year-round patterns in bacterial production ... -...

Inland Waters (2013) 3, pp. 235-252

© International Society of Limnology 2013

DOI: 10.5268/IW-3.2.536

235

Article

Year-round patterns in bacterial production and biomass in Lake Simcoe, Ontario, Canada: are heterotrophic bacteria a significant contributor to low hypolimnetic oxygen?C.J. Quinn1*, R.L. North2, and P.J. Dillon3 1 Trent University, Graduate Studies: Environmental and Life Studies, 1600 West Bank Drive, Peterborough, K9J 7B8, ON, Canada2 University of Saskatchewan, 112 Science Place, Saskatoon, S7N 5E2, SK, Canada3 Trent University, Department of Chemistry, 1600 West Bank Drive, Peterborough, K9J 7B8, ON, Canada* Corresponding author email: [email protected]

Received 6 July 2012; accepted 29 March 2013; published 30 April 2013

Abstract

Bacteria serve an important function in aquatic environments and are associated with nutrient regeneration, carbon (C) metabolism, and secondary production. This is the first study to measure heterotrophic bacterial production, abundance, biomass, and biovolume in Lake Simcoe, Ontario. Excessive phosphorus loading resulting in low hypolimnetic oxygen concentrations has impacted the cold-water fishery in Lake Simcoe. We tested the hypothesis that bacteria contributed to dissolved oxygen declines in Lake Simcoe and examined the environmental factors impacting bacterial activity. Spatial and temporal variations in the microbial community were measured from June 2010 to July 2011. A dual-iso-tope method (3H-TdR and 14C-leu) was used, resulting in mean (± standard deviation) annual epilimnetic bacterial production estimates of 0.130 ± 0.173 and 0.268 ± 0.304 µg C L−1 h−1, respectively. The mean annual bacterial abundance was 1.49 ± 1.53 cells × 109 L−1, with a mean biomass of 1.21 ± 1.34 µg C L−1 and a mean biovolume of 0.0043 ± 0.0030 µm3 cell−1. These estimates had distinct seasonal patterns, with consistently lower bacterial activity in the winter relative to the spring, summer, and fall. Differences between epilimnetic and hypolimnetic bacterial activity were inconsistent for the 4 bacterial parameters measured. Lake temperature, chlorophyll a, and dissolved organic C concentrations were the most significant factors influencing the annual epilimnetic patterns in the examined bacterial parameters. Annual bacterial production was low in Lake Simcoe and does not seem to be a major contributor to the low hypolimnetic oxygen concentrations in the lake.

Key words: 3H-thymidine, 14C-leucine, bacterial activity, dual isotope, eutrophication, hypolimnetic oxygen, microbial, phosphorus

Introduction

Bacterial production, which is the rate of synthesis of biomass by heterotrophic bacterioplankton (Fuhrman and Azam 1980, 1982), is an important parameter that provides information on food web dynamics in lakes. For example, the bacterial biomass is potentially available to grazers (Chin-Leo and Kirchman 1988), thus contributing to carbon (C) flow through the aquatic food web (Smith and Azam 1992). Degradation of organic matter by

bacteria is an oxygen demanding process, however, and high bacterial activity can contribute significantly to low hypolimnetic oxygen conditions (Goosen et al. 1995). Bacterial production rates have also been used as an indicator of bacterial response to changing environmental conditions (Chin-Leo and Kirchman 1988, Cole and Pace 1995). Many factors can limit bacterial production and abundance in lakes, such as nutrient availability (e.g., organic C), temperature, viral lysis (Weinbauer and Peduzzi 1995), and predation and grazing by zooplankton

236

DOI: 10.5268/IW-3.2.536

CJ Quinn et al.


(Gardner et al. 1986, Pace and Cole 1996) and dreissenid mussels (Frischer et al. 2000). Competition for resources between bacteria and phytoplankton can also limit bacterial production and abundance (Caron 1994, Kirchman 1994).

In oligotrophic lakes, bacteria are the major competitor for dissolved nutrients such as phosphorus (P) and nitrogen (N; Cotner and Wetzel 1992). Although bacteria are superior competitors for P, they also depend on dissolved organic carbon (DOC) from primary production (autoch-thonous DOC; Olsen et al. 2002) and terrestrial sources (allochthonous DOC; Smith and Prairie 2004). Chrost et al. (2000) found that autochthonous C was much more mi-crobiologically labile than allochthonous C. Labile DOC released by phytoplankton (extracellular organic carbon [EOC]; Meyer et al. 1987, Moran and Hodson 1990, Chrost et al. 2000), and aquatic macrophytes (Murray and Hodson 1985), and allochthonous DOC released from sources such as agriculture and sewage are the primary energy sources for bacterial production in lakes (Meyer et al. 1987, Moran and Hodson 1990). Dependence on DOC by bacteria may explain why bacteria and phytoplankton can coexist (Currie and Kalff 1984). Bacteria differ from most other heterotrophic organisms in that they have the ability to assimilate dissolved organic matter (DOM; Cole and Pace 1995), which can be utilized by bacterivores and grazers (Cole and Pace 1995, Chrost et al. 2000), and thus are a significant link to higher consumers (Ducklow et al. 1986).

Lake Simcoe is under the threat of eutrophication due to high total P (TP) loading from the watershed, which has resulted in the excess growth of phytoplankton and submerged aquatic vegetation (Young et al. 2011). Decom-position of this plant and algal material by bacteria can contribute to declines in hypolimnetic oxygen (O2) concen-trations (Goosen et al. 1995), but their relative contribution to the process is poorly understood. The saturation concen-tration of dissolved O2 in a lake is inversely related to water temperature; however, photosynthesis and respiration can significantly alter lake O2 levels (Horne and Goldman 1994). In Lake Simcoe, end-of-summer hypolim-netic dissolved O2 concentrations have increased from ap-proximately 2 mg L−1 in the 1980s to 5 mg L−1 in 2008 (Young et al. 2011); however, these values are still well below the target concentration of 7 mg L−1 required for natural cold-water fish reproduction (Young et al. 2011). Models developed for Lake Simcoe (Young et al. 2011) suggest that chlorophyll a (Chl-a) concentrations do not solely account for the low hypolimnetic O2 concentrations.

Although there have been numerous bacterial studies in aquatic systems (e.g., Jonas et al. 1988, Kirchman 2001, Sturluson et al. 2008), few have been conducted over a sufficient length of time to allow seasonal variation to be

clearly identified. In general, bacterial studies have focused on the spring, coinciding with the spring phyto-plankton bloom, and the summer, when primary and secondary production rates are generally maximal (Hwang and Heath 1997, DeBruyn et al. 2004, Sturluson et al. 2008). Here, we provide the first assessment of hetero-trophic bacterial activity in Lake Simcoe and investigate whether bacteria contributed to hypolimnetic O2 declines. This study also enhances our understanding of the importance of bacteria in the aquatic food web of temperate lakes on a year-round basis. We performed spatially extensive sampling over an entire year to assess seasonal and spatial variation in bacteria and to identify the environmental factors that impact bacterial activity.

Study site

Lake Simcoe is the largest inland lake in Southern Ontario, excluding the Laurentian Great Lakes. Located at 44º25ʹN; 79º20ʹW, it has a volume of 1.03 × 1010 m3, a surface area of 722 km2, and a mean depth of 14.2 m (Nicholls 1995). With average open-water TP concentra-tions of about 14 µg L−1 in 2010 (North et al. 2013), Lake Simcoe is considered oligo-mesotrophic. Its major inflows from the south and east drain heavily urbanized and agri-cultural areas, and it outflows into Lake Couchiching to the north (Young et al. 2010). The lake freezes over completely most winters from late December to early April (Hawryshyn et al. 2012). It is a dimictic lake that is typically stratified from June to October; however, the length of stratification has increased since the 1980s, possibly due to climate change (Stainsby et al. 2011).

Methods

Field sampling

Monthly sampling occurred from June to December 2010 and February to July 2011 at 10 lake stations (Fig. 1). To have a continuous year-round dataset, we collected additional biweekly to monthly samples between January and June 2011 from the Beaverton Water Treatment Plant (WTP) unchlorinated intake pipe (Fig. 1, Table 1). The intake of the pipe was located at a 7.6 m deep site approxi-mately 986 m from the shoreline, while the pipe itself was suspended 3.05 m above the lake bottom. Sampling seasons were defined as winter (under ice: Jan–Mar), spring (Apr–May), summer (Jun–Oct), and fall (Nov–Dec). In this year-round study, thermal stratification was observed during the summer at stations >20 m deep. The entire water column was mixed from top to bottom the rest of the year with the exception of slight winter stratifi-cation under ice.

DOI: 10.5268/IW-3.2.536

237Year-round patterns in bacterial production and biomass in Lake Simcoe


Monthly lake water samples were collected during 2-day cruises from approximately 07:00 to 19:00 h; however, the order the stations were sampled and the approximate time of water collection varied month to month, depending on weather conditions. Epilimnetic integrated water samples were collected from the surface to 1 m above the thermocline (defined by a change in water temperature gradient >1 °C m−1; Wetzel 2001), and hypolimnetic samples were collected from a discrete depth 2 m above the lake bottom. Water samples were collected in acid-washed 20 L carboys that were protected from exposure to direct sunlight and temperature fluctua-tions until they were processed the same evening. Under-ice sampling was conducted in a similar manner at

the same 10 lake stations, although ice conditions dictated the frequency each station was sampled throughout the winter. Winter water samples were collected with a Van Dorn sampler at discrete depths ranging from 0 to 7 m through a 30 cm sampling hole. For sampling at the Beaverton WTP intake pipe, 20 L of raw, unchlorinated water was collected directly from the intake pipe between 09:00 and 11:00 h.

