1

The fisheries and limnology of Oneida Lake 2019

New York Federal Aid in Sport Fish Restoration Job 2-2, Study 2

F-63-R

Prepared by:

J.R. Jackson, A.J. VanDeValk, T.E. Brooking, K.T. Holeck, C. Hotaling, and L.G. Rudstam

Cornell Biological Field Station

Department of Natural Resources Cornell University

900 Shackelton Point Rd. Bridgeport, N. Y. 13030

www.dnr.cornell.edu/fieldst/cbfs.htm.

April 2020

Corresponding author: Randy Jackson: jrj26@cornell.edu

2

Table of contents

Abstract 2 Introduction 4 Methods 6 Oneida Lake in 2019 6 Limnology 6 Fish community changes (Gill net) 14 Walleye 15 Yellow perch 22 White perch 25 Smallmouth bass 27 Open water forage fish 30 Lake sturgeon 31 Double-crested cormorants 32 Round goby 33 Nearshore sampling 35

Fyke nets 35 Spring electrofishing with black bass population assessments 38 Shoreline seining 47

Creel survey 50 Annual abbreviated summer survey 50

Recommendations for management and future research directions 58 Literature Cited 64

Appendix 1: The fall and rise of the chain pickerel in Oneida Lake 69 (VanDeValk et al.)

Appendix 2: Summary of Underwater Video Surveys of Round Goby 75

Abundance in Oneida Lake, 2017-2019 (Brooking et al.)

Appendix 3: Data collection methods 90

Appendix 4: Standard data tables 95

Abstract

Oneida Lake is New York State’s 3rd most heavily fished lake. Walleye have historically received the majority of targeted effort, with black bass increasing in importance in recent years. Long-term monitoring of the fisheries and limnology of Oneida Lake has captured a series of changes in recent decades that have resulted in pronounced changes in the lake’s physical and biological characteristics, including reductions in nutrient inputs resulting from the Great Lakes Basin water quality agreements;

3

establishment of invasive dreissenid mussels resulting in increases in water clarity; increases in summer water temperatures and decreases in duration of ice cover; establishment of a breeding population of double-crested cormorants; and increases in populations of white perch and gizzard shad. Round goby were first reported by anglers in 2013, appeared in our standard surveys in 2014 and increased in abundance through 2016. An apparent die-off of round goby occurred over the winter of 2016-2017, but the population appears to be increasing back to peak densities. Long-term analyses show significant changes in production and abundance of many important sport and prey fish species as a result of these changes, including increases in black bass populations and concomitant growth in the black bass fishery. New analyses focus on the period since 2007, following the establishment of quagga mussels, with the goal of differentiating long-term from ongoing changes. Analyses of data since the year 2007 suggest that the lake has reset at a new, mesotrophic state, with resulting changes in fish populations. Increased water clarity has resulted in increases in littoral macrophytes and a shift of some of the lake’s productivity from pelagic to littoral production. While zooplankton densities did not decrease immediately following establishment of dreissenids, recent data indicate that a significant decrease in Daphnia densities is now occurring. Bythotrephes longimanus was discovered in Oneida Lake in 2019 and could contribute to further declines in Daphnia. Establishment of double-crested cormorants contributed to declines in yellow perch and walleye populations, but even with an aggressive management program, both populations remained below historic levels through 2016. A 2019 mark-recapture estimate of the adult walleye population resulted in an estimate of a population of one million fish, the first time the population has reached that size since 1986. The adult yellow perch population has also doubled in the last three years. These increases appear, in part, to result from decreases in the white perch population, contributing to improved larval survival and year classes at levels above what was observed through much of the 2000s. Angler catch and harvest rates of walleye in 2019 were the highest observed since we began regular creel surveys in 2012 (0.72 and 0.21/hr. respectively) and 60,000 walleye were harvested. A large walleye year class from 2016 will recruit into the fishery in 2020 and current harvest levels should be sustainable without reducing population size in the near future, The smallmouth and largemouth bass populations exhibited increases during the 1990s, suggesting that current lake conditions are more favorable for black bass. Spring electrofishing catch rates of both largemouth and smallmouth bass are within the middle third of those observed in other New York waters, and observed length-at-age exceed the mean for most ages of both largemouth and smallmouth bass. Both electrofishing and gill net samples suggest the smallmouth bass population has declined in recent years, but not to the levels seen prior to the 1990s increases. This decline co-occurred with disease outbreaks in the falls of 2017 and 2018, but a cause and effect relationship cannot be positively established. Black bass angler catch rates of 0.46/hr in 2019 were well within the range observed since 2012. Oneida Lake continues to support quality sustainable fisheries for walleye, yellow perch and black bass. Continued monitoring of the lake’s limnology and fish populations will help keep track of the impacts of round goby establishment and guide management to ensure that the lake’s fisheries will continue to satisfy anglers and benefit the regional economy.

4

Introduction Oneida Lake is the largest lake by area entirely within the borders of New York State and is third only to lakes Ontario and Erie in total angling effort, with an estimated 649,000 angler days in 2017 (Responsive Management 2019). Effort estimates conducted by the Cornell Biological Field Station have produced annual levels of open water effort in excess of 165,000 boat hours in every year since 2010 (see below). Angling on Oneida Lake generates at location revenues of over 7.7 million dollars annually, and as such represents an important resource both locally and across the state (Responsive Management 2019). Traditionally, walleye Sander vitreus has been the primary focus of the Oneida Lake fishery, with yellow perch Perca flavescens and black bass (smallmouth bass Micropterus dolomieu and largemouth bass M. salmoides) also providing popular fisheries. The walleye population is intensively managed on Oneida Lake, including annual stockings of 150 million walleye fry, management of double-crested cormorants Phalacrocorax auritus, and angling regulations that have been imposed and relaxed with the goals of retaining both a high walleye yield and a yellow perch population capable of providing forage for walleye and larger fish attractive to anglers (Forney 1980). Angling regulations are based on intensive monitoring of the walleye and yellow perch populations and predicted walleye recruitment. Oneida Lake has been the subject of research by the Cornell Biological Field Station (CBFS) since its establishment in 1956. Work on Oneida Lake is an important part of the collaboration between Cornell’s Department of Natural Resources and the New York State Department of Environmental Conservation’s Bureau of Fisheries (NYSDEC). Research and monitoring on Oneida Lake is designed to encompass a range of trophic levels, from nutrients to fish and anglers, and these data are used to improve our understanding of the interactions between the ecosystem and the fishery in Oneida Lake. During the time span that data have been collected on Oneida Lake, a series of perturbations have resulted in fundamental changes in the lake and how it functions. Events that have resulted in demonstrable impacts on Oneida Lake’s dynamics include the Great Lakes Water Quality Agreement of 1972 (with amendments in 1983 and 1987), establishment of a nesting colony of double-crested cormorants in the 1980s, invasion by zebra mussels Dreissena polymorpha in the early 1990s (followed by the invasion by quagga mussels Dreissena rostriformis bugensis in the mid-2000s), and most recently invasion by round goby Neogobius melanostomus in the mid-2010s. This report provides a summary of the standard monitoring data for 2019, with analyses relating data to previous years, along with an appendix with standardized methods for data collection and standard data tables. Over the course of our sampling on Oneida Lake, the occurrence of shifts in conditions over the long-term is well-documented, particularly decreased productivity and increased water clarity resulting from international water quality agreements and

5

establishment of dreissenid mussels. In recent reports, we presented analyses of trends in lake biology over more recent years (2000-2016) to assess whether the lake continues to demonstrate changing trends in physical and biological features (Jackson et al. 2017). The shorter time frame for trend analyses was adopted with the intention of allowing detection of responses to recent perturbations free from the influence of more well-established shifts in lake dynamics associated with water quality improvements, double-crested cormorants and the initial colonization of the lake by zebra mussels. Most recently, we used the results of change point analyses on limnological parameters to explore what time frame was most appropriate for identifying ongoing changes (Jackson 2018). Change point analyses identified a shift in soluble reactive phosphorus (SRP) in 1989 correlated with regional water quality plans, shifts in chlorophyll-a and water clarity in the years 1992 and 1993 which are correlated with establishment of zebra mussels, and shifts in Daphnia and total zooplankton biomass in 2007 and 2008 which correlate with the initiation of replacement of zebra mussels by quagga mussels. We therefore have amended our trend analyses to cover the period 2007-present in order to ensure sensitivity to ongoing changes. This period spans the years during which quagga mussels became the dominant dreissenid and is also characterized by ongoing cormorant management efforts and consistent walleye harvest regulations. With this approach, we hope to be able to separate documented trends that were a result of past changes from those that may suggest a response to new changes (e.g., round goby, Bythotrephes longimanus). Several of our data sets are available on the web through the Knowledge Network for Biocomplexity (http://knb.ecoinformatics.org/index.jsp) a data repository that is also a member node of the National Science Foundation DataONE portal (www.dataone.org). A single search for “Oneida Lake” as of March 2019 showed the ten available data sets, which include limnology, phytoplankton, zooplankton, benthos, mussels, ice cover, walleye, yellow perch, gill net and trawl catches. In early 2016, we published a book with summaries and analyses of many of the long-term data sets from our work on Oneida Lake (Rudstam et al. 2016). Collection of data to maintain the long-term database and directed studies aimed at understanding the effects of ecosystem change on the fish populations were continued in 2019 by the Department of Natural Resources of Cornell University as part of the activities of CBFS. Funding was provided by NYSDEC through the Federal Aid in Sport Fish Restoration Program and from the CBFS endowment.

6

Methods Most of the data presented in this report result from continuation of long-term sampling protocols established at the outset of the CBFS studies on Oneida Lake. Detailed methods for both limnological and fisheries surveys can be found in Appendix 3. In instances where models have been developed to estimate some index parameters or newly established or modified methods are employed, they are presented in association with the results utilizing them.

RESULTS - ONEIDA LAKE IN 2019 Limnology In the winter of 2018-2019, first complete lake ice cover was observed on January 12, 2019. Unlike many recent years, there were no break-up and refreeze events following initial ice formation. Ice out occurred April 9, 2019. The total complete ice cover duration of 86 days was 2 days shorter than the long-term average (Appendix Table A1). Long-term records of the lake’s winter ice season exhibit trends towards being shorter, and are consistent with regional and global warming patterns. Date of first complete ice cover since 1976 is trending later (linear regression: df = 41; F-ratio = 4.85; r2 = 0.11; p = 0.03). Ice off date exhibits no trend (linear regression: df = 42; F-ratio = 0.57; r2 = 0.01; p = 0.45). Recent patterns of multiple break-up and refreeze events following initial ice formation have contributed to a significant decline in the number of days the lake has complete ice cover (Figure 1; linear regression: df = 42; F-ratio = 4.79; r2 = 0.10; p = 0.03). On average, duration of complete ice cover on Oneida Lake has decreased 6 days/decade since 1976. A comparison of mean duration of ice cover for the first 15 years of our data record (1975-1989) and the last 15 years (2004-2018) revealed a significant reduction (early mean = 99.5 days, recent mean = 81.5 days; student’s t-test: t-ratio = -2.55, p = 0.02).

7

Figure 1. Days of complete ice cover, Oneida Lake, New York, 1975-2019.

June-August water temperatures at 2 m depth averaged 73.0 °F (22.8 °C) in 2019, 1.3 degrees above the long-term average of 71.7 °F (22.1°C) (Appendix Table A1). June-August water temperatures have exceeded the long-term average in every year since 2009. Summer water temperature trends are consistent with patterns observed regionally and globally. Average June-August water temperatures since 1975 exhibit a strong and significant increase (Figure 2; linear regression: df = 43; F-ratio = 35.86; r2 = 0.45; p < 0.0001). On average over the period 1975-2019, summer water temperatures have increased at a rate of 0.1° F/year (0.05°C).

8

Figure 2. Average daily water temperature at 2 m depth at the Shackelton Point station from June 1 to August 31, Oneida Lake, New York, 1975-2019.

Mean annual Secchi depth in 2019 was 3.3 m, equal to the long-term average of 3.3 m (Figure 3, Appendix Table A1). The mean chlorophyll-a concentration of 4.7 µg/L was well below the long term average of 6.7 µg/L (Figure 4). High water clarity and low chlorophyll-a concentrations have been typical since 1992, when zebra mussels became abundant in Oneida Lake (Zhu et al. 2006). In approximately 2005, quagga mussels entered Oneida Lake, and by 2008 began displacing zebra mussels. Quagga mussels have colonized softer substrates considered uncolonizable by zebra mussels, creating potential for increases in dreissenid biomass lakewide (Hetherington et al. 2019). Initially, displacement of zebra mussels by quagga mussels resulted in an increase in total dreissenid biomass at sites historically sampled for zebra mussels, but after the early post-quagga peak, dreissenid biomass declined to levels observed when only zebra mussels were present (Figure 5). Predation on smaller mussels by round goby has impacted total densities of mussels, suggesting that round goby may have potential to reduce mussel populations through reductions in survival to larger sizes and declines in total mussel biomass are evident in the most recent years of sampling (CBFS unpublished data). Chlorophyll-a concentrations have not exhibited a significant change since the arrival of quagga mussels, remaining below the long-term average (Table 1). Water clarity has remained stable over the same time period (Table 1). Soluble reactive phosphorus (SRP) concentrations in 2019 were less than 30% the long-term average at 3.0 µg/L, and total phosphorus (TP) was just over half the long-term average (Figure 6). Neither SRP nor TP concentrations show significant trends over the period since 2007 when quagga mussels began to become dominant (Table 1). Following water quality

9

improvement efforts in the 1970s and 1980s and establishment of dreissenids, the productivity of the lake is overall typical of a mesotrophic lake (Carlson 1977; Wetzel 2001; Idrisi et al. 2001; Zhu et al. 2006).

Figure 3. Time trends in May-October Secchi disc measurements in Oneida Lake, New York, 1975-2019.

Figure 4. Time trends in May-October Chl-a concentration in Oneida Lake, New York, 1975-2019.

10

Figure 5. Time trends in fall dreissenid mussel biomass at standardized sites in Oneida Lake, New York, 1992-2019.

Figure 6. Time trends in May-October phosphorus concentration in Oneida Lake, New York, 1975-2019.

11

Total zooplankton biomass was near the long-term average, but Daphnia spp. biomass was only half of the long-term average, consistent with recent years (Figure 7; Appendix Table A1). Analyses of the years since 2007 do not indicate significant trends in Daphnia spp. biomass or total zooplankton biomass (Table 1). Whereas Daphnia spp. have typically accounted for 40-65% of total zooplankton biomass over much of our data series, in recent years their contribution has fallen below 35%, and in 2019 was at 22% (Figure 8). The decline in representation of Daphnia spp. in the Oneida Lake zooplankton community since 2007 is significant (Table 1). Observed reductions in Daphnia spp. and the shift to a more copepod dominated zooplankton community is of concern, as Daphnia spp. are a critical food for supporting growth of early life stages of fish. Bythotrephes longimanus colonized Oneida Lake in late summer 2019, and have potential to alter the zooplankton community through predation on Daphnia spp., but for 2019 they represented only a small fraction of zooplankton biomass (annual average 0.87 ug/L). Bythotrephes were common in the diets of juvenile fish after their arrival. We will continue to monitor the zooplankton community response to this new arrival and look for changes in growth rates of planktivorous fishes.

Figure 7. Time trends in May-October total zooplankton biomass (upper solid line, with squares) and Daphnia biomass (lower dashed line, with circles) in Oneida Lake, New York, 1975-2019.

12

Figure 8. Time trend in May-October percentage of total zooplankton biomass represented by Daphnia spp. in Oneida Lake, New York, 1975-2019.

Table 1. Recent trends (2007-2019) in lower trophic level parameters in Oneida Lake, New York. Significance levels are based on simple linear regression. Data are presented in Appendix Table A1. Trend indicates direction (+ or -) over time, with r2 and p reported for regressions. Significant trends indicated by bold type.

Variable

2007-2019

Trend r2 p

Secchi depth (m) - 0.11 0.27

Chlorophyll-a (ug/L) + 0.01 0.70

SRP (ug/L) - 0.22 0.13

TP (ug/L) Zooplankton biomass (ug/L) Daphnia biomass (Ug/L) Daphnia percent biomass

- + - -

0.16 0.13 0.06 0.47

0.18 0.23 0.42 0.01

Our analyses of long-term trends in physical and limnological features in Oneida Lake show a lake at a lower state of productivity than when limnological studies were initiated in the 1970s. Reduced nutrient loading, combined with grazing by introduced dreissenid mussels, have led to reduced chlorophyll-a concentrations as compared to the 1970s and 1980s. While Oneida Lake was once classified as a eutrophic lake, it now possesses characteristics of a mesotrophic lake. However, summer bluegreen

13

algal blooms still occur, sometimes causing beach closings, and attracting media attention. Bluegreen bloom production is variable across years, but 2019 was relatively mild (CBFS database). Ongoing studies are directed at determining if bluegreen blooms have increased in recent years, as perceived by much of the public, or whether the perception simply reflects more coverage in the media. While we observed large bluegreen blooms in 2014 and 2017, no statistically significant trends in blooms are detectable over the long term or short term (Figure 9; simple linear regression, 1996-2018: annual mean (May-September) p=0.62, July/August mean p=0.60, maximum p=0.99; 2009-2018: annual mean p=0.18, July/August mean p=0.43, maximum p=0.48).While Daphnia spp. biomass did not initially decrease as a result of decreased productivity or establishment of zebra mussels, densities have been below the long-term average since 2005, and this may have implications for production of fish.

Figure 9. Time trend in Cyanophycota (“bluegreen algae”) biovolume in Oneida Lake, New York, 1975-2018.

Hemimysis anomala was found in Oneida Lake in 2009 (Brooking et al. 2010), but no Hemimysis have been observed in fish diets since 2013. While still present in the lake, there is no indication that Hemimysis will become abundant in Oneida Lake. However, Hemimysis have expanded east from the lake into the Erie Canal (Brown et al. 2014). Water temperatures in recent years continue to be above long-term averages and ice duration has been significantly declining. Increases in water temperatures are predicted to increase periods of summer stratification, resulting in anoxic bottom waters and increased phosphorus release from the sediments (Hetherington et al. 2015), but to date we have not seen increases in TP or SRP. Reduced duration of safe ice for

14

anglers will impact winter fishing effort and may reduce harvest of yellow perch, given that historically about 50% of annual yellow perch harvest has been through the ice. Fish Community Changes (Gill net) Gill net catches in Oneida Lake are typically dominated by yellow perch, white perch Morone americana and walleye. These three species have represented over 70% of the total gill net catch in every year since sampling started in 1957. While white perch were the most frequently captured species in gill nets in 2015, and were typically over 25% of the catch in years after 2001, they declined in representation to only 9.6% of the total catch in 2016. In 2017, white perch increased to 20% of total gill net catch, and have remained near 20% of the catch since, representing 21.2% of catch in 2019 (Figure 10). Yellow perch were the most common species in the catch in 2019, and represented 51.4% of the total catch, similar to the long-term average of 53.2%. Yellow perch catches in 2018 and 2019 were the highest percent representation observed since 1999. Walleye were the second most common species in 2019 gill nets, representing 22.2% of catch, similar to 2018, and above the long- term average of 19.4%. Total number of fish caught in the standard gill nets in 2019 was 1,976, near the long-term average of 1,940, but higher than all but two of the years since 1989. Other common species such as gizzard shad Dorosoma cepedianum, and white sucker Catastomus commersonii were captured in numbers within the range of recent years. Smallmouth bass catches were the lowest observed since 1985. Freshwater drum Aplodinotus grunniens catches remained below the levels observed from 1988-2013. Lake sturgeon Acipenser fulvescens catches were relatively high for the last decade, due to the re-initiation of stocking in 2014 and recruitment of young sturgeon into the percid nets. Redhorse (keyed out specimens have been greater redhorse Moxostoma valenciennesi) catches were the highest on record at nine.

Figure 10. Proportion of three major fish species in standard gill net sets in Oneida Lake, New York, 1957-2019.

15

Walleye We assess the walleye population in Oneida Lake at several life stages: as larvae (lengths of 9 to 13 mm) with Miller high-speed samplers; as juveniles in the spring, summer and fall with bottom trawls; and as juveniles, sub-adults and adults with gill nets in the summer, supported with mark-recapture for adult fish (age-4 and older) at regular intervals (currently every 3 years, conducted in 2019). The abundance estimate of adult walleye (age-4 and older) for 2019 is based on a mark-recapture estimate (Figure 11; Appendix Table A2). Between April 11 and 25, 15,410 walleye were fin clipped at the Oneida Fish Cultural Station (OFCS; 7,050 males and 8,360 females). A small number of fish were clipped in other parts of the lake during yellow perch netting. The fin-clipped sample was dominated by age-4 and 5 fish (52% of sample). Recaptures were obtained throughout the 2019 field season through both standard and targeted sampling. Recapture efforts produced 1,622 adult walleye (gill net – 209; 18 ft trawl – 12; 40 ft trawl – 116; electrofishing – 1,255). Combined, recapture efforts produced 24 recaptures (gill net – 3; 18 ft trawl – 2; 40 ft trawl – 1; electrofishing – 16). Pooled across all gears, the population estimate for age-4 and older walleye was 1,000,402 fish (Figure 11; 95% CI – 681,101-1,460,867). The most common age class was age 5 (2014 year class), representing 31% of the population. The 2015 year class represented 28% of the population, while year classes from 2012 and earlier combined to represent 27% of the population.

In recent reports we have noted reduced harvest of walleye following establishment of round goby. Prior to round goby, annual mortality rates of adult walleye between recent consecutive mark-recapture estimates was typically 20%. With an estimated harvest reduction of 60% in 2016 and continued harvest reductions below pre-2016 levels of 20% or more in 2017 and 2018, we predicted that the adult walleye population estimated by the 2019 mark-recapture could increase to over 600,000 fish. The annual mortality rate of adult walleye between the 2016 and 2019 mark-recapture estimates was estimated at 12%, reflecting the reduced harvest we observed. However, reduced mortality alone does not account for the sharp increase of the adult walleye population to over one million fish. The other factor underlying the unexpectedly large increase in the adult walleye population is apparent underestimates of abundance at age 3 of recent year classes. Abundance at age 3 is estimated independently from mark-recapture estimates using gill net and trawl catches adjusted for catchabilities estimated by Irwin et al. (2008). Abundances of age 4 and older fish are estimated based on mark-recapture estimates. For all three year classes recruiting into the adult population between the 2016 and 2019 mark-recapture estimates, abundance at age 4 was higher than estimated abundance at age 3, ranging from 30% higher for the 2013 year class to 76% higher for the 2015 year class. While we have seen increases in estimated abundance between ages 3 and 4 in the past, we haven’t seen them consecutively or of such magnitude before. We will need to reassess our estimation methods for subadult walleye if this becomes a regular phenomenon.

16

Estimated abundance of the 2016 year class at age 3 is the highest observed since 1994, and this cohort could add substantially to the adult walleye population in 2020. The adult walleye population has exhibited a significant increasing trend since 2007 (Table 2).

Figure 11. Density of adult walleye in Oneida Lake, New York, 1957-2019. Table 2. Recent trends (2007-2019) in measurements of walleye abundance in Oneida Lake, New York. Significance levels are based on simple linear regression. Data are presented in Appendix Tables A2 and A4. Trend indicates direction (+ or -) over time, with r2 and p reported for regressions. Significant trends indicated by bold type.

Variable

2007-2019

Trend r2 p

Adult (age 4+) population size + 0.48 0.01 Larval density - 0.02 0.68 October 1 age-0 density + 0.01 0.69 Spring age-1 density - 0.01 0.82

We predict future walleye recruitment using the average of catches in trawls and gill nets of age-1 and age-2 walleye (Appendix Table A2). We estimate density of age-1 to 3 walleye from the average of the estimates from the trawl and the gill net using the age and gear specific catchabilities derived by Irwin et al. (2008) and predict future recruitment using the catchability-adjusted catches of age-1 and age-2 walleye (see Appendix Table A2). The “best” model (determined using the Akaike Information

17

Criterion) given the data for year classes 1957- 2004 includes the natural logarithm of age-1 and age-2 walleye abundance:

Ln(Age-4) = -0.059 + 0.239 Ln(Age-1) + 0.593 Ln(Age-2) (1) where Age-1, Age-2 and Age-4 are densities of walleye age classes in fish/ha. Based on these relationships, our prediction for recruitment to age-4 in 2020 of walleye produced by the 2016 year class is 270,600 fish. Recruitment to age-4 of walleye produced by the 2017 year class in 2021 is predicted to be 65,300 fish.

