992, Alsop and Wood, 1997;McGeer et al., 2000), but lower than peak AAS (~7.3 mg O2

per kg−0.95 min−1) at 20°C in Australian O. mykiss (Chenet al., 2015). Likewise, our FAS values were bracketed byvalues obtained in previous laboratory studies. At 24°C, FAS(2.13 ± 0.33) was greater than that reported at 25°C (1.8) forWestern Australian O. mykiss (Chen et al., 2015), but com-pared with FAS values for juvenile rainbow trout (1.8–5.8) at13°C (Scarabello et al., 1992, Alsop and Wood, 1997,McGeer et al., 2000), our FAS at 13°C (3.32 ± 0.41) was inthe middle of the range.

To place the present data for Californian O. mykiss intoperspective, we have compared (Fig. 3) their reaction normwith those published for juveniles of northern O. mykiss (datafrom Fry, 1948) and Australian hatchery-selected O. mykiss

(data from Chen et al., 2015), as well as adult northern popula-tions of selected Pacific salmon populations (data from Leeet al., 2003a and Eliason et al., 2011). Among the native O.mykiss populations, the Lower Tuolumne River juvenileCalifornian O. mykiss are likely to experience the highest tem-peratures during summer (up to 26°C), although the introducedAustralian O. mykiss population had experienced selectiontemperatures ≥25°C (Chen et al., 2015). Notably, AAS at 24°C for Tuolumne River O. mykiss is greater than other O.mykiss populations and only bettered by the Chilko sockeyesalmon population, one of several sockeye salmon populationsthat are known to have a peak AAS at the modal temperaturefor their upstream spawning migration (Eliason et al., 2011,2013). Thus, the present data are in line with evidence of intra-specific matching of metabolic rate to local water temperatures

Figure 2: The relationship between tail beat frequency (TBF; in hertz)and metabolic rate (MR; in mg O2 kg

−0.95 min−1) measured whenTuolumne River Oncorhynchus mykiss were swimming continuouslyin a swim tunnel at 14 (A) or 20°C (B). The continuous black linerepresents the linear regression based on the data for n = 7 fish at14°C and n = 5 fish at 20°C. The vertical dashed lines represent theestimated TBF (2.94 Hz at 14°C and 3.40 Hz at 20°C) taken fromvideos of O. mykiss maintaining station in a water current in theirnormal Tuolumne River habitat. At 14°C, the relationship betweenTBF and MR followed the equation MR = 0.75TBF + 1.05, withdegrees of freedom (d.f.) = 41, P < 0.001, residual standard error(RSE) = 1.27 and r2 = 0.35. According to this formula, the MR for theTBF measured in the river (2.943 Hz) at 14°C was estimated to be3.26 mg O2 kg

−0.95 min−1. At 20°C, the relationship between TBF andMR followed the equation MR = 1.04TBF + 1.89, with d.f. = 15,P = 0.009, RSE = 1.29 and r2 = 0.33. According to this formula, the MRfor the TBF measured in the river at 20°C (3.402 Hz) was estimated tobe 5.43 mg O2 kg

−0.95 min−1.

Figure 3: Absolute aerobic scope (AAS) for three strains ofOncorhynchus mykiss, i.e. a northern strain (Fry, 1948), an Australianstrain (Chen et al., 2015) and the California strain reported in thismanuscript, compared with AAS measurements of Chehalis Cohosalmon (Oncorhynchus kisutch Walbaum 1792) and Gates Creek,Weaver Creek (Oncorhynchus nerka Walbaum 1792; Lee et al., 2003a)and Chilko Creek sockeye salmon (Eliason et al., 2011). The best-fitline of the relationship between AAS and temperature of the speciesand populations from other publications was predicted using asecond-order polynomial linear regression performed on the raw data(Lee et al., 2003a; Chen et al., 2015) or data extracted from plots (Fry,1948; Eliason et al., 2011) from the original publications. Coefficientestimates from the linear regression analysis were then used todetermine the peak aerobic scope and the temperaturescorresponding to the peak and 95% of peak AAS.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7


within the Oncorhynchus genus. Although the peak AAS of theAustralian O. mykiss population was 50% greater than for theother two O. mykiss populations, Tuolumne River O. mykisshad the broadest and highest thermal window (17.8–24.6°C)among the O. mykiss populations (20.5–22.4°C from Fry,1948; and 12.8– 18.6°C from Chen et al. 2015).

Whether the matching of Tuolumne River O. mykiss meta-bolic performance to local habitat temperatures is a result ofthermal acclimation or local adaption, as in the WesternAustralian O. mykiss, will need study well beyond the presentwork. Thermal acclimation usually results in fish performingbetter at the new temperature. For example, thermal acclima-tion offsets the effect of acute warming on RMR in 5–7 gMount Shasta and Eagle Lake O. mykiss (2.3–2.8 mg O2

kg−0.95 min−1 at 14°C and 2.9–3.1 mg O2 kg−0.95 min−1 at25°C; Myrick and Cech, 2000), which would normally doubleRMR over this temperature range, as observed here. Warmacclimation can also increase upper thermal tolerance limits,as it did for four anadromous Great Lakes populations ofjuvenile O. mykiss (Bidgood and Berst, 1969). Given that ourfish were captured at and, presumably, thermally acclimatizedto between 14 and 17°C, it would be of interest to test wildfish with a warmer thermal acclimation history. But evenwithout thermal acclimation, the present data suggest thatTuolumne River O. mykiss and those for northern O. mykiss(Fry, 1948; see Fig. 3) have the aerobic capacity temporarily,if not regularly, to exploit temperatures well above 18°C,which is the upper thermal limit suggested by EPA guidancedocuments (US Environmental Protection Agency, 2003).

Nevertheless, we caution that such local tailoring may notbe evident in all salmonid species. For example, the thermalphysiology of Atlantic salmon (Salmo salar Linnaeus 1758)from northern and southern extremes of their European rangedid not show any major difference (Anttila et al., 2014). Allthe same, a sub-species of redband trout (O. mykiss gairdneri),which are apparently adapted to high summer temperatures ofNorth American desert streams (Narum et al., 2010; Narumand Campbell, 2015), are likewise capable of high levels ofswimming performance up to 24°C (Rodnick et al., 2004) andhigher swimming performance for a warm vs. a cool creekpopulation (Gamperl et al., 2002). Redband trout have beenobserved actively feeding at 27–28°C (Sonski, 1984; Behnke,2010), but thermal selection of wild redband trout is centredbetween 13 (Gamperl et al., 2002) and 17°C (Dauwalter et al.,2015). How O. mykiss behaviourally exploit the steep sum-mer thermal gradient in the Tuolumne River below the LaGrange Diversion Dam (from 12 to 26°C over 25 km; HDREngineering, Inc., 2014) is another unknown. Even withoutthese important details, Tuolumne River O. mykiss appearphysiologically to be tolerant of the thermal extremes theyexperience.

The capacity of a fish to deliver oxygen to support activ-ities in water of varying quality is a concept originally intro-duced for fishes >60 years ago (Fry, 1947). The oxygen- and

capacity-limited thermal tolerance hypothesis broadens thisconcept and provides a mechanistic explanation (Pörtner,2001; Pörtner and Farrell, 2008; Deutsch et al., 2015), but iscurrently under debate (Overgaard et al., 2012; Clark et al.,2013; Farrell, 2013; Pörtner and Giomi, 2013; Ern et al.,2014; Norin et al., 2014). An accepted fact is that a metabolicload from an environmental factor (e.g. temperature) canincrease the oxygen cost for living (i.e. RMR). Consequently,like all other temperature studies with fish, the magnitude ofthe 2.5-fold increase in RMR observed here over a 12°C tem-perature range (between 13 and 25°C) was expected.However, temperature did not limit MMR, which increasedlinearly with acute warming, and the peak MMR was notresolved. The statistical models, which were based on individ-ual responses and 1°C temperature increments from 13 to25°C, predicted a peak AAS at 21.2°C for Tuolumne RiverO. mykiss and a FAS >2.0 up to 23°C. As the allocation ofenergy and trade-offs are recognized and fundamental tenantsof ecological physiology, especially in fishes (Sokolova et al.,2012), we suggest that being able to at least double RMR hasecological relevance for two behaviours that are likely to influ-ence survival of O. mykiss, maintaining station in a flowingriver and processing a large meal.

Snorkeling in the Tuolumne River provided visual obser-vations of O. mykiss maintaining station in the river currentfor prolonged periods that were punctuated by hiding underthe river bank and by darting behaviours to capture preyand to protect their position. Maintaining station required asteady TBF similar to the situation in the swim tunnel respir-ometer, which allowed us to estimate a metabolic cost ofmaintaining station in typical Tuolumne River habitats at 14and 20°C (a 1.5- to 2-fold increase in RMR) and the aerobicscope available for additional activities (1.7–2 times RMR).Although darting behaviours are likely to be fuelled anaer-obically, O. mykiss must (and were clearly able to) repay thepost-exercise excess oxygen debt (Lee et al., 2003b) whilemaintaining station in the river current. The rapid recoveryof RMR after exhaustive exercise in the swim tunnel suggeststhat O. mykiss had the capacity to repay post-exercise excessoxygen debt rapidly at temperatures as high as 24°C.Although digestion of a meal at high temperatures proceedsmore rapidly and with a higher peak metabolic rate, the totaloxygen cost of the meal remains similar. Thus, fish can the-oretically eat more frequently and potentially grow faster ata higher temperature provided there is a sufficient FAS fordigestion within the overall AAS. Given that peak metabolicrate during digestion of a typical meal for a salmonid doesnot necessarily double RMR at the temperatures used here(e.g. Jobling, 1981; Alsop and Wood, 1997; Fu et al., 2005;Luo and Xie, 2008), an FAS value of 2 should be a reason-able index, and all O. mykiss tested had this capacity up to23°C. Indeed, the fish were apparently feeding well in the riv-er, given a high condition factor (1.1 SEM = 0.01), the faecaldeposits found in the swim tunnel and two fish regurgitatingmeals when tested at 25°C. Meal regurgitation would beconsistent with an oxygen limitation, given that aquatic

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8


hypoxia impairs digestion in O. mykiss (Eliason and Farrell,2014). Indeed, feeding and growth are suppressed at supra-optimal temperatures (Hokanson et al., 1977; Brett and Groves,1979; Elliott, 1982; Myrick and Cech, 2000, 2001). Takentogether, these data suggest that Tuolumne River O. mykisswere doing well in their habitat and had the aerobic capacityto do so.

Our metabolic measurements, which show good quantita-tive agreement with controlled laboratory O. mykiss studies,represent a major challenge to the use of a single thermal cri-terion to regulate O. mykiss habitat when determining conser-vation criteria along the entire Pacific coast and perhapselsewhere. The 7DADM of 18°C for O. mykiss draws heavilyon a growth study performed in Minnesota (Hokanson et al.,1977). Therefore, it will be important to examine whether thepeak AAS at 21.2°C for Tuolumne River O. mykiss is asso-ciated with a peak growth rate. In this regard, the peakgrowth rate of another Californian O. mykiss population (theMount Shasta strain) occurred at acclimation temperatures(19–22°C; Myrick and Cech, 2000) above the 7DADM andwithin the thermal window for 95% peak AAS for TuolumneRiver O. mykiss. The Mount Shasta O. mykiss strain alsostopped growing at 25°C, the same temperature at which FASfor Tuolumne River O. mykiss approached 2. In contrast, theCalifornian Eagle Lake O. mykiss strain grew fastest at 19°Cand lost weight at 25°C (Myrick and Cech, 2000). Thus, theMount Shasta and Tuolumne River O. mykiss populationsare better able to acclimate thermally to temperatures >20°Cthan the Eagle Lake strain. With clear evidence that aCalifornia strain of O. mykiss can grow faster at acclimationtemperatures >18°C and that strains may differ in their theoptimal temperature for growth by as much as 3°C, there is aprecedent that local populations of O. mykiss can performwell above 18°C. Our findings also highlight the need forfuture experiments that consider replicate populations fromthroughout the species range to assess how widespread intra-specific variation in aerobic scope in O. mykiss might be.Continual development and refinement of the metrics used tobest inform regulatory criteria should be an ongoing pursuit,particularly if regional standards are to be implemented and ifthe criteria move away from what may now be consideredconservative. Probabilistic modelling approaches associatedwith a diversity of water temperature standards should bedeveloped in order for managers to understand the balancebetween standards that are conservative compared with thosethat are more risky.

The capacity to balance the essential environmentalrequirements of aquatic biota with human requirements isbecoming increasingly challenging across the globe because ofrecent increases in severe drought and record high temperatureoccurrences, a trend that climate change models project willcontinue. We suggest that broadly applying regulatory criteria,such as the 18°C 7DADM criterion for Pacific Northwest O.mykiss populations, to all North American O. mykiss is nolonger realistic and, in the present case, overly conservative.

The high degree of thermal plasticity discovered here for theTuolumne River O. mykiss population, which corresponds tolocal thermal conditions, adds to the accumulating evidence ofthe capacity for local adaptation among populations withinthe Oncorhynchus genus, including O. mykiss. Importantly,this work clearly illustrates that, owing to thermal plasticity,broad application of a single temperature criterion for fishprotection and conservation is not scientifically supported,especially for fish populations at the extreme limits of the spe-cies’ indigenous range.

Supplementary materialSupplementary material is available at ConservationPhysiology online.

AcknowledgementsWe thank Stillwater Sciences and FISHBIO for fish collectionassistance, and the Turlock Irrigation District for the use offacilities. We also thank Steve Boyd, John Devine, JennaBorovansky, Jason Guignard, Andrea Fuller and RonYoshiyama. Fieldwork was supported by Trinh Nguyen,Sarah Baird, James Raybould, Mike Kersten, Jeremy Pombo,Shane Tate, Patrick Cuthbert and Mike Phillips. All experi-ments followed ethical guidelines approved by the Universityof California Davis’ Institutional Animal Care and UseCommittee (protocol no. 18196). Tuolumne River O. mykisswere collected under National Marine Fisheries ServiceSection 10 permit no. 17913 and California Fish and WildlifeScientific Collecting Permit Amendments. We thank Ken Zilligand Dr Jamilynn Poletto for valuable manuscript feedback.

FundingThis work was supported by the Turlock Irrigation District,the Modesto Irrigation District, the City and County of SanFrancisco and UC Davis Agricultural Experiment Station(grant 2098-H to N.A.F.). A.P.F. holds a Canada ResearchChair and is funded by National Sciences and Energy ResearchCouncil of Canada.

ReferencesAlsop D, Wood C (1997) The interactive effects of feeding and exercise

on oxygen consumption, swimming performance and proteinusage in juvenile rainbow trout (Oncorhynchus mykiss). J Exp Biol10: 2337–2346.

Angilletta MJ (2009) Thermal Adaptation: a Theoretical and EmpiricalSynthesis. Oxford University Press, Oxford, UK.

Anttila K, Couturier CS, Overli O, Johnsen A, Marthinsen G, Nilsson GE,Farrell AP (2014) Atlantic salmon show capability for cardiac accli-mation to warm temperatures. Nat Commun 5: 4252.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9


Barrett RDH, Paccard A, Healy TM, Bergek S, Schulte PM, Schluter D,Rogers SW (2011) Rapid evolution of cold tolerance in stickleback.Proc Biol Sci 278: 233–238.

Behnke R (2010) Trout and Salmon of North America. The Free Press,New York, NY.

Bidgood BF, Berst AH (1969) Lethal temperatures for Great Lakes rain-bow trout. J Fish Res Board Can 26: 456–459.

Blunden J, Arndt DS (2015) State of the climate in 2014. Bul AmMeteor Soc 96: S1–S267.

Brett JR, Groves TDD (1979) Physiological Energetics. In Hoar WS,Randall DJ, Brett JR, eds, Fish Physiology Volume VIII Bioenergeticsand Growth, Academic Press, New York, NY, pp 279–352.

California Department of Water Resources (2015) Reservoir water stor-age. http://cdec.water.ca.gov/cgi-progs/reservoirs/STORAGEW

Cech JJ Jr (1990) Respirometry. In CB Schreck, PB Moyle, eds. Methods forFish Biology. American Fisheries Society, Bethesda, MD, pp 335–362.

Chabot D, Steffensen JF, Farrell AP (2016) The determination of stand-ard metabolic rate in fishes. J Fish Biol 88: 81–12.

Chen JQ, Snow M, Lawrence CS, Church AR, Narum SR, Devlin RH,Farrell AP (2015) Selection for upper thermal tolerance in rainbowtrout (Oncorhynchus mykiss Walbaum). J Exp Biol 218: 803–812.

Clark TD, Sandblom E, Jutfelt F (2013) Aerobic scope measurements offishes in an era of climate change: respirometry, relevance andrecommendations. J Exp Biol 216: 2771–2782.

Crocker CE, Cech JJ Jr (1997) Effects of environmental hypoxia on oxy-gen consumption rate and swimming activity in juvenile whitesturgeon, Acipenser transmontanus, in relation to temperature andlife intervals. Environ Biol Fish 50: 759–769.