Water chemistry

Depth-profiled temperature and O2 concentrations were recorded at each lake station throughout the year using a YSI meter (model 6600 V2; Yellow Springs Instruments, Yellow Springs, Ohio, USA) with an ROX optical O2 probe that was calibrated daily. Chl-a concentrations were determined using lake water prefiltered through 200 μm nylon (Nitex) filters. Size-fractionated subsamples were filtered through a series of polycarbonate membranes (20 and 2 μm). In each step, the filtrate was filtered onto glass fiber filters (GFF, nominal pore-size 0.7 μm) and stored in the dark at −20 °C. Each filter was passively extracted with 90% acetone in the freezer. A fluorometer (Turner Designs 10-AU; Turner Designs, Sunnyvale, California, USA) that was calibrated yearly with pure Chl-a was used to determine the pheophytin-corrected Chl-a concentra-tions (Smith et al. 1999).

Standard colourimetric methods were used to analyze whole water (prescreened through a 200 μm nylon filter) for TP and filtered (0.2 μm) total dissolved P (TDP; OMOE 2007). Soluble reactive P (SRP) samples were filtered (0.2 μm) and analyzed according to Stainton et al. (1977) with a detection limit of 0.372 μg L−1 P. Samples were also filtered (0.45 μm) and analyzed for DOC using a Shimadzu TOC-VCPH analyzer (Chiyoda-ku, Tokyo, Japan).

Fig. 1. Lake Simcoe with locations of the 10 lake sampling stations and the Beaverton Water Treatment Plant (WTP) intake pipe.

Station Name Station Code Depth (m) Zmix (m)*Cooks Bay 6 C6 12.1 7Cooks Bay 9 C9 18.6 8East Simcoe 50 E50 10 10.1East Simcoe 51 E51 12.6 10.9Kempenfelt Bay 42 K42 39 16Kempenfelt Bay 45 K45 32.5 4Main Basin 66 M66 31 11North Simcoe 32 N32 6 6South Simcoe 15 S15 20 11T2 T2 8 6Beaverton Water Treatment Plant WTP 4.6*Zmix values were based on temperature profiles from July and August 2010.

Table 1. Station name and code as well as water depth and mixing depth of the 10 lake stations and the Beaverton WTP.

238

DOI: 10.5268/IW-3.2.536

CJ Quinn et al.


Bacterial production

Bacterial production was measured using simultaneous incorporation of methyl-3H thymidine and 14C-leucine (Chin-Leo and Kirchman 1988). Water samples (5 mL) were dispensed into ethanol-rinsed glass scintillation vials along with 20 nM of methyl-3H thymidine (80 Curies mmol−1, PerkinElmer) and 50 nM of 14C-leucine (300 milliCuries mmol−1, PerkinElmer).

Saturation experiments were performed each year to determine the saturating concentrations of leucine and thymidine in the same manner as described in the paragraph below, with the exception of the amount of each isotope added into each vial. To validate the choice of a saturating concentration of leucine and thymidine, final concentrations of 0, 20, 40, 50, 80, 150, 200, and 250 nM; and 0, 10, 20, 40, 80, 100, and 120 nM, respectively, were added to individual vials that contained 5 mL prefiltered (200 μm Nitex) lake water collected from the epilimnion of stations E50 and K42 (Fig. 1) on 24 July 2010 and 14 July 2011, respectively. No thymidine was added to leucine saturating experiment vials, and no leucine was added to vials for thymidine saturating experiments. Final concentrations of 50 nM 14C-leucine and 20 nM of 3H-thymidine were found to be saturating and were used for all subsequent assays.

Fresh (<24 h) working solutions of thymidine and leucine were sterilized by filtration (0.2 μm) before addition to subsamples within an hour; 5 mL of 10% (weight/weight) trichloroacetic acid (TCA, Sigma; St. Louis, Missouri, USA), was added to one of the 4 subsamples for immediate termination after both isotopes were added. Terminated “time-zero” subsamples served as blanks to control for abiotic adsorption of the working solution. Time-zero subsamples were carried through the incubation procedure in parallel with the live subsamples. Samples from each station contained one time-zero subsample and 3 live subsamples. All subsamples were incubated in the dark for 1 h at in situ collection temperature, then terminated by placing them on ice for 5 min followed by the addition of 5 mL of ice-cold 10% (wt/wt) TCA and left to sit for 5 min before filtering. Subsamples were filtered through a cellulose nitrate membrane filter (25 mm diameter, 0.2 μm, VWR; Radnor, Pennsylvania, USA) followed by a 2-fold rinse with 3 mL of ice-cold 10% TCA and 10 mL of prefiltered lake water from the respective stations, followed by 3 mL of 95% ethanol. Cold TCA-insoluble material was retained on the filters and was measured with a Packard Tri-Carb 1600CA liquid scintillation counter (Perkin Elmer; Waltham, Mas-sachusetts, USA). Thymidine incorporation was converted to cell production (cells L−1 h−1) by the calibration factors 2.0 × 109 cells nmol−1, 3H incorporated, and 0.04 μm3 cell−1

and 1 × 106 conversion to C production (Bell 1993). Leucine incorporation was converted to protein production using the fractions 100/7.3 mole percent of leucine in protein, 131.2 g molecular weight of leucine, and 0.86 g of C per protein (Kirchman 2001). On a few occasions, bacterial production rates were measured to be below zero; these were reported as zero values.

Bacterial abundance, biomass, and biovolume

At each station, a 20 mL subsample of prescreened lake water (200 μm) was collected for bacterial abundance and biomass determination. Subsamples were preserved in glass scintillation vials with 5% paraformaldehyde solution (Sigma 95% powder paraformaldehyde) and kept at 4 °C until subsamples could be processed (within 6 months). Bacterial abundance was determined using an epifluorescence microscope (Zeiss Axioscope; Thornwood, New York, USA). Aliquots (1 mL) were stained with DNA-DAPI (4’6’-diamidino-2-phenylindole) then filtered onto 0.2 μm pore-size black polycarbonate membrane filter backed with a 0.45 μm white filter. For each measurement of bacterial abundance, at least 200 cells or 40 fields of view, whichever came first, were counted. Northern Eclipse Imaging System (Mississauga, Ontario, Canada) was used to measure at least 200 cells per sample to determine the biomass of the sample. Using the measurements from Northern Eclipse, biovolume (μm3 cell−1) estimates were calculated using Bratbak’s (1993) formula. Total bacterial biomass (mg C L−1) was calculated by multiplying biovolume with a carbon conversion factor of 200 fg C μm−3 (Bratbak 1993) followed by the multipli-cation with bacterial abundance.

Statistical analyses

SYSTAT 13 (Chicago, Illinois, USA) was used for all data analysis. Transformations (log10 for all parameters) were applied to ensure that the data met the assumptions of normality. One-way analysis of variance (ANOVA) tests were performed followed by Tukey-Kramer post hoc tests with a significance value of p < 0.05 to determine the differences among seasons and to examine spatial differences for both epilimnetic and hypolimnetic values separately for all bacterial parameters measured. Pearson correlations were performed using log10 transformed means, calculated by averaging the 3 replicates for each bacterial parameter within each site, to examine the rela-tionships among bacterial parameters and with associated water quality data. Pearson correlation coefficients (r) and Bonferroni probabilities (significance value of p < 0.05) are reported. Based on an ANOVA test, samples from the WTP were not significantly different from epilimnetic

DOI: 10.5268/IW-3.2.536



samples from nearby stations E50 and E51 (TdR incorpo-ration F[2,33] = 1.692, p = 0.20 and leucine incorporation F[2, 48] = 0.517, p = 0.60; Fig. 1); therefore, the WTP was treated as an 11th epilimnetic lake station in all analyses.

Results

Spatial distribution of bacteria in Lake Simcoe

The epilimnetic data were used to examine the spatial dis-tribution of bacteria across the lake. There were few differences in annual bacterial production, abundance, biomass, and biovolume among the 11 stations sampled. Annually, the nearshore station T2 had significantly higher bacterial production (for both leucine and thymidine incorporated) and abundance than the other 10 stations (Tukey-Kramer test p < 0.05). Overall, there were no significant differences among stations for annual bacterial biomass or biovolume (Tukey-Kramer test p > 0.05) with the exception of station T2, where biomass was 4.1 times higher than at the WTP (p < 0.05). Due to lack of spatial differences across the 11 stations, seasonal analyses were performed by combining all stations.

Seasonality of bacterial production in Lake Simcoe

The seasonal patterns in bacterial production were measured from the incorporation rates of 3H-thymidine and 14C-leucine. The mean annual epilimnetic bacterial production rates for 3H-TdR and 14C-leucine incorporation were 0.130 ± 0.173 (n = 241) and 0.268 ± 0.304 mg C L−1 h−1 (n = 292), respectively (Table 2). Estimates from leucine incorporation for annual epilimnetic samples were significantly higher than the thymidine estimates (F[1, 501] = 21.962, p < 0.0005); however, both methods illustrated similar temporal trends (Fig. 2a and b; Table 3). Significant differences were also observed in bacterial production measured from the incorporation rates of 3H-thymidine compared with 14C-leucine incorporation in the winter (F[1, 93] = 6.567, p = 0.012), summer (F[1, 225] = 48.549, p < 0.0005), and fall (F[1, 88]= 13.186, p < 0.0005). No significant difference was observed between the 2 methods in the spring (F[1, 109] = 0.874, p = 0.352).