Adult walleye length-at-age is determined from fish collected in fall (mid-September and later). For much of the data series, samples were primarily collected by bottom trawl, but electrofishing samples were integrated into length-at-age estimates during mark-recapture years and during all years since 2010. Aging was conducted using scales through 2009, after which otoliths have been used. Despite decreases in observed length-at-age in 2019 compared to previous years, significant increasing trends in length at ages 4-6 have been observed over the period 2007-2019, (linear regression: age 4: df = 11; F-ratio = 6.21; r2 = 0.36; p = 0.04; age 5: df = 11; F-ratio = 4.68; r2 = 0.30; p = 0.05; age 6: df = 11; F-ratio = 10.40; r2 = 0.49; p = 0.01; Figure 12). Walleye growth has historically been dependent on availability of yellow perch, with gizzard shad and white perch providing additional forage in recent decades. Increases in growth in the 1990s and early 2000s were attributed to increased availability of food resulting from relatively regular production of summer-hatched gizzard shad year classes, which began in the late 1980s and early 1990s (He et al. 2005). Recent increases in length at age may be partly associated with the establishment of round goby (see below). Establishment of round goby in Lake Erie did not result in changes in walleye growth (Johnson et al. 2005), and improved condition of only the largest walleye (>550 mm) was observed in Lake Ontario (Crane et al. 2015). Lake Erie walleye tend to rely more heavily on pelagic prey (Johnson et al. 2005) than in Oneida Lake where walleye have traditionally depended on demersal yellow perch, so it is possible round goby could play a more significant role in walleye diets in Oneida Lake than observed in Lake Erie.

18

Figure 12. Observed length-at-age for ages 4-6 walleye from fall trawls and fall electrofishing in Oneida Lake, 1961-2019.

Between May 3, 2019 and May 7, 2019 the OFCS stocked 191.2 million walleye fry into Oneida Lake. The CBFS Miller sampler estimate of larval walleye density is conducted together with our first estimate of yellow perch larvae (when yellow perch average 8 mm length). In a subset of past years, walleye were assessed earlier, when average walleye lengths were approximately 9 mm (9.4 mm, range 9.0-10.2 mm, N=18). In years when both the 9 mm survey and the yellow perch 8 mm survey were conducted, the larval walleye estimates from the two surveys were correlated (r2 = 0.58, p = 0.01, N=10). With one outlier removed (2002, when few stocked walleye larvae survived a cold period after stocking), the correlation improves (r2 = 0.88, p = 0.0002, N=9). The equation is WDYP = 203.6 + 0.722 WD9mm (2) where WDYP is walleye density at the 8 mm yellow perch survey and WD9mm is walleye density at the 9 mm survey, both in fish/ha. Our walleye larval index (Appendix Table A4) is the number of walleye larvae at the time of the 8 mm yellow perch survey, either measured directly or calculated from the 9 mm survey with this equation. The walleye larval abundance in 2019 was 1,962 fish/ha, above the long-term average of 1,570 larvae/ha (33 years, 1966 – 2019; Figure 13). There is no trend in larval walleye abundance since 2007 (Table 2).

19

Figure 13. Time trends in density of larval walleye, Oneida Lake, New York, 1961-2019.

Age-0 walleye are monitored with bottom trawl surveys at 10 standard stations during 3 out of every 4 weeks from July through October. Catch per unit effort is converted to density in fish/ha assuming that each trawl samples an area of 0.1 ha and there is no avoidance by young fish. The 2019 age-0 fall walleye density estimate was 3.8 fish/ha on October 1. Average length on October 1 was 152 mm. Age-0 density in 2019 was well below the long-term average of 28.4 (Figure 14, Appendix Table A4). Since 2007, there is no significant trend in fall age-0 walleye density (Table 2).

Figure 14. Time trends in density of age-0 walleye on October 1 based on bottom trawls, Oneida Lake, New York, 1961-2019.

20

During recent years, poor walleye year classes were common, and “good” years represent much smaller year classes than observed prior to the 1990s when zebra mussels first established in the lake. The three largest year classes since 2000 (as measured by trawling) ranged from 14.3-19.2 fish/ha, whereas catches of 30 or more fish/ha were occurring at least every five years prior to 1992 (Figure 13). However, by modern standards, we have observed relatively high fall walleye densities in alternate years since 2014. Among the factors identified as contributing to the decline of the adult walleye population through the 1990s was increased mortality of walleye fry prior to becoming demersal and recruiting into our trawl. The increased mortality was attributed to greater vulnerability to predation under clear water conditions created by the establishment of zebra mussels. For years when larval sampling was conducted daily mortality (Z) of walleye fry was significantly higher following establishment of zebra mussels (Z = 0.07) than in the preceding years (Z = 0.04) (student’s t-test, df = 33, F-ratio = 36.25, p < 0.0001). Diet analyses have identified white perch as an important larval predator, and the occurrence of relatively strong walleye year classes co-occurred with declining trends in white perch abundance (see below). Analysis of walleye larval Z as a function of white perch abundance reveals a significant relationship (Figure 15; years 1966-1971, 1992, 1998-2019 simple linear regression: df = 32, F-ratio = 21.47, r2 = 0.40, p < 0.0001). Given the numerous factors that can influence larval survival, the strength of this relationship is evidence that white perch likely play a large role in determining survival of larval walleye. Intuitively, the influence of white perch is influenced by water clarity, but our larval data do not include years in the 1980s with both high white perch densities and the lower water clarity of the pre-zebra mussel years, so we cannot explore that relationship quantitatively.

21

Figure 15. Relationship between daily mortality of larval walleye (Z) and catch of adult white perch in Oneida Lake, 1966-2019 (not inclusive). The walleye year class density on October 1, 2018 was the highest observed since 2010, at 13.0/ha (Figure 14), and captures of yearlings from this year class in May 2019 trawling were the highest observed since 2010 (Figure 16). Densities of yearling walleye in the spring show no trend since 2007 (Table 2).

Figure 16. Time trends in density of yearling walleye in May based on bottom trawls, Oneida Lake, New York, 1961-2019.

22

The adult walleye population in Oneida Lake declined through the 1990s, and the population remained below 500,000 fish despite aggressive cormorant management and a daily harvest limit of three fish. Reduced harvest and increased early survival of walleye have now contributed to the adult walleye population reaching the one million fish mark for the first time since 1986. What appears to be a very strong 2016 year class recruiting into the fishery in 2020 should help maintain the population at this level or higher. Survival of age-0 walleye was low through most of the 2000s compared to years prior to establishment of zebra mussels, with higher mortality between the larval and demersal stages. Reduced first year survival may be attributable to higher predation mortality experienced as a result of clearer water following establishment of zebra mussels. Declines in the white perch population (see below) beginning in 2014 appear to have allowed increased survival of larval walleye and contributed to relatively strong year classes in alternate years beginning in 2014. The 2018 year class currently appears strong and should help support the adult walleye population at high levels in the immediate future. Yellow perch Adult yellow perch numbers are estimated from the catches in standard gill nets adjusted for estimates of catchability (Irwin et al. 2008). Catches of adult yellow perch in 2019 produced a population estimate of 1,991,300, more than 55% higher than the average population sizes estimated since 2000 (Figure 17, Appendix Table A5). Catches over the last four summers reflect recovery from a population low observed in 2014. No trend in adult yellow perch population size is detectable since 2007 (Table 3).

Figure 17. Time trends in age 3+ yellow perch densities (#/ha), Oneida Lake, New York, 1961-2019.

23

Table 3. Recent trends (2007-2019) in measurements of yellow perch abundance at various stages in Oneida Lake, New York. Significance levels are based on simple linear regression. Data are presented in Appendix Tables A5. Trend indicates direction (+ or -) over time, with r2 and p reported for regressions. Significant trends indicated in bold type.

2007-2019

Variable Trend

r2

p

Adult (age 3+) population size + 0.12 0.24 Larval density - 0.18 0.14 October 15 age-0 density - 0.11 0.26 October 15 mean length + 0.03 0.33 Spring age-1 density + 0.01 0.83 Summer age-1 density + 0.45 0.02

We measure the abundance of yellow perch at the larval stage (two surveys - 8 and 18 mm), and as juveniles in bottom trawls through the summer, and again as yearlings in trawls centered on May 1 (Appendix Table A6). We use the decline in catches in the bottom trawl to estimate age-0 yellow perch abundance on October 15 (or take the average of the three October trawl samples in years where summer catches do not exhibit a significant decline). Larval yellow perch density in 2019 was 96,760/ha, well above the long-term average of 67,200, and the highest observed since 2009 (Figure 18). Fall density of age-0 yellow perch was 561/ha, below the long-term average of 983 (Figure 19). Spring yearling catches of yellow perch from the 2018 year class indicated a density of 129/ha, higher than all but one year since 2008 (Figure 20). Long-term numbers show that the yellow perch population has exhibited a significant decline in larval production, fall age-0 densities and summer catches of age-1 fish (Irwin et al. 2009; Rudstam et al. 2016)). As with walleye, the last decade has shown some moderation of the declining trends observed over the long-term, with no significant trends in abundance for larvae, fall age-0 density or spring age-1 density, but a significant increase in summer age-1 densities (Table 3). In as much as yellow perch still represent the primary forage for adult walleye prior to gizzard shad recruiting to their diets in late summer and fall, mortality of age-0 yellow perch may remain high with the current high adult walleye population size. Establishment of round goby (see below) may provide an alternate forage for walleye and could ultimately benefit yellow perch recruitment.

24

Figure 18. Time trends in larval yellow perch densities (#/ha), Oneida Lake, New York, 1961-2019.

Figure 19. Time trends in fall age-0 yellow perch densities (#/ha), Oneida Lake, New York, 1961-2019.

25

Figure 20. Time trends in spring age-1 yellow perch densities (#/ha), Oneida Lake, New York, 1961-2019. White perch Based on gill net catches, the white perch population in Oneida Lake increased sharply through the late 1990s and early 2000s (Figure 21). White perch catches in gill nets exceeded yellow perch in 2007, 2009-2011, and 2015 (Appendix Table A7). Recruitment is variable, but the white perch recruitment index is suggestive of successful year classes at least once every three years from the early 1990s through 2004, but only two large year classes have been observed since 2005 (Figures 21 and 22, Appendix Table A7). Catches of white perch in gill nets followed a declining pattern from 2009-2016, but have increased steadily since, although still below the numbers of 2009. We have seen a significant decline in adult catches in gill nets since 2007 (linear regression: df = 11; F-ratio = 9.04; r2 = 0.45; p = 0.01). There has been no trend in age-0 white perch catches in bottom trawls since 2007 (linear regression: df = 11; F-ratio = 1.34; r2 = 0.11; p = 0.27). After five years of low production of age-0 white perch, the 2017 year class produced the highest catches observed since 2011. The 2018 year class was small, but the 2019 year class was the second largest observed since 2011 (Figure 22). White perch diets are similar to yellow perch, although they appear to feed more on larval and juvenile fish. Increases in white perch could therefore be part of the explanation for increased early mortality of larval percids, and their decline appears to contribute to improved survival of larval walleye (see above). Given the relatively small year classes of white perch produced most years since 2005, it is reasonable to expect a decline in adult numbers over the next several years after the 2012 year class works

26

through (the 2015 year class is large by recent standards, but less than half the size of the 2012 year class), which may benefit survival of larval percids (see VanDeValk et al. 2016 for assessment of white perch recruitment in Oneida Lake).

Figure 21. Time trends in gill net catches of white perch, Oneida Lake, New York, 1961-2019.

27

Figure 22. Time trends in age-0 white perch densities (#/ha), Oneida Lake, New York, 1961-2019.

Smallmouth bass Smallmouth bass have become an important sport fish in Oneida Lake, and can also have large effects on littoral fish communities when abundant (VanderZanden et al. 1999; Lepak et al. 2006). Opening of a spring catch-and-release fishery in 2007 was met with some concern about potential impacts on young-of-year bass production, but our studies of potential impacts of spring fishing showed catches of age-0 smallmouth bass over the first six years following the opening of spring fishing were significantly higher than those from the six years preceding the regulation change (Jackson et al. 2015). Following a high catch of young-of-year smallmouth bass in 2014, the trawl catch in 2015 was among the lowest observed since 2000. No age-0 smallmouth bass were captured in trawls in 2019 (Figure 23). It is possible that a shift to foraging on round goby has resulted in a habitat shift by age-0 smallmouth bass, and the trawl may no longer be a reliable index of year class strength. Catches of age-0 smallmouth bass in fyke nets in 2019 were the second lowest observed since fyke netting started in 2008, and fyke nets reflect four small year classes in a row after record catches in 2015 (Figure 24). Catches of adult smallmouth bass in standard gill nets in 2019 were the lowest observed since 1985, and continue a declining trend that started in 2012 (Figure 25). Over the period 2007-2019, we have observed no trends in young-of-year or adult catches, although the decline in adult gill net catches is approaching significance (young-of-year trawl - df = 11; F-ratio = 0.86; r2 = 0.07; p = 0.37; young-of-year fyke net

28

- df = 10; F-ratio = 1.03; r2 = 0.09; p = 0.83; adult - df = 11; F-ratio = 3.83; r2 = 0.26; p = 0.08). In the fall of 2017, bacterial lesions were observed on adult smallmouth bass by both anglers and CBFS staff. Prevalence was high in some localities, but no mortalities were reported, and the condition had cleared by the early sampling season in 2018. More severe presentations were observed in fall 2018, again by both anglers and CBFS staff. Mortalities were reported by anglers, but it is unknown what the extent of the die-off was. Anglers reported “many” dead bass, but lakewide sampling by CBFS staff and cormorant monitoring by DEC staff at the same time period did not detect large numbers of dead bass. Fish sent to the Cornell Veterinary School were diagnosed with a combination of LMBv, bacterial infections and heavy helminth loads (Rod Getchell personal communication). No similar outbreak of observable disease in smallmouth bass was detected in 2019. While declines in both adult gill net catches and young-of-year production initiated prior to the disease outbreak, we cannot discount the possibility that mortality from these events was greater than observed and that the adult population has been impacted by the disease. Continued monitoring of this situation will continue in 2020, and subsequent population surveys should shed light on whether the mortality affected population size. The series of relatively small smallmouth bass year classes observed over the last four years coincides with establishment of round goby. Round goby do pose a potential threat to smallmouth bass as egg predators, and we will continue to monitor year class production to determine if there is a negative impact of round goby on smallmouth bass recruitment (Steinhart et al. 2004b). We have not seen a decline in young-of-year largemouth bass over the same time period, and it is not clear why round goby would have differential impacts on the two species unless spawning habitat selection exposes nests to different predation risk. Changes in lake conditions, likely both clearer water facilitating foraging and warmer summer water temperatures contributing to increased year class success, have allowed the smallmouth bass population to reach a higher level than observed in the 1960s-1980s. Despite recent declines, we anticipate they will continue to be an abundant and important species in the lake’s ecology and fisheries, and it is possible that round goby could enhance smallmouth bass growth and condition. Lake Erie smallmouth bass showed improved condition at all ages after establishment of round goby (Steinhart et al. 2004a; Johnson et al. 2005).

29

Figure 23. Time trends of age-0 smallmouth bass catches in bottom trawls, Oneida Lake, New York, 1960-2019.

Figure 24. Time trends of age-0 smallmouth bass catches in fall fyke net samples (combined catch from ¼ in and ½ in mesh nets), Oneida Lake, New York, 2008-2019.

30

Figure 25. Time trends of adult smallmouth bass catches in gill nets, Oneida Lake, New York, 1960-2019.

Open water forage fish (gizzard shad and emerald shiner) Pelagic fish biomass is estimated in the fall using hydroacoustics with supporting small-mesh gill nets and mid-water trawling. Total pelagic fish density in 2019 was estimated to be 5,900 fish/ha (long-term average 11,200) of which 3,900 fish/ha were age-0 emerald shiner Notropis atherinoides, 1,500/ha were age-1 and older emerald shiner, and 600/ha were age-0 gizzard shad (Figure 26; Appendix Table A8). Gizzard shad density was the lowest observed since 1998 and only 23% of the long-term average. Biomass was estimated at 10.1 kg/ha (long-term average 27.0), of which 4.7 kg/ha was gizzard shad and 5.4 kg/ha was age-0 and adult emerald shiner. Pelagic fish surveys indicated alewife Alosa pseudoharengus were present in the lake in 2017 at a density of 250 fish/ha (this density was only exceeded in 1996), but no alewife were captured in 2018 or 2019. Alewife have been captured in pelagic fish surveys on Oneida Lake before (1996, 2003, 2013), but no resident population appears to have established. Gizzard shad were the most common diet item of walleye in October in 2019, accounting for 35% of identifiable diet items (Appendix Table A3). While gizzard shad were the most common diet item, their frequency of occurrence was the lowest observed since 2005. Walleye diets included 31% round goby in 2019, the highest proportion to date. Emerald shiner accounted for 4% of fall walleye diets (see Appendix Table A3). Neither observed density nor biomass of gizzard shad show a significant

31

trend over the period 2007-2019 (density - df = 11, F-ratio = 0.26; r2 = 0.02, p = 0.62; biomass – df = 11, F-ratio = 0.07; r2 = 0.01, p = 0.80). Similarly, no trends have been observed in emerald shiner density or abundance (young-of-year density - df = 11; F-ratio = 0.22; r2 = 0.02; p = 0.65; young-of-year biomass -biomass – df = 11, F-ratio = 2.93; r2 = 0.21; p = 0.12; adult density - df = 11; F-ratio = 0.26; r2 = 0.02; p = 0.62; adult biomass – df = 11, F-ratio = 0.80; r2 = 0.07; p = 0.39).

Figure 26. Time trends in biomass of open water forage fish from hydroacoustic estimates, Oneida Lake, New York, 1993-2019.

Lake sturgeon Lake sturgeon Acipenser fulvescens catches from directed sampling with large mesh gill nets in May 2019 (0.33/hr) were well within the historical range observed since sampling was initiated in 2002 (Figure 27; Appendix Table A9). June catch was 0.20/hr (more than twice the rate of any year since 2013). Prior to a stocking of 500 fish in 2014 (stockings have continued annually since), no stockings of lake sturgeon had been conducted since 2004, so catches might be expected to increase as the more recently stocked fish recruit into the gear, and we have seen increases following low catches in

32

2017. Length and weight data from collected lake sturgeon still indicate a population of fish in excellent condition that are growing at high rates. The largest lake sturgeon we have collected so far (73 in TL, 139 lb, estimated to be age 20) was captured in 2019. Our records now include evidence of successful natural reproduction and survival of some young from five years, 2011 (1 fish captured), 2012 (1), 2014 (3), 2015 (3), and 2016 (1). There is every reason to believe that some spawning now takes place annually in Fish Creek and possibly other tributaries to the lake.

Figure 27. Time trends in spring large mesh gill net catches of lake sturgeon, Oneida Lake, New York, 1993-2019.

Double-crested cormorants Double-crested cormorant management evolved through the 2000s, with seasonal management of nesting success and fall migrants during the early 2000s; near complete removal of cormorants from the lake with no successful nesting allowed from 2004-2009; nest control and fall hazing from 2010-2013; and full season hazing 2014-2019. Management since 2010 has been conducted by NYSDEC following a period of management administered by the USDA Wildlife Services on Oneida Lake (2004-2009). Funding for the USDA program was lost prior to the 2010 season. The initial seasons after cessation of the USDA program saw low summer numbers of cormorants and no efforts to nest, but fall hazing was conducted by NYSDEC to push migrants off the lake. In 2013, summer cormorant counts and nesting efforts (12 nests produced chicks) exhibited increases and NYSDEC began implementing a full season hazing program in 2014 with a target goal of no more than 100 cormorants on the lake and no successful nesting.

33

The NYSDEC conducted weekly counts beginning April 30, 2019 and continuing through October 30, 2019. The average weekly count over the season was 562 birds (similar to 523 in 2018). During the spring and summer (pre-fall migration, April through the end of July), counts averaged 131 birds (88 in 2018), and during the migration (August onwards) counts averaged 853 birds (957 in 2018). Counts near or over 1,000 were typical throughout the migration period, with an October 8 peak of 1,358. Fall numbers on Oneida Lake have increased since management of the Lake Ontario colony on Little Galloo Island was ended by a Federal court injunction of the USFWS depredation permit in 2016. Analyses of diets were conducted on 203 birds collected from April 30 through October 30. Of the 1,452 diet items, 1,323 were identifiable to species. Of the identifiable items recovered from stomachs, 88% were gizzard shad, 25% were round goby, 17% yellow perch and 15% emerald shiner. No other species accounted for more than 2% of observed diets. Twenty-three walleye were found in cormorant diets, accounting for less than 2% of identifiable food items. Use of round goby in 2019 was a decrease from 2018. Use of gizzard shad in the fall exceeded use of round goby despite a small gizzard shad year class, contrary to 2018 results when round goby use remained high in the fall despite abundant gizzard shad. By weight, yellow perch made up 34% of cormorant diets, walleye 30%, and gizzard shad 14%, with no other species accounting for more than 10%. Our analyses of cormorant diets over a 15-year span have found positive selection for schooling, soft-bodied prey such as gizzard shad when they are available, so buffering of potential impacts on percids by fall migrating cormorants has been realized in years when gizzard shad reproduce successfully (DeBruyne et al. 2013). Round goby may ultimately act as a buffer for more valued sport fishes, but at the current densities of round goby, incidental percid consumption by cormorants feeding on the bottom is a concern. Evidence from other systems suggests that round goby can dominate cormorant diets when abundant (Johnson et al. 2010).

Round goby Round goby were confirmed in the Oneida River at the last barrier before Oneida Lake as early as 2010, but no confirmed reports came from within the lake until 2013, when anglers found them in yellow perch stomachs. No sampling by CBFS produced round gobies in 2013 and none were observed in fish diets. By late July 2014, round gobies began to show up in standard trawl samples and by mid-August were encountered regularly throughout the lake. Trawl catches in 2015 indicated that the round goby population was expanding. After a peak catch of 10 round goby in a trawl sample from 2014, catches peaked at 618 on September 15, 2015, 707 on September 19, 2016, 940 on October 3, 2017 (Figure 28). Round goby catches declined in 2017 after an apparent winter die off. In 2019, trawl catches peaked at 392 on August 28.

34

Figure 28. Catches of round goby in standard bottom trawling in Oneida Lake, New York, 2014-2019. Low catches of round goby early in 2017, combined with reports and direct observations of concentrations of dead gobies prior to ice on suggest a die off of round goby in the winter of 2016-2017. Overwinter predation may also be a factor in determining round goby survival. Consistently low catches through 2018 suggested that round goby had not yet recovered to pre-2017 levels. Sampling in 2019 indicated that the round goby population was continuing to increase. It is uncertain if the 2017 die off was an isolated event related to disease or some other factor or if temperatures in Oneida Lake will result in occasional winter die offs in the future. While several standard gears used in our sampling capture round goby, and there is general agreement among gears in relative abundance of round goby across years, density estimates from gears such as the trawl and seining differ substantially (see Appendix 2). A small-scale video survey was conducted in 2017, and lakewide video surveys were conducted in 2018 and 2019. Results from these surveys suggest that the trawl consistently underestimates round goby density and the video technique may be the best way to obtain reliable density estimates of round goby in all habitats in which they occur (see Appendix 2). We will continue to assess our captures of round goby in the various gears to try and determine which provides the most reliable measure of round goby abundance. In 2019, round goby occurred in 11% of white perch diets, 11% of yellow perch, 6% of walleye, and 12% of smallmouth bass as determined from standard gill nets. These numbers reflect increased use by all species from levels observed in 2018 (Figure 29). Preliminary analyses suggest round goby may be having a negative impact on benthic

35

invertebrate densities in Oneida Lake (CBFS unpublished data). Additionally, dreissenid mussel length-frequencies from 2015 to 2019 indicate reduced densities of mussels in the size range preferred by round goby, suggesting potential impacts of round goby on mussel recruitment to larger size classes. This is now beginning to be evidenced in reduced mussel biomass (see above). If round goby follow the pattern observed in other systems, we might expect to see an expanding population over the next few years, but there is some question if overwinter mortality and predation may act to keep densities down. Continued sampling should allow us to detect any responses in terms of fish diets, growth and angler catch rates. We should be well-positioned to detect any longer-term changes in the limnology of the lake and its fish community given both pelagic and littoral sampling have been underway for some time in advance of the arrival of round gobies. In Lake Erie, round goby were detected in the diets of all piscivorous species examined, but significant increases in growth were only documented for smallmouth bass, yellow perch and burbot Lota lota (Steinhart et al. 2004; Johnson et al. 2005).

Figure 29. Occurrence of round goby in the diets of predators in Oneida Lake, New York, 2015-2019.

