PROJECT QUINTE ANNUAL REPORT 2014 Monitoring Report

PROJECT QUINTE ANNUAL REPORT 2014

Bay of Quinte RAP Restoration Council / Project Quinte

Monitoring Report #25 Summer 2016 BAY OF QUINTE REMEDIAL ACTION PLAN MONITORING REPORT #25

PROJECT QUINTE ANNUAL REPORT 2014

prepared by

Project Quinte members

in support of the Bay of Quinte Remedial Action Plan

Bay of Quinte Remedial Action Plan Kingston, Ontario, Canada.

Summer 2016

Editors Note: This report does not constitute publication. Many of the results are preliminary findings. The information has been provided to assist and guide the Bay of Quinte Remedial Action Plan. The information and findings cannot be used in any manner or quoted without the consent of the individual authors. Individual authors should be contacted prior to any other proposed application of the data herein. CONTENTS

Preface 4

Introduction 6

Water temperature monitoring in the Bay of Quinte, 2014 9 S.M. Larocque and S.E. Doka

Point source phosphorus loadings: 1965 to 2014 21 P. Kinstler and A. Morley

Phytoplankton and microbial food web interactions at a long-term monitoring station in the Bay of Quinte: Belleville, 2014 25 M. Munawar, R. Rozon, M. Fitzpatrick and H. Niblock

Zooplankton in the Bay of Quinte – 2014 39 K.L. Bowen, R. Rozon and W.J.S. Currie

Fish populations in the Bay of Quinte, 2014 57 J. A. Hoyle PREFACE

BAY OF QUINTE REMEDIAL ACTION PLAN MONITORING REPORT #25

2014 PROJECT QUINTE ANNUAL REPORT (Summer 2016)

In 1985, the Great Lakes Water Quality Board of the International Joint Commission (IJC) identified 42 Areas of Concern in the Great Lakes basin where the beneficial uses were impaired. The Board recommended that the appropriate jurisdictions and government agencies prepare, submit and implement a Remedial Action Plan (RAP) in each area to restore the water uses. The Bay of Quinte was designated as one of the Areas of concern. Ten of 14 beneficial uses described in Annex 2 of the Great lakes Water Quality Agreement (revised 1987) are impaired. The impaired uses include beach postings, eutrophication or undesirable algae, restrictions on fish consumption, taste and odour problems in drinking water, etc. The contributing factors are excessive phosphorus loadings, persistent toxic contaminants, bacteriological contamination, as well as alterations and destruction of shorelines, wetland and fish habitat. Project Quinte is a long-term, multi-agency research and monitoring project. The project’s original objectives included studying, comparing, and evaluating the Bay of Quinte limnological attributes (biological, chemical and physical) before and after phosphorus control was implemented at municipal sewage treatment plants thus reducing phosphorus loads to the bay. Project Quinte was launched over 40 years ago in 1972 and is still operating (Minns et al., 2011, 2012). The time between 1972 and 1977 has been referred to as the “pre-phosphorus control” period, while post-1977 is called “post-phosphorus control” period. Changes in water quality, as well as aquatic communities (e.g. phytoplankton, zooplankton, benthic and fish) between pre- and post-control periods have been compared. Phosphorus control strategies were assessed. Finally, long-term ecosystem responses within the Bay of Quinte were examined and described. Another focus of Project Quinte is to assess changes in lower trophic level productivity in response to the invasion of the bay by Zebra Mussels that began in 1994. Since 2000, microbial loop assessment, consisting of the enumeration of bacteria, autotrophic picoplankton and ciliates, has been included in the program. Also, size fractionated primary productivity was included in the monitoring to assess the contributions of various size fractions of phytoplankton. The response to invasion of the bay by both Zebra Mussels and other non-native species of zooplankton and fish and how changes are transferred up the food chain is currently being addressed. Clearly, managing both the fishery as well as phosphorus loadings during alterations in the food chain presents a challenge that requires the yearly productivity assessment provided by Project Quinte. This report is the 25th in the Bay of Quinte RAP monitoring report series. It is a window looking into the status of the Bay by providing information about projects and results currently being conducted

4 in the Bay in support of the Bay of Quinte Remedial Action Plan. The report presents preliminary findings and data. It does not necessarily represent the views or policies of the sponsoring agencies. The information and data contained in the report, more or less, serves as background reference material. The information has been compiled and reported so that:

1. the Bay of Quinte RAP Restoration Council and QWC can formulate abatement and remedial options, 2. remedial actions can be monitored and progress reported, and 3. researchers can be informed about the current status of the health of the Bay of Quinte

The report has been prepared as part of the Bay of Quinte RAP under the auspices of the Cana- da-Ontario Great Lakes Remedial Action Plan. Financial assistance and technical sponsorship for the investigations and research was provided by Fisheries and Oceans Canada, Environment Canada, the Ontario Ministry of the Environment and the Ontario Ministry of Natural Resources. Through this preface the Bay of Quinte RAP Restoration Council notifies potential users that this report does not constitute publication.

References

Minns, C.K., Munawar, M., Koops, M.A., Millard, E.S., 2011. Long-term ecosystem studies in the Bay of Quinte, Lake Ontario, 1972- 2008. A prospectus. Aquat. Ecosyst. Health Mgmt. 14(1), 3-8. Minns, C.K., Munawar, M., Koops, M.A., 2012. Preface. Aquat. Ecosyst. Health Mgmt. 14(1), 369.

Bay of Quinte Information

Symposium “Ecosystem Health and Recovery of the Bay of Quinte, Lake Ontario: Past, Present and Future”, May 9, 2010. Organized by: Aquatic Eco- system Health & Management Society as a special session at the International Association for Great Lakes Research’s 53rd annual

conference on Great Lakes Research. Toronto, Ontario.

Primary Publications

Besides the series of annual reports, of which this is the 25th, the following publications have been printed on the Bay of Quinte: Minns, C.K., Hurley, D.A., Nicholls, K.H. (Eds.), 1986. Special publication: “Project Quinte: Point-Source Phosphorus Control and Eco- system Response in the Bay of Quinte, Lake Ontario”. Canadian Journal of Fisheries and Aquatic Sciences, 86. Aquatic Ecosystem Health & Management Society, 2011. Special issue: “Ecosystem Health and Recovery of the Bay of Quinte, Lake Ontario”. Aquat. Ecosyst. Health Mgmt. 14(1). Aquatic Ecosystem Health & Management Society, 2012. Special issue: “Ecosystem Health and Recovery of the Bay of Quinte, Lake Ontario: Part II”. Aquat. Ecosyst. Health Mgmt. 15(4).

5 INTRODUCTION

Project Quinte is a co-operative, multi-agency, research and monitoring project between the federal (De- partment of Fisheries and Oceans) and provincial governments (Ontario Ministry of Natural Resources, Ontario Ministry of the Environment) that has investigated the long-term effects of the reduction in point-source phosphorus (P) loadings (Minns et al., 1986), food-chain influences (Nicholls and Hurley, 1989), and more recently Zebra Mussel colonization (RAP monitoring reports Nos.7-12) on trophic dynamics of the entire bay ecosystem. The Bay of Quinte is one of 42 severely impaired ecosystems (Area of Concern - AOC) on the Great Lakes. Annex 2 of the Revised Great Lakes Water Quality Agreement of 1978 outlined a three stage Remedial Action Plan process that called for the identification of impaired beneficial uses (Stage I), their causes and a plan to be implemented (stage II) to restore these uses. The third and final stage of the RAP process requires monitoring to measure the success of RAP implementation that should ultimately lead to delisting the Bay of Quinte as an AOC. Project Quinte has been invaluable to stage three of the Remedial Action Plan (RAP), supporting an essential, continuing program of research and monitoring in the bay. Project Quinte has contributed long-term data that was used to produce comprehensive assessments of the status of two impaired beneficial uses, phytoplankton (Nicholls, Monitoring Report #11) and zooplankton (Johannsson and Nicholls, Monitoring Report #12). Project Quinte presented the first evidence of the impact of Zebra Mussels on the Bay of Quinte ecosystem in the 1995 monitoring report. Since that time we have observed variable impact, both spatially and inter-annually, of this invader on physical, chemical and biological properties of the bay (e.g. transparency, phosphorus, and lower trophic level biomass and production). There are no other Great Lakes studies that have the benefit of the extensive multi-trophic level database that exists for the Bay of Quinte to determine ecosystem level impacts of Dreissenid Mussels. We stand to learn a great deal about the impacts of this exotic species on fisheries productivity in the Bay of Quinte that may be of use in other ecosystems where background data is not as extensive. The scope of the work in Project Quinte is rare in freshwater ecosystem research, now encompass- ing multi-year, multi-trophic level data from bacteria to fish. Long-term data sets for biomass, species composition and production for all trophic levels provide a unique opportunity to model trophic interactions in the bay and determine how various factors are impacting important fish populations. This report provides a detailed summary of the results and highlights for the 2014 field season, prepared by various scientists studying the bay. Please contact these individuals directly if further ex- planation or detail is required.

Link to the Remedial Action Plan

Progress towards restoration goals must be measured in order to delist impaired beneficial uses. The

6 ongoing monitoring provided by Project Quinte is invaluable in assessing restoration progress as the federal government attempts to delist this AOC. The restoration goals originally set as part of the stage II development were considered to be the interim. Project Quinte has been instrumental in developing more realistic goals for phytoplankton, zooplankton, benthos and other related beneficial uses. Infor- mation on listing criteria and delisting goals has been summarized in the report, “Bay of Quinte RAP Monitoring and Delisting Strategy IBU Assessment Statements 2003” prepared for the Bay of Quinte Restoration Council under contract to Murray German Consulting and Fred Stride Environmental.

WATER TEMPERATURE MONITORING IN THE BAY OF QUINTE, 2014

S.M. Larocque and S.E. Doka

Fish Habitat Science Section, Great Lakes Laboratory for Fisheries and Aquatic Sciences Fisheries and Oceans Canada 867 Lakeshore Road Burlington, Ontario, L7S 1A1

Since 2001, Great Lakes Laboratory for Fisheries and Aquatic Sciences (GLLFAS) has monitored water temperatures in a maximum of 16 locations in the Bay of Quinte area (Leisti and Doka, 2007). In 2014, data was collected at the same six offshore (i.e. mid-channel) and three nearshore stations monitored in 2013 (Figure 1). Offshore stations have been grouped into Upper Bay (Belleville and Napanee), Middle Bay (Hay Bay and Glenora) and Lower Bay (Conway and Lennox) sites. At offshore stations, temperature was monitored throughout the water column at set logger depths. Nearshore stations were in shallow locations to capture coastal temperature dynamics.

Figure 1. Temperature logger locations in the Bay of Quinte in 2014.

At all offshore stations, temperature loggers were fixed onto spar buoys at approximately 1-m below the water surface (Figure 2); the buoys provided some shading for the logger so temperatures should be reflective of the ambient conditions at this depth without being subject to UV heating. For sites that are equal to or shallower than 12-m deep (e.g. Belleville, Napanee and Hay Bay), all loggers were attached to the spar buoy line at depths listed in Table 1, using the deployment set-up illustrated

Larocque & Doka 9 in Figure 2A. At sites deeper than 12 m (e.g. Glenora, Conway and Lennox), U-moorings were used to ensure that the loggers were consistently positioned at a set height above bottom or depth below surface (Figure 2B). On U-moorings, loggers deployed between 1- and 9-m deep were attached to the spar buoy line, which was suspended vertically in the water column by the tension between the fl oating spar buoy and a weight on the lake bottom. An additional weight was placed below the deepest logger on the spar buoy line to fi x the loggers at a consistent position in the water column. Loggers, deeper than 9 m, were deployed on a second line and suspended in the water column by a submersed buoy.

Figure 2. Offshore mooring methods used at (A) sites less than 12-m deep and (B) the U mooring deployed at offshore sites deeper than 12 m.

Nearshore loggers were deployed approximately 30 cm from the lake bottom, using a submerged fl oat connected to a large weight resting on the bottom, in water depths ranging from 1.5 to 2 m during spring deployment. Actual water depths varied throughout the season as water levels fl uctuated but the logger typically remained above the sediments to avoid burial and 1.5-m deployment depths avoided complete dewatering of the loggers during low seasonal water levels, typically in late summer. Water temperatures were recorded at ½-hour intervals for the full year at nearshore sites and from early May to early November at offshore sites. Onset HOBO (U22-001) loggers were used at all stations except at Glenora, Conway and Lennox where the U-mooring fi ttings only accommodated the smaller Onset Tidbit (UTBI-001) loggers at depths of 9 m and greater (Figure 2B). A summary of the approx-

10 Water temperature imated depths and number of offshore loggers deployed in the Bay of Quinte in 2014 can be found in Table 1.

Table 1. Relative depth of temperature loggers at all offshore (mid-channel) stations in the Bay of Quinte.

Offshore Station Temperature monitoring depth (m) in 2014 Belleville 1* 5 Napanee 1 5 Hay Bay 1 3 6 9 12 Glenora 1 3 6 9 12* 15 18 Conway 1 3 6 9 12 15 18 21 24 27* 30 Lennox 1 5 10 15 20 25 30 35 40 45 50 55 60 * Loggers malfunctioned, no data

Air temperatures in 2014 were generally similar to recent historical mean monthly air temperatures except for February and March temperatures, which were cooler on average (Figure 3). Comparing the 2014 mean monthly air temperatures against the 1981 to 2010 climate normals for the Trenton A weather station, February and March were 3.3 to 4.6 °C cooler in 2014 than historic norms, and the remaining months ranged from -2.2 to 2.3 °C cooler or warmer than historical norms (Government of Canada, 2015).

