international association for ecologyangelo.berkeley.edu/wp-content/uploads/hot-rocks... ·...

Springer and International Association for Ecology are collaborating with JSTOR to digitize, preserve and extend access toOecologia.

http://www.jstor.org

International Association for Ecology

Hot Rocks or No Hot Rocks: Overnight Retreat Availability and Selection by a Diurnal Lizard Author(s): John L. Sabo Source: Oecologia, Vol. 136, No. 3 (Aug., 2003), pp. 329-335Published by: in cooperation with Springer International Association for EcologyStable URL: http://www.jstor.org/stable/4223681Accessed: 13-04-2015 00:10 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of contentin a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.For more information about JSTOR, please contact [email protected].

This content downloaded from 169.229.32.36 on Mon, 13 Apr 2015 00:10:14 UTCAll use subject to JSTOR Terms and Conditions

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=springer

http://www.jstor.org/action/showPublisher?publisherCode=intecol

http://www.jstor.org/stable/4223681

http://www.jstor.org/page/info/about/policies/terms.jsp


Oecologia (2003) 136:329-335 DOI 10. 1007/s00442-003-1292-6

John L. Sabo

Hot rocks or no hot rocks: overnight retreat availability and selection by a diurnal lizard

Received: 25 June 2002 / Accepted: 18 April 2003 / Published online: 7 June 2003 ? Springer-Verlag 2003

Abstract I used radio telemetry to determine the effects of substrate size and composition on overnight retreat site selection by western fence lizards (Sceloporus occidentalis). In watersheds of northern California (USA), these lizards occupy two habitat types differing in substrate characteristics: rocky cobble bars found in the dry, active channels of rivers and grassy upland meadows. Rocky substrates, found almost exclusively on cobble bars, provided warmer potential retreat sites than all available retreat sites on meadows during the first 5 h of inactivity. Only cobble and sand substrates provided retreats with temperatures within the preferred daily active range (32- 36?C) during the inactive period for these lizards (1900- 0900 hours). Females on a cobble bar used rocks as retreats on >90% of nights during the breeding season whereas females on a meadow used wood (>70% of nights) and burrows (>25% of nights). In contrast to females, cobble bar males used rocks significantly less frequently (<70%) and slept in the open air significantly more frequently (25% vs. <1%). Cobble bar females further, showed a significant preference for cobbles 15 cm thick, whereas the rocks used by males did not differ significantly in thickness from those measured in ran- domly placed transects. Rocks 15 cm thick were the warmest retreats commonly available on this habitat type. Thus, thermal microenvironments available to and chosen by gravid female lizards differ considerably between river and non-river habitats.

Keywords Habitat selection * Retreat site Sceloporus occidentalis . Thermoregulation . Riparian

Introduction

Substrate size determines both the texture and suitability of many aquatic and terrestrial habitats. This is especially true in stream ecosystems (Cummins and Lauff 1969; Hynes 1970; McAuliffe 1984; Minshall 1984; Mackay 1992; Allan 1995). Many sessile organisms (e.g., macro- phytes and freshwater mussels) occur only in depositional stream habitats (Hynes 1970; Strayer and Ralley 1993). By contrast, high levels of fine sediments reduce the survivorship of the eggs of many species of stream fishes (Chapman 1988; Jones et al. 1999) and substrates composed of mostly fine grains expose the larvae of aquatic insects to higher rates of predation by reducing refuge availability (Power 1992). However, despite much interest in the effects of substrate size on aquatic organisms in stream settings, little is currently known about the effect of sediment size on terrestrial organisms that occupy seasonally exposed sand- or cobble bar habitats in near-river settings.

One potentially important property of rocky substrates in exposed active channel habitats is heat conduction and storage. Many ectotherms maintain relatively constant body temperatures (Tb) by selecting microhabitats that provide environmental temperatures (Te) within a much narrower preferred range (Tp), (Cowles and Bogert 1944; Porter and Tracy 1983; Porter et al. 1973; Casey 1981; Wilimer 1982). Moreover, variation in microclimate may determine habitat selection not only during periods of activity (Huey 1982, 1991; Grant and Dunham 1988), but during periods of inactivity as well (Christian et al. 1984; Huey et al. 1989; Webb and Shine 1998, 2000; Keamey 2002). This is true not only for ectotherms, but some endotherms as well (e.g., bats, Maloney et al. 1999; Adam and Hayes 2000). One source of microclimate variation at night is differential radiation of stored heat from substrates of various sizes and composition. For example, medium-sized rocks typically store heat longer and provide warmer overnight retreats than burrows in soil (Huey et al. 1989). Heat storage by rocks in active channel habitats may thus, provide a valuable thermal

J. L. Sabo (E) Department of Integrative Biology, University of California Berkeley, Berkeley, CA 94720-3140, USA e-mail: [email protected]

Present address: J. L. Sabo, Department of Biology, Arizona State University, P.O. Box 871501, Tempe, AZ 85287-1501 USA



330

resource to ectotherms seeking refuge from cooler air temperatures at night.

