Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions

JGLR-01355; No. of pages: 14; 4C: Journal of Great Lakes Research xxx (2018) xxx–xxx

Contents lists available at ScienceDirect

Journal of Great Lakes Research journal homepage: www.elsevier.com/locate/jglr

Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions Bart T. De Stasio ⁎, Ashley E. Beranek 1, Michael B. Schrimpf 2 Department of Biology, Lawrence University, 711 E. Boldt Way, Appleton, WI 54911, USA

a r t i c l e

i n f o

Article history: Received 31 October 2017 Received in revised form 12 May 2018 Accepted 21 May 2018 Available online xxxx Communicated by Jerry Kaster Keywords: Dreissena Bythotrephes Trophic interaction Transfer efficiency Cyanobacteria Zooplankton

a b s t r a c t As the largest freshwater estuary in the Laurentian Great Lakes, Green Bay, Lake Michigan (USA) is an important ecosystem presenting both challenges and opportunities for investigating changes in the face of multiple anthropogenic stressors. We collected new data from 2000 to 2007 to assess changes in lower food web interactions after establishment of invasive species (Bythotrephes longimanus and Morone americana in 1988 and Dreissena polymorpha in 1993) and nutrient reductions (1990s). Phytoplankton and zooplankton biomass and composition, as well as primary productivity and zooplankton community grazing rates, were determined along the previously well-studied trophic gradient from the shallow Lower bay to the stratified, open-water Middle bay. A clear trophic gradient still occurred during 2000–2007, with higher nutrients, phytoplankton and zooplankton in Lower bay compared to Middle bay. Phytoplankton abundance and cyanobacteria dominance increased significantly compared to earlier studies. However, integrated primary productivity did not change significantly at either Lower or Middle bay. Zooplankton standing stock decreased in Lower bay, driven primarily by reductions of bosminids, chydorids, and cyclopoid copepods, but did not change in Middle bay. Zooplankton community grazing rates did not change significantly, but shifts in magnitude and seasonality of net phytoplankton growth rates are consistent with increased phytoplankton standing stocks. Changes in zooplankton composition indicate increased predation by invertebrates and decreased fish predation. Shifts in both bottom-up and top-down factors have occurred, with Lower and Middle bay regions more eutrophic and similar to each other as a result of changes in this highly productive Great Lakes embayment. © 2018 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved.

Introduction The Laurentian Great Lakes have been increasingly exposed to multiple stressors in recent decades, including changes in nutrient loading, climate change, and biological invasions (Stow, 2014; Vanderploeg et al., 2015; Cotner et al., 2017). Depending on the nature of the stressor, each can create bottom-up or top-down effects that can change food web interactions and function. For example, increased nutrient loading during the middle of the 20th century led to strong bottom-up effects, leading to eutrophication of the Great Lakes (Schindler and Vallentyne, 2008; Egan, 2017). In the 1980s, invasion of North America by the predatory cladoceran Bythotrephes longimanus resulted in strong top-down effects causing changes in crustacean zooplankton ⁎ Corresponding author. E-mail addresses: [email protected] (B.T. De Stasio), [email protected] (A.E. Beranek), [email protected] (M.B. Schrimpf). 1 Current address: Wisconsin Department of Natural Resources, PO Box 7921, Madison, WI, USA 53707-7921. 2 Current address: Department of Ecology and Evolution, Stony Brook University, 650 LSB, Stony Brook, NY, USA 11794-5245.

communities and lower food web interactions (Lehman and Caceres, 1993; Barbiero and Tuchman, 2004). Invasion of the Great Lakes by white perch (Morone americana) led to major shifts in planktivory and top-down effects due to declines of yellow perch in Lake Erie (Bur and Klarer, 1991) and effects on minnows, walleye and white bass in the Bay of Quinte, Lake Ontario (Schaeffer and Margraf, 1987). Understanding how ecosystems respond to such stressors has become a major goal of Great Lakes research. Green Bay of Lake Michigan is the largest embayment, and one of the most productive ecosystems of the Laurentian Great Lakes (Bertrand et al., 1976; Klump et al., 2009). Extensive studies during the 1970s and 1980s demonstrated that it was heavily influenced by excessive nutrient loading, resulting in strong bottom-up effects driving a trophic gradient for phytoplankton (Sager and Richman, 1991), zooplankton (Richman et al., 1984), and fish (Smith and Magnuson, 1990). As occurred in many of the Great Lakes, this system was also stressed by biological invasions in the 1980s and 1990s. The spiny water flea Bythotrephes longimanus and white perch Morone americana invaded by 1988 (Jin and Sprules, 1990; Cochran and Hesse, 1994), followed by the zebra mussel Dreissena polymorpha in 1992–93 (Kraft, 1993). During this same period, nutrient reduction efforts became more

https://doi.org/10.1016/j.jglr.2018.05.020 0380-1330/© 2018 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved.

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

2

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

effective as well, reducing phosphorus loading from the main source, the Fox River, by approximately 1.5% per year (Qualls et al., 2013). Given the combination of nutrient reductions and invasion by D. polymorpha it was expected that phytoplankton abundance and overall productivity of the system would decrease, as had occurred in other Great Lakes (Padilla et al., 1996). However, basic water quality conditions have not noticeably improved following these changes (De Stasio et al., 2008, 2014; Qualls et al., 2013), raising the question of why this system has responded differently than other Great Lakes locations to invasions and remediation efforts. Probably the most well-known and documented response of Great Lakes ecosystems to biological invasions has been the example of D. polymorpha, where decreases in phytoplankton and increases in water transparency occurred due to high filtering capacity and rapid population growth of mussels (Lavrentyev et al., 1995; Barbiero et al., 2006; Fishman et al., 2009; Fahnenstiel et al., 2016). However, recent work shows that dreissenid impacts are context dependent (Sarnelle et al., 2005; Qualls et al., 2007; Vanderploeg et al., 2014), and some systems exhibit increased cyanobacteria following invasion (Vanderploeg et al., 2001; Strayer, 2009). These situations, where responses differ from expected changes, have led to multiple hypotheses about mechanisms leading to increased cyanobacteria blooms. These include increased light penetration following water clearing by mussels (leading to favorable conditions for light-tolerant cyanobacteria like Microcystis), and/or selective predation on competitive algae that leads to increased cyanobacteria dominance (Fishman et al., 2010). There is also evidence that responses depend on starting trophic condition of water bodies, with largest negative effects on the cyanobacteria Microcystis observed in more oligotrophic systems and positive effects in more eutrophic conditions (Sarnelle et al., 2005). Another possibility is that recycling of nutrients by mussels may improve conditions for cyanobacteria growth (Arnott and Vanni, 1996; Vanderploeg et al., 2002, 2014). Although zebra mussels are known to decrease the ratio of nitrogen to phosphorus due to their higher retention of nitrogen, the overall release of nitrogen to the water in eutrophic systems with high phosphorus concentrations may be more important than the ratios. This could help explain the recent increased dominance of non nitrogen fixing groups like Microcystis during late summer in eutrophic systems like lower Green Bay. Understanding responses in the Green Bay ecosystem is complex because of the invasions by D. polymorpha, B. longimanus, and M. americana along with nutrient reduction efforts, and requires examining changes in multiple trophic levels to assess the relative roles of bottom-up and top-down factors. Only recently have the effects of B. longimanus in productive systems like Green Bay been examined, demonstrating that this invertebrate predator may have had stronger impacts than previously suspected in nearshore regions and embayments (Pothoven and Höök, 2014; Merkle and De Stasio, 2018). Here we report results of studies on lower food web dynamics in Green Bay during the years 2000–2007 along the previously well studied trophic gradient. These data are then compared to earlier published and unpublished data to assess changes in phytoplankton and zooplankton biomass and productivity, as well as grazing interactions from before and after the invasions and nutrient reductions in Green Bay, Lake Michigan. Our analyses of recent changes in lower food web interactions provide new insights into the relative importance of various forces that are affecting this and other ecosystems of the Laurentian Great Lakes. Methods Green Bay was sampled during the summers of 2000, 2004, 2005, 2006, and 2007 at five locations established during previous studies along a trophic gradient (Fig. 1; Richman et al., 1984; De Stasio and Richman, 1998; Sager and Richman, 1991). Sites sampled range from shallow Lower bay sites with hypereutrophic conditions to a deeper

mesotrophic Middle bay region (maximum depths: GB1A = 1.5 m, GB2 = 3 m, GB3 = 7 m, GB4 = 11 m, GB6 = 15 m). The region south of Long Tail Point and Point Sable are often referred to as the “inner bay” and included stations GB1A and GB2. This area is highly influenced by river water from the lower Fox River, and has water residence times on the order of a few months, depending on river inflows (Klump et al., 2009). Station GB2 corresponds to the “Lower bay” site used in earlier work (Richman and Sager, 1990; Richman et al., 1990; Sager and Richman, 1990, 1991) and GB6 represents “Middle bay” for comparisons of our data with the studies from 1986 to 1988 and 1990–1992. Our samples were collected approximately biweekly each year from June through August, as was done earlier. As reported elsewhere (De Stasio et al., 2008, 2014), standard measures of physical and chemical limnological features were obtained on each date (Secchi depth, vertical profiles of photon flux density, temperature, dissolved oxygen, pH, conductivity, oxidative-reductive potential). Water clarity was measured on each date using a black and white Secchi disk (0.20 m diameter). Vertical profiles of photon flux density were determined at each location with 2π underwater and incident PAR quantum sensors (LI-192S and LI-190) and datalogger (Model LI-1000, LI-COR Co., Lincoln, NE), while other parameters were measured with a multiparameter data sonde (Model DS5, Hydrolab, Loveland, USA). For chlorophyll a (chl-a), phytoplankton composition and primary productivity analyses, duplicate integrated samples were collected from the top 4 m of the water column using a submersible pump (or to just above the bottom at sites shallower than 4 m). Water for chl-a analysis and productivity determinations was transported in opaque bottles kept on ice in the dark until returned to the laboratory later the same day, while phytoplankton samples were preserved in 1% Lugol's solution. In the laboratory chl-a concentration was determined using the standard acetone extraction procedure (Wetzel and Likens, 1991). As in earlier work, replicate subsamples (15–50 mL) for phytoplankton identification and enumeration were examined using settling chambers viewed on an inverted microscope or on permanent slides made by filtering subsamples onto membrane filters (0.45 μm pore size) under low vacuum. Filters were cleared with immersion oil, sealed with Permount and enumerated at 100–500× magnification. Cell linear dimensions were determined with an ocular micrometer and used to estimate cell biovolume based on published relationships between linear dimensions and volume (Wetzel and Likens, 1991). Biovolume was converted to dry weight of biomass assuming 106 μm3 is equivalent to 0.22 μg dry weight (Lind, 1985; Rocha and Duncan, 1985). In 2006 and 2007 phytoplankton community photosynthetic rates were also determined at all five stations using standard 14C-uptake methodology employing photosynthesis versus light intensity curves (P vs. I) according to the procedure of Fee (1998). Duplicate light bottles and one “DCMU” bottle were incubated for ca. 3 h at each of four light intensities and at ambient epilimnetic temperatures. All bottles (50 mL Pyrex) received 0.5 mL (148 kBq) of [14C]NaHCO3 while DCMU bottles also received 0.5 mL of 0.005 M DCMU (Diuron, 3-(3,4dichlorophenyl)-1,1-dimethyl urea) as a photosynthetic inhibitor (Legendre et al., 1983). Uptake of 14C was determined with standard methodology using liquid scintillation counting as performed in Sager and Richman (1991). Total alkalinity and pH were determined for each sample to compute total dissolved inorganic carbon. Estimates of photosynthetic parameters were obtained from P vs. I curves using the curve-fitting programs provided by Fee (1998). Field data on incident solar irradiance, light penetration, chlorophyll, and mixing depths were used with the programs to calculate daily areal photosynthetic rates (∑P = mg C/m2/d), volumetric rates at optimal light intensity (Popt = mg C/m3/h), and maximum biomass-specific rates of photosynthesis (PBm = mg C/mg chl-a/h). Comparison data are available for Lower and Middle bay locations for earlier years. Photosynthetic rates from 1986 to 1988 and 1990–1992 were corrected as explained in Millard and Sager (1994) for differences in calculation methodology between the Fee program approach we employed and earlier methods