Offshore temperature results

At all offshore stations, the 1-m depth temperatures varied more in the spring (April - May) and fall (September - November) from turnover, with seasonal standard deviation values ranging between 2.51 and 4.45, while summer (June - August) seasonal temperature had standard deviation values ranging between 1.48 to 3.14 (Table 2). The minimum temperatures during the ice-free deployment (> 0°C) ranged between 4.51°C and 8.49°C, at Lennox and Hay Bay, respectively, and occurred either within the first 5 days after deployment or the last few days before retrieval of the loggers as these were not full year de- ployments but during the ice-free season. The maximum temperatures at most locations were recorded on June 30 or in early July of 2014 (Table 2). Typically, temperatures across the Bay of Quinte increased from deployment in early May, peaked in July, and then generally declined through September until logger retrieval in November (Figure 4); although depending on their location, loggers experienced differing levels of temperature variation. Prior to grouping the 1-m depth temperature data of offshore stations into Upper (Napanee site only; logger at Belleville 1-m depth malfunctioned), Middle (Hay Bay and Glenora) and Lower (Conway and Lennox) bays, daily mean temperatures were compared within each group using a paired t-test. Daily mean temperatures at 1-m depth in Hay Bay was significantly warmer than Glenora (n=189, p=0.0008) and there was no significant difference in daily mean water temperature between Conway and Lennox (n=197, p=0.98).

Larocque & Doka 11 Figure 3. Mean monthly air temperatures from the Trenton A weather station from 2010 to 2014, including the 1981-2010 climate normals.

Table 2. Summary of temperatures (°C) at 1-m depth at offshore stations in 2014. Minimum and maximum temperatures are listed with the date when they occurred during deployment. The mean and standard deviation are reported for data available in the spring (Apr-May), summer (Jun-Aug) and fall (Sep-Nov). The 1-m depth logger at Belleville malfunctioned and data was not available.

Belleville Napanee Hay Bay Glenora Conway Lennox Minimum - 8.42 8.49 6.03 5.36 4.51 Date of Minimum - 03-Nov 01-May 04-May 04-May 01-May Maximum - 26.79 25.72 25.33 23.95 24.03 Date of Maximum - 30-Jun 03-Jul 30-Jun 30-Jun 30-Jun Spring Mean - 15.24 14.96 11.51 10 9.88 Spring SD - 3.51 3.44 3.15 2.94 2.72 Summer Mean - 22.76 22.03 19.22 19 19.09 Summer SD - 1.58 1.48 3.12 3.14 2.7 Fall Mean - 16.6 16.76 16.95 17.35 17.08 Fall SD - 4.45 4.05 2.63 2.51 2.82

In general, the shallower Upper Bay site warmed faster in the spring and summer, but cooled slightly faster in the fall than the Middle and Lower Bay stations (Figure 4). The 1-m depth temperatures in the Middle and Lower bays were more similar to each other than to the more protected and shallower Upper Bay (i.e., Napanee). Over the entire deployment period, the mean and standard deviation of daily mean

12 Water Temperature temperatures were warmest in Upper Bay (19.3 ± 0.34 °C), followed by Middle Bay (18.00 ± 0.29 °C) and Lower Bay (16.83 ± 0.31 °C). The differences in temperature between the bays are principally due to thermal inertia and illustrate the heat storage capacity of waters with varying depths and exposures to the main Lake dynamics.

Figure 4. Mean daily offshore water temperatures measured at 1-m depth in 2014. Upper Bay is Napa- nee (logger malfunction at Belleville station), Middle Bay is the mean of Hay Bay and Glenora, and Lower Bay is the mean of Conway and Lennox.

The offshore stations in the Middle (Hay Bay and Glenora) and Lower Bay (Conway and Lennox) are sufficiently deep (12 m or deeper) to observe stratification of the water column throughout the year (Figures 5 to 8). Hay Bay began to stratify in early May and was isothermal again by mid-late September with a temperature range of generally less than 1 °C at all depths until logger retrieval in November (Fig- ure 5). The 12-m temperature on the day before retrieval, November 6, 2014, was 9.1 °C. Stratification occurred at this site between 3- and 12-m throughout the year until turnover. The maximum difference between the 1-m and 12-m temperatures (a proxy of epi- to hypolimnion difference) was 17.3 °C and occurred on July 3, 2014. The water column at Glenora began stratifying in early May and was isothermal by Oct 9th, where the temperature range across all depths was less than 1.5 °C until the loggers were retrieved (Figure 6). The day before logger retrieval, November 4, 2014, the temperature at 18-m was 12.4 °C. The maximum temperature difference between the 1-m and 18-m depth loggers at this site was 17.6 °C on June 29, 2014.

Larocque & Doka 13 Figure 5. Mean daily offshore temperatures measured at different depths at the Hay Bay Station (at the confluence between Hay Bay and Middle Bay) in 2014.

Figure 6. Mean daily offshore temperatures measured at different depth at Glenora Station (Lower - Middle Bay transition) in 2014.

14 Water Temperature In the Lower Bay, stratification was just starting to occur at Conway by the time of logger deployment (Figure 7) but frequent upwellings or mixing in the spring delayed thermocline development. From October 9th until logger retrieval in November, water was isothermal with less than 1 °C temperature difference across all depths. The day before logger retrieval, November 4, 2014, the temperature at 30-m was 12.5 °C but maximum temperature at this depth was 15.7 °C when waters became isothermal. The maximum difference between 1- and 30-m depth temperatures was 17.4 °C on June 28, 2014. The Lennox station located in the Adolphus Reach is twice as deep as Conway and closer to Lake Ontario proper than the other stations (Figure 1). Lennox began to stratify in May and was isothermal again by November 1, 2014, with a temperature range of less than 1 °C at all depths until logger retrieval shortly after (Figure 8). The thermocline predominated between 15 and 20-m depth until September in which the thermocline/metalimnion expanded as deeper layers continued warming but surface layers cooled. The 60-m depth temperature on the day before retrieval, November 6, 2014, was 7.3 °C. The pattern of turnover in early November was similar to that of 2013 (Leisti et al., 2015), with a sharp rise in bottom temperatures to equal the steadily declining surface temperatures as the lake became isothermal. The maximum difference between 1- and 60-m depth temperatures was 18.1 °C on June 29, 2014.

Nearshore temperature results

Nearshore temperatures were logged throughout the year in the Upper Bay Napanee station and in the Middle Bay at both the Hay Bay and Carnachan Bay nearshore stations (Figure 1). All sites were deployed in approximately 1.5 m of water but these depths then varied throughout the season as water levels changed. At the beginning of the year, water temperatures were consistently above 1 °C (approximate date of ice-off) by April 3rd in Carnachan Bay, April 10th at Hay Bay and April 8th at Napanee. Tem- peratures peaked by July and generally followed air temperatures patterns, albeit with a dampened response because of thermal inertia (Figure 9). Water temperatures were below 1 °C (approximate date of ice-on) by mid-November (November 19th for Carnachan Bay and November18th for Hay Bay). There was no data for Napanee from October 15th onwards as the loggers were not located upon retrieval. Temperatures in Carnachan Bay were generally cooler than at Napanee or Hay Bay in the spring, very similar during the summer, and then were slightly warmer, by less than 1 °C , in the fall (Table 3). The maximum difference in daily mean temperatures between Napanee and Carnachan Bay was when Napanee was 12.6 °C warmer than Carnachan Bay on May 29, 2014, while Hay Bay was 12.0 °C warmer than Carnachan Bay on the same day. The cooler temperatures in Carnachan Bay in the late spring are consistent with previous years and are due to upwelling and mixing with the generally colder Lake Ontario waters by large seiches and diffusion between the Lower and Middle bay areas.

Larocque & Doka 15 Figure 7. Mean daily offshore temperatures measured at different depths at Conway Station (Lower Bay) in 2014.

Figure 8. Mean daily offshore temperatures measured at different depths at Lennox Station (Lower Bay) in 2014.

16 Water Temperature Figure 9. 2014 mean daily water temperatures at nearshore stations of Bay of Quinte.

Table 3. Summary of temperatures at nearshore sites of the Bay of Quinte in 2014. Approximate dates when water froze over and thawed are listed. Maximum temperatures are listed with the date when they occurred during the deployment. The mean and standard deviation (SD) are reported for data available in the spring (Apr-May), summer (Jun-Aug), fall (Sep-Nov) and winter (December – March). Napanee only logged until Oct 15, 2014, and does not represent a full fall or winter (values marked with an asterisk).

Napanee Hay Bay Carnachan Bay Date of Ice-off (> 1 °C) 08-Apr 10-Apr 03-Apr Date of Ice-on (< 1 °C) - 18-Nov 19-Nov Maximum 28.3 27.7 28.15 Date of Maximum 01-Jul 01-Jul 01-Jul Spring Mean 10.97 11.38 9.84 Spring SD 6.62 7.25 5.29 Summer Mean 22.9 22.67 22.39 Summer SD 2.03 1.99 2.36 Fall Mean 18.71* 12.01 12.81 Fall SD 3.61* 4.05 6.99 Winter Mean 0.07* 0.68 0.27 Winter SD 0.08* 0.47 0.48

Larocque & Doka 17 Discussion

Bay of Quinte temperatures and oxygen have been analyzed both in the past (Minns and Johnson, 1986; Minns, 2009) and more recently (Minns et al., 2011). In the 2011 paper, the offshore/nearshore temperature dataset during the open water seasons from 2001 to 2008, and earlier data, was used to assess the influence that climate, morphometry, nutrient loading and Dreissenids have had on Quinte’s temperature and oxygen temporal trends and spatial patterns. The analysis revealed a slight climate warming signa- ture, with maximum summer surface waters increasing by almost 1 °C on average from 1972 to 2008. A detailed analysis of thermal structure in major embayments and the whole Bay of Quinte are underway and coarse statistical temperature models have been used for various assessments including habitat analyses for the Remedial Action Plan and for potential delisting of this Area of Concern (Gert- zen et al., 2012a; 2012b) and for modelling hydrodynamic conditions in the Bay of Quinte (Oveisy et al., 2015). Fish are bioenergetically and behaviourally affected by temperature at various stages of their life history, from spawning and egg survival to growth and migration (Hanson et al., 1997). Thermal dynamics were modelled in the Bay, and incorporated into a spatiotemporal habitat suitability assessment for different thermal guilds (Gertzen et al., 2012a). Specifically, thermal habitat has been included in fish habitat mapping and population modelling activities and delisting has been recommended, based on trends and various spatial comparisons (Gertzen et al., 2012b). For a different application, Oveisy et al. (2015) have incorporated water temperature into a three-dimensional hydrodynamic model of the Bay of Quinte to understand horizontal transport, dilution of nutrient rich inflows and water exchange to Lake Ontario. The model helps provide insight for watershed management, in terms of areas to target for nutrient discharges (e.g., wastewater plumes) and nutrient load reductions to improve water quality. Inter-annual variability in temperature will influence variation in production from year to year for different fish species. We will be incorporating this information into population models for some key fisheries in the eastern end of Lake Ontario so that the potential relative contribution of the Bay is quantified. Lastly, these data can contribute to other long-term datasets, like that at Belleville Water Treatment Facility. Fisheries and Oceans Canada (DFO) field sampling datasets can help identify trends in warming that may or may not be occurring spatially across the Bay. These data can also contribute to local and regional, hydrodynamic and climate modelling underway that can assist future management of restoration activities, coastal areas and fisheries.

Acknowledgements

Thanks go to Sunci Avlijas, Stephen James, Kathy Leisti, Janet Mossman, Dave Reddick and Abby Wynia for logger preparation and nearshore deployment/retrieval. Offshore logger support was provided by Joe Gabriele, Jeremy Hicks, Benoit Lalonde, Adam Morden, Charles Talbot, and Carl Yanch of Environment Canada aboard the CCG and DFO Science vessel, the Kelso. Funding for the Bay of Quinte Fish Habitat Assessment project, of which this is a subproject, was provided by the Great Lakes Action Plan, administered and led by Environment Canada which funds

18 Water Temperature research and monitoring in Areas of Concern and is geared towards delisting.

References

Government of Canada, 2015. Climate, Daily Data Report for September 2014: Trenton A, Ontario. http://climate.weather.gc.ca/clima- teData/dailydata_e.html?timeframe=2&Prov=ON&StationID=5126&dlyRange=1935-01-01%7C2015-09-08&Year=2014&Mon th=9&cmdB1=Go# (Accessed September 9, 2015). Gertzen, E.L., Doka, S.E., Minns, C.K., Moore, J.E., Bakelaar, C.N., 2012a. Effects of water levels and water level regulation on the supply of suitable spawning habitat for eight fish guilds in the Bay of Quinte, Lake Ontario. Aquat. Ecosyst. Health Mgmt. 15(4), 397-409. Gertzen, E.L., MacEachern, J.T., Doka, S.E., 2012b. BUI #14-4: Loss of Fish and Wildlife Habitat (FWH-4). Bay of Quinte AOC BUI Status Report. Informal Report to Bay of Quinte RAP Council. Oct 2012. Hanson, P.C., Johnson, T.B., Schindler, D.E., Kitchell, J.F., 1997. Bioenergetics model 3.0 for Windows. University of Wisconsin, Sea Grant Institute, Technical Report WISCU-T-97-001, Madison, USA. Leisti, K.E., Doka, S.E., 2007. Water Temperature Monitoring in the Bay of Quinte. In: Monitoring Report #16. Project Quinte Annual Report 2005, pp. 7-11. Bay of Quinte Remedial Action Plan, Kingston, Ontario, Canada. Leisti, K.E., Avlijas, S. , Doka, S.E., 2015. Water Temperature Monitoring in the Bay of Quinte. In: Monitoring Report #24. Project Quinte Annual Report 2013, pp. 8-20. Bay of Quinte Remedial Action Plan, Kingston, Ontario, Canada. Minns, C.K., Johnson, M.G., 1986. Temperature and Oxygen Conditions and Oxygen Depletion in the Bay of Quinte, Lake Ontario. In: Minns, C.K., Hurley, D.A., Nicholls, K.H. (Ed.), Project Quinte: point-source phosphorus control and ecosystem response in the Bay of Quinte, Lake Ontario, pp. 40-49. Can. Spec. Publ. Fish. Aquat. Sci. 86. Minns, K., 2009. Temperature and oxygen regimes in the Bay of Quinte 1972 – 2008. In: Monitoring Report #18, Project Quinte Annual Report 2007, pp. 8-12. Bay of Quinte Remedial Action Plan, Kingston, Ontario, Canada. Minns, C.K., Moore, J.E., Doka, S.E., St. John, M.A., 2011. Temporal trends and spatial patterns in the temperature and oxygen regimes in the Bay of Quinte, Lake Ontario, 1972-2008. Aquat. Ecosyst. Health Mgmt. 14(1), 9-20. Oveisy, A., Boegman, L., Rao, Y.R., 2015. A model of the three-dimensional hydrodynamics, transport and flushing in the Bay of Quinte. Journal of Great Lakes Research. 41(2), 536-548.