Dauwalter DC, Fesenmyer KA, Bjork R (2015) Using aerial imagery tocharacterize redband trout habitat in a remote desert landscape.Trans Am Fish Soc 8487: 1322–1339.

Dettinger M, Cayan DR (2014) Drought and the California Delta—amatter of extremes. San Francisco Estuary and Watershed Sci 12:1–6.

Deutsch C, Ferrel A, Seibel B, Pörtner HO, Huey RB (2015)Ecophysiology. Climate change tightens a metabolic constraint onmarine habitats. Science 348: 1132–1135.

Donelson JM, Munday PL, McCormick MI, Pitcher CR (2012) Rapidtransgenerational acclimation of a tropical reef fish to climatechange. Nat Clim Change 2: 30–32.

Eliason E, Farrell AP (2014) Effect of hypoxia on specific dynamicaction and postprandial cardiovascular physiology in rainbow trout(Oncorhynchus mykiss). Comp Biochem Physiol A Mol Integr Physiol171: 44–50.

Eliason EJ, Clark TD, Hague MJ, Hanson LM, Gallagher ZS, Jeffries KM,Gale MK, Patterson DA, Hinch SG, Farell AP (2011) Differences inthermal tolerance among sockeye salmon populations. Science332: 109–112.

Eliason EJ, Wilson SM, Farrell AP, Cooke SJ, Hinch SG (2013) Low car-diac and aerobic scope in a costal population of sockeye salmon,Oncorhynchus nerka, with a short upriver migration. J Fish Biol 82:2104–2112.

Elliott JM (1982) The effects of temperature and ration size on thegrowth and energetics of salmonids in captivity. Comp BiochemPhysiol B Comp Biochem 73: 81–91.

Ern R, Huong DTT, Phuong NT, Wang T, Bayley M (2014) Oxygen deliv-ery does not limit thermal tolerance in a tropical eurythermalcrustacean. J Exp Biol 217: 809–814.

Fangue NA, Hofmeister M, Schulte PM (2006) Intraspecific variation inthermal tolerance and heat shock protein gene expression in com-mon killifish, Fundulus heteroclitus. J Exp Biol 209: 2859–2872.

Farrell AP (2013) Aerobic scope and its optimum temperature: clarify-ing their usefulness and limitations – correspondence on J. Exp.Biol. 216, 2771–2772. J Exp Biol 216: 4493–4494.

Farrell AP (2016) Pragmatic perspective on aerobic scope: peaking,plummeting, pejus and apportioning. J Fish Biol 88: 322–343.

Farrell AP, Franklin CE (2016) Recognizing thermal plasticity in fish.Science 351: 132–133.

Farrell AP, Hinch SG, Cooke SJ, Patterson DA, Crossin GT, Lapointe M,Mathes MT (2008) Pacific salmon in hot water: applying aerobicscope models and biotelemetry to predict the success of spawn-ing migrations. Physiol Biochem Zool 81: 697–708.

Fry FEJ (1947) Effects of the environment on animal activity.Publications of the Ontario Fisheries Research Laboratory 55: 1–62.

Fry FEJ (1948) Temperature relations of salmonids. Proc Can CommFreshwater Fish Res 1st Meeting. App. D. 1–6.

Fry FEJ (1971) The effect of environment factors on the physiology offish. Fish Physiol 6: 1–98.

Fu SJ, Xie XJ, Cao ZD (2005) Effect of meal size on postprandial meta-bolic response in southern catfish (Silurus meridionalis). CompBiochem Physiol A Mol Integr Physiol 140: 445–451.

Gamperl AK, Rodnick KJ, Faust HA, Venn EC, Bennett MT, CrawshawLI, Keeley ER, Powell MS, Li HW (2002) Metabolism, swimmingperformance, and tissue biochemistry of high desert redbandtrout (Oncorhynchus mykiss ssp.): evidence for phenotypic differ-ences in physiological function. Physiol Biochem Zool 75:413–431.

Gardiner NM, Munday PL, Nilsson GE (2010) Counter-gradient vari-ation in respiratory performance of coral reef fishes at elevatedtemperatures. PLoS One 5: e13299.

HDR Engineering, Inc. (2014) In-river diurnal temperature variation.Don Pedro Project FERC No. 2299.

Hochachka PW, Somero GN (2002) Biochemical Adaptation. OxfordUniversity Press, Oxford, UK.

Hokanson KEF, Kleiner CF, Thorslund TW (1977) Effects of constanttemperatures and diel temperature fluctuations on specific growth

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10


and mortality rates and yield of juvenile rainbow trout, Salmogairdneri. J Fish Res Board Can 34: 639–648.

Ineno T, Tsuchida S, Kanda M, Watabe S (2005) Thermal tolerance of arainbow trout Oncorhynchus mykiss strain selected by high-temperature breeding. Fish Sci 71: 767–775.

Jain KE, Farrell AP (2003) Influence of seasonal temperature on therepeat swimming performance of rainbow trout Oncorhynchusmykiss. J Exp Biol 206: 3569–3579.

Jain KE, Hamilton JC, Farrell AP (1997) Use of a ramp velocity test tomeasure critical swimming speed in rainbow trout (Oncorhynchusmykiss). Comp Biochem Physiol A Mol Integr Physiol 117: 441–444.

Jobling M (1981) The influences of feeding on the metabolic rate offishes: a short review. J Fish Biol 18: 385–400.

Jobling M, Baardvik BM (1994) The influence of environmental manipula-tions on inter–and intra–individual variation in food acquisition andgrowth performance of Arctic charr, Salvelinus alpinus. J Fish Biol 44:1069–1087.

Killen SS, Costa I, Brown JA, Gamperl AK (2007) Little left in the tank:metabolic scaling in marine teleosts and its implications for aer-obic scope. Proc Biol Sci 274: 431–438.

Lassalle G, Rochard E (2009) Impact of twenty-first century climatechange on diadromous fish spread over Europe, North Africa andthe Middle East. Glob Change Biol 15: 1072–1089.

Lee CG, Farrell AP, Lotto A, MacNutt MJ, Hinch SG, Healey MC (2003a)The effect of temperature on swimming performance and oxygenconsumption in adult sockeye (Oncorhynchus nerka) and coho(O. kisutch) salmon stocks. J Exp Biol 206: 3239–3251.

Lee CG, Farrell AP, Lotto A, Hinch SG, Healey MC (2003b) Excess post-exercise oxygen consumption in adult sockeye (Oncorhynchus ner-ka) and coho (O. kisutch) salmon following critical speed swim-ming. J Exp Biol 206: 3253–3260.

Lucas J, Schouman A, Plyphout L, Cousin X, LeFrancois C (2014)Allometric relationship between body mass and aerobic metabol-ism in zebrafish Danio rerio. J Fish Biol 84: 1171–1178.

Luo XP, Xie XJ (2008) Effects of temperature on the specific dynamicaction of the southern catfish, Silurus meridionalis. Comp BiochemPhysiol A Mol Integr Physiol 149: 150–156.

McGeer JC, Szebedinsky C, McDonald DG, Wood CM (2000) Effects ofchronic sublethal exposure to waterborne Cu, Cd or Zn in rainbowtrout. I: Iono-regulatory disturbance and metabolic costs. AquatToxicol 50: 231–243.

McKenzie DJ, Estivales G, Svendsen JC, Steffensen JF, Agnèse JF(2013) Local adaptation to altitude underlies divergent thermalphysiology in tropical killifishes of the genus Aphyosemion. PLoSOne 8: e54345.

Molony BW (2001). Environmental requirements and tolerances ofrainbow trout (Oncorhynchus mykiss) and brown trout (Salmo trut-ta) with special reference to Western Australia: a review. FisheriesResearch Reports No. 130. Western Australia, p 28.

Molony BW, Church AR, Maguire GB (2004) A comparison of the heattolerance and growth of a selected and non-selected line of rain-bow trout, Oncorhynchus mykiss, in Western Australia. Aquaculture241: 655–665.

Morrissy NM (1973) Comparison of strains of Salmo gairdneriRichardson from New South Wales, Victoria and Western Australia.Aust Soc Limn Bull 5: 11–20.

Moyle PB, Katz JVE, Quinones RM (2011) Rapid decline of California’snative inland fishes: a status assessment. Biol Conserv 144: 2414–2423.

Myrick CA, Cech JJ Jr (2000) Temperature influences on California rainbowtrout physiological performance. Fish Physiol Biochem 22: 245–254.

Myrick CA, Cech JJ Jr (2001) Temperature Effects on Chinook Salmonand Steelhead: a Review Focusing on California’s Central ValleyPopulations. Bay-Delta Modeling Forum, Technical Publication 01-1.

Narum SR, Campbell NR (2015) Transcriptomic response to heat stressamong ecologically divergent populations of redband trout. BMCGenomics 16: 103.

Narum SR, Campbell NR, Kozfkay CC, Meyer KA (2010) Adaptation ofredband trout in desert and montane environments. Mol Ecol 19:4622–4637.

Nilsson GE, Crawley N, Lunde IG, Munday PI (2009) Elevated tempera-ture reduces the respiratory scope of coral reef fishes. GlobChange Biol 15: 1405–1412.

NOAA National Centers for Environmental Information (2016) State ofthe Climate: Global Analysis for December 2015. http://www.ncdc.noaa.gov/sotc/global/201512.

Norin T, Clark TD (2016) Measurement and relevance of maximummetabolic rate in fishes. J Fish Biol 88: 122–151.

Norin T, Malte H, Clark TD (2014) Aerobic scope does not predict theperformance of a tropical eurythermal fish at elevated tempera-tures. J Exp Biol 217: 244–251.

Overgaard J, Andersen JL, Findsen A, Pedersen PBM, Hansen K,Ozolina K, Wang T (2012) Aerobic scope and cardiovascular oxy-gen transport is not compromised at high temperatures in thetoad Rhinella marina. J Exp Biol 215: 3519–3526.

Pörtner HO (2001) Climate change and temperature-dependent bio-geography: oxygen limitation of thermal tolerance in animals.Naturwissenschaften 88: 137–146.

Pörtner HO, Farrell AP (2008) Physiology and climate change. Science322: 690–692.

Portner HO, Giomi F (2013) Nothing in experimental biology makessense except in the light of ecology and evolution – correspond-ence on J Exp Biol 216: 2771–2782. J Exp Biol 216: 4494–4495.

Pörtner HO, Knust R (2007) Climate change affects marine fishes throughthe oxygen limitation of thermal tolerance. Science 315: 95–97.

R Core Development Team (2013) R: a Language and Environment forStatistical Computing. R Foundation for Statistical Computing,Vienna. http://www.R-project.org

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11


Reidy SP, Nelson JA, Tang Y, Kerr SR (1995) Post-exercise metabolicrate in Atlantic cod and its dependence upon the method ofexhaustion. J Fish Biol 47: 377–386.

Reyes KK (2008) Reviewed distribution maps for Oncorhynchus mykiss(Rainbow trout), with modelled year 2100 native range map basedon IPCC A2 emissions scenario. www.aquamaps.org, version ofAugust 2013.

Rodgers GG, Tenzing P, Clark TD (2016) Experimental methods inaquatic respirometry: the importance of mixing devices andaccounting for background respirometry. J Fish Biol 88: 65–80.

Rodnick KJ, Gamperl AK, Lizars KR, Bennett MT, Rausch RN, Keeley ER(2004) Thermal tolerance and metabolic physiology among red-band trout populations in south-eastern Oregon. J Fish Biol 64:310–335.

Rummer JL, Couturier CS, Stecyk JAW, Gardiner NM, Kinch JP, NilssonGE, Munday PI (2014) Life on the edge: thermal optima for aerobicscope of equatorial reef fishes are close to current day tempera-tures. Glob Change Biol 20: 1055–1066.

Scarabello M, Heigenhauser GJF, Wood CM (1992) Gas exchange,metabolite status and excess postexercise oxygen consumptionafter repetitive bouts of exhaustive exercise in juvenile rainbowtrout. J Exp Biol 167: 155–169.

Sokolova IM, Frederich M, Bagwe R, Lannig G, Sukhotin AA (2012)Energy homeostasis as an integrative tool for assessing limits of

environmental stress tolerance in aquatic invertebrates. MarEnviron Res 79: 1–15.

Sonski AJ (1984) Comparison of heat tolerances of redband trout,Firehole River rainbow trout and Wytheville rainbow trout. TexasParks and Wildlife Department, Fisheries Research Station.

Sousa PM, Trigo RM, Aizpurua P, Nieto R, Gimeno L, Garcia-Herrera R(2011) Trends and extremes of drought indices throughout the 20thcentury in the Mediterranean. Nat Hazard Earth Sys Sci 11: 33–51.

Steffensen J (1989) Some errors in respirometry of aquatic breathers:how to avoid and correct for them. Fish Physiol Biochem 6: 49–59.

Svendsen MBS, Bushnell PG, Steffensen JF (2016) Design and setup ofintermittent-flow respirometry system for aquatic organisms J FishBiol 88: 26–50.

Swain DL, Tsiang M, Haugen M, Singh D, Charland A, Rajaratnam B,Diffenbaugh NS (2014) The extraordinary California drought of2013/2014: character, context, and the role of climate change. BullAm Meteorol Soc 95: S3.

Turlock Irrigation District and Modesto Irrigation District (2011) Pre-application document, No. 2299. Turlock, CA.

US Environmental Protection Agency (2003) EPA Region 10 Guidance forPacific Northwest State and Tribal Temperature Water QualityStandards. http://www.epa.gov/region10/pdf/water/final_temperature_guidance_2003.pdf

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12




ATTACHMENT D

OXYGEN- AND CAPACITY-LIMITED THERMAL TOLERANCE: BRIDGING ECOLOGY AND PHYSIOLOGY

COMMENTARY

Oxygen- and capacity-limited thermal tolerance: bridging ecologyand physiologyHans-O. Portner*, Christian Bock and Felix C. Mark

ABSTRACTObservations of climate impacts on ecosystems highlight the need foran understanding of organismal thermal ranges and their implicationsat the ecosystem level. Where changes in aquatic animal populationshave been observed, the integrative concept of oxygen- and capacity-limited thermal tolerance (OCLTT) has successfully characterisedthe onset of thermal limits to performance and field abundance.The OCLTT concept addresses the molecular to whole-animalmechanisms that define thermal constraints on the capacity foroxygen supply to the organism in relation to oxygen demand. Theresulting ‘total excess aerobic power budget’ supports an animal’sperformance (e.g. comprising motor activity, reproduction and growth)within an individual’s thermal range. The aerobic power budget is oftenapproximated through measurements of aerobic scope for activity(i.e. themaximum difference between resting and the highest exercise-induced rate of oxygen consumption), whereas most animals in thefield rely on lower (i.e. routine) modes of activity. At thermal limits,OCLTT also integrates protective mechanisms that extend time-limitedtolerance to temperature extremes – mechanisms such aschaperones, anaerobic metabolism and antioxidative defence. Here,we briefly summarise the OCLTT concept and update it by addressingthe role of routinemetabolism.We highlight potential pitfalls in applyingthe concept and discuss the variables measured that led to thedevelopment of OCLTT.We propose that OCLTTexplains why thermalvulnerability is highest at the whole-animal level and lowest at themolecular level. We also discuss how OCLTT captures the thermalconstraints on the evolution of aquatic animal life and supports anunderstanding of the benefits of transitioning from water to land.

KEY WORDS: Organisational complexity, Sublethal thermal limits,Aerobic power budget, Aerobic performance, Oxygen supply,Oxygen demand, Temperature adaptation, Water breather, Airbreather

IntroductionGiven the impacts of climate warming on ecosystems, it is criticalthat we increase our understanding of organismal thermal ranges,responses and tolerances. Our understanding has long beeninsufficient, as studies have often focused on estimates of criticalthermal maxima (CTmax) or lethal limits (LT50; see Glossary). Infishes, these upper and lower limits and the range between themcorrelate to varying degrees with latitude, and probably also withlatitude-associated temperature regimes (re-assessed by Pörtner andPeck, 2010). However, lethal limits are often more extreme thanthe temperatures that an animal will experience in its environment.

Thus, there is a variable ‘safety margin’ between ambienttemperature extremes and lethal temperatures (Sunday et al.,2012, 2014). Negative effects of changing temperature may occurwithin this margin, impacting ecology and therefore requiringidentification. The physiological mechanisms causing heat or chillcoma and death have been investigated for more than a century,thanks to the desire to identify the primary mechanism oftemperature-associated death, yet a comprehensive mechanism-based understanding of this process has not been established.