Nearshore fish community Fall fyke nets Catches in fyke nets in 2019 produced similar results to past years, with a total of 29 species represented. Overall catches of most species were within the range of past years, although catches of young-of-year Lepomis spp. were the highest since 2013 in

36

the ¼ in mesh net (Appendix Tables A10 and A11). Catches in the 1/4 in mesh nets averaged 324 fish/net-night, the second highest we have observed, in large part due to the catch of young Lepomis. Annual catches in this mesh are often driven by production of young-of-year Lepomis. Other commonly caught species were round goby at 15/net-night, young-of-year yellow perch at 27/net-night and rock bass Ambloplites rupestris at 5/net-night. No other species were caught at rates higher than 3/net-night. Catches in the ½ in mesh nets averaged 45 fish/net-night, the lowest catch rate observed since 2014. The most commonly caught species were age 1+ yellow perch at 15/net night and pumpkinseed at 7/net-night. No other species was captured at a rate of more than 4/net-night. The ½ in mesh nets provide a good sample of adult sunfish, and species composition as represented by this gear has been fairly consistent over time (Figure 30). Young-of-year largemouth bass in both meshes in 2019 produced catches of 3.2/net-night, the highest seen since 2016 (Figure 31). Young-of-year smallmouth bass catches were the second lowest observed since sampling was initiated in 2008 at 2.0/net-night, extending to four the run of small year classes as measured by fyke nets (Figure 31). The fyke nets continue to produce catches of littoral species not represented in the traditional gears used in our long-term studies (but spring electrofishing, initiated in 2011, now provides an additional littoral sample and annual beach seining was initiated in 2015). They have provided our only index of young-of-year largemouth bass production prior to the initiation of regular shoreline seining in 2015, and also show potential as an index for sunfish (Figure 28) and esocids. Fyke nets do not show potential as an index of adult bass, and an index of largemouth bass, in particular, requires shoreline electrofishing. Given that potential behavioral shifts in young-of-year smallmouth bass could compromise the reliability of our trawl index, fyke nets may also become our primary gear for indexing their production over the long term (beach seining will also serve as a more traditional index, but has not been conducted as consistently in the last decade as fyke net sampling).

37

Figure 30. Catch rate of age 1 and older centrarchids in ½” mesh fyke net samples, Oneida Lake, New York, 2007-2019.

Figure 31. Catch rate of young-of-year black bass in fyke net samples (combined catches from ¼: and ½” meshes), Oneida Lake, New York, 2008-2019.

38

Spring electrofishing Spring electrofishing was not scheduled for 2019 because it was a walleye mark- recapture year. While no community samples were collected, predator runs were conducted at all eight sites in order to monitor declining trends in smallmouth bass catches in other gears (see below). Black bass population assessments Our spring electrofishing surveys are similar to NYSDEC black bass surveys (Brooking et al. 2018). Surveys are conducted two of every three years (excluding walleye mark-recapture years). No survey was scheduled for 2019, but due to the smallmouth bass disease outbreak in the falls of 2017 and 2018 and concerns about declines in the smallmouth bass population indicated by gill net catches and the 2018 electrofishing survey, a partial survey comprised of only predator runs at our eight standard sites was conducted to assess bass catch rates and population structure. Yearling data are not presented as they typically are derived from catches in all fish runs. We report 2019 adult black bass results below with an emphasis on the most recent years. Catch rates Catches of both black bass species have been variable across years, with largemouth bass varying by a factor of 2.5 over the seven years of our surveys, while smallmouth bass catches varied by a factor of 2.7. Adult largemouth bass catch rate declined by 55% between 2018 and 2019, but were only modestly lower than rates observed from 2011-2015 (Figure 32). Adult smallmouth bass catch rates declined by 50% between 2017 and 2018, but 2019 reflected a 25% increase from 2018 (Figure 31; note 2019 reflects predator runs only and 26% of the 2018 catch came from all fish runs). There are no statistically significant differences in catch rates for either species among years (Wilcoxon/Kruskal-Wallis test, largemouth bass – df = 6, X2 = 7.1, p = 0.32; smallmouth bass – df = 6, X2 = 3.8, p = 0.70). A 50% decline in adult smallmouth bass catch rates observed in 2018 coincided with a disease outbreak involving LMBv in the fall of 2017. However, another disease outbreak in the fall of 2018, which anecdotally resulted in more severe lesions as well as observed incidents of mortality did not result in further declines in electrofishing catch rates, and catches increased between 2018 and 2019. Combined with declines in smallmouth bass gill net catches (see above), we have two indicators of reduced smallmouth bass population size, but patterns in catch rates are not entirely consistent with the disease as a primary factor in those declines, as gill net catches began to decline before observations of fish presenting disease. The decline in largemouth bass catch rates between 2018 and 2019 would not appear to be attributable to disease as we had no observations of the disease affecting largemouth bass.

39

Figure 32. Spring electrofishing catch rates of stock size (200 mm) and larger smallmouth and largemouth bass in Oneida Lake, New York, 2011-2019. Population structure Largemouth bass Size distributions of largemouth bass captured in spring electrofishing samples have been highly variable across the years 2011-2019 (Table 4, Figure 33). Proportional stock density (PSD) has also been highly variable, ranging from 26.7 to 76.7 across our years of sampling. Proportional stock density was 76.7 in 2019, reflecting low catches of stock sized fish. Largemouth bass in Oneida Lake typically recruit into the stock size class at age 2 or 3, and these results reflect high inter-annual variability in abundance of these age groups or the effectiveness with which they are sampled. Annual variations in age-0 largemouth bass catches in fyke nets and beach seines do not appear consistent with observed variations in stock size catches 2-3 years later. These results suggest that survival from age-0 to stock size is variable, or that presence of these subadult size classes inshore at time of sampling is inconsistent.

40

Table 4. Catch rates (fish/h; 95% CI) by length category, PSD and RSD-P for largemouth bass collected in spring electrofishing samples in Oneida Lake, New York, 2011-2019.

Length category catch-per-unit effort (fish/h) and 95% confidence

interval

Year Number

captured

Total

effort

(h)

All fish

Yearlings

(≤ 150 mm) –

all fish runs

only

≥ Stock size

(200 mm)

≥Quality

size

(300 mm)

≥ Preferred

size

(380 mm)

PSD

RSD-P

2011 127 12 10.6 5.0 – 16.1

0.8 -0.1 – 1.6

9.6 4.5 – 14.7

7.1 3.7 – 10.4

3.3 1.8 – 4.7

72.6 33.3

2012 120 12 9.8 3.4 - 16.3

1.3 -0.9 – 3.4

8.0 2.9 – 13.1

5.8 2.2 – 9.3

2.8 1.3 – 4.2

71.9 34.4

2014 146 12 12.1 4.6 - 19.7

0.3 -0.3 – 0.8

11.8 4.5 – 19.0

7.1 3.2 – 11.0

3.0 1.1 – 4.9

60.7 26.4

2015 113 12 9.4 4.3 – 14.6

0.0

9.4 4.2 – 14.6

8.6 4.2 – 13.0

4.3 2.6 – 6.1

91.2 46.0

2017 269 12 22.4 7.2 – 37.6

0.5 -0.2 – 1.2

18.4 5.7 – 31.1

4.9 2.4 – 7.5

3.5 1.4 – 5.6

26.7 19.0

2018 211 12 14.7 3.5 – 25.8

0.8 -0.4 – 1.9

16.7 6.1 – 27.2

9.1 3.1 – 15.1

1.3 0.4 – 2.1

54.9 8.3

20191 60 7.5 0.9 – 14.4

5.8 1.5 – 10.0

1.1 0.2 – 2.1

76.7 15.0

1Predator runs only (8 hr total effort).

41

Figure 33. Length-frequencies of largemouth bass from spring shoreline electrofishing samples in Oneida Lake, New York, 2017-2019. Age structure of largemouth bass captured in spring electrofishing samples confirms suggestions of variable recruitment indicated by size structure analyses (Table 5). Catches in 2019 suggested a small 2017 year class, but continued good representation of the 2016 year class, both consistent with age-0 fyke net samples. The 2015 year class dominated the 2019 adult catch, despite being poorly represented as age 2 in 2017.

0

20

60 110 160 210 260 310 360 410 460 510Nu

mb

er

Length class (10 mm)

2017

0

10

20

30

60 110 160 210 260 310 360 410 460 510

Nu

mb

er

Length class (10 mm)

Stock Quality Preferred2018

42

Table 5. Percent composition by age of largemouth bass collected in spring electrofishing samples in Oneida Lake, New York, 2011-2019.

Age 2011

N=83

2012

N=113

2014

N=105

2015

N=76

2017

N=100

2018

N=175

2019

N=60

1 7.2 8.8 1.9 24.0 4.0

2 6.0 13.3 16.2 13.0 26.9 8.3

3 14.5 20.4 28.6 9.2 27.0 40.0 26.7

4 15.7 11.5 10.5 18.4 1.0 17.1 31.7

5 9.6 6.2 8.6 22.4 4.0 4.0 15.0

6 18.1 10.6 5.7 5.3 12.0 2.3 6.7

7 16.9 8.0 11.4 3.9 7.0 2.9 5.0

8 7.2 14.2 7.6 11.8 4.0 5.0

9 2.4 4.4 6.7 21.1 5.0 1.1 1.7

10 1.2 2.7 1.0 7.9 1.0 0.6

11+ 1.2 2.0 2.0 1.1

Smallmouth bass Size distributions of smallmouth bass captured in spring electrofishing samples were variable across the years 2011-2019 with poor representation of smaller size classes in 2014, 2015, 2018 and 2019, similar to that observed with largemouth bass. (Table 6, Figure 34). Proportional stock density has ranged between 37% and 69%. Annual variations in age-0 smallmouth bass catches in fyke nets and beach seines do not appear consistent with observed variations in stock size catches 2-3 years later. While the 2013 year class produced the lowest fyke net catches we’ve observed, the 2014 year class produced among the highest catches on record and both year classes produced comparably low catches as age 1 in electrofishing samples. Despite high yearling catches in 2017, catches of stock size fish were low in 2018. As with largemouth bass, catches of subadult smallmouth bass in spring electrofishing samples are highly variable and frequently inconsistent with other measures of year class strength, suggesting vulnerability of non-spawning bass to spring sampling may vary with factors not related to abundance.

43

Table 6. Catch rates (fish/h) by length category, PSD and RSD-P for smallmouth bass collected in spring electrofishing samples in Oneida Lake, New York, 2011-2019.

Length category catch-per-unit effort (fish/h) and 95% confidence interval

Year Number captured

Total effort (h)

All fish

Yearlings (≤ 150 mm) – all fish runs

only

≥ Stock size

(200 mm)

≥Quality size

(300 mm)

≥ Preferred size

(380 mm)

PSD

RSD-P

2011 39 12 3.3

1.0 – 5.5 1.3

-0.03 – 2.5 1.8

0.06 – 3.4 1.5

0.3 – 2.7 1.3

0.2 – 2.5

66.6 59.3

2012 56 12 4.7 1.9 – 7.5

2.0 0.9 – 3.9

3.6 1.3 – 5.9

1.8 0.9 – 2.8

1.5 0.7 – 2.3

51.2 41.9

2014 45 12 3.8 1.4 – 6.3

0.0

3.7 1.2 – 6.1

2.0 -0.2 – 4.2

1.7 -0.6 – 3.9

55.8 46.5

2015 52 12 4.0 2.0 – 6.1

0.0

4.3 2.3 – 6.2

3.3 1.6 – 5.0

1.8 0.6 – 3.1

84.0 44.0

2017 86 12 7.2 0.8 – 13.5

8.3 -2.2 – 18.7

3.2 1.6 – 4.7

1.2 0.4 – 1.9

1.1 0.3 – 1.8

36.8 34.2

2018 20 12 1.7 0.9 – 2.4

0.3 -0.3 – 0.8

1.6 0.8 – 2.3

0.9 0.3 – 1.6

0.4 0.1 – 0.7

52.6 26.3

20191 19 2.0 0.4 – 3.6

1.4 -0.2 – 3.0

0.6 -0.4 – 1.6

68.8 31.1

1Predator runs only (8 hr total effort).

44

Figure 34. Length-frequencies of smallmouth bass from spring shoreline electrofishing samples in Oneida Lake, New York, 2017-2019. Age structure of smallmouth bass captured in spring electrofishing samples confirm suggestions of variable recruitment and/or vulnerability of younger fish to spring electrofishing indicated by size structure analyses (Table 7). The 2017 year class was well-represented in 2019 samples, but 2018 samples at age 1 suggested a poor year class. The 2019 sample produced no fish older than age 6, the first time this has happened (although sample size was low). Age-4 fish were the most common in the 2019 sample, while age 2 or 3 fish have typically been most abundant in other years.

0

5

10

15

60 110 160 210 260 310 360 410 460 510

Nu

mb

er

Length class (10 mm)

2017

45

Table 7. Percent composition by age of smallmouth bass collected in spring electrofishing samples in Oneida Lake, New York, 2011-2019.

Age 2011 N=30

2012 N=49

2014 N=44

2015 N=51

2017 N=73

2018 N=18

2019 N=18

1 36.7 14.3 2.3 3.9 58.9 5.6 11.1 2 16.7 30.6 22.7 7.8 16.4 38.9 22.2 3 3.3 14.3 29.5 29.4 6.8 27.8 22.2 4 10.0 8.2 4.5 15.7 11.1 27.8 5 6.7 8.2 2.3 5.9 4.1 11.1 6 6.7 6.1 6.8 7.8 5.5 5.6 5.6 7 6.7 10.8 20.5 13.7 5.5 5.6 8 3.3 6.8 9.8 1.4 9 6.7 2.0 3.9 1.4

10 3.3 2.0 4.5 2.0 11+ 4.1

Growth and condition Largemouth bass Largemouth bass typically recruit into the stock size class at age 2, although some fish

had not achieved a length of 200 mm by the spring of their second year (Table 8). Fish

typically did not reach quality length until age 4, and preferred length until age 6.

Observed lengths-at-age in 2019 were consistent with previous years, but did reflect

increases in length of age-7 fish. Time trends in mean length-at-age were positive for

all ages except ages 6 and 8, and increases in mean length with year were significant

for ages-2 and 4 largemouth bass (linear regression – age 1 r2 = .26, p = 0.38; age 2 r2

= 0.67, p = 0.04; age 3 r2 =0 .40, p = 0.13; age 4 r2 = 0.80, p = 0.01; age 5 r2 = 0.38, p =

0.14; age 6 r2 = 0.01, p = 0.85; age 7 r2 = 0.60, p = 0.04; age 8 r2 = 0.51, p = 0.07).

46

Table 8. Length-at-age for largemouth bass (± 1SE) collected in spring electrofishing

samples in Oneida Lake, New York, 2011-2019.

Year

1

2

3

Age

4

5

6

7

8

2011 127 (±7) 206 (±11) 275 (±9) 310 (±7) 353 (±6) 389 (±5) 406 (±6) 430 (±9)

2012 121 (±6) 165 (±7) 256 (±6) 316 (±7) 337 (±7) 374 (±5) 391 (±7) 409 (±6)

2014 145 (±12) 229 (±6) 299 (±5) 328 (±8) 363 (±6) 391 (±7) 407 (±6) 428 (±8)

2015 296 (±11) 325 (±7) 359 (±4) 384 (±9) 407 (±12) 421 (±8)

2017 149 (±4) 235 (±7) 287 (±6) 345 (±25) 369 (±9) 405 (±5) 412 (±8) 410 (±11)

2018 130 (±8) 253 (±3) 300 (±4) 335 (±4) 365 (±11) 378 (±11) 411 (±18) 410 (±11)

2019 244(±10) 290(±7) 336(±4) 357(±6) 373(±4) 437(±12) 393(±4)

Condition of largemouth bass, as measured by relative weight, was at or above 100% for all size classes in all years when fish were weighed with the exception of preferred size fish in 2014 and 2019 (Table 9). Relative weights of largemouth bass collected in 2019 were consistent for stock-size fish but lower than all previous years for quality and preferred-size bass. Table 9. Relative weights by size class (± 1SE) of largemouth bass collected in spring electrofishing samples in Oneida Lake, New York, 2014-2019.

Year Stock

Size class Quality

Preferred

2014 116.0 (±1.4) 106.9 (±1.9) 99.0 (±1.4) 2015 115.8 (±3.2) 104.6 (±1.8) 100.1 (±1.2) 2017 119.7 (±1.1) 111.4 (±3.1) 102.5 (±1.5) 2018 118.0 (±1.2) 109.4 (±1.1) 101.0 (2.5) 2019 118.7 (±2.8) 100.2 (±1.8) 95.3 (±1.8)

Smallmouth bass Smallmouth bass typically recruit into the stock size class by age 2 (Table 10).

Smallmouth bass typically attain quality length at age 4 (although some age-3 fish reach

this length) and preferred length at age 5. Memorable lengths are frequently attained by

age 7. Yearling smallmouth bass are typically smaller than yearling largemouth bass

but by age 5 smallmouth bass are often larger than largemouth bass and this length-at-

age advantage persists at older ages. Lengths-at-age observed in 2019 reflected

increases over previous years for fish younger than age 4. Nonsignificant increases in

47

smallmouth bass length-at-age with time were exhibited for ages 1-3, variable in all

other ages except for age-7, which shows a significant increase (linear regression – age

1 r2 = .35, p = 0.16; age 2 r2 = 0.40, p = 0.13; age 3 r2 = 0.47, p = 0.09; age 4 r2 = 0.01,

p = 0.97; age 5 r2 = 0.03, p = 0.73; age 6 r2 < 0.01, p = 0.95; age 7 r2 = 0.88, p = 0.01;

age 8 r2 = 0.05, p = 0.72).

Table 10. Length-at-age for smallmouth bass (± 1SE) collected in spring electrofishing

samples in Oneida Lake, New York, 2011-2019.

Year 1

2

3

Age 4

5

6

7

8

2011 108 (±5) 212(±12) 235 (n=1) 349 (±14) 383 (±4) 421 (±19) 419 (±22) 410 (n=1)

2012 125 (±6) 209 (±6) 250 (±12) 308 (±23) 395 (±4) 411 (±8) 422 (±5)

2014 93 (n=1) 191 (±8) 248 (±12) 367 (±15) 427 (n=1) 423 (±12) 431 (±8) 447 (±3)

2015 89 (4) 191 (±4) 291 (±9) 313 (±9) 366 (±9) 382 (±10) 428 (±9) 451 (±7)

2017 130 (±2) 224(±11) 234 (±7) 402 (2) 422 (±3) 434 (±6) 471 (n=1)

2018 132 (n=1) 225 (±9) 292 (±16) 339 (±8) 432 (n=1) 434 (±6) 405 (n=1)

2019 151 (±3) 243 (±8) 317 (±16) 339 (±18) 401 (±14) 401 (n=1)

Condition of smallmouth bass, as measured by relative weight, increased relative to previous years for stock and preferred-size fish (Table 11). Table 11. Relative weights by size class (± 1SE) of smallmouth bass collected in spring electrofishing samples in Oneida Lake, New York, 2014-2019.

Year Stock

Size class Quality

Preferred

Memorable

2014 110.5 (±21) 83.0 (±7.1) 94.0 (±1.8) 95.9 (±3.7) 2015 108.8 (±1.4) 106.6 (±2.8) 102.4 (±2.1) 89.3 (±2.7) 2017 115.0 (±2.6) 106.8 (±3.9) 95.9 (n=1) 2018 118.7 (±3.2) 102.9 (±1.8) 85.0 (±13.3) 84.4 (±2.9) 2019 124.0 (±6.7) 101.7 (±4.2) 107.8 (± 4.2)

Shoreline seining To address potential shifts in habitat use by young-of-year yellow perch from offshore areas where they have been indexed by our trawl samples to inshore areas, we implemented a seine survey in 2015 (Fetzer 2013, Fetzer et al. 2015). Daytime seining with a 75 ft beach seine with ¼ in mesh is conducted monthly from July through September at 9 sites with noncontinuous long-term data available.

48

Seining in 2019 collected 24 species, lower than the diversity represented in fyke net samples (29 species). Seine samples in 2019 were dominated by young-of-year yellow perch in July, with catches of 161/haul. Yellow perch catches declined rapidly to 58 in August and 21 in September. Round goby was the most commonly caught species in September at 74/haul. Young-of-year Lepomis spp. were captured in both August and September at the highest rates during the current phase of annual seining, consistent with indications from fyke nets that sunfish reproduction was relatively high in 2019. Fall seine catches of yellow perch indicated densities 52% lower than trawl catches in October in 2019, the first time nearshore densities were lower than offshore densities since we initiated regular annual seining in 2015 (Figures 35 and 36; Fetzer et al. 2015).

Figure 35. Fall density (#/ha) of young-of-year yellow perch from bottom trawl samples and shoreline seining, in Oneida Lake, New York, 1961-2019.

49

Figure 36. Ratio of fall trawl and seine densities of young-of-year yellow perch in Oneida Lake, New York, 1961-2019.

Seining provides a more traditional measure of young-of-year black bass production than fyke nets, and may ultimately be our best measure of trends once a longer continuous series is available. Comparisons of relative year class size at age 0 do not agree well between the two gears (Figures 37 and 38). Fyke net catches indicate a large 2015 year class of smallmouth bass followed by four weak years. The seine samples indicate comparable year classes of smallmouth bass from 2015-2018, followed by a weak year class in 2019. Fyke nets indicate a missing year class of largemouth bass in 2017 but this year class was within the range of previous years based on beach seine samples.

Figure 37. Catch rate of young-of-year black bass in beach seine samples, Oneida Lake, New York, 2015-2019.

50

Figure 38. Catch rate of young-of-year black bass in fyke net samples, Oneida Lake, New York, 2008-2019. Creel survey The current approach is to conduct an abbreviated summer survey each year, basing effort on boat counts from a fixed location in May, June and July and basing catch and harvest on exit interviews conducted at three locations during June and July (see Appendix 3 for detailed methods). A complementary full open water roving survey is conducted every fifth year (most recently in 2018). Annual abbreviated summer survey - angler effort Total open water effort was estimated at 176,364 boat hours, a 4% increase from 2018 effort (Figure 39; Appendix Table A13). This represents a reversal of a three year declining trend, but remained below levels observed from 2010-2015.

51

Figure 39. Open water angler effort (boat hours) on Oneida Lake, 2002-2007, 2010-2019.

Annual abbreviated summer survey - species sought Total number of access point interviews conducted during June and July 2019 was 270. Of these anglers, 142 (53%) strictly sought walleye, while 81 (30%) sought only black bass. Anglers who sought some combination of walleye, black bass, yellow perch and panfish comprised the rest of the sample. Of anglers seeking black bass, 11 (14% of bass anglers) indicated they were fishing in a tournament. This level of tournament participation among black bass anglers is low compared to previous years, which typically encountered tournament anglers at a rate of 20% or higher of all black bass anglers. It is likely that annual variability in tournament participation effort, and by extension black bass effort, is in part driven by whether random interview schedules place creel clerks at the one ramp where tournament weigh-ins are commonly held on weekend afternoons. However, the declines in black bass anglers and tournament participation over the last two years might suggest a drop off in the number of bass tournaments on the lake the last two years. Two walleye anglers also reported they were fishing in a tournament (1.4% of all walleye anglers).

Annual abbreviated summer survey - catch and harvest rates and 2019 walleye harvest estimate Festa et al. (1987), based on a survey of walleye fisheries in New York, suggested that walleye catch rates of 0.10-0.25/hr were characteristic of good to very good fisheries,

52

with catch rates exceeding 0.25/hr considered excellent. For targeted catch rates, rates exceeding 0.20/hr were above average and rates approaching 0.50/hr were considered excellent. Estimated catch rate for walleye (all anglers) from access surveys in the 2019 open water season was 0.39/hr in June and 0.49 in July (mean targeted catch rate for the June was 0.72/hr and for July 0.72/hr). Smallmouth bass catch rate (all anglers) was 0.24/hr in June and 0.16/hr in July (mean targeted catch rate was 0.54/hr in June and 0.38/hr in July). Walleye catch rates for all anglers increased 26% from rates observed in 2018, and were the highest during the modern survey period (Figure 40). Targeted catch rates for walleye were the highest observed in the modern survey period, and were 31% higher than the next highest year (2012, Figure 41). Targeted catch rates for black bass were well within the range observed during the current survey period (Figures 40, 41).

Figure 40. Overall angler catch rates for walleye and smallmouth bass in Oneida Lake, 2012-2019.