In this year-round study, thermal stratification was observed at deep stations (>20 m) during the summer, and slight inverse stratification was observed under ice in the winter. During stratification, the lakewide epilimnetic bacterial production rates were significantly higher than the hypolimnetic rates (TdR: incorporation, F[1, 154] = 38.146, p < 0.0005; leucine incorporation, F[1, 205] = 38.312, p < 0.0005; Fig. 2a,b). Bacterial production exhibited a distinct temporal pattern; maximum production values

were reached during the stratified summer (Jun: 0.486 ± 0.242 and Jul: 0.614 ± 0.441 mg C L−1 h−1) with the leucine method, while the lowest rates were measured during the winter, under ice (0.035 ± 0.042 mg C L−1 h−1; Fig. 2b; Table 2 and 3). With the thymidine method, maximum production was observed in the isothermal spring (May: 0.441 ± 0.256 mg C L−1 h−1; Fig. 2a; Table 2 and 3), likely because June samples were not measured. After peaking in June (May for TdR), production declined and remained low until the following April, with the exception of a second smaller peak in March (0.049 ± 0.053 mg C L−1 h−1), just before ice break-up (Fig. 2a and b).

All annual and seasonal correlation analyses were conducted using the leucine bacterial data only due to its larger sample size. When bacterial parameters were summarized as annual means, there were no significant correlations between bacterial production and the other bacterial activity parameters (Table 4); however, bacterial production was positively correlated with biomass in the fall (r = 0.537, p = 0.039). No significant correlations were observed in winter, spring, or summer (Table 4).

Annual, epilimnetic bacterial production was positively correlated with surface water temperature (r = 0.517, p = <0.0005) and DOC concentration (r = 0.207, p = 0.049), but negatively correlated with Chl-a concentra-tions (20–200 and 2–20 μm; r = −0.238, p = 0.022, and r = −0.202, p = 0.050, respectively) and TDP concentra-tions (r = −0.278, p = 0.008; Table 5). In the stratified summer, bacterial production was negatively correlated with Chl-a concentrations (20–200 μm; r = −0.304, p = 0.028; Table 5). There were no significant correlations of bacterial production with any of the water quality or en-vironmental parameters in the winter (Table 5).

At the beginning of summer, water temperatures reached approximately 17 ºC, and the relationship between bacterial production and temperature started to uncouple and was not observed again until the end of summer when temperatures began to decline. Further analysis of these relationships showed that log10-trans-formed bacterial production rates were correlated with water temperature, but only at temperatures below 17 °C (Fig. 3a). During the period of time when water temperature was above 17 ºC, however, log10-transformed bacterial production rates showed a strong relationship with TDP concentrations (Fig. 3b).

Seasonality of bacterial abundance, biomass, and biovolume

At the annual scale, there were no significant differences in bacterial abundance (F[1, 133] = 0.094, p = 0.760; Fig. 2c) or biomass (F[1, 133] = 0.274, p = 0.602; Fig. 2d) between the epilimnion and the hypolimnion; however, hypolim-

240

DOI: 10.5268/IW-3.2.536

CJ Quinn et al.


Uni

tsEp

ilim

netic

Hyp

olim

netic

Ann

ual

Win

ter

Spri

ngSu

mm

erFa

llW

inte

rSu

mm

er

BP

(TdR

)μg

C L

−1 h

−10.

130

± 0.

173

(2

41)

0.01

5 ±

0.03

0

(48)

0.36

2 ±

0.25

9

(45)

0.11

6 ±

0.09

0

(103

)0.

051

± 0.

028

(4

5)0.

004

± 0.

008

(2

1)0.

034

± 0.

050

(5

3)

BP

(Leu

)μg

C L

−1 h

−10.

268

± 0.

304

(2

92)

0.03

5 ±

0.04

2

(48)

0.36

8 ±

0.22

0

(45)

0.36

5 ±

0.34

7

(154

)0.

081

± 0.

048

(4

5)0.

031

± 0.

029

(2

1)0.

067

± 0.

083

(5

3)

BA

cells

L−1

1.49

× 1

09 ±

1.53

× 1

09 (1

05)

3.89

× 1

08 ±

2.44

× 1

08 (1

6)

1.57

× 1

09 ±

1.33

× 1

09 (1

5)

1.56

× 1

09 ±1

.31

× 10

9 (5

9)

2.29

× 1

09 ±

2.54

× 1

09 (1

5)

3.71

× 1

08 ±

1.42

× 1

08 (7

)

1.54

× 1

09 ±

9.68

× 1

08 (2

3)

BV

μm3 c

ell−1

0.00

43 ±

0.0

030

(105

)0.

0032

± 0

.002

2 (1

6)0.

0033

± 0

.000

7 (1

5)0.

0050

± 0

.003

7 (5

9)0.

0037

± 0

.000

9 (1

5)0.

0039

±0.

0023

(7

)0.

0052

± 0

.002

5 (2

3)

BB

μg C

L−1

1.21

± 1

.34

(1

05)

0.23

± 0

.15

(1

6)0.

93 ±

0.6

9

(15)

1.41

± 1

.25

(5

9)1.

72 ±

2.1

5

(15)

0.28

± 0

.17

(7

)1.

49 ±

1.0

1

(23)

Tem

pera

ture

ºC13

.9 ±

8.4

(1

06)

0.7

± 0.

9

(16)

7.2

± 2.

4

(15)

20.6

± 2

.9

(60)

7.9

± 1.

8

(15)

1.8

± 1.

1

(7)

10.6

± 2

.1

(18)

O2

mg

L−19.

6 ±

2.7

(9

7)10

.4 ±

3.7

(1

2)12

.6 ±

0.6

(1

2)8.

4 ±

2.3

(5

8)11

.2 ±

0.7

(1

5)8.

9 ±

4.2

(7

)7.

6 ±

2.8

(2

2)C

hl-a

(20–

200

μm)

μg L

−11.

9 ±

2.1

(1

06)

3.0

± 4.

6

(16)

1.6

± 1.

5

(15)

1.5

± 1.

0

(60)

2.6

± 1.

4

(15)

NA

0.6

± 0.

5

(24)

Chl

-a(2

–20

μm)

μg L

−11.

4 ±

1.7

(1

06)

2.5

± 3.

8

(16)

1.3

± 1.

3

(15)

1.1

± 0.

8

(60)

1.4

± 0.

7

(15)

NA

0.4

± 0.

4

(14)

Chl

-a(0

.7–2

μm

)μg

L−1

0.6

± 0.

4

(105

)0.

3 ±

0.2

(1

6)0.

4 ±

0.2

(1

5)0.

7 ±

0.4

(5

9)0.

9 ±

0.4

(1

5)N

A0.

2 ±

0.2

(1

4)

DO

CμM

339.

5 ±

21.0

(1

05)

325.

4 ±

9.5

(1

6)33

2.0

± 13

.6

(15)

343.

5 ±

24.4

(5

9)34

6.1

± 11

.3

(15)

NA

314.

3 ±1

5.6

(2

4)

TPμM

0.27

± 0

.06

(1

06)

0.27

± 0

.03

(1

6)0.

27 ±

0.0

7 (1

5)0.

27 ±

0.0

7 (6

0)0.

30 ±

0.0

7 (1

5)N

A0.

30 ±

0.1

3 (2

4)

TDP

μM0.

15 ±

0.0

5

(104

)0.

19 ±

0.0

4

(16)

0.14

± 0

.02

(14)

0.13

± 0

.04

(59)

0.18

± 0

.04

(15)

NA

0.20

± 0

.13

(24)

SRP

μM0.

06 ±

0.0

4

(104

)0.

07 ±

0.0

3

(16)

0.06

± 0

.03

(15)

0.05

± 0

.04

(58)

0.08

± 0

.05

(15)

NA

0.09

± 0

.11

(24)

Tab

le 2

. Mea

n (±

SD; w

ith n

val

ue in

par

enth

eses

) mea

sure

men

ts o

f bac

teria

l act

ivity

(bac

teria

l pro

duct

ion

[BP]

, bac

teria

l abu

ndan

ce [B

A],

bact

eria

l bio

volu

me

[BV

], an

d ba

cter

ial b

iom

ass

[BB

]) a

nd e

nviro

nmen

tal f

acto

rs (

tem

pera

ture

, O2,

Chl

-a, p

hosp

horu

s, an

d D

OC

con

cent

ratio

ns)

for

11 s

ampl

ing

stat

ions

in L

ake

Sim

coe

from

Jun

e 20

10 to

Jul

y 20

11; h

ypol

imne

tic

sam

ples

wer

e co

llect

ed o

nly

whe

n st

ratifi

catio

n w

as p

rese

nt.

DOI: 10.5268/IW-3.2.536



netic bacterial biovolume was significantly higher than epilimnetic biovolume (F[1, 133] = 3.391, p = 0.049; Fig. 2e).

Annual epilimnetic mean abundance and biomass were 1.49 × 109 ± 1.53 × 109 (n = 105; Table 2) and 1.10 ± 1.25 μg C L−1 (n = 105; Table 2), respectively. Distinct temporal patterns were evident in bacterial biomass, with biomass increasing during the spring to peak in June (2.02 ± 1.66 µg C L−1) and again in November (2.34 ± 2.42 µg C L−1), followed by a decrease in December (0.459 ± 0.307 µg C

L−1; Fig. 2d). Bacterial abundance peaked in November (3.12 × 109 ± 2.77 × 109 cells L−1) and then decreased in December (6.43 × 108 ± 4.30 × 108 cells L−1; Fig. 2c). There were no significant differences among spring, summer, and fall for abundance and biomass (Table 3); however, bacterial abundance and biomass were signifi-cantly lower in the winter relative to the other 3 seasons (Table 3). Epilimnetic annual mean bacterial biovolume was 0.0043 ± 0.0032 μm3 cell−1 (n = 105; Fig. 2e; Table 2).