Because temperature has a pervasive influence on all levels ofbiological organisation (Hochachka and Somero, 2002), researchshould address how mechanisms across these levels combine toshape the thermal limitations of an organism in the context of theecosystem. The oxygen- and capacity-limited thermal tolerance(OCLTT) concept (Box 1), developed over the last two decades, hasbeen proposed to meet these challenges and to provide a frameworkexplaining how physiological mechanisms co-define an animal’sfundamental and realised thermal niches (see Glossary), with afocus on critical life stages (for early summaries of OCLTT, seePörtner, 2001, 2002; for thermal niches, see Pörtner et al., 2010;Deutsch et al., 2015; Payne et al., 2016). The basic idea underlyingthe OCLTT is that once temperatures approach limiting values,constraints on the capacity of an animal to supply oxygen to tissuesto meet demand cause a progressive decline in performance (e.g.Pörtner and Giomi, 2013; Giomi et al., 2014), with consequences atthe ecosystem level (e.g. Del Raye and Weng, 2015; Payne et al.,2016). OCLTT considers that most routine performances are fuelledsustainably by aerobic metabolism in excess of standard metabolicrate (SMR) and largely exclude anaerobic metabolism.

The aim of this Commentary is to summarise (and update) the keyelements of the OCLTT concept. We first discuss the use of theOCLTT to understand species’ responses to climate change. We thenhighlight pitfalls that can result from (over-)simplification, fromdifferent uses of terms and from combining OCLTT with traditionalconcepts (e.g. CTmax), especially when bypassing the transition fromsublethal to lethal thermal limits (see Glossary). We summarise thephysiological variables that were measured when developing theOCLTT concept and that should be tested further in order to assessOCLTT, and discuss different understandings of the term ‘capacity’.Finally, we consider the evolutionary modulation of the OCLTT.

Using OCLTT to understand species’ responses to climatechangeThe limits of a species’ realised niche are thought to determine itslarge-scale temperature-dependent biogeography, as well as theanimals’ responses to warming. Combined with consideration ofOCLTT, such principles may allow us to explain the currentlyobserved biogeographical shifts of marine animals (Poloczanskaet al., 2013). Individuals that undergo biogeographical shifts haveexperienced non-lethal thermal constraints; however, organisms thatstay behind may eventually be lethally affected. Both processes

Section of Integrative Ecophysiology, Biosciences, Alfred-Wegener-Institute,Bremerhaven D-27570, Germany.

*Author for correspondence ([email protected])

H.-O.P., 0000-0001-6535-6575

2685

© 2017. Published by The Company of Biologists Ltd | Journal of Experimental Biology (2017) 220, 2685-2696 doi:10.1242/jeb.134585

Journa

lofEx

perimen

talB

iology

contribute to local extinction (e.g. Jones et al., 2009). Whileindividuals survive non-lethal thermal constraints, the resultingreductions in available energy will jeopardise reproduction and thuspopulation survival. Furthermore, the borders of the realised nichealso depend on species interactions (e.g. changing availability of preyorganisms). Current knowledge suggests that the borders of thermalniches can shift as a result of acclimation of individuals betweenseasons or adaptation over generations; this might allow organisms totolerate rising temperatures without changes in distribution. However,data reveal ongoing distribution shifts for different species, indicatingthat the capacities for acclimation and adaptation are limited or thatthese processes are too slow to prevent biogeographical shifts(evidence reviewed and assessed by IPCC AR5: Poloczanska et al.,2014; Pörtner et al., 2014).OCLTT has been used to explain the physiology underpinning the

climate responses of individual species, as observed in the field, atbiogeographical borders of critical life stages or under extremeseasonal conditions. Information on species-specific thermal nichescan be derived from data on temperature-dependent steady-state animalperformance (see Glossary) under the routine conditions of a species –relevant performances can range from steady-state swimming ofsalmon during spawning migrations (Eliason et al., 2011), to growth(including feeding) rates of benthic and demersal fish (Pörtner andKnust, 2007), to ventilatory and motor activities as in amphipods(Jakob et al., 2016) – and, more generally, they relate to the scope ofroutine oxygen demand and associated shifts in species abundance(see Pörtner and Knust, 2007) or biogeographical boundaries (Deutschet al., 2015). With the possible exception of salmon migratingupstream (Fig. 1), routine performance in the centre of the thermalrange usually does not fully exploit the available aerobic power budget(sensu Guderley and Pörtner, 2010; see Glossary).As a result of acclimation or local adaptation and the associated

trade-offs, the resulting total thermal performance curves (relatingto aerobic power budget and encompassing various performances)differ between species and even between populations of thesame species (Box 2, Fig. 1A), leading to different metabolic andperformance characteristics across latitudes (Pörtner, 2006; Pörtneret al., 2008; Schröer et al., 2009). As a result of phenotypic diversityand plasticity, a species’ biogeographical range is made up ofoverlapping niches of populations or individuals (Fig. 1A). Inaddition, differences in thermal curves and optima for individualperformances may result from individuals being in differentphysiological modes when performing (e.g. due to differentfeeding or hormonal status), such that trade-offs in energy

List of symbols and abbreviationsCaO2 arterial oxygen contentCvO2 venous oxygen contentCO cardiac outputCTmax critical thermal maximumLT50 lethal temperature causing 50% mortalityMO2 oxygen consumption rateOCLTT oxygen- and capacity-limited thermal tolerancePO2 partial pressure of O2

PaO2 arterial partial pressure of O2

PvO2 venous partial pressure of O2

SMR standard metabolic rateTc critical temperatureTd denaturation temperatureTopt optimum temperatureTp pejus temperature

GlossaryActive thermal toleranceThis occurs in the range of temperatures permanently tolerated. It involvesaerobic performance and associated aerobic metabolism fuelling theenergy demands of maintenance and additional functions (e.g. growth,reproduction) and behaviours (e.g. roaming, foraging, mating).Aerobic power budgetThe full amount of excess aerobic energy available above maintenancethat is recruited from mitochondrial metabolism. It encompasses and istraditionally estimated from aerobic scope for exercise (see below).However, muscles may not be able to fully exploit that aerobic powerbudget, or may push energy demand beyond routine power budget,through anaerobic contributions and transient mobilisation of functionalreserves. Furthermore, trade-offs in energy allocation may occur,affecting the balance between behaviours and exercise, growth ofreproductive and somatic tissue and repair processes.Aerobic scopeThe difference between resting and the highest exercise-induced rate ofoxygen consumption. In brief, aerobic scope for exercise is a measure ofaerobic power budget with the need to consider the complexities andfunctional constraints discussed under ‘aerobic power budget’.Critical thermal maximum (CTmax)The high temperature extreme leading to the onset of spasms(unorganised locomotion), close to the lethal temperature.Functional capacityThe ability to routinely and permanently maintain a certain rate offunctioning, supporting a specific level and kind of performance asneeded under routine conditions at the ecosystem level.Functional reserveAdditional performance capacity activated by hormonal action, e.g.catecholamines (fight and flight response).Functional scopeThe ability to increase the rate of a specific function or set of functionsabove those at rest, supporting a specific level and kind of performanceas needed under routine conditions at the ecosystem level.Fundamental thermal niche (sensu Hutchinson)The temperature range within which physiological functioning of aspecies allows tolerance under resting conditions (covering active andpassive ranges). Temperature effects may be influenced by specificeffects of other environmental factors.Lethal temperature limit (LT50)The temperature extreme (cold or warm) causing 50%mortality. It shouldbe noted that the value of LT50 found is influenced by the experimentalprotocol, especially the duration of exposure to step-wise increased ordecreased temperature.Passive thermal toleranceThe range of temperatures sustained passively by an organism throughexploitation of residual aerobic and anaerobic metabolism, antioxidativedefence, metabolic depression and the heat shock response. As theseresources are depleted over time and feeding is constrained, passivetolerance is time limited.Realised thermal niche (sensu Hutchinson)The range of temperatures within which physiological functioningsustains Darwinian fitness and persistence of a species under routineconditions, including species interactions. Temperature effects may beinfluenced by specific effects of other environmental factors.Steady-state routine performanceThe rate of performance (feeding, behaviours, reproduction) that anorganism displays routinely and permanently to maintain fitness in itsnatural environment.Sublethal limitsConstraints in maintaining a functional rate under changing environmentalconditions, e.g. warming, with negative implications at the ecosystem level.SymmorphosisAccording to this theory, the components of an organism match itsoverall functional scope, building on a quantitative match of design andfunctional parameters. The functional capacity of a complex system suchas an animal’s body must cope with the highest functional demand (afterWeibel et al., 1991).

2686

COMMENTARY Journal of Experimental Biology (2017) 220, 2685-2696 doi:10.1242/jeb.134585

Journa

lofEx

perimen

talB

iology

budgets may occur (see Holt and Jorgensen, 2015; for an earlyexample of how starvation modifies thermal performance andoptima in salmonids, see Brett, 1971).OCLTT principles are presently integrated into models

predicting the effects of climate on species’ distributions and theconsequences for ecosystems (e.g. Jones and Cheung, 2014;Deutsch et al., 2015). The concept has recently been extended toincorporate the combined effects of various climate change-associated drivers, such as ocean warming combined withacidification and hypoxia (Pörtner, 2010, 2012). It should alsobe noted that the pattern of acclimatisation to one climate-relateddriver such as temperature can be modified by the combinedeffects of multiple drivers (Anttila et al., 2015), indicating that

highly complex mechanisms shape sublethal and lethal thermalconstraints.

Complexity shaping thermal limits: pitfalls when addressingOCLTTEarly in the development of the OCLTT concept it was suggestedthat thermal constraints are first noticeable at the highest level oforganisational complexity (i.e. the whole animal), before affectinglower hierarchical levels (e.g. cellular and molecular levels; Pörtner,2002). Recently, Storch et al. (2014) developed a ‘complexityindex’ to compare the largely different thermal limits found acrossmarine organism domains. They found a relationship betweensublethal (and lethal) thermal limits and the number of body and cell

Box 1. The OCLTT conceptSelected indicators of oxygen- and capacity-limited thermal tolerance (OCLTT) provide a systems view of multiple interlinked parameters characterising thethermal range of an aquatic animal species and its aerobic window (at steady state) (e.g. Pörtner, 2002, 2012; Pörtner et al., 2010, updated according tofindings by Deutsch et al., 2015). The example given in the figure is for awarm temperate aquatic animal. The solid line shows the principal pattern of (mixed)body fluid PO2 against temperature in Maja squinado (Frederich and Pörtner, 2000). Note that temperature-dependent patterns of body fluid PO2 are notuniform between species and do not closely follow temperature-dependent changes in metabolic rate or performance. The range of active thermal tolerance(seeGlossary) is limited on both sides by pejus temperatures (Tp; the box lists processes supporting active tolerance, which become constrained beyond Tp).Towards warm and cold extremes, the transition to passive thermal tolerance (see Glossary) is indicated by a decline in (venous) PO2 (solid black arrows),causing oxidative stress (Heise et al., 2006), heat shock response and, finally, transition to anaerobic metabolism (e.g. Kyprianou et al., 2010; Pörtner andKnust, 2007; dashed black arrows). The model proposes that, in a systemic to molecular hierarchy of thermal tolerance thresholds, these progressivetransitions from sublethal to acutely lethal conditions [characterised by critical thermal maximum (CTmax), cold shock, denaturation (at denaturationtemperature, Td)] involve feedback between whole-organism and molecular levels. Blue arrows indicate the link between oxidative stress, heat-inducedmolecular damage and heat shock protein expression (for further details on these interactions, see Kassahn et al., 2009). This whole-organism feedbackmay narrow molecular thermal windows, such that Td is reached at lesser extremes of temperature (red arrows shifting upper and lower Td). The passivetolerance range is a component of the niche used routinely by organisms experiencing extreme temperatures (e.g. in the intertidal zone). Rather thanwidening the active thermal range at a cost, theyminimise metabolic costs and tolerate extremes anaerobically. Extended thermal tolerance is then achievedby protective mechanisms such as metabolic depression, anaerobic metabolism beyond the critical temperature (Tc), antioxidative defence and the use ofchaperones such as heat shock proteins (e.g. Tomanek and Somero, 2002).

Cardiovascular scope and costScope for growth, metabolism

Mitochondria: capacity and densityBody condition

Scope for behaviour, exercise

Aerobicwindow(steadystate)

DenaturationCold shockFreezing0

Anaerobicmetabolism

Body fluid PO

2

(as in Maja squinado)

Tc Tc

Td Td

Tp Tp

Anaerobicmetabolism

DenaturationCTmax?

Heat shockproteins ∆PO2

∆PO2

Oxidativedamage

Active

PassiveOxidativedamage

Heat shockproteins

Passive – Active – Passive rangeShort – Long – Short-term tolerance

2687


Journa

lofEx

perimen

talB

iology

compartments (separated by structure or function) as a proxy ofcomplexity. The comparatively low thermal limit to animal life,i.e. above about 45°C (or somewhat less in water), would result fromfunctional integration of a large number of compartments at thewhole-organism level (Storch et al., 2014). This immediately rulesout the possibility that one individual key protein or mechanism hasan exclusive role in whole-organism limitation. Instead, constraintsfelt at lower levels of biological organisation should be embeddedinto the whole-organism context (see Pörtner, 2012). In heat-tolerant microbes, proteins do function up to 120°C, and there is noreason to assume that animal proteins could not evolve heat limitsabove the thermal limit of animal life.For the benefit of optimum functioning, whole-organism and

molecular thermal ranges would be interdependent on evolutionarytime scales, with molecular thermal ranges being somewhat widerthanwhole-animal thermal ranges (Pörtner et al., 2012). Lower whole-animal thermal limits would promote the functional optimisation ofproteinswithin the low range of animal body temperatures, resulting inmolecular limits beyond but close to whole-organism thermal limits(e.g. Somero, 2010). At the same time, whole-animal constraints canfeed back to the protein level, e.g. through oxidative stress (seeKassahn et al., 2009). Consequently, whole-organism limits ‘trickledown’ to limits at lower organisational levels, such that individualmolecular or organellar functions may become limited at less extremetemperatures in situ than when extracted from the whole-organism ortissue context (Pörtner et al., 2012; cf. Iftikar and Hickey, 2013; Leoet al., 2017). Characterising the role of aerobic metabolism andunderlying mechanisms in thermal limitation (Schulte, 2015) thusrequires considering how these mechanisms interact with others

(e.g. antioxidative defence) and whether whole-organism phenomenafeed back to these mechanisms (Box 1). This level of complexity mayexplain why thermal biology has not had a coherent framework andalso why experimental work building on reductionist hypotheses (e.g.the idea that thermal damage to one kind of protein causes whole-organism heat death) comes with potential pitfalls. Althoughexperiments must necessarily be reductionist, researchers shouldstrive to embed experimental findings into concepts that capture thefull complexity of the mechanisms involved, in an ecological context.

Matches or mismatches in oxygen supply and demand affect alltissues and cells, and thus the largest conceivable number of bodycompartments in an animal (Storch et al., 2014). Despite theunderlying role of functional complexity in OCLTT-inducedsublethal thermal constraints, however, some recent studies haveexclusively focused on LT50 in an attempt to investigate the OCLTTconcept; for example, by asking whether oxygen availability canshift LT50 or whether a maximally stimulated and exploitedcardiovascular system has the capacity to supply oxygen until thispoint. LT50 and CTmax are conventional measures of ultimatetolerance limits (e.g. Lutterschmidt and Hutchison, 1997). Theselimits lie at the edge of or outside the range of aerobic power budget(Zakhartsev et al., 2003; Pörtner and Knust, 2007; Chen et al.,2015), beyond critical limits (Tc, where there is a transition toanaerobic metabolism; Box 1) and close to the co-evolvingdenaturation temperature (Td) (see Farrell, 2009, and below). Suchtesting raises concerns, as it bypasses the sublethal thresholds at thecore of OCLTT, such as pejus temperature (Tp, the onset of capacitylimitation and hypoxaemia) and Tc, and their ecological relevance(primarily of Tp), which has been demonstrated in field studies

40

A

B

35

30

25

2013 14 15 16

a

b c

d

Average temperature (°C)

Spawners:

Large adults:

Juveniles:

Eggs, early larvae:

Aerobic thermal window

Sequ

ence

of l

ife s

tage

sG

row

th ra

te (µ

m y

ear–

1 )

17 18 19

Thermal constraints on salmon spawning migrations, Fraser RiverSensitivity of spawning North Sea cod to warming winter

Summer losses of large eelpout in the Wadden SeaSouthern distribution boundaries of Atlantic codHeat-induced migration of Baikal amphipods to deeper watersAbundance of water fleas in relation to seasonal temperature

Different thermal ranges of competing sardines and anchoviesMovement of juvenile Atlantic salmon to cooler water

High temperature affecting crustacean larvaeHigh temperature constraining cod embryo development

Fig. 1. Dynamics of thermal windows betweenpopulations and during ontogeny. (A) Temperature-dependent growth rates (the increment per year) readfrom otoliths in field populations of banded morwong,Cheilodactylus spectabilis, around Australia and NewZealand (after Neuheimer et al., 2011). The overallgrowth curvemay well reflect the combination of differentgrowth curves of individual (hypothetical) populations(a–d), indicating phenotypic diversity of thermal optimaand ranges resulting from local acclimatisation oradaptation, as seen in cod,Gadus morhua. Trade-offs inenergy budget result, and these lead to shifts in growthrate and performance capacities between populations(e.g. Portner et al., 2008; see Box 2). (B) Ontogeneticdynamics of thermal windows [curves indicate changesin upper and lower oxygen- and capacity-limited thermaltolerance (OCLTT) limits over time and between lifestages; thermal window widths are indicated by arrows;Portner and Farrell, 2008] in relation to observedecosystem/life-history constraints or to field phenomenaexplained by laboratory evidence for the respective lifestage (after Perry et al., 2005; Breau et al., 2011; Eliasonet al., 2011, 2013; Deutsch et al., 2015; Dahlke et al.,2016; Jakob et al., 2016; Portner and Knust, 2007;Portner et al., 2008; Storch et al., 2011; Schwerin et al.,2010; Takasuka et al., 2007).