Larocque & Doka 19

POINT SOURCE PHOSPHORUS LOADINGS: 1965 TO 2014

P. Kinstler and A. Morley

Ontario Ministry of the Environment and Climate Change 1259 Gardiners Road, Unit 3, P.O. Box 22032 Kingston, Ontario K7M 8S5

The 1965 to 2014 daily average total phosphorus loads to the Bay of Quinte from the six local municipal sewage treatment plants (STP) were derived from routine Ontario Ministry of the Environment and Climate Change (MOECC) monitoring data (Table 1). Daily average phosphorus loadings are reported as kilograms per day (kg d-1). Loadings are calculated as the monthly average effluent phosphorus concentration multiplied by the monthly average effluent flow from the STP for the period of record. In 2014, the estimated daily average total phosphorus loads from STPs bordering the Bay of Quinte were 5.6 kg d-1 for the year and 5.3 kg d-1 for the May to October period. Since 1991, low total phosphorus inputs have been maintained by controlling variability in wastewater flow, enhancing phosphorus removal capabilities and expanding treatment capacity. To manage long-term phosphorus loading, the MOECC advised municipalities bordering the Bay of Quinte that a Bay of Quinte RAP “phosphorus load cap” would be applied to STPs. Initially, the “phosphorus load cap” was to be calculated as the product of a total phosphorus effluent concentration of 0.3 mg -1l and the hydraulic capacity for STPs bordering the Bay of Quite. This “phosphorus load cap” would be used post-1995 to measure and assess performance. Since 1995, most of the municipal STPs have voluntarily reduced total phosphorus inputs or have made upgrades where the “phosphorus load cap” has been incorporated into the STP’s operating Environmental Compliance Approval (ECA). Similarly, since 1995 the MOECC has advised municipalities in the Bay of Quinte watershed that a Bay of Quinte RAP “phosphorus load cap” applies to STPs in the Bay of Quinte watershed. The watershed “phosphorus load cap” has been initially established as the product of a total phosphorus effluent concentration of 0.5 mg l-1 and the STP hydraulic capacity. Most STPs in the watershed have voluntarily reduced total phosphorus inputs or have had upgrades where the “phosphorus load cap” has been incorporated into the STP’s operating ECA. Phosphorus loading from bypassing events is not specifically reported in the data as presented. Appreciable bypassing continues to occur at some STPs adding to the phosphorus loading to the Bay of Quinte. Continued efforts are required to control wastewater influent flow variability and reduce bypassing to effectively “cap” phosphorus inputs to the Bay of Quinte.

Kinstler & Morley 21 Table 1. Daily average for total phosphorus loadings from municipal STPs bordering the Bay of Quinte (total P kg d-1) from 1965 to 2014.

Period T CFB B D N P TOTAL 1965-72 53 13 110 2.7 20 15 214 1973-75 47 5.9 73 2.3 17 11 156 1976-77 43 2.1 39 2.3 63 9.1 158 1978-86 Jan-Dec 7.5 1.9 31 0.9 24 2.4 68 May-Oct 6.3 1.7 25 0.7 24 1.9 60 1987 Jan –Dec 5.9 2.7 16 0.6 11 2.1 38 May-Oct 6.4 2 15 0.5 6.2 1.8 32 1988 Jan-Dec 6.9 4.7 15 0.7 9.2 2.2 42 May-Oct 6.3 1.7 11 0.5 12.4 1.9 34 1989 Jan-Dec 6 * 13 0.7 5.6 1.4 26.7 May-Oct 5.3 * 11 0.6 5.9 1 23.8 1990 Jan-Dec 5.9 * 14 1 4.1 1.8 26.8 May-Oct 6.5 * 9 0.6 3.8 1.5 21.4 1991 Jan-Dec 6.8 * 9.2 1 2.6 1.7 21.3 May-Oct 5.2 * 4.6 0.8 2.3 0.9 13.8 1992 Jan-Dec 5.9 * 12.3 1.4 2.5 1.5 23.6 May-Oct 5.1 * 9.6 0.9 2.3 0.8 18.7 1993 Jan-Dec 6.1 * 12.1 0.9 3 1 23.1 May-Oct 4.6 * 10.1 0.3 2.8 0.7 18.5 1994 Jan-Dec 4.2 * 18.5 0.5 2 1.3 26.5 May-Oct 3.3 * 12.1 0.4 1.4 0.9 18.1 1995 Jan-Dec 4.2ª * 14.3 0.3 1.5 1.2 21.5 May-Oct 3.3ª * 14.2 0.2 1.1 0.9 19.7 1996 Jan-Dec 3.4 * 18.4 0.4 1.4 1.7 25.3 May-Oct 4.4 * 14.9 0.3 1.2 1.1 21.9 1997 Jan-Dec 4 * 16.7 0.5 1.8 1.4 24.3 May-Oct 3.2 * 13.2 0.4 1.4 0.6 18.8 1998 Jan-Dec 2.3 0.7 11.6 0.4 1.5 1.1 16.9 May-Oct 2.3 0.6 11.8 0.4 1.3 0.6 16.4 1999 Jan-Dec 3.3 0.6 6.8 0.7 1.6 1.3 13.6 May-Oct 3.5 0.5 7 0.6 1.3 0.8 13.2 2000 Jan-Dec 2 0.7 6.1 0.6 1.5 1.5 12.4 May-Oct 2.5 0.8 7.3 0.5 1.4 0.8 13.4 2001 Jan-Dec 1.5 0.5 4.8 0.2 1.3 1.8 10.1 May-Oct 1.4 0.5 3.5 0.1 1.3 1 7.8 2002 Jan-Dec 1.7 0.5 7 0.9 1.3 1.2 12.6 May-Oct 1.9 0.4 9.2 0.4 1.4 1.2 14.5 2003 Jan-Dec 2.2 0.7 13 0.2 4 2.2 22.3 May-Oct 2.8 0.8 4 0.2 1 0.5 9.3 T=Trenton, CFB=Canadian Forces Base Trenton, B=Belleville, D=Deseronto, N=Napanee, P=Picton * CFB Trenton loadings not available. ª No 1995 records available for Trenton, 1994 loadings substituted to estimate total loading. ᵇ In October 2009, Belleville reported bulking as a result of a hydraulic overload caused by heavy rain.

22 P Loadings Period T CFB B D N P TOTAL 2004 Jan-Dec 3.3 0.4 7 0.2 1.4 0.9 13.2 May-Oct 3.2 0.4 2 0.2 0.9 0.7 7.4 2005 Jan-Dec 2.9 0.4 5.5 0.2 1.3 0.8 11.1 May-Oct 2.8 0.3 3.8 0.1 1.1 0.5 8.6 2006 Jan-Dec 4.8 0.8 8.9 0.3 1.7 2 18.5 May-Oct 4 0.3 5.7 0.3 1.4 1.2 12.9 2007 Jan-Dec 3.6 0.3 1.6 0.3 1.4 1.4 8.6 May-Oct 3.4 0.2 1.2 0.2 1.1 0.9 7 2008 Jan-Dec 4.1 0.5 1.4 0.3 1.6 1.2 9.1 May-Oct 3.8 0.3 1.5 0.1 1.3 0.8 7.8 2009 Jan-Dec 2 0.4 21.5ᵇ 0.2 1.3 0.9 26.3 May-Oct 2.1 0.3 27.9ᵇ 0.1 1 0.9 32.3 2010 Jan-Dec 1.3 0.2 10.6 0.1 1 0.7 13.9 May-Oct 1.4 0.3 9.9 0.2 0.7 0.5 13 2011 Jan-Dec 1.8 0.3 2.6 0.1 1.2 0.6 6.2 May-Oct 2.1 0.3 2.3 0.1 1 0.2 6 2012 Jan-Dec 1.1 0.2 1.3 0.1 1.1 0.2 4 May-Oct 1.5 0.3 1.5 0.1 1 0.2 4.6 2013 Jan-Dec 1.5 0.3 2 0.1 0.9 0.1 4.9 May-Oct 1.4 0.3 1.5 0.1 0.8 0.1 4.2 2014 Jan-Dec 1.5 0.6 1.9 0.2 1.3 0.1 5.6 May-Oct 1.6 0.4 2 0.1 1.1 0.1 5.3 T=Trenton, CFB=Canadian Forces Base Trenton, B=Belleville, D=Deseronto, N=Napanee, P=Picton * CFB Trenton loadings not available. ª No 1995 records available for Trenton, 1994 loadings substituted to estimate total loading. ᵇ In October 2009, Belleville reported bulking as a result of a hydraulic overload caused by heavy rain.

Kinstler & Morley 23

PHYTOPLANKTON AND MICROBIAL FOOD WEB INTERACTIONS AT A LONG-TERM MONITORING STATION IN THE BAY OF QUINTE: BELLEVILLE, 2014

M. Munawar, R. Rozon, M. Fitzpatrick and H. Niblock

Great Lakes Laboratory for Fisheries and Aquatic Sciences, Fisheries and Oceans Canada, Burlington, ON L7S 1A1 Correspondence: [email protected]

Introduction

This annual report is based on the data collected under the auspices of Project Quinte, the long-term research and monitoring program maintained by Fisheries & Oceans Canada in conjunction with Environ- ment Canada, the Ontario Ministry of the Environment and the Ontario Ministry of Natural Resources. Project Quinte began in 1972 in response to widespread concerns over eutrophication and focused on monitoring the response of phytoplankton and zooplankton communities to phosphorus load reductions. Beginning in 2000, the microbial food web, including bacteria, autotrophic picoplankton, heterotrophic nanoflagellates and ciliates was added to the regular suite of parameters in order to provide a more ho- listic assessment of the lower trophic levels. This is an interim report for the 2014 sampling period (May – October) which provides details of the structure and function of the microbial-planktonic food web at the Belleville monitoring site. In addition, long-term changes in total phytoplankton biomass as well as total phosphorus and chlorophyll a as they relate to Remedial Action Plan targets are also discussed.

Methods

A total of 13 sampling events were conducted bi-weekly from May 6 – October 23, 2014 at the long- term monitoring site near Belleville (Fig. 1). Integrated (0 – 5 m) whole water samples were collected from our research vessel, the Murray J, and transported on ice to our laboratory in Burlington, Ontario for further analysis. Various physical measurements were performed on site including temperature (YSI ExoSonde) and water column irradiance (Li-Cor Quantum sensor). Total phosphorus and nitrate + nitrite were analyzed in accordance with the standard protocols of the National Laboratory for Environmental Testing (NLET). Chlorophyll a concentrations were determined by filtering up to 1 litre of water through Whatman GF/C filters followed by cold acetone pigment extraction and spectrophotometric analysis (Strickland & Parsons, 1968) Microbial loop samples were preserved in 1.6% formaldehyde and processed using DAPI staining and epifluorescent microscopyto enumerate the bacteria, heterotrophic nanoflagellates (HNF) and autotrophic picoplankton (APP). Details of this technique are given in Munawar et al. (1994). Ciliate samples were preserved with acidified Lugol’s iodine and prior to analysis post-fixed with Bouin’s fluid.

Munawar et al. 25 Enumeration and identification followed the Quantitative Protargal Staining technique (Montagnes and Lynn, 1993). Phytoplankton samples were fixed with acidified Lugol’s iodine upon collection. Enumeration and measurement followed the HPMA (2-hydroxypropyl methacrylate) technique described by Crumpton (1987) and is broadly compatible with the Utermöhl (1958) inverted microscope technique. Size fractionated primary productivity was estimated by 14-Carbon uptake for three size categories of phytoplankton (<2 µm, 2-20 µm, >20 µm) following the standard technique of Munawar and Mun- awar (1996).

Figure 1. Map of the Bay of Quinte showing long-term monitoring sites at Belleville (B), Napanee (N), Hay Bay (HB) and Conway (C). Only the Belleville data is reported on here.

Results and discussion

Physical characteristics

Project Quinte has included long-term monitoring of several physical and chemical parameters including temperature, water column irradiance (light penetration) and nutrients in addition to measures of primary productivity and algal standing crop. Surface water temperatures at the Belleville site ranged from 10.6 °C to 25.2 °C between May and October 2014. Likewise, the vertical attenuation coefficient (Ԑpar), a measure of light penetration through the water column, was between 0.8 and 1.3 m‑1 indicating relatively low water clarity, especially during the summer months.

Nutrients

Phosphorus abatement was implemented in the Bay of Quinte in 1978 in response to concerns over

26 Phytoplankton & Microbial Food Web eutrophication. Later, when the Remedial Action Plan was developed, a target concentration for total phosphorus (TP) of 30 μg l‑1 was established. During 2014, we observed that TP concentrations ranged from 11 – 71 μg l‑1 (Fig. 2a) with a mean of 31.5 μg l‑1 and exceeded the target on 7 of 13 occasions. Ni- trate + nitrite concentrations were between 8 and 258 μg l‑1 with the highest levels observed in the early spring and the lowest levels in the summer (Fig. 2b). From mid-July to early October, nitrate + nitrite concentrations were below 20 μg l‑1 indicating a significant drawdown of nitrogen by photosynthesizing algae which is characteristic of a eutrophic environment.

a) Total phosphorus (μg l-1)

b) Nitrate + nitrite (μg l-1) 400

300

200

100

c) Chlorophyll a (μg l-1) 20

Figure 2. Total phosphorus, nitrate + nitrite and chlorophyll a concentrations observed at Belleville during the 2014 monitoring season. All units are μg l‑1. Remedial Action Plan (RAP) targets for total phosphorus and chlorophyll a are indicated on the respective graphs.

Munawar et al. 27 Chlorophyll a

Chlorophyll a is often used as an indicator of the algal standing crop and a target concentration of 12 – 15 μg l‑1 was established for the Bay of Quinte. In 2014, we observed that chlorophyll a ranged from 3.4 – 15.7 μg l‑1 (Fig. 2c) exceeding the (upper) target only once. The average chlorophyll a concentration of 8.1 μg l‑1 is considered indicative of mesotrophic to eutrophic conditions (Munawar et al., 2012; Vollenweider et al., 1974).