Preferred overnight retreat characteristics may further differ between males and females, or juvenile and reproductive individuals of the same species, depending on daily energetic requirements. For example, females may choose warmer retreat sites than males of the same species during the mating season to increase overnight digestive efficiency and energy allocation to reproduc- tion. Warmer overnight retreats may in this way allow individual females to increase their seasonal clutch production. By contrast, males may have higher daily energy expenditures, especially if defending territories (Marler et al. 1995; Marler and Moore 1988; 1989; 1991), and may choose cooler retreat sites to avoid further metabolic costs during periods of inactivity. However, few studies have addressed the effect of thermoregulation on habitat selection during periods of inactivity (Huey 1982; Christian et al. 1984; Huey et al. 1989; Kearney 2002), and fewer have examined male/female differences in retreat site selection.

I studied the effect of rock size in exposed, near-river habitats on overnight retreat selection by the common western fence lizard (Sceloporus occidentalis). Rocky riparian habitats consist of large fields of cobbles and boulders, in a variety of sizes, which are almost entirely absent from meadows and other upland habitats. Rocky substrates typically have larger thermal inertia and a greater capacity for heat storage than soil or other fine sediments and wood (Huey et al. 1989). I therefore hypothesized that: (1) "hot rocks" would provide warmer overnight retreats on cobble bars than in meadows, (2) gravid females would actively select cobble sizes providing the warmest retreats within the limits of their critical thermal maximum temperature (Tn), and (3) males would choose cooler overnight retreats than females to minimize metabolic expenditures during inactivity.

Study site

The study took place on one cobble bar and one meadow within the watershed of the SF Eel River (39?44'N, 123?39'W) in Mendocino County, California, at an average elevation of 300 m. Cobble bars are rocky habitats along the river margin, averaging 0.54 ha (range 0.28-1.04) in area. Mendocino County experiences a Mediterranean climate with up to 150 cm of rain during the wet season (October-April), and summer drought (May-September). Cobble bars are exposed portions of the winter active channel providing temporary habitat for a variety of invertebrates, reptiles, amphibians and small mammals during summer drought months. Sediment size distributions vary within and among cobble bars, ranging from sand to boulders (median diameter>1 m). However, most cobble bars with appreciable numbers of lizards are characterized by large numbers of intermediate sized cobbles ranging from 30-50 cm median diameter and between 10- 20 cm thick ("minor" axis).

Grassy meadows 2.3?1.8 ha in area (range 0.3-4.5) occur on upland river terraces at distances of 3-50 m from the river. These habitats lack exposed cobbles, but are instead covered by soil, grasses and forbs (Kotanen 1997). Meadows are surrounded by old-growth Douglas fir (Pseudotsuga menziesii), oak (Quercus and Lithocarpus spp.) and coastal redwood (Sequoia sempervirens) forest, whereas cobble bars are bordered by river on the downslope side and by forest or meadow on the upslope side.

In summer, western fence lizards, western skinks (Eumeces skilatonianus) and northern alligator lizards (Elgaria coeruleus) are common on both cobble bars and in meadows. Cobble bars support populations of the sagebrush lizard (S. graciosus) in addition to fence lizards. Sagebrush lizards are very rarely found in meadows. Fence lizards occur largely around the perimeter of both meadow and cobble bar habitats, although this pattern is less pronounced on cobble bars, especially those with interior vegetation (J. Sabo and A. Amacher, unpublished data). Surface temperatures fluctuate widely in both habitats during most sunny days (15-60'C). During peak temperatures, lizards retreat to shade, cast by the adjacent forest, trees and shrubs on the cobble bar, or under rocks.