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

3

Fig. 1. Map of southern Green Bay and stations sampled during 2000–2007. Station GB2 corresponds to Lower bay and station GB6 represents Middle bay.

used in Green Bay. In addition, daily volumetric photosynthetic rates and algal carbon biomass were used to estimate phytoplankton population growth rates, as performed by Sager and Richman (1991) for the same sites sampled prior to invasion. To facilitate comparisons across trophic levels, photosynthetic rates also were converted from carbon units to dry weight with a conversion factor of 0.5 (Lind, 1985; Rocha and Duncan, 1985). The zooplankton community was sampled with a Clark-Bumpus sampler with a 153-μm mesh net. Replicate oblique tows from 0.5 m above the bottom to just below the surface were performed, ensuring that at least 400 L of water was filtered. Zooplankton samples were preserved in 4% sugar-Formalin solution. Replicate subsamples were examined using a Wild M-5A dissecting microscope at 25×–50× magnification and enumerated using a circular counting tray. Crustacean zooplankton were identified to species using standard taxonomic references (Balcer et al., 1984; Pennak, 1989) and measured with an ocular micrometer (at least 50 individuals per species) to convert counts into dry weight biomass estimates using conversions in Wetzel and Likens (1991) and Bottrell et al. (1976). Microcrustacean daily productivity (mg/m3/day) was estimated using the production-to-biomass (P/B) relationships from Shuter and Ing (1997): Log(P/B) = a + b × T, where T is temperature (°C) and a and b are coefficients from separate regression analyses for Cladocera, Cyclopoida and Calanoida. Temperature (T) was calculated as the average water column temperature from vertical temperature profiles for each day at each station. While this approach is driven by temperature and does not include changes in food quality, it is a standard method for estimating zooplankton productivity in the Great Lakes. Considering the shifts in phytoplankton composition reported below this approach may overestimate zooplankton production because of declines in food quality. Comparisons among time periods

focused on major microcrustacean groups grazing on algae, so larger predatory zooplankton were excluded from the analysis (i.e. Bythotrephes and Leptodora). Zooplankton feeding experiments were conducted at stations GB2 and GB6 on each date in 2006 and 2007 using a modification of the Lampert and Taylor (1985) Haney-Trap method. These sites correspond to Lower and Middle bay locations used in earlier studies (Richman et al., 1990; Sager and Richman, 1990, 1991). Water samples were collected between 1000 h and 1400 h at multiple depths at each site (GB2: 0 m and 2 m; GB6: 0 m, 5 m, and 10 m) with a Schindler Trap (23-L volume) modified to collect whole water samples (i.e. net replaced with a stopper). Feeding experiments were run on board ship in 3-L plastic containers kept in the dark. Containers received 3 mL of 14 C-labelled Chlamydomonas reinhardti suspension, resulting in an increase of approximately 150 cells/mL over ambient phytoplankton concentrations. After 20 min zooplankton were filtered through a 153-μm plankton mesh cup and washed in three successive baths of soda water to remove un-ingested radioactive algae and to anesthetize the zooplankton. Animals were collected on membrane filters (0.45-μm, Gelman Brand) and placed into plastic scintillation vials. Replicate subsamples (20 mL) from the containers were filtered on to membrane filters (0.45-μm) and used to determine radioactivity in the phytoplankton and dissolved fractions. Radioactivity was measured with a Beckman 230 liquid scintillation system. Community grazing rates (mL/L/h) were calculated using the equations in Lampert and Taylor (1985) and also converted to animal specific rates by dividing by animal abundance per liter. Richman et al. (1990) showed no diel patterns in grazing rates for these locations, so the short-term (hourly) rates we determined were multiplied by 24 for conversion to daily rates. For determination of the impact of grazing on phytoplankton population growth,

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

4

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

community grazing rates were converted into phytoplankton loss rates per day by dividing by 103 (Sager and Richman, 1991). We used the method of Sager and Richman (1991) to calculate the difference between photosynthesis based population growth rates and loss rates due to grazing, representing phytoplankton adjusted population growth rates. Difference values were determined from mean water column estimates of growth and grazing rates for each site on each date. As mentioned above, we performed comparisons of published and unpublished data collected in 1982–1992 during the studies by Richman and Sager (Richman et al., 1984, 1990; Richman and Sager, 1990; Sager and Richman, 1990, 1991; De Stasio and Richman, 1998) with our data collected in 2000–2007. These two time periods for which data are available bracket the times when invasions occurred by B. longimanus (1988), M. americana (1988), and D. polymorpha (1992–1993) and also when nutrient reductions occurred (1990s; Qualls et al., 2013). For convenience we refer to the time periods as pre-invasion and post-invasion, with reminders in places that nutrient reductions also occurred between these periods. We also present time series data on total phosphorus and nitrate concentrations in Lower and Middle bay regions collected from 1986 to 2007 by the Green Bay Metropolitan Sewerage District (adapted from Qualls et al., 2013). To examine changes in potential planktivory by fish over time we obtained data from annual fish trawl surveys for Lower bay (Long Tail Point and Point Sable station means) and Middle bay (Pensaukee and West of Little Sturgeon station means; T. Paoli, Wisconsin Department of Natural Resources, personal communication). We examined data for heteroscedasticity and normality, using data transformations where appropriate. Parametric tests (t-test, Analysis of Variance and Tukey's HSD post hoc tests) were employed if possible, with corresponding non-parametric tests used if parametric test assumptions could not be fulfilled. All statistical analyses were performed using PAST (Paleontological Statistics Package, version 3.1; Hammer et al., 2001).

Results Spatial Patterns during 2000–2007 Phytoplankton: A strong trophic gradient for phytoplankton communities occurred during 2000–2007 (Fig. 2). In each year studied mean summer phytoplankton biomass was highest at GB1A and then decreased northwards along the bay to GB6 (Fig. 2a). Phytoplankton biomass varied among years, but densities at GB1A ranged from around 1 g/m3 dry weight (DW) to N3.5 g DW/m3. At GB6, the station farthest from the Fox River, mean summer biomass was between 0.2 and 1.0 g DW/m3. Summer phytoplankton community composition during this time period consisted primarily of cyanobacteria and diatoms, with relative dominance of these groups shifting among locations (Fig. 2c). At GB1A cyanobacteria accounted for N50% of total phytoplankton biomass, with diatoms comprising ca. 40%. Chlorophyte taxa represented 10% or less of total biomass at all sites. Relative density of cyanobacteria decreased with increasing distance from the Fox River, representing b30% of total biomass at GB6 (Fig. 2c). Phytoplankton community assemblage shifted along the spatial gradient, with Middle bay (e.g. GB6) exhibiting over 50% diatoms and another 10% due to other taxa, primarily chrysophytes (Fig. 2b; De Stasio et al., 2014). At all sites summer cyanobacteria community composition was dominated by Microcystis, ranging from approximately 50% to 70% of biomass (Fig. 2d). The remainder of cyanobacteria biomass primarily consisted of Aphanizomenon and Dolichospermum (formerly Anabaena). Rates of integral photosynthesis in 2006 & 2007 decreased along the spatial gradient from Lower to Middle bay (Fig. 2b), similar to the wellestablished gradient of phytoplankton productivity documented in earlier studies (Richman et al., 1984). There were significant differences in productivity among sites in both 2006 and 2007 (repeated measures ANOVA: for 2006, F4,15 = 4.478, p = 0.019; for 2007, F4,15 = 6.021, p =

Fig. 2. Phytoplankton community characteristics at five stations (GB1A–GB6) in Green Bay during 2000–2007. Mean summer values for (a) total phytoplankton biomass (mean ± 1 SE), (b) areal integrated photosynthesis rates (mean ± 1 SE), (c) relative dominance of major phytoplankton taxa, and (d) relative contribution of taxa to cyanobacteria biomass.