Parameter1 Seasonal ANOVA results Post-hocTukey-Kramer test2

BP TdR Epi F[3, 209]= 92.10, p< 0.0005 WaSpbSucFd

BP TdR Hypo F[1, 59] = 22.93, p< 0.0005 WaSub

BP Leu Epi F[3, 273] = 76.77, p< 0.0005 WaSpbSubFc

BP Leu Hypo F[1, 65] = 1.38, p = 0.245 WaSua

BA Epi F[3, 101] = 6.87, p< 0.0005 WaSpbSubFb

BA Hypo F[1, 28] = 7.17, p = 0.012 WaSub

BV Epi F[3, 101] = 2.92, p = 0.038 WaSpaSuabFb

BV Hypo F[1, 28] = 3.48, p = 0.073 WaSub

BB Epi F[3, 101] = 9.19, p< 0.0005 WaSpbSubFb

BB Hypo F[1, 28] = 10.10, p = 0.004 WaSub

1 TdR=Thymidine, Epi=Epilimnion, Hypo=Hypolimnion, Leu=Leucine2 W: Winter; Sp: Spring; Su: Summer; F: Fall

Table 3. Summary statistics for one-way ANOVA seasonal comparisons of bacterial production (BP), abundance (BA), biovolume (BV), and biomass (BB) during the period of June 2010 to July 2011 in Lake Simcoe. Dissimilar superscript letters denote significant differences (p < 0.05) among seasons; hypolimnetic samples were collected only when stratification was present.

Epilimnetic BP Hypolimnetic BPAnnual Winter Spring Summer Fall Winter Summer

BA 0.146 (97) 0.095 (16) −0.318 (15) −0.083 (51) 0.480 * (15) −0.331 (7) −0.276 (17)BV −0.103 (97) −0.238 (16) −0.007 (15) −0.255 * (51) 0.246 (15) 0.226 (7) 0.134 (17)BB 0.109 (97) −0.104 (16) −0.353 (15) −0.167 (51) 0.537 * (15) −0.054 (7) −0.219 ( 17)

Epilimnetic BA Hypolimnetic BAAnnual Winter Spring Summer Fall Winter Summer

BV 0.019 (105) −0.157 (16) −0.664** (15) −0.014 (59) −0.131 (15) −0.046 (7) −0.001 (23)BB 0.907 *** (105) 0.774 *** (16) 0.986 *** (15) 0.865 *** (59) 0.975 *** (15) 0.651 (7) 0.948 *** (23)

Epilimnetic BV Hypolimnetic BVAnnual Winter Spring Summer Fall Winter Summer

BB 0.438 *** (105) 0.503 * (16) −0.532 * (15) 0.490 *** (59) 0.092 (15) 0.729 * (7) 0.318 ( 23)Note: Bacterial biomass (BB) is calculated from bacterial abundance (BA) and biovolume (BV) thus BB is autocorre-lated with both BA and BV.

Table 4. Pearson correlation coefficients (with n values in parentheses) for the correlations between bacterial production (BP), bacterial abundance (BA), bacterial biovolume (BV), and bacterial biomass (BB) for 11 stations sampled in Lake Simcoe for the period of June 2010 to July 2011; correlation p-values are significant at *0.05, **0.01, and ***0.001.

242

DOI: 10.5268/IW-3.2.536

CJ Quinn et al.


Fig. 2. Temporal variation in bacterial production in the epilimnion (solid line) and hypolimnion (dashed line) for (a) 3H-thymidine incorpo-rated (n = 315) and (b) 14C-leucine incorporated (n = 366) samples for the period of July 2010 to July 2011, inclusive. Temporal variation in epilimnetic and hypolimnetic estimates of (c) bacterial abundance (n = 135), (d) bacterial biomass (n = 135), and (e) bacterial biovolume (n = 135) for the period of June 2010 to July 2011, inclusive. Each data point represents the overall mean of all stations sampled calculated by averaging the means of the 3 replicates for each station sampled during that week. Error bars represent the standard deviation of the mean. Sample size (n) refers to entire dataset on figure; 11 stations, 3 replicates, and 2 depths.

DOI: 10.5268/IW-3.2.536



The lowest biovolume was measured in the winter, under ice (0.0032 ± 0.0022 μm3 cell−1; Table 3). Similar to bacterial production, the maximum biovolume was observed in June (0.0081 ± 0.0060 μm3 cell−1). During the spring, bacterial abundance and biovolume were negatively correlated (r = −0.664, p = 0.007; Table 4).

Due to autocorrelation resulting from bacterial biomass calculations based on measured values of bacterial abundance and biovolume, significant positive relation-ships were observed between bacterial abundance and biovolume with biomass (r = 0.907, p < 0.001, and r = 0.438, p < 0.001, respectively; Table 4).

Fig. 3. Temporal variation in epilimnetic bacterial production (n = 292) and (a) surface water temperature (n = 300) for the period of June 2010 to July 2011; horizontal line represents 17 ºC, above which the temperature and bacterial production relationship starts to uncouple, and (b) TDP (n = 295) for the period of June 2010 to July 2011 as well as an insert for the period of June to October 2010 (summer stratified period) when water temperatures were approximately 17 ºC. Each data point represents the mean of the epilimnetic samples calculated by averaging the means of the 3 replicates for all stations sampled during that week. Error bars represent the standard deviation of the mean. Sample size (n) refers to the epilimnetic dataset, including 3 replicates for bacterial production and one for surface temperature and TDP.

244

DOI: 10.5268/IW-3.2.536

CJ Quinn et al.


Annual epilimnetic bacterial abundance was positively correlated with temperature (r = 0.254, p = 0.010), Chl-a (0.7–2 μm: r = 0.305, p = 0.002), and DOC concentrations (r = 0.226, p = 0.021; Table 5). In the stratified summer period, epilimnetic bacterial abundance was positively correlated with Chl-a (20–200 μm: r = 0.265, p = 0.043; 2–20 μm: r = 0.277, p = 0.034; 0.7–2 μm: r = 0.287, p = 0.029). Annual bacterial biomass was positively correlated with temperature (r = 0.329, p = 0.0007) and negatively correlated with TDP (r = −0.213, p = 0.031; Table 5). Annual epilimnetic bacterial biovolumes were positively correlated with temperature (r = 0.241, p = 0.014) and negatively correlated with Chl-a (20-200 μm: r = −0.302, p = 0.002; 2–20 µm: r = −0.361, p = 0.0002; 0.7–2 μm: r = −0.282, p = 0.004), DOC (r = −0.331, p = 0.0006), TP (r = −0.218, p = 0.026), and TDP (r = −0.228, p = 0.021) concentra-tions (Table 5).

At the seasonal scale, epilimnetic bacterial biovolumes were positively correlated with hypolimnetic O2 concen-trations in the spring and summer (r = 0.591, p = 0.043 and r = 0.395, p = 0.002, respectively; Table 5). Epilimnetic bacterial biovolume was negatively correlated with all size fractions of Chl-a (20–200 μm: r = −0.419, p = 0.0009; 2–20 μm: r = −0.437, p = 0.0005; and 0.7–2 μm: r = −0.483, p = 0.0001), DOC (r = −0.523, p <0.0005), and TP (r = −0.253, p = 0.050) concentrations during the summer (Table 5).

Discussion

We did not find any relationship between the bacterial parameters measured and hypolimnetic O2 concentrations, indicating that bacteria do not play a dominant role in O2 consuming processes in Lake Simcoe. This suggests that the low hypolimnetic O2 concentrations at the end-of-summer in Lake Simcoe are not directly related to hetero-trophic bacterial production and may be more strongly related to biomass produced by autotrophs. Bacterial production accounted for 11.6 and 24% of primary production in Lake Simcoe using thymidine and leucine incorporation, respectively (Kim 2013).

During the period of this study (June 2010 to July 2011), there was limited spatial variation in Lake Simcoe’s microbial community. Conversely, distinct seasonal patterns were observed with bacterial production, abundance, biomass, and biovolume consistently lower in the winter relative to the spring, summer, and fall. Annual differences between epilimnetic and hypolimnetic samples were inconsistent for the 4 bacterial parameters measured. Lake temperature, Chl-a, and DOC concentrations were strongly correlated with annual bacterial parameters, suggesting these factors affect bacteria in the epilimnion

of Lake Simcoe. Higher water temperatures were associated with greater bacterial production, biomass, and biovolume. Higher Chl-a concentrations of size fraction 20–200 μm were associated with decreased bacterial production, while Chl-a of size fraction 0.7–2 μm was negatively correlated with bacterial abundance and biovolume. In addition, higher DOC concentrations were associated with stimulated bacterial production and abundance, but decreased cell biovolume.

Both the 3H-thymidine and 14C-leucine incorporation methods illustrated similar seasonal and spatial patterns of bacterial production rates. This agreement between independent, simultaneous estimates provides confidence that the observed variations reflect real changes in the rates of bacterial production. The 3H-thymidine and 14C-leucine incorporation methods are commonly in agreement (Chin-Leo and Kirchman 1988, Jonas et al. 1988, van Looij and Riemann 1993). The leucine method uses the incorporation of L-leu into protein to provide a direct estimate of the growth rate of biomass and thus the flux of C into bacterial biomass.