Materials and methods

Location and characterization of retreat sites

I used radio telemetry to locate overnight retreat sites of western fence lizards on a single cobble bar and meadow every night between 8 June and 21 July 1999. This period encompasses most of the breeding and egg-laying season. I captured six males and six females from the cobble bar site, and five females from the meadow site. All lizards were transported to a holding facility and detained for 24 h in 50-1 plastic containers. Each lizard was measured (snout-vent length, nearest millimeter) and weighed (nearest milligram). Initial sizes (see Table 1) of females did not differ significantly between habitats in length (t=-0.23, df=9, P >0.8) or mass (t=-1.01, dft9, P>0.3). On cobble bars, females were significantly larger than males in terms of length (sign test P<0.05), but not mass (P>0.2). All five meadow females were gravid at the time of capture, whereas 3/6 cobble bar females were not. Three females on cobble bars appeared gaunt and had excess skin in the abdominal region, suggesting that these females were post-gravid.

Using a non-toxic, epoxy adhesive, I glued radio transmitters (1.3 g; Hohohil, Carp, ON) to all 17 animals. Transmitters were attached dorsally, above the pelvic region to minimize interference with locomotion. Following transmitter attachment, all lizards were released at their capture site. On the same night (8 June), and every night thereafter until 21 July, I located the positions of overnight retreats for each lizard using a handheld radio receiver (Wildlife Materials, Champaign, Ill.). All retreats were marked with labeled wire flags. Telemetry was carried out either at night (after 2300 hours), or in the early morning before sunrise (0500- 0700 hours) when surface temperatures were <20'C, well below the preferred activity range for fence lizards (34-360C). No changes in retreat site location were observed between consecutive night and morning passes.

I continued tracking lizards until all females laid eggs (see Table 1), at which time I recollected the lizards, removed their transmitters and returned them to flagged locations of capture. Over



331

Table 1 Mass and clutch dates of female lizards on cobble bar and meadow habitats

Codea Sex Date Initial snout-vent Initial mass Final mass Egg laying: Egg laying: No. of captured length (mm) (g) b (g) b June July clutches

Cobble bar 669 F 8 June 69 15.9 11.0 18 June - l 687 a F 18 June 72 17.6/18.3 14.2/13.3 25 June 17 July 2 230 a F 25 June 73 16.2 13.2 - 11 July 1 209 a F 12 July 72 16.8 11.25 - 1S July 1 789 F 8 June 69 12.1 (14.4)d 11.0 - 17 July 1 370 F 8 June 68 11.8 13.5 - - -

350c F 8 June 72 17.1 - - - - 330c F 8 June 72 16.8 - - - - 687C F 8 June 66 11.8 - - - - 392 M 8 June 62 10.8 - - - -

451 M 8 June 65 12.7 - - - - 471 M 8 June 63 10.8 - - - -

512 M 8 June 64 11.3 - - - -

706 M 8 June 62 11.3 - - - - 807 M 8 June 66 11.8 - - - -

Meadow 127 F 9 June 70 15.8 13.3 17 June - 1 150 F 9 June 71 15.7 10.1 25 June - 1 169 F 9 June 72 16.8 10.8 25 June - 1 209 F 9 June 63 12.1 8.0 25 June - 1 230C F 9 June 71 16.0 - - - -

the course of the study I observed four female mortalities, some of which could be attributed to predation by snakes (e.g., Sabo and Ku, in press). No males died over the course of the study. All transmitters from deceased individuals were attached to visibly gravid females such that a total of 15 females were followed over the 5- to 6-week period of the study.

I recorded the substrate type and size of all overnight retreats. Retreats were classified as cobble, wood or sand substrates, debris (leaves and grass), burrow, or open air (no retreat). For wood retreats, I recorded whether lizards were on the ground or in trees (under bark or in tree holes). I categorized frequencies of microhabitat use (four classes) by females in cobble bars and meadows and males in cobble bars and tested the null hypothesis that frequencies of habitat use were independent among classes of lizards using x2 analysis. For cobble retreats I measured all linear dimensions, including length (maximum diameter), width (or "median diameter") and thickness (minimum diameter) to the nearest millimeter. Finally, I characterized available retreats on cobble bars by recording the substrate type (sand, vegetation, debris, wood or cobble) and cobble thickness for retreats located at 0.5-m intervals along three 30-m transects across portions of the cobble bar heavily used by lizards at night. I used a lizard model, similar in size to a female S. occidentalis, to identify cobbles with adequate size or interstitial space to provide a lizard refuge. All rocks too small or lacking adequate interstitial space to cover this model entirely were rejected as available retreat sites. Available retreats were compared against retreats selected by males and females on the cobble bar in a x2 analysis. Analysis of the standardized residuals (Zar 1996) then allowed me to quantify contributions of individual cells to the overall variation, and thus, preference or avoidance of particular cobble sizes by each sex.