2688


Journa

lofEx

perimen

talB

iology

(e.g. Pörtner andKnust, 2007; Eliason et al., 2011; Jakob et al., 2016).By addressing the effect of oxygen on lethality, such studies alsooverlook the contribution and capacity of mechanisms contributing topassive thermal tolerance (see Glossary; e.g. heat shock response,anaerobic metabolism, metabolic depression and antioxidativedefence) as well as its time limitation beyond Tp and Tc, beforelethal effects set in at around Td (Box 1; Pörtner, 2010, 2012; cf. Pecket al., 2009). More generally, passive thermal tolerance starts oncehypoxaemia and/or Q10-dependent rate limitations (in the cold)constrain routine performance beyond Tp (Box 1).Considering the step-by-step development of OCLTT from upper

Tp to Tc – and the increasing involvement of further mechanismslimiting thermal tolerance – leads to the prediction that any effectsof experimentally increasing oxygen availability are progressivelyreduced and may become small at CTmax or LT50; such systematicinvestigation of oxygen effects across thermal thresholds would be

useful. The existence of a whole-organism to molecular hierarchy ofthermal sensitivity also suggests that increasing oxygen availabilitymay not cause large shifts in CTmax or LT50, as oxygen may not(fully) alleviate the extreme (e.g. denaturation) limits that co-evolved at lower levels of organisation. For example, under ambienthyperoxia, excess oxygen may alleviate thermal stress by reducingthe costs of oxygen supply, thereby causing shifts in sublethal butnot necessarily in lethal limits (e.g. Mark et al., 2002; Pörtner et al.,2006; Ekström et al., 2016). By contrast, ambient hypoxia orreduced oxygen supply capacity would exacerbate thermal stressand, because of stronger feedback from whole-organism tomolecular levels, e.g. through enhanced oxidative stress, mayreduce upper CTmax or LT50 to some extent.

In line with the prediction that reduced oxygen supply capacitymay reduce CTmax, recent studies in fish have identified a role forhaematocrit in thermal tolerance. In support of OCLTT, Beers and

Box 2. Performance and OCLTTWithin the thermal range constrained by the PO2 profile (Box 1), the kinetic stimulation of metabolic processes and energy-dependent functions by warming(and their inhibition upon cooling) leads to an asymmetric whole-organism total performance curve supported by aerobic power budget. The curve has afunctional optimum close to the upper Tp. The power budget is shared between growth, immune response, reproduction and exercise during differentbehaviours (e.g. migration, competition, foraging) at different routine steady-state levels (e.g. sit and wait, roaming). Thermal limitation begins with reducedperformance, and tolerance becomes progressively more time limited once functional scope falls below a limiting threshold beyond Tp. At optimumtemperature (Topt), maximised steady-state functional scope results from optimum oxygen supply at baseline oxygen demand, which rises exponentiallyabove Topt (dashed exponential curves, see Fig. 2). Capacity constraints lead to a mismatch in oxygen supply and demand beyond Tp. Depending on thelevel of routine performance and associated energy demand, Tp and Tc are dynamic within species. They are influenced by oxygen demand versus supplyand by the temperature dependence of underlying metabolic costs (dashed exponential lines linking upper and lower Tp). The figure illustrates shifts in upperand lower Tp depending on functional rates and associated metabolic demands. Different species use different routine functional rates, and their Tp valuesare adjusted during their evolution to match ambient conditions. Both ambient hypoxia and elevated carbon dioxide levels may modify thermal windows(Walther et al., 2009; Zittier et al., 2012) and performance optima (green dashed lines). The graph depicts acute performance levels and limitations inresponse to short-term temperature fluctuations; during acclimatisation or adaptation (double-headed arrows), these may lead to shifts and changing widthsof thermal windows (Fig. 1). The exact position of CTmax in relation to Tc and Td is unknown.

Thermal acclimatisation and adaptation shift thermal limits through changing membrane composition or capacities of enzymes and mitochondria, orthrough mechanisms protecting molecular integrity (influencing Td; see Pörtner, 2012). For example, in the warmth, reduced oxygen demand and increasedTp and Tc are expected following a decrease in tissue mitochondrial density, capacity and proton leakage costs. The resulting unidirectional shifts of bothupper and lower tolerance thresholds, and the changing width of the thermal window involve molecular and cellular adjustments shaping the metaboliccapacity of tissues and the functional capacity of relevant energy-consuming processes and their maintenance costs.

Stea

dy-s

tate

rout

ine

perfo

rman

ce le

vels

(—)

at d

iffer

ent m

etab

olic

rate

s ( e

.g.

Onset of denaturationCTmax?

T (°C)

Protection

Td

Tc

Tc CO2hypoxia

TpEnduranceOnset of anaerobiosis

Loss of performance, abundanceCapacity

Ecosystem examples

Routines in different species:Evolution of metabolic capacities shifts TpWithin-species

functional states:ExerciseRoamingSit & wait

Pacific salmon (spawning migration)Atlantic cod (demersal)North Sea eelpout (benthic)

Tp

Topt

)

Passive – Active – Passive rangeShort – Long – Short-term tolerance

2689


Journa

lofEx

perimen

talB

iology

Sidell (2011) found a positive correlation between haematocrit andCTmax across Antarctic fish species. A recent study in sea bassmanipulated the haematocrit and found a small but significantdecline in CTmax at low haematocrit (Wang et al., 2014). However,CTmax and haematocrit were not correlated, possibly as a result ofthe effects of reduced sensitivity to oxygen close to CTmax, asdiscussed above, combined with high data variability. The specificmechanisms causing the shift in CTmax or the reasons for the highvariability in the relationship between CTmax and haematocritremain insufficiently explored (see below). Overall, sublethalthermal constraints are more likely than lethal limits to beresponsive to changing oxygen availability, and are likely to bemore closely related to tissue functional capacity (see Glossary) andenergy budget.

Variables indicative of OCLTTIn light of the above discussion, it seems prudent to avoid focusingexclusively on CTmax or LT50 in tests of the OCLTT concept. Instead,we should specifically identify sublethal thermal constraints fromrespiratory and metabolic variables under resting or routineconditions (Table 1). Breakpoints in the temperature dependence ofthese variables by their nature are ‘softer’ indicators of thermallimitation than a ‘hard’ endpoint such as lethal collapse. For example,warming causes SMR to rise exponentially until a breakpointtemperature beyond which SMR no longer increases (indicating theTc to be surpassed) (e.g. Melzner et al., 2006; Giomi and Pörtner,2013; see Fig. 2). Increasing ambient oxygen levels can lower theslope of the exponential rise in SMR and cause significantly loweredoxygen consumption rates at high temperatures – as seen in restingfish (Mark et al., 2002) or in amphibious crabs exposed to air (Giomiet al., 2014) – thereby increasing Tc.Excess oxygen leads to reduced blood flow and thus lowers the

cost of cardiovascular activity. This implies that, conversely, lowerambient oxygen levels cause metabolism to rise more strongly withincreasing temperature, as a result of increased cardiovascularcirculation. Similarly, anaemia can cause an increase in cardiacoutput (as in anaemic sea bass; see Wang et al., 2014); however, thecost increment in cardiovascular activity may remain small. Byreducing viscosity (Farrell, 1991), a lower haematocrit maycompensate for the cost increment, balancing the oxygen shortagecaused by the reduced haematocrit. Generally, the patterns of heartrate and cardiac output indicate sublethal thermal limitation as theydo not increase sufficiently to match thewarming-induced rise in O2

demand and to keep the aerobic power budget large, a lag setting inwell below CTmax (e.g. Wang et al., 2014). Haematocrit may thus bebetter correlated with sublethal constraints than with CTmax (e.g.Buckley et al., 2014).So how might variables underlying the OCLTT best be

investigated? Analyses of the OCLTT should mimic naturalconditions and consider routine activities displayed by the animalin the field, as well as minimising stress phenomena that wouldtransiently mobilise functional reserves (see Glossary), e.g. throughrelease of catecholamines, which stimulate cardiovascular circulation,glycogenolysis or anaerobic metabolism. Such stimulation supportstime-limited thermal tolerance but has negative consequences forother components of the energy budget, e.g. growth. It should benoted that not all species display continuous motor activity; thus,measurements of steady-state aerobic scope (see Glossary; Farrell,2013) for exercise may not always be possible when testing OCLTT.Tissue oxygenation and oxygen supply to sensitive aerobic organssuch as the heart (Ekström et al., 2017) or liver may closely traceOCLTT under routine conditions, but such estimates are usually not

available. In fish, measurements of venous rather than arterialPO2 appear appropriate to indicate thermal constraints, because ofvenous perfusion of the heart in most species (Farrell and Clutterham,2003; Lannig et al., 2004; Ekström et al., 2016; Farrell, 2009) and theproximity of venous blood to tissues; these analyses would ideallybe complemented by those of venous oxygen content (CvO2). Incrustaceans, measurements of oxygen partial pressure in (mixed

Table 1. Variables analysed and interpreted as indicators of OCLTT inanimals

Parameter Topt Tp Tc Td

Maximum aerobic scope: maximum growth,maximum exercise

✓

COmax (exercise) ✓Reduced performance* (exercise, growth, CO) ✓PvO2 /CvO2 ✓ ✓BP PvO2 /CvO2 ✓ ✓BP ventilation ✓ ✓BP heart rate ✓ ✓BP stroke volume ✓ ✓BP MO2 ✓ ✓COmax (rest, routine) ✓Mitochondrial functioning (permeabilised fibres) ✓? ✓?Anaerobic end products (especially succinate) ✓Cardiac arrhythmia/bradycardia ✓CTmax ✓?

See Box 1 and 2, and Storch et al. (2014). Indicators are mostly respiratoryparameters that have been used to assess different oxygen- and capacity-limited thermal tolerance (OCLTT)-related terms and thresholds (Box 1, Box 2;Topt, Tp, Tc, Td), as indicated with a tick. A general conclusion from availablestudies is that the physiological condition of the experimental animals needs tobe well defined and to match (long- and short-term) ecosystem conditions.Otherwise, trade-offs in aerobic energy (power) budget may haveconsequences for individual performances and their thermal constraints. Notethat the assessment benefits from an integrative analysis of various processesand taxon- or even species-specific patterns, as not all indicators may displayobvious thresholds in all taxa or species. MO2, oxygen consumption rate; CO,cardiac output; BP, breakpoint; PvO2, venous partial pressure of O2; CvO2,venous oxygen content; ?, proof of concept needed. *Ideally, total aerobicperformance (not directly measurable because of trade-offs in energy budget),but often referred to as growth (Fig. 1) or exercise.

SMR

Temperature

High capacity for oxygen supply,low Q10, larger scope:

eurythermy Thermal window

of functional scope

BP

BP

Low capacity for oxygensupply, high Q10, lower

scope: stenothermy

Fig. 2. Conceptual graph illustrating the thermal operating ranges of twohypothetical systems with different temperature-dependent capacities.The schematic diagram shows how low- and high-capacity systems (green andblue curves, respectively) have different baseline costs (standard metabolicrate, SMR) and ranges of operation if exposed to workload (e.g. due towarming) until reaching different capacity limits (under routine conditions,indicated by breakpoints, BP). Through similar principles and differences inthermal responses (Q10), OCLTT distinguishes stenotherms from eurytherms(here exemplified through narrow versus wide temperature ranges anddifferences in the temperature dependence of the workloads; see Portner,2006). In each case, the thermal window width of functional scope isrepresented by the length of the dashed arrows.

2690


Journa

lofEx

perimen

talB

iology

arterial and venous) haemolymph and of succinate concentrations intissues can reveal the development of extreme hypoxaemia atthermal extremes (Frederich and Pörtner, 2000). The accumulation ofanaerobic metabolites (such as succinate) beyond Tc indicatesoxygen-deficient mitochondria. The ecologically importanttransition phases from earliest sublethal limitation (at Tp) to Tcprecede CTmax (see Zakhartsev et al., 2003) – but their mechanisticlink to CTmax clearly deserves further study. In the future, continuousrecordings of tissue oxygenation may also support such analyses.Regulatory responses are likely to involve various molecular

factors (Kassahn et al., 2009). For example, hypoxia-induciblefactor 1 (HIF-1) contributes to improving anaerobic capacity as wellas oxygen supply through erythropoesis and may, thereby, enhancethe capacity for heat and cold endurance. Cold adaptation may besupported by thyroid hormones (Little et al., 2013). Depending onthe animal phylum, these changes contribute to adjustments ofoxygen-transport capacity by both ventilation and circulation(invertebrates) or mainly circulation (fishes), as well as toadjustments in mitochondrial density, capacity, oxygen demandand energy budget (all tissues).

Addressing capacity, performance and energy budgetIt should be noted that the term ‘capacity’ is used in different ways.The original OCLTT literature emphasises the interdependence andtrade-offs between the rates and capacities of functions supplyingand consuming oxygen and associated energy (e.g. cardiovascularsystems, as reflected in cardiac output; or mitochondrial ATPsynthesis or transmembrane ion transport in all tissues), their baselinecosts (e.g. resting contractile cardiac activity, mitochondrial protonleakage or transmembrane ion leakage) and the resulting performanceat various levels of biological organisation, in unstressed animals atrest or during routine activities in variable or stable climates.However, recent studies intending to test OCLTT have focused on thecardiovascular system and pushed it to its limits to determine whetherit has the maximum capacity to provide sufficient oxygen to theorganism until LT50 (Gräns et al., 2014; Wang et al., 2014). Thisapproach probably activates functional reserves (see above), which ispossible only transiently. At first sight, both views may appear valid,yet OCLTT emphasises sublethal thermal limitation and whole-organism consequences and trade-offs under resting and routineconditions. The concept addresses the subtle links and constraints inbaseline costs, net functional scope (see Glossary) and resultingwhole-organism performance capacity (see Fig. 2), as well as anyshifts in net energy allocation to routine activities, without transientactivation of functional reserves.Conceptually, functional properties of the fish heart illustrate core

aspects of the term ‘capacity’ and the associated functional andthermal limits, as considered in the OCLTT concept for species withdifferent modes of life and under different temperature regimes.Oxygen consumption of the body (MO2, excluding gas exchange viathe skin) equals cardiac output (CO) multiplied by the difference inarterial and venous oxygen concentration (Fick’s principle):

_MO2¼ CO " ðCaO2

$ CvO2Þ: ð1Þ

Cardiac output is the product of stroke volume and heart rate, andreflects the ‘functional capacity’ of the heart. Stroke volume is theblood volume pumped by the fish ventricle during one contraction.Any increase in oxygen demand (e.g. during routine exercise) isreflected in an increase in cardiac output. For most fishes, thecontribution of stroke volume to oxygen supply is usually higher

than that of heart rate, e.g. during sustained (routine) swimming(Farrell, 1991).

Functional capacity has three interdependent aspects: (1) howmuch performance can be achieved, depending on (2) how therespective system is set up and (3) at what cost. A high-capacitycardiovascular system comprises a larger heart with higher restingcardiac output than a low-capacity system (Farrell, 1991). Activefishes usually have a larger relative heart mass (Santer, 1985;Farrell, 1991). Accordingly, exercise training can produce isometriccardiac growth as seen in rainbow trout (Farrell et al., 1990). Cardiacoutput depends on the volume of the heart and the pressuregenerated by wall tension. When higher cardiac output is needed(e.g. during warming), heart rate and, possibly, stroke volumeincrease, at an energetic cost (i.e. increased oxygen consumption).A larger stroke volume entails an increase in ventricle radius andgreater filling during diastole.