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

0.006). In both years the two Lower bay sites in the “inner bay” region south of Long Tail Point and Point Sable (GB1A and GB2, Fig. 1) exhibited the highest productivity, with lower values recorded at the three remaining sites. Mean summer rates in 2006 for the inner bay were between 2.8 and 3.8 g C/m2/d and ranged from 1.0 to 1.9 for GB3 through GB6. In 2007 overall productivity values were lower than in 2006, with inner bay sites closest to the Fox River between 1.8 and 2.1 g C/m2/d and remaining sites ranging from 0.6–0.9 g C/m2/d (Fig. 2b). Productivity at the most northern site in Middle bay (GB6) was significantly lower than at both inner bay sites in both years (Tukey's HSD, df = 20, p ≤ 0.05 for both comparisons). Mean photosynthesis rate at GB4 also was significantly lower than at GB1A in 2006 (Tukey's HSD, Q = 3.243, df = 20, p = 0.043). Zooplankton: Total microcrustacean zooplankton biomass during 2000–2007 did not exhibit a consistent pattern along the trophic gradient (Fig. 3a). Zooplankton mean summer dry weight reached as high as 2600 mg DW/m3 at GB2 in 2000, demonstrating the large potential for secondary production in this largest embayment of the Great Lakes. Summer mean biomass at GB1A at the mouth of the Fox River was typically lower than at other locations, usually 1000 mg DW/m3 or less. Middle bay (GB6) biomass concentrations varied less than at other stations, with a mean of 1128 mg DW/m3 (SE = 61.67) for the postinvasion period. Dreissena veligers were found at all sites sampled and were generally in higher abundances at stations GB1A-GB3 than further away from the Fox River (Fig. 3c). Mean summer biomass of veligers ranged between 20 and 30 mg DW/m3 at GB2 and GB3 in 2000, but was typically b15 mg DW/m3 at all other sites and years. There was high temporal variability within sites, and also high variability across years at a given site. The predatory cladoceran Bythotrephes longimanus was not collected at any stations in 2000, but in all other years it was observed at every station except GB6 and GB4 in 2004 (Fig. 3d). Bythotrephes biomass varied across years, but summer mean values typically increased from GB1A to GB3 or GB4. Biomass in 2004 changed from ca. 20 mg DW/m3 at GB1A to 93 mg DW/m3 at GB3. Zooplankton productivity generally increased from GB1A to GB6 (Fig. 3b). Closest to the Fox River mean summer production was b500 mg DW/m2/day

5

in all years examined. Production estimates increased along the trophic gradient, reaching just under 2000 mg DW/m2/day in Middle bay at station GB6. Grazing by zooplankton measured in 2006 and 2007 at Lower bay (GB2) and Middle bay (GB6) showed that rates did not vary consistently along the gradient (Table 1). Both community grazing rates and individual rates were higher in 2006 than 2007, with the zooplankton community filtering between 177 and 250 mL/L/day in 2006 (i.e. 17–25% of the water). In 2007 b 10% of the water was filtered by zooplankton. On an individual basis, grazing rates were 2.5–3.4 mL/animal/day in 2006 and ca. 1 mL/animal/day in 2007. Biomass estimates of major zooplankton groups during the postinvasion period were negatively related to the dominant cyanobacteria group Microcystis (Fig. 4). Daphnia biomass reached almost 2500 mg DW/m3 when Microcystis was in low abundance, but decreased quickly with increasing biomass of this group (Fig. 4a). Very little Daphnia biomass occurred when Microcystis biomass exceeded approximately 2000 mg DW/m3. Similar, although weaker, relationships occurred with the other zooplankton groups. Total biomass of bosminids and chydorid cladocera appeared to be less affected than Daphnia, exhibiting larger biomass values than Daphnia at Microcystis levels above 2000 mg DW/m3 (Fig. 4b). Cyclopoid and calanoid copepods, groups known to be able to feed selectively and avoid cyanobacteria, showed a less obvious negative relationship with increasing Microcystis biomass (Fig. 4c and d). Phytoplankton Growth and Grazing Losses: Photosynthesis based population growth rates of phytoplankton were generally higher than loss rates from zooplankton grazing at both Lower and Middle bay sites during 2006 and 2007 (Fig. 5). Population growth rates in both years were highest in June and then decreased through the summer, with peak values of 1.0 per day in Lower bay and approximately 0.5–0.6 per day in Middle bay. Grazing losses were typically lower than growth rates. Differences between growth and loss terms were greatest for Lower bay where grazing exceeded growth on only a single date in 2006. In Middle bay grazing was higher than growth rate on a single date in late June 2006, and from mid July–early August 2007.

Fig. 3. Zooplankton community characteristics at five stations (GB1A–GB6) in Green Bay during 2000–2007. Mean (±1 SE) summer values for (a) total zooplankton biomass, (b) areal integrated production rates, (c) dreissenid veliger biomass, and (d) Bythotrephes longimanus biomass. Dates on which no B. longimanus were observed are indicated by zero values.

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

6

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

Table 1 Summer mean zooplankton community grazing rates (mL/L/d) and animal specific (i.e. individual) grazing rates (mL/animal/d) for Lower and Middle bay of Green Bay from time periods prior to invasion and after invasions. Standard error values are shown in parentheses. Means are based on four sampling dates in 1986 (Middle bay), 1987, 1988 and five dates in 1986 (Lower bay), 2006, and 2007. Measure

Time period

Year

Lower bay

Middle bay

Community rates

Pre-invasiona

1986 1987 1988 2006 2007 1986 1987 1988 2006 2007

115.44 120.48 99.6 177.62 91.66 1.44 0.48 0.96 3.41 1.17

88.08 82.08 113.52 249.50 93.59 4.08 1.92 2.64 2.51 1.24

Post-invasionb Individual rates

Pre-invasiona

Post-invasionb a b

(51.84) (78.96) (113.52) (88.89) (30.95) (0.32) (0.24) (0.36) (1.86) (0.48)

(13.92) (28.32) (44.16) (145.67) (17.94) (1.92) (0.48) (1.56) (1.60) (0.64)

Data for pre-invasion period from Sager and Richman (1990). Data for post-invasion period from this study.

During 2006, there were generally inverse relationships between phytoplankton growth rates and grazing losses, with periods of increase in one coinciding with decreases in the other. This also occurred in 2007 at Middle bay and at Lower bay during late summer. Pre-invasion and post-invasion comparisons Water clarity and nutrient concentrations: There was a significant overall effect of time on water clarity, as measured by Secchi depth (F14,69 = 5.303, p b 0.0001; Fig. 6a, b), but Secchi depth was not significantly different at either Lower or Middle bay during 2000–2007 compared to years before the dreissenid invasion in 1993 (LSD post hoc tests, p N 0.05). However, Secchi depths at Lower bay were significantly shallower in years after 1988 when B. longimanus and M. americana invaded Green Bay (LSD post hoc tests, p b 0.001). Prior to the invasions average Secchi depths exceeded 1 m in Lower bay whereas values were ca. 0.5 m during the post-invasion years. Total phosphorus concentrations in Lower bay were generally higher throughout time than in Middle bay, with concentrations peaking during 2001–2003 (Fig. 6c, d). Concentrations in Lower bay varied between 0.11 and 0.18 mg/L in 1986–2000, then exceeded 0.2 mg/L from 2001 to 2003, and decreased down to 0.10 mg/L by 2007. A similar pattern occurred in Middle bay, but at lower concentrations levels overall. Concentrations were 0.03–0.08 in 1986–2000, 0.15–0.20 during 2001–2003, and then 0.1–0.2 in 2004–2007 (Fig. 6c). Total phosphorus concentrations did not show obvious changes during the 1990s, the period of phosphorus loading reductions documented in the literature (Qualls et al., 2013). Nitrate concentrations were more variable than total phosphorus, but did show some of the same trends (Fig. 6e, f). Nitrate levels were typically greater in Lower bay than Middle bay by a factor of at least 1.5–2.0 in most years. Concentrations were highest in 2002–2004, reaching over 0.3 mg/L in Lower bay and ca. 2.0 mg/L in Middle bay. Similar to total phosphorus, concentrations of nitrate decreased after this period. There were noticeable increases in nitrate in 1993, the first year of dreissenid establishment, in both Lower and Middle bay. A similar but less pronounced spike in that year was also observed for total phosphorus (Fig. 6c, d). Phytoplankton Community: Biomass of phytoplankton following the invasions and phosphorus loading reductions was greater than in earlier years (Fig. 7). As we have documented previously, at both Lower and Middle bay locations, summer chl-a concentrations were significantly higher during the post-invasion period (De Stasio et al., 2008, 2014). During 1986–1988, mean chl-a in Lower bay ranged from 15 to 23 mg/m3 whereas in later years levels reached as high as 93 mg/m3. Similarly, Middle bay values increased from 4 to 6 mg/m3 to 17–28 mg/m3. Similar trends were observed for total phytoplankton biomass based on microscope counts (Fig. 7c, d) with ca. 0.5 g DW/m3 in

Fig. 4. Microcystis biomass compared to biomass of a) Daphnia, b) bosminids and chydorids combined, c) cyclopoid copepods and d) calanoid copepods at five stations (GB1A–GB6) in Green Bay for five years after invasions (2000, 2004, 2005, 2006, and 2007).

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

Fig. 5. Mean population rates of phytoplankton photosynthesis (solid lines) and zooplankton community grazing rates (dashed lines) during 2006 (a and c) and 2007 (b and d) for Lower bay (c and d) and Middle bay (a and b). Error bars indicate ±1 SE.

Lower bay during the early time period and concentrations often between 1.0 and 1.9 g DW/m3 in later years. These results are consistent with our previous analyses based on chlorophyll, microscope counts and electronic particle counts demonstrating higher standing stocks of phytoplankton during the post-invasion periods compared to before invasion by zebra mussels (De Stasio et al., 2008, 2014). In addition, those studies documented a clear regime shift in phytoplankton community composition following the invasions, with increased dominance by cyanobacteria, especially Microcystis. Photosynthesis rates were similar between the pre-invasion and post-invasion time periods using data available for representative sites for Lower bay (GB2) and Middle bay (GB6; Fig. 7e, f). There was no significant difference in mean summer photosynthesis rate in Lower bay between the pre-invasion and post-invasion periods (t-test, t = 1.1189, df = 6, p = 0.45). Likewise, photosynthesis rates at Middle bay were not different during the post-invasion period compared to years before zebra mussels arrived (t = 0.003, df = 4, p = 0.998). Zooplankton Community: Zooplankton biomass and composition changed between the time periods examined (Fig. 8). Lower bay