Characterizing thermal profiles of overnight retreats

To characterize standardized thermal profiles of potential retreat sites for lizards in cobble bar and meadow habitats I set up an experimental garden of substrates called "lizardhenge (Fig. 1) following the approach used by Huey et al. (1989) and Keamey (2002). This garden, located in a meadow within 100 m of the meadow on which telemetry took place, consisted of cobble, wood and sand substrates evenly spaced on a level, 100-m2 plot within a

Lizardhenge

OA E I B5 810 0

Fig. 1 Spatial layout of experimental garden of substrates ("lizardhenge") showing relative positions of cobbles (small circles), boulders (big circles), wood (horizontal rectangles), sand (S), burrows (B) and shaded bulb for I m open air (OA) temperature. Numbers are thickness (cobbles, boulders and wood) or depth (burrows) in centimeters

meadow, but cleared of all standing vegetation. This garden was sunlit for 11 h (0800-1900 hours) on most sampling days, mimicking natural sunlight in portions of most cobble bars and meadows heavily used by lizards. Lizardhenge was constructed on a flat meadow to ensure equal contact of all retreats with a uniform substratum and to minimize potentially confounding effects of heat storage by adjacent stones. Sandstone cobbles of four thicknesses (5, 10, 15 and 20 cm) were arranged in a randomized Latin square design ( n=7 each). In addition to these cobbles, three boulders (25, 30, and 40 cm thick), four half sections of tree ( P. menziesii) trunks (10 and 25 cm thick, n=2 each), one large stump (40 cm thick) and two containers of sand (30x40 cm and 10 cm deep) were placed along the perimeter of the smaller cobble garden. I measured 24-h thermal profiles under all of these substrates using copper-wire



332

thermocouples multiplexed to an outdoor weather logger (OWL; EME Systems, Berkeley, Calif.). One thermocouple was placed under the center of each stone or piece of wood. Finally, I also recorded profiles for air (1 m shaded bulb) and two nearby gopher (Thomomys bottae) burrows. Thermal profiles from lizardhenge provided predictions about the relative thermal characteristics of female retreat sites in meadows and cobble bars as well as predictions about the cobble size(s) maximizing heat storage and subsequent release time at night.

In addition, I used the OWL to record 24-h thermal profiles of actual retreats selected by lizards on meadow and cobble bar sites. Profiles were measured on a single day for each habitat (26 July and 1 August, cobble bar and meadow, respectively).

Results

Temperature profiles of retreats on lizardhenge

Average temperature profiles were warmer during the early evening (1900-2400 hours) for medium sized cobbles (15 cm thick), than for air, wood, burrows and sand (Fig. 2a). Only cobbles and sand gave temperatures within the preferred daily active range (32-36?C) during the period of inactivity (1900-0900 hours on most cobble bars and meadows). Of these two retreat types, cobble profiles remained above the lower threshold of this range (i.e., Tp ln = 32?C) over the longest period of time during hours of inactivity (Fig. 2a).

Thermal profiles of cobbles varied considerably with cobble thickness (Fig. 2b). Heat retention, defined as minimum night-time temperature, increased with substrate size. Single observations of larger boulders (30 and 40 cm thick) suggest further that the maximum daily temperature underneath cobbles, decreased with increas- ing rock thickness. Finally, temperatures are very near the upper Tmax (45?C) (Huey 1982) for S. occidentalis at the start of the inactive period (1900 hours) underneath cobbles < 15 cm thick. Temperature profiles for the center of 15- cm-thick cobbles cross this upper threshold almost precisely at 1900 hours, the onset of the period of inactivity (Fig. 2).

Overnight retreat-site selection

Microhabitat use differed between females in meadow and cobble bar habitats as well as between males and females on cobble bars (Fig. 3; X2=412.5, df-5, P<0.001). Cobble bar females used cobbles on >90% of nights whereas meadow females used wood or gopher (T. bottae) burrows on a majority of nights (72% and 15% wood and burrows, respectively). Male lizards at the cobble bar site used cobbles on a majority (68%) of nights, but also were found exposed in the open air, either under shrubs or among leaf litter and debris on 25% of nights. Standard- ized residuals were highest for cells corresponding to meadow females using wood, meadow females using burrows, cobble bar males in open air and cobble bar females under cobbles (12.0, 6.4, 5.6, 5.3, respectively) and lowest for meadow females under cobbles, and

a. Substrate types Sand .