Morphologically, the fish heart can be simplified as a spherewith defined volume (V ) and diameter (D, wall thickness) of themyocardial muscle. Theworkload (W ) of a contracting sphere equals:

W ¼ p " V ; ð2Þ

where p is pressure.Energy requirements (oxygen demand) of the heart are mainlydetermined by wall tension (T ) as defined by Laplace’s law:

T ¼ p " r=2D; ð3Þ

where r is the ventricle radius. The equation illustrates that a thickermyocardial muscle produces higher pressure and then output, at theexpense of higher baseline costs due to larger, mitochondria-richtissue mass. The energy turnover during contraction is related to thecontractile shortening of heart muscle fibres. In the assumedspherical heart, the length of a circular heart fibre may equal theheart circumference, L=2πr, with a shrinking radius representingthe contractile shortening of the muscle fibre. Because V=4/3πr3,the same absolute value of contractile shortening in a large, high-capacity heart (with a large radius) will produce a stroke volumelarger than that of a small, low-capacity heart, at rest and duringexercise. Similarly, for the same stroke volume, larger hearts need asmaller contraction of the muscle fibre than smaller hearts, resultingin a lower increase in relative cellular oxygen/energy demand for thesame increase in performance, and a larger functional and energyreserve in larger hearts to maximise performance. For the sameincrease in workload, a small-volume heart will thus be stimulatedto a greater extent and limited sooner than a large-volume heart(Fig. 2), a conclusion supported by the larger maximum strokevolumes of isolated perfused hearts of active fish (larger hearts)versus sluggish fish (smaller hearts; Farrell, 1991).

According to OCLTT, the capacity of the heart plays a role indetermining thermal tolerance and associated energy turnover. Asoutlined above, excess oxygen can cause reduced blood flow andvisibly lower whole-organism oxygen demand in the warmth,possibly by reducing the rising cost of cardiovascular activity.Increasing temperature induces increased cardiac output (e.g.Lannig et al., 2004; Farrell, 2009; Franklin et al., 2013), buildingon different baseline costs of maintenance and with differentexponential slopes in hearts of different sizes according to a species’active or passive lifestyle (Fig. 2).

Even within-species variability as seen in European sea bassindicates that larger hearts in fish with lower SMR support highertemperature tolerance and faster recovery from exercise than smallerhearts in fish with higher SMR (Ozolina et al., 2016). This

2691


Journa

lofEx

perimen

talB

iology

emphasises that there is poorly understood variability in the patternsdepicted in Fig. 2. The following hypothetical picture emerges: thecomparison of sluggish versus active fish appears analogous to thatof cold stenothermal versus cold eurythermal fish (or ectotherms ingeneral). To meet the same absolute or relative increase in oxygendemand, a low-capacity, low-cost system as in a sluggish fish (or acold-adapted stenotherm) would experience a stronger stimulus thana higher-capacity system as in an active fish (or in a temperateeurytherm), causing a greater percentage cost increment in thesluggish fish/stenotherm and thus contributing to a higher Q10 andearlier thermal limitation under routine conditions. This pattern isalso mirrored in the low-capacity, low-cost mitochondria seenin cold-adapted stenotherms versus high-capacity, high-costmitochondria as in cold-adapted eurytherms (Pörtner, 2006;Fig. 2). In line with these findings, more active, mobile Antarcticstenotherms are indeed more heat tolerant than sessile sluggish ones(Peck et al., 2009; for the role of cold adaptation and eurythermy inthe evolution of high-energy turnover endotherms, see Pörtner,2004; Clarke and Pörtner, 2010). Further observations are also inline with these emerging principles. Population-specific adaptationto various temperature regimes in salmon involves different heartsizes and adrenoceptor densities (Eliason et al., 2011, 2013).Acclimation of individual fish to temperature also involves changesin cardiac performance of fish. Acclimation to warming in salmoncauses an increase in maximum heart rate, meeting the risingbaseline cost (Anttila et al., 2014).The mechanical picture drawn from this simplified approach will

thus be modified by potential cellular or morphological differences,such as in the oxidative capacity of mitochondria (see Pörtner,2006), pacemaker activity, and size and capillarisation of the heart,or blood viscosity. For example, a low contribution of blood oxygentransport to aerobic scope is compensated for to some extent by theevolution of relatively large hearts, as in Antarctic icefishes (Farrell,1991). The interplay of all of these factors will shape thecontribution of the cardiovascular system to the species-specificoxygen and capacity limitation of the whole organism.In general, maintenance costs (measured as SMR) are relatively

low within the optimal thermal range but rise exponentially towardsthe upper limit of thermal tolerance, constraining functional(aerobic) scope. Although for some performances (like growth orreproduction) or routine activities (roaming and feeding) aerobicscope is not fully exploited, rising maintenance costs will stillintroduce constraints on aerobic power budget. Thus, functionalscope, e.g. of the heart, is highest at the thermal optimum (Topt)when maintenance costs are still relatively low. At temperaturesbelow Topt, functional scope is depressed by cooling more thanmaintenance costs are, finally leading to the failure of oxygensupply to meet demand as seen in warm temperate animals atcritically low temperatures (Frederich and Pörtner, 2000). High-capacity systems (e.g. tuna), while having a higher baseline cost(Fig. 2), come with the benefit of easily buffering demand underroutine conditions, e.g. during warming or exercise or both, with asmaller percentage increment in cost and limitation setting in athigher temperatures than for the same condition in a low-capacitysystem (as in hagfish or in Antarctic icefish, considering the loss ofhaemoglobin in the latter). Here, baseline costs are lower butpercentage increments are higher, and the system runs into capacitylimitations at lesser extremes. Because of the interdependence ofcapacity and cost, the percentage increment of cost per degree ofwarming is thus highest in energy-saving, low-capacity systems,such as polar or winter stenotherms (e.g. Pörtner, 2006; Wittmannet al., 2008; Pörtner et al., 2013), emphasising a link between energy

turnover, mode of life and the level of eurythermy (see Pörtner,2004; Peck et al., 2009; Clarke and Pörtner, 2010). It should benoted that temperate-zone animals may be able to exploit theenergetic benefits of being either winter stenotherms or springand summer eurytherms through seasonal acclimatisation (e.g.Wittmann et al., 2008).

Looking at capacity just in terms of its maximum exploitabilitythus misses the role of underlying design and its plasticity underroutine conditions, as well as the subtleties in the functionaltransitions and limitations. In a living animal, an early subtleindication of capacity limitation can be the presence of a breakpointtemperature (Fig. 2). This more complex approach to capacitycaptures the progressive development of thermal limitation from theearliest constraints to lethal temperatures (Box 1), as well as thedifference between stenotherms and eurytherms (Pörtner, 2006).

In this context, measurements of aerobic scope for exercise as anestimate of aerobic power budget have to be interpreted verycarefully, as analyses of aerobic scope using critical swimmingspeed (Ucrit) protocols in fish can include exploitation of non-sustainable short-term functional reserves that rely on hormonal(adrenergic) stimulation or anaerobic processes, beyond the onset ofkick-and-glide swimming (Lurman et al., 2007). As the degree ofmobilisation of anaerobic reserves can have a strong behaviouralcomponent (Peake and Farrell, 2006), the use of fatigue-basedexercise protocols may overestimate aerobic capacity, thereby againmissing earliest functional constraints at pejus limits (Table 1).

It is also important to note that Tp and Tc may shift depending onthe routine performances used and their steady-state energy demandat the ecosystem level (Box 1). For some species in a specific lifephase (e.g. spawning migrations in salmon), Tp and Tc are bestdetermined during full exploitation of aerobic scope for exercise.For other species and life phases (and more widely), this should beduring their lower levels of routine activities (e.g. Atlantic cod;Deutsch et al., 2015). In species regularly experiencing temperatureextremes (such as at low tide in the intertidal zone), their capacity toexploit the passive tolerance range may become important inshaping fitness. Ideally, for assessing ecologically relevant Tp andTc values, mode of life and associated energy demand, life phaseand habitat challenges require consideration.

Evolutionary modulation of OCLTTThe OCLTT framework has identified phenomena of thermallimitation in various animal phyla against ecological andevolutionary backgrounds (e.g. Pörtner et al., 2005; Knoll et al.,2007). It has been suggested to be an early evolutionary principle inanimals that has been modified according to life stage (see Pörtnerand Farrell, 2008) or climate zone (Pörtner, 2006; Beers and Sidell,2011; Pörtner et al., 2013), or during the evolution of air breathing(Giomi et al., 2014).

OCLTT varies with ontogeny, the associated development oforgan functioning, metabolic plasticity and the organ’s resultingcapacity and body size. Thermal windows are typically narrowduring early life and adult spawning stages and wider duringjuvenile and young adult stages (Pörtner and Farrell, 2008; Polettoet al., 2017; Fig. 1B). Such bottlenecks constrain where earlyand spawning life stages can live, and expose them to strongevolutionary pressures, leading to adaptive changes with functionalconsequences for the next life stage. These are virtually unexplored.Knowing the life history of a species in the context of habitatfluctuations is thus relevant to fully identify evolutionarybottlenecks and their consequences for physiology andbiogeography.

2692


Journa

lofEx

perimen

talB

iology

In permanently oxygen-rich polar waters, adaptation may havealleviated thermal constraints on the cold side of the thermalwindow (Wittmann et al., 2012; Pörtner et al., 2013), leavingresidual cold limitation to kinetic constraints on functional capacity.Improved cold tolerance was facilitated by excess oxygen dissolvedin cold water, which supported low metabolic rates by allowing alowered oxygen supply capacity and cost. However, this wouldcome at the price of enhanced heat intolerance (Pörtner et al., 2013).In addition, excess oxygen supply at stable low temperatures mayhave enabled the loss of myoglobin and haemoglobin functions inAntarctic icefish, which lowered oxygen supply capacity andincreased heat intolerance even further (Beers and Sidell, 2011).The situation is less clear for the evolutionary adaptation to

breathing air, which has 30-fold higher oxygen levels than water. Thetransition to terrestrial life often required the evolution of completelynew gas-exchange systems (lungs, trachea), while convective‘blood’-bound oxygen supply to tissues persisted in most taxa(except for some adult insects). The symmorphosis principle (seeGlossary; Weibel et al., 1991) suggests that these new convectiveoxygen-supply systems also evolved with their capacity limits set tocope with temperature extremes. In lower vertebrates, some evidencein fact indicates a progressively limited scope for oxygen uptake bythe lungs towards higher temperatures (see fig. 6 in Pörtner, 2002;Jackson, 2007). In contrast, crustaceans still use their gills when in airand are therefore suitable models to investigate the potential benefitsof air breathing for thermal tolerance. In fact, findings in amphibiouscrabs corroborate that oxygen supply costs are reduced in air and thatthis causes enhanced heat tolerance (Giomi et al., 2014). Insects mayalso have exploited this route; while aquatic larvae are subject toOCLTT principles (Verberk and Calosi, 2012), the tracheal oxygensupply to tissues in terrestrial adults supports elevated heat toleranceat minimised oxygen-supply costs.Recent studies of thermal tolerance in air breathers may not have

considered this diverse evolutionary background, or the pitfallsmentioned above. A study of the tropical toad Rhinella marinareported evidence for thermal acclimation at two temperatures(Seebacher and Franklin, 2011) but did not explore the relatedchanges in steady-state cardiovascular functions or in tissueoxygenation (or venous oxygen tensions as determined by Pörtneret al., 1991). Such evidence is required when investigating a role foroxygen in thermal limitation. A later study of the same species byOvergaard et al. (2012) found a typical exponential increase inresting oxygen consumption, which entered the steep phase beyond30°C, in line with a transition to pejus range (see Fig. 2). However,activity was enforced and non-steady state; and data variability washigh and may not have provided sufficient resolution to determinelimiting thresholds. A later study on python (Fobian et al., 2014)drew conclusions from measurements of temperature-dependentarterial oxygen tensions, which remained high even at hightemperatures. It needs to be considered that arterial oxygen valuesare often not suitable to investigate OCLTT, as also seen in fish, forexample, where arterial PO2 (PaO2) remained high when venousPO2 (PvO2) fell with rising temperatures (see Sartoris et al., 2003). Inline with OCLTT, however, the python developed a warming-induced reduction in aerobic scope, which was determined as thedifference between metabolic rates of fasting and digesting snakes.The drop in scope was paralleled by a levelling off in temperature-dependent arterial PO2 and oxygen consumption in digesting snakes(see fig.1 of Fobian et al., 2014). As a corollary, future studiesaddressing sublethal constraints in oxygen supply towards extremetemperatures in air require careful consideration of best practices forstudying OCLTT (Pörtner, 2014).

Further study of the role of OCLTT in animal evolution andontogeny is needed. For example, the mechanisms of thermaladaptation and limitation have been poorly investigated across alllife stages of a species or in both aquatic and terrestrial animals fromthe sub-tropics and the tropics. These organisms are probablyadapted to ambient temperatures closer to metazoan heat limitsof about 45°C, suggesting a steep transition from optimumtemperatures via sublethal to lethal limits. This makes it moredifficult to accurately identify sublethal thresholds, analogous to thestudy of cold limitation in (sub-)polar organisms. Thermaladaptation to the warmth would cause a down-regulation ofmetabolic rate at high oxygen diffusivity, which might alleviateoxygen-dependent constraints. Nonetheless, some data in tropicalfish indicate constraints on aerobic scope for exercise at hightemperatures (Munday et al., 2009). That said, the available data arepresently too limited to clearly identify typical patterns ofcomplexity limits in both tropical and subtropical aquatic andterrestrial ectotherms.

ConclusionsThe ad hoc mixing of OCLTT with classical concepts in thermalbiology such as CTmax requires care, as such studies tend tooverlook the subtle limits to aerobic metabolism and performance,and the systemic to molecular hierarchy of thermal tolerance.

For understanding ecological patterns, the functional rates oforganisms need to be explained. Darwinian fitness does not dependon one performance trait only, but various traits come together intheir additive energy cost. Fitness is thus related to routinemetabolism, reflecting how aerobic power budget is used bydifferent performances simultaneously and at the required minimumlevel and above. Therefore, OCLTT should not be simplified tocomprise aerobic scope for exercise only. OCLTT is about a causeand effect understanding for different performances and their role infitness and share in energy budget of a specific life stage. Such causeand effect understanding is at the core of physiological studies in anecological context. This principal understanding is also an assetwhen making predictions on the fate of populations in a distantfuture.

The OCLTT concept explains the first line of thermal limitation atthe whole-organism level in animals, and may represent anevolutionary constraint that was modified depending on life stageand climate, and during transition to life in air. Neglecting to considerthe links between levels of biological organisation will lead toinsufficient explanations of thermal limits that fall behind what theOCLTT concept has already achieved. It is possible that thecombination of OCLTT and molecular limits shape lethal limitsand their response to oxygen availability. As experiments can rarelyresolve all facets of complex phenomena, addressing such complexityrequires the parallel development of theoretical background andexperimental approaches at multiple levels; current theories ofevolutionary biology illustrate these requirements (e.g. Angilletta,2009). Thermal physiology should strive to interpret experimentalresults in light of an organism’s ontogeny and ecology, usingcomprehensive, ecologically relevant concepts (Bozinovic andPörtner, 2015). Conversely, reductionist lines of interpretationshould remain coherent with the widest possible conceptualframework.

Sublethal thermal limitation according to OCLTT can also varydepending on activity level (e.g. resting, roaming or high rates ofsteady-state energy use); thus, it is important to consider whichsituation is typical for a species and its critical life stage(s) inthe wild. Whether limitations to routine metabolic scope set

2693


Journa

lofEx

perimen

talB

iology

biogeographical borders and, thus, limits to the realised niche needsto be investigated more widely. In the context of thermal biologyand climate change, we require integrative, ecosystem-oriented andevolutionary modes of interpretation, ideally combining field andexperimental data. With this aim, the OCLTT concept hasconnected ecological and physiological findings in animals, andwe hope that these connections will be developed further in thefuture.

AcknowledgementsWe appreciate the lively discussion of the OCLTT concept and its implications forconventional measures of thermal tolerance with T. Wang and his colleagues. Wegratefully acknowledge the constructive comments made by our reviewers, as wellas, last but not least, the many contributions to the field by the members of ourIntegrative Ecophysiology section.

Competing interestsThe authors declare no competing or financial interests.

FundingSupported by the PACES (Polar regions And Coasts in the changing Earth System)program of the Alfred-Wegener-Institute, Helmholtz Centre for Polar and MarineResearch, and Deutsche Forschungsgemeinschaft [Po 278, Ma 4271].

ReferencesAngilletta, M. J. (2009). Thermal Adaptation: A Theoretical and EmpiricalSynthesis. Oxford University Press.

Anttila, K., Couturier, C. S., Øverli, Ø., Johnsen, A., Marthinsen, G., Nilsson,G. E. and Farrell, A. P. (2014). Atlantic salmon show capability for cardiacacclimation to warm temperatures. Nat. Commun.

Anttila, K., Lewis, M., Prokkola, J. M., Kanerva, M., Seppanen, E., Kolari, I. andNikinmaa, M. (2015). Warm acclimation and oxygen depletion induce species-specific responses in salmonids. J. Exp. Biol. 218, 1471-1477.

Beers, J. M. and Sidell, B. D. (2011). Thermal tolerance of Antarctic notothenioidfishes correlates with level of circulating hemoglobin. Physiol. Biochem. Zool. 84,353-362.

Bozinovic, F. and Portner, H. O. (2015). Physiological ecology meets climatechange. Ecol. Evol. 5, 1025-1030.