7

zooplankton biomass was highest prior to invasions, reaching close to 4000 mg DW/m3 dry weight in some years. During 2000–2007 biomass was significantly lower, with values approximately 1000 mg DW/m3 or less (Mann-Whitney test, U = 0, p = 0.02). The decrease in total biomass was due to significantly lower biomass of both cyclopoid copepods (U = 1, p = 0.037), as well as bosminid and chydorid cladocerans (U = 0, p = 0.02; Fig. 8b). There were similar and non-significantly different biomasses of Daphnia before and after the invasion (U = 9, p = 0.90), averaging 611 mg DW/m3 (SE = 116.7) across all years. Both D. mendotae and D. retrocurva were observed in all years, but the largerbodied species D. pulicaria was documented only during the postinvasion period. Calanoid copepods comprised a very small proportion of zooplankton biomass in all years studied. In Middle bay there was no significant change in zooplankton biomass between the two time periods, ranging from 1012 to 1875 mg DW/m3 (Fig. 8ab; Mann-Whitney test, U = 10, p = 0.90). Cyclopoid copepod biomass decreased following the invasion (U = 0, p = 0.02), but there were no significant changes in either the combination of bosminids and chydorids (U = 4, p = 0.18) or Daphnia (U = 9, p = 0.90). All three species of Daphnia that occurred in Lower bay also were observed in Middle bay during the post-invasion period. Before invasion, Lower Bay zooplankton biomass was significantly higher than that found in Middle bay (U = 0, p = 0.012) whereas in years after invasion there was no significant difference between Lower and Middle bay (U = 8, p = 0.885). Zooplankton integrated productivity consistently differed between Lower and Middle bay (Fig. 8c, d). Secondary production at Lower bay was significantly lower than at Middle bay (2 way repeated measures ANOVA; location effect, F1,4 = 46.54, p = 0.002). Mean zooplankton production at Lower bay was 999 mg DW/m2/d (SE = 146.0) and was approximately half the mean value observed at Middle bay, 2060 mg DW/m2/d (SE = 170.1). The decreases in zooplankton volumetric biomass observed at Lower bay following the invasions made little difference in the integrated productivity comparison among regions due in large part to the deeper water column depth in Middle bay (GB2 = 3 m, GB6 = 15 m). As a result, there was no significant effect of time period on these productivity relationships (F1,4 = 1.54, p = 0.282; interaction effect, F1,4 = −0.002, p = 1) indicating that overall water column zooplankton production was not affected by the invasions. Grazing by zooplankton did not change appreciably following the invasions (Table 1), despite observed changes in zooplankton community biomass and composition. Community grazing rates ranged between 3.67 and 10.4 mL/L/h across all locations and years examined, but there was not a significant effect of either time period (2 way ANOVA;

Fig. 6. Time series of available data for Middle bay (top panels) and Lower bay (bottom panels) during 1986–2007 for summer mean (±1 SE) estimates for Secchi depth (a and b), total phosphorus (c and d), and nitrate concentrations (e and f).

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

8

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

Fig. 7. Time series of available data for Middle bay (top panels) and Lower bay (bottom panels) during 1986–2007 for summer mean (±1 SE) estimates for chlorophyll a (a and b), total phytoplankton biomass (c and d), and areal integrated photosynthesis rates (e and f). Vertical solid lines indicate time of dreissenid invasion, separating pre-invasion and post-invasion periods. Years for which no data are available are indicated by “ND.”

F1,14 = 1.52, p = 0.24) or location (F1,14 = 0.17, p = 0.68). These rates indicate that generally around 10% of the water is swept clear of particles by the zooplankton community daily, but in 2006 there were slightly higher grazing rates (17–25%). Animal specific grazing rates did not vary significantly with all years and locations combined, with values ranging from 0.48 to 4.0 mL/animal/d. While individual rates prior to invasion tended to be higher in Middle than Lower bay, there was no significant overall effect of location (F1,14 = 0.11, p = 0.74) or time period (F1,14 = 1.50, p = 0.24). Trophic Interactions: Comparing ratios of standing stock of zooplankton to phytoplankton biomass (Z:P) shows that in both Lower and Middle bay there was a trend after invasions to smaller values (Fig. 8e, f). Mean ratios in Lower bay ranged from 0.69 to 0.88 in years prior to the invasions while averages after the invasions varied from 0.13 to 0.44 (Fig. 8f). There were also decreases in ratios at Middle bay, with a range of 0.7–2.44 before the invasions to 0.2–0.99 following

the invasions (Fig. 8e). Variability of Z:P ratios within years were often high, reflecting shifts in both phytoplankton and zooplankton populations during the summer. As in many eutrophic lake systems, phytoplankton composition in Green Bay shifts to blooms of cyanobacteria in late summer and zooplankton composition shifts to groups with smaller body sizes (Richman and Sager, 1990; De Stasio et al., 2014). Accordingly, there was no significant effect of time period on Z:P ratios for either Lower bay (ANOVA on log transformed values; F1,6 = 3.0, p = 0.134) or Middle bay (F1,6 = 3.92, p = 0.09). Middle bay ratios were shown to be significantly higher than Lower bay values during the pre-invasion period (Sager and Richman, 1990) but there was less of a difference in ratios between Lower and Middle bay in years after the invasions. Combining our post-invasion data with previously published data from Sager and Richman (1991) demonstrates that phytoplankton population growth rates adjusted for grazing losses tended to be lower

Fig. 8. Time series of available data for Middle bay (top panels) and Lower bay (bottom panels) during 1986–2007 for summer mean (±1 SE) estimates for zooplankton dry weight biomass and composition of major microcrustacean zooplankton (a and b), zooplankton areal integrated production rates (c and d), and ratios of zooplankton to phytoplankton biomass (e and f). Vertical solid lines indicate time of dreissenid invasion, separating pre-invasion and post-invasion periods. Years for which no data are available are indicated by “ND.”

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

9

following invasions at both Lower and Middle bay (Fig. 9). In Lower bay adjusted population growth rate after invasion was approximately half the pre-invasion mean, decreasing from 0.65/day to 0.33/day, while mean growth rate in Middle bay dropped from 0.06/day to 0.04/day. There was not a significant effect of time on adjusted growth rates due to high variability in both time periods (2 way ANOVA; F1,38 = 3.055, p = 0.089). There was a significant effect of location (F1,38 = 21.53, p = 0.00004), with significantly greater values at Lower than Middle bay before the invasion (Tukey's test, Q = 6.62, p = 0.0003). While not significant, mean adjusted growth rate at Middle bay was smaller than at Lower bay (Tukey's test, Q = 2.75, p = 0.068). Fish communities: Annual index trawl data demonstrate a clear decline in potential planktivory by fish in both Lower and Middle bay between the 1980s compared to later years (Fig. 10). During the 1980s yellow perch were the most abundant planktivore in both areas, with 2000–5000 fish captured per trawl hour. Yellow perch catch decreased to 500–1000 fish/h after 1991, with only a few sporadic years with higher abundances. The two other dominant plantivores, alewive and smelt, were primarily found in Middle bay at abundances of 2000–4000/h (Fig. 10b). These two species also declined during the early 1990s, with catch rates since the late 1990s of b500 fish/h. Although white perch was first collected in 1988 in Lower bay, b100 fish/h were caught until 1992. After that time white perch catch varied between 500 and 7000 fish/h (Fig. 10c). Discussion

Fig. 9. Summer mean values for phytoplankton growth rate at Middle bay (a) and Lower bay (b) showing 14C-based photosysthesis rate (column total), loss rate due to zooplankton community grazing (shaded area) and adjusted phytoplankton population growth (white area) for pre-invasion (1986, 1987, 1988) and post-invasion periods (2006, 2007). Error bars indicate ±1 SE of adjusted phytoplankton population growth rates.

Our comparison of interactions between phytoplankton and zooplankton in the Green Bay ecosystem before and after invasions by Bythotrephes longimanus, Morone americana and Dreissena polymorpha, and phosphorus loading reductions, demonstrate the contextdependent nature of ecosystem changes in response to multiple stressors. In contrast to the expected response, Green Bay phytoplankton biomass during summer was significantly greater and exhibited greater dominance by cyanobacteria during years following invasive

Fig. 10. Time series of fish index trawl catch per effort data (number/trawl hour) at Middle bay (top panels) and Lower bay (bottom panels) during 1980–2010 for yellow perch and white perch (a and c), and alewive and smelt (b and d).

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

10

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

species establishment and nutrient loading reductions compared to the pre-invasion time period. Increases in total phosphorus and nitrate concentrations in the early 2000s help to explain the phytoplankton increases. The changes in phytoplankton biomass and composition occurred despite no significant change in mean integral photosynthesis rates, indicating shifts in the difference between phytoplankton population growth and loss rates. Shifts in seasonality of grazing and zooplankton biomass resulted in decreased summer grazing impact relative to photosynthesis rates. Phytoplankton mean summer population growth rates adjusted for grazing losses did not differ significantly during postinvasion compared to earlier years, but there was a trend towards decreased adjusted growth rates in Lower bay. Shifts towards lower overall phytoplankton growth rates, reduced grazing losses, decreasing individual grazing rates across the summer, and decreased Z:P ratios are consistent with the increased dominance of cyanobacteria, which typically have slower population growth rates and are more resistant to grazing. In addition, significant decreases in zooplankton biomass and shifts in community composition in Lower bay, but not in Middle bay, likely reflect changes in top-down effects from the predatory cladoceran Bythotrephes longimanus, as well as decreased planktivory by fish. Trophic gradient, AOC status and lower food web interactions The lower Green Bay ecosystem has historically been driven strongly by nutrient inputs from the lower Fox River, resulting in obvious gradients from the shallow, well-mixed lower bay to the deeper upper bay (Bertrand et al., 1976). Large inflows of nutrients resulted in degraded ecosystem functions and designation of the lower Fox River and inner Green Bay as an Area of Concern (AOC) by the International Joint Commission (Persson et al., 1988). The dilution of nutrients at increasing distances from the mouth of the Fox River resulted in a strong, welldocumented trophic gradient in terms of total phosphorus, phytoplankton, and zooplankton biomass, all of which decreased along the gradient (Richman et al., 1984). Following invasions and nutrient loading reductions a strong trophic gradient still exists in terms of total phosphorus (Fig. 6), phytoplankton biomass and productivity (Fig. 2; De Stasio et al., 2008, 2014; Yurista et al., 2015). In fact, phytoplankton biomass significantly increased at both Lower and Middle bay locations compared to the earlier period, resulting in a gradient along the bay similar to that which occurred before these events, but at higher biomass levels. These increases in productivity do not correspond to changes in climate patterns in the bay. There have not been any obvious changes in climatic conditions such as temperature, precipitation, or cloud cover between the time periods (http://www.aos.wisc.edu/~sco/clim-history/7cities/green_bay. html#Temperature; accessed on 11 May 2018). Higher standing stock of phytoplankton following the invasions in Green Bay also indicates that a greater amount of phytoplankton production presumably settles out of the water column. This higher flux of material to sediments likely contributes to increased frequency and extent of bottom hypoxia in southern Green Bay (Klump et al., 2009; Hamidi et al., 2015; Labuhn and Klump, 2016). Primary productivity continues to decrease along the gradient as well, with post-invasion Lower bay rates significantly greater than Middle bay estimates (Fig. 2b). Other recent primary productivity studies along the gradient documented similar rates in both Lower bay and Middle bay regions (Althouse et al., 2014; Labuhn and Klump, 2016). These rates are 2–3 times greater than those obtained for other Great Lakes following invasion by dreissenid mussels (Lake Erie: DePew et al., 2006; Ostrom et al., 2005; Lake Huron: Fahnenstiel et al., 1995), demonstrating the continued highly eutrophic condition of lower Green Bay. The persistence and magnification of the trophic gradient means that remediation efforts have not been as effective as anticipated in the AOC, with most beneficial use impairments unresolved (Qualls