60 . ~~~~~Cobble- - 60 Burrow

Air T - c

40 / -

a)~~~~~~~ r p

0

20

b _______InactivityA

X- b. Cobble thickness a) 60 .n5 20

110 -301- E 15 ......140..... ,r~~ L~~rnax

40 T

20

0 __________Inactivity

08 12 16 20 00 04 08

Time (h)

Fig. 2 Average thermal profiles of representative overnight retreats from lizardhenge for the undersides of various substrate types (a) and representative rock substrates of varying thickness (b). Substrate types in a are as follows: sand (Sand, n=2), 15-cm-thick cobbles (Cobble, n=5), a 10-cm burrow (Burrow, n=l), a 45-cm- thick log (Log, n=l), and OA temperature in shade (Air, n=l). Cobble thicknesses in b were 5 cm (5, n=5), 10 cm (10, n=5), 15 cm (15, n=5), 20 cm (20, n=5), 30 cm (30, n=1) and 40 cm (40, n=1). Vertical lines represent the period of inactivity. Dotted horizontal line indicates the critical thermal maximum (Tc ), for Sceloporus occidentalis (45'C) and shaded horizontal area indicates the preferred thermal range during activity (Tp) for S. occidentalis

-0 1.00

..) ' , ,ACobble bar 0.75Cobble bar T

C D 0.75 v X\Meadow__

CO .50

o 0.25 0

a. 0.00 Cobble Air Sand Burrow Wood

Retreat type

Fig. 3 Overnight microhabitat use by females on cobble bars (black), males on cobble bars (gray), and females on meadows (hatched). Bars represent average proportion of nights spent under a particular retreat substrate per individual. Error bars are 1 SEM

cobble bar males and females under wood (-7.7, -5.4, -4.8, respectively).

In addition to a higher use of cobbles by females than males at the cobble bar site, only females appeared to select cobble retreats by size (Fig. 4). The size distribu-



333

Do 0.75 a) =) C0.50 A Xvailable cobbles

?0.25 0 CL

0 .0 0 ----- -- ... --- --- 0 5 10 15 20 25 30 35 40

Cobble thickness (cm)

Fig. 4 Size selection of overnight cobble retreats by females (black) and males (gray) on cobble bars. Bars represent the proportion of nights spent under cobbles of a given thickness (cm) for all females summed over the entire 7-week observation period. The solid line represents the availability of cobbles of different thicknesses in three replicate transects across the same cobble bar

2 Cobble Cobble-

5Q 05 Sand-

O 5Air o 50 Meadow- Inactivity

40 ,

E 20

10 r__ __ _ r __r __r _

08 12 16 20 24 04 08

Time (h)

Fig. 5 Thermal characteristics of retreats chosen by lizards. Lines represent average profiles of female cobble retreats (y Cobble, n=12), male cobble retreats (, Cobble, n=1 1), male sand retreats on cobble bars (d Sand, n=2), male OA retreats on cobble bars (c Air, n=4), and female "tree" retreats on meadows (y Meadow, n=5). The shaded horizontal area indicates the Tp for S. occidentalis. All profiles are from a single 24-h period on similar sunny days (26 July, 2 August 1999, cobble bar and meadow, respectively). For abbreviations, see Figs. 1 and 2

1.2 A AB ABC BC C IQ 1.0

C 0.8

E 0.6 ~~~0.4 ~ ~ ~ ~ ~ EO.2

FC MC MS MA FM

Fig. 6 Time (h) within the Tp, during inactivity (1900-0800 hours) for females under cobbles (FC), males under cobbles (MC), males under sand (MS), males in open air on cobble bars (MA) and females in trees on meadows (FM). Sample sizes are as in Fig. 5. Groups with entirely different letters are significantly different (P<0.05; Fisher's LSD post-hoc test)

tion of cobble retreats differed significantly from the distribution of available cobbles for females (X2=40.7; df=7; P<0.001), but not males (x2=7.1; dft7; P>0.3). Examination of the standardized residuals revealed that the use of 15-cm cobbles by females resulted in the highest residual value (3.44) and contributed at least 33% of the residual variation. Residual values for females using 5-cm and 10-cm cobbles were the lowest in the analysis (-2.6, -1.9, respectively).