Primary productivity

Primary productivity ranged from 7 to 67 mg C m‑3 h‑1 during the 2014 field season (Fig. 3a) with the lowest level reported in late May and the highest level in mid-July. Our size fractionation experiments revealed that larger-sized net plankton (> 20 µm) were dominant from early July until early October (37 – 59%), with the only exception being a peak in nanoplankton in mid-July. By way of comparison, primary productivity in the offshore, oligotrophic waters of eastern Lake Ontario (Station 81, near Kings- ton, ON) ranged from 1.2 – 9.9 mg C m‑3 h‑1 during the same time frame with relatively equal contributions from the nano- (2 – 20 µm) and pico- (< 2 µm) size classes early in the year, and an increase in the net (> 20 µm) production later in the year (M. Munawar, unpubl.). The much higher rates of production by larger sized algae observed in the Bay of Quinte is characteristic of a nutrient enriched environment.

Phytoplankton biomass and composition

Unique to the Bay of Quinte RAP, a target for phytoplankton biomass (4 – 5 g m‑3) was also established to provide a more reliable measurement of the algal standing crop. During 2014, phytoplankton biomass at the Belleville site ranged from 0.4 – 6.1 g m‑3 (Fig. 3b), exceeding the target range only once on Sep- tember 9th. Not coincidentally, this was also the only sampling date on which we observed algal bloom conditions. The September 2014 bloom was composed of filamentous diatoms Aulacoseira ambigua and A. muzzanensis (44.5% and 14.6%, respectively), and to a lesser extent Fragilaria crotonensis (8.2%). Throughout 2014, Aulacoseira spp. (Diatomeae), Cryptomonas erosa, (Cryptophceae) as well as Uroglena sp. (Chrysophyceae) composed >30% of the biomass in 7 of 13 observations. Cyanophyta (blue-green algae) were most prevalent from the mid-summer to early fall contributing 0.7 – 37.5% of the total biomass and containing several filamentous and colonial forms including species of Dolichos- permum (formerly Anabaena), Aphanocapsa and Microcystis. Of the dominant cyanobacterial forms, only Microcystis aeruginosa (observed July 3, July 29) and Dolichosmpermum crassa (Sept 29) are known to produce algal toxins. The average phytoplankton biomass observed at Belleville (1.9 g m‑3) was the lowest on record surpassing the previous record set in 2000 (2.3 g m‑3) and undoubtedly reflects the colder temperatures and higher precipitation experienced in the first half of the year. In general, the phytoplankton assemblage had a mixture of diatoms, blue-greens and other algae that reflect cultural eutrophication although the biomass was more indicative of mesotrophic conditions.

28 Phytoplankton & Microbial Food Web a) Size fractionated primary productivity (mg C m-3 h-1) and % size composition 100

0 100%

75%

50% % <2 % 2-20

25% % >20

b) Phytoplankton biomass (g m-3) and % taxonomic composition 8

0 100%

75% Dinophyceae Cryptophyceae 50% Diatomae Chrysophyceae 25% Euglenophyta Chlorophyta 0% Cyanophyta

c) Microbial loop biomass (g m-3) and % composition

100%

75%

50% Ciliates HNF APP 25% Bacteria

Figure 3. a) Primary productivity (mg C m‑3 h‑1) and % size composition; b) phytoplankton biomass (g m‑3) and % taxonomic composition, and c) microbial loop biomass (g m‑3) and % composition observed at the Belleville site during 2014.

Munawar et al. 29 Species composition (2013 and 2014)

A comparison of species composition and biodiversity between common species during 2013 and 2014 is presented in Tables 1a,1b, 1c. During 2014 the ecology of phytoplankton of the Bay was dominated by diatoms Aulocoseira spp, Syndera spp., Cyclostephanos spp., Fragilaria capucina var. mesolepta were most common species (>5% to the total biomass). Aulacoseira was absent only twice in two samples out of 13 samples on June 3rd and July 3rd. Despite the fact that mean phytoplankton biomass declined from 3.0 g m‑3 in 2013 to 1.9 g m‑3 in 2014, a similar mixture of dominant species were observed in both years (Table 1). These included various species of Diatomeae (e.g. Aulacoseira ambigua, A. granulata, A. muzzanensis, Fragilaria capucina), Cyanophyta (Aphanocapsa deliccatissima, Dolichospermum sp., Microcystis aeruginosa) and Cryptophyceae (Cryptomonas erosa). One of the more surprising findings was that two genera of Chrysophyceae (Uroglena and Mallomonas) were found to be prevalent in June 2014 although total biomass was low (1.6 g m‑3) on that date and the larger filamentous diatoms (e.g. Aulacoseira sp.) were noticeably absent. Another noticeable difference in 2014 compared to 2013 was the presence of centric diatoms including Cyclostephanos tholiformis and Cyclotella ocellata throughout the year.

Microbial loop

Microbial loop biomass (composed of bacteria, autotrophic picoplankton, heterotrophic nanoflagellates, and ciliates) ranged from a low of 3.2 g m‑3 in late October to a high of 7.5 g m‑3 in mid-July (Fig. 3c). On average, microbial loop biomass was more than double the phytoplankton biomass. Bacteria contributed 30 - 69% of the microbial loop biomass; APP accounted for 1 – 45%, HNF: 7 – 68% and ciliates: 1 – 2%.

Microbial – planktonic food web

Considering the combined phytoplankton and microbial communities, total autotrophic biomass (phytoplankton + APP) was 0.8 – 9.0 g m‑3 compared to 2.8 – 6.1 g m‑3 for heterotrophs (bacteria, HNF, ciliates). On an average, autotrophs accounted for 43% of the microbial food web biomass compared to 56% for heterotrophs. While the problem of eutrophication is generally regarded as an excess of autotrophic production driven by nutrient enrichment, our results show that heterotrophic bacteria and nanoflagellates have a very important role in food web interactions. It is worth noting that bacterial biomass during 2014 exceeded phytoplankton biomass on 11 of 13 sampling dates and was on average (2.7 g m‑3) more than 40% higher than phytoplankton biomass. Our results suggest that bacterial decomposition may have been an important factor regulating phytoplankton biomass during 2014.

30 Phytoplankton & Microbial Food Web Table 1a. Phytoplankton species representing >5% of the total biomass observed during the spring of 2013 and 2014.

2013 2014 Date Species % Date Species % May-09 Aulacoseira muzzanensis 50.9 May-06 Fragilaria capucina var. mesolepta 19.8 Synedra filiformis 14 Chlamydomonas sp. 14.2 S. delicatissima 8.7 Aulacoseira ambigua 11.2 Synedra ulna 10 Cocconeis placentula var. lineata 7.3 Cryptomonas erosa 5.2 May-23 Aulacoseira muzzanensis 38.1 May-21 Aulacoseira ambigua 19 Synedra filiformis 10.7 Synedra ulna var. acus 10.5 Actinocyclus normanii 8.8 S. ostenfeldii 9.8 Aulacoseira granulata 5.9 Cyclostephanos tholiformis 7.6 Fragilaria crotonensis 5.1 S. filiformis 6.5 Cryptomonas erosa 5.3 Jun-04 Aulacoseira ambigua 38.5 Jun-03 Uroglena sp. 44.2 A. muzzanensis 17.5 Mallomonas sp. 12.9 A. granulata 14.3 Navicula tripunctata 7.6 Peridinium sp. 10.6 Cyclostephanos tholiformis 7.3 C. dubius 6.3 Jun-19 Aulacoseira granulata 42.6 Jun-18 Aulacoseira muzzanensis 26.9 A. muzzanensis 17.3 A. granulata 10.8 Peridinium umbonatum 13.5 Cyclostephanos dubius 9.9 Synedra ulna var. chaseana 7.1 Cryptomonas erosa 6.5 Asterionella formosa 6.3 Cyclotella ocellata 5.9

Table 1b. Phytoplankton species representing >5% of the total biomass observed during the summer of 2013 and 2014.

2013 2014 Date Species % Date Species % Jul-03 Peridinium umbonatum 27.9 Jul-03 Cryptomonas erosa 35.1 Aulacoseira muzzanensis 23.1 Peridinium cunningtonii 27.1 Aulacoseira granulata 9.7 Microcystis aeruginosa 7.5 Cryptomonas erosa 9.3 Coelastrum microporum 6.9 Melosira varians 7.5 Aulacoseira islandica 5.9 Jul-16 Aulacoseira ambigua 48.8 Jul-15 Aulacoseira granulata 37.1 Microcystis aeruginosa 19.6 A. ambigua 23.7 Cryptomonas erosa 5.8 Aphanothece nidulans 5.7

Munawar et al. 31 2013 2014 Date Species % Date Species % Jul-31 Aulacoseira ambigua 29.2 Jul-29 Aulacoseira ambigua 44.5 A. crenulata 19.2 Microcystis aeruginosa 14.3 Fragilaria capucina var. mesolepta 17 Aulacoseira granulata 9.3 F. crotonensis 5.1 Aphanothece nidulans 5.1 Aug-13 Dolichospermum crassa 22 Aug-12 Cryptomonas erosa 56.6 Mougeotia sp. 18.2 Aphanocapsa holsatica 10.4 Cyclostephanos dubius 14.8 A. delicatissima 6.7 Fragilaria capucina var. vaucheriae 8.8 Rhodomonas minuta var. nanno- 5.1 Pediastrum duplex 5.1 planctica Aug-27 Aulacoseira ambigua 65.5 Aug-26 Dolichospermum circinalis 24.9 Aphanocapsa holsatica 7.4 Pediastrum boryanum var. longi- 15.6 Microcystis aeruginosa 5.8 corne Mallomonas sp. 9.9 Chlamydomonas sp. 8.4 Rhodomonas minuta var. nanno- 8 planctica Gymnodinium sp. 5.5 Sep-10 Aulacoseira crenulata 19.6 Sep-09 Aulacoseira ambigua 44.5 Fragilaria crotonensis 17.5 A. muzzanensis 14.6 Aulacoseira ambigua 8.5 Fragilaria crotonensis 8.2 Dolichospermum planctonica 8.2 A. islandica 5.9 Cyclotella comta var. bodanica 6.7 A. alpigena 5.6 Microcystis aeruginosa 5.9

Long term trends in total phosphorus, chlorophyll a and phytoplankton

Project Quinte began in 1972 in response to widespread concerns over eutrophication, both within the Bay and throughout the Great Lakes basin. At the time, phosphorus loads into the Bay exceeded 200 kg d‑1 and controlling point-source phosphorus loadings was deemed to be the best course of action to alleviate eutrophication (e.g. Vollenweider et al., 1974; Minns et al., 1986). Project Quinte was unique in that it not only informed the discussion and policy over nutrient controls under Annex 4 of the Great Lakes Water Quality Agreement, but it was also intended to be the case study for measuring long-term ecosystemic response to phosphorus abatement. In 1978, point-source phosphorus loads were cut to < 80 kg d‑1 and have been steadily reduced since. The current recommended cap is 15 kg d‑1 (Minns and Moore, 2004) and in 2014, P loadings into the Bay were estimated to be 5.3 kg d‑1 during the May to Oc- tober period (Kinstler & Morely, 2016). Similarly, the total phosphorus concentration (seasonal average) was reduced from ≈ 90 μg l‑1 in the early 1970s to ≈ 50 μg l‑1 following P abatement to current levels which were near the target of 30 μg l‑1 although TP concentrations appear to have been on an upswing over the past 6 years, despite a drop in 2014 relative to 2013 (Fig. 4). While point-source phosphorus loads continue to be reduced, TP concentrations have not necessarily followed the same trend since TP

32 Phytoplankton & Microbial Food Web is influenced by many other non-point factors including tributary loadings, storm sewer overflows and sediment re-suspension (e.g., Munawar et al., 2012).

Table 1c. Phytoplankton species representing >5% of the total biomass observed during the fall of 2013 and 2014.

2013 2014 Date Species % Date Species % Sep-24 Aulacoseira ambigua 31.2 Sep-23 Aulacoseira ambigua 17 Fragilaria crotonensis 16.6 Aphanocapsa holsatica 9 Microcystis aeruginosa 7 Oocystis parva 8.5 A. granulata 6.6 Dolichospermum crassa 7.4 Stephanodiscus alpinus 5.5 Chrysochromulina parva 6.4 Lagerheimia ciliata 6.1 Oct-09 Aulacoseira ambigua 36.9 Oct-07 Aulacoseira ambigua 44.5 A. granulata 25.4 Aphanothece nidulans 7.5 Microcystis aeruginosa 14.7 Cryptomonas erosa 6.1 Fragilaria crotonensis 6 Gomphonema truncatum 5.4 Oct-30 Aulacoseira ambigua 64.2 Oct-23 Aulacoseira ambigua 15.5 Cryptomonas erosa 6.3 Cryptomonas erosa 13 Cyclotella ocellata 9 Asterionella formosa 8.8 Rhodomonas minuta var. nanno- 8.1 planctica Eudorina elegans 5.1

Reductions in phosphorus loadings were predicted to reduce the algal standing crop. Chlorophyll a concentrations have declined from ≈ 40 μg l‑1 (on average) prior to phosphorus abatement to current levels that are at or near the target of 12 – 15 μg l‑1 although there has been considerable variability (Fig. 4). Likewise, phytoplankton biomass has declined from an annual average > 10 g m‑3 in the early 1970s to levels that have generally been consistent with the RAP target of 4 – 5 g m‑3. It is worth noting however, that meeting the RAP targets for both chlorophyll a and phytoplankton biomass still results in persistently eutrophic conditions (see Munawar et al., 2012, for a full discussion). Both chlorophyll a (8.1 μg l‑1) and phytoplankton biomass (1.9 g m‑3) during 2014 were the lowest on record which may have more to do with extreme weather events than general improvements in ecosystem health. There is not enough room in the current report to provide a full discussion of the composition of the phytoplankton community however, such a discussion is critical to assessing and understanding the Ben- eficial Use Impairments of “Eutrophication” and “Degradation of phytoplankton and zooplankton communities”. Nicholls and Carney (2011) considered the long-term data from 1972 – 2008 and observed that while most of the dominant taxa declined since P abatement, the blue-green Microcystis actually showed an increase in biomass following the establishment of dreissenid mussels in the bay during the mid-1990s. Having said that, diatoms, notably Aulacosiera spp., have consistently been reported as the

Munawar et al. 33 largest contributor to total phytoplankton biomass followed by species of Cyanophyta (Dolichosper- mum spp.) as shown in Fig. 5. Both Diatomeae and Cyanophyta blooms have contributed to the current eutrophication status and impairment of the Bay of Quinte over the years and further research into the complexity and composition of individual bloom events is warranted.

a) Total phosphorus (μg l-1)

100 90 80 70 60 50 40 30 20 10 0 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014

b) Chlorophyll a (μg l-1) 45 40 35 30 25 20 15 10 5 0 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014

Figure 4. Total phosphorus and chlorophyll a trends in the Bay of Quinte at the Belleville long-term monitoring station. Values are the seasonal weighted mean and units are μg l-1. The RAP target is indicated by the gray bar.