53

Figure 41. Targeted angler catch rates for walleye and smallmouth bass in Oneida Lake, 2012-2019. Open water harvest rate for walleye for the June/July 2019 period used to predict harvest was 0.21/hr. Estimated total harvest of walleye for the 2019 open water season was 57,824 fish. The 2019 harvest reflected continued increases from the low observed in 2016 and a return to levels observed prior to establishment of round goby (2012-2014 average 58,124; Figure 42). Smallmouth bass harvest rates have remained consistent with previous years at 0.01/hr for the June/July period. The low harvest rate of smallmouth bass is typical of a catch-and-release fishery.

54

Figure 42. Open water walleye harvest in Oneida Lake, 2011-2019.

Creel data from 2016 indicated sharp declines from previous years in walleye catch rates and harvest. Data from 2019 show an improvement in catch rates for walleye to pre-round goby levels. Walleye catch rates in Oneida Lake were previously shown to be sensitive to the availability of natural forage for walleye, with increased forage resulting in reduced catch rates (VanDeValk et al. 2005). If round goby contribute to the same relationship, and winter die-offs are not a regular phenomenon, their increasing densities could result in declines in the quality of the walleye fishery on Oneida Lake. Black bass catch rates have been variable since 2015, but generally below rates observed from 2012-2014

Annual abbreviated summer survey - angler opinion and behavior survey As a complement to the catch and harvest rate data collected from angler interviews, we added additional survey questions in 2013 directed at assessing angler opinions about the quality of the Oneida Lake fishery and its management as well as angler activities in areas of recent or current management concern. Five questions were developed with NYSDEC staff and incorporated into angler interviews. Some questions were designed to allow tracking of opinions over time, and we report here the results of the first five years of the program. Question 1. On a scale of 1 to 5, with 5 being very satisfied, how satisfied are you with the overall quality of fishing on Oneida Lake? From 2013 through 2015, anglers indicated a high level of satisfaction with the quality of angling on Oneida Lake, with average scores of 4 or higher (Figure 43). Satisfaction scores fell to 3.2 in 2016, but rebounded to 3.9 in 2017 and 2018. Satisfaction dropped slightly to 3.7 in 2019 despite high walleye catch rates.

55

Figure 43. Angler level of satisfaction with overall quality of fishing on Oneida Lake from access interviews. 2013 survey N=489, 2014 N=475, 2015 N=320, 2016 N=276, 2017 N=427, 2018 N=390, 2019 N=270.

Question 2. How often do you typically fish Oneida Lake for walleye? Participants in the 2018 creel survey reported they fished Oneida Lake for walleye an average of 27.8 days/year. This question was new to the 2018 survey, so no time trends are yet available. The 2019 creel survey inadvertently used old forms, so no answers to this question were collected in 2019. Question 3. Do you fish any other waterbodies in New York State for walleye? If yes, how often?

Of the 322 anglers responding in 2018, 37% indicated that they fished other waters than Oneida Lake for walleye. Of those who did, they reported spending an average of 9.4 days fishing other waters. Taken with question 2, these results indicate that Oneida walleye anglers spend about a third of their total fishing time on other waterbodies. This question was new to the 2018 survey, so no time trends are yet available. The 2019 creel survey inadvertently used old forms, so no answers to this question were collected in 2019.

56

Question 4. What is your opinion about the current daily possession limit for walleye on Oneida Lake? a – continue as is, b – change to statewide possession limit. Results indicated that a majority of anglers from all years of interviews felt the current three fish possession limit for walleye should continue (Figure 44). Interestingly, the percentage of anglers expressing this opinion increased steadily in each year of the survey before dropping slightly in 2018 and 2019.

Figure 44. Angler opinions on the current daily possession limit for walleye on Oneida Lake (3 fish) as opposed to a change to the statewide limit (5 fish). 2013 N=399, 2014 N=338, 2015 N=320, 2016 N=281, 2017 N=427, 2018 N=322, 2019=270..

57

Question 5. On this fishing trip, which of the following have you fished with? a – artificial lures, b – natural baits, c – both. Artificial lures were the most commonly used technique in all years of interviews (Figure 45). Natural baits were commonly used and 14-26% of anglers have reported using both, likely a function of the popularity of worm harnesses and jigs tipped with worms in the Oneida Lake walleye fishery. There was an increase in use of artificial lures concurrent with a reduction in worm-oriented baits in 2016 which reversed in 2017-2019. Of natural baits used in 2019, interviews revealed 130 anglers using worms, 21 using baitfish, and 1 using crayfish. Use of baitfish was noticeably higher than recent years.

Figure 45. Percentage use of artificial lures and natural baits by anglers on Oneida Lake. 2013 N=736, 2014 N=475, 2015 N=320, 2016 N=281, 2017 N=426, 2018 N=381, 2019 N=270.

58

Question 6. On a scale of 1 to 5, with 5 being very satisfied, how satisfied are you with the job the DEC Bureau of Fisheries does managing Oneida Lake? Anglers indicated a high level of satisfaction with DEC’s management of Oneida Lake in 2019, with an average satisfaction score of 4.2 out of 5 (Figure 46). Interestingly, as in past years, satisfaction with DEC management was higher than satisfaction with the fishery itself..

Figure 46. Angler level of satisfaction with management of Oneida Lake by the DEC Bureau of Fisheries. 2013 N=899, 2014 N=438, 2015 N=320, 2016 N=281, 2017 N=426, 2018 N=362, 2019 N=270.

Recommendations for management and future research directions. Over the duration of our research on Oneida Lake, we have identified several ecological changes, many ongoing, that have affected the fish community and fisheries. These have included warming water temperatures, species invasions, and increased water clarity resulting from dreissenid mussels establishment and reduced nutrient inputs. The data collected in 2019 are consistent with previous results showing that the lake has undergone changes in physical characteristics and productivity at the lower trophic levels. Water temperatures and ice duration continued to reflect warmer conditions than when studies were first initiated, and water clarity remained well above levels observed in the earliest years of our studies. Oneida Lake presently fits the overall characteristics of a mesotrophic system, with reduced primary production from the early decades of our

59

studies when the lake was classified as eutrophic. Productivity has shifted from the pelagic to the littoral zones, including increases in littoral macrophytes and benthic algae production (Cecala et al. 2008), with concomitant increases in abundance of nearshore fish species. We are now seeing signs of reduced Daphnia spp. production, a typical response to dreissenid colonization, but one that was not evident in the first decade following mussel establishment. In 2019, Bythotrephes longimanus was discovered in Oneida Lake. This predacious zooplankter could exert additional pressure on Daphnia in the lake and contribute to further declines. Continued declines in Daphnia densities and conversion of the zooplankton community to one dominated by copepods and Bythotrephes longimanus may have implications for planktivores and planktivorous life stages of important sport fishes. Clearer water conditions are correlated with reduced survival of pelagic walleye and yellow perch fry, resulting in lower average year class size and recruitment to subadult stages than was typical of the lake before major ecological changes were observed. Further reductions in Daphnia combined with apparent reduction in mussel biomass due to predation by round goby could lead to reductions in water clarity which may lead to increased survival of larval percids. We have seen indications of increased survival of larval walleye and yellow perch concurrent with declines in the white perch population over much of the last decade and reduced water clarity could act synergistically with reduced predator abundance to allow larger year classes of walleye and yellow perch than we saw through most of the 2000s. Double-crested cormorant predation on subadult percids resulted in decreases in recruitment to the fishery, and the establishment of a management program contributed to increases in adult walleye numbers, but we had seen only modest increases in yellow perch numbers as double-crested cormorant management continued. However, the adult yellow perch population has doubled since 2016 as a result of two large year classes. Hazing efforts contribute to keeping predation pressure by double-crested cormorants at levels that are well below those observed prior to any management being implemented, and gizzard shad commonly buffer much of the impact in the fall. Round goby, if a high density population persists, may provide further buffering for percids from double-crested cormorant predation. At present, double-crested cormorant management efforts have reduced impacts on sport fish species. While summer numbers of double-crested cormorants are below levels that threaten sport fish production, high fall migrant numbers have resulted from cessation of management of populations further north, and likely increase impacts on young percids despite availability of gizzard shad and round goby. Round goby, established in the lake in 2014, were numerically more common in double-crested cormorant diets than any other species in 2016. They were the most commonly consumed species again in 2018, even when an abundant gizzard shad year class became available in the fall. However, in 2019 gizzard shad were again the most common diet item in double-crested cormorants, despite one of the smallest gizzard shad year classes seen in decades and continued increases in round goby numbers after a decline over the winter of 2016-2017. Round goby may provide an additional buffer for sport fish species against double-crested cormorant predation, and one that would be available all season, as opposed to the late-summer and fall availability of gizzard shad. Additional years of

60

data will be required to determine if round goby die-offs such as observed in the winter of 2016-2017 will be an isolated or regular event and the availability of larger round goby to double-crested cormorants early in the growing season will likely determine their value as a buffer for sport fish. Harvest of walleye in 2016 dropped sharply from previous years, and we attributed this to round goby providing plentiful natural forage for walleye. Harvest increased in 2017 and 2018 from the level observed in 2016, but did not recover fully until the 2019 season to levels seen prior to round goby. A large walleye year class from 2014 recruited into the fishery in 2018, and combined with the reduced harvest, we predicted the 2019 walleye mark-recapture population estimate would be well over 600,000 fish. The estimate, in fact, indicated the adult walleye population had exceeded one million fish for the first time since 1986. The unexpected magnitude of the increase stems, in part, from systematic underestimation of abundance at age 3 of the year classes that recruited into the population after the 2016 mark-recapture. A large year class from 2016 will recruit into the fishery in 2020. The 2018 year class also has potential to produce a solid number of recruits. For the near term, it appears that the adult walleye population could remain near or above one million fish, and it may be time to consider lifting the three fish daily regulation on Oneida Lake and shift to the statewide regulation of five fish. Consistently poor white perch year classes in recent years have resulted in a decline in the white perch adult population. This has resulted in enhanced survival of both yellow perch and walleye larvae, which has contributed to population growth of both species. A large 2012 year class of white perch has resulted in increased catches of white perch in gill nets and the 2017 year class may also be large by recent standards. The 2019 year class appears larger than those that led to the declining adult population, but not as large as the year classes that contributed to the population peak in the late 2000s. If the white perch population returns to those levels we would expect to see negative impacts on year class strength of percids. Smallmouth bass and largemouth bass have benefited from changes in the lake, and their populations have reached higher levels than were observed in the 1960s and 1970s. The presence of round goby as potential nest predators of bass eggs and larvae will need to be monitored, but based on record catches of young smallmouth bass in fyke nets in 2015 and of largemouth bass young in 2016, there were no early indications of a detrimental impact of round goby on production of young bass. Smallmouth bass have produced three relatively small year classes in a row, but we have not observed reductions in largemouth bass year class production. We will continue to assess potential impacts of round goby on black bass year class production. An outbreak of bacterial lesions in the fall of 2017 affected smallmouth bass on an unknown scale. While prevalent in some locations, there were no reports of mortalities. The condition repeated in the fall of 2018, and anecdotally appeared to lead to more severe presentations in smallmouth bass, and mortalities were reported by anglers, although the extent is unknown. Affected fish were diagnosed with a combination of LMBv, bacterial lesions and heavy helminth loads. No outbreak was detected or

61

reported in 2019. Concurrent with the disease outbreak we saw reduced electrofishing and gill net catches of adult smallmouth bass. Largemouth bass electrofishing catch rates dropped in 2019, but no reports or observations indicated they were affected by the disease. The timing of reductions in the smallmouth bass population and the progress of the disease is not entirely consistent, so the ultimate causative factors underlying the decrease are unknown. We will continue to monitor adult bass populations and year class formation. The round goby is now established in the lake, and we may see several responses in our data series. Growth of adult walleye, yellow perch and black bass may increase if gobies provide an abundant food resource. The round goby spawns multiple times in a growing season, and may provide a prey resource for young-of-year walleye and black bass throughout the summer which could improve growth, and presumably survival. We have seen indications of improved young-of-year smallmouth bass growth with round goby appearing in diets. However, round goby can negatively impact angler catch rates by providing an abundant food source for piscivores Perhaps more importantly, we now see reductions in both density and biomass of dreissenid mussels

which we attribute to predation by round goby. Continuing analysis and monitoring of the Oneida Lake data set should give us information on the response to these ongoing ecological changes that are relevant not only to Oneida Lake, but also to the northeastern US and southeastern Canada. Recommendation for current management: Fisheries management on Oneida Lake includes stocking of walleye larvae, size and creel limits for walleyes, black bass, and other species, and control of double-crested cormorants. We recommend maintaining cormorant management and annual stocking at current levels in the future. The need for continuation of special walleye regulations should be considered. Stocking of walleye larvae. Stocking maintains a consistent supply of walleye larvae to the lake and makes walleye less sensitive to potential increases in egg predation from a future abundant round goby population. Our best estimates suggest that the number of naturally produced walleye larvae in the lake is about 33% of the numbers stocked. Due to the covid-19 outbreak, it will not be possible to produce enough fry to stock Oneida Lake in 2020. Given the current adult population, we do not anticipate this will result in any jeopardy to the sustainability of the fishery. Size and creel limits for walleye. The adult walleye population was estimated at one million fish in the 2019 mark-recapture study. Even under historic harvest rates, we feel that the population will exhibit continued growth in 2020 as a result of a large 2016 year class. We suggest maintaining the current size limit for Oneida Lake walleye at 15 inches. The three fish creel limit is a conservative approach to reduce impacts of poor walleye recruitment and enjoys strong support from anglers, but at current population levels implementation of the statewide five fish limit should be considered. Increased early survival due to decreased water clarity, decreased white perch abundance and expansion of round goby, providing both an additional food source for young walleye

62

and an additional buffer against predation on young walleye, could result in regular levels of recruitment that would strain the predator-prey balance observed for so many years. Liberalization of the bag limit may help ease an increasing predator demand on prey fish populations, particularly yellow perch. Double-crested cormorant control. We have observed an increase in the walleye and yellow perch populations concomitant with more intensive double-crested cormorant control. This suggests that removing double-crested cormorants does increase percid recruitment to the fishery. More intensive double-crested cormorant control by APHIS was conducted from 2004-2010. Our data do show that a rebuilding of summer double-crested cormorant numbers will likely reduce subadult walleye and yellow perch survival, and potentially reduce populations (DeBruyne 2013, Coleman et al. 2016). Continued efforts should be made to find a workable approach to limiting cormorant numbers and preventing the rebuilding of a large summer population. Fall migrant numbers have increased due to a court injunction on management of the Lake Ontario population. Consistently high fall numbers over an extended period of time are a concern, but effective action will require restoration of the depredation permit to allow control of double-crested cormorant colonies that produce fall migrants to Oneida Lake. Given the results and discussion in this report, we recommend the following research and monitoring activities in 2020: 1) Continue standard sampling program. This program includes limnological surveys, two larval fish sampling surveys (8 and 18 mm yellow perch surveys), a spring centrarchid survey, 15 standard gill nets, trawl surveys three weeks of every four from mid-July through October, pelagic prey fish survey with acoustics, midwater trawl and pelagic gill nets at the end of August, fyke net sampling for nearshore fish in September, and large mesh gill nets for lake sturgeon. 2) Maintain attention to the importance of changing spatial distributions of age-0 yellow perch. Continue summer seining to complement trawling in order to sample yellow perch in both offshore and near shore habitats. Replace one weekly trawl sample per month with a seine survey one week per month from July to September. Combined trawling and seining may also provide a potential inshore/offshore index of round goby. Seining will also provide a more traditional index of black bass year class production than the fyke nets we currently rely on. 3) Continue diet analyses of double-crested cormorants in coordination with NYSDEC. 4) Implement a standardized video-based assessment of round goby densities in early summer for comparison to trawl and seine catches. The lakewide surveys conducted in 2018 and 2019 suggest that video is the best available technique for estimating round goby density, but the scale of those surveys is not logistically sustainable over the long

63

term. Use those data to develop a survey with a reduced number of sites (~30) and incorporate that into our standardized sampling schedule.

64

Literature Cited Brooking, T.E., J.R. Jackson, L.G. Rudstam, and A.J. VanDeValk. 2014. Habitat mapping of Oneida and Canadarago Lakes. Final Report, Study 2, Job 4. New York State Department of Environmental Conservation, Albany, New York. Brooking, T. E., J. Loukmas, J. R. Jackson, A. J. VanDeValk. 2018. Black bass and sunfish sampling manual for lakes and ponds in NY. Sportfish Restoration Grant F-63-R, Job 2-2.3. NYS Department of Environmental Conservation. Albany, NY. 62 pp. Brooking, T.E., L.G. Rudstam, S.D. Krueger, J.R. Jackson, A.B.Welsh, W.W. Fetzer. 2010. First occurrence of the mysid Hemimysis anomala in an inland lake in North America, Oneida Lake, NY. Journal of Great Lakes Research 36(2010):577-581. Brown, M., B. Boscarion, J. Roellke, E. Stapylton, and A. Driller-Colangelo. 2014. Fifteen miles on the erie Canal: the spread of Hemimysis anaomala G.O. Sras, 1907 (bloody red shrimp) in the New York State canal system. BioInvasion Records 3:261-267. Carlson, R.E. 1977. A trophic state index for lakes. Limnology and Oceanography 22:361-369. Cecala, R.K., C.M. Mayer, K.L. Schultz, and e.L. Mills. 2008. Increased benthic algal production in response to the invasive zebra mussel (Dreissena polymorpha) in a productive ecosystem, Oneida Lake, New York. Journal of Integrative Plant Biology 50:1452-1466. Chevalier, J. R. 1973. Cannibalism as a factor in first-year survival of walleyes in Oneida Lake. Transactions of the American Fisheries Society 103:739-744. Coleman, J. T. H. 2009. Diving behavior, predator-prey dynamics, and management efficacy of Double-crested Cormorants in New York State. PhD thesis, Cornell University, Ithaca, New York. Coleman, J.T.H., R.L. DeBruyne, L.G. Rudstam, J.R. Jackson, A.J. VanDeValk, T.E. Brooking, C.M. Adams, and M.E. Richmond. 2016. Evaluating the influence of double-crested cormorants on walleye and yellow perch populations in Oneida Lake, New York. Pages 397-424 in Rudstam, L.G., E.L. Mills, J.R. Jackson, and D.J. Stewart, editors. 2016. Oneida Lake: Long-term dynamics of a managed ecosystem and its fishery. American Fisheries Society, Bethesda, Maryland.

Crane, N.A., J.M. Farrell, D.W. Einhouse, J.R. Lantry, and J.L. Markham. 2015. Trends in body composition of native piscivores following invasion of Lake Erie and Ontario by the round goby. Freshwater Biology 60:111-124.

65

DeBruyne, R.L., J.T.H. Coleman, J.R. Jackson, L.G. Rudstam, and A.J. VanDeValk. 2013. Analysis of prey selection by double-crested cormorants: a 15-year diet study in Oneida Lake, New York. Transactions of the American Fisheries Society 142:430-446. Festa, P.J., J.L. Forney, and R.T. Colesante. 1987. Walleye management in New York State: A plan for restoration and enhancement. New York State Department of Environmental Conservation, Albany. Fetzer, W.W. 2013. Disentangling the effects of multiple ecosystem changes on fish population and community dynamics. Ph.D. Thesis, Cornell University. Fetzer, W.W., T.E. Brooking, J.R. Jackson and L.G. Rudstam. 2011. Over-winter mortality of gizzard shad: evaluation of starvation and cold temperature shock. Transactions of the American Fisheries Society 140:1460-1471. Fetzer, W.W., M.M. Luebs, J.R. Jackson, and L.G. Rudstam. 2015. Intraspecific niche partitioning and ecosystem state drive carbon pathways supporting lake food webs. Ecosystems 18:1440-1454. Forney, J. L. 1980. Evolution of a management strategy for the walleye in Oneida Lake, New York. New York Fish and Game Journal 27:105-141. Forney, J. L., L. G. Rudstam, D. M. Green, and D. L. Stang. 1994. Percid sampling manual. Fish sampling manual. New York Department of Environmental Conservation, Albany, New York. Hetherington, A.L., R.L. Schneider, L.G. Rudstam, G. Gal, A.T. DeGaetano, and M.T. Walter. 2015. Modeling climate change impacts on the thermal dynamics of ploymictic Oneida Lake, New York. Ecological Modeling 300:1-11. Hetherington, A.L., Rudstam, L.G., Schneider, R.L., Holeck, K.T., Hotaling, C.W., Cooper, J.E., Jackson, J.R. 2019. Invader invaded: population dynamics of zebra mussels (Dreissena polymorpha) and quagga mussels (Dreissena rostriformus bugensis) in polymictic Oneida Lake, NY, USA (1992-2013). Biological Invasions 1529-1544. He, J.X., L.G. Rudstam, J.L. Forney, A.J. VanDeValk, and D.J. Stewart. 2005. Long-term patterns in growth of Oneida Lake walleye: a multivariate and stage explicit approach for applying the von Bertalanffy function. Journal of Fish Biology 66:1459-1470. Idrisi, N., E. L. Mills, L. G. Rudstam, and D. J. Stewart. 2001. Impact of zebra mussels, Dreissena polymorpha, on the pelagic lower trophic levels of Oneida Lake, New York. Canadian Journal of Fisheries and Aquatic Sciences 58:1430-1441.

66

Irwin, B. J., T. J. Treska, L. G. Rudstam, P. J. Sullivan, J. R. Jackson, A. J. VanDeValk, and J. L. Forney. 2008. Estimating walleye (Sander vitreus) density, gear catchability, and mortality using three fishery-independent data sets for Oneida Lake, New York. Canadian Journal of Fisheries and Aquatic Sciences 65:1366-1378. Irwin, B.J., L.G. Rudstam, J.R. Jackson, A.J. VanDeValk, J.L. Forney, and D.G. Fitzgerald. 2009. Depensatory mortality, density-dependent growth, and delayed compensation: disentangling the interplay of mortality, growth, and density during early life stages of yellow perch. Transactions of the American Fisheries Society 139:99-110. Jackson, J.R., D.W. Einhouse, A.J. VanDeValk, and T.E. Brooking. 2015. Year-class production of black bass before and after opening of a spring catch-and-release season in New York: case studies from three lakes. Pages 181-191 in M.D. Tringali, J.M. Long, T.W. Birdsong, and M.S. Allen, editors. Black bass diversity: multidisciplinary science for conservation. American Fisheries Society, Symposium 82, Bethesda, Maryland. Jackson, J.R., A.J. VanDeValk, T.E. Brooking, K.T. Holeck, C. Hotaling, and L.G. Rudstam,. 2017. The fisheries and limnology of Oneida Lake 2016. New York State Department of Environmental Conservation. 91 pp. Jackson, J.R., A.J. VanDeValk, T.E. Brooking, K.T. Holeck, C. Hotaling, and L.G. Rudstam,. 2018. The fisheries and limnology of Oneida Lake 2017. New York State Department of Environmental Conservation. 105 pp. Johnson, J.H., R.M. Ross, R.D. McCullough, and A. Mathers. 2010. Diet shift of double-crested cormorants in eastern Lake Ontario associated with the expansion of the invasive round goby. Journal of Great Lakes research 36:242-247. Johnson, T. B. and D. O. Evans. 1991. Behavior, energetics and associated mortality of young-of-the-year white perch (Morone americana) and yellow perch (Perca flavescens) under simulated winter conditions. Canadian Journal of Fisheries and Aquatic Sciences 48:672-680. Johnson, T.B., D.B. Bunnell, and C.T. Knight.2005. A potential new energy pathway in central Lake Erie; the round goby connection. Journal of Great Lakes Research 31(Suppl. 2):238-251. Krueger, S. D., J. R. Jackson, A. J. VanDeValk, and L. G. Rudstam. 2009. The Oneida Lake Creel Survey, 2002-2007. New York State Department of Environmental Conservation, Albany, NY. Lepak, J. M., C. E. Kraft, and B. C. Weldel. 2006. Rapid food web recovery in response to removal of an introduced apex predator. Canadian Journal of Fisheries and Aquatic Sciences 63:569-575.