Although the thymidine method has been used most extensively (Hollibaugh 1994, Servais 1995), estimating bacterial C production with the thymidine method relies on the knowledge of 3 conversion factors: cell production per mole, cell biovolume, and C content per cell volume (Jørgensen 1992). Discrepancies within these values have been acknowledged to contribute to differences between bacterial production from study to study and even differences between methods within a single study. Our leucine-based estimates were 1.4 times higher than the thy-midine-based estimates (data from July 2010–May 2011), which may be attributed to unbalanced growth in bacteria that can occur when the various growth rates of macromo-lecular syntheses are uncoupled; this generally occurs when bacteria are shifting from one growth rate to another (Chin-Leo and Kirchman 1990). Changes in temperature (Shiah and Ducklow 1997), nutrient concentrations (Tibbles 1996), and bacterial growth rates (Chin-Leo and Kirchman 1988) have been shown to modify the ratio between leucine and thymidine; however, neither our seasonal nor monthly data exhibited a clear pattern that can account for the differences between the 2 methods. Nutrient (DOC, TP, SRP) and Chl-a concentrations did not differ between the winter and spring seasons (p > 0.05; data not shown), implying there were no significant changes in growth rates from the winter to spring, and that balanced growth (i.e., growth when macromolecular syntheses’ various growth rates are coupled, and thus there are no changes in growth rate) was observed. Changes in temperature and nutrients may have contributed to unbalanced growth during the summer, fall, and winter, however, leading to discrepancies between the 2 methods.

DOI: 10.5268/IW-3.2.536



Chemical Parameter

Bacterial Parameter1 Annual Winter Spring Summer Fall

Epi. Temp. BP 0.517 *** (98) −0.040 (16) 0.116 (15) 0.203 (52) 0.210 (15)BA 0.254 ** (105) −0.361 (16) 0.108 (15) −0.139 (59) 0.379 (15)BB 0.329 *** (105) −0.359 (16) 0.077 (15) −0.223 * (59) 0.352 (15)BV 0.241 ** (105) −0.185 (16) −0.202 (15) −0.201 (59) −0.126 (15)

Hypo O2 BP 0.004 (89) 0.259 (12) −0.362 (12) 0.273 * (50) 0.043 (15)BA −0.146 (96) −0.117 (12) −0.439 (12) −0.254 * (57) −0.249 (15)BB −0.065 (96) −0.067 (12) −0.369 (12) −0.023 (57) −0.245 (15)BV 0.154 (96) 0.058 (12) 0.591 * (12) 0.395 *** (57) 0.027 (15)

Chl-a BP −0.238 * (98) −0.147 (16) 0.318 (15) −0.304 * (52) 0.080 (15)(20–200 μm) BA 0.126 (105) −0.320 (16) −0.022 (15) 0.265 * (59) 0.359 (15)

BB −0.014 (105) −0.293 (16) −0.023 (15) 0.020 (59) 0.351 (15)BV −0.302 *** (105) −0.021 (16) 0.009 (15) −0.419 *** (59) −0.042 (15)

Chl-a BP −0.202 * (98) −0.202 (16) 0.291 (15) −0.264 * (52) 0.448 (15)(2–20 μm) BA 0.140 (105) −0.302 (16) −0.022 (15) 0.277 * (59) 0.636 ** (15)

BB −0.027 (105) −0.254 (16) −0.014 (15) 0.022 (59) 0.587 ** (15)BV −0.361 *** (105) 0.017 (16) 0.048 (15) −0.437 *** (59) −0.186 (15)

Chl-a BP −0.058 (97) −0.303 (16) −0.126 (15) −0.226 (51) 0.469 * (15)(0.7–2 μm) BA 0.305 *** (104) −0.344 (16) −0.212 (15) 0.287 * (58) 0.631 ** (15)

BB 0.155 (104) −0.344 (16) −0.166 (15) 0.007 (58) 0.623 ** (15)BV −0.282 ** (104) −0.068 (16) 0.338 (15) −0.483 *** (58) −0.050 (15)

DOC BP 0.207 * (97) 0.416 (16) 0.483 * (15) 0.078 (51) 0.425 (15)BA 0.226 * (104) −0.114 (16) 0.031 (15) 0.137 (58) 0.331 (15)BB 0.064 (104) −0.130 (16) −0.025 (15) −0.143 (58) 0.303 (15)BV −0.331 *** (104) −0.047 (16) −0.273 (15) −0.523 *** (58) −0.135 (15)

TP BP −0.070 (98) 0.066 (16) 0.363 (15) −0.010 (52) −0.330 (15)BA 0.043 (105) 0.036 (16) −0.304 (15) 0.086 (59) 0.232 (15)BB −0.053 (105) 0.027 (16) −0.275 (15) −0.053 (59) 0.129 (15)BV −0.218 * (105) −0.007 (16) 0.315 (15) −0.253 * (59) −0.468 * (15)

TDP BP −0.278 ** (96) −0.512 (16) 0.286 (14) 0.179 (51) −0.664 ** (15)

BA −0.130 (103) 0.302 (16) 0.451 (14) −0.048 (58) −0.266 (15)BB −0.513 * (103) 0.318 (16) 0.425 (14) −0.122 (58) −0.308 (15)BV −0.228 * (103) 0.084 (16) −0.388 (14) −0.160 (58) −0.181 (15)

SRP BP −0.133 (96) −0.206 (16) −0.008 (15) 0.115 (50) −0.236 (15)BA −0.163 (103) 0.309 (16) 0.100 (15) −0.228 (57) −0.066 (15)BB −0.201 * (103) 0.499 * (16) 0.067 (15) −0.241 ** (57) −0.080 (15)BV −0.129 (103) 0.357 * (16) −0.211 (15) −0.085 (57) −0.063 (15)

1BP: Bacterial production (leucine incorporated); BA: Bacterial abundance; BB: Bacterial biomass; BV: Bacterial biovolume

Table 5. Coefficients (with n value in parentheses) for the correlations between the epilimnetic bacterial parameters and the examined environ-mental factors (epilimnetic temperature, hypolimnetic O2, Chl-a, phosphorus, and DOC concentrations) for 11 sampling stations in Lake Simcoe for the period of June 2010 to July 2011; correlation p-values are significant at *0.05, **0.01, and ***0.001.

246

DOI: 10.5268/IW-3.2.536

CJ Quinn et al.


Loc

atio

nPr

oduc

tion

TdR

met

hod

μg C

L−1

h−1

Prod

uctio

n L

eu m

etho

dμg

C L

−1 h

−1

Abu

ndan

ce

cells

L−1

Bio

volu

me

μm3 c

ell−1

Bio

mas

s μg

C L

−1Sa

mpl

ing

peri

odM

etho

dsR

efer

ence

Lake

Si

mco

e0.

116

0.36

51.

56 E

× 1

090.

0050

1.41

Jun–

Oct

- DA

PI m

etho

d- E

piflu

ores

cenc

e m

icro

scop

e- 3 H

-TdR

(20

nM),

14C

-Leu

cine

(

50 n

M)

- Filt

ratio

n m

etho

d

Cur

rent

stud

y

Lake

Erie

0.

05–0

.91

5.2–

13.1

× 1

09- 3 H

-TdR

- Epi

fluor

esce

nce

mic

rosc

ope

Hw

ang

and

Hea

th

1997

1.74

1.26

01.

8–4.

6 ×

109

Jul–

Aug

- Acr

idin

e O

rang

e M

etho

d- E

piflu

ores

cenc

e m

icro

scop

e- D

ark

upta

ke- 2

0 nM

TdR

, 40

nM L

eu- F

iltra

tion

met

hod

Wilh

elm

and

Sm

ith

2000

Lake

Erie

(e

aste

rn

basi

n)

0.08

83.

90 ×

109

Jul

- 3 H-L

euci

ne (2

0 nM

)- M

icro

cent

rifug

e m

etho

dD

eBru

yn e

t al.

2004

Lake

M

ichi

gan

0.11

51.

13 ×

109

May

–Jun

- Acr

idin

e O

rang

e M

etho

d- E

piflu

ores

cenc

e m

icro

scop

e- 3 H

-Leu

cine

(20

nM)

- Filt

ratio

n m

etho

d

Bid

dand

a an

d C

otne

r 20

02

Lake

Su

perio

r 2.

44 ×

109

Jun–

Aug

, O

ct.

- DA

PI m

etho

d- E

piflu

ores

cenc

e m

icro

scop

eTa

pper

and

Hic

ks

1998

0.00

5–0.

044

5.67

–9.9

8 ×

108

- DA

PI m

etho

d- E

piflu

ores

cenc

e m

icro

scop

e- 3 H

-Leu

cine

inco

rpor

ated

Hea

th a

nd M

unaw

ar

2004

0.02

70.7

- 3 H-L

euci

ne in

corp

orat

edM

unaw

ar e

t al.

2009

Aug

- Use

d fr

esh

wei

ght t

o co

nver

t

to b

iom

ass

Lake

B

aika

l0.

6–1.

5 ×

109

0.03

–0.1

045.

88–2

1.15

St

rašk

rábo

vá e

t al.

2005

Tab

le 6

. Com

paris

on o

f bac

teria

l par

amet

ers i

n La

ke S

imco

e ve

rsus

pub

lishe

d va

lues

for o

ther

tem

pera

te fr

eshw

ater

lake

s.

DOI: 10.5268/IW-3.2.536



Do summer heterotrophic bacterial parameters represent the maximum seasonal values in temperate large lakes?