et al., 2013). Our results indicate that lower food web function in Green Bay changed following the invasions, and management efforts based on relationships developed using pre-invasion conditions, or data from ecosystems with different lower food web structure, might be less effective at solving water quality impairments. While some recent analyses have supported the use of predictive correlations under certain conditions (e.g. Stow and Cha, 2013), in Green Bay predicting chlorophyll content or harmful algal bloom dominance based on phosphorus loading or concentration will most likely be inaccurate. De Stasio et al. (2014) show that in Lower bay the relationship between total phosphorus and fraction of total phytoplankton composed of cyanobacteria did not follow predictive correlations published by Trimbee and Prepas (1987) or Downing et al. (2001). The contribution of cyanobacteria to total phytoplankton biomass changed following invasions, presumably because of changes in trophic interactions (e.g. Higgins et al., 2010, 2011). Zooplankton biomass no longer follows the same gradient as observed in earlier studies and biomass does not decrease along the gradient (Fig. 3a). Zooplankton biomass significantly decreased following the invasions in Lower bay and remained similar in Middle bay, indicating a greater relative shift in ecosystem function in Lower bay compared to Middle bay (Fig. 8). Integrated water column zooplankton productivity is significantly greater at Middle bay than Lower Bay, as it was before the invasions (Figs. 3, 8). However, this difference between Lower and Middle bay is primarily driven by the greater depth at Middle bay and shifts in species composition, not differences in total zooplankton biomass on a volumetric basis. Overall, the observed increases in phytoplankton standing stock did not result in higher zooplankton abundance or productivity, signaling changes in phytoplankton-zooplankton interactions following the invasions. Responses to invasions likely result from changes in both bottom-up and top-down forces in aquatic food webs (Higgins et al., 2011), and our data from the Green Bay lower food web support a similar conclusion. Bottom-up vs. top-down effects and phytoplankton communities A cornerstone of our understanding of freshwater food webs is that biomass of phytoplankton is typically determined by nutrient loading, especially the amount of phosphorus delivered to a system (Vollenweider, 1968; Schindler et al., 2008). In Green Bay, postinvasion increases in phytoplankton can partly be explained by increased nutrient concentrations (Fig. 6). There have not been increases in nutrient loading from the lower Fox River, the main source of nutrients to the bay; in fact, data indicate there were overall reductions in phosphorus flux at the mouth of the Fox River of about 1.5% per year between 1988 and 2010 (Qualls et al., 2013). In spite of these loading reductions, total phosphorus increased during the time period examined in our study, as did nitrate concentrations in the Lower bay region. It is still unclear why nutrient concentrations in the bay have not decreased following reductions in phosphorus loading to the system (Qualls et al., 2013). Along with increased nutrient concentrations phytoplankton biomass also increased and shifted to a community dominated by cyanobacteria (Fig. 2; De Stasio et al., 2014, 2008), which is consistent with increased concentrations of total phosphorus. Taken together, these trends suggest that nutrient cycling and dynamics likely changed in Lower bay following the invasions. Establishment of dreissenid mussels in other Great Lakes is known to have changed nutrient dynamics, and this helps explain the observed shifts in nutrient concentrations and phytoplankton communities in Green Bay (De Stasio et al., 2014). Shifts to dominance by cyanobacteria have occurred in a number of Great Lakes embayments following invasion by dreissenid mussels. Studies in Saginaw Bay (Bierman et al., 2005; Sayers et al., 2016), Bay of Quinte on Lake Ontario (Nicholls and Carney, 2011; Shimoda et al., 2016), and western Lake Erie (Vanderploeg et al., 2001; Reavie et al., 2014) all documented increased intensity and frequency of cyanobacteria blooms. Some studies have

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

correlated the phytoplankton community shifts with changes in nutrient cycling by zebra mussels (Heath et al., 1995; Gardner et al., 1995). Based on modeling of zebra mussel effects, cyanobacteria likely increased in Saginaw Bay because of increased nutrient recycling, as well as selective rejection of cyanobacteria by mussels (Bierman et al., 2005; Fishman et al., 2009). Such shifts in nutrient recycling, rejection of cyanobacteria, and differential grazing on possible competitors (like small chlorophytes) should lead to phytoplankton communities dominated by cyanobacteria, which generally have slower population growth rates due to lower production per biomass ratios (Reynolds, 1984). Our observations following invasion by zebra mussels in Lower bay show decreases in phytoplankton population growth rates before adjustments for grazing effects, consistent with this idea (Fig. 9). The decreases in population growth rates are apparently offset by increased biomass of phytoplankton (Fig. 7d), resulting in no significant change in integrated primary productivity (Fig. 7f). In spite of shifts to greater dominance by cyanobacteria, 8%–25% of phytoplankton are removed per day by zooplankton communities across the trophic gradient, with interannual variability greater than among-site differences (Table 1). These grazing rates are not significantly different than those measured before the invasions (Sager and Richman, 1990, 1991), and are similar to those recorded for other eutrophic systems in the Great Lakes. Summer microcrustacean zooplankton grazing rates in the range of 6%–15% have been documented in Lake Erie (Wu and Culver, 1991), Saginaw Bay of Lake Huron (Bridgeman et al., 1995), and Lake Champlain (Levine et al., 1999). Effects of dreissenid invasions on grazing in the Great Lakes have varied, with no significant changes observed in Lake Erie, and decreased grazing rates due to decreases in zooplankton biomass in Saginaw Bay. During the post-invasion period in Green Bay large Daphnia occurred early in the season, followed by small cladocerans and cyclopoid copepods in August when the cyanobacterium Microcystis increased in abundance. Zooplankton community grazing rates were higher in June than August (Fig. 5), whereas during the pre-invasion period the highest rates were observed during late summer (Richman et al., 1990). This change is timing of grazing impacts by large-bodied grazers has likely been driven by more intense blooms of inedible cyanobacteria during late summer along with probable decreases in planktivory by fish in early summer (Fig. 10). Higgins et al. (2014) indicate that grazing by large-bodied zooplankton has the potential to alter total ecosystem autotrophic productivity by removing algae and allowing greater light penetration to benthos. However, trophic status of the system is important for determining this interaction, and in eutrophic lakes, similar to Lower bay, effects of grazing by zooplankton are diminished. This is consistent with what is seen in the Green Bay system. Althouse et al. (2014) demonstrate that planktonic production in Lower bay accounts for 95% of total production whereas benthic production increases along the gradient (at least at shallower, nearshore sites) reaching approximately 45% of total production in Middle bay. However, the steep-sided bathymetry at Middle bay limits the overall relative contribution of benthos to total production, resulting in the majority of total production being derived from the plankton. This same bathymetry and stratification, leading to hypolimnetic hypoxia of Middle bay regions (Hamidi et al., 2015), likely limits the influence of grazing by dreissenid mussels on phytoplankton productivity (Cha et al., 2013). Overall, dreissenid mussels in Lower and Middle bay probably contribute little to grazing on phytoplankton in the water column. The only study of mussel densities in the bay documented b1 individual/m2 in the silty, soft bottom Lower bay and an average of 1300 mussels/m2 in shallow nearshore regions of Middle bay (Fettes, 2001). Based on these densities, and filtering rates from Walz (1978), mussels would potentially clear b1% of the water in Lower bay and at most 14% nearshore in Middle bay. As mentioned above, the steep-sided bathymetry in Middle bay would effectively reduce this grazing impact in the Middle bay open water sites such as GB4 and GB6.

11

Zooplankton communities, grazing, and planktivory Zooplankton biomass significantly decreased in Lower bay following the invasions and became more similar to that in Middle bay (Fig. 8). Despite these decreases, the observed mean zooplankton biomass concentrations of approximately 1000 mg/m3 remain higher than those in most embayments on the Great Lakes (Saginaw Bay: 50–100 mg/m3, Pothoven and Höök, 2014; Bay of Qunite: 50–350 mg/m3, Bowen and Johannsson, 2011; Lake Ontario embayments: 200–300 mg/m3, Hall et al., 2003). A few other studies have shown decreased zooplankton biomass following dreissenid mussel invasion, but in many cases multiple invasions and changes in conditions confounded results. Adlerstein et al. (2014) showed decreased zooplankton biomass in Saginaw Bay following zebra mussel invasion, most likely due to a combination of reduced phytoplankton availability and direct predation by zebra mussels. Decreased zooplankton production was recorded in Lake Ontario (Stewart et al., 2010) and in the Bay of Quinte, Lake Ontario, where Johannsson and Bowen (2012) also showed significantly decreased zooplankton production, concluding that changes were driven by multiple factors related to predation by fish and the invertebrate predator Cercopagis (e.g. improved light environment and increased fish abundance), as well as changes in temperature and nutrient levels. Johannsson et al. (2000) demonstrated a similar trend of decreased zooplankton production for Lake Erie. In both cases reduced phytoplankton density was the primary driver leading to improved light climate for predation. This was not the case in Green Bay as mean summer light penetration and water clarity decreased following invasions (Fig. 6; De Stasio et al., 2008, 2014). Given these findings, observed decreases in zooplankton biomass in Green Bay possibly could have been due to degraded feeding conditions, increased planktivory or a combination of these bottom-up and top-down factors. Data from earlier studies demonstrated that changes in zooplankton biomass along the trophic gradient occurred because energy transfer in Lower bay was significantly lower than in other regions. This was due to a Lower bay phytoplankton composition dominated by cyanobacteria that interfered with zooplankton grazing (Sager and Richman, 1990). Reduced standing stocks of zooplankton relative to phytoplankton (Z: P) occurred in Lower bay, especially during late summer. In Middle and Upper bay, larger and more efficient zooplankton grazers previously fed upon a phytoplankton community that was smaller and of higher nutritional quality. This allowed for more efficient transfer of energy into zooplankton, and also supported the conclusion that topdown effects were more important than bottom-up effects in Middle and Upper bay regions (Sager and Richman, 1991). Following the invasions, Lower and Middle bay regions have become more similar in terms of lower food web interactions, with lower proportions of phytoplankton moving into zooplankton (Fig. 8e, f). Phytoplankton biomass and cyanobacteria dominance increased significantly at both locations, and zooplankton biomass decreased significantly at Lower bay, both of which resulted in lower Z:P ratios. Similar negative relationships between Z:P biomass ratios and cyanobacteria dominance are well known and have been documented in a recent comparison across 173 eutrophic lakes (Heathcote et al., 2016); however, in Green Bay two trends are likely driving these changes. Both decreased quality of phytoplankton as food for zooplankton, and increased invertebrate predation on zooplankton seem to have become more important following the invasions. Food quality clearly still affects zooplankton populations in Green Bay. Individual grazing rates in the post-invasion period decline during summer, as cyanobacteria populations become dominant (Fig. 5). Strong negative relationships occur between cladoceran biomass and biomass of Microcystis, the most abundant cyanobacteria (Fig. 4). The small cladocerans (i.e. bosminid and chydorid combined) sustain higher biomass than Daphnia when Microcystis biomass is high (Fig. 4b), consistent with their greater ability to feed during cyanobacteria bloom conditions (Richman and Sager, 1990; Fulton, 1988). Prior to invasions Sager and Richman (1990) showed that individual grazing rates were