Breau, C., Cunjak, R. A. and Peake, S. J. (2011). Behaviour during elevated watertemperatures: can physiology explain movement of juvenile Atlantic salmon tocool water? J. An. Ecol. 80, 844-853.

Brett, J. R. (1971). Energetic responses of salmon to temperature: a study of somethermal relations in the physiology and freshwater ecology of sockeye salmon(Oncorhynchus nerka). Am. Zool. 11, 99-113.

Buckley, B. A., Hedrick, M. S. and Hillman, S. S. (2014). Cardiovascular oxygentransport limitations to thermal niche expansion and the role of environmental PO2

in Antarctic notothenioid fishes. Physiol. Biochem. Zool. 87, 499-506.Chen, Z., Snow, M., Lawrence, C. S., Church, A. R., Narum, S. R., Devlin, R. H.and Farrell, A. P. (2015). Selection for upper thermal tolerance in rainbow trout(Oncorhynchus mykiss Walbaum). J. Exp. Biol. 218, 803-812.

Clarke, A. and Portner, H. O. (2010). Temperature, metabolic power and theevolution of endothermy. Biol. Rev. 85, 703-727.

Dahlke, F. T., Leo, E., Mark, F. C., Portner, H. O., Bickmeyer, U., Frickenhaus, S.and Storch, D. (2016). Effects of ocean acidification increase embryonicsensitivity to thermal extremes in Atlantic cod, Gadus morhua. Glob. ChangeBiol. 23, 1499-1510.

Del Raye, G. and Weng, K. C. (2015). An aerobic scope-based habitat suitabilityindex for predicting the effects of multi-dimensional climate change stressors onmarine teleosts. Deep-Sea Res. II 113, 280-290.

Deutsch, C., Ferrel, A., Seibel, B., Portner, H.-O. and Huey, R. B. (2015). Climatechange tightens a metabolic constraint on marine habitats. Science 348,1132-1135.

Ekstrom, A., Brijs, J., Clark, T. D., Grans, A., Jutfelt, F. and Sandblom, E. (2016).Cardiac oxygen limitation during an acute thermal challenge in the Europeanperch: effects of chronic environmental warming and experimental hyperoxia.Am. J. Physiol. 311, R440-R449.

Ekstrom, A., Axelsson, M., Grans, A., Brijs, J. and Sandblom, E. (2017).Influence of the coronary circulation on thermal tolerance and cardiacperformance during warming in rainbow trout. Am. J. Physiol. Regul. Integr.Comp. Physiol. 312, R549-R558.

Eliason, E. J., Clark, T. D., Hague, M. J., Hanson, L. M., Gallagher, Z. S., Jeffries,K. M., Gale, M. K., Patterson, D. A., Hinch, S. and Farrell, A. P. (2011).Differences in thermal tolerance among sockeye salmon populations. Science332, 109-112.

Eliason, E. J., Wilson, S. M., Farrell, A. P., Cooke, S. J. and Hinch, S. G. (2013).Low cardiac and aerobic scope in a coastal population of sockeye salmonOncorhynchus nerka with a short upriver migration. J. Fish Biol. 82, 2104-2112.

Farrell, A. P. (1991). From hagfish to tuna: a perspective on cardiac function in fish.Physiol. Zool. 64, 1137-1164.

Farrell, A. P. (2009). Environment, antecedents and climate change: lessons fromthe study of temperature physiology and river migration of salmonids. J. Exp. Biol.212, 3771-3780.

Farrell, A. P. (2013). Aerobic scope and its optimum temperature: clarifying theirusefulness and limitations. J. Exp. Biol. 216, 4493-4494.

Farrell, A. P. and Clutterham, S. M. (2003). On-line venous oxygen tensions inrainbow trout during graded exercise at two acclimation temperatures. J. Exp. Biol.206, 487-496.

Farrell, A. P., Johansen, J. A., Steffensen, J. F., Moyes, C. D., West, T. G. andSuarez, R. K. (1990). Effects of exercise training and coronary ablation onswimming performance, heart size and cardiac enzymes in rainbow trout,Oncorhynchus mykiss. Can. J. Zool. 68, 1174-1179.

Fobian, D., Overgaard, J. and Wang, T. (2014). Oxygen transport is notcompromised at high temperature in pythons. J. Exp. Biol. 217, 3958-3961.

Franklin, C. E., Farrell, A. P., Altimiras, J. and Axelsson, M. (2013). Thermaldependence of cardiac function in arctic fish: implications of a warming world.J. Exp. Biol. 216, 4251-4255.

Frederich, M. and Portner, H.-O. (2000). Oxygen limitation of thermal tolerancedefined by cardiac and ventilatory performance in the spider crabMaja squinado.Am. J. Physiol. 279, R1531-R1538.

Giomi, F. and Portner, H.-O. (2013). A role for haemolymph oxygen capacity in heattolerance of eurythermal crabs. Front. Physiol. 4, 110.

Giomi, F., Fusi, M., Barausse, A., Mostert, B., Portner, H.-O. and Cannicci, S.(2014). Improved heat tolerance in air drives the recurrent evolution of air-breathing. Proc. R. Soc. B 218, 20132927.

Grans, A., Jutfelt, F., Sandblom, E., Jonsson, E., Wiklander, K., Seth, H.,Olsson, C., Dupont, S., Ortega-Martinez, O., Einarsdottir, I. et al. (2014).Aerobic scope fails to explain the detrimental effects on growth resulting fromwarming and elevated CO2 in Atlantic halibut. J. Exp. Biol. 217, 711-717.

Guderley, H. Portner, H.-O. (2010). Metabolic power budgeting and adaptivestrategies in zoology: examples from scallops and fish. Can. J. Zool. 88, 753-763.

Heise, K., Puntarulo, S., Nikinmaa, M., Abele, D. and Portner, H.-O. (2006).Oxidative stress during stressful heat exposure and recovery in the North Seaeelpout (Zoarces viviparus). J. Exp. Biol. 209, 353-363.

Hochachka, P. W. and Somero, G. N. (2002). Biochemical Adaptation: Mechanismand Process in Physiological Evolution. Oxford: University Press.

Holt, R. E. and Jorgensen, C. (2015). Climate change in fish: effects of respiratoryconstraints on optimal life history and behaviour. Biol. Lett. 11, 20141032.

Iftikar, F. I. and Hickey, A. J. R. (2013). Do mitochondria limit hot fish hearts?Understanding the role of mitochondrial function with heat stress in Notolabruscelidotus. PLoS ONE 8, e64120.

Jackson, D. C. (2007). Temperature and hypoxia in ectothermic tetrapods.J. Therm. Biol. 32, 125-133.

Jakob, L., Axenov-Gribanov, D. V., Gurkov, A. N., Ginzburg, M., Bedulina, D. S.,Timofeyev, M. A., Luckenbach, T., Lucassen, M., Sartoris, F. J. and Portner,H. O. (2016). Lake Baikal amphipods under climate change: thermal constraintsand ecological consequences. Ecosphere 7, e01308.

Jones, M. C. and Cheung, W. W. L. (2014). Multi-model ensemble projections ofclimate change effects on global marine biodiversity. ICES J. Mar. Sci. 72,741-752.

Jones, S. J., Mieszkowska, N. and Wethey, D. S. (2009). Linking thermaltolerances and biogeography: Mytilus edulis (l.) at its southern limit on the eastcoast of the United States. Biol. Bull. 217, 73-85.

Kassahn, K., Crozier, R. H., Portner, H. O. and Caley, M. J. (2009). Animalperformance and stress: responses and tolerance limits at different levels ofbiological organisation. Biol. Rev. 84, 277-292.

Knoll, A. H., Bambach, R. K., Payne, J. L., Pruss, S. and Fischer, W. W. (2007).Paleophysiology and end-Permian mass extinction. Earth Planet. Sci. Lett. 256,295-313.

Kyprianou, T.-D., Portner, H. O., Anestis, A., Kostoglou, B. and Michaelidis, B.(2010). Metabolic andmolecular stress responses of gilthead seam breamSparusaurata during exposure to low ambient temperature: an analysis of mechanismsunderlying the winter syndrome. J. Comp. Physiol. B 180, 1005-1018.

Lannig, G., Bock, C., Sartoris, F. J. and Portner, H. O. (2004). Oxygen limitation ofthermal tolerance in cod, Gadus morhua L. studied by non-invasive NMRtechniques and on-line venous oxygen monitoring. Am. J. Physiol. 287,R902-R910.

Leo, E., Kunz, K. L., Schmidt, M., Storch, D., Portner, H.-O. and Mark, F. C.(2017). Mitochondrial acclimation potential to ocean acidification and warming ofPolar cod (Boreogadus saida) and Atlantic cod (Gadus morhua). Front. Zool. 14,21.

Little, A. G., Kunisue, T., Kannan, K. and Seebacher, F. (2013). Thyroid hormoneactions are temperature-specific and regulate thermal acclimation in zebrafish(Danio rerio). BMC Biol. 11, 26.

2694


Journa

lofEx

perimen

talB

iology

Lurman, G. J., Bock, C. H. and Portner, H.-O. (2007). An examination of themetabolic processes under-pinning critical swimming in Atlantic cod (Gadusmorhua L) using in vivo 31P-NMR spectroscopy. J. Exp. Biol. 210, 3749-3756.

Lutterschmidt, W. I. and Hutchison, V. H. (1997). The critical thermal maximum:history and critique. Can. J. Zool. 75, 1561-1574.

Mark, F. C., Bock, C. and Portner, H.-O. (2002). Oxygen limited thermal tolerancein Antarctic fish investigated by MRI and 31P-MRS. Am. J. Physiol. 283,R1254-R1262.

Melzner, F., Bock, C. and Portner, H.-O. (2006). Critical temperatures in thecephalopod Sepia officinalis investigated using in vivo 31P-NMR spectroscopy.J. Exp. Biol. 209, 891-906.

Munday, P. L., Crawley, N. and Nilsson, G. E. (2009). Interacting effects ofelevated temperature and ocean acidification on the aerobic performance of coralreef fishes. Mar. Ecol. Prog. Ser. 388, 235-242.

Neuheimer, A. B., Thresher, R. E., Lyle, J. M. and Semmens, J. M. (2011).Tolerance limit for fish growth exceeded by warming waters. Nat. Clim. Change 1,110-113.

Overgaard, J., Andersen, J. L., Findsen, A., Pedersen, P. B. M., Hansen, K.,Ozolina, K. and Wang, T. (2012). Aerobic scope and cardiovascular oxygentransport is not compromised at high temperatures in the toad Rhinella marina.J. Exp. Biol. 215, 3519-3526.

Ozolina, K., Shiels, H. A., Ollivier, H. and Claireaux, G. (2016). Intraspecificindividual variation of temperature tolerance associated with oxygen demand inthe European sea bass (Dicentrarchus labrax). Conserv. Physiol. 4, cov060.

Payne, N. L., Smith, J. A., van der Meulen, D. E., Taylor, M. D., Watanabe, Y. Y.,Takahashi, A., Marzullo, T. A., Gray, C. A., Cadiou, G. and Suthers, I. M. (2016).Temperature dependence of fish performance in the wild: links with speciesbiogeography and physiological thermal tolerance. Funct. Ecol. 30, 903-912.

Peake, S. J. and Farrell, A. P. (2006). Fatigue is a behavioural response inrespirometer-confined smallmouth bass. J. Fish Biol. 68, 1742-1755.

Peck, L. S., Clark, M. S., Morley, S. A., Massey, A. andRossetti, H. (2009). Animaltemperature limits and ecological relevance: effects of size, activity and rates ofchange. Funct. Ecol. 23, 248-256.

Perry, A. L., Low, P. J., Ellis, J. R. andReynolds, J. D. (2005). Climate change anddistribution shifts in marine fishes. Science 308, 1912-1915.

Portner, H.-O. (2001). Climate change and temperature-dependent biogeography:oxygen limitation of thermal tolerance in animals. Naturwissenschaften 88,137-146.

Portner, H.-O. (2002). Climate variations and the physiological basis of temperaturedependent biogeography: systemic to molecular hierarchy of thermal tolerance inanimals. Comp. Biochem. Physiol. A 132, 739-761.

Portner, H.-O. (2004). Climate variability and the energetic pathways of evolution:the origin of endothermy in mammals and birds. Physiol. Biochem. Zool. 77,959-981.

Portner, H.-O. (2006). Climate dependent evolution of Antarctic ectotherms: anintegrative analysis (EASIZ, SCAR). Deep Sea Res. II 53, 1071-1104.

Portner, H.-O. (2010). Oxygen- and capacity-limitation of thermal tolerance: amatrixfor integrating climate-related stressor effects in marine ecosystems. J. Exp. Biol.213, 881-893.

Portner, H.-O. (2012). Integrating climate-related stressor effects on marineorganisms: unifying principles linking molecule to ecosystem-level changes.Mar. Ecol. Progr. Ser. 470, 273-290.

Portner, H.-O. (2014). How to and how not to investigate the oxygen and capacitylimitation of thermal tolerance (OCLTT) and aerobic scope. J. Exp. Biol. 217,4432-4435.

Portner, H.-O. and Farrell, A. P. (2008). Physiology and climate change. Science322, 690-692.

Portner, H.-O. and Giomi, F. (2013). Nothing in experimental biology makes senseexcept in the light of ecology and evolution. J. Exp. Biol. 216, 4494-4495.

Portner, H.-O. and Knust, R. (2007). Climate change affects marine fishes throughthe oxygen limitation of thermal tolerance. Science 315, 95-97.

Portner, H.-O. and Peck, M. (2010). Climate change impacts on fish and fisheries:towards a cause and effect understanding. J. Fish Biol. 77, 1745-1779.

Portner, H.-O., MacLatchy, L. M. and Toews, D. P. (1991). Metabolic responses ofthe toad Bufo marinus to environmental hypoxia: An analysis of the critical Po2.Physiol. Zool. 64, 836-849.

Portner, H.-O., Langenbuch, M. and Michaelidis, B. (2005). Synergistic effects ofincreased CO2, temperature and hypoxia onmarine animals: from Earth history toglobal change. J. Geophys. Res. 110, C09S10.

Portner, H.-O., Peck, L. S. andHirse, T. (2006). Hyperoxia alleviates thermal stressin the Antarctic bivalve, Laternula elliptica: evidence for oxygen limited thermaltolerance. Polar Biol. 29, 688-693.

Portner, H.-O., Bock, C., Knust, R., Lannig, G., Lucassen, M., Mark, F. C. andSartoris, F. J. (2008). Cod and climate in a latitudinal cline: physiological analysesof climate effects in marine fishes. Clim. Res. 37, 253-270.

Portner, H.-O., Schulte, P. M., Wood, C. M. and Schiemer, F. (2010). Nichedimensions and limits in fishes: An integrative view. Illustrating the role ofphysiology in understanding ecological realities. Physiol. Biochem. Zool. 83,808-826.

Portner, H.-O., Peck, L. S. and Somero, G. N. (2012). Mechanisms definingthermal limits and adaptation in marine ectotherms: an integrative view. InAntarctic Ecosystems: An ExtremeEnvironment in a ChangingWorld, 1st edn (ed.A. D. Rogers, N. M. Johnston, E. J. Murphy and A. Clarke), pp. 360-396.Chichester: Blackwell Publishing Ltd.

Portner, H.-O., Walther, K. and Wittmann, A. (2013). Excess oxygen in polarevolution: a whole organism perspective. In Adaptation and Evolution in MarineEnvironments, Vol. 2 (ed. C. Verde and G. di Prisco), pp. 67-87. Heidelberg:Springer.

Portner,H.-O., Karl, D.M., Boyd, P.W., Cheung,W.W. L., Lluch-Cota, S. E., Nojiri,Y., Schmidt, D. N. and Zavialov, P. O. (2014). Ocean systems. InClimate Change2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects.Contribution of Working Group II to the Fifth Assessment Report of theIntergovernmental Panel on Climate Change (ed. C.B. Field, V.R. Barros, D.J.Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O.Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R.Mastrandrea and L.L. White), pp. 411-484. Cambridge, NY, USA: CambridgeUniversity Press.

Poletto, J. B., Cocherell, D. E., Baird, S. E., Nguyen, T. X., Cabrera-Stagno, V.,Farrell, A. P. and Fangue, N. A. (2017). Unusual aerobic performance at hightemperatures in juvenile Chinook salmon, Oncorhynchus tshawytscha. Cons.Physiol. 5, cow067.

Poloczanska, E. S., Brown, C. J., Sydeman,W. J., Kiessling,W., Schoeman, D. S.,Moore, P. J., Brander,K., Bruno, J. F.,Buckley, L.B., Burrows,M. T. et al. (2013).Global imprint of climate change on marine life. Nat. Clim. Change 3, 919-925.