34 phytopLankton & microbiaL FooD Web 20

100%

75%

50%

25%

Cyanophyta Chlorophyta Euglenophyceae Chrysophyceae Diatomeae Cryptophyta Dinophyceae

Figure 5. Long-term trends in phytoplankton biomass (g m‑3) and composition (% biomass) at the Bel- leville monitoring site.

Summary

As part of its ongoing commitments to Project Quinte, Fisheries & Oceans Canada surveyed the phytoplankton and microbial communities of the Bay of Quinte during 2014. This work included microscop- ic assessments of the phytoplankton, microbial loop (autotrophic picoplankton, bacteria, heterotrophic

Munawar et al. 35 nanoflagellates) and ciliate communities in addition to radioisotope measurements of primary productivity. Regular measurements of temperature, water column irradiance, nutrient levels (total phosphorus, nitrate + nitrite), and chlorophyll a were also included. During 2014, we found that seasonal averages of total phosphorus (31.5 μg l‑1), chlorophyll a (8.1 μg l‑1) and phytoplankton biomass (1.9 g m‑3) were below Remedial Action Plan targets, and that chlorophyll a and phytoplankton were at their lowest recorded levels. During the course of our 13 sampling events, only 1 algal bloom was observed during late summer and was dominated by species of Aulacoseira (A. ambigua, A. muzzanensis, A. islandica, A. alpigena), filamentous diatoms which are strongly associated with phosphorus enrichment and eutrophic conditions. Cyanophyta blooms were notably absent from our observations in 2014, however some toxigenic species (Microcystis aeruginosa, Dolichosmpermum crassa) were part of the dominant flora during the summer and early fall. While these results may appear to be encouraging, we suspect that they were driven by extreme weather events and are not part of a general trend. We also found that >50% of the biomass of the microbial – planktonic food web was heterotrophic primarily bacteria; we are concerned that the uptick in bacteria may have negative implications for ecosystem health. The long-term monitoring record maintained since 1972 indicates a reduction in chlorophyll a and phytoplankton biomass but both have still trended towards eutrophic conditions. In 2014, measures of algal standing crop were anomalously low and these results should be interpreted with caution.

Acknowledgements

We thank Robert Bonnell and Ashley Bedford of DFO, as well as our summer students (Joe Marquis, Velina Milkova, Curtis Mosier, Sonya Oetterich, Janine Weber) for the long hours in the field and the lab. We also thank Jennifer Lorimer of the AEHMS for assistance with data analysis and technical editing.

References

Crumpton, W.G., 1987. A simple and reliable method for making permanent mounts of phytoplankton for light and fluorescence micros- copy. Limnol. Oceanogr. 32, 1154-1159.

Kinstler, P., Morely, A., 2016. Point Source Phosphorus Loadings 1965 to 2014. In: Monitoring Report #26. Project Quinte Annual Report 2014, pp. #-#. Bay of Quinte Remedial Action Plan, Kingston, ON, Canada. (this volume) Minns, C.K., Owen, G.E., Johnson, M.G., 1986. Nutrient loads and budgets in the Bay of Quinte, Lake Ontario, 1951-81. In: Minns, C.K., Hurley, D.A., Nicholls, K.H. (Eds), Project Quinte: point-source phosphorus control and ecosystem response in the Bay of Quinte, Lake Ontario. Can. Spec. Publ. Fish. Aquat. Sci. 86, 59-76. Minns, C.K., Moore, J.E., 2004. Modelling phosphorus management in the Bay of Quinte, Lake Ontario in the past, 1972 to 2001, and in the future. Can. Manuscr. Rep. Fish. Aquat. Sci. 2695, v+42p. Montagnes, D.J.S., Lynn, D.H., 1993. A quantitative protargol stain (QPS) for ciliates and other protists. In: Kemp, P.F., Sherr, B.F., Sherr, E.B., Cole, E.J. (Eds), Handbook of methods in aquatic microbial ecology. Lewis Publishers, Boca Raton, FL. Munawar, M., Munawar, I.F., Weisse, T., Leppard, G.G., Legner, M., 1994. The significance and future potential of using microbes for assessing ecosystem health: The Great Lakes example. J. Aquat. Ecosyst. Health. 3, 295-310.

36 Phytoplankton & Microbial Food Web Munawar, M., Munawar, IF., 1996. Phytoplankton dynamics in the North American Great Lakes, Vol. 1. Lakes Ontario, Erie and St. Clair. SPB Academic Publishing, Amsterdam, The Netherlands. Munawar, M., Fitzpatrick, M., Munawar, I.F., Niblock, H., Kane, D., 2012. Assessing ecosystem health impairments using a battery of ecological indicators: Bay of Quinte, Lake Ontario example. Aquat. Ecosyst. Health Mgmt. 15(4), 430-441. Nicholls, K.H., Heintsch, L., Carney, E., 2002. Univariate step-trend and Multivariate assessments of the apparent effects of P loading reductions and zebra mussels on the phytoplankton of the Bay of Quinte, Lake Ontario. J. Great Lakes Res. 28(1), 15-31. Strickland, J.D.H., Parsons, T.R., 1968. A Practical Handbook of Seawater Analysis. Fisheries Research Board of Canada, Bulletin 167, 71–75. Utermöhl, H., 1958. Zur vervolkommnung der quantitativen phytoplankton-methodik. (The improvement of quantitative phytoplankton methodology. In German.) Mitt. Internat. Verein. Limnol. 9, 1-38. Vollenweider, R.A., Munawar, M., Stadelmann, P., 1974. A comparative review of phytoplankton and primary production in the Lauren- tian Great Lakes. J. Fish. Res. Board Can. 31, 739-762.

Munawar et al. 37

ZOOPLANKTON IN THE BAY OF QUINTE – 2014

K.L. Bowen, R. Rozon and W.J.S. Currie

Great Lakes Laboratory for Fisheries and Aquatic Sciences Department of Fisheries and Oceans 867 Lakeshore Road, P.O. Box 5050 Burlington, Ontario, L7S 1A1

Introduction

The zooplankton community in the Bay of Quinte is dynamic. It changes in structure down the length of the Bay from the shallow, eutrophic habitat in the upper bay, represented by the stations at Belleville (B), through eutrophic but slightly deeper conditions in the middle bay near Hay Bay (HB), to a mesotrophic to oligotrophic environment in the mouth of the Bay at Conway (C). Station depth increases from 5 m at B, to 12 m at HB and 32 m at C. However, depth drops off fairly rapidly at C, and in both 2013 and 2014, the mooring was located slightly closer to shore, at a depth of only about 26 m. Thermal stratification occurs throughout the summer at C, sporadically at HB and not at all at B. Zooplankton community structure and productivity is controlled not only by the physical environment but also by the type and quantity of available food. This is determined by nutrient conditions, the structure of the fish community which is the primary source of mortality in this system, and the introduction of new species which either compete for food or alter predation on or within the zooplankton community. Monitoring of zooplankton in the Bay of Quinte tracks species composition, size structure, abundance, biomass and productivity in an effort to understand the response of the community to changes in the controlling factors and assess its ‘health’ as a component of the ecosystem. This report presents data from the 2014 field year in the context of previous data.

Methods

In 2014, zooplankton samples were collected at four primary monitoring stations, B, N, HB and C. Samples were collected on alternate weeks from early May until the end of October for B and HB, but only monthly at N and C due to budgetary and logistical constraints. An extra sample was collected at C in late September in an attempt to capture a possible peak in veligers. At B, N and HB, discrete samples were collected through the water column with a 41-l Schindler-Patalas trap fitted with 64-µm mesh. Three depths were sampled at B and N (1 m, 2 m and 3 m) and five at HB (1 m, 2 m, 3 m, 6 m and 9 m). A single composite sample was constructed for each station-date by combining 50% of the sample from each depth. At C, an epilimnetic (epi) and a metalimnetic-hypolimnetic (MH) vertical net haul were taken with a 64‑µm, 40‑cm diameter Wisconsin-style zooplankton closing net. If the water column was unstratified,

Bowen et al. 39 the epi net was taken from 0-20 m. This occurred in May, June, late September and October. In July, Aug and early September, the bottom of the epilimnion was at 9, 5 and 16 m, respectively. The MH net was taken from the bottom of the epi sample to 1 m off bottom. Densities in the total water column were estimated by weighting the densities from each stratum according to the proportion of the water column that stratum occupied. This assumed a station depth of 28 m. To better quantify the spiny water flea Bythotrephes, additional total water column net hauls were taken with a 153‑µm, 50‑cm diameter Wisconsin-style zooplankton net at N, HB and P (Picton) in late August, and at B, N, HB, GL (Glenora) and C in late October. These samples were counted in their en- tirety for Bythotrephes and Cercopagis. For the remaining samples, a minimum of 400 individual zooplankters and all loose eggs within a subsample aliquot were enumerated from each. The counting method is described in detail in Bowen and Johannsson (2011). Cladoceran mean lengths were estimated over the June 01 to October 06 period. Seasonally-weighted mean (SWM) densities, biomass and total production of zooplankton were calculated over the longer May 01 to October 31 sampling season, based on measured lengths and length weight regression equations (Bowen and Johannsson, 2011). Production was estimated by the egg-ratio method of Paloheimo (1974) for the more abundant taxa, and P/B relationships for the less dominant taxa (Johannsson and Bowen, 2012). The May 01 to October 31 time period is a departure from previous reports where the May 01 to October 6 period was used. The shorter season better represented sampling in the early years of Project Quinte, but since the mid-1990s we have sampled until late October. Bio- mass and production were recalculated for all previous years to reflect this longer period, extrapolating where necessary. Based on 2000 to 2014 data, extending the season from October 6 to October 31 increased total seasonal production by 3.7% at B, 4.1% at HB and 3.2% at C. During the planning for the 2013 Lake Ontario sampling year, there was some dispute that the equation previously used to estimate veliger weights (Hillbricht-Ilkowska and Stanczykowska, 1969) may have in fact represented wet weight. We developed a new length-dry weight regression equation in 2015 (Conway et al., in prep), and all veliger biomass values in this report use this new equation. Be- tween 2008 and 2014, veliger biomass values using the new equation were 71% and 63% of the original values at B and C, respectively.

Rotifers

Rotifer sampling began in 2000 by collecting 1 litre of water from each depth indicated above using a Van Dorn sampler. For each station-date at B and HB, water was pooled and filtered through 20-µm mesh. Rotifers were narcotized using carbonated water and preserved as above. Each year, a seasonal composite sample for each station was made by combining 50% of the sample from each date. Enumer- ation and identification of rotifers in these seasonal composites were performed using a weighted counting method and limits similar to that of the zooplankton (Bowen and Johannsson, 2011). The biovolume (mm3) for each individual was calculated by using formulae of Ruttner-Kolisko (in McCauley, 1984).

40 Zooplankton Results

Belleville

The zooplankton seasonally-weighted mean (SWM) biomass at B in 2014 was 113 mg m‑3, which is similar to the 2008 to 2013 mean of 121 mg m‑3 (Figure 1A). This represents a recovery from the extremely low value of 58 mg m‑3 observed the previous year, but still below the 20 year average of 150 mg m‑3. Biomass at B has been highly variable over the years, but was generally greater in the 1982 to 1991 period, and again in 2001. The lowest values were seen in 1992, 1997, 2000 and 2013. This variability has often been driven by fluctuations in cladoceran biomass, which in 2014 averaged 93 mg m‑3 or 82% of the total (Table 1). Cyclopoid copepod SWM biomass was 10.9 mg m‑3 in 2014, a minor decrease compared to the 2008 to 2013 period when it averaged 11.9 mg m‑3 or 10%. Cyclopoids were more abundant prior to the arrival of dreissenids (1979-1994), when they averaged 47 mg m‑3, or 18% of the total. Calanoid copepods have usually comprised <2% to 4% of the biomass at B. The contribution by dreissenid veligers has been variable since the species invaded in the mid-1990s. In 2014, veliger biomass was 8.3 mg m‑3 (7.3%), very similar to the 1995 to 2013 mean of 7.2 ± 1.1 mg m‑3. Veliger biomass peaked at 21 mg m‑3 or 19% in 2010, but was only 0.2% of the total biomass in 2001. In 2014, rotifers only contributed an additional 1.2 mg m‑3 to SWM biomass, slightly below the 2000 to 2013 mean of 2.9 mg m‑3. The dominant rotifer taxa in 2014 in terms of biomass were Polyarthra vulgaris, Keratella cochlearis, P. major and Trichocerca cylindrica (Figure 2A). Bay of Quinte zooplankton biomass shows strong seasonal patterns, which are influenced by water temperature, phytoplankton abundance, and fish predation. Biomass at B is typically low in early May (Figure 3A). In 2014, biomass began increasing in early June with the appearance of Bosmina, Eubos- mina coregoni and with Daphnia retrocurva appearing throughout July. The cause of the precipitous biomass decline in mid-July and subsequent rapid recovery is unknown. Biomass reached its highest level in early July (521 mg m‑3) when D. retrocurva and Bosmina peaked. A second smaller peak occurred in late July (392 mg m‑3) when D. retrocurva, Bosmina and E. coregoni were abundant. Daphnia and Eubosmina began to decline in early August and remained low for the rest of the season. From late August through early October, biomass averaged 49 mg m‑3, and dropped to 27 mg m‑3 by late October. Bosmina was again the dominant taxon in the fall, with much lower levels of the cladoceran Chydorus sphaericus, veligers and cyclopoid copepodites. Total zooplankton seasonal production has fluctuated in recent years and generally mimics biomass (Figure 1D). These variations are largely driven by changes in cladoceran production, such as the extremely high Daphnia reproduction rates at B in 2012. In 2014, total seasonal dry production was 2 432 mg m‑3 for the 01 May to 31 Oct period, the second lowest level since 2006, and below the 2002-2013 mean of 3 163 mg m‑3. When rotifers were excluded, cladocerans contributed 88% of total production in 2014 (Table 2), and the bosminids (Bosmina + Eubosmina) together contributed 53%. Cyclopoids and calanoids each made up 5% or less of the total, while veligers made up 6.2% of the

Bowen et al. 41 total production. Rotifer production in 2014 was 44.3 mg m-3, representing 1.3% of total zooplankton + rotifer production.