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

12

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

significantly higher at Middle bay than Lower bay, indicating more efficient feeding on the phytoplankton community at Middle bay. Our data from years after invasions show that individual grazing rates are more similar, and not significantly different, between the two locations (Table 1). This is consistent with increased cyanobacteria dominance in both areas and greater similarity in phytoplankton communities along the gradient, resulting in reduced food quality and degraded feeding conditions for zooplankton. Reduced resource use efficiency (RUE) is common when cyanobacteria increase in abundance, especially if Microcystis is the dominant group (Filstrup et al., 2014). Decreased RUE is also consistent with the observed lower Z:P biomass ratios and increased cyanobacteria biomass in Green Bay following invasions. Changes in zooplankton community composition following the invasions suggest that predation also may have been affected. Zooplankton shifts were due mostly to decreased biomass of small cladocerans and cyclopoid copepods in Lower Bay (Fig. 8b). This change in small-bodied zooplankton biomass is consistent with increased predation by invertebrate predators, like Bythotrephes and Leptodora kindtii. Following establishment in Green Bay Bythotrephes were more abundant in Lower bay, the same area where small cladocerans decreased (Fig. 3d). Bythotrephes biomass in Lower bay was approximately 10× higher than that reported in Saginaw Bay, Lake Huron (Pothoven and Höök, 2014), indicating a high potential for affecting zooplankton biomass. Lower bay is well mixed and zooplankton do not show signs of diel vertical migration (Richman et al., 1990; B. De Stasio, unpublished data), therefore zooplankton prey in the total water column are accessible to predatory invertebrates like Bythotrephes. Interestingly, no Bythotrephes were observed in 2000 at any of the stations sampled, and in that year zooplankton exhibited a gradient in biomass with highest values in Lower bay, similar to gradients observed prior to invasions. Also, total zooplankton biomass decreases in mid-summer, corresponding to higher Bythotrephes and Leptodora biomass during that time (Merkle and De Stasio, 2018). Bythotrephes are known also to have reduced zooplankton abundance in Lake Huron, in conjunction with reductions in food availability from dreissenid mussels (Vanderploeg et al., 2012). Increases in another invertebrate predator, Leptodora kindtii, has also been associated with decreased abundance of small cladocerans like Bosmina (Branstrator and Lehman, 1991). Small cladocera like Bosmina and Chydorus are able to feed on cyanobacteria more effectively than larger cladocerans (Gliwicz, 1977; Haney, 1973; Fulton, 1988), and their declines would have reduced grazing pressure on cyanobacteria, consistent with the observed increased dominance of cyanobacteria after the invasions. These probable changes in invertebrate planktivory are consistent with shifts in other cladocerans in Green Bay as well. The smaller Daphnia retrocurva drastically decreased in abundance after the invasions while D. galeata and D. pulicaria increased in relative abundance. This is similar to zooplankton community shifts observed in Lake Michigan following invasion by Bythotrephes (Lehman and Caceres, 1993). However, these changes in Daphnia communities are also consistent with decreased predation by visual predators. Reductions in abundance of fish collected in annual trawl sampling in Green Bay indicate probable decreases in planktivory by fish (Fig. 10). Catches of yellow perch, alewife and smelt after 2000 were 10–15% of 1980–1988 abundances. Similar declines in alewife in Saginaw Bay were related to decreased Bosmina longirostris densities (Pothoven et al., 2013), perhaps through release of Bythotrephes from predation and increased invertebrate predation on smaller cladocerans. White perch, M. americana, increased in abundance in Lower bay after 1992, but this species has a more diverse diet than yellow perch, alewife and smelt. They have been shown to specialize on fish eggs and minnows (Schaeffer and Margraf, 1987), but will also select larger predatory zooplankton like B. longimanus in some lakes (Bur and Klarer, 1991). Given these feeding traits it is unlikely that planktivory on zooplankton by white perch offset decreases due to declines of yellow perch, alewive and smelt. Further studies on the interplay of changes in fish and invertebrate

predation and possible cascading effects through food webs would help expand our understanding of ecosystem functioning. Conclusions Invasion of Green Bay by dreissenid mussels, the invertebrate predator Bythotrephes longimanus, and white perch (Morone americana) have likely resulted in significant changes in lower food interactions of this important bay of Lake Michigan. Nutrient concentrations have not decreased following reductions in phosphorus loading to the system, and phytoplankton biomass has increased during the summer period, with increased dominance of Microcystis. These increases in biomass have offset slower population growth rates of cyanobacteria, resulting in no change in integrated primary productivity. Average summer zooplankton community grazing rates have not changed, but seasonality of grazing has shifted and much of the late summer cyanobacteria blooms remains uneaten by zooplankton. Poor feeding conditions for zooplankton exist in both Lower and Middle bay, and the amount of zooplankton relative to phytoplankton biomass has decreased in both areas, indicating lower trophic transfer efficiencies. Zooplankton biomass has decreased in Lower bay, especially for small cladocerans, and is likely due to a combination of poor feeding conditions and increased predation by Bythotrephes. Shifts in both bottom-up and top-down factors have occurred, and Lower and Middle bay regions now are more eutrophic and similar to each other as a result of changes in this important Great Lakes ecosystem. Acknowledgements This work was funded by the University of Wisconsin Sea Grant Institute under grants from the National Sea Grant College Program, National Oceanic and Atmospheric Administration, U.S. Department of Commerce, and from the State of Wisconsin (Federal grant number: NA16RG2257, project number: R/LR-93). Additional funding was provided through the American Philosophical Society Franklin Grant Program (APS FR-2004), and the Excellence in Science Fund from Lawrence University. We thank J. Beyer, R. Boeckman, W. Daniels, T. Haas, T. Hallidayschult, K. Nockleby, B. Pauli, and T. Reed for assistance with sampling and data management. Editorial assistance was kindly provided by E. De Stasio and two anonymous reviewers. Access to previously unpublished data was generously provided by T. R. Harvey, D. Branstrattor, T. Paoli, S. Richman, and P. Sager. References Adlerstein, S., Nalepa, T.F., Vanderploeg, H.A., Fahnenstiel, G.L., 2014. Trends in phytoplankton, zooplankton, and macroinvertebrates in Saginaw Bay relative to zebra mussel (Dreissena polymorpha) colonization: A generalized linear model approach. In: Nalepa, T.F., Schloesser, D.W. (Eds.), Quagga and Zebra Mussels: Biology, Impacts, and Control, 2nd ed. CRC Press, New York, pp. 525–543. Althouse, B., Higgins, S., Vander Zanden, M.J., 2014. Benthic and planktonic primary production along a nutrient gradient in Green Bay, Lake Michigan, USA. Freshw. Sci. 33, 487–498. Arnott, D.L., Vanni, M.J., 1996. Nitrogen and phosphorus recycling by the zebra mussel (Dreissena polymorpha) in the western basin of Lake Erie. Can. J. Fish. Aquat. Sci. 53, 646–659. Balcer, M.D., Korda, N.L., Dodson, S.I., 1984. Zooplankton of the Great Lakes: A Guide to the Identification and Ecology of the Common Crustacean Species. The University of Wisconsin Press, Madison. Barbiero, R.P., Tuchman, M.L., 2004. Changes in the crustacean communities of Lakes Michigan, Huron, and Erie following the invasion of the predatory cladoceran Bythotrephes longimanus. Can. J. Fish. Aquat. Sci. 61, 2111–2125. Barbiero, R.P., Rockwell, D.C., Warren, G.J., Tuchman, M.L., 2006. Changes in spring phytoplankton communities and nutrient dynamics in the eastern basin of Lake Erie since the invasion of Dreissena spp. Can. J. Fish. Aquat. Sci. 63, 1549–1563. Bertrand, G., Lang, J., Ross, J., 1976. The Green Bay Watershed: Past/Present/Future. University of Wisconsin Sea Grant College Program, Madison, WI. Bierman, V.J., Kaur, J., Depinto, J.V., Feist, T.J., Dilks, D.W., 2005. Modeling the role of zebra mussels in the proliferation of blue-green algae in Saginaw Bay, Lake Huron. J. Great Lakes Res. 31, 32–55. Bottrell, H.H., Duncan, A., Gliwicz, Z.M., Grygierek, E., Herzig, A., Hillbrichtilkowska, A., Kurosawa, H., Larsson, P., Weglenska, T., 1976. Review of some problems in zooplankton production studies. Nor. J. Zool. 24, 419–456. Bowen, K.L., Johannsson, O.E., 2011. Changes in zooplankton biomass in the Bay of Quinte with the arrival of the mussels, Dreissena polymorpha and D. rostiformis bugensis, and