Because comparable bacterial studies are lacking lakes of a similar size to Lake Simcoe, we compared our bacterial data to those from larger temperate lakes. Our bacterial production rates and abundances were within the ranges presented in the literature for heterotrophic bacteria in temperate freshwater ecosystems (Table 6). In contrast, bacterial biovolume and biomass in Lake Simcoe were lower than those reported for other large freshwater systems such as Lake Superior (Munawar et al. 2009) and Lake Baikal (Straškrábová et al. 2005; Table 6).

The lowest bacterial production rates in Lake Simcoe were measured under ice during the winter while the maximum values consistently occurred in the summer, although a smaller second peak was observed in March under ice. At low temperatures, bacteria require increased substrate concentrations to grow and respire at rates comparable to those under high temperatures (Pomeroy and Wiebe 2001). The March bacterial production peak coincided with a temporary increase in Chl-a concentrations, suggesting enhanced substrate levels may have compensated for the bacterial metabolism temperature suppression (Biddanda and Cotner 2002). Further research is required to determine if this bacterial production pulse was an annual event or an anomaly. Although there were no significant differences found among bacterial abundances in the spring, summer, and fall, we did observe a November peak in abundance. These results indicate that sampling in the summer will capture peak bacterial production rates but will not be representative of the fall to spring period when production is reduced and will miss the fall abundance peak.

In the Mediterranean Sea, bacterial abundance was found to be lowest during the well-mixed periods, which Trabelsi and Rassoulzadegan (2011) partly attributed to high viral abundance during lake mixing. To date, no viral work has been conducted in Lake Simcoe. Grazing by protozoans also plays a role in controlling bacteria (Sherr et al. 1983, Chrost et al. 2000, Kalff 2002) and can be an important channel for transferring bacterial biomass to upper trophic levels. In the Laurentian Great Lakes (lakes Huron and Michigan), average protozoan abundance is 4–15 cells mL−1, with the lowest abundance occurring in the winter and the highest abundances measured from May–June and October–November (Carrick and Fahnenstiel 1990).

Protozoan abundance has not been reported for Lake Simcoe, but seasonal patterns in bacterial abundance

reflect the protozoan seasonal dynamics reported for lakes Huron and Michigan, with low abundances in the winter and high abundances in the spring to early summer and late summer to early fall. In July, when both bacterial abundance and biomass decreased, bacterial production was at its maximum and biovolume exhibited a small peak. This indicates fewer cells were being produced but their mean size had increased. Bacterial biovolume in Lake Simcoe was lower than that reported in other freshwater studies (Table 6). Novitsky and Morita (1976) reported a decrease in bacterial biovolume of starved cells, which may have been the case during the winter in Lake Simcoe when the smallest biovolumes were measured. In Lake Erie, maximum bacterial biovolumes were observed in May and July, while minimum biovolumes occurred in June and August (Hwang and Heath 1997). In contrast, maximum bacterial biovolume for Lake Simcoe was observed in June, when bacterial production was at a maximum. Increased temperature and C availability during this time may have allowed production of larger bacterial cells.

Factors controlling bacterial production in Lake Simcoe

Bacterial production rates are affected by both abiotic factors (e.g., temperature, substrate availability, and nutrient concentrations; Jonas et al. 1988, Kirchman 1990, Pomeroy and Wiebe 1993), and biotic factors (e.g., grazing; Button 1994, Sherr and Sherr 2002). Numerous studies have shown that temperature and bacterial production are positively correlated (Servais 1989, Findlay et al. 1991, Coveney and Wetzel 1995), as was seen in Lake Simcoe. In some temperate lakes, bacterial production and temperature were correlated below a certain critical level (10 ºC for Lake Michigan, Scavia and Laird 1987; 14 ºC for Mirror Lake, Ochs et al. 1995). At temperatures above this critical level, bacterial production may be limited by nutrient concentrations (Morris and Lewis 1992) and substrate quality (Kirchman 1990), as well as bacterial grazing and viral lysis. Thus, temperature may also regulate nutrient competition between bacteria and phytoplankton (Pomeroy and Wiebe 1988, Ameryk et al. 2005). In Lake Simcoe, this critical temperature seems to be 17 °C, above which bacterial production and water temperatures were uncoupled. Grazing by protozoa could be responsible for the decoupling between bacterial production and water temperature seen here at temperatures above 17 ºC. At the same time that protozoa peaked in lakes Huron and Michigan, bacterial production in Lake Simcoe was tightly coupled with TDP.

248

DOI: 10.5268/IW-3.2.536

CJ Quinn et al.


Annual bacterial production was significantly correlated with Chl-a (20–200 μm) and DOC concentra-tions. DOC and Chl-a concentrations are useful variables for estimating the supply of organic matter (Chrost et al. 2000) to heterotrophic bacteria. As Chl-a concentrations increased in Lake Simcoe, bacterial production decreased, likely a result of decreased P (TDP and SRP) availability for bacteria due to uptake by phytoplankton. One source of labile DOC is through the release by phy-toplankton (Chrost et al. 2000). Similar to the findings of Biddanda and Cotner (2002), our DOC concentrations were positively correlated with Chl-a concentrations, particularly the 0.7–2 μm size fraction. Annual bacterial abundance and biovolume were also significantly correlated with both Chl-a and DOC concentrations, implying that organic matter from phytoplankton seem to provide an important substrate for not only bacterial production, but abundance and biovolume as well.

Given the importance of P as a limiting nutrient to phytoplankton, combined with the strong role it plays in the management of Lake Simcoe (Young et al. 2011), we compared trends in bacterial parameters with the different forms of P. With increasing TP and TDP concentrations, bacterial cells were smaller, which may be attributed to an increase in phytoplankton biomass. Stronger competition between bacteria and phytoplankton for P may occur during periods of increased P concentrations; thus, P may be limiting for both phytoplankton and bacteria in Lake Simcoe. Although bacterial production was occurring, bacterial cell size may have been limited by availability of P during the summer.

With increasing P concentrations there could be an increase in predator populations, which could also contribute to smaller bacterial cell size. Bacteria are selectively grazed by protozoan bacterivores, which tend to select the largest, fastest growing cells (Sherr et al. 1992). In eutrophic systems there should be a positive re-lationship between frequency of larger, rapidly growing cells and protozoan bacterivory (Sherr et al. 1992). Zooplankton and dreissenid mussels can also selectively graze larger bacterial cells, thus eliminating larger cells at a faster rate than smaller ones in the water column. When protozoans are the dominant grazers, however, zooplankton and dreissenids feed on smaller bacterial cells (Kalff 2002). In addition, viral lysis is known to account for a significant fraction of bacterial mortality in eutrophic systems (Weinbauer and Peduzzi 1995). Overall, in Lake Simcoe the relationship between P con-centrations and small-sized bacteria may be attributed to increased competition for nutrients by phytoplankton or changes in grazing dynamics protozoans, zooplankton, and dreissenids.

Relationship between bacterial activity and lake trophic status

Within Lake Simcoe, a range in trophic status can be observed from eutrophic conditions in Cook’s Bay (21.8 mg L−1 annual ice-free mean TP concentration in 2010), to the oligotrophic main basin and Kempenfelt Bay (12.0 and 12.5 mg L−1, respectively, annual ice-free mean TP concentrations in 2010; Wetzel 2001, North et al. 2013). Trophic status of a lake can influence bacterial activity measurements; however, the lack of spatial variation in bacterial activity from our sampling sites may be due to the limited range and overall low bacterial activity in Lake Simcoe. Chrost et al. (2000) found that autochthonous C was much more microbiologically labile than allochtho-nous C. They also observed that the highest proportion of labile DOC was found in enriched lakes. Labile DOC that is released by phytoplankton (EOC; Meyer et al. 1987, Moran and Hodson 1990, Chrost et al. 2000), aquatic macrophytes (Murray and Hodson 1985), and allochtho-nous DOC from sources such as agriculture and sewage are the most significant energy sources for bacterial production (Meyer et al. 1987, Moran and Hodson 1990).

Due to the range in trophic status across the lake, variation in bacterial activity was expected, but was not observed. The lack of spatial differences may be due to competition with phytoplankton, increased grazing pressure, and free versus attached bacteria. Eutrophic Cook’s Bay has high biomass of both phytoplankton (North et al. 2013) and submerged aquatic vegetation (SAV; Depew et al. 2011, Ginn 2011). SAV has high nutrient demands and also serves as habitat for other organisms including bacteria. This may result in numerous bacteria attached to the vegetation while free-floating bacteria dominate at other sites; the former would not be included in measurements made in this study. Perhaps as a result of such high primary producer biomass, higher zooplankton abundances have been reported in Cook’s Bay relative to other areas of the lake (North et al. 2013). This increased grazing pressure may be contributing to the low bacterial abundance. An additional, but untested, con-tributing factor to bacterial mortality could be high viral lysis rates, typical of eutrophic systems (Weinbauer and Peduzzi 1995).

Why was there no relationship between bacterial production and measures of abundance?

In contrast to previous studies (Biddanda and Cotner 2002, DeBruyn et al. 2004), epilimnetic bacterial production in Lake Simcoe was not significantly correlated to bacterial abundance, biomass, or biovolume.

DOI: 10.5268/IW-3.2.536



The lack of relationships may be attributed to a change in the proportion of metabolically active bacteria (DeBruyn et al. 2004). The DAPI stain used in this study was a DNA-specific, fluorescing stain commonly used to stain bacteria; however, it does not differentiate between living and dead cells (Madigan et al. 2000), whereas bacterial production only measures viable cells.