Figure 1. Trends in seasonally-weighted mean dry biomass (A-C) and total seasonal production (D-F) of zooplankton groups at B, HB and C. These are volumetric estimates based on measured lengths from May 1 to October 31. Time stanzas are indicated by the arrows: implementation of phosphorus control (PC) and invasion by dreissenid mussels (DM). Note the different scale used for C. Rotifer sampling began in 2000, but rotifers were not sampled at C in 2013 or 2014. * indicates no samples were collected that year.

42 ZoopLankton Figure 2. Mean May to October dry biomass of dominant rotifer groups in the Bay of Quinte from 2000 to 2014 at B and HB, and 2000 to 2009 at C.

boWen et aL. 43 Table 1. Distribution of zooplankton SWM biomass (May 1 - Oct. 31) amongst the taxonomic groups in the Bay of Quinte at B, N, HB and C in 2014.

The May-October SWM zooplankton density at B was 112.3 no. l-1 in 2014. Numerically the most dominant taxa were Bosmina sp. (29%), veligers (18%), E. coregoni (11%), Ceriodaphnia (10%), D. retrocurva (8%), cyclopoid copepodids (8%), cyclopoid nauplii larvae (7%) and C. sphaericus (3%) (Table 3). D. galeata mendotae was less abundant, but it was still important given its larger size and high productivity. Most of the cyclopoids found at all stations were juveniles not identifi ed to species, and the most dominant adult cyclopoids at B were Diacyclops thomasi in the spring and Tropocyclops extensus in the summer and early fall. Neither the invasive cladoceran Cercopagis pengoi nor the spiny water fl ea Bythotrephes were found at B (Figure 4). The B June 1- October 6 mean cladoceran length has fl uctuated widely in the last decade. The 2014 mean cladoceran length at B was 0.42 mm, the highest since 2007 and nearly identical to mean values from 1972-2013.

Napanee

Station N was sampled monthly in 2014, yielding a total of 7 dates - the same days as C. Densities, biomass, production, community composition and seasonal trends at N were more similar to HB than B. Biomass at N averaged 78.8 mg m-3 across the season and were comprised of 76% cladocerans, 14% cyclopoids and 5% each of calanoids and veligers (Table 1). Biomass peaks at N were considerably lower than at B, reaching 203 mg m-3 on 14-Jul (Figure 3B). July biomass was dominated by D. galeata, D. retrocurva and E. coregoni. After August, Daphnia declined and were replaced by Bosmina, C. sphaericus and veligers. A smaller peak of 85 mg m-3 was seen on 8-Sep. Veligers represented between 2.5% and 12% of the total biomass in 2014. Daphnia represented over 50% of the total biomass in July, whereas Bosminids represented 74% of the total biomass on 8-Sept. Calanoids and other cladocerans remained at fairly low levels throughout the season.

44 ZoopLankton Figure 3. May to October 2014 seasonal trends in volumetric dry biomass at B, Napanee, HB and C and in the Bay of Quinte. Bosminids include Eubosmina and Bosmina and Daphnia includes D. galeata mendotae and D. retrocurva. “Other Herb. Clad.” represents the remaining herbivorous cladocerans, particularly Ceriodaphnia and Chydorus sphaericus. Note the reduced sampling frequency at C and Napanee and the different scale at C.

Total seasonal production was 1371 mg m-3 at N, and was comprised of 83% cladocerans (Table 2). Total SWM zooplankton density was 89.6 no. l-1, and was dominated by the same taxa as B (Table 3). Only on one occasion was the invasive Cercopagis pengoi found at Napanee (23-Sept), with a density of 0.008 no. l-1. Bythotrephes were not found at N in 2014.

Hay Bay

In 2014, SWM zooplankton biomass at HB (95 mg m-3) was the lowest ever recorded at HB. Similar low years at HB were 1975 and 2001 (each 112 mg m-3). Prior to 2014, zooplankton biomass at HB was relatively stable (2004 to 2013 mean of 151 mg m-3; Figure 1B), with a 20 year mean of 161.5 mg m-3. Before 1991, variability was caused by changes in both cladoceran and cyclopoid biomass, whereas in recent years cyclopoids have been lower and more stable. Mean cladoceran biomass in 2014 was 74 mg m-3 representing 78% of the total biomass. Although this represents the lowest level of Cladoc- era since 1975, it is nearly identical to the proportion of cladoceran biomass compared to the previous year. This group has always been dominant in the Bay of Quinte, and Lake Ontario in general, in the 1980s and much of the 1990s. Mean cyclopoid biomass was stable between 2012 (22 mg m-3) and 2013 boWen et aL. 45 (29 mg m-3), but fell in 2014 (15 mg m-3) only representing 16 % of the total biomass. Veliger biomass peaked at HB in 2008 (22 mg m-3) and has shown a declining trend since. Excluding 2008, veliger biomass has averaged 5.2 mg m-3 since 1995 when it had recently invaded. The 2014 mean was only 2.6 mg m-3 (2.7%). Biomass contributions by both calanoids and rotifers were small (3.0 mg m-3 (3.2%) and 2.5 mg m-3 (2.5%), respectively). Rotifer biomass has generally declined since 2010 when the large rotifer Asplanchna was common (Figure 2B). The dominant rotifers in 2014 were P. vulgaris, Asplanch- na sp., Conochilus unicornis and K. cochlearis. There were several biomass peaks through the season at HB, the highest reaching 337 mg m-3 on 12-Aug and 180 mg m-3 in mid-July (Figure 3C). Bosmina was one of the dominant taxa through most of the sampling season, except in mid-August when it was largely replaced by the somewhat larger E. coregoni. The cladocerans D. galeata mendotae and D. retrocurva began climbing in June, with a peak on August 12th dominated by D. galeata mendotae. The herbivorous cladoceran Ceriodaphnia appeared in mid-July, and C. sphaericus in August and September. Veliger larvae were present at low levels throughout the sampling season, and showed small increases in early summer and again in early fall (max. 7.5 mg m-3). Cyclopoid biomass remained at fairly low levels through the season, with an August peak (71.1 mg m-3) largely driven by copepodids (juveniles).

Table 2. Distribution of zooplankton total seasonal production (May 1 - Oct. 31) amongst the taxonomic groups in the Bay of Quinte at B, N, HB and C in 2014.

46 ZoopLankton Table 3. 2014 seasonal mean abundances of major zooplankton taxa and percent contribution to total density at stations B, N, HB and C. (no. l-1). Shaded values indicate dominant taxa (>5%).

boWen et aL. 47 Figure 4. Seasonal weighted mean density (SWM) of the invasive predatory cladocerans Cercopagis and Bythotrephes in the Bay of Quinte. Note that 2010 to 2012 was not sampled at C, and 2009 was not sampled at HB.

Total seasonal production has shown small fl uctuations at HB since 2000, largely driven by changes in cladoceran production (Figure 1E). In 2014, production fell to 1 228 mg m-3. This was nearly half of the 2002-2012 mean of 2 355 mg m-3. In 2014, cladocerans, cyclopoids and veligers contributed 86%, 9% and 4%, respectively, to the total production. Bosminids contributed 40% of the production total. Rotifers contributed only 58 mg m-3 at HB in 2014, or 4.5% of the total zooplankton + rotifer production. The SWM density at HB in 2014 was 77.8 no. l-1. Not surprisingly, the small taxa were numerically dominant – Bosmina (26%), followed by cyclopoid copepodids (12%), C. sphaericus (11%), veligers (10%) and cyclopoid nauplii (9%) (Table 3). Other important, larger taxa included Ceriodaphnia, D. retrocurva, D. galeata mendotae, E. coregoni, and the copepod T. extensus. Densities of the invasive cladoceran Cercopagis pengoi remained low at HB in 2014, and the spiny water fl ea Bythotrephes was not found (Figure 4). Small cladoceran taxa dominated the HB zooplankton community in 2014, and mean cladoceran length was 0.43 mm. This was slightly higher than the other values seen over the last decade. Values were also generally higher in the late 1980s and 1990s.

48 ZoopLankton Figure 5. Biomass comparison of dominant zooplankton taxa in epilimnetic (epi) vs deep (meta-hypo) 64-µm plankton net hauls at C. Cladoceran taxa include Bosminids (Bosmina and Eubosmina), Daphnia, Ceriodaphnia and Cercopagis. Cyclopoid copepods include Diacyclops thomasi, Meso- cyclops edax, Tropocyclops extensus and copepodids (juveniles), and calanoid copepods include Leptodiaptomus sicilis, Skistodiaptomus oregonensis, Leptodiaptomus minutus, Eurytemora and copepodids. Note the different scales and extreme values. boWen et aL. 49 Conway

C was sampled in 2014 a total of 7 times. The SWM zooplankton biomass at C in 2014 was 60.9 mg m‑3 (Figure 1C), which is slightly above the 20 year mean of 46.2 mg m‑3. Although variable over the 40 year time series, biomass at this mesotrophic lower bay station tends to be much lower than in the upper and middle reaches of Quinte. Biomass at C has been fairly consistent between 2008-2009 and 2013-2014, and higher than in the low periods of 1999 to 2001 and 2004 to 2007 (17 to 40 mg m‑3). This rise has largely been due to an increase in veliger biomass in the latter years. For example, veligers rose from 8% of the total biomass in 2007 to 48% in 2009. In 2014, veligers contributed 17 mg m‑3 or 28% of the total. Cladocerans contributed 31 mg m‑3 (50%) whereas cyclopoids contributed about 11 mg m‑3 (17%). In 2013, mean cyclopoid biomass reached its highest level (25.1 mg m‑3 or 32%) since the mid-1990s, but fell in 2014 to10.5 mg m‑3 (17%). This biomass value is still slightly higher than most annual means in the 2000s. In contrast, calanoids remained only a small contributor at C in 2014, adding only about 2.8 mg m‑3 to total biomass (5%). Rotifers were not collected at C. Biomass peaked at C only once in 2014, reaching 141 mg m‑3 on 11-Aug (Figure 3D). June biomass was dominated by veligers, Bosmina and cyclopoids, particularly juveniles, and some D. retrocurva. These taxa remained present throughout the summer and declined near the end of the season. The dominant adult cyclopoid was Mesocyclops edax in mid-summer, and D. thomasi as well as T. extensus later in the season. In September, biomass fell to 23 mg m‑3 and remained low for the rest of the season, dominated by cyclopoids and veligers. From August until October, veligers represented between 14% and 48% of the total biomass at C in 2014. D. retrocurva and D. galeata mendotae peaked in August, representing 29% of total biomass. Calanoids and other cladocerans remained at fairly low levels throughout the season, with the exception of a peak in Holopedium gibberum on 11-Aug. Throughout Project Quinte, total seasonal production at C has consistently been much lower than the Upper and Middle Bay stations (Figure 1F). Production to biomass (P:B) ratios tend to decline from the upper to lower bays, and in 2014, these values were 21.5 at B, 17.4 at N, 14.0 at HB and 11.8 at C (Table 2). In 2014, production at C was 650 mg m‑3, representing a decline relative to the previous year. Since 2008, levels have been consistently higher than most years since 1992, although 2014 had the lowest production since 2007. As with biomass, this recent upswing has largely been due to the increased contribution by veligers. In 2014, veligers contributed 43% of the total production, with cladocerans and cyclopoids contributing 46% and 9% respectively. Rotifer data were not available at C in 2014. The SWM density at C in 2014 was 67.7 no. l‑1. Veliger larvae numerically dominated at C (57%), a proportion much higher than at the other stations (10-20%). In contrast to the other stations, Bosmina only comprised 7% of the total at C. Other important, larger taxa included Ceriodaphnia, D. galeata, E. coregoni, D. retrocurva, and D. thomasi (Table 3). As at HB, small cladoceran taxa dominated the C zooplankton community in 2014, and mean cladoceran length was 0.48 mm. The invasive Cercopagis pengoi was found at C in low numbers, averaging 0.10 no. l‑1 across the season. The highest numbers (0.57 no. l‑1) were seen on 14-Jul. The spiny water flea Bythotrephes, which was only found in Septem-

50 Zooplankton ber and October at C, were considerably less abundant than the previous year and were found at a mean density of only 1.4 x10-4 no. l‑1 (Figure 4). In 2014, a comparison between Epi and Meta-Hypo (MH) sampling depth was done at C to examine the depth preference of different species. This was done to examine sampling bias prior to 2013 when Schindler-Patalas samples tended to over-represent the epilimnion (Figure 5). The bosminids, Daphnia, Ceriodaphnia sp., and T. extensus showed a strong preference for EPI, as was expected based on known ecology of these species. The copepodids, the cyclopoid D. thomasi, and the calanoids Skistodiaptomus oregonensis and Leptodiaptomus sicilis demonstrated a seasonal progression and preference toward the MH in the later part of the year when the epilimnion was warm. Veligers tended to prefer the Epi except on 11-Aug; the opposite was true for the cyclopoid M. edax. Given that Schindler traps were used more frequently in the epi as compared to deeper depths, this could result in an over estimation of bosminids and cladocerans while cyclopoids and calanoids might be underestimated in data prior to 2013.