67

Mayer, C.M., L.E. Burlakova, P. Eklov, D. Fitzgerald, A.Y. Karatayev, S.A. Ludsin, S. Millard, E.L. Mills, A.P. Ostapenya, L.G. rudstam, B. Zhu, and T.V. Zhukova. 2014. The benthification of freshwater lakes: exotic mussles turning ecosystems upside down. Page 575-585 in T.F. Nalepa and D.W. Schloesser, editors. Quagga and zebra mussels: biology, impacts, and control. Second edition. CRC Press, Boca raton, Florida. Responsive Management. 2019. New York angler effort and expenditures 2017. Harrisonberg, Virginia. https://www.dec.ny.gov/docs/fish_marine_pdf/nyas17rpt1.pdf Rudstam, L. G., D. M. Green, J. L. Forney, D. L. Stang, and J. T. Evans. 1996. Evidence of biotic interactions between walleye and yellow perch in New York State lakes. Annales Zoologica Fennici 33:443-449. Rudstam, L. G., A. J. VanDeValk, C. M. Adams, J. T. H. Coleman, J. L. Forney, and M. E. Richmond. 2004. Cormorant predation and the population dynamics of walleye and yellow perch in Oneida Lake. Ecological Applications 14:149-163. Rudstam, L.G., J.R. Jackson, T.E. Brooking, S.D. Krueger, J.L. Forney, W.W. Fetzer, R. DeBruyne, E.L. Mills, J. Swan, A. Seifert, and K. Holeck. 2009. Oneida Lake and its fishery in 2008. New York State Department of Environmental Cnservation, Albany. Rudstam, L.G., E.L. Mills, J.R. Jackson, and D.J. Stewart, editors. 2016. Oneida Lake: Long-term dynamics of a managed ecosystem and its fishery. American Fisheries Society, Bethesda, Maryland. Rudstam, L.G., J.R. Jackson, A.J. VanDeValk, T.E. Brooking, W.W. Fetzer, B.J. Irwin, and J.L. Forney. 2016. Walleye and yellow perch in Oneida Lake. Pages 319-354 in Rudstam, L.G., E.L. Mills, J.R. Jackson, and D.J. Stewart, editors. 2016. Oneida Lake: Long-term dynamics of a managed ecosystem and its fishery. American Fisheries Society, Bethesda, Maryland. Steinhart, G.B., E.A. Marschall, and R.A. Stein. 2004b. Round goby predation on smallmouth bass offspring in nests during simulated catch-and-release angling. Transactions of the American Fisheries Society 133:121-131. Steinhart, G.B., R.A. Stein, and E.A. Marschall. 2004a. High growth rate of young-of-year smallmouth bass in Lake Erie: a result of the round goby invasion? Journal of Great Lakes research 30:381-389. VanderZanden, M. J., J. M. Casselman, and J. B. Rasmussen. 1999. Stable isotope evidence for the food web consequences of species invasions in lakes. Nature 401:464-467.

68

VanDeValk, A.J., J.L. Forney, J.R. Jackson, L.G. Rudstam, T.E. Brooking, and S.D. Krueger. 2005. Angler catch rates and catchability of walleye in Oneida Lake, New York. North American Journal of Fisheries Management 25:1441-1447. VanDeValk, A.J., J.L. Forney, T.E. Brooking, J.R. Jackson, and L.G. Rudstam. 2016. First-year density and growth as they relate to recruitment of white perch to the adult stock in Oneida Lake, New York, 1968–2011. Transactions of the American Fisheries Society 145:416-426. Wetzel, R. G. 2001. Limnology. Lake and river ecosystems. 3rd edition. Academic Press, San Diego, CA, USA. Zhu, B., D. G. Fitzgerald, C. M. Mayer, L. G. Rudstam, and E. L. Mills. 2006. Alteration of ecosystem function by zebra mussels in Oneida Lake: Impacts on submerged macrophytes. Ecosystems 9:1017-1028.

69

Appendix 1:

The fall and rise of the chain pickerel in Oneida Lake

T. VanDeValk, T. Brooking and R. Jackson

Prior to the completion of the Barge Canal in 1916, the chain pickerel Esox niger rivaled the

walleye Sander vitreus as the top piscivore in Oneida Lake (Adams and Hankinson 1928). The

combination of lake level stabilization and draining of marshes for development resulted in the

loss of spawning and nursery habitat for esocids (Forney 1980). By the 1940s, the walleye was

clearly the predominant piscivore with the smallmouth bass Micropterus dolomieu a distant

second (Forney 1980). Once a popular Oneida Lake sport fish (Adams and Hankinson 1916),

the chain pickerel was no longer a component of the angler’s creel. A three-year creel survey

conducted from 1957 - 1960 indicated catches of chain pickerel by Oneida Lake anglers were

scarce (Grosslein 1958, 1960).

Catches of chain pickerel in Cornell Field Station standard fisheries surveys were generally low

before the late 1990s. Most standard sampling surveys, initiated in 1956, rarely sampled chain

pickerel probably because surveys were designed to monitor offshore walleye and yellow perch

Perca flavescens populations. Two standard surveys that sampled littoral habitats and caught

chain pickerel were seine surveys (young of year, yoy) and trap-net surveys (age-1 and older).

A 75 foot beach seine was pulled at each of 9 standard sites in July (32 years) and August (30

years) since 1959. A 6 foot trap net described by Forney et al. (1994) was fished in the same

location near Shackelton Point in the spring (N = 29, most years since 1985) and in the fall (N =

39, 1957 - 1960 and most years since 1980). Spring trap nets were typically fished for about 8

days (range, 3 - 15 days) in April - May and fall trap nets were fished for about 11 days (range,

3 - 24 days) in October and sometimes into early November. Catch per effort (cpe) for each

survey was calculated by dividing the total chain pickerel catch by the number of net-nights.

Angler catches were monitored from 1994 – 1998 (angler diary program), 2002 – 2007 (roving

creel surveys) and 2011 – 2019 (access point creel surveys). Whole lake roving creel surveys

and access point creel surveys were conducted in 2002, 2006, 2013 and 2018. Angler catch

rates (#/angler-h) of chain pickerel from the two methods were closely correlated (N = 4, r =

0.98, p = 0.02) so access point data were used for comparisons to other gears for 2013 and

2018. Correlation analysis was used to compare seasonal catches within the same gear.

Simple linear regression was used to assess relationships among gear and to lake level.

The catch of yoy chain pickerel in beach seine surveys was zero for all years surveyed prior to

2004 except 1963 when 1 yoy chain pickerel was captured. From 2004 – 2016, yoy chain

pickerel were caught in 7 of the 10 years surveyed (no seine surveys were conducted in 2010,

2012 and 2013), followed by none being caught the last 3 years (Figure 1, Panel A). Year

classes that appeared in seine surveys generally contributed to the trap-net catch 1 to 2 years

later. Observed increases in yoy catches of chain pickerel in seine surveys since 2000 cannot

70

be explained by spring lake levels. In New York State, chain pickerel typically spawn in April

(Underhill 1949). Young of year chain pickerel catches in seine surveys were not related to

mean April lake levels measured at Cleveland, NY (New York State Canal Corporation).

0

2

4

6

8

1990 2000 2010 2020

Se

ine

ca

tch

Year

0

2

4

6

8

10

12

1990 2000 2010 2020

Tra

p-n

et

cp

e

Year

Spring

Fall

Figure 1. Mean July - August (July only in 2004 and August only in 2006) yoy chain pickerel

catch in beach seine surveys since 1990 (Panel A) and age-1 and older chain pickerel cpe in

spring and fall trap-net surveys since 1990 (Panel B). Note only 1 yoy chain pickerel was

caught in years seine surveys were conducted between 1959 and 1989.

From 1957 – 1960, fall trap-net surveys caught chain pickerel in 2 of the 4 years surveyed (cpe

< 0.50 for both years). After 1960, no standard trap-net surveys were conducted until 1980 (fall

survey) and 1985 (spring survey). However, beginning in 1969 and occurring every other year

through 1977, spring trap nets were set for percid mark-recapture studies in up to 8 locations

around Shackelton Point from April into May. Catches and effort from all sites were pooled, and

the resulting chain pickerel cpe ranged from 0.00 – 0.03, except for 1969 when the cpe was

0.13. While not comparable to standard trap-net survey catch rates, these data indicate age-1

and older chain pickerel abundance remained low from when standard trap-net surveys were

terminated in 1960 until the 1980s when surveys were resumed. Since 1980, no age-1 and

older chain pickerel were caught in either standard survey until 1998 (fall trap net) and 1999

(spring trap net). In both datasets, catches increased slowly between 2000 and 2010 but

exhibited higher catches in some of the subsequent years, especially in spring trap-net surveys

(Figure 1, Panel B). Spring and fall trap-net catches were correlated (1985 - 2019, r = 0.45, p =

0.02).

A

B

71

Angler catch rates of chain pickerel, based on all trips, ranged between 0.00/angler-h to

0.40/angler-h (mean = 0.11/angler-h). Catch rates in recent years (since 2011) have averaged

0.20/angler-h. Despite the recent increase in chain pickerel abundance, their popularity as a

sport fish did not keep pace. Of the 7,190 interviews conducted since 2011, only 26 angling

parties (0.3%) reported chain pickerel as one of the species they were targeting. Prior to

completion of the canal, chain pickerel ranked fifth in Oneida Lake angler sport fish preference

(Adams and Hankinson 1916). Catches in both trap-net datasets were significantly and

positively related to angler catch rates of chain pickerel (Figure 2).

ACR = 0.031STC + 0.04R² = 0.68, p<0.001

0.0

0.1

0.2

0.3

0.4

0.5

0 2 4 6 8 10 12

An

gle

r c

atc

h r

ate

Spring trap-net cpe

A

ACR = 0.037FTC + 0.06R² = 0.16, p=0.05

0.0

0.1

0.2

0.3

0.4

0.5

0 1 2 3 4 5 6

An

gle

r c

atc

h r

ate

Fall trap-net cpe

B

72

Figure 2. Angler catch rate (ACR) versus spring trap-net cpe (STC, Panel A) and fall

trap-net cpe (FTC, Panel B).

As Oneida Lake changed physically and biologically, new surveys employing different

gears were implemented to track the effects on fish populations. In 2007 a standard

fyke-net survey (described in the main report) was implemented to track changes in the

littoral fish community, and in 2011 nighttime shoreline electrofishing surveys (described

in the attached report) were initiated to track changes in Centrarchid sp. abundances.

Fyke-net cpe was related only to fall trap-net cpe (F(1, 9 )= 7.55, p = 0.02, R2 = 0.40). The

mean annual fyke-net cpe was less than 0.5 for 12 of the 13 years surveyed, which does

not allow for year to year variation in chain pickerel abundances. Pope et al. (2009)

recommends conducting fyke-net surveys in spring when fish are most active (around

spawning season) so our September surveys may not sample esocids effectively.

Based on only 6 years of comparisons, the electrofishing catch rate (#/h) was

significantly and positively related to the fall trap-net cpe (F(1, 4) = 9.64, p = 0.04, R2 =

0.63), but not the spring trap-net cpe. As we accumulate additional years of

electrofishing data for comparisons, we expect relations with catches in other gears to

improve.

The recovery of the chain pickerel population in Oneida Lake is likely associated with the

arrival of dreissenid mussels. Zebra mussels Dreissena polymorpha were first identified

in Oneida Lake in 1991 and by 1992 peaked at 30,000/m2 (Hetherington et al. 2019). In

2005, the presence of quagga mussels Dreissena rostriformis bugensis was

documented and by 2010 replaced zebra mussels as the most abundant dreissenid in

the lake (Hetherington et al. 2019). Dreissenid mussels can increase submerged

aquatic vegetation (sav) by relocating nutrients from the water column to the sediments

through their feces and pseudofeces, and by removing suspended particulate matter

from the water column through grazing thereby increasing light penetration (Fitzgerald et

al. 2016). Increased growth of sav should increase suitable habitat for littoral fishes

(Irwin et al. 2016) like esocids. Submerged aquatic vegetation surveys indicated

coverage doubled in the shallower west end of Oneida Lake from 1990 to 2002 and

maximum depth of sav almost doubled lake-wide (Fitzgerald et al. 2016). The timing of

the chain pickerel increase relative to the arrival of dreissenid mussels is consistent with

our expectations. Both the expansion of sav in the 1990s (Fitzgerald et al. 2016) and

subsequent increase in age-1 and older chain pickerel abundance likely operate on a

timescale of approximately a decade or longer. Chain pickerel mature at age 4 or 5

(Smith 1985) and probably require a generation or two for population expansion, so the

7-8 year lag from when dreissenid mussels first arrived in 1991 to when chain pickerel

first started appearing in fish surveys seems reasonable.

The draining of marshes may have played a more important role in the chain pickerel

collapse than did lake level control. While completion of the Caughdenoy Dam in 1910

likely impacted lake water levels to some extent, the stabilization of lake levels was not

attainable until the 1950s when gates were installed in the Caughdenoy Dam (Schneider

et al. 2016). The chain pickerel population had already decreased by then (Forney

73

1980). The expansion of sav to deeper waters since dreissenid mussel arrival probably

protected areas of sav from winter draw-down thereby increasing chain pickerel

spawning and nursery habitat to a level allowing population recovery.

The chain pickerel population has increased in recent decades since their collapse in the

first half of the 20th century. The increase is likely due to expansion of suitable spawning

and nursery habitat associated with increased water clarity. As a result, this esocid now

offers angling opportunities that were not available for decades after spawning habitat

was lost. Chain pickerel diets were recently described (Jackson et al. 2019), and length

at age data collection is ongoing. The preferred aging structure for chain pickerel is the

otolith (Bauerlien et al. 2018) and most of our chain pickerel are caught in non-lethal

gear so accumulation of age data is slow. Cornell Field Station staff will continue to

monitor the chain pickerel population as well as angler catch rates to further assess the

status of this once abundant species and its effect on the Oneida Lake fish community

and fishery.

Literature cited

Adams, C. C. and T. L. Hankinson. 1916. Notes on Oneida Lake fish and fisheries.

Transactions of the American Fisheries Society 45:154-169.

Adams, C. C. and T. L. Hankinson. 1928. The ecology and economics of Oneida Lake

fish. Roosevelt Wild Life Annals 1:235-548.

Bauerlien, C. J., M. R. Cornett, E. A. Zielonka, D. P. Crane, and J. S. Bulak. 2018.

Precision of calcified structures used for estimating age of chain pickerel. North

American Journal of Fisheries Management 38:930-939.

Fitzgerald, D. G., B. Zhu, E. L. Mills, L. G. Rudstam, S. B. Hoskins, D. E. Haddad, N. R.

Burtch, J. T. H. Coleman, and D. L. Crabtree. 2016. Dynamics of aquatic vegetation in

Oneida Lake, 1915-2005. Pages 181-200 in Rudstam et al., editors. Oneida Lake: long-

term dynamics of a managed ecosystem and its fishery. American Fisheries Society,

Bethesda, Maryland.

Forney, J. L. 1980. Evolution of a management strategy for the walleye in Oneida Lake,

New York. New York Fish and Game Journal 27:105-141.

Forney, J. L., L. G. Rudstam, and D. L. Stang. 1994. Percid sampling manual. New

York State Department of Environmental Conservation, Albany, New York.

Grosslein, M. D. 1958. Creel census on Oneida Lake. New York Federal Aid in Sport Fish Restoration, Project No. F-17-R-2, Job 1. Albany, New York.

Grosslein, M. D. 1961. Estimation of angler harvest on Oneida Lake, New York. Doctoral dissertation. Cornell University, Ithaca, New York.

Hetherington, A. L., L. G. Rudstam, R. L. Schneider, K. T. Holeck, C. W. Hotaling, J. E.

Cooper, and J. R. Jackson. 2019. Invader invaded: population dynamics of zebra

74

mussels (Dreissena polymorpha) and quagga mussels (Dreissena rostriformis bugensis)

in polymictic Oneida Lake, NY, USA (1992–2013). Biological Invasions 21:1529-1544.

Irwin, B. J., L. G. Rudstam, J. R. Jackson, A. J. VanDeValk, and J. L. Forney. 2016.

Long-term trends in the fish community of Oneida Lake. Pages 375-396 in Rudstam et

al., editors. Oneida Lake: long-term dynamics of a managed ecosystem and its fishery.

American Fisheries Society, Bethesda, Maryland.

Jackson, J. R., A. J. VanDeValk, T. E. Brooking, K. T. Holeck, C. Hotaling, and L. G. Rudstam. 2019. The fisheries and limnology of Oneida Lake 2018. New York Federal Aid in Sport Fish Restoration, Job 2-2, Study 2 F-63-R. Albany, New York.

Pope, K. L., R. M. Neumann, and S. D. Bryan. 2009. Warmwater fish in small standing waters. Pages 13-28 in Bonar et al., editors. Standard methods for sampling North American freshwater fishes. American Fisheries Society, Bethesda, Maryland.

Schneider, R. L., M. S. Mayer, and T. C. Hall. 2016. Watershed, climate, and lake level

manipulations. Pages 53-70 in Rudstam et al., editors. Oneida Lake: long-term

dynamics of a managed ecosystem and its fishery. American Fisheries Society,

Bethesda, Maryland.

Smith, C. L. 1985. The inland fishes of New York State. New York State Department of

Environmental Conservation, Albany, NY.

Underhill, A. H. 1949. Studies on the development, growth and maturity of chain

pickerel, Esox niger Lesueur. The Journal of Wildlife Management 13:377-391.

75

Appendix 2:

Summary of Underwater Video Surveys of

Round Goby Abundance in Oneida Lake,

2017-2019

Thomas E. Brooking, James R. Jackson, Allyson S. Jones, Tia A.

Offner, Olive Onyekwelu, Amy Li, Anthony J. VanDeValk, and Lars G.

Rudstam

April 2020

Cornell University

76

Abstract Round gobies were surveyed using underwater video in 2017-2019. Mean round goby densities lake wide were estimated at 4,000, 3,200, and 12,900 fish/ac (1.0, 0.8, and 3.2 fish/m2) from 2017-2019, respectively. The 2017 estimate is an extrapolation to the whole lake from a more limited sampling area and therefore less comparable with the 2018 and 2019 estimates. Variation was high, with gobies only found in 15-33% of sites sampled. Round goby densities in rock/gravel/coble substrate were significantly higher than in mud/sand/silt substrate in all 3 years. We were not able to detect any significant differences attributable to depth or plant coverage. For areas with dense plant coverage, this may well result from visibility issues rather than reflecting habitat use. Round goby densities were significantly higher at sites with higher mussel coverage. Densities estimated from the video surveys were 18-41 times higher than densities estimated by beach seines, and 490-2,025 times higher than densities estimated with bottom trawls done at the same time of year. However, relative abundance between 5 different sampling methods (including the video) was in general agreement between years. All 5 methods indicated a decline in round goby abundance from 2017 to 2018 and all 5 gears indicated an increase in abundance from 2018 to 2019, although the magnitude of the changes were sometimes different.

Introduction

Round gobies (Neogobius melanostomus) were reported by anglers in Oneida Lake in

2013 but first appeared in sampling gear in 2014. In 2015 they were common in bottom

trawl samples, and by 2016 had become abundant in several different sampling gears.

While these gears provided an index of round goby relative abundance, a reliable

estimate of actual density was lacking. Standard sampling gears were not effective in

sampling round gobies in their preferred habitat (rocky substrates), some surveys lacked

lake wide spatial and depth coverage, and avoidance or attraction of gobies was

suspected to bias estimates. In order to directly relate round goby abundance to impacts

on mussels, benthic invertebrates and fish, a density estimate was needed. Beginning in

2017, we initiated underwater camera surveys for round goby in Oneida Lake in order to

obtain more reliable density estimates. Cornell University summer interns funded

through the NYSDEC Federal Aid in Sport Fish Restoration Act funding, the John and

Janet Forney Intern Endowment, and the Doris Duke Foundation performed these

studies under the guidance of Cornell Biological Field Station researchers. While the

inaugural study in 2017 resulted in a summary report of findings and a Senior Honors

Thesis (Jones 2017; 2019), the expanded lake wide studies in 2018 and 2019 did not

and are summarized here.

Methods

We used an underwater camera mounted on a raised frame to count the number of

gobies in a known area of lake bottom. We used GoPro Hero 4 and Hero 5 cameras to

record a video of each site. Camera settings were critical for achieving sufficient contrast

and light in deeper areas of the lake. The settings used in 2018 and 2019 were 2.7k

resolution, 60 f/s frame rate, and 6400 ISO sensitivity setting. Videos were collected for

7.5-10 minutes at each site. It was necessary to download the camera files onto a laptop

77

or portable drive in the boat as the camera internal hard drive would reach capacity.

Extra fully charged batteries were necessary as well and sometimes had to be changed

in between sites.

The camera was mounted on a ½” tubular PVC frame with 4 legs designed for the

camera face to be 2.8 ft (0.85 m) from the bottom (in 2017), facing straight down. In

2017 the frame was white and included a square bottom frame measuring 1.7 ft X 1.7 ft

= 2.9 ft2 (0.53 m X 0.52 m = 0.28 m2). Fish were also counted in the entire viewing area

which was 3.5 ft X 2.7 ft = 9.5 ft2 (1.07 m X 0.82 m = 0.88 m2). In 2018 and 2019, the

bottom square frame was removed leaving just 4 legs touching the bottom and the PVC

was painted black, to decrease avoidance or attraction by gobies (Figure 1). The camera

was lowered to 2.0 ft (0.60 m) from bottom to improve visibility and the viewing area in

2018-2019 was 5.8 ft2 (0.54 m2). A measuring bar of ¼ in metal was placed on the

bottom of one leg in the field of view and marked every 0.4 in (10 mm) to allow for

estimation of fish length. In 2018 we collected side-looking videos by attaching a second

camera on the frame to quantify avoidance or attraction.

In 2017, 38 sites were sampled resulting in 33 sites with useable data, and all sites were

within 3.7 mi (6 km) of Shackelton Point. A concurrent project in 2017 looked at density

of gobies in relation to zebra mussel abundance (Li 2017). In 2018 and 2019, we

78

sampled 100 sites (same sites used both years) which resulted in 98 useable videos in

both years. Sites were randomly chosen a priori and were stratified, both spatially and by

bottom substrate. Spatial stratification was done by dividing the lake into 7 geographic

areas with approximately equal coverage (14-15 sites) in each area (2018 and 2019

only). Bottom substrate was stratified based on the bottom substrate map by Greeson

(1971) with the following number of randomly selected sites in each quadrant: cobble –

2, gravel – 1, sand – 3, and mud/silt/clay – 8 (Figures 2 and 3).

Figure 2. Map used in 2018-2019 to distribute sites in 7 areas of the lake, and 4 bottom classifications. Small grid blocks were numbered, and sites chosen using a random number generator. Site selection was double stratified by substrate and depth.

79

Bottom substrate was determined using Ekman grabs in 2017 and from visual analysis

of the video in 2018-2019 and was categorized as soft bottom (sand/silt/mud), or hard

bottom (rock/cobble/gravel) in analyses. Table 1 indicates the percentage of each site

falling within those categories, with comparisons to 2017. Actual depth was recorded

from the boat depth finder.

Figure 3. Spatial distribution of sites used in 2017 (top panel), and in 2018 and 2019 (bottom panel).

2017

2018-2019

80

Table 1. Collection methods and site characteristics, 2017-2019 camera surveys.

Year 2017 2018 2019

Mean collection date (range) 6/29 (6/22-7/12) 7/2 (6/18-7/12) 7/8 (7/2-7/15)

Camera height from bottom 2.8 ft (0.85 m) 2.0 ft (0.60 m) 2.0 ft (0.60 m)

Bottom frame white PVC bottom frame no frame no frame

Frame size 1.7 ft x 1.7 ft no frame no frame

Video area 2.9 ft2 5.8 ft2 5.8 ft2

# sites 38 sites (2 vid per site) 100 100

# useable sites 33 sites (2 vid per site) 98 98

geographic area Fisher Bay to Lakeport whole lake whole lake

% shallow 0-13 ft (0-4 m) 52 30 30

% medium 13-26 ft (4-8 m) 30 36 36

% deep >26 ft (8 m) 18 35 35

Recorded videos were analyzed later by projection onto a large TV or computer screen.

In 2017, gobies were counted by taking 7 screen shots obtained from 4 to 10 minutes

after deployment and averaged for each video, then the mean of the two videos was

used for the average of that site. Additionally, a subset of 8 sites with 15 or more gobies

was chosen in 2017 and counted every minute from 2-10 minutes to analyze

avoidance/attraction, comparing the number of round gobies inside vs. outside the

bottom frame. In 2018-19, a screenshot was counted at the 4, 5, 6, and 7-minute marks

and averaged to calculate the mean count for each site. In all 3 years, the video footage

was used to assist identification of gobies using motion, as they were often difficult to

detect from the still frame image. Bottom substrate, percent vegetation, and percent

mussel coverage were visually estimated for each site from the video and still frame.

Additionally, in 2017 and 2018 length measurements were taken on gobies using image

measurement software from the still images.

Mean round goby counts were divided by the viewing area to calculate the round goby

density at each site. In 2017, lake wide density was estimated using the proportion of the

bottom in different depth-substrate combinations. In 2018-2019, lake wide density was

estimated using the average of all sites, which assumed the site selection method

accurately represented the proportion of bottom substrates in the lake. Additionally, in

2018-2019 we compared round goby density by depth, bottom substrate classification,

percent plant coverage, and percent mussel coverage.