Thermal characteristics of retreats selected by lizards

Average temperatures during inactivity (1900-0800 hours) differed significantly among retreats used by lizards on meadow and cobble bar habitats (Fig. 5; F=6.0, dft4, P<0.001). Temperatures under cobbles used by males and females were warmer than those of open air retreats used by males on cobble bar habitats (P<0.001, Fisher's LSD post-hoc test). Further, temperatures under cobbles used by females on cobble bars were warmer than those in tree holes used by meadow females (Fisher's LSD, P<0.01).

Only cobble and sand retreats on cobble bars provided overnight temperatures within Tp (Fig. 6). Retreats selected by cobble bar females provided longer time intervals at temperatures within Tp than all retreats chosen by meadow females (F=3.1, d,f4, P<0.005; Fisher's LSD). Males spending the night in the open air and females in wood retreats on meadows experienced temperatures below the lower threshold of Tp for the entire period of inactivity (Figs. 5, 6). Cobble retreats (of all sizes) chosen by female lizards remained within Tp for a significantly longer duration than air retreats chosen by males (P<0.01, Fisher's LSD). However, though time within Tp for cobble retreats chosen by male and female lizards did not differ significantly (P=0.19), time within Tp for cobble and air retreats chosen by male lizards likewise did not differ significantly (P=0.13). Thus on cobble bars, male cobble retreats appear to be interme-

diate to female cobble and male air retreats with respect to time within Tp during inactive periods.

Discussion

Sediment size and substrate texture are critical determi- nants of the productivity and diversity of stream ecosystems (Hynes 1970; Minshall 1984; Dudley and D'Antonio 1991; Allan 1995). However, very little is known about how particle size distributions in exposed active channel habitats affect the performance, abundance or diversity of terrestrial organisms. Cobbles provide an important thermal resource for female western fence lizards living in exposed portions of active river channels. Retreat types used by females on cobble bars and meadows differ significantly and have significantly different thermal



334

properties at night (e.g., average temperature and time within Tp). Further, females but not males on cobble bars actively chose cobbles that offered one of the warmest microenvironments during inactivity.

Male and female retreat site selection on cobble bars

Retreat site selection may vary between males and females of the same species as a result of differences in energetic demands or behavior (Huey et al. 1989). Satiated (Huey 1982) or gravid (Wilimer 1982) individuals should select warmer retreat sites than hungry or non- reproductive animals to maximize net energy gain or gestation. In this study, retreats used by male and gravid female S. occidentalis differed in several respects. First, females on cobble bars chose cobbles as retreats nearly exclusively, even though other retreats were available (trees, sand, or no retreat). Females spent a higher proportion of nights under cobbles than males (93 vs. 68%), but males spent more nights (25%) in the open air than females (2%). Second, females more carefully selected cobble retreats based on substrate thickness. Females chose cobbles 15 cm thick significantly more frequently than expected by chance alone, while the sizes of cobbles used by males did not differ from those available on this particular cobble bar.

Cobbles provide the warmest overnight retreats during the early part of the inactive period (1900-2400 hours, Fig. 2a). Thermal profiles from lizardhenge suggest that cobble retreats (15 cm thick) provide temperatures above T1M,n (e.g., 32?C) for >4 h during inactivity. Open air retreat sites are never within this range over the same period. Thus, on at least 23% of the nights during the breeding season females are choosing warmer retreat substrates than males (cobble vs. air).

Cues for female retreat-site selection

Thermal profiles from lizardhenge suggest that cobbles 20-30 cm thick should provide retreats with temperatures higher than T'min for the longest interval during the period of inactivity. These results mirror those of Huey et al. (1989) for basalt rocks near Eagle Lake, California. At Eagle Lake, cobbles 20-30 cm thick provided the warmest overnight retreats for garter snakes (Thamnophis elegans), and these snakes showed a significant preference for cobbles in this size class. At the SF Eel River, large cobbles (>25 cm thick) were scarce (Fig. 4). Thus, the largest available retreats on cobble bars consisted of cobbles 20-25 cm thick. This size class of cobbles maintains temperatures >TP,n for a longer portion of the inactive period than cobbles in all other available (smaller) size classes. However, female western fence lizards chose cobbles slightly smaller (15 cm thick) than those offering the warmest overnight environment (20 cm). Garter snakes (T. couchii, T. elegans and T.

sirtalis) are very common on cobble bar habitats, and the latter of these three species are likely predators of Sceloporus. In addition to garter snakes, a number of other snake species likely to consume lizards are common on cobble bar habitats, including gopher snakes (Pituo- phis melanoleucus), juvenile western rattlesnakes (Cro- talus viridis) and racers (Coluber constrictor). Moreover, snake predation on female (but not male) fence lizards was observed during this study (Sabo and Ku, in press). This suggests the intriguing hypothesis that retreat site selection by lizards is limited by selection of larger rocks by potential predators.