Other researchers have found that a significant proportion of aquatic heterotrophic bacterial populations can be in an inactive state, especially under starvation conditions (Cole 1982, Goulder 1991, Smith and del Giorgio 2003), which could explain why bacterial production, abundance, biomass, and biovolume were sig-nificantly lower in winter than in other seasons. The addition of limiting nutrients may stimulate inactive bacteria to become active, thus increasing production rates while maintaining the same bacterial abundance (DeBruyn et al. 2004). There may be evidence of this occurring in Lake Simcoe; during fall mixis, with increased P concentra-tions in the water column, bacterial production was strongly correlated to both bacterial abundance and biomass.

Additionally, the lack of relationship between bacterial production and the other bacterial parameters could be due to grazing pressure on bacteria by protozoans and viral lysis (Chin-Leo and Kirchman 1990). Grazing by protozoans plays an important role in controlling bacterial numbers (Sherr et al. 1983, Chrost et al. 2000, Kalff, 2002), therefore, abundance is less responsive to environ-mental conditions (nutrients and temperature) and more responsive to grazing than bacterial production. In addition, viral lysis can affect bacterial production and abundance relationships because viral lysis redistributes nutrients to bacteria, which can be associated with an increase in bacterial production but a decrease in bacterial abundance.

Our study has provided insights on bacterial activity in Lake Simcoe, but additional work is needed to more fully describe bacteria in Lake Simcoe; in particular, multivari-ate analysis may help to understand the environmental factors affecting bacterial activity and seasonal dynamics. Future bacterial studies should also investigate hetero-trophic respiration rates in the system on a seasonal and spatial basis, and characterize the influence of viral abundances and activity on bacteria. It is also unknown what influence protozoan and dreissenid grazing have on bacterial populations in the lake.

Conclusions

This study consisted of extensive temporal and spatial measurements of bacterial activity in Lake Simcoe to determine if bacteria contribute to the low hypolimnetic O2 concentrations in the lake, as well as to investigate the

factors influencing annual epilimnetic patterns in the 4 bacterial parameters measured. Annual bacterial production was low in Lake Simcoe relative to other systems, and we found no evidence for heterotrophic bacterial activity contributing to low hypolimnetic O2 conditions. Although thymidine and leucine incorporation methods resulted in significantly different bacterial production estimates, the resultant temporal and spatial patterns were similar between the 2 methods. We found that bacterial production rates were lowest under the ice during the winter, while the maximum values consistently occurred during the open-water season. Maximum bacterial abundance occurred in November in this study, thus sampling exclusively in the spring and summer may result in an underestimation of bacterial abundance.

In addition, by not sampling winter abundances, researchers may be over-estimating annual abundance rates. Annual differences between the mean epilimnetic and hypolimnetic samples for the bacterial parameters were inconsistent as one was not consistently higher than the other. Due to the lack of spatial differences in bacterial activity across the length of the lake, future work may not require such extensive spatial sampling. In Lake Simcoe, bacterial activity was positively correlated with high lake temperature and high DOC concentrations with the exception of bacterial biovolume, which was negatively correlated with DOC, most likely due to the abundance of smaller cells being produced. To our knowledge, ours is the first investigation of heterotrophic bacterial activity in Lake Simcoe, and we hope our work will initiate further research on this important component of the Lake Simcoe ecosystem.

Acknowledgements

We thank Environment Canada’s Lake Simcoe Clean-Up Fund (LSCUF) and an NSERC DG to Peter J. Dillon for funding. We thank Lewis Molot for his support and guidance throughout this project as well as Clayton Williams and Victoria Jackson for their assistance with bacterial parameter calculations. Technical facilitation/assistance was provided by David Evans, Victoria Kopf, and Tedy Ozersky (OMNR-Trent), Veronique Hiriart-Baer and technical operations staff (Environment Canada), Brian Ginn, Ryan MacLean, and Lance Aspden (LSRCA), Courtney English, Kamila Baranowska, Joe Findeis, and Shufen Xu (Trent University), Emilija Cvetanovska and Lisa Kim (University of Waterloo), and Quinn’s Marina. We would also like to thank William D. Taylor and Ralph E.H. Smith (University of Waterloo) for use of their epif-luorescence microscope and R.E.H. Smith for his support throughout this project. We thank the two anonymous reviewers for taking the time to read this manuscript.

250

DOI: 10.5268/IW-3.2.536

CJ Quinn et al.


References

Ameryk A, Podgórska B, Witek Z. 2005. The dependence between bacterial production and environmental conditions in the Gulf of Gdańsk. Oceanologia. 47:27–45.

Bell RT. 1993. Estimating production of heterotrophic bacterioplank-ton via incorporation of tritiated thymidine. In: Kemp PF, Sherr BF, Sherr EB, Cole JJ, editors. Handbook of methods in aquatic microbial ecology. Boca Raton (FL): Lewis Publishers. p. 495–503.

Biddanda BA, Cotner JB. 2002. Love handles in aquatic ecosystems: the role of dissolved organic carbon drawdown, resuspended sediments, and terrigenous inputs in the carbon balance of Lake Michigan. Ecosystems. 5:431–445.

Bratbak G. 1993. Microscope methods for measuring bacterial biovolume: epifluorescence microscopy, scanning electron microscopy and transmission electron microscopy. In: Kemp PF, Sherr BF, Sherr EB, Cole JJ, editors. Handbook of methods in aquatic microbial ecology. Boca Raton (FL): Lewis Publishers. p. 309–318.

Button DK. 1994. The physical base of marine bacterial ecology. Microb Ecol. 28:273–285.

Caron DA. 1994. Inorganic nutrients, bacteria, and the microbial loop. Microb Ecol. 28:295–298.

Carrick HJ, Fahnenstiel GL. 1990. Planktonic protozoa in Lakes Huron and Michigan: seasonal abundance and composition of ciliates and dinoflagellates. J Great Lakes Res. 16:319–329.

Chin-Leo G, Kirchman DL. 1988. Estimating bacterial production in marine waters from the simultaneous incorporation of thymidine and leucine. Appl Environ Microb. 54:1934–1939.

Chin-Leo G, Kirchman DL. 1990. Unbalanced growth in natural assemblages of marine bacterioplankton. Mar Ecol Prog Ser. 63:1–8.

Chrost RJ, Koton M, Siuda W. 2000. Bacterial secondary production and bacterial biomass in four Mazurian Lakes of differing trophic status. Pol J Environ Stud. 9:255–266.

Cole JJ. 1982. Interactions between bacteria and algae in aquatic ecosystems. Annu Rev Ecol Syst. 13:21–314.

Cole JJ, Pace ML. 1995. Why measure bacterial production? A reply to the comment by Jahnke and Craven. Limnol Oceanogr. 40:441–444.

Cotner JB, Wetzel RG. 1992. Uptake of dissolved inorganic and organic phosphorus compounds by phytoplankton and bacterio-plankton. Limnol Oceanogr. 37:232–243.

Coveney MF, Wetzel RG. 1995. Biomass, production, and specific growth rate of bacterioplankton and coupling to phytoplankton in an oligotrophic lake. Limnol Oceanogr. 40:1187–1200.

Currie DJ, Kalff J. 1984. A comparison of the abilities of freshwater algae and bacteria to acquire and retain phosphorus. Limnol Oceanogr. 29:298–310.

DeBruyn J, Leigh-Bell JA, McKay RM, Bourbonniere RA, Wilhelm SW. 2004. Microbial distributions and the impact of phosphorus on bacterial activity in Lake Erie. J Great Lakes Res. 30:166–183.

Depew DC, Houben AJ, Ozersky T, Hecky RE, Guildford SJ. 2011.

Submerged aquatic vegetation in Cook’s Bay, Lake Simcoe: Assessment of changes in response to increased water transparency. J Great Lakes Res. 37:72–82.

Ducklow HW, Purcbe DA, Williams PJ, Davies JM. 1986. Bacterio-plankton: a sink for carbon in a coastal marine plankton community. Science. 232:865–867.

Findlay S, Pace ML, Lints D, Cole JJ, Caraco NF, Peierls B. 1991. Weak coupling of bacterial and algal production in a heterotrophic ecosystem, the Hudson estuary. Limnol Oceanogr. 36:268–278.

Frischer ME, Nierzwicki-Bauer SA, Parsons RH, Vathanodorn K, Waitkus KR. 2000. Interactions between zebra mussels (Dreissena polymorpha) and microbial communities. Can J Fish Aquat Sci. 57:591–599.

Fuhrman JA, Azam F. 1980. Bacterioplankton secondary production estimates for coastal waters of British Columbia, Antarctica, and California. Appl Environ Microb. 39:1085–1095.

Fuhrman JA, Azam F. 1982. Thymidine incorporation as a measure of heterotrophic bacterioplankton production in marine surface waters: evaluation and field results. Mar Biol. 66:109–120.

Gardner WS, Chandler JF, Laird GA, Scavia P. 1986. Microbial response to amino acid additions in Lake Michigan: Grazer control and substrate limitation of bacterial population. J Great Lakes Res. 12:161–174.

Ginn BK. 2011. Distribution and limnological drivers of submerged aquatic plant communities in Lake Simcoe (Ontario, Canada): Utility of macrophytes as bioindicators of lake trophic status. J Great Lakes Res. 37:83–89.

Goosen NK, van Rijswijk P, Brockmann U. 1995. Comparison of het-erotrophic bacterial production rates in early spring in the turbid estuaries of the Scheldt and the Elbe. Hydrobiologia. 311:31–42.