Results

2017 Study

Round goby density in 2017 at 33 sites around Shackelton Point ranged from 0-27,500

fish/ac (0-6.8 fish/m2) depending on substrate type (Jones 2019). Gobies were only

found at 11 of 33 sites (33%). The highest density of gobies was in shallow rock

81

substrates, with less in sand substrates and zero in mud substrates. When expanded to

the whole lake based on percentage of each substrate type present (Fitzgerald et al.

2016), the mean lake wide density was estimated to be 4,000 fish/ac (1.0 fish/m2), with a

biomass of 33 lb/ac (3.7 g/m2; Table 2). Bottom trawling and beach seining (Jackson et

al. 2019) estimated round goby density was 8.1 and 240 fish/ac (0.002 and 0.06 fish/m2),

respectively, or only 0.2% and 5.6%, respectively, of the lake wide densities estimated

from the video analysis. Comparison of size structure of round gobies indicated that

video under-represented small gobies compared to the trawl.

Table 2. Results of camera survey in 2017 (Jones 2019). All densities are in fish/m2 and

all biomass are in g/m2. SE= standard error, N = number of sites.

Depth Rock Sand Mud/Silt 2017 Trawl

2017 Seine

Density SE N Density SE N Density SE N Shallow (0-

4m) 6.8 7.5 3 2.8 1.8 12 0 0 2

Medium (4-7m) - - - 1.2 0.8 6 0 0 4

Deep (7m-) - - - 0 - 1 0 0 5

Biomass SE N Biomass SE N Biomass SE N Shallow (0-

4m) 24.8 11.6 3 10.1 1.3 12 0 0 2

Medium (4-7m) - - - 6.7 1.8 6 0 0 4

Deep (7m-) - - - 0 - 1 0 0 5

All substrates All

sites July

Density Biomass Density Density Whole lake 1.0 3.7 0.002 0.06

Jones (2019) found the densities inside and outside the frame were correlated (r2=0.64,

P<0.001) and there were no significant differences between density estimates inside and

outside the frame (two tailed paired t-test, P= 0.62). The proportion of maximum

abundance increased with time inside the frame (linear regression, r2=0.65, P=0.009;

Figure 4) but not outside the frame (r2=0.006, P=0.84). Jones also checked for parallax

problems with measurement outside the square, and there is quite a marked effect the

further out from the square the measurements are taken. Within the square frame the

difference was approximately 10%.

82

In 2017, a concurrent project analyzed goby abundance compared to mussel and

invertebrate abundance (Li 2017). They found that mussel abundance had not changed

at that time, but mean size of mussels was larger in sites with round goby, and that total

benthic macroinvertebrate counts declined as goby density increased.

2018 and 2019 Study

In 2018 and 2019 at 98 useable sites stratified across the lake, round gobies were found

at 15 sites in 2018 (15%) and at 32 sites in 2019 (33%). The average of all sites was

used to expand to the whole lake based on the assumption that substrates and depths

were accurately represented by all sites. The mean lake wide density was estimated to

be 3,200 fish/ac and 12,900 fish/ac (0.8 and 3.2 fish/m2) in 2018 and 2019, respectively.

The 2019 density was significantly higher than in 2018 (Wilcoxon signed rank test,

p=0.002). The densities in rock/cobble/gravel substrate increased to 25,500 and 32,400

fish/ac (6.3 and 8.0 fish/m2), respectively. In mud/sand/silt substrate, densities were

lower at 1,200 and 3,200 fish/ac (0.3 and 0.8 fish/m2) in 2018 and 2019, respectively

(Table 3; Figure 5). Mean density estimated from bottom trawling (Jackson et al. 2019)

was 0.02-0.08% of the video estimate, and mean density estimated from beach seining

in July was 2.5-4.7% of the video estimate (Figure 6).

Figure 4. Linear regression analysis comparing mean proportion of maximum abundance of round gobies against time, inside the bottom frame (left panel) and outside the frame (right panel).

y = 4.8557x + 19.322R² = 0.65p=0.0001

0

10

20

30

40

50

60

70

80

1 3 5 7 9

Pro

port

ion o

f M

axim

um

Abundance

Time (minutes)

Inside frame vs. time

y = -0.2374x + 43.977R² = 0.006

p=0.84

0

10

20

30

40

50

60

70

1 2 3 4 5 6 7 8 9 10

Pro

port

ion o

f M

axiu

mu A

bundance

Time (minutes)

Outside frame vs. time

83

Table 3. Results of camera survey in 2018-2019. Beach seine is from 9 seine hauls in

July, bottom trawl is average of first 2 samples in July (10 sites each). Densities are in

fish/m2. SE= standard error, N = # of sites.

2018 Video 2018 Trawl

2018 Seine 2019

2019 Trawl

2019 Seine

Substrate Density SE N Density Density Density SE N Density Density

Mud/Sand/Silt 0.3 0.1 85 0.8 0.3 65

Rock/Gravel/Cobble 6.3 1.5 7 8.0 1.7 33

Whole lake 0.8 0.3 98 0.0004 0.02 3.2 0.7 98 0.003 0.15

Round goby density in 2018 and 2019 was compared across depths, mussel coverage,

and plant coverage (Table 4). Though the highest densities of round goby were

observed at 23 ft (7 m) and shallower both years, no significant differences across depth

were found (Figure 7; Wilcoxon signed rank test, p = 0.12 in 2018, p = 0.10 in 2019).

When compared to mussel coverage at each site, there were several significant

increases in round goby density as the percent coverage of mussels increased. In 2018,

21-40% was significantly higher than 0-20%, 41-60% was significantly higher than 0-

20%, but no other pairs were different (Wilcoxon signed rank test comparing all pairs;

Figure 8). In 2019, 21-40%, 61-80% and 81-100% were significantly higher than 0-20%

Figure 5. Overall round goby density

(average of all sites) in 2018 and 2019,

with standard error bars.

0

1

2

3

4

5

2018 2019

De

nsi

ty (

fish

/m2)

Year

Round goby density Error bars are one standard error

2019 significantly higher(Wilcoxin test, p=0.002)

Figure 6. Densities (fish/m2) of round gobies from three

gears. Video is average of all useable sites (98), seine is

from 9 beach seine hauls in July, and bottom trawl is the

average of the first two July samples (10 sites each).

0.81

3.22

0.02 0.150.0004 0.0025

0

0.5

1

1.5

2

2.5

3

3.5

2018 2019

Den

sity

(fi

sh/m

2)

Year

Round goby density from 3 gears

Video

Seine

Trawl

84

(all p<0.02); 81-100% was higher than 41-60% (p=0.03). When round goby density was

compared to plant coverage, no significant trends were found (Figure 9).

Table 4. Round goby density by site depth, mussel coverage, and plant coverage in

2018 and 2019. All densities are in fish/m2. SE= standard error, N = number of sites.

2018 2019

Site depth, ft Density SE n Density SE n

3.3 (1 m) 0 0 3 0 0 3

6.6 (2 m) 3.0 1.4 10 2.6 1.5 10

9.9 (3 m) 0.2 0.2 6 0.4 0.4 6

13.3 (4 m) 1.8 1.1 13 6.1 2.1 13

16.4 (5 m) 0.04 0.04 12 8.7 3.9 12

19.7 (6 m) 1.0 1.0 9 3.7 2.9 9

23.0 (7 m) 0.7 0.7 5 5.9 3.0 5

26.2 (8 m) 0 0 8 0.7 0.7 8

29.5 (9 m) 0 0 6 1.5 1.5 6

32.8 (10 m) 1.2 1.1 9 1.8 1.2 9

36.1 (11 m) 0 0 9 0.5 0.5 9

39.4 (12 m) 0 0 6 0 0 6

42.6 (13 m) 0 0 2 6.3 6.3 2

Mussel Coverage Density SE n Density SE n

0-20% 0.4 0.2 65 0.9 0.3 63

21-40% 1.6 0.9 14 4.2 1.5 13

41-60% 2.0 1.5 9 4.0 2.8 7

61-80% 1.6 1.2 8 8.1 2.8 12

81-100% 0 0 2 26.5 7.2 3

Plant Coverage Density SE n Density SE n

0-25% 0.9 0. 90 3.2 0.8 80

26-50% 0 0 2 5.7 3.1 5

51-75% - - 0 4.4 2.7 7

76-100% 0 0 6 0.5 0.5 6

85

Figure 7. Round goby density by depth, with standard error bars. Differences were not significant (Wilcoxon signed rank test; 2018 p=0.12, 2019 p=0.10).

-1

1

3

5

7

9

11

13

15

1 2 3 4 5 6 7 8 9 10 11 12 13

De

nsi

ty (

fish

/m2)

Depth (m)

Round goby density vs. depthError bars are one standard error

2018

2019

Figure 8. Round goby density in different categories of mussel coverage. Mussel coverage was visually estimated from videos and still images at each site.

0

5

10

15

20

25

30

35

0-20% 21-40% 41-60% 61-80% 81-100%

De

nsi

ty (

fish

/m2)

Mussel coverage (%)

Round goby density vs. mussel coverageError bars are one standard error

2018

2019

86

Biomass of round goby in Oneida Lake was estimated at 33 lb/ac (37 kg/ha) in 2017

(Jones 2019). In 2018 and 2019, biomass was estimated using the mean length of

gobies in the July beach seine and first two July bottom trawl surveys (2.4 in or 61.0 mm

in 2018, 2.3 in or 58.6 mm in 2019) resulting in mean weight of 0.11 oz (3.0 g) in 2018,

and 0.09 oz (2.63 g) in 2019. Resulting biomass estimates using the density from the

video surveys were 21 lb/ac (24 kg/ha) in 2018 and 75 lb/ac (84 kg/ha) in 2019.

Conclusions and Discussion

Mean round goby densities lake wide were estimated at 3,200 -12,900 fish/ac (0.8-3.2

fish/m2) from 2017-2019. The 2017 lake-wide estimate of 4,000 fish/ac (1.0 fish/m2) is

less comparable to the 2018 and 2019 estimates due to limited spatial and substrate

stratification. Round goby densities in rock/gravel/coble substrate were significantly

higher than in mud/sand/silt substrate in all 3 years. We were not able to detect any

significant differences attributable to depth or plant coverage. For areas with dense plant

coverage, this may well result from visibility issues rather than reflecting habitat use.

Round goby densities were significantly higher at sites with higher mussel coverage.

Densities estimated from the video surveys were much higher than densities estimated

by beach seines and bottom trawls. The video estimates were 18-41 times higher than

beach seine estimates, and 490-2,025 times higher than bottom trawl estimates done at

the same time of year. Beach seining sites are inshore in <5 ft (1.5 m) deep with many

Figure 9. Round goby density by % plant coverage (no sites fell into 51-75% class). No significant differences were found using parametric or nonparametric tests.

0 00

1

2

3

4

5

6

7

8

9

10

0-25% 26-50% 51-75% 76-100%

Go

by

de

nsi

ty (

fish

/m2)

Plant coverage (%)

Goby density by plant coverageError bars are one standard error

2018

2019

No

sit

es

87

cobble and gravel sites, whereas trawl sites are offshore in >18 ft (5.5 m) deep on mostly

mud and silt bottoms, which may account for some of the discrepancies.

Gobies were only found at 15-33% of video sites sampled, compared to 100% of beach

seine sites, and 40-65% of trawl sites. Area sampled annually by the July video survey

(98 sites) was 0.013 ac (53 m2), compared to 1.74 ac (7,042 m2) in the 9-site July beach

seine survey, and 2.47 ac (10,000 m2) in the 10-site July trawl survey.

Andres et al. (2020) used a similar video survey in Cayuga Lake, NY in July-August

2017 which resulted in an unadjusted round goby density of 17,900 fish/ac (4.43 fish/m2)

during summer in <60 ft (18.4 m) of water. After adjustment for detection probability and

effective sampling area (due to strong attraction of gobies to the camera structure), they

estimated the true density at 7,400 fish/ac (1.82 fish/m2), or 41% of the video-observed

density. If the same over-estimation applies to our video survey, densities would have

been 1,300-5,300 fish/ac (0.33-1.32 fish/m2), which are still 7-17 times higher than the

beach seine estimate, and 200-825 times higher than the bottom trawl estimate. The

Oneida Lake adjusted densities are about 18-73% of the density found in Cayuga Lake.

Unadjusted biomass of round goby in Oneida Lake ranged from 21-75 lb/ac (24-84

kg/ha) in 2017-2019. Summer round goby biomass (adjusted) in Cayuga Lake was 109

lb/ac (122 kg/ha) in 2017, though that was in <60 ft (18.4 m) of water (Andres et al.

2020). Densities reported from Lake Ontario were 36 lb/ac (40 kg/ha; Pennuto et al.

2012) and 45-100 lb/ac (50-112 kg/ha; Taraborreli et al. 2010). In the western basin of

Lake Erie, Johnson et al. (2005) reported round goby density of 64 lb/ac (72 kg/ha).

Goby biomass in Oneida Lake was in the lower range of estimates found elsewhere and

may still have some room for expansion, based on biomass found in other lakes.

Although results from other sampling gears provided very different estimates of round

goby absolute density, there was general agreement between different gears as indices

of abundance. Figure 10 shows the round goby catch in 5 different sampling methods

including the video survey, equalized by expressing each catch as the percentage of the

maximum catch in that sampling method. All 5 sampling methods indicated a decline in

round goby abundance from 2017-2018 and all indicated an increase in abundance from

2018-2019, although the magnitude of the changes were sometimes different. Although

other methods appear to be able to provide indices of changes in abundance, at present

only videos appear to produce reliable density estimates, though these may be biased

high due to attraction of round gobies to the camera structure. Differences between

sampling methods are due both to habitat accessibility of traditional gears (rocky

bottoms are difficult to sample with traditional gear), and low catchability of round goby in

traditional gear as the estimates of round goby from video are higher even in the same

habitat.

88

Acknowledgements

The authors would like to thank the agencies who provided intern and staff funding

including the NYSDEC Federal Aid in Sport Fish Restoration Fund, the John and Janet

Forney Intern Endowment, and the Doris Duke Foundation. Many other interns from the

Intern Classes of 2017, 2018, and 2019 assisted interns and staff with completing these

studies.

Literature Cited

Andres, K. J. Andres, S. A. Sethi, E. Duskey, J. M. Lepak, A. N. Rice, B. J. Estabrook, K.

B. Fitzpatrick, E. George, B. Marcy‐Quay, M. R. Paufve, K. Perkins, and A. E.

Scofield. 2020. Seasonal habitat use indicates that depth may mediate the

potential for invasive round goby impacts in inland lakes. Freshwater Biology.

2020 00:1–11.

Fitzgerald, D. G., B. Zhu, L. G. Rudstam, S. B. Hoskins, D. E. Haddad, N. R. Burtch, J.

T. Coleman, D. L. Crabtree, and E. L. Mills. 2016. Dynamics of aquatic

Figure 10. Round goby catches in 5 different survey gears including the video survey, equalized

by expressing each catch as the percentage of the maximum catch in that gear. For gear

descriptions, see Jackson et al. (2019).

0

20

40

60

80

100

120

2012 2013 2014 2015 2016 2017 2018 2019

% o

f m

axim

um

cat

ch in

th

at g

ear

Year

Goby relative abundanceexpressed as % of maximum catch in each gear

Fyke Net (small mesh only, Sept.)

Seine (avg of 3 surveys, July-Sept)

GN diet (avg of WP, YP, June-Sept)

Trawl (July-Oct)

Annual Average

Video survey (July, 33-100 sites)

89

vegetation in Oneida Lake, 1915-2005: A response to ecosystem change. Pages

181-199 in L. G. Rudstam, E.L. Mills, J. R. Jackson, and D. J. Stewart, editors.

Oneida Lake: Long-term dynamics of a managed ecosystem and its fisheries.

American Fisheries Society, Bethesda, Maryland.

Jackson, J. R., A. J. VanDeValk, T. E. Brooking, K.T. Holeck, C. Hotaling, and L. G.

Rudstam. 2019. The Fisheries and Limnology of Oneida Lake 2018. Cornell

University Biological Field Station, Bridgeport, NY. 124 pp.

Johnson, T. B., M. Allen, L. D. Corkum, and V. A. Lee. 2005. Comparison of methods

needed to estimate population size of round gobies (Neogobius melanostomus)

in western Lake Erie. Journal of Great Lakes Research 31:78–86.

Jones, A. S. 2019. Surveying Neogobius melanostomus populations within Oneida Lake

with underwater video. Honors Thesis, Cornell University, Environmental Science

and Sustainability. Cornell University Biological Field Station, Bridgeport, NY. 23

pp.

Jones, A. S. 2017. Surveying Neogobius melanostomus populations within Oneida Lake

with underwater video. Intern Project 2017. Cornell University Biological Field

Station, Bridgeport, NY. 15 pp.

Li, A. 2018. Impact of round goby invasion on benthic macroinvertebrate populations in

Oneida Lake. Intern report 2017. Cornell University Biological Field Station,

Bridgeport, NY. 22 pp.

Pennuto, C. M., E. T. Howell, and J. C. Makarewicz. 2012. Relationships among round

gobies, Dreissena mussels, and benthic algae in the south nearshore of Lake

Ontario. Journal of Great Lakes Research 38:154–160.

Taraborelli, A. C., M. G. Fox, T. B. Johnson, and T. Schaner. 2010. Round goby

(Neogobius melanostomus) population structure, biomass, prey consumption and

mortality from predation in the Bay of Quinte, Lake Ontario. Journal of Great

Lakes Research 36:625–632.

90

Appendix 3: Data collection methods

Limnology. Zooplankton samples are collected weekly (May-October) from 1-5 sites with a 153 um mesh nylon net (0.5 m diameter) using a vertical tow from 0.5 m above the bottom to the water surface. Samples are preserved in 70% ethyl alcohol (8% sugar-formalin solution 1975-1996). Zooplankton are identified, counted, and measured (to the nearest µm) using a digitizing tablet and microscope (1998 – present). Previous methods include use of a dissecting microscope and calipers (1975-1982), and a touch screen setup with computer-assisted plankton analysis system WSAM (1983-1997) (Hambright and Fridman 1994). Mean May-October biomass is calculated from weekly averages using the length – weight regressions in Watkins et al. (2011). These values are in dry weight. Integrated water samples for total phosphorous (TP) and soluble reactive phosphorous (SRP) are collected using a 1.9 cm inside diameter Nalgene tube lowered to a depth of 1 meter above bottom, and frozen for later analysis. In the lab, a 50 mL aliquot of unfiltered water is analyzed for TP using the persulfate digestion method (Menzel and Corwin 1965). For SRP, lake water is filtered through a Whatman 934-AH glass fiber filter and a 50 mL aliquot is analyzed using the molybdate method of Strickland and Parsons (1972). For chlorophyll-a measurements, lake water (up to 2.0 L) is filtered through Whatman 934-AH glass fiber filters and the filters are assayed using the acetone extraction method (Strickland and Parsons 1972). Annual averages are calculated as the average of weekly values collected at 1 to 5 stations from May to October. All 5 stations are included when available, except for Secchi depth from the shallow station (Three Mile Bay) because the Secchi disk is sometimes observed on the bottom. Beginning in 2010, one site (Buoy 117) was dropped from standard sampling and a new sampling protocol for water chemistry was developed. Four sites were sampled each week, and on week 1 water samples were processed by individual stations as in 1975-2009. On weeks 2-4, water from all four sites was pooled for analysis. This rotation was maintained throughout the sampling season. Samples were pooled for water chemistry only, not zooplankton. Beginning in 2009, nutrients samples were analyzed at the Upstate Freshwater Institute (UFI). This EPA approved laboratory uses SM 18-20 4500-P E for TP and SRP, and SM 20 4500-SiO2 C for SRS (APHA). Beginning in 2013, chlorophyll were analyzed with a Turner Design fluorometer after extractions following the EPA standard operating procedure LG405 with the exception that all samples are run during the winter after the completion of the field season.

APHA. 1998. Standard Methods for the Examination of Water and Wastewater. 20th edition.

EPA LG405 Standard Operating Procedure for In Vitro Determination of Chlorophyll a in Freshwater Phytoplankton by Fluorescence

91

Larval fish surveys: Miller high-speed sampler surveys are designed to estimate abundance of larval walleye and yellow perch. Larval walleye and yellow perch are sampled when yellow perch reach approximately 8 mm and again at approximately 18 mm. For each survey, the lake is divided into two or more horizontal and vertical depth strata and samples taken at a total of 46 randomly selected sites within designated strata. At each site, four Miller samplers are towed simultaneously at different depths and catches are pooled by stratum. Distance towed is about 1.6 km at a speed of 3.6 m/s. Larval fish captured are identified, counted, and measured. Density estimates are calculated for each strata based on catch and volume of water strained. Catches of yellow perch in the 18 mm survey are adjusted for size-specific gear avoidance (Noble 1970). Gill net surveys: Standard gill net catches provide an index of the adult walleye and yellow perch populations as well as relative abundance estimates of various other species. A variable mesh multifilament gill net is fished overnight at a different standard site each week for 15 consecutive weeks starting in the beginning of June and continuing through mid-September. The net consists of four gangs 45.75 m long by 1.83 m deep sewn together to form one 183 m long net. Each gang consists of six 7.6 m panels with 38, 51, 64, 76, 89 and 102 mm stretch mesh. The net is set around sunset, fished on the bottom, and retrieved in the morning at about 0730. The time fished varies somewhat with season but has been identical for each location each year. All fish (or a subsample of at least 60 individuals of a species) are measured (total length in mm), weighed (g), sexed, stomach contents recorded, and scales taken. Large mesh gill nets were used to monitor sturgeon reproductive status and abundance and growth in 4 different substrate types. Variable (152, 203, 254, and 305 mm stretch mesh) mesh monofilament gill nets 61 m in length were set for approximately 4 hours at 12 sites monthly in May and June. All sturgeon caught were examined for tags, measured, weighed, a fin ray section removed for age determination, diet recorded using gastric lavage, tagged with both a Carlin dangler tag and PIT tag, and released. Trawl surveys: The catch in trawls provides an estimate of year class abundance for young-of-year (age-0) and yearling walleye as well as prey species, primarily young yellow perch. Trawling begins around the middle of July when age-0 yellow perch become demersal (at about 1 g in weight) and weekly surveys continue until three October surveys are completed. A 5.5 m otter trawl is towed for 5 minutes, sampling approximately 0.10 ha per haul. Ten standard sites are sampled in each survey. Age-0 fish are identified, counted, total weight by species recorded to the nearest gram, and a subsample of fish measured for length. Lengths are recorded and scale samples taken on all older fish. A series of three trawl surveys at the same sites centered around May 1 is also conducted to assess age-1 walleye and yellow perch abundance. Hydroacoustic surveys: Pelagic fish biomass is estimated in the end of August–beginning of September using hydroacoustics. Surveys are conducted using a

92

123 kHz split beam unit (Biosonics DT-X, pulse length 0.4 ms, 7.8o beam width) along a set of approximately 8 transects from the east to the west ends of the lake. Surveys are typically conducted during two consecutive nights starting one hour after sunset. Acoustic data are analyzed with EchoView (v5.4 in 2016). Echograms are checked for problems associated with poor bottom detection, bubbles from waves, echoes from macrophytes, and other sources of noise. Questionable areas are removed from the analysis. Attempts are made to sample as close to the bottom as possible by re-defining the bottom at high magnification when needed. All densities are calculated from in situ backscattering cross section (average for targets larger than –60dB) and echo integration according to the standard operating procedure for Great Lakes acoustics (Parker-Stetter et al. 2009). Noise level at 16 m, the maximum depth in Oneida Lake is estimated to be –85 dB (uncompensated TS) thus satisfying a 15 dB signal to noise ratio throughout the water column for the smallest targets included in the analysis as recommended by Rudstam et al. (2009). Analyses are conducted using each transect as cluster of elementary sampling units of 500 m. Cluster analysis was used to estimate mean density and standard error using standard formulas (Scheaffer et al. 2006) and a program available on the web site “Acoustics Unpacked” (www.acousticsunpacked.org, Sullivan and Rudstam 2008). Fish are sampled in association with acoustic surveys using a midwater fry trawl and fine mesh gill nets. These gears are used to assess the species composition of young fish in the pelagic zone. The trawl measures 2 m x 2 m at the mouth and is mounted in a metal frame. The first 2 m of the net is comprised of 12.7 mm stretch mesh, the next 2 m of 6.4 mm stretch mesh, and the cod end of the net consists of a 0.5 m plankton net and bucket with 1 mm mesh. At each site, one haul divided into 2.5 minutes at 4.3 to 6.1 m depth and 2.5 minutes at 2 to 3.8 m depth (determined from rope angles) and a second 5 minute haul at the surface (sampling the top 2 m of the water column) are conducted. Two trawl hauls are completed at each of 10 sites, and fish are preserved in formalin and returned to the lab for species identification, enumeration, and measurement. Fine mesh gill nets, 21 m long, are set either on bottom or suspended from the surface. Each gill net consists of seven 3 m wide by 6 m deep panels of different mesh sizes (6.2, 8.0, 10.0, 12.5, 15.0, 18.7 and 25.0 mm bar mesh). Paired (1 surface and 1 bottom) gill nets are set at each of 4 deep stations, and 4 shallow stations are sampled with only 1 net that samples the entire water column. Acoustic density estimates are apportioned to emerald shiners, gizzard shad, and other fish based on catches in vertical gill nets and midwater trawls after accounting for the relative length selectivity and effort of the two gears. Fish in the top 2 m of the water column that are not surveyed with acoustics are accounted for by calculating the proportion of gizzard shad and emerald shiners caught in the top 2 m in vertical gillnets set compared to the rest of the water column. Fyke Net Surveys: We sample 18 sites around the lake representative of nearshore habitat types. Sites were selected to represent the common

93

substrates in the nearshore in the proportions they occur and distributed around both shores of the lake as evenly as possible while still achieving substrate representation. Each site is sampled via approximately 24 hour sets of a fyke net comprised of a 0.9 m x 1.5 m frame fitted with 12.7 mm (1/2”) delta knotless mesh. In 2008, we concurrently sampled 14 sites with a fyke net comprised of a 0.9 m x 1.5 m frame fitted with 5 mm (1/4”) delta knotless mesh From 2009 onwards, all 18 sites were sampled with nets of both mesh sizes. Sampling is typically conducted in September of each year. Annual abbreviated summer survey – methodology: Analyses of seasonal patterns in walleye catch and harvest rates estimated during roving creel surveys for the entire open water season indicated a good predictive relationship existed between rates observed in June and July and full open water season rates. The full open water season walleye catch rate can be predicted by the relationship:

CR = 0.700(JJCR) + 0.043 where CR is the catch rate predicted for the entire open water season and JJCR is the mean of June and July catch rates (r2 = 0.94; p = 0.0002). The open water walleye harvest rate was predicted by the relationship:

HR = 0.761(JJHR) + 0.012 where HR is the harvest rate predicted for the entire open water season and JJHR is the mean of June and July harvest rates (r2 = 0.85; p = 0.003). Beginning with the 2016 season, we also predict full open water season based on counts from a subset of months. We found that effort estimates from May, June and July predicted total season effort, accounting for 96% of the variability observed over 11 seasons of full effort data. The relationship was;

OWEFFORT = 31,302.9 + (1.2(ME+JnE+JuE))

where OWEFFORT is effort predicted for the entire open water season and ME, JnE and JuE are May, June and July effort, respectively (r2 = 0.96; p < 0.0001). Effort is estimated by fixed point boat counts conducted from a tower on the CBFS property. Counts are conducted at two random times on two randomly selected weekdays and both weekend days through the counting season and effort in boat hours calculated following the methods described in Krueger et al. (2009). Boat hours are converted to angler hours by multiplying boat hours by the average party size calculated from exit interviews in June and July. Exit interviews are conducted on two randomly selected weekdays and both weekend days during either a morning shift (0800-1400) or afternoon shift (1400-2000),

94

also randomly selected. Exit interviews are conducted at three boat launches, South Shore Boat Launch, Godfrey Point Boat Launch and Oneida Shores, and location for each day is randomly selected. Catch and harvest rates are calculated using the ratio of means following methods described by Krueger et al. (2009).