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx the predatory cladoceran, Cercopagis pengoi: 1975 to 2008. Aquat. Ecosyst. Health Manag. 14, 44–55. Branstrator, D.K., Lehman, J.T., 1991. Invertebrate predation in Lake Michigan: regulation of Bosmina longirostris by Leptodora kindtii. Limnol. Oceanogr. 36, 483–495. Bridgeman, T.B., Fahnenstiel, G.L., Lang, G.A., Nalepa, T.F., 1995. Zooplankton grazing during the zebra mussel (Dreissena polymorpha) colonization of Saginaw Bay, Lake Huron. J. Great Lakes Res. 21, 567–573. Bur, M.T., Klarer, D.M., 1991. Prey selection for the exotic cladoceran Bythotrephes cederstroemi by selected Lake Erie fishes. J. Great Lakes Res. 17, 85–93. Cha, Y., Stow, C.A., Bernhardt, E.S., 2013. Impacts of dreissenid mussel invasions on chlorophyll and total phosphorus in 25 lakes in the USA. Freshw. Biol. 58, 192–206. Cochran, P.A., Hesse, P.J., 1994. Observations on the white perch (Morone americana) early in its invasion of Wisconsin. Trans. Wis. Acad. Sci. Arts Lett. 82, 23–58. Cotner, J.B., Weinke, A.D., Biddanda, B.A., 2017. Great Lakes: science can keep them great. J. Great Lakes Res. 43, 916–919. De Stasio, B.T., Richman, S., 1998. Phytoplankton spatial and temporal distributions in Green Bay, Lake Michigan, prior to colonization by the zebra mussel (Dreissena polymorpha). J. Great Lakes Res. 24, 620–628. De Stasio, B.T., Schrimpf, M.B., Beranek, A.E., Daniels, W.C., 2008. Increased chlorophyll a, phytoplankton abundance, and cyanobacteria occurrence following invasion of Green Bay, Lake Michigan by dreissenid mussels. Aquat. Invasions 3, 21–27. De Stasio, B., Schrimpf, M., Cornwell, B., 2014. Phytoplankton communities in Green Bay, Lake Michigan after invasion by dreissenid mussels: increased dominance by cyanobacteria. Diversity 6, 681–704. Depew, D.C., Guildford, S.J., Smith, R.E.H., 2006. Nearshore-offshore comparison of chlorophyll a and phytoplankton production in the dreissenid-colonized eastern basin of Lake Erie. Can. J. Fish. Aquat. Sci. 63, 1115–1129. Downing, J.A., Watson, S.B., McCauley, E., 2001. Predicting cyanobacteria dominance in lakes. Can. J. Fish. Aquat. Sci. 58, 1905–1908. Egan, D., 2017. The Death and Life of the Great Lakes. W.W. Norton and Company, Inc., New York, NY. Fahnenstiel, G.L., Bridgeman, T.B., Lang, G.A., McCormick, M.J., Nalepa, T.F., 1995. Phytoplankton productivity in Saginaw Bay, Lake Huron: effects of zebra mussel (Dreissena polymorpha) colonization. J. Great Lakes Res. 21, 465–475. Fahnenstiel, G.L., Sayers, M.J., Shuchman, R.A., Yousef, F., Pothoven, S.A., 2016. Lake-wide phytoplankton production and abundance in the Upper Great Lakes: 2010–2013. J. Great Lakes Res. 42, 619–629. Fee, E.J., 1998. Computer Programs for Calculating In Situ Phytoplankton Photosynthesis; Canadian Technical Report of Fisheries and Aquatic Sciences; No. 1740; Published 26 July 1990, revised (WWW version) February 1998; Catalog Number 97–6/1740E ISSN 0706—6457. Fettes, A.L., 2001. The Bottom of the Bay Then and Now: Investigations Into Both the Zebra Mussels Colonizing Green Bay and the Pre-invasion Macroinvertebrate Community. (MS thesis). University of Wisconsin, Green Bay. Filstrup, C.T., Hillebrand, H., Heathcote, A.J., Harpole, W.S., Downing, J.A., 2014. Cyanobacteria dominance influences resource use efficiency and community turnover in phytoplankton and zooplankton communities. Ecol. Lett. 17, 464–474. Fishman, D.B., Adlerstein, S.A., Vanderploeg, H.A., Fahnenstiel, G.L., Scavia, D., 2009. Causes of phytoplankton changes in Saginaw Bay, Lake Huron, during the zebra mussel invasion. J. Great Lakes Res. 35, 482–495. Fishman, D.B., Adlerstein, S.A., Vanderploeg, H.A., Fahnenstiel, G.L., Scavia, D., 2010. Phytoplankton community composition of Saginaw Bay, Lake Huron, during the zebra mussel (Dreissena polymorpha) invasion: a multivariate analysis. J. Great Lakes Res. 36, 9–19. Fulton, R.S., 1988. Resistance to blue-green algal toxins by Bosmina longirostris. J. Plankton Res. 10, 771–778. Gardner, W.S., Cavaletto, J.F., Johengen, T.H., Johnson, J.R., Heath, R.T., Cotner, J.B., 1995. Effects of the zebra mussel, Dreissena polymorpha, on community nitrogen dynamics in Saginaw Bay, Lake Huron. J. Great Lakes Res. 21, 529–544. Gliwicz, Z.M., 1977. Food size selection and seasonal succession of filter feeding zooplankton in an eutrophic lake. Ekol. Polska A 25, 179–225. Hall, S.R., Schneider, C.P., Lary, S.J., Arrhenius, F., Pauliukonis, N.K., Mills, E.L., Rudstam, L.G., 2003. A comparison of total phosphorus, chlorophyll a, and zooplankton in embayment, nearshore, and offshore habitats of Lake Ontario. J. Great Lakes Res. 29, 54–69. Hamidi, S.A., Bravo, H.R., Klump, J.V., Waples, J.T., 2015. The role of circulation and heat fluxes in the formation of stratification leading to hypoxia in Green Bay, Lake Michigan. J. Great Lakes Res. 41, 1024–1036. Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: paleontological statistics software package for education and data analysis. Palaeontol. Electron. 4, 1–9. Haney, J.F., 1973. An in situ examination of the grazing activities of natural zooplankton communities. Arch. Hydrobiol. Beih. Ergeb. Limnol. 72, 87–132. Heath, R.T., Fahnenstiel, G.L., Gardner, W.S., Cavaletto, J.F., Hwang, S.J., 1995. Ecosystemlevel effects of zebra mussels (Dreissena polymorpha): an enclosure experiment in Saginaw Bay, Lake Huron. J. Great Lakes Res. 21, 501–516. Heathcote, A.J., Filstrup, C.T., Kendall, D., Downing, J.A., 2016. Biomass pyramids in lake plankton: influence of cyanobacteria size and abundance. Inland Waters 6, 250–257. Higgins, S.N., Vander Zanden, M.J., 2010. What a difference a species makes: a meta-analysis of dreissenid mussel impacts on freshwater ecosystems. Ecol. Monogr. 80, 179–196. Higgins, S.N., Vander Zanden, M.J., Joppa, L.N., Vadeboncoeur, Y., 2011. The effect of dreissenid invasions on chlorophyll and the chlorophyll: total phosphorus ratio in north-temperate lakes. Can. J. Fish. Aquat. Sci. 68, 319–329. Higgins, S.N., Althouse, B., Devlin, S.P., Vadeboncoeur, Y., Vander Zanden, M.J., 2014. Potential for large-bodied zooplankton and dreissenids to alter the productivity and autotrophic structure of lakes. Ecology 95, 2257–2267. Jin, E.H., Sprules, W.G., 1990. Distribution and abundance of Bythotrephes cederstroemii (Cladocera: Cercopagidae) in the St. Lawrence Great Lakes. Verh. Internat. Verein. Limnol. 24, 383–385.