Alternatively, garter snakes and fence lizards have very different activity patterns and metabolic demands, both of which may influence retreat site selection. S. occidentalis are diurnal lizards, active on most days. In contrast, garter snakes (T. elegans) may remain inactive over periods >24 h, which include both the maximum and minimum temperature of the retreat (Huey et al. 1989). Retreat selection by garter snakes, but not fence lizards, is limited by maximum daytime temperatures of retreats such that snakes may choose larger rocks than lizards to maintain lower Tb during the day. Moreover, many snake species consume prey which are large relative to their body mass (Cundall and Greene 2000). Large prey require a longer processing time and thus, longer periods within temperatures optimal for digestion (Td). By contrast, prey and typical gut mass are small relative to body mass for fence lizards (J. Sabo, unpublished data), and these animals consequently should require less time within Td (29-37?C for S. occidentalis; Harwood 1979) to complete the digestion of a day's forage. This suggests an alternate hypothesis for avoidance of the warmest cobble retreats by female lizards. Specifically, female fence lizards along the SF Eel River may choose retreats that offer the best of both worlds: those that complete digestion early in the evening and minimize further metabolic expenditures until the following activity period. Both would likely increase energy allocation to clutch mass and potentially increase clutch production. Our observations of gravid females tentatively support this hypothesis (Table 1); only females on cobble bars (with warmer retreats) produced more than one clutch during our observations. However, further study and larger sample sizes are necessary to more definitively evaluate the hypothetical link between night-time microhabitat selection, enhanced digestive efficiency and increased clutch production.

In conclusion, between-habitat differences in retreat site composition or size lead to significant differences in the thermal environment experienced by female westen fence lizards at night, and thus differences in the potential time available for digestion and metabolism of prey consumed during the day. Further, on cobble bar habitats female western fence lizards select the size of rocky retreat sites more carefully than males and this selection appears to be strongly tied to temperature.

Acknowledgements This study was funded by NSF grant DEB-FD 97-00834 to J. L. S. and Mary E. Power and two Graduate Research



335

Grants from the Department of Integrative of Biology, University of California Berkeley. I am extremely grateful for the help of M. Ku and A. Su for long hours behind the radio receiver. I thank the Museum of Vertebrate Zoology and M. Power for supplying the radio receiver and outdoor weather logger, respectively. G. Gilchrist, H. Greene, R. Huey, S. Kupferberg, W. Porter, D. Roberts, J. Rodriguez, V. Vredenburg and K. Zamudio helped with the conceptual development of this project. J. Finlay, R. Huey, S. Kuchta, D. Miles, M. Power, W. Sousa ,W. Getz and three anonymous reviewers provided comments that improved earlier drafts of this paper. Finally, I thank P. Steele and the California Natural Reserve System for providing and maintaining a protected research site at the Angelo Coast Range Preserve.

Refernces

Adam MD, Hayes JP (2000) Use of bridges as night roosts by bats in the Oregon Coast Range. J Mammal 81:402-407

Allan JD (1995) Stream ecology. Chapman and Hall, New York Casey TM (1981) Behavioral mechanisms of thermoregulation. In:

Heinrich B (ed) Insect thermoregulation. Wiley, New York, pp 79-1 14

Chapman DW (1988) Critical review of variable used to define effects of fines in redds of large salmonids. Trans Am Fish Soc 117:1-21

Christian KA, Tracy CR, Porter WP (1984) Physiological and ecological consequences of sleeping-site selection by the Galapagos land iguana (Colonophus pallidus). Ecology 65:752-758

Cowles RB, Bogert CM (1944) A preliminary study of the thermal requirements of desert reptiles. Bull Am Mus Nat Hist 83:256- 296

Cummins KW, Lauff GH (1969) The influence of substrate particle size on the microdistribution of the stream benthos. Hydrobio- logia 34:134-181

Cundall D, Greene HW (2000) In: Schwenk K (ed) Feeding: form, function and evolution on tetrapod vertebrates. Academic Press, San Diego, Calif., pp 293-333

Dudley TL, D'Antonio CM (1991) The effects of substrate texture, grazing, and disturbance on macroalgal establishment in streams. Ecology 72:297-309

Grant BW, Dunham AE (1988) Thermally imposed time constraints on the activity of the desert lizard Sceloporus meriami. Ecology 69:167-176