95

Appendix 4:

Standard Data Tables

96

Table A1. Physical, chemical and biological characteristics of Oneida Lake since 1975. Secchi depth (m), chlorophyll-a (µg/L), total phosphorous (TP, µg/L) soluble reactive phosphorous (SRP, µg/L), total zooplankton biomass (µg/L), and Daphnia spp. biomass (µg/L) are averages of weekly data from 1 to 5 stations from May to October. Ice freeze day (day since Dec 1), ice duration and ice out day (day of year) are noted at CBFS and refer to the year of ice break-up. The lake was not completely frozen over in the winter of 2001. Summer temperature (oC) is the average temperature from June to Aug measured every hour at 2 m depth at a site near Shackelton Point.

Year Secchi Chl-a SRP TP Zoopl.

Biomass Daphnia Biomass

First Freeze Day

Ice Duration

Ice Out Day Sum Temp

1975 2.8 9.0 17.6 45.9 211 107 no data no data 87 22.2 1976 2.8 9.9 3.3 29.5 241 163 19 99 87 20.6 1977 2.7 11.2 5.2 36.2 209 53 3 118 90 20.9 1978 2.9 7.7 16.5 42.5 116 73 15 121 105 22.0 1979 3.3 7.6 29.0 56.9 226 101 29 96 94 19.8 1980 2.6 12.7 10.2 45.2 257 126 35 91 95 20.4 1981 2.3 11.7 13.8 31.3 243 45 15 95 76 21.6 1982 2.2 9.0 15.2 48.0 260 93 20 118 107 20.8 1983 2.6 8.0 21.7 38.6 261 107 13 74 87 22.3 1984 2.3 9.2 14.7 30.1 231 104 21 111 101 21.6 1985 2.2 10.5 11.6 38.3 261 82 40 79 88 20.4 1986 2.4 10.3 27.5 67.1 304 178 19 104 92 20.4 1987 2.9 6.5 7.0 27.6 178 97 35 86 90 21.7 1988 2.7 9.5 16.6 34.6 248 99 34 91 94 20.8 1989 3.4 5.2 7.8 24.1 185 81 16 102 84 21.9 1990 2.4 9.5 4.8 22.0 221 65 5 107 81 21.7 1991 2.4 11.7 4.6 23.2 188 67 31 78 78 23.0 1992 2.8 7.1 1.8 20.1 315 196 25 93 102 20.2 1993 3.9 5.1 5.9 15.8 157 64 24 99 105 21.4 1994 3.7 6.6 6.2 30.4 193 103 27 113 109 22.0 1995 4.9 3.2 10.0 22.9 207 140 39 75 97 23.2 1996 3.6 5.5 6.0 19.9 222 128 32 100 101 22.0 1997 3.6 5.3 3.3 14.7 300 135 39 88 96 21.6 1998 3.0 5.3 5.3 21.5 161 57 48 58 86 22.5 1999 3.3 6.0 6.3 15.2 206 82 33 94 96 23.3 2000 2.9 6.5 4.4 18.1 154 85 45 63 77 21.3 2001 3.6 5.3 10.4 27.8 237 101 12 117 103 22.4 2002 3.7 4.8 7.0 27.2 162 75 no freeze 0 no freeze 23.0 2003 3.8 6.5 9.8 27.3 209 92 10 104 105 22.2 2004 3.4 7.7 10.8 29.0 233 99 21 90 95 21.5 2005 4.2 3.8 16.4 29.4 259 116 26 97 98 24.2 2006 3.1 7.3 10.6 29.2 209 77 18 72 91 22.9 2007 3.5 5.8 6.4 20.9 185 68 54 71 94 22.6 2008 4.2 3.8 12.2 24.6 165 45 19 83 92 22.5 2009 3.7 4.0 24.4 117 40 24 85 81 21.7 2010 4.4 2.6 11.1 28.5 139 48 23 85 81 23.4 2011 3.2 4.9 7.9 30.5 97 22 17 107 94 23.3 2012 4.7 2.9 15.3 31.5 183 63 46 21 53 24.1

97

Year Secchi Chl-a SRP TP Zoopl.

Biomass Daphnia Biomass

First Freeze Day

Ice Duration

Ice Out Day Sum Temp

2013 3.2 3.9 10.7 30.9 209 64 30 80 91 22.8 2014 3.3 4.2 11.1 28.0 176 43 30 118 103 22.3 2015 3.5 3.3 6.7 22.8 214 46 37 99 105 22.6 2016 4.3 3.7 7.8 24.5 167 42 45 74 57 23.4 2017 3.1 5.8 10.2 25.1 181 47 20 63 90 22.4 2018 3.5 4.5 4.9 15.8 152 38 27 98 94 23.3 2019 3.3 4.7 3.0 17.8 1961 441 42 86 99 22.8

Avg 3.3 6.7 10.2 29.2 205 84 27.0 88.3 92.0 22.1

1 Shackelton Point site only.

98

Table A2. Walleye age-specific density estimates in Oneida Lake since 1957 (in fish/ha). Age 1, 2 and 3 are estimated from the average of trawl and gill net estimates using catchabilities in Irwin et al. (2008). Bold values are from mark-recapture estimates. Densities of walleyes for intervening years were approximated from the distribution of mortality between successive population estimates (for 2017 and 2018 estimates annual mortality of 12% was used). Estimates from 1978-1987 and 1992-1994 from Irwin et al. (2008).

Year Age 1 Age 2 Age 3 Age 4 Age 5 Age 6 Age ≥ 7

Total (Age-≥4)

1957 no

data no

data no

data 0.4 6.22 0.97 4.62 12.21 1958 9.18 1.55 4.72 37.82 0.6 6.12 5.59 50.13 1959 0.60 12.23 3.72 2.69 34.12 0.27 9.47 46.54 1960 4.94 3.62 22.70 1.8 2.36 15.74 4.8 24.7 1961 27.87 18.72 4.76 20.82 2.45 2.14 14.9 40.31 1962 15.84 14.49 12.62 3.15 13.93 1.71 10.35 29.14 1963 24.58 16.31 17.92 13.26 2.56 15.32 7.61 38.75 1964 34.61 19.28 10.36 9.03 8.85 1.71 15.29 34.88 1965 43.68 19.15 12.78 8.69 5.53 7.18 10.53 31.93 1966 22.09 31.64 15.05 11.61 7.1 4.52 14.48 37.71 1967 3.99 19.27 29.23 10.29 8.17 3.77 10.42 32.66 1968 22.35 2.89 20.67 17.37 5.66 3.88 8.88 35.79 1969 93.66 31.11 4.87 12.74 13.83 4.65 9.44 40.65 1970 3.10 37.77 10.75 1.18 8.41 9.53 9.05 28.16 1971 4.07 0.53 8.00 9.53 1.01 5.48 12.1 28.12 1972 80.32 9.21 1.54 23.09 6.19 0.86 11.42 41.55 1973 0.65 43.68 4.58 1.41 12.63 3.63 7.17 24.84 1974 6.08 2.18 47.64 2.65 0.37 2.52 3.48 9.02 1975 1.56 3.68 1.08 29.91 2.6 0.36 5.88 38.76 1976 92.71 3.61 3.23 1.08 27.76 2.11 5.06 36.00 1977 0.70 55.05 2.56 1.92 0.49 15.08 3.9 21.39 1978 36.75 0.96 31.26 1.56 1.64 0.36 16.67 20.24 1979 3.35 30.20 1.04 22.17 1.24 1.23 11.27 35.91 1980 2.48 4.41 22.30 0.98 14.45 0.81 8.2 24.44 1981 39.70 4.52 5.71 21.39 0.64 9.45 5.87 37.35 1982 26.88 22.72 3.87 3.53 17.5 0.45 10.37 31.85 1983 14.32 33.19 30.66 2.58 2.88 12.55 7.12 25.13 1984 9.79 9.43 20.79 13.77 2.13 2.06 13.33 31.28 1985 10.23 6.85 14.32 26.89 11.68 1.55 10.27 50.4 1986 15.01 9.07 7.03 9.96 22.41 8.7 7.77 48.84 1987 3.09 13.46 6.31 7.89 8.05 16.75 11.13 43.82 1988 105.80 2.14 12.59 10.34 5.02 9.66 22.32 47.34 1989 3.88 50.90 2.80 8.16 7.58 7.68 13.29 36.71 1990 7.98 8.29 49.85 1.16 5.99 5.41 14.54 27.1

99

Year Age 1 Age 2 Age 3 Age 4 Age 5 Age 6 Age ≥ 7

Total (Age-≥4)

1991 12.22 7.31 3.91 19.75 2.78 4.84 13.15 40.52 1992 45.62 9.25 6.74 1.79 16.71 1.15 10.45 30.1 1993 3.55 26.66 2.87 1.8 1.42 11.57 7.57 22.36 1994 8.64 2.40 23.14 2.29 1.19 0.98 12.83 17.29 1995 5.81 5.31 1.14 5.65 1.56 0.65 6.22 14.08 1996 9.66 2.65 2.78 1.32 5.52 0.92 4.87 12.63 1997 3.67 4.82 3.01 1.57 1.2 5.39 4.00 12.16 1998 22.17 1.43 4.16 0.59 1.50 1.17 7.33 10.59 1999 13.65 7.68 1.88 1.54 0.57 1.44 6.86 10.42 2000 9.58 11.81 6.41 0.33 1.82 0.63 7.17 9.95 2001 7.26 12.47 5.99 3.9 0.38 2.15 6.9 13.33 2002 32.13 9.23 8.39 3.65 3.25 0.88 7.54 15.33 2003 10.87 14.43 3.65 2.78 3.85 2.71 8.32 17.66 2004 6.39 12.94 12.19 4.74 3.92 2.59 6.89 18.15 2005 8.52 1.59 4.65 6.15 4.97 5.53 6.05 22.71 2006 5.53 9.68 1.17 1.17 6.60 2.75 9.48 20.00 2007 9.24 6.67 4.41 1.27 1.03 7.09 9.28 18.67 2008 2.07 2.32 0.87 4.77 1.12 0.91 14.46 21.27 2009 0.11 5.51 2.67 3.32 4.21 0.99 13.58 22.11 2010 1.81 5.13 3.69 4.52 2.94 3.72 12.88 24.06 2011 8.37 2.73 6.95 2.60 3.38 2.20 12.40 20.57 2012 3.95 7.65 2.49 4.50 1.94 2.52 10.90 19.86 2013 5.54 6.43 9.71 2.25 3.36 1.45 9.93 16.99 2014 9.40 10.98 4.36 8.98 1.91 2.86 9.67 23.42 2015 11.43 10.32 5.64 3.29 8.25 1.63 10.65 24.36 2016 12.18 10.23 6.57 2.34 3.29 7.52 7.58 20.73 2017 32.96 13.53 10.33 8.54 2.06 2.90 13.29 26.78 2018 7.01 20.58 7.76 17.00 7.51 1.81 14.24 40.56 2019 11.29 3.52 16.76 13.71 14.96 6.61 13.05 48.33

100

Table A3. Fish observed in stomachs of yearling and older walleye taken by trawls and electrofishing during October and November in Oneida Lake since 1971, expressed as numbers per kg of walleye (YP-yellow perch; Morone-white perch or white bass; Gizz-gizzard shad; ES-emerald shiner; RG-round goby)..

Year #

examined %

empty YP Morone Gizz ES

RG Other UID Total

1971 240 37 3.92 0.01 0.00 0.00 0.00 0.06 1.59 5.58 1972 163 58 1.02 0.10 0.00 0.00 0.00 0.62 0.89 2.63 1973 295 32 0.69 1.36 0.00 0.00 0.00 0.43 1.35 3.83 1974 228 27 2.11 1.15 0.01 0.11 0.00 0.38 1.76 5.52 1975 204 68 0.20 0.13 0.00 0.02 0.00 0.17 0.24 0.76 1976 156 36 1.31 0.89 0.00 0.16 0.00 0.75 1.17 4.28 1977 70 19 3.14 1.25 0.00 0.00 0.00 0.14 0.89 5.42 1978 85 56 0.51 0.12 0.00 0.00 0.00 0.47 0.74 1.84 1981 88 66 1.52 0.16 0.00 0.00 0.00 0.00 0.56 2.24 1982 122 11 0.38 5.27 0.00 0.00 0.00 0.00 0.54 6.19 1983 117 62 0.19 0.79 0.00 0.00 0.00 0.00 0.30 1.28 1984 148 59 0.21 0.45 0.97 0.00 0.00 0.07 0.46 2.16 1985 151 50 1.60 0.04 0.36 0.00 0.00 0.13 0.44 2.57 1986 193 45 1.60 0.16 0.05 0.00 0.00 0.15 0.49 2.45 1987 194 23 0.05 0.64 1.96 0.00 0.00 0.02 0.54 3.21 1988 180 55 0.36 0.00 0.30 0.00 0.00 0.07 0.33 1.06 1989 193 26 0.00 0.18 5.42 0.00 0.00 0.03 0.83 6.46 1990 179 28 0.03 0.00 4.91 0.01 0.00 0.00 0.66 5.61 1991 137 20 0.02 0.01 3.81 0.00 0.00 0.10 0.77 4.71 1992 65 58 0.17 0.02 0.22 0.00 0.00 0.07 0.32 0.80 1993 134 25 2.13 0.51 0.01 0.42 0.00 0.81 1.28 5.16 1994 120 55 0.36 0.06 0.71 0.17 0.00 0.04 0.75 2.09 1995 86 45 0.44 0.35 0.06 0.02 0.00 0.13 0.67 1.67 1996 184 32 0.85 0.37 0.10 0.07 0.00 0.52 1.39 3.30 1997 75 45 0.28 0.36 0.00 0.02 0.00 0.26 1.15 2.07 1998 78 40 0.28 0.15 0.00 0.10 0.00 0.18 0.66 1.37 1999 64 42 0.25 0.03 0.25 0.75 0.00 0.03 0.61 1.92 2000 134 21 0.04 0.28 2.32 0.01 0.00 0.01 0.92 3.58 2001 123 28 0.40 0.18 0.88 0.17 0.00 0.24 0.36 2.23 2002 83 41 0.03 0.04 1.03 0.16 0.00 0.03 0.31 1.60 2003 183 39 0.84 0.09 0.36 0.04 0.00 0.21 0.52 2.06 2004 135 13 0.30 0.38 2.36 0.57 0.00 0.06 0.91 4.58 2005 134 30 1.08 0.11 0.70 0.31 0.00 0.13 0.52 2.85 2006 110 25 0.37 0.29 2.50 0.15 0.00 0.09 0.51 3.91 2007 264 50 0.87 0.00 0.67 0.02 0.00 0.08 0.45 2.09 2008 324 16 0.58 0.08 3.54 0.02 0.00 0.08 1.39 5.69 2009 308 44 1.21 0.05 1.63 0.02 0.00 0.05 0.26 3.21 2010 164 13 0.03 0.01 3.93 0.04 0.00 0.01 1.12 5.15 2011 207 37 0.22 0.58 0.93 0.15 0.00 0.08 0.65 2.60 2012 206 21 0.03 0.00 2.25 0.03 0.00 0.02 0.93 3.25 2013 234 63 0.14 0.03 3.12 0.01 0.00 0.58 0.45 4.33 2014 196 30 0.49 0.09 1.84 0.00 0.00 0.07 1.08 3.56 2015 152 14 0.13 0.00 2.91 0.37 0.06 0.49 1.88 5.80 2016 150 21 0.31 0.01 1.79 0.22 0.38 0.03 0.97 3.71 2017 202 12 0.17 0.08 5.76 0.01 2.05 0.08 1.18 9.33 2018 220 15 0.37 0.00 5.86 0.27 0.14 0.07 1.04 7.75 2019 219 38 0.32 0.05 0.63 0.07 0.56 0,17 0,91 2.71

101

Table A4. Young-of-year and age-1 walleye density estimates and mean lengths in Oneida Lake since 1961. Larval walleye density (at the time of the 8 mm perch survey) are from Miller sampler surveys at that time or calculated from the 9 mm larval walleye survey. Age-0 walleye densities (#/ha) and mean lengths (TL, mm) on October 1 are from trawl surveys surrounding the Oct 1 date (5 dates, 50 trawls), and age-1 walleye densities (#/ha) and mean lengths on May 1 are from trawl surveys around May 1 (3 dates, 30 trawls). Densities calculated based on area swept (0.1 ha per trawl) assuming no avoidance.

Year Class

Larval Density

Oct 1 Age 0

Density Oct 1 Age 0 Length

May 1 Age 1 Density

May 1 Age 1 Length

1961 114.5 141 1962 135.9 143 44.2 159 1963 98.5 124 37.9 154 1964 80.6 138 73.4 161 1965 79.4 154 133.0 164 1966 1,348 6.3 139 9.0 148 1967 967 82.4 127 1968 1,580 219.0 144 184.2 164 1969 559 50.0 143 17.0 161 1970 2,271 25.8 121 24.5 167 1971 309 42.0 167 124 181 1972 1,599 6.0 121 12.5 156 1973 222 1.6 164 4.5 174 1974 1,464 14.8 100 6 144 1975 1,362 148.4 171 59 185 1976 2,327 1.6 133 1.5 159 1977 660 71.6 137 108 168 1978 14.6 123 1979 4.6 146 1980 17.8 155 28.0 165 1981 57.8 149 1982 22.4 162 27.7 176 1983 28.0 155 25.5 167 1984 6.0 133 26.3 151 1985 31.0 141 31.5 159 1986 5.4 140 3.8 165 1987 29.8 177 25.0 187 1988 10.4 142 2.2 146 1989 3.0 160 17.0 154 1990 14.4 174 15.0 178 1991 46.7 173 44.0 175 1992 333 12.4 150 5.0 157 1993 10.4 147 13.0 168 1994 11.4 131 11.5 163 1995 13.6 135 11.3 166

102

Year Class

Larval Density

Oct 1 Age 0

Density Oct 1 Age 0 Length

May 1 Age 1 Density

May 1 Age 1 Length

1996 1.8 150 5.0 168 1997 8.0 159 0.7 142 1998 275 2.4 207 3.0 189 1999 1,773 12.4 144 2.7 122 2000 1,208 3.0 177 14.3 181 2001 2,541 19.2 153 33.0 154 2002 213 3.7 174 12.3 179 2003 986 2.5 139 19.5 167 2004 3,196 15.3 151 21.0 162 2005 8,106 5.6 107 13.3 143 2006 1,304 2.2 163 9.3 173 2007 942 7.5 132 3.3 183 2008 5,102 5.4 130 3.67 132 2009 957 1.6 154 1.67 144 2010 898 14.3 132 27.5 162 2011 251 2.5 183 0.0 - 2012 2,694 1.0 104 0.3 140 2013 530 1.7 167 3.0 175 2014 1457 11.8 146 0.7 133 2015 1660 2.2 138 1.0 122 2016 1546 9.0 122 5.0 137 2017 581 5.8 149 0.7 180 2018 1,779 13.0 136 11.0 162 2019 1,962 3.8 152

103

Table A5. Yellow perch density estimates in Oneida Lake since 1961. Data are from mark-recapture (bold) or based on the catch in gill nets using size specific net selectivity.