13

Johannsson, O.E., Bowen, K.L., 2012. Zooplankton production in the Bay of Quinte 1975–2008: relationships with primary production, habitat, planktivory, and aquatic invasive species (Dreissena spp. and Cercopagis pengoi). Can. J. Fish. Aquat. Sci. 69, 2046–2063. Johannsson, O.E., Dermott, R., Graham, D.M., Dahl, J.A., Millard, E.S., Myles, D.D., Leblanc, J., 2000. Benthic and pelagic secondary production in Lake Erie after the invasion of Dreissena spp. with implications for fish production. J. Great Lakes Res. 26, 31–54. Klump, J.V., Fitzgerald, S.A., Waples, J.T., 2009. Benthic biogeochemical cycling, nutrient stoichiometry, and carbon and nitrogen mass balances in a eutrophic freshwater bay. Limnol. Oceanogr. 54, 692–712. Kraft, C., 1993. Zebra Mussel Update #18. University of Wisconsin Sea Grant Advisory Services, Green Bay. Labuhn, S., Klump, J.V., 2016. Estimating summertime epilimnetic primary production via in situ monitoring in an eutrophic freshwater embayment, Green Bay, Lake Michigan. J. Great Lakes Res. 42, 1026–1035. Lampert, W., Taylor, B.E., 1985. Zooplankton grazing in a eutrophic lake - implications of diel vertical migration. Ecology 66, 68–82. Lavrentyev, P.J., Gardner, W.S., Cavaletto, J.F., Beaver, J.R., 1995. Effects of the zebra mussel (Dreissena polymorpha Pallas) on protozoa and phytoplankton from Saginaw Bay, Lake Huron. J. Great Lakes Res. 21, 545–557. Legendre, L., Demers, S., Yentsch, C.M., Yentsch, C.S., 1983. The 14C method: patterns of dark CO2 fixation and DCMU correction to replace the dark bottle. Limnol. Oceanogr. 28, 996–1003. Lehman, J.T., Caceres, C.E., 1993. Food-web responses to species invasion by a predatory invertebrate: Bythotrephes in Lake Michigan. Limnol. Oceanogr. 38, 879–891. Levine, S.N., Borchardt, M.A., Braner, M., Shambaugh, A.D., 1999. The impact of zooplankton grazing on phytoplankton species composition and biomass in Lake Champlain (USA-Canada). J. Great Lakes Res. 25, 61–77. Lind, O.T., 1985. Handbook of Common Methods in Limnology. 2nd ed. Kendall/Hunt Publishing Co., Dubuque, IA. Merkle, C.A., De Stasio, B.T., 2018. Bythotrephes longimanus in shallow, nearshore waters: interactions with Leptodora kindtii, impacts on zooplankton, and implications for secondary dispersal from southern Green Bay, Lake Michigan. J. Great Lakes Res. https:// doi.org/10.1016/j.jglr.2018.02.013. Millard, E.S., Sager, P.E., 1994. Comparison of phosphorus, light climate, and photosynthesis between two culturally eutrophied bays: Green Bay, Lake Michigan, and the Bay of Quinte, Lake Ontario. Can. J. Fish. Aquat. Sci. 51, 2579–2590. Nicholls, K.H., Carney, E.C., 2011. The phytoplankton of the Bay of Quinte, 1972–2008: point-source phosphorus loading control, dreissenid mussel establishment, and a proposed community reference. Aquat. Ecosyst. Health Manag. 14, 33–43. Ostrom, N.E., Russ, M.E., Field, A., Piwinski, L., Twiss, M.R., Carrick, H.J., 2005. Ratios of community respiration to photosynthesis and rates of primary production in Lake Erie via oxygen isotope techniques. J. Great Lakes Res. 31, 138–153. Padilla, D.K., Adolph, S.C., Cottingham, K.L., Schneider, D.W., 1996. Predicting the consequences of dreissenid mussels on a pelagic food web. Ecol. Model. 85, 129–144. Pennak, R.W., 1989. Fresh-Water Invertebrates of the United States: Protozoa to Mollusca. 3rd ed. John Wiley and Sons, Inc., New York, NY. Persson, L., Harris, V., Lukas, C., Christie, J., Harris, H.J., Meyers, L., Sullivan, J., Allen, P., Baba, R., 1988. Lower Green Bay Remedial Action Plan for the Lower Fox River and Lower Green Bay Area of Concern. Wisconsin Department of Natural Resources, Madison, WI, p. 319. Pothoven, S.A., Höök, T.O., 2014. Predatory demands of Bythotrephes and Leptodora in Saginaw Bay, Lake Huron. J. Great Lakes Res. 40, 106–112. Pothoven, S.A., Höök, T.O., Nalepa, T.F., Thomas, M.V., Dyble, J., 2013. Changes in zooplankton community structure associated with the disappearance of invasive alewife in Saginaw Bay, Lake Huron. Aquat. Ecol. 47, 1–12. Qualls, T.M., Dolan, D.M., Reed, T., Zorn, M.E., Kennedy, J., 2007. Analysis of the impacts of the zebra mussel, Dreissena polymorpha, on nutrients, water clarity, and the chlorophyll-phosphorus relationship in lower Green Bay. J. Great Lakes Res. 33 (3), 617–626. Qualls, T., Harris, H.J., Harris, V., 2013. The State of the Bay: The Condition of the Bay of Green Bay/Lake Michigan 2013. University of Wisconsin Sea Grant Institute, Madison, WI. Reavie, E.D., Barbiero, R.P., Allinger, L.E., Warren, G.J., 2014. Phytoplankton trends in the Great Lakes, 2001–2011. J. Great Lakes Res. 40, 618–639. Reynolds, C.S., 1984. The Ecology of Freshwater Phytoplankton. Cambridge. Richman, S., Sager, P.E., 1990. Patterns of phytoplankton-zooplankton interaction along a trophic gradient: II. Biomass and size distribution. Verh. Internat. Verein. Limnol. 24, 401–405. Richman, S., Sager, P.E., Banta, G., Harvey, T.R., De Stasio, B.T., 1984. Phytoplankton standing stock, size distribution, species composition and productivity along a trophic gradient in Green Bay, Lake Michigan. Verh. Intern. Verein. Limnol. 22, 460–469. Richman, S., Branstrator, D.K., Huber-Villegas, M., 1990. Impact of zooplankton grazing on phytoplankton along a trophic gradient. In: Tilzer, M.M., Serruya, C. (Eds.), Large Lakes: Ecological Structure and Function. Springer-Verlag, Berlin, pp. 592–614. Rocha, O., Duncan, A., 1985. The relationship between cell carbon and cell volume in freshwater algal species used in zooplanktonic studies. J. Plankton Res. 7, 279–294. Sager, P.E., Richman, S., 1990. Patterns of phytoplankton-zooplankton interactions along a trophic gradient: I. Production and utilization. Verh. Internat. Verein. Limnol. 24, 393–396. Sager, P.E., Richman, S., 1991. Functional interaction of phytoplankton and zooplankton along the trophic gradient in Green Bay, Lake Michigan. Can. J. Fish. Aquat. Sci. 48, 116–122. Sarnelle, O., Raikow, D.F., Wilson, A.E., Hamilton, S.K., Knoll, L.B., 2005. Complex interactions between the zebra mussel, Dreissena polymorpha, and the harmful phytoplankter, Microcystis aeruginosa. Limnol. Oceanogr. 50, 896–904.

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020

14

B.T. De Stasio et al. / Journal of Great Lakes Research xxx (2018) xxx–xxx

Sayers, M., Fahnenstiel, G.L., Shuchman, R.A., Whitley, M., 2016. Cyanobacteria blooms in three eutrophic basins of the Great Lakes: a comparative analysis using satellite remote sensing. Int. J. Remote Sens. 37, 4148–4171. Schaeffer, J.S., Margraf, F.J., 1987. Predation on fish eggs by white perch, Morone americana, in western Lake Erie. Environ. Biol. Fish 18, 77–80. Schindler, D.W., Vallentyne, J.R., 2008. The Algal Bowl. The University of Alberta Press, Edmonton, Canada. Schindler, D.W., Hecky, R.E., Findlay, D.L., Stainton, M.P., Parker, B.R., Paterson, M.J., Beaty, K.G., Lyng, M., Kasian, S.E.M., 2008. Eutrophication of lakes cannot be controlled by reducing nitrogen input: results of a 37-year whole-ecosystem experiment. Proc. Natl. Acad. Sci. U. S. A. 105, 11254–11258. Shimoda, Y., Watson, S.B., Palmer, M.E., Koops, M.A., Mugalingam, S., Morley, A., Arhonditsis, G.B., 2016. Delineation of the role of nutrient variability and dreissenids (Mollusca, Bivalvia) on phytoplankton dynamics in the Bay of Quinte, Ontario, Canada. Harmful Algae 55, 121–136. Shuter, B.J., Ing, K.K., 1997. Factors affecting the production of zooplankton in lakes. Can. J. Fish. Aquat. Sci. 54, 359–377. Smith, D., Magnuson, J.J., 1990. Biomas size structure and energy transfer efficiency across the productivity gradient in a freshwater estuary - Green Bay, Lake Michigan. Verh. Internat. Verein. Limnol. 24, 405–410. Stewart, T.J., Johannsson, O.E., Holeck, K., Sprules, W.G., O'Gorman, R., 2010. The Lake Ontario zooplankton community before (1987–1991) and after (2001–2005) invasioninduced ecosystem change. J. Great Lakes Res. 36, 596–605. Stow, C.A., 2014. The news from Saginaw Bay: where the mussels are strong, the walleye are good-looking, and all the phosphorus is above average. J. Great Lakes Res. 40, 1–3. Stow, C.A., Cha, Y., 2013. Are chlorophyll a-total phosphorus correlations useful for inference and prediction? Environ. Sci. Technol. 47, 3768–3773. Strayer, D.L., 2009. Twenty years of zebra mussels: lessons from the mollusk that made headlines. Front. Ecol. Environ. 7, 135–141. Trimbee, A.M., Prepas, E.E., 1987. Evaluation of total phosphorus as a predictor of the relative biomass of blue-green-algae with emphasis on Alberta lakes. Can. J. Fish. Aquat. Sci. 44, 1337–1342.

Vanderploeg, H.A., Johengen, T.H., Fahnenstiel, G.L., Nalepa, T.F., Liebig, J.R., Carmichael, W.W., Agy, M.A., 2001. Zebra mussel (Dreissena polymorpha) selective filtration promoted toxic Microcystis blooms in Saginaw Bay (Lake Huron) and Lake Erie. Can. J. Fish. Aquat. Sci. 58, 1208–1221. Vanderploeg, H.A., Nalepa, T.F., Jude, D.J., Mills, E.L., Holeck, K.T., Liebig, J.R., Grigorovich, I.A., Ojaveer, H., 2002. Dispersal and emerging ecological impacts of Ponto-Caspian species in the Laurentian Great Lakes. Can. J. Fish. Aquat. Sci. 59, 1209. Vanderploeg, H.A., Pothoven, S.A., Fahnenstiel, G.L., Cavaletto, J.F., Liebig, J.R., Stow, C.A., Nalepa, T.F., Madenjian, C.P., Bunnell, D.B., 2012. Seasonal zooplankton dynamics in Lake Michigan: disentangling impacts of resource limitation, ecosystem engineering, and predation during a critical ecosystem transition. J. Great Lakes Res. 38, 336–352. Vanderploeg, H.A., Wilson, A.E., Johengen, T.H., Bressie, J.D., Sarnelle, O., Liebig, J.R., Robinson, S.D., Horst, G.P., 2014. Role of selective grazing by dreissenid mussels in promoting toxic Microcystis blooms and other changes in phytoplankton composition in the Great Lakes. In: Nalepa, T.F., Schloesser, D.W. (Eds.), Quagga and Zebra Mussels: Biology, Impacts, and Control, 2nd ed. CRC Press, New York, pp. 509–523. Vanderploeg, H.A., Bunnell, D.B., Carrick, H.J., Höök, T.O., 2015. Complex interactions in Lake Michigan's rapidly changing ecosystem. J. Great Lakes Res. 41 (Supplement 3), 1–6. Vollenweider, R.A., 1968. Scientific Fundamentals of the Eutrophication of Lakes and Flowing Waters, with Particular Reference to Nitrogen and Phosphorus as Factors in Eutrophication. OECD Technical Report, DAS/CS/68.27. Walz, N., 1978. The energy balance of the freshwater mussel Dreissena polymorpha Pallas in laboratory experiments and in Lake Constance. I. Pattern of activity, feeding and assimilation efficiency. Arch. Hydrobiol. Suppl. 55, 83–105. Wetzel, R.G., Likens, G.E., 1991. Limnological Analyses. 2nd ed. Springer-Verlag, New York. Wu, L., Culver, D.A., 1991. Zooplankton grazing and phytoplankton abundance: an assessment before and after invasion of Dreissena polymorpha. J. Great Lakes Res. 17, 425–436. Yurista, P.M., Kelly, J.R., Cotter, A.M., Miller, S.E., Van Alstine, J.D., 2015. Green Bay: spatial variation in water quality, and landscape correlations. J. Great Lakes Res. 41, 560–572.

Please cite this article as: De Stasio, B.T., et al., Zooplankton-phytoplankton interactions in Green Bay, Lake Michigan: Lower food web responses to biological invasions, J. Great Lakes Res. (2018), https://doi.org/10.1016/j.jglr.2018.05.020