Poloczanska, E. S., Hoegh-Guldberg,O., Cheung,W., Portner, H.O. andBurrows,M. (2014). Cross-chapter box on observed global responses of marinebiogeography, abundance, and phenology to climate change. In Climate Change2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects.Contribution of Working Group II to the Fifth Assessment Report of theIntergovernmental Panel on Climate Change (ed. C.B. Field, V.R. Barros, D.J.Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O.Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R.Mastrandrea and L.L. White), pp. 123-127. Cambridge, NY, USA: CambridgeUniversity Press.

Santer, R. M. (1985). Morphology and innervation of the fish heart. Adv. Anat.Embryol. Cell. Biol. 89, 1-102.

Sartoris, F. J., Bock, C., Serendero, I., Lannig, G. and Portner, H. O. (2003).Temperature-dependent changes in energy metabolism, intracellular pH andblood oxygen tension in the Atlantic cod, Gadus morhua. J. Fish Biol. 62,1239-1253.

Schroer, M., Wittmann, A. C., Gruner, N., Steeger, H.-U., Bock, C., Paul, R. andPortner, H.-O. (2009). Oxygen limited thermal tolerance and performance in thelugworm Arenicola marina: a latitudinal comparison. J. Exp. Mar. Biol. Eol. 372,22-30.

Schulte, P. M. (2015). The effects of temperature on aerobic metabolism: towards amechanistic understanding of the responses of ectotherms to a changingenvironment. J. Exp. Biol. 218, 1856-1866.

Schwerin, S., Zeis, B., Horn, W., Horn, H. and Paul, R. J. (2010). Hemoglobinconcentration in Daphnia (D. galeata-hyalina) from the epilimnion is related to thestate of nutrition and the degree of protein homeostasis. Limnol. Oceanogr. 55,639-652.

Seebacher, F. and Franklin, C. E. (2011). Physiology of invasion: cane toads areconstrained by thermal effects on physiological mechanisms that supportlocomotor performance. J. Exp. Biol. 214, 1437-1444.

Somero, G. N. (2010). The physiology of climate change: how potentials foracclimatization and genetic adaptation will determine ‘winners’ and ‘losers’.J. Exp. Biol. 213, 912-920.

Storch, D., Fernandez, M., Navarrete, S. A. and Portner, H.-O. (2011). Thermaltolerance of larval stages of the Chilean kelp crab Taliepus dentatus. Mar. Ecol.Progr. Ser. 429, 157-167.

Storch, D., Menzel, L., Frickenhaus, S. and Portner, H.-O. (2014). Climatesensitivity across the domains of life: Limits to evolutionary adaptation shapespecies interactions. Global Change Biol. 20, 3059-3067.

Sunday, J. M., Bates, A. E. and Dulvy, N. K. (2012). Thermal tolerance and theglobal redistribution of animals. Nat. Clim. Change 2, 686-690.

Sunday, J. M., Bates, A. E., Kearney, M. R., Colwell, R. K., Dulvy, N. K., Longino,J. T. and Huey, R. B. (2014). Thermal-safety margins and the necessity ofthermoregulatory behavior across latitude and elevation. Proc. Natl. Acad. Sci.USA 111, 5610-5615.

Takasuka, A., Oozeki, Y. and Aoki, I. (2007). Optimal growth temperaturehypothesis: why do anchovy flourish and sardine collapse or vice versa under thesame ocean regime? Can. J. Fish. Aqu. Sci. 64, 768-776.

Tomanek, L. and Somero, G. N. (2002). Interspecific- and acclimation-inducedvariation in levels of heat-shock proteins 70 (hsp70) and 90 (hsp90) andheat-shock transcription factor-1 (HSF1) in congeneric marine snails (genusTegula): implications for regulation of hsp gene expression. J. Exp. Biol. 205,677-685.

2695


Journa

lofEx

perimen

talB

iology

Verberk, W. C. E. P. and Calosi, P. (2012). Oxygen limits heat tolerance and drivesheat hardening in the aquatic nymphs of the gill breathing damselfly Calopteryxvirgo (Linnaeus, 1758). J. Therm. Biol. 37, 224-229.

Walther, K., Sartoris, F. J., Bock, C. and Portner, H. O. (2009). Impact ofanthropogenic ocean acidification on thermal tolerance of the spider crab Hyasaraneus. Biogeosciences 6, 2207-2215.

Wang, T., Lefevre, S., Iversen, N. K., Findorf, I., Buchanan, R. and McKenzie,D. J. (2014). Anaemia only causes a small reduction in the upper criticaltemperature of sea bass: is oxygen delivery the limiting factor for tolerance ofacute warming in fishes? J. Exp. Biol. 217, 4275-4278.

Weibel, E. R., Taylor, C. R. and Hoppeler, H. (1991). The concept ofsymmorphosis: a testable hypothesis of structure-function relationship. Proc.Natl. Acad. Sci. USA 88, 10357-10361.

Wittmann, A., Schroer, M., Bock, C., Steeger, H.-U., Paul, R. and Portner, H. O.(2008). Indicators of oxygen- and capacity-limited thermal tolerance in thelugworm Arenicola marina. Clim. Res. 37, 227-240.

Wittmann, A. C., Portner, H. O. and Sartoris, F. J. (2012). A role for oxygendelivery and extracellular magnesium in limiting cold tolerance of the sub-antarcticstone crab Paralomis granulosa? Physiol. Biochem. Zool. 85, 285-298.

Zakhartsev, M. V., De Wachter, B., Sartoris, F. J., Portner, H. O. and Blust, R.(2003). Thermal physiology of the common eelpout (Zoarces viviparus). J. Comp.Physiol. B 173, 365-378.

Zittier, Z. M. C., Hirse, T. and Portner, H.-O. (2012). The synergistic effects ofincreasing temperature and CO2 levels on activity capacity and acid–basebalance in the spider crab, Hyas araneus. Mar. Biol. 160, 2049-2062.

2696


Journa

lofEx

perimen

talB

iology

ATTACHMENT D

Analysis of Diurnal Temperature Variation Along the

Tuolumne River

ATTACHMENT D

Analysis of Diurnal Temperature Variation Along the Tuolumne River

On page 2 and 3 of the April 11, 2019, comments on FERC’ DEIS for the Don Pedro and La Grange projects, NMFS discusses the relationship of the EPA (2003) 7DADM temperature “criterion” of 18°C for juvenile O. mykiss to the site-specific metabolic rate studies on the Tuolumne River reported in Farrell et al (2017) and Verhille et al. (2016). On page 3, NMFS suggests that while its study of anadromous fish introduction to the upper Tuolumne River (Boughton et al. 2018) indicates that the upper Tuolumne temperatures may not meet the EPA (2003) “criterion”, the EPA (2003) 7DADM “criterion” of 18°C may not necessarily apply to the upper Tuolumne River. NMFS states: “The upper Tuolumne basin experiences high diurnal variation that make it a potential candidate for a temperature metric that does not use the 7DADM metric”. NMFS did not provide any temperature data or analysis to describe the referenced “diurnal temperature variation” occurring in the upper Tuolumne River which qualifies for development of “site-specific criteria”. The Districts have proceeded to conduct a comparative analysis of diurnal water temperature variation in relevant reaches of the upper and lower Tuolumne River, each corresponding to the middle and lower sections of available O. mykiss habitat. These locations are described below:

RM 90.8 and 85.9 of the upper Tuolumne River, which are, respectively, seven and 11 miles downstream of Lumsden Falls;

RM 47 and 40 of the lower Tuolumne River, which are, respectively, five and 12 miles downstream of La Grange Diversion Dam.

The assessment of diurnal temperature variations used temperature data and analyses provided in the La Grange and Don Pedro license applications filed with FERC in October 2017. Figure 1 below compares the amount of diurnal temperature variation occurring at each of the four locations. As shown in the plot, the diurnal temperature variation experienced in both locations in the lower Tuolumne River exceed the diurnal temperature variation experienced in the upper Tuolumne River. The upper Tuolumne River diurnal temperature variation was deemed by NMFS to be large enough to qualify as a reach where EPA (2003) metric would not necessarily apply and a site-specific criteria should be developed. Given the substantially larger diurnal temperature variation experienced in the lower Tuolumne River, it follows that NMFS recognizes that the EPA (2003) temperature metric may not be appropriate for the lower Tuolumne River, and a site-specific metric is warranted.

Figure 1. Diurnal temperature variation in the Upper and Lower Tuolumne River.

ATTACHMENT E

Response to NMFS Comments

on FERC DEIS, Page 6-7

ATTACHMENT E Regarding DEIS p.3-169

Response to NMFS Comments on FERC DEIS, page 6-7

NMFS claims that the Anchor QEA (2017) report (Anchor report) demonstrates that a Hybrid Downstream Collection Facility (NMFS’ proposed Hybrid Concept) is “feasible”, would collect downstream migrating salmonids at flows up to 4,500 cfs, and would meet the three “Feasibility Factors” outlined in the Districts’ review of the Anchor report which the Districts filed with FERC on March 15, 2018. NMFS also comments that the Districts’ rationale for finding that the Hybrid Concept falls short of meeting the defined Feasibility Factors is based upon procedural arguments and not founded on engineering or constructability factors. NMFS’ comments lack technical merit and, contrary to NMFS’ assertions, the Districts’ rationale for finding the Hybrid Concept to be infeasible is based on substantive engineering and constructability factors that were either under-evaluated or not evaluated at all in the Anchor report. The Districts’ technical review demonstrates why the Hybrid Concept endorsed by NMFS does not meet minimum engineering, constructability, or operational requirements (Feasibility Factor 1), cannot operate without interfering with existing uses (Feasibility Factor 2), and is highly unlikely to be able to meet usual and customary fish passage performance standards (Feasibility Factor 3). The Hybrid Concept has been determined to be infeasible based upon the Districts’ detailed assessment of the content provided in the Anchor report. The full review can be found in Attachment B to the Districts’ March 15, 2018, submittal to FERC. A summary of the findings of the Districts’ review related to the Feasibility Factors is provided below. Feasibility Factor 1: Ability to Meet Engineering, Constructability, and Operational Constraints: Numerous fundamental engineering deficiencies, spatial discrepancies, and operational concerns underlay the Hybrid Concept. A primary weakness of the Hybrid Concept is the failure of the proposal to meet fundamental engineering criteria that must be considered when proposing to build a large dam in a remote location on the Tuolumne River. As one example, the Hybrid Concept lacks even the most rudimentary description of site geology, seismicity, or assessment of the potential foundation conditions at the dam site. There is no description or any schematic representation of how the structure will be secured to competent rock, or even if there is expected to be competent rock, in the river bed. This represents a significant engineering, constructability, and cost estimating deficiency. This alone renders the Hybrid Concept to be no more than an unproven concept not yet investigated at the feasibility level. Another example, essential to the initial planning of a new dam, is the requirement to make certain it will safely pass without failure or significant damage what engineers refer to as the “design flood”. At a minimum the design flood for the Hybrid Concept would be the 100-year event, and likely a more extreme event due to the amount of recreation activity that occurs in the vicinity and

immediately downstream of the proposed dam. As stated on page 44 of Anchor report, the proposed Hybrid Concept is configured to safely pass a flood event of 11,200 cubic feet per second (cfs). As shown in the Districts’ review, this flow would be expected to be exceeded, on average, about 3 to 4 days each year. The minimum “design flood” event at the proposed dam site has a flow magnitude of over 110,000 cfs, or ten times greater than that currently planned to be accommodated by the Hybrid Concept. This is one of the fatal flaws in planning for the Hybrid Concept. A dam needing to safely pass a flow of 110,000 cfs without significant damage is likely to look substantially different, and have substantially greater cost, than the current Hybrid Concept. The flood hydrology of the proposed dam site, in addition to other standard hydraulic design considerations, is not adequately treated in the Anchor report. As mentioned, the 100-year flood recurrence would be the minimum “design flood” to be safely passed, and the actual “design flood” for the Hybrid Concept likely to be required by regulatory bodies would be significantly greater. To be considered an actual engineering concept, the proposed Hybrid Concept must depict and describe how the facility would respond to the design flood event. Such an assessment would include an analysis of the structural stability of the dam and its components under the flood event and the potential for erosion and scour to the surrounding landscape, including the river bed at and downstream of the dam when the facility is overtopped. These are important factors in selecting a spillway or dam shape that can accommodate flood flows. Other hydraulic design considerations that are not addressed by the Hybrid Concept are provided in Attachment B of the Districts’ March 15, 2018, submittal to FERC (see pages 20 through 23). A proposed dam and its associated components, which in this case include all the fish collection and transport facilities, must be able to physically fit into the site proposed to be used. Even without considering modifications that would have to be made to handle a realistic design flood, the Hybrid Concept would not fit within the site identified in Anchor report without major abutment and bank excavations, none of which are depicted on the report’s schematics or included in the cost estimates provided in the report. The reason for the lack of fit is the Anchor report appears to be based on a gross level of site topography information. The Districts obtained detailed site topography when considering recreation-related facilities very near the dam site at Wards Ferry Bridge. Using the available site survey, the Districts developed more detailed schematics and illustrations of the proposed Hybrid Concept and attempted to fit the facility into existing topography of the dam site. In brief, the steep hillslopes and narrow canyon walls do not provide the space needed to accommodate the structural, fish passage, and accessibility elements of the Hybrid Concept, even without considering any needed modifications to the concept to accommodate a design flood. In contrast to what is depicted in the Anchor report, the fish screening facility would end up being embedded deep into the left hillslope, the foundation and hydraulic conveyance elements of the dam are obstructed by canyon walls (thereby requiring substantial rock excavations to fit the facility), and proposed access roads don’t actually fall upon the existing ground. This all would add up to the need for significant additional slope excavations, rock stabilization, embankments, and retaining walls which are neither illustrated as part of the Hybrid Concept nor included in the construction cost opinion or an assessment of constructability. To fit within the existing site constraints, the concept would need to be scaled down, thus reducing its fish collection and screening capacity well below the stated 4,500 cfs.

Planning for dam construction requires that substantial thought be given to the construction process and overall constructability. One example of the lack of consideration of constructability demonstrated in the Anchor report relates to the fact that vehicular access to the dam site is quite limited. Access to the current site for the purposes of construction and/or operation would have to occur via the Wards Ferry Road -- an existing steep-grade, one-lane road currently in poor condition. The road does not meet current standards with regard to size, alignment, or safety for large vehicles. To accommodate heavy construction vehicle traffic, the entire several mile long Wards Ferry Road would require a comprehensive series of improvements to support the activities required by the proposed concept. This consideration is not adequately or seriously evaluated in the Anchor report or in the construction sequencing or cost opinion. There are also a number of operational issues associated with the Hybrid Concept that are not recognized or discussed in the Anchor report. Operational concerns with the Hybrid Concept range from the inability to control flows through lower fish collection facility outlets or over the crest of the facilities due to the lack of details with regard to the critical water return structure to be built on the left bank. In addition, experience has shown that the hydraulic performance and reliability of similar facilities are heavily influenced by the amount of debris present within the water body. In practice, such facilities incorporate primary, secondary, and tertiary debris management systems to reduce the impacts that large and small debris have on the dewatering screens and fish collection and holding elements. Insufficient debris management strategies result in poor hydraulic conditions and greater fish mortality. The debris management systems, and the importance associated with operation of these systems, are not sufficiently depicted to adequately assess if operation of the Hybrid Concept is workable without major interferences caused by debris. As evidenced by the example issues discussed above, the Districts’ critique of, and concerns with, the Hybrid Concept are not based on “procedural arguments”. Feasibility Factor 2: Ability to Operate without Interference with Existing Uses One of the existing uses of the area under consideration for the Hybrid Concept is whitewater boating in the Wild & Scenic River reach that abuts the Don Pedro Project upstream boundary. The Anchor report clearly describes the location of the Tuolumne River Wild & Scenic river reach and states that alternatives that create backwater impacts to the Wild and Scenic portion of the Tuolumne River upstream of the Don Pedro Project boundary are considered infeasible under existing federal statutes. Therefore, the Anchor report acknowledges that any concept which results in such an impact is, by definition, infeasible. Nonetheless, the Anchor report fails to evaluate whether and to what degree the Hybrid Concept would impact the Wild & Scenic reach. This is a potentially fatal flaw. Having clearly identified the possibility of a fatal flaw, the probability that the Hybrid Concept may contain the flaw must be examined before the concept is otherwise considered for development as a fish passage concept. A brief backwater analysis was performed as part of the Districts’ review. Results indicate that construction and operation of a structure at the proposed dam site with the dimensions and conveyance characteristics of the Hybrid Concept, to the extent depicted in the Anchor report, will likely cause frequent hydraulic backwater effects on the Wild and Scenic reach. Such impacts may occur any time the hydraulic capacity of the proposed fish passage facility is exceeded and water is directed over the crest of

the structure (elevation 835 feet). While the Districts’ brief study of this issue is preliminary, the fact that it was not examined at all in the Anchor report is a serious flaw, and serves as another demonstration that the evaluation of the Hybrid Concept in the Anchor report falls short of even being considered a reconnaissance-level study, let alone a full feasibility investigation. Feasibility Factor 3: Ability to Meet Usual and Customary Fish Passage Performance Standards Fish passage performance cannot be deemed to be met by merely taking a downstream fish passage technology that appears to have worked satisfactorily in one environment and applying it to a completely different site, in this case, to the head of Don Pedro Reservoir. Using this line of reasoning does not provide an adequate determination of “feasibility” to meet the usual and customary fish passage performance metrics and goals. The application of downstream fish passage technologies must carefully consider the physical, biological, and operational conditions of the environment within which they are to be implemented. As elaborated in Kock et al. (2019), similar downstream fish passage technologies perform very differently in different environments. A floating surface collector operating in one reservoir may have a high collection efficiency whereas a floating surface collector of similar size, shape, and hydraulic capacity operating in another reservoir may have very poor collection facility. Such is the case when considering the applicability to the Hybrid Concept of downstream fish passage technologies installed at Rocky Reach on the Columbia River, Wells Dam on the Columbia River, and the Coanda screens at Butte Creek. These technologies take advantage of very specific hydraulic, structural, and fish behavioral conditions exhibited in the environment within which they are applied. These same conditions are unlikely to exist at the head of Don Pedro Reservoir and therefore they are not adequate analogues for determination of fish passage performance. A discussion and assessment of each fish passage technology proposed for incorporation into the Hybrid Concept is provided in the Districts’ March 15, 2018, submittal to FERC. A summary of conclusions relative to the potential passage performance of the Hybrid Concept follows:

The performance of downstream fish passage facilities relies on an understanding of fish migrational cues, behavioral response to hydraulics associated with installed structures, and the effects of environmental conditions such as water quality, light, and concentrations of predators. Even where extensive studies have been conducted of downstream fish passage, researchers have acknowledged that there remains a poor understanding of the complex interplay of factors affecting the success or failure of collection and passage facilities. The Don Pedro Project presents additional challenges regarding the life history and behavioral responses of downstream migrants since site-specific information on the behaviors of Central Valley (CV) spring-run Chinook and steelhead in the Tuolumne River are not available.