Discussion

Overall, the zooplankton community in the Bay of Quinte has fluctuated markedly over time. Total zooplankton densities and biomass in the upper and middle bay are currently at low levels compared to the average over the last twenty years, and are particularly diminished relative to the 1980s and 1990s. This decline was most apparent at HB in 2014. In contrast, zooplankton biomass values in the lower bay in 2013 and 2014 were above the 20 year average, and are more similar to those seen in the 1980s. Howev- er, community composition has shifted at C, and high populations of veligers in recent years have largely offset the loss of cladocerans and cyclopoids. Domination by veligers reflects the large populations of adult quagga mussels in the lower bay and in Lake Ontario (Wilson et al., 2006, Dermott and Bonnell, 2012). Adult quaggas show a marked increase in 2008 and 2009 in the lower Bay of Quinte relative to the previous decade, which corresponds to the increase in veligers at C starting in 2008 (Dermott and Bonnell, 2012). Veligers tend to be much less important in the upper and middle reaches of Quinte, which is expected given the relatively low population of adults in these areas. The role of veligers in the planktonic food web needs further study in Lake Ontario and its embayments. Community composition at B, N and HB has been relatively consistent in last decade, with the zooplankton dominated by herbivorous cladocerans, especially bosminids, Daphnia, Ceriodaphnia and Chydorus. Densities of Bosmina and Eubosmina tend to be highest in the upper bay and decline down the length of the Bay. Cyclopoids have dropped markedly over the last 15 years compared to the 1980s and 1990s, especially in the middle and lower reaches where they were more dominant, but the small juvenile stages are still numerically important. Cyclopoid populations recovered somewhat at C in 2013, reflecting a lakewide increase observed during the large-scale 2013 Lake Ontario CSMI program (CSMI 2013 Coordinating Committee, 2014). However, this change may have been short-lived, as cyclopoid populations in 2014 were only slightly above the 1999 to 2009 average of 7.1 mg m‑3. Cyclopoids such as D. thomasi and M. edax are predatory as adults, consuming soft-bodied rotifers, copepod nauplii

Bowen et al. 51 and copepodids and small cladocerans (McQueen, 1969; Lane, 1978), and are in turn preyed upon by planktivorous fish. The food web ramifications of the reductions in cyclopoids in Quinte are unknown. Calanoid copepod abundance tends to be low in Quinte, but this group is proportionally more dominant in the lower reaches. This is not surprising given that most calanoids prefer cooler, deeper waters. Skistodiaptomus oregonensis was the most dominant calanoid in Quinte in 2014. While this species is usually more abundant offshore than in the littoral zone, it tends to be found in the epilimnion (Balcer et al., 1984). This suggests that it can tolerate the warm temperatures in of the Bay of Quinte. It was also one of the most common calanoids in central and eastern Lake Erie in August 1998 (Barbiero et al., 2001) and in the epilimnion of Lake Ontario during the early fall in 2003 and 2008 (Rudstam et al., 2015). Although numerically abundant, rotifer biomass in Quinte remains at low but near average levels. This group generally contributes less than 10% of total biomass and seasonal production. Unfortunately long-term trends in rotifer populations cannot be determined as their enumeration only began in 2000. The rotifers dominating the Bay are ubiquitous species found throughout the Great Lakes (Makarewicz and Lewis, 2015; Barbiero and Warren, 2011). Despite the recent declines in biomass in the Bay of Quinte, it is still an extremely productive system relative to Lake Ontario. Lake Ontario has also seen precipitous declines in zooplankton during the last decade (Rudstam et al., 2015). During the 2008 CSMI lakewide monitoring program, nearshore epilimnetic biomass values were an order of magnitude or more lower than in Quinte, averaging only 3.7, 6.8 and 23.2 mg m‑3 in spring, summer and fall, respectively (Rudstam et al., 2015). Zooplankton populations in the Bay of Quinte are still indicative of an eutrophic environment, although the system may be edging toward mesotrophy in recent years. Seasonal zooplankton succession patterns followed expected trends in 2014, with bosminids and Daphnia dominating from June to September. Cyclopoid species shifted from D. thomasi dominating early in the year, to M. edax and T. extensus later in the summer and fall. Community composition tends to be fairly low and stable during May, September and October. The reason for the precipitous short- term decline in zooplankton at B in mid-July is unknown and is not matched at other sites. It is possible that this is an artifact of sampling error during that cruise. The reasons for zooplankton population fluctuations over the past 39 years have been proposed to include changes in phosphorus loading, water temperature, water clarity, invasive species, variations in the phytoplankton community, and fish predation (Bowen and Johannsson, 2011, Bowen and Gerlofsma 2012, Johannsson and Bowen, 2012, Nicholls and Carney, 2011). Zooplankton biomass and production dropped precipitously in 1992 at all three stations, and has generally remained lower in the last two decades than during the peak pre-dreissenid period of 1982 to 1991. Biomass and community structure was relatively stable at B between 2008 and 2014, although biomass dropped markedly in 2013 due to a decline in cladocerans. Cladoceran production has also fluctuated at this station in recent years. In 2013, planktivorous fish densities appeared to be very high at B as shown by the Ontario Ministry of Natural Resources and Forestry bottom trawling program (OMNRF, 2016). This program indicates that in 2013,

52 Zooplankton counts of Alewife, White Perch and Gizzard Shad were at their highest levels in the last twelve years, and zooplankton biomass, particularly Daphnia, was supressed. Other planktivores in the upper bay include shiners and the juveniles of many species such as yellow perch and even walleye. Planktivore biomass at B was much reduced in 2014, with the lowest catches in the trawl and gill nets observed since 1972. The largest declines were seen in Alewife, Gizzard Shad and White Perch, although Yellow Perch numbers were up. Low predation rates may have contributed to the recovery of zooplankton observed at B in 2014, although the fact that zooplankton biomass only reached “average” levels suggests that other factors were at play. The extremely cold winters in 2013-2014 may have also influenced zooplankton populations throughout the Bay, but the mechanisms for this are unclear. Unlike the winters, the May to Sept. mean air temperatures at the Trenton airport were very close to the 1972 to 2015 average (Me- teorological Service of Canada, http://climate.weather.gc.ca/climate_data/daily_data_e.html?Station- ID=5126, accessed May 6, 2016). Total planktivore densities in the middle bay were only at moderate levels in 2014 and appeared to be relatively unchanged relative to the previous year (OMNRF, 2016). Therefore it seems unlikely that planktivory played a large role in the precipitous decline of zooplankton at this station in 2014. Plank- tivory at C also appeared be relatively low in 2014, especially in terms of adult Alewife caught in the OMNRF gillnets. Trawl catches of Alewife and Rainbow Smelt were at the highest level since 2011. Total zooplankton production and the production to biomass (P:B) ratio was lowest at C. This is in part due to the warmer temperatures in the upper bay which enhance production, especially early in the summer. The upper bay is also largely comprised of herbivorous cladocerans, and production for a given biomass tends to be higher for this group than for copepods and veligers. Cladoceran populations can rise and fall very quickly to take advantage of suddenly appearing food sources such as blooms of algae, largely due to their ability to reproduce rapidly by cloning (parthenogenesis). Changes in both the abundance and composition of algal communities also influence production of herbivorous zooplankton such as Daphnia. Blooms of inedible, potentially toxic cyanophytes and other large colonial algae frequently occur in the upper and middle reaches of the Bay of Quinte in late summer and early fall (Munawar et al., 2012; Nicholls and Carney, 2011). In addition to potentially interfering with the filtering apparatus of cladocerans (e.g., de Bernardi and Giussani, 1990), these filamentous “bloom” taxa may outcompete smaller, more edible phytoplankton and cause food limitation in zooplankton. On the other hand, large numbers of cladocerans and their competitors, dreissenid mussels, have the ability to “graze down” desired forms of phytoplankton and can contribute to improved water clarity. There is some evidence that bottom-up impacts (phytoplankton food supply) were limiting zooplankton populations in 2014. Size fractionated primary productivity (SFPP) reached its lowest mean level at B since measurements started in 2000 (Munawar et al., 2016). The 2014 SFPP value at HB was also the lowest since 2010 when regular measurements began. While mean total phosphorus (TP) as measured by NLET were fairly consistent over the last few years at B and HB, mean uncorrected chlorophyll a levels as measured by GLLFAS were among the lowest values at both stations since 1977

Bowen et al. 53 (Munawar et al., 2016). Conversely at C, TP reached its highest seasonal mean value since NLET began measurements in 1988, and the mean chlorophyll a value was also among the highest recorded. Anec- dotally, the presence of periodic algal blooms in the lower bay have been noted by the GLLFAS field crew in the last few years. This increase in algal biomass may be supporting the higher zooplankton biomass and production, particularly in veligers, observed since 2008. In summary, zooplankton densities and biomass in the Bay of Quinte are currently at low levels at B and HB, but have increased in recent years in the lower bay due to high numbers of veligers. The system is dominated by cladocerans including bosminids and Daphnia, and veligers are also numerically dominant in the lower bay. Cyclopoids have declined in the last decade or so, but are still an important part of the community, especially in the middle and lower reaches. Zooplankton populations and community composition are determined by a number of factors, including both top-down and bottom-up effects. These include changes in the abundance of planktivorous fishes such as Alewife, invertebrate predators, and changes in phytoplankton primary production rates, biomass and community composition, such as domination by inedible filamentous cyanophytes. Furthermore, fluctuations in temperature may also influence zooplankton and phytoplankton populations in the Bay.

References

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, WI. Barbiero, R.P., Warren, G.J., 2011. Rotifer communities in the Laurentian Great Lakes, 1983–2006 and factors affecting their composition. J. Great Lakes Res. 37, 528–540. Barbiero, R.P., Little, R.E., Tuchman M.L., 2001. Results from the U.S. EPA’s Biological Open Water Surveillance Program of the Lau- rentian Great Lakes: III. Crustacean Zooplankton. J. Great Lakes Res. 27(2), 167–184. Bowen, K.L, Gerlofsma, J. 2012. Zooplankton in the Bay of Quinte – 2009 and 2010. In: Monitoring Report #21. Project Quinte Annual Report 2010, pp. 75-98. Bay of Quinte Remedial Action Plan, Kingston, ON, Canada. 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 the predatory cladoceran Cercopagis pengoi: 1975-2008. Aquat. Ecosyst. Health Manage. 14, 44-55 de Bernardi, R., Giussani, G., 1990. Are blue-green algae a suitable food for zooplankton? An overview. Hydrobiologia 200/201, 29-41. Conway, A.J., Bowen, K.L., Currie, W.J.S., In Prep. Could Dreissenid veligers be the lost biomass of invaded lakes? Limnology and Oceanography Letters. CSMI 2013 Coordinating Committee, 2014. Lake Ontario 2013 CSMI Progress Report. http://www.dec.ny.gov/docs/water_pdf/ csmi2013progrpt2.pdf. Dermott, R., Bonnell, R., 2012. Benthic Fauna of the Bay of Quinte in 2010. In: Monitoring Report #21. Project Quinte Annual Report 2010, pp. 99-115. Bay of Quinte Remedial Action Plan, Kingston, ON, Canada. Hillbricht-Ilkowska, A., Stanczykowska, A., 1969. The production and standing crop of planktonic larvae of Dreissena polymorpha Pall. in two Mazurian lakes. Pol. Arch. Hydrobiol. 16, 193-203.

54 Zooplankton 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. Lane, P.A., 1978. The role of invertebrate predation in structuring zooplankton communities. Int. Ver. Theor. Angew. Limnol. Verh. 20(1), 480-485. Makarewicz, J.C., Lewis, T.W., 2015. Long-term changes in Lake Ontario rotifer abundance and composition: A response to Cercopagis predation? J. Great Lakes Res. 41, 192-199. McCauley, E., 1984. The estimation of the abundance and biomass of zooplankton in samples. In: Downing, J.A., Rigler, F.H. (Eds.), A Manual on Methods for the Assessment of Secondary Productivity in Fresh Waters, Second Edition, pp. 228–265. Blackwell Sci- entific Publications, Oxford. McQueen, D.J., 1969. Reduction of zooplankton standing stocks by predaceous Cyclops bicuspidatus thomasi in Marion Lake, British Columbia. F. Fish Res. Board Can. 26(6), 1605-1618. Munawar, M., Fitzpatrick, M., Niblock, H., Lorimer, J., 2012. The spatial distribution of microbial-planktonic communities and size fractionated primary Productivity in the Bay of Quinte, 2010: A brief overview. In: Monitoring Report #21. Project Quinte Annual Report 2011, pp. 69-74. Bay of Quinte Remedial Action Plan, Kingston, ON, Canada. Munawar, M., Fitzpatrick, M., Niblock, H., Lorimer, J., 2016. The spatial distribution of microbial-planktonic communities and size fractionated primary Productivity in the Bay of Quinte, 2014: A brief overview. In: Monitoring Report #24. Project Quinte Annual Report 2014, pp. XXX. Bay of Quinte Remedial Action Plan, Kingston, ON, Canada. 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 Mgmt. 14(1), 33–43. Ontario Ministry of Natural Resources and Forestry (OMNRF), 2016. Lake Ontario Fish Communities and Fisheries: 2015 Annual Report of the Lake Ontario Management Unit. Ontario Ministry of Natural Resources and Forestry, Picton, Ontario, Canada. Paloheimo, J.E., 1974. Calculation of instantaneous birth rate. Limnol. Oceanogr. 19, 692-694. Rudstam, L.G., Holeck, K.T., Bowen, K.L., Watkins, J.M., Weidel, B.C., Luckey, F.J., 2015. Lake Ontario zooplankton in 2003 and 2008: Community changes and vertical redistribution. Aquat. Ecosyst. Health Mgmt. 18(1), 43-62. Wilson, K.A., Howell, E.T., Jackson. D.A., 2006. Replacement of Zebra Mussels by Quagga Mussels in the Canadian Nearshore of Lake Ontario: the Importance of Substrate, Round Goby Abundance, and Upwelling Frequency. J. Great Lakes Res. 32, 11–28.

Bowen et al. 55

FISH POPULATIONS IN THE BAY OF QUINTE, 2014

J. A. Hoyle

Ontario Ministry of Natural Resources, Lake Ontario Management Unit Glenora Fisheries Station, 41 Hatchery Lane, Picton, Ontario, K0K 2T0

Introduction

This report updates long-term abundance trends of Bay of Quinte fish populations to 2014. Fish community sampling programs have existed in the Bay of Quinte since the 1950s, initially using gill nets and later adding trap nets and bottom trawls. While each gear has unique bias with respect to habitat and species selectivity, collectively, these programs provide a comprehensive data set within which the long-term dynamics of the Bay of Quinte fish community can be examined. Observed major changes in species dominance and abundance have been related to large-scale ecological events associated with nutrient levels and invasive species. Hurley and Christie (1977) described a Bay of Quinte fish community depreciated by the impacts of cultural eutrophication in the 1960s and 1970s. Hurley (1986) evaluated the impact of point-source phosphorus control implementation in municipal sewage treatment plants, during the winter of 1977-78, on Bay of Quinte fish populations. Hoyle et al. (2012) documented major changes in the fish community following dreissenid mussel invasion in the early 1990s, and related these changes to increased water clarity and the return of submerged aquatic macrophytes to vast areas of the Bay. Round Goby invaded the Bay of Quinte in 1999 and by 2003 provided the missing food-web link between dreissenid mussels and piscivores but have not led to major changes in fish species dominance to date.