Harwood RH (1979) The effect of temperature on the digestive efficiency of three species of lizards, Cnemidophorus tigris, Gerhonotus multicarinatus, and Sceloporus occidentalis. Comp Biochem Physiol 63A:417-433

Huey RB (1982) Temperature, physiology, and ecology of reptiles. In: Gans C, Pough FH (eds) Biology of the Reptilia, vol 12. Academic Press, London, pp 25-91

Huey RB (1991) Physiological consequences of habitat selection. Am Nat 137:S91-S115

Huey RB, Peterson CR, Arnold SJ, Porter WP (1989) Hot rocks and not-so-hot rocks: retreat-site selection by garter snakes and its thermal consequences. Ecology 70: 931 - 944

Hynes HBN (1970) The ecology of running waters. University of Toronto Press, Toronto

Jones EBD, Helfman GS, Harper JO, Bolstad PV (1999) Effects of riparian forest removal on fish assemblages in southern Appalachian streams. Conserv Biol 13:1454-1465

Kearney M (2002) Hot rocks or much-too-hot rocks: seasonal patterns of retreat-site selection by a nocturnal ectotherm. J Therm Biol 27:205-218

Kotanen PM (1997) Effects of experimental soil disturbance on revegetation by natives and exotics in coastal Californian meadows. J Appl Ecol 34:631-644

Mackay RJ (1992) Colonization by lotic macroinvertebrates-a review of processes and patterns. Can J Fish Aquat Sci 49:617- 628

Maloney SK, Bronner GN, Buffenstein R (1999) Thermoregulation in the Angolan free-tailed bat Mops condylurus: a small mammal that uses hot roosts. Physiol Biochem Zool 72:385- 396

Marler CA, Moore MC (1988) Evolutionary costs of aggression revealed by testosterone manipulations in free-living male lizards. Behav Ecol Sociobiol 23:21-26

Marler CA, Moore MC (1989) Time and energy costs of aggression in testosterone-implanted free-living male mountain spiny lizards (Sceloporus jarrovi). Physiol Zool 62:1334-1350

Marler CA, Moore MC (1991) Supplementary feeding compensates for testosterone-induced costs of aggression in male mountain spiny lizards, Sceloporus jarrovii. Anim Behav 42:209-219

Marler CA, Walsberg G, White ML, Moore M (1995) Increased energy-expenditure due to increased territorial defense in male lizards after phenotypic manipulation. Behav Ecol Sociobiol 37:225-231

McAuliffe JR (1984) Competition for space, disturbance and the structure of a benthic stream community. Ecology 65:894-908

Minshall GW (1984) Aquatic insect-substratum relationships. In: Resh VH, Rosenberg DM (eds) The ecology of aquatic insects. Praeger, New York, pp 358-400

Porter WP, Tracy CR (1983) Biophysical analyses of energetics, time-space utitlization, and distribution limits. In: Huey RB, Pianka ER, Schoener TW (eds) Lizard ecology: studies of a model organism. Harvard University Press, Cambridge, Mass., pp 55-83

Porter WP, Mitchell JP, Beckman WA, De Witt CB (1973) Behavioral implications of a mechanistic ecology: thermal and behavioral modeling of desert ectotherms and their microen- vironment. Oecologia 13:1-54

Power ME (1992) Habitat heterogeneity and the functional- significance of fish in river food webs. Ecology 73:1675-1688

Sabo JL, Ku M (in press) Failed predation between a snake and Sceloporus occidentalis (western fence lizard). Herpetol Rev

Strayer DL, Ralley J (1993) Microhabitat use by an assemblage of stream-dwelling Unionaceans (Bivalvia), including 2 rare species of Alasmidonta. J N Am Benthol Soc 12:247-258

Webb JK, Shine R (1998) Using thermal physiology to predict retreat-site selection by an endangered snake species. Biol Conserv 86:233-242

Webb JK, Shine R (2000) Paving the way for habitat restoration: can artificial rocks restore degraded habitats of endangered reptiles? Biol Conserv 92:93-99

Wilimer PG (1982) Microclimate and the environmental physiology of insects. In: Berridge MJ, Treherne JE, Wigglesworth VB (eds) Advances in insect physiology. Academic Press, New York, pp 1-57

Zar JH (1996) Biostatistical analysis. Prentice Hall, Englewood Cliffs, N.J.



international association for ecologyangelo.berkeley.edu/wp-content/uploads/hot-rocks... ·...

Documents