The unique hydraulic and operational characteristics of the impoundment created by the proposed Hybrid Concept are not realistically considered in the Anchor report. The proposed hydraulic design characteristics and overall operating environment is very

different from that experienced on the Columbia River and therefore collection performance may vary significantly from experience gained in the Pacific Northwest.

It appears that during approximately 70 percent of the spring outmigration period the vertical slots would be operating without any attraction flow that is supposed to be provided by the low-level outlets. At river flows up to 11,200 cfs, up to 6,950 cfs could be conveyed through the low level outlets, which is almost 30 times less flow than the Columbia River facilities which are deemed to be examples used to support performance of the Hybrid Concept. The distribution of flow patterns across the face of the proposed Hybrid Concept cannot be assumed to mimic the concentration of flow that contributes to the success of the smolt bypass slots at Wells Dam. Results from a literature search on the use of Coanda-style screens indicate that Coanda’s have not been used in fish collection or bypass facilities in applications where flows exceed about 200 cfs, their capacity at the Forks of Butte hydroelectric diversion. The proposed Hybrid Concept incorporates five sections of Coanda-style screens to dewater 1,350 cfs and return it to Don Pedro Reservoir, more than six times greater than any existing facility. Therefore, this would be an experimental application of the Coanda screen technology. The Anchor report states that the expected collection efficiency would be 100 percent during normal operating conditions. This is unrealistic. There is a high level of certainty that collection efficiencies will be far less than the expectations presented in the Anchor report. The performance of the Hybrid Concept assumes that all fish migrating downstream are collected when in actuality the total fish passage efficiency of such a structure must take into consideration numerous factors such as timing of fry and juvenile outmigrations, river temperatures, velocity profiles, predation, residualization, and disease. The high level of expected performance stated in the Anchor report is unlikely to occur in practice for reasons summarized above and described in more detail in the Districts’ March 15, 2018, filing with FERC.

References Cited Anchor QEA. 2017. Conceptual Engineering Plans for Fish Passage at La Grange and Don Pedro

Dams on the Tuolumne River, prepared by Anchor QEA on behalf of NOAA Fisheries. October 2017.

Kock, Tobias, Nick Veretto, Nick Ackerman, Russel Perry, John Beeman, Michael Garello, and Scott Fielding. 2019. Assessment of operational and structural factors influencing performance of fish collectors in forebays of high-head dams. Transactions of the American Fisheries Society.

Turlock and Modesto Irrigation Districts. 2017. Attachment B to Districts’ Reply to Comments: Technical Review Summary of the Document “Conceptual Engineering Plans for Fish Passage at La Grange and Don Pedro Dams on the Tuolumne River, Prepared by Anchor QEA on Behalf of NOAA Fisheries.”

ATTACHMENT F

Additional Examples of Large-Scale and Experimental Predator

Suppression Programs

ATTACHMENT F Table 1. Selected examples of large-scale management programs in North America to remove or suppress predatory fish to benefit native fish species.

Program Protected Species

Suppressed Species Methods Program Goals Impact, Successes and Failures Source:

Northern Pikeminnow Predator Control Program [Columbia and Snake Rivers, Washington and Oregon]

Native salmonids

Northern pikeminnow (Ptychocheilus oregonensis)

Sport-reward fishery; pay anglers for fish >9" ($4-$8/fish) (Continuing)

Site specific gill nets (1994 – 2002) near areas w/ high levels of predation (e.g. hatchery release points and dams)

Trained Anglers at dams to target larger pikeminnow (1991 – 2002, 2006)

Reduce the average size of predators; curtail the number of larger, older fish

In 2016, the top 20 anglers caught an average 4,250 fish per angler and averaged $36,000

Since 1990, over 4.6 million northern pikeminnow have been removed

Estimate that predation on juvenile salmonids has been cut by 40%

Have not detected any signs of compensatory response in smallmouth bass and walleye.

Their website states: "Results indicate the [sport-reward] program is successful".

Dam angling and gill net fisheries were not very successful (discontinued in 2002)

http://www.pikeminnow.org/info.html http://www.dfw.state.or.us/fish/OSCRP/CRI/Pikeminnow.asp

Upper Colorado River Endangered Fish Recovery Program (Non-native Fish Management) [Upper Colorado River Basin tributaries, Wyoming, Colorado and Utah]

Bonytail (Gila elegans)

Humpback chub (G. cypha)

Colorado pikeminnow (Ptychocheilus Lucius)

Razorback sucker (Xyrauchen texanus)

Smallmouth bass (Micropterus dolomieu)

Northern pike (Esox Lucius)

Walleye (Sander vitreus)

Boat/raft electrofishing of 700+ miles annually (non-native fish are euthanized)

Target spawning aggregations of non-natives

Holding nonnative fishing tournaments at reservoirs

Evaluating strategic flow releases to disrupt reproduction

Considering reservoir draining and rotenone treatments

Evaluating effectiveness of exclusion barriers to limit access to nursery habitats by nonnative fishes

Planning to install nets on the reservoir spillways to avoid reintroductions

Considering production of sterile hybrid sportfish

Considering "must-kill" regulations for non-native sportfish

Overall Goals: Restoring and managing river flows and habitat, boosting wild populations with hatchery raised native fish, and non-native predator control efforts. Predator Objectives: reduce the numbers of the "worst of the worst" non-native predators to a level where they are no longer a threat. Develop sport fisheries with species that are compatible with endangered fish recovery.

Flows are managed in all Upper Basin rivers to benefit pikeminnow and chub

Management of non-native sport fish has been underway 10+ years

Removed 20,000+ nonnative channel catfish, 10,000 nonnative sunfish and bass, and 200,000 nonnative minnows.

Controlling non-native predators will require more than just mechanically removing them from rivers (USFWS website).

Walleye continue to increase in abundance and distribution, but in San Juan River researchers report declines in both juvenile and adult nonnative channel cat-fish in river reaches where the greatest amount of removal occurs.

An assessment of the native fish populations before and after removals in Middle Green River (2004-2008) found that that smallmouth bass removal did not produce a definitive response in YOY native fishes, but this may be because of the short duration and inconsistent removal efforts (Skorupski et al. 2012).

https://www.fws.gov/ecological-services/highlights/06292015.html http://www.coloradoriverrecovery.org; http://wildlife.utah.gov/fishing/nonnative/ http://www.fws.gov/coloradoriverrecovery/ http://www.fws.gov/mountain-prairie/crrip/doc/smb08Resmtg2.ppt

Strong public outreach message that all of these actions are needed to achieve recovery

In the past three years, the programs have spent more than $2.5 million per year managing nonnative fish species (Briefing Book 2016).

Mechanical Removal of Non-Native Fishes in the Colorado River [Mainstem Colorado River, Arizona]

Humpback chub (G. cypha)

Some key non-natives: Rainbow trout

(Oncorhynchus mykiss)

Brown trout (Salmo trutta)

Black bullhead (Ameiurus melas)

Electrofishing surveys (6 trips per year from 2003-2006). Non-native fish are euthanized and given to Colorado River tribal communities for use as fertilizer

Reducing non-native fish densities

Between 2003 and 2006, rainbow trout population in the Colorado River near the Little Colorado River was reduced by more than 80%.

Densities of non-native fish returned to similar levels observed in early phases of the experiment after fishing stopped. Thus, follow up was needed.

http://www.azgfd.gov/w_c/research_mechanical_removal.shtml http://www.usgs.gov/newsroom/article.asp?ID=2206

Great Lakes Restoration Initiative Action Plan (Invasive Species Program) [Lake Superior, Lake Michigan, Lake Huron, Lake Erie and Lake Ontario]

Native fishes Sea Lamprey (Petromyzon marinus)

Silvery carp (Hypophthalmichthys molitrix)

Grass carp (Ctenopharyngodon idella)

Bighead carp (Hypophthalmichthys nobilis)

Block pathways for invasions Conduct early detection

monitoring Conduct rapid response actions Implement control projects for

targeted invasive species (Ex. chemical pheromones, tributary barriers and traps for sea lampreys)

Develop/enhance technologies to prevent and control the spread (ex. seismic pressure “water guns” and carbon dioxide barriers for Asian carp

Develop invasive species collaboratives to support rapid responses

Invasive Species Goals: 1) Prevent new introductions of invasive species 2) Control established invasive species 3) Develop invasive species control technologies and refine management techniques

Sea lamprey program is "One of the longest-running and most effective invasive control technology programs".

Success largely due to a multi-year effort to test ~ 6,000 chemical compounds that could be used to effectively control lampreys without harming other species.

GLRI is working to refine control techniques and develop targeted control methods for other invasive species.

https://www.glri.us (Great Lakes Interagency Task Force 2014)

Yellowstone’s Native Fish Conservation Plan [Yellowstone Lake, Montana]

Yellowstone cutthroat trout (Oncorhynchus clarkii bouvieri)

Lake trout (Salvelinus namaycush)

Live entrapment nets, Tagging of sentinel fish Gill netting Catch-and-kill regulations Targeting larger individuals

Goal: conserve the Yellowstone cutthroat trout in Yellowstone Lake by increased netting of non-native lake trout. Objective: reduce the population by 25% each year until it collapses to an insignificant level

Over 1.7 million lake trout have been removed since 1994. CPUE rose between 2002-2012, suggesting that the

population has been increasing faster than the fish are being removed. However, increases in netting effort since 2012 has decreased CPUE.

NPS acknowledges, "With current technology, lake trout probably cannot be eliminated… However, ongoing management of the problem can significantly reduce lake trout population growth and maintain the cutthroat trout population..."

https://www.nps.gov/yell/learn/nature/laketroutinfo.htm https://parkplanning.nps.gov/projectHome.cfm?projectID=30504

Table 2. Example experimental studies examining to survival responses to nonnative fish removal conducted in California.

Watershed Protected Species Predator Species Methods Used

Research Questions Results: Conclusions

Mokelumne River at the Woodbridge Irrigation District Dam (WIDD), California, USA (Sabal et al. 2016) [East Bay Municipal Utility District, University of California, Santa Cruz, Cramer Fish Sciences, NOAA Fisheries]

juvenile Chinook Salmon (Oncorhynchus tshawytscha)

Striped Bass (STB; Morone saxatilis)

Largemouth Bass (Micropterus salmoides)

Spotted Bass (Micropterus punctulatus)

2009-2013: diet and abundance - surveyed 10 sites

using single-pass boat e-fishing STB removals and survival of paired

marked-recaptured releases of juvenile salmon before and after removal (4 e-fishing passes within a block net area)

Diet energetic analysis to estimate the population-level impact of STB predation

Before–after impact (BACI) with survival based on screw trap catch of salmon (impact = removal event). Predator removals with boat e-fishing (10 removal events over 3 years)

(1) Is the PCC of juvenile salmon by STB greater at altered habitats?; (2) Do STB aggregate at these habitats?; and (3) Within an area of high predation, what is population- level impact of STB on an emigrating salmon population?

Juvenile Chinook salmon were the dominant prey type consumed by STB caught at WIDD, but not at other locations.

STB aggregated at WIDD (eightfold increase in CPUE) compared to other habitats.

Based on energetic analysis, estimated that 7.9–13.1% of the emigrating salmon were consumed.

The BACI detected a 26–29% increase in the survival of juvenile Chinook Salmon after STB removal

The synergistic effect of habitat modification and a nonnative predator can exacerbate the mortality of juvenile salmon

North Fork Mokelumne River downstream of the Delta Cross Channel (DCC), California, USA (Cavallo et al. 2013) [EBMUD, University of California, Santa Cruz, Cramer Fish Sciences]


Fish species identified as potentially predatory (Moyle 2002)

15 and 30 May 2010: BACI in a 1.6 km predator-removal

reach (impact) and a 2.0 km control reach using 8 paired groups of acoustic-tagged fish before and after predator removal and before/after DCC opening.

two paired groups at the beginning (consecutive day) of the removal, 2 paired groups after the first removal and 2 paired groups after the second predator removal.

Removals with boat e-fishing three-pass depletion without block nets.

What is the relative effects of experimental reductions of potential predators and a flow pulse on the survival of emigrating juvenile Chinook salmon in the Delta?

Estimated 91% of predators vulnerable to e-fishing were captured first removal, and 83% second removal.

Survival improved significantly after the first removal, but estimated survival decreased to pre-impact levels after the second removal.

The DDC opening decreased emigration time and increased survival significantly through the impact reach by creating unusually high flows.

short-term study conducted in one month, but results demonstrate predator control and flow increases can be effective to enhance juvenile salmon survival. They concluded that a sustained effort over time (with daily rather than weekly removals) may be necessary to benefit salmon survival.

San Joaquin River between Port of Stockton and Lathrop, California, USA1 [University of Washington, NOAA Fisheries]


non-native fish predators

2014 and 2015: Predator population sizes from e-fishing

and hydroacoustics boat surveys (one or two multibeam sonars)

Acoustic telemetry to examine predator behaviors

Diet of predators using DNA barcoding Impacts of removals examined by

relocating predators to create low and high density reaches and determined relative predation rates with predation event recorders (PER) before and after relocation events

1) examine the abundance, distribution, and movement of predators, 2) quantify the magnitude of predation with genetic analysis of guts, 3) manipulate the density of predators to assess the influence on the predation rates, and 4) determine how predation may be influenced by environmental features.

Found differences in relative abundance, movement patterns, and smolt consumption rates among species.

2,846 predators relocated, but removals and additions had "negligible influence" on predation rates.

Predation appears to be influenced by water velocity and other environmental conditions.

Salmonid DNA was found in stomachs of 19.2% of channel catfish, 6.0% of STB, 4.8% of white catfish, and 2.8% of the largemouth bass.

Preliminary estimates suggest STB were eating 1 to 160 smolts per 1-km reach (wide variation).

TBD

1 Related talks at the 2016 Biennial Bay-Delta Science Conference. http://scienceconf2016.deltacouncil.ca.gov/sites/default/files/2016-11-17-NonNative-Predator-Fish-I-and-II.pdf

Great Lakes Interagency Task Force. 2014. The Great Lakes Restoration Initiative: Action Plan II. September 2014. Skorupski, J. A., Jr., M. J. Breen, and L. Monroe. 2012. Native Fish Response to Nonnative Fish Removal from 2005-2008 in the

Middle Green River, Utah. Publication Number 12-24. Prepared for: Upper Colorado River Basin Endangered Fish Recovery Program, Project Number: 144, Cooperative Agreement Number: 09-FG-40-2845, Salt Lake City, UT. February 2012.

CERTIFICATE OF SERVICE

I hereby certify that I have this day served the foregoing document upon each person

designated on the official service list compiled by the Secretary in these proceedings, in accordance

with Rule 2010 of the Commission’s Rules of Practice and Procedure, 18 C.F.R. § 385.2010.

Dated at Washington, D.C., this 15th day of August, 2019.

/s/ Kimberly Ognisty Kimberly Ognisty 1700 K Street N.W. Washington, D.C. 20006 (202) 282-5217 [email protected]

AmericasActive:12288517.3

via electronic filing...summary of hdr’s assessment of anchor qea (2017) report.” in attachment...

Documents