Sampling programs

Long-term fish community sampling programs, first initiated on the Bay of Quinte in the late 1950s (Hurley, 1986), provide intensive geographic coverage from the mouth of the Trent River in the upper bay to Lake Ontario in the lower bay. Gill nets and bottom trawls sampled offshore habitats while trap nets were used nearshore. The nearshore is distinguished from the offshore at approximately 5 m depth. Bottom trawls sampled all sizes of fish, particularly smaller ones, whereas gill nets and trap nets more effectively sampled large-bodied fish. Here, only summer (June-September) data are presented. De- tailed sampling methods have been reported elsewhere (e.g., Hoyle, 2012).

Data Analysis and Presentation

In this summary report, only the catches of dominant species, as ranked in Hoyle et al. (2012), are included. Species-specific catches, by number, were averaged across geographic area of the Bay (except

Hoyle 57 for trap nets; only upper Bay of Quinte is reported here since the lower bay is not sampled annually), and then summarized and reported as an overall mean across years prior to 1978 (phosphorus time-stanza) and from 1978-1994 (post-phosphorus time-stanza). After 1994 (dreissenid mussel time-stanza), mean annual catches were reported. These results are reported in Tables 1, 2 and 3 for gill nets, bottom trawls, and trap nets, respectively.

Walleye

Walleye is one of the most important members of the Bay of Quinte fish community. As the dominant piscivore, it plays a pivotal role in fish community trophic structure. The Bay of Quinte recreational fishery is centered on Walleye. Abundance trends in gill nets (juvenile and adult fish) and bottom trawls (YOY) are illustrated in Fig. 1 and Fig. 2. Walleye abundance in gill nets has been relatively steady in the dreissenid mussel time-stanza (Fig. 1) and should remain so given recent recruitment trends (Fig. 2). The 2014 Walleye year-class was exceptionally strong.

Ecosystem health indices

Indices have been developed based on trap netting to evaluate fish community and ecosystem health (Fig. 3 and Fig. 4).

Piscivore biomass

A proportion of the fish community biomass comprised of piscivores (PPB) greater than 0.20 reflects a healthy trophic structure (e.g., predator vs. prey balance in the fish community; Hoyle et al., 2012). PPB in 2014 was 0.35.

Index of Biotic Integrity

An Index of Biotic Integrity (IBI) was developed that combined 11 fish community metrics describing attributes of diversity, trophic structure and abundance (Hoyle and Yuille, In Press). The IBI is a measure of overall ecosystem health. IBI increased from 47 in the phosphorus time-stanza to 63 post-phosphorus control and finally to 72 in the dreissenid time-stanza (Fig. 4). IBI in 2014 was 73.

58 Fish Populations Figure 1. Walleye catch in gill nets in the Bay of Quinte from 1972-2014.

Figure 2. Young-of-the-year (YOY) Walleye catch in bottom trawls in the Bay of Quinte, 1972-2014 (no trawling in 1989). hoyLe 59 Figure 3. Proportion of the fi sh community biomass comprised of piscivores in trap nets in the Bay of Quinte (upper Bay only) from 1969-2014 (not all years sampled). Dashed line indicates 0.2 target PPB value.

Figure 4. Index of Biotic Integrity (measures with trap nets) in the Bay of Quinte (upper Bay only) from 1969-2014 (not all years sampled). Dashed lines represent mean IBI for time-stanza.

60 Fish popuLations ------0.417 3.861 10.544 0.475 0.05 0.033 0.083 - 0.017 - 0.017 - - 0.775 Round Goby 0.091 3.617 4.25 5.979 6.106 5.283 3.333 4.867 6.433 2.433 2.258 4.172 3.614 9.425 4.381 2.428 2.856 1.739 2.822 1.353 2.681 2.45 3.943 - Freshwa Drum ter 0.013 0.002 - - - - 0.056 ------0.083 - - - 0.056 - - - - 0.01 Large - mouth Bass 40.032 19.518 8.292 0.083 0.139 0.206 5.528 0.417 0.667 2.167 0.667 0.072 14 0.25 0.442 1.297 - 1.056 29.478 3.736 1.792 - 3.514 Gizzard Shad 1.881 6.504 6.542 5.271 5.772 4.85 4.478 3.333 4.125 4.908 2.244 4.169 1.533 2.4 1.775 2.461 2.628 4.206 2.1 3.267 2.158 2.99 3.561 White Sucker 0.005 - - 0.083 0.222 0.556 0.278 0.583 2.375 1.25 0.167 0.111 0.833 2.167 1.778 1.056 1.85 2.222 2.319 0.389 3.778 1.44 1.173 Bluegill 0.479 16.385 11.833 9.979 6.072 6.217 5.178 4.267 2.633 4.208 3.9 3.653 2.603 4.667 3.311 3.839 4.033 4.578 3.253 2.306 6.831 2.92 4.814 Walleye 0.701 0.257 0.125 0.667 0.778 4.806 3.528 4.333 6.058 3.1 0.503 2.153 0.236 1.292 0.433 0.583 0.472 0.306 0.528 0.333 0.833 0.39 1.573 Pumpkin - seed 2.228 2.079 1.625 1 1.65 2.306 2.75 2.458 2.558 1.875 0.728 0.928 0.453 1.258 0.503 0.389 0.222 0 0.072 0.167 0.389 0.19 1.076 Brown Brown Bullhead 222.559 143.982 24.736 8.563 16.756 19.65 10.644 5.392 9.167 3.967 2.556 6.872 23.233 6.6 7.7 24.042 29.122 68.367 62.981 39.525 30.072 20.55 21.025 Alewife 134.78 51.352 28.5 12.813 15.5 16.872 22.672 16.35 7.5 13.933 23.019 49.217 26.542 75.908 45.5 49.744 44.231 31.267 32.111 48.725 67.364 6.06 31.691 White Perch Yellow Yellow Perch 55.652 108.703 108.583 106.583 144.417 183.493 162.083 129.083 133.417 124.167 91.494 76.272 91.25 80.083 45.567 62.658 74.256 89.806 55.847 21.606 19.464 44.09 92.211 - (1978-1994) time-stanzas, annual catches from 1995-2014, and the mean for 1995-2014. rus mean mean mean Time-stanza Post-phospho 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 1995-2014 Phosphorus Table 1. Catch per Table gill net for major Bay of Quinte fish species. Shown are mean catches for phosphorus (1972-1977) and post-phosphorus

Hoyle 61 ------0.15 1.69 106.11 35.23 72.46 17.42 72.55 35.92 33.13 51.06 89.14 72.00 35.26 30.13 32.61 Round Goby 0.01 1.81 5.39 6.24 3.79 8.19 3.82 2.39 18.90 8.97 6.31 6.50 34.87 37.57 85.58 5.97 6.63 12.64 6.01 6.86 7.49 2.42 13.83 - Freshwa Drum ter 0.01 - 0.40 0.08 0.59 0.05 0.46 0.19 0.29 0.97 0.41 0.10 0.68 0.53 0.02 0.73 0.84 1.77 1.75 2.32 1.36 0.01 0.68 Large - mouth Bass 29.11 65.50 183.51 18.56 14.23 7.99 62.77 168.93 12.15 26.20 8.62 53.50 10.73 15.18 8.89 47.62 31.48 57.32 88.10 142.90 91.62 0.02 52.52 Gizzard Shad 0.20 4.80 1.28 1.74 2.08 2.54 2.07 5.69 46.30 10.06 4.47 5.55 1.82 1.55 5.64 1.30 0.32 1.70 2.01 1.50 2.21 1.69 5.08 White Sucker - 0.00 0.13 0.15 0.34 0.29 0.05 11.31 10.56 1.23 1.73 0.14 0.88 1.06 2.00 1.75 0.98 1.02 2.38 0.84 0.70 0.71 1.91 Bluegill 0.07 9.28 10.45 5.04 3.00 2.68 3.45 1.63 5.71 2.20 4.48 4.39 2.68 3.92 5.64 10.61 4.69 4.75 7.32 3.49 1.78 9.07 4.85 Walleye 0.08 0.58 4.45 11.56 12.78 10.49 29.26 62.81 32.25 13.88 19.07 11.29 5.96 18.46 12.07 16.34 9.18 11.04 8.94 4.88 13.38 5.81 15.70 Pumpkin - seed 8.42 5.80 14.41 8.42 7.98 18.25 19.11 10.89 21.60 10.05 8.43 10.10 7.53 12.07 6.44 1.80 2.05 1.96 0.84 1.63 4.74 2.76 8.55 Brown Brown Bullhead 388.87 443.15 139.11 41.29 60.16 30.20 294.03 261.93 216.12 44.84 38.97 63.23 75.22 156.00 320.29 319.64 258.63 394.99 416.25 224.21 295.25 254.60 195.27 Alewife 318.32 95.30 192.33 204.96 298.47 44.01 229.92 126.28 10.53 141.05 76.28 713.83 137.13 445.28 89.04 190.71 405.91 46.39 265.18 96.72 374.38 7.47 204.82 White Perch Yellow Yellow Perch 15.84 90.95 312.28 99.36 205.92 263.41 441.81 307.43 373.09 429.37 211.61 253.92 236.61 315.96 457.53 263.73 181.14 161.74 355.76 89.72 157.34 291.47 270.49 - (1978-1994; no trawling in 1989) time-stanzas, annual catches from 1995-2014, and the mean for 1995-2014. rus mean mean mean Time-stanza Post-phospho 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 1995-2014 Phosphorus Table 2. Catch per Table trawl for major Bay of Quinte fish species. Shown are mean catches for phosphorus (1972-1977) and post-phosphorus

62 Fish Populations ------Round Goby 0.02 0.55 6.36 3.31 3.81 2.14 4.36 1.25 1.17 1.89 1.97 1.67 2.19 0.94 2.46 0.94 - Freshwa Drum ter 0.02 0.07 2.47 6.11 7.92 6.08 2.75 4.53 5.39 4.33 4.25 10.39 2.72 3.58 4.99 4.33 Large - mouth Bass 0.12 1.27 1.11 1.44 2.00 0.06 20.42 0.39 1.00 0.06 0.64 0.14 0.33 0.25 2.15 0.06 Gizzard Shad 1.93 2.07 1.03 1.47 1.72 1.25 1.11 0.44 0.92 0.64 0.44 0.42 0.72 0.72 0.90 0.86 White Sucker 0.83 0.64 169.58 142.64 66.25 75.19 44.44 63.92 159.11 71.75 61.50 136.03 74.92 75.81 91.90 53.56 Bluegill 0.48 12.24 3.17 2.47 2.22 2.56 2.14 1.61 2.50 1.75 2.53 2.36 1.44 1.33 2.59 7.56 Walleye 24.40 19.94 89.39 73.08 26.94 15.33 15.97 18.61 18.14 23.42 29.08 37.53 28.11 15.25 31.20 14.72 Pumpkin - seed 18.08 20.33 167.67 95.83 37.33 20.83 17.89 7.25 6.42 2.56 10.56 13.69 7.11 6.08 31.42 15.28 Brown Brown Bullhead 546.11 50.25 ------Alewife 279.74 100.19 2.19 2.89 7.69 3.67 2.75 4.61 4.31 3.86 1.69 3.75 3.58 19.42 0.19 4.66 White Perch Yellow Yellow Perch 19.95 11.06 3.75 3.42 1.94 0.83 1.00 4.72 7.00 2.64 6.11 6.25 1.31 2.69 4.94 3.59 - 1971, 1976) and post-phosphorus (1978-1979, 1981-1983, 1985-1988) time-stanzas, annual catches from 2001-2005 and 2007-2014 and the mean for years 2001-2005, 2007-2014. rus mean mean mean Time-stanza Post-phospho 2001 2002 2003 2004 2005 2007 2008 2009 2010 2011 2012 2013 2014 2001-2014 Phosphorus Table 3. Catch per trap net (upper Bay of Quinte Table only) for major Bay of Quinte fish species. Shown are mean catches for phosphorus (1969-

Hoyle 63 References

Hurley, D.A., 1986. Fish populations in the Bay of Quinte, Lake Ontario, before and after phosphorus control. In: Minns, C.K., Hurley, D.A., Nichols, K.H. (Eds.), In: Project Quinte: point source phosphorus control and ecosystem response in the Bay of Quinte, Lake Ontario. Canadian Special Publication of Fisheries and Aquatic Science 86, 201-214. Hurley, D.A., Christie, W.J., 1977. Depreciation of the warmwater fish community in the Bay of Quinte, Lake Ontario. Journal of the Fisheries Research Board of Canada. 34, 1849-1860. Hoyle, J.A., Bowlby, J.N., Brousseau, C., Johnson, T., Morrison, B.J., Randall, R., 2012. Fish Community Structure in the Bay of Quinte, Lake Ontario: The Influence of Nutrient Levels and Invasive Species. Aquat. Ecosyst. Health Mgmt. 15(4), 370-384. Hoyle, J.A., 2012. Bay of Quinte Fish, 2010. In: Monitoring Report #21. Project Quinte Annual Report 2010, pp. 116-131. Bay of Quinte Remedial Action Plan. Kingston, Ontario, Canada. Hoyle, J.A., Yuille, M.J., In press. Nearshore Fish Community Assessment on Lake Ontario and the St. Lawrence River: A Trap Net-Based Index of Biotic Integrity. Journal of Great Lakes Research 42(3).

64 Fish Populations Edited by

M. Munawar Fisheries & Oceans Canada Burlington, Ontario

J. Lorimer Aquatic Ecosystem Health & Management Society Burlington, Ontario

Cover Photo Credits R. Rozon J. Hoyle