Goulder R. 1991. Metabolic activity of freshwater bacteria. Sci Prog. 75:73–91.

Hawryshyn J, Ruhland KM, Quinlan R, Smol JP. 2012. Long-term water quality changes in a multiple-stressor system: a diatom-based paleolimnological study of Lake Simcoe (Ontario, Canada). Can J Fish Aquat Sci. 69:24–40.

Hollibaugh JT. 1994. Relationship between thymidine metabolism, bacterioplankton community metabolic capabilities, and sources of organic matter. Microb Ecol. 28:117–131.

Horne AJ, Goldman CR. 1994. Limnology. 2nd ed. New York (NY): McGraw-Hill.

Hwang SJ, Heath RT. 1997. Bacterial productivity and protistan bacterivory in coastal and offshore communities of Lake Erie. Can J Fish Aquat Sci. 54:788–799.

Jonas RB, Tuttle JH, Stoner DL, Ducklow HW. 1988. Dual-label radi-oisotope method for simultaneously measuring bacterial production and metabolism in natural waters. Appl Environ Microb. 54:791–798.

Jørgensen NOG. 1992. Incorporation of [3H] leucine and [3H] valine into protein of freshwater bacteria: field application. Appl Environ Microb. 58:3647–3653.

Kalff J. 2002. Limnology: Inland water ecosystems. New Jersey (NJ): Prentice Hall.

DOI: 10.5268/IW-3.2.536



Kim TY. 2013. Primary production by phytoplankton in Lake Simcoe, 2010–2011 [master’s thesis]. Waterloo (ON): University of Waterloo.

Kirchman DL. 1990. Limitation of bacterial growth by dissolved organic matter in the subarctic Pacific. Mar Ecol Prog Ser. 62:47–54.

Kirchman DL. 1994. The uptake of inorganic nutrients by hetero-trophic bacteria. Micro Ecol. 28:255–271.

Kirchman D. 2001. Measuring bacterial biomass production and growth rates from leucine incorporation in natural aquatic environ-ments. Method Microbiol. 30:227–236.

Madigan MT, Martinko JM, Parker J. 2000. Brock: Biology of micro-organisms. 9th ed. Saddle River (NJ): Prentice Hall.

Meyer J, Edwards RT, Risley R. 1987. Bacterial growth on dissolved organic carbon from a blackwater river. Micro Ecol. 13:13–29.

Moran MA, Hodson RE. 1990. Bacterial production on humic and nonhumic components of dissolved organic carbon. Limnol Oceanogr. 35:1744–1756.

Morris DP, Lewis WMJ. 1992. Nutrient limitation of bacterioplankton growth in Lake Dillon, Colorado. Limnol Oceanogr. 37:1179–1192.

Munawar M, Munawar IF, Fitzpatrick M, Niblock H, Lorimer J. 2009. The base of the food web at the top of the Great Lakes: structure and function of the microbial food web of Lake Superior. In Munawar M and Munawar IF, editors. State of Lake Superior. New Delhi: Goodwood Books. p. 289–318.

Murray RE, Hodson RE. 1985. Annual cycle of bacterial secondary production in five aquatic habitats of the Okefenokee Swamp Ecosystem. Appl Environ Microb. 49:650–655.

Nicholls KH. 1995. Some recent water quality trends in Lake Simcoe, Ontario: Implications for basin planning and limnological research. Can Wat Resour J. 20:213–226.

North RL, Barton D, Crowe AS, Dillon PJ, Dolson RML, Evans DO, Ginn BK, Håkanson L, Hawryshyn J, Jarjanazi H, et. al. 2013. The state of Lake Simcoe (Ontario, Canada): the effects of multiple stressors on phosphorus and oxygen dynamics. Inland Waters. 3(1):51–74.

Novitsky JA, Morita RY. 1976. Morphological characterization of small cells resulting from nutrient starvation of a pyschrophilic marine vibrio. Appl Environ Microbiol. 32:617–622.

Ochs CA, Cole JJ, Likens GE. 1995. Population dynamics of bacterio-plankton in an oligotrophic lake. J Plankton Res.17:365–391.

Olsen LK, Reinertsen H, Vadstein O. 2002. Can phosphorus limitation inhibit dissolved organic carbon consumption in aquatic microbial food webs? A study of three food web structures in microcosms. Microb Ecol. 43:353–366.

[OMOE] Ontario Ministry of the Environment. 2007. The determina-tion of total phosphorus in water by colourimetry. Toronto (ON).

Pace ML, Cole JJ. 1996. Regulation of bacteria by resources and predation tested in whole-lake experiments. Limnol Oceanogr. 41:1448–1460.

Pomeroy LR, Wiebe WJ. 1988. Energetics of microbial food webs. Hydrobiologia. 15:7–18.

Pomeroy LR, Wiebe WJ. 1993. Energy sources for microbial food

webs. Mar Microb Food Webs. 7:101–118.Pomeroy LR, Wiebe WJ. 2001. Temperature and substrates as

interactive limiting factors for marine heterotrophic bacteria. Aquat Microb Ecol. 23:187–204.

Scavia D, Laird GA. 1987. Bacterioplankton in Lake Michigan: Dynamics, controls, and significance to carbon flux. Limnol Oceanogr. 32:1017–1033.

Servais P. 1989. Bacterioplankton biomass and production in the river Meuse (Belgium). Hydrobiologia. 174:99–110.

Servais P. 1995. Measurements of the incorporation rates of four amino acids into proteins for estimating bacterial production. Microb Ecol. 29:115–128.

Sherr BF, Sherr EB, Berman T. 1983. Grazing growth and ammonium excretion rates of a heterotrophic microflagellate fed with four species of bacteria. Appl Environ Microb. 45:1196–1201.

Sherr BF, Sherr EB, McDaniel J. 1992. Effect of protistan grazing on the frequency of dividing cells in bacterioplankton assemblages. Appl Environ Microb. 58:2381–2385.

Sherr EB, Sherr BF. 2002. Significance of predation by protists in aquatic microbial food webs. Anton Leeuw Int J G. 81:293–308.

Shiah FK, Ducklow HW. 1997. Bacterioplankton growth responses to temperature and chlorophyll variations in estuaries measured by thymidine:leucine incorporation ratio. Aquat Microb Ecol. 13:151–159.

Smith DC, Azam F. 1992. A simple, economical method for measuring bacterial protein synthesis rates in seawater using 3H-leucine. Mar Microb Food Webs. 6:107–114.

Smith EM, del Giorgio PA. 2003. Low fractions of active bacteria in natural aquatic communities? Aquat Microb Ecol. 31:203–208.

Smith REH, Furgal JA, Charlton MN, Greenberg BMH, Hiriart VP, Marwood CA. 1999. The attenuation of ultraviolet radiation in a large lake with low dissolved organic matter concentrations. Can J Fish Aquat Sci. 56:1351–1361.

Stainsby EA, Winter JG, Jarjanazi H, Paterson AM, Evans DO, Young JD. 2011. Changes in thermal stability of Lake Simcoe from 1980 to 2008. J Great Lakes Res. 37:55–62.

Stainton MP, Capel MJ, Armstrong FAJ. 1977. The chemical analysis of freshwater, 2nd ed. Winnipeg (MB): Department of Fisheries and Oceans Canada.

Straškrábová V, Izmest’yeva LR, Maksimova EA, Fietz S, Nedoma J, Boroves J, Kobanova GI, Shchetinina EV, Pislegina EV. 2005. Primary production and microbial activity in the euphotic zone of Lake Baikal (Southern Basin) during late winter. Glob Planet Change. 46:57–73.

Sturluson M, Nielsen TG, Wassmann P. 2008. Bacterial abundance, biomass and production during spring blooms in the northern Barents Sea. Deep-Sea Res. II, 55:2186–2198.

Tapper MA, Hicks RE. 1998. Temperate viruses and lysogeny in Lake Superior bacterioplankton. Limnol. Oceanogr. 43:95–103.

Tibbles BJ. 1996. Effects of temperature on the incorporation of leucine and thymidine by bacterioplankton and bacterial isolates. Aquat Microb Ecol. 11:239–250.

Trabelsi A, Rassoulzadegan F. 2011. Effect of bacterial community

252

DOI: 10.5268/IW-3.2.536

CJ Quinn et al.


dynamics on DOC seasonal changes in the north-western Mediter-ranean Sea. J Plankton Res. 6:1–14.

Van Looij A, Riemann B. 1993. Measurements of bacterial production in coastal marine environments using leucine: application of a kinetic approach to correct for isotope dilution. Mar Ecol Prog Ser. 102:97–104.

Weinbauer MG, Peduzzi P. 1995. Significance of viruses versus heter-otrophic nanoflagellates for controlling bacterial abundance in the northern Adriatic Sea. J Plankton Res. 17:1851–1856.

Wetzel RG. 2001. Limnology: Lake and river ecosystems. New York (NY): Academic Press.

Wilhelm SW, Smith REH. 2000. Bacterial carbon production in Lake Erie is influenced by viruses and solar radiation. Can J Fish Aquat Sci. 57:317–326.

Young JD, Landre AL, Winter JG, Jarjanazi H, Kingston J. 2010. Lake Simcoe water quality update. Toronto (ON): Ontario Ministry of the Environment.

Young JD, Winter JG, Molot L. 2011. A re-evaluation of the empirical relationships connecting dissolved oxygen and phosphorus loading after dreissenid mussel invasion in Lake Simcoe. J Great Lakes Res. 37:7–14.

year-round patterns in bacterial production ... -...

Documents