Year Density (#/ha) at age Total (age3+) 2 3 4 5 6 >6

1961 51.2 11.3 56.7 9.4 10.2 17.5 105.1 1962 18.9 38.4 27.8 40.2 12.8 6.5 125.7 1963 15.6 26.7 40.1 32.7 33.5 16.1 149.1 1964 10.5 11.3 45.0 44.4 32.8 12.4 146.0 1965 11.3 44.2 12.7 67.2 41.4 14.1 179.7 1966 34.3 19.6 28.6 20.9 39.2 12.7 120.9 1967 1.4 50.1 28.1 27.2 20.4 20.3 146.0 1968 37.0 3.5 70.2 16.0 17.1 24.6 131.4 1969 33.2 21.7 7.3 54.3 18.9 18.9 121.2 1970 6.7 48.0 23.1 7.5 61.9 37.9 178.4 1971 1.9 7.7 52.6 17.1 3.0 30.2 110.7 1972 41.5 1.8 7.6 26.9 9.0 11.9 57.1 1973 4.6 144.4 3.9 7.2 17.7 26.2 199.5 1974 No gill netting 1975 39.0 0.9 5.7 61.3 2.5 15.0 85.5 1976 5.3 56.5 2.8 11.2 51.2 14.4 136.1 1977 2.7 12.9 40.0 0.5 2.2 24.2 79.7 1978 19.7 3.9 8.6 41.7 3.6 28.8 86.6 1979 99.1 12.5 5.4 6.1 33.9 10.3 68.1 1980 4.9 179.2 16.3 8.6 14.5 41.3 260.0 1981 16.0 16.3 134.4 23.2 3.7 24.9 202.5 1982 31.2 10.3 10.6 99.6 4.3 8.0 132.8 1983 2.8 27.7 8.2 5.2 54.4 5.8 101.4 1984 18.6 12.6 48.3 17.2 10.3 36.0 124.5 1985 29.8 7.6 5.0 22.2 3.3 12.2 50.3 1986 29.5 24.0 10.3 8.1 28.9 9.0 80.3 1987 15.4 31.7 29.0 11.1 5.0 35.7 112.5 1988 10.0 15.5 24.7 18.9 4.3 21.9 85.4 1989 27.8 7.1 18.6 31.0 23.5 24.2 104.4 1990 8.7 33.5 2.9 5.8 17.2 18.0 77.4 1991 3.4 3.7 18.5 5.9 9.0 22.3 59.4 1992 47.9 5.5 5.2 18.4 6.5 10.0 45.5 1993 29.5 28.2 7.5 4.8 13.7 10.8 65.1 1994 1.7 10.4 8.9 1.5 0.8 6.5 28.1 1995 13.9 4.3 16.1 5.9 1.4 4.0 31.7 1996 26.4 10.7 4.0 8.5 3.6 3.8 30.6 1997 21.3 26.3 7.0 1.4 2.7 1.7 39.0 1998 13.2 23.9 22.0 10.4 4.2 3.9 64.3 1999 4.3 10.5 13.1 8.9 2.7 1.7 37.0 2000 20.3 8.9 15.2 19.4 10.5 7.3 61.4

104

Year Density (#/ha) at age Total (age3+)

2 3 4 5 6 >6

2001 3.7 21.5 7.1 4.8 5.8 6.5 45.7 2002 5.7 7.9 46.0 11.5 10.7 24.6 100.8 2003 1.7 2.3 7.1 21.7 6.1 9.8 47.0 2004 3.4 5.4 5.5 8.3 17.0 19.5 55.7 2005 2.9 13.4 12.4 4.9 9.2 32.6 72.5 2006 15.5 11.0 12.6 8.1 2.9 18.5 53.0 2007 38.2 15.0 7.1 6.5 3.5 11.7 43.8 2008 14.7 41.7 16.0 5.6 3.7 13.4 80.4 2009 8.3 14.8 12.1 3 5.8 3.4 39.1 2010 17.5 14.2 5.5 11.8 9.0 7.9 48.3 2011 33.2 8.5 12.9 6.5 13.8 7.9 49.5 2012 15.0 7.7 19.9 17.6 4.2 18.8 68.2 2013 58.1 23.4 8.1 11.4 18.0 19.7 80.6 2014 47.3 9.0 5.2 2.2 5.4 6.9 28.7 2015 20.0 24.9 8.8 4.2 1.0 5.7 44.6 2016 43.4 9.0 25.9 4.3 1.6 3.1 43.9 2017 13.8 11.0 6.8 12.1 4.4 21.1 55.5 2018 79.9 38.9 14.5 7.3 12.1 15.5 88.3 2019 49.9 43.4 30.6 12.4 2.3 5.2 93.9

105

Table A6. Young-of-year and age-1 yellow perch density estimates and mean lengths in Oneida Lake since 1961. Larval yellow perch densities (at 18 mm, #/ha) are estimated from Miller sampler surveys. Age-0 yellow perch densities (#/ha), age-0 mean lengths (TL, mm) are estimates for October 15 obtained from regression analysis of weekly catches throughout the season. Age-1 yellow perch density are from trawl surveys around May 1 and from mid-July through October (#/ha). Age-1 yellow perch mean lengths are from spring trawl surveys centered on May 1 since 1961.

Year class

Larval density October age-0 Age-1

density mean length

density spring

mean length

density summer

1961 2,850 60 19.4 1962 4,260 73 486 76 186.8 1963 780 60 71 15.8 1964 3,520 71 849 73 585.9 1965 140,100 2,610 60 30 2.0 1966 40,200 170 73 25 74 39.3 1967 61,200 2,240 72 136.5 1968 141,800 6,700 67 598 75 57.2 1969 69,200 210 65 2 0.5 1970 80,000 930 77 158 85 44.5 1971 216,400 3,520 57 52 62 30.5 1972 120,700 100 67 4 77 0.8 1973 16,600 510 86 63 90 46.0 1974 32,000 320 72 33 74 9.3 1975 188,700 450 65 5 75 4.3 1976 46,600 180 72 12 77 4.8 1977 65,200 4,140 69 3385 70 241.5 1978 180 73 13.5 1979 103,200 360 75 6.4 1980 131,600 500 81 118 81 100.9 1981 208,200 2,590 57 4.6 1982 353,400 980 63 25 68 10.6 1983 45,600 710 79 95 79 26.3 1984 16,000 810 71 103 73 32.1 1985 91,100 2,700 68 174 74 29.8 1986 14,600 70 82 2 84 1.8 1987 3,700 220 68 128 70 97.9 1988 76,200 220 81 19 83 4.5 1989 3,700 20 81 17 82 13.1 1990 117,000 460 73 184 70 121.9 1991 34,000 340 82 705 84 166.5 1992 60,800 100 73 13 79 5.4 1993 32,800 320 85 70 84 56.4 1994 21,800 280 83 281 83 30.1

106

Year class

Larval density October age-0 Age-1

density

mean length

density spring

mean length

density summer

1995 15,100 90 90 373 89 58.5 1996 43,600 80 80 74 81 24.3 1997 4,600 30 80 23 80 17.5 1998 57,100 700 83 457 84 99.3 1999 42,100 1,080 81 18 84 17.8 2000 19,300 140 78 73 79 7.2 2001 36,200 270 84 466 86 6.3 2002 23,400 1,660 76 380 80 17.6 2003 68,500 60 85 38 84 5.3 2004 60,700 180 86 36 84 5.7 2005 36,300 410 93 280 93 13.5 2006 58,502 240 79 117 79 19.6 2007 135,990 1,842 81 243 85 6.2 2008 67,420 720 73 104 76 9.1 2009 112,712 1,454 74 1.67 88 6.9 2010 62,853 829 72 47 78 17.8 2011 4,713 21 72 16 73 2.7 2012 22,297 18 88 20 84 28.8 2013 6,654 131 84 62 84 13.6 2014 12,035 427 78 19 77 8.42 2015 28,162 58 82 22 80 22.5 2016 33,477 1,270 74 435 77 46.6 2017 18,665 518 78 30 78 18.1 2018 27,060 842 70 129 78 36.1 2019 96,760 561 86

107

Table A7. Relative abundance of white perch year classes at successive stages of development in Oneida Lake since 1961. Age-0 white perch abundance represented by the calculated density from area swept in trawls in August-September, Age-0 Length is from October trawls, Age-1 Spring is from the CPUE in spring trawls, and age-1 and older are catches in standard gill nets. These values are data for the year of collection. The recruitment index (RI) is the sum of the gill net catch at age 2 and 3 of fish born that year (Fitzgerald et al. 2006). For example the RI value for years 1961 is the sum of the gill net catch of age 2 in 1963 and age 3 in 1964. Bold RI numbers for 1971 and 1972 includes an extrapolation of gill net catches for 1974 when gill nets were not used (see Fitzgerald et al. 2006).

Year Age-0 Age-0 Length

Age-1 Spring

Age-1 Age-2 Age-3 Age-4 Age-5 Age-6 Age7+ RI sum GN

1961 1,236 84 2 9 20 94 6 8 39 10 178 1962 316 94 0 15 2 34 66 10 13 114 140 1963 49 68 0 5 28 5 83 62 15 12 198 1964 56 79 0.4 0 55 5 36 20 62 54 54 232 1965 963 77 9.3 6 7 59 9 56 43 74 9 254 1966 1,318 78 0.0 0 19 5 28 8 54 19 5 133 1967 132 85 0.0 0 3 35 9 15 26 19 16 107 1968 12 86 0 0 6 18 6 20 22 7 72 1969 81 80 0.0 0 0 5 10 21 6 23 3 65 1970 263 81 0.0 0 4 16 20 46 37 56 25 179 1971 85 78 0.0 1 0 3 7 2 9 23 69 45 1972 30 84 0.5 0 11 3 0 0 2 8 9 24 1973 2,106 85 0.0 0 6 14 1 1 0 6 551 28 1974 355 72 3.0 No Gill Net Data 15 1975 208 87 0.0 0 240 5 143 14 2 11 3 415 1976 314 64 0.0 0 4 311 5 101 39 41 8 501 1977 957 77 0.0 0 1 11 128 4 52 11 517 207 1978 40 85 12.0 23 0 2 3 53 1 18 12 100 1979 1,740 78 1 224 8 17 1 228 30 6 509 1980 6,432 79 0 8 293 0 1 3 48 59 353 1981 278 75 2.0 0 1 4 775 28 22 132 10 962 1982 4,824 75 0 21 5 10 411 8 31 29 486 1983 6,669 80 0.0 0 0 38 5 6 343 28 249 420 1984 359 72 0.0 0 6 10 141 10 13 244 297 424 1985 102 70 0.5 1 31 23 12 212 15 372 38 666 1986 17 72 0.5 4 142 218 29 26 195 309 15 923 1987 5,259 62 0.0 1 27 155 31 11 11 69 17 305 1988 4 81 0.0 0 1 11 7 0 3 8 3 30 1989 886 78 0.0 3 4 14 34 4 0 8 8 67 1990 74 72 0.0 2 0 13 19 56 18 17 2 125 1991 86 90 0.0 6 4 3 1 4 19 19 6 56 1992 48 70 0.0 0 0 4 1 10 4 59 0 78 1993 797 79 0.0 0 3 2 2 1 18 40 15 66 1994 66 80 0.0 1 0 3 3 0 2 31 19 40 1995 613 97 0.5 2 4 0 6 2 4 18 243 36 1996 54 87 14.7 2 2 11 3 8 0 14 13 40 1997 956 76 0.0 0 155 17 8 1 5 14 415 200 1998 125 85 0.0 87 0 88 4 10 2 6 202 197 1999 8 97 0.7 40 315 13 122 9 0 4 132 502 2000 590 78 0.7 2 50 100 4 47 2 3 211 208 2001 59 84 2.3 6 56 152 211 14 55 0 283 494 2002 1,145 82 8.3 32 122 76 65 120 7 26 72 448 2003 59 84 5.3 0 106 89 46 52 36 17 5 346 2004 1,413 79 0.0 0 33 177 61 38 27 40 590 376

108

Year Age-0 Age-0 Length

Age-1 Spring

Age-1 Age-2 Age-3 Age-4 Age-5 Age-6 Age7+ RI sum GN

2005 81 98 1.3 44 1 39 227 40 53 17 210 421 2006 137 76 1.7 16 261 4 32 214 50 10 115 587 2007 140 81 0.0 12 111 329 20 34 198 67 78 771 2008 335 75 1.0 5 16 99 126 11 42 74 163 373 2009 196 73 0.7 32 39 99 138 277 38 150 22 773 2010 48 89 2.5 4 79 39 45 71 184 67 93 489 2011 1,181 76 0 22 10 84 32 28 84 282 5 541 2012 144 100 0 0 25 12 68 22 33 216 270 376 2013 7 86 0.3 44 0 68 47 97 29 261 50 546 2014 49 81 0.0 3 108 5 25 20 64 97 34 322 2015 19 88 0.0 0 26 162 17 35 37 168 115 445 2016 15 93 0.0 7 12 24 40 3 6 26 286 118 2017 896 74 0.3 37 36 22 27 77 7 43 250 2018 42 83 0.3 0 82 79 9 37 54 43 305 2019 350 76 0 8 204 65 24 35 82 418

109

Table A8. Abundance and biomass of pelagic fish (emerald shiners (ES), gizzard shad, and Alosa spp. (blueback herring and alewife) in Oneida Lake since 1994.

Abundance (fish/ha) Biomass (kg/ha)

Year ES

Age-0 ES

Age1+ Gizzard

shad Alosa spp.

Sum ES Age-0

ES Age1+

Gizzard shad

Alosa spp.

Sum

1994 3,589 1,352 2,515 607 7,457 2.2 4.1 15.6 6.1 28.0 1995 350 792 538 575 2,255 0.6 3.2 17.2 9.3 30.3 1996 2,909 280 22 492 3,704 2.3 1.0 1.3 5.3 10.0 1997 16,936 1,760 101 14 18,811 15.0 6.4 0.6 0.2 22.3 1998 2,254 5,668 41 3 7,966 1.0 16.1 0.4 0 17.5 1999 7,539 4,093 726 0 12,358 6.1 10.0 8.7 0 24.8 2000 3,463 1,836 1,936 0 7,235 3.4 5.2 6.3 0 14.9 2001 16,112 2,441 2,458 0 21,010 15.2 9.0 23.5 0 47.8 2002 20,529 2,516 2,924 0 25,969 9.4 7.2 5.6 0 22.2 2003 2,645 8,149 2,474 0 13,268 1.7 23.3 13.8 0 38.7 2004 9,057 1,407 2,664 0 13,128 10.1 4.9 14.6 0 29.6 2005 2,597 1,307 2,215 0 6,119 2.6 4.9 47.8 0 55.3

2006 2,651 666 1,716 0 5,033 2.6 2.3 15.5 0 20.4

2007 417 215 1,431 0 2,065 0.6 0.3 14.3 0 15.2

2008 7,900 381 2,073 0 10,354 6.6 1.3 14.6 0 22.5

2009 1,001 1,521 5,969 5 8,496 0.7 4.4 18.8 0.5 24.0

2010 8,350 2,032 7,643 26 18,051 7.0 6.3 24.7 0.2 38.2

2011 35,918 4,067 4,679 0 44,664 23.4 11.6 15.0 0 50.0

2012 2,749 1,224 2,773 0 6,746 3.2 3.8 29.9 0.0 36.9

2013 738 106 1,236 5 2,085 0.7 0.4 7.0 0.1 8.0

2014 2,031 774 4,949 0 7,754 3.0 1.1 11.4 0.0 16.0

2015 3,163 1,206 1,143 0 5,512 4.4 1.7 6.8 0.0 12.9

2016 7,257 2,766 1,539 1 11,563 6.4 2.4 16.5 0.1 25.4

2017 4,312 1,644 4,527 247 10,730 5.2 2.0 24.9 1.7 33.8

2018 2,896 1,104 6,089 0 10,089 2.6 1.0 27.1 0.0 30.7

2019 3,852 1,468 616 0 5,936 3.9 1.5 4.7 0.0 10.1

110

Table A9. Catch/hour of lake sturgeon in large mesh gill nets at 12 standard sites in Oneida Lake since 2002.

Year Month May June July August September October November

2002 - 0.39 0.35 0.10 0.10 0.16 - 2003 0.32 0.17 0.09 0.09 0.56 - - 2004 0.35 0.39 0.08 0.37 0.15 - - 2005 0.18 0.11 - - - - - 2006 0.31 0.11 - - - - 0.06 2007 0.30 0.11 - - - 0.07 - 2008 0.17 0.13 - - - - - 2009 0.20 0.14 - - - - - 2010 0.20 0.04 - - - - - 2011 0.34 0.04 2012 0.40 0.15 2013 0.20 0.19 2014 0.09 0.06 2015 0.24 0.03 2016 0.34 0.06 2017 0.11 0.05 2018 0.47 0.09 2019 0.33 0.20

111

Table A10a. Catches in ½” mesh fyke nets, Oneida Lake, 2007-2015 (n=18 sites). Mean Catch/Net

Scientific Name

2007

2008

2009

2010

2011

2012

2013

2014

2015

Family Lepisosteidae Longnose gar (Adult) <0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Family Amiidae Bowfin (Adult) 0.1 0.2 0.2 0.1 0.3 0.4 0.2 0.1 0.1 Family Clupeidae Gizzard shad (All) 0.2 0.4 0.3 4.1 35.3 3.4 25.1 0.5 0.9 Family Cyprinidae Common carp (All) 0.0 <0.1 0.2 0.0 <0.1 0.0 <0.1 0.0 0.0 Golden shiner (All) 0.2 0.5 0.3 0.2 0.2 0.0 0.1 0.1 0.0 Emerald shiner (All) 0.0 0.0 0.0 0.0 0.0 <0.1 <0.1 0.1 0.0 Spottail shiner (All) 0.1 0.0 0.2 0.2 0.0 0.1 0.0 0.0 0.2 Bluntnose minnow (All) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Family Catostomidae Longnose sucker (All) <0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 White sucker (All) 0.6 0.8 0.4 0.8 0.7 0.4 0.4 0.2 0.4 Creek chubsucker (All) <0.1 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 Greater redhorse (All) <0.1 <0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Family Ictaluridae Yellow bullhead (All) 0.2 0.4 0.5 0.4 0.2 0.2 0.1 0.3 0.1 Brown bullhead (All) 0.8 0.9 0.8 0.6 0.5 0.3 1.8 0.5 0.9 Family Esocidae Chain pickerel (All) 0.4 <0.1 <0.1 0.7 0.6 0.5 1.0 0.5 0.8 Family Gadidae Burbot (Adult) <0.1 <0.1 0.1 0.1 0.0 <0.1 <0.1 0.1 0.1 Family Fundulidae Banded killifish (All) <0.1 0.0 0.0 0.0 <0.1 0.0 0.0 0.0 0.0 Family Percichthyidae White perch (YOY) 1.4 0.1 5.4 <0.1 4.5 0.0 0.0 0.1 0.9 White perch (Adult) <0.1 <0.1 <0.1 <0.1 0.4 <0.1 0.3 0.1 0.0 Family Centrarchidae Rock bass (Adult) 3.1 3.0 2.6 3.3 2.0 2.5 3.7 3.3 2.8 Green sunfish (Adult) 0.2 0.1 0.2 0.3 0.4 0.4 0.4 0.1 0.1 Pumpkinseed (Adult) 9.5 13.4 16.8 7.0 7.5 4.5 10.1 7.0 6.4 Bluegill (Adult) 2.6 3.8 9.3 3.5 4.3 4.4 19.5 5.2 1.7 (YOY Lepomis - <75mm) 3.3 1.1 2.1 3.2 10.7 2.9 3.0 1.0 4.7 Smallmouth bass (YOY) 11.6 4.8 2.4 1.3 1.0 1.0 0.4 1.9 14.3 Smallmouth bass (Adult) 0.0 0.1 0.0 0.2 0.1 0.1 0.0 0.1 0.1 Largemouth bass (YOY) 1.2 2.2 0.7 0.8 0.8 1.0 1.3 2.9 1.8

112

Largemouth bass (Adult) 0.1 0.0 0.1 <0.1 <0.1 0.1 0.2 0.1 0.0 Black crappie (All) 2.4 0.9 0.9 0.8 3.3 0.9 6.2 2.2 1.4 Family Percidae Yellow perch (YOY) 18.2 1.5 0.5 0.3 1.0 6.2 0.3 0.8 4.5 Yellow perch (Adult) 16.0 26.1 24.5 19.5 19.6 16.3 24.5 14.9 25.7 Logperch (All) 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Tesselated darter (All) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Walleye (YOY) 0.2 0.4 0.1 0.2 0.1 0.1 <0.1 0.1 0.1 Walleye (Adult) 0.5 0.5 0.8 0.5 1.6 1.1 1.0 1.4 0.7 Family Sciaenidae Freshwater drum (Adult) <0.1 <0.1 0.1 0.3 0.4 0.4 0.4 0.2 0.1 Family Gobiidae Round goby 0.0 0.0 0.0 0.0 0.0 0.0 0.0 <0.1 0.8

113

Table A10b. Catches in ½” mesh fyke nets, Oneida Lake, 2016- (n=18 sites).

Scientific Name

Mean Catch/Net

2016

2017

2018

2019

Family Lepisosteidae Longnose gar (Adult) 0.1 0.0 0.0 0.1 Family Amiidae Bowfin (Adult) 0.1 0.3 0.6 0.2 Family Clupeidae Alewife 0.0 0.0 0.0 0.0 Gizzard shad (All) 0.1 2.5 7.6 1.2 Family Cyprinidae Common carp (All) 0.1 0.0 0.1 0.0 Golden shiner (All) 0.1 0.0 0.3 0.2 Emerald shiner (All) 0.0 0.0 0.0 0.0 Spottail shiner (All) 0.1 0.0 0.0 0.1 Bluntnose minnow (All) 0.0 0.0 0.0 0.0 Family Catostomidae Longnose sucker (All) 0.0 0.0 0.0 0.0 White sucker (All) 0.7 0.2 0.2 0.6 Creek chubsucker (All) 0.3 0.0 0.0 0.0 Greater redhorse (All) 0.0 0.0 0.0 0.0 Family Ictaluridae Yellow bullhead (All) 0.3 1.1 0.9 0.9 Brown bullhead (All) 0.4 1.1 1.3 0.3 Family Esocidae Chain pickerel (All) 0.8 0.1 0.2 0.3 Family Gadidae Burbot (Adult) 0.0 0.0 0.0 0.0 Family Fundulidae Banded killifish (All) 0.1 0.0 0.0 0.0 Family Percichthyidae White perch (YOY) 0.0 4.2 0.0 0.9 White perch (Adult) 0.1 0.1 0.0 0.1 Family Centrarchidae Rock bass (Adult) 3.1 2.7 3.3 0.7 Green sunfish (Adult) 0.1 0.1 0.1 0.1 Pumpkinseed (Adult) 9.1 14.3 15.3 6.7 Bluegill (Adult) 5.1 10.6 8.7 2.2 (YOY Lepomis - <75mm) 7.7 0.2 0.4 4.1 Smallmouth bass (YOY) 3.0 3.2 2.0 1.1 Smallmouth bass (Adult) 0.3 0.0 0.2 0.0 Largemouth bass (YOY) 1.7 0.1 0.9 1.3 Largemouth bass (Adult) 0.1 1.2 0.1 0.0

114

Scientific Name

Mean Catch/Net

2016

2017

2018

2019

Black crappie 1.5 2.9 1.4 1.1

Family Percidae Yellow perch (YOY) 0.4 0.0 0.2 15.2 Yellow perch (Adult) 26.7 23.4 21.8 1.4 Logperch (All) 0.0 0.0 0.0 0.0 Tesselated darter (All) 0.0 0.0 0.0 0.0 Walleye (YOY) 0.4 0.1 0.2 2.1 Walleye (Adult) 0.7 1.1 1.1 0.0 Family Sciaenidae Freshwater drum (Adult)

0.2 0.4 0.3 0.7

Family Gobiidae Round goby 1.6 0.4 0.3 0.7

115

Table A11a. Catches in 1/4” mesh fyke nets, Oneida Lake, 2008-2015 (2008: n=14 sites; 2009- : n=18 sites). Scientific Name

Mean Catch/Net

2008

2009

2010

2011

2012

2013

2014

2015

Family Lepisosteidae Longnose gar (Adult) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 Family Amiidae Bowfin (Adult) 0.1 0.3 0.0 <0.1 <0.1 0.0 0.0 0.0 Family Clupeidae Gizzard shad (All) 0.1 0.4 0.1 465.1 5.1 168.9 0.3 0.1 Family Cyprinidae Common carp (All) 0.0 <0.1 0.0 0.0 0.0 0.0 0.0 0.0 Golden shiner (All) 0.1 <0.1 0.1 0.4 0.1 0.2 0.1 0.0 Emerald shiner (All) 0.1 <0.1 <0.1 0.2 0.1 5.2 0.1 0.7 Spottail shiner (All) 0.0 0.8 0.1 2.8 0.2 0.7 0.0 0.0 Bluntnose minnow (All) 0.6 7.1 0.3 0.3 0.1 0.7 0.1 0.0 Family Catostomidae Longnose sucker (All) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 White sucker (All) 0.0 0.1 <0.1 0.1 <0.1 <0.1 0.0 0.2 Creek chubsucker (All) 0.0 <0.1 0.0 0.0 0.0 0.0 0.0 0.0 Greater redhorse (All) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Family Ictaluridae Yellow bullhead (All) 0.6 0.3 0.0 0.1 0.2 0.6 0.4 0.3 Brown bullhead (All) 0.5 0.2 0.2 0.2 0.3 0.4 0.3 0.3 Family Esocidae Chain pickerel (All) 0.0 <0.1 0.1 0.1 0.1 <0.1 0.1 0.1 Family Gadidae Burbot (Adult) 0.2 0.0 0.0 0.0 0.2 0.1 0.2 0.0 Family Fundulidae Banded killifish (All) 0.4 1.2 5.1 0.3 1.8 3.5 0.1 2.1 Family Percichthyidae White perch (YOY) 0.0 0.1 0.0 6.0 0.0 0.0 0.1 0.2 White perch (Adult) 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 Family Centrarchidae Rock bass (Adult) 3.6 2.5 3.3 3.0 3.6 2.9 4.2 2.6 Green sunfish (Adult) 0.3 0.3 <0.1 0.3 0.5 0.4 0.2 0.2 Pumpkinseed (Adult) 2.4 1.7 4.3 3.1 1.4 2.3 5.8 1.7 Bluegill (Adult) 2.4 4.4 5.3 3.5 3.5 10.0 4.7 0.6 (YOY - <75mm) 42.9 237.1 21.1 139.0 38.1 1009.7 89.3 176.3 Smallmouth bass (YOY) 11.2 2.5 2.5 5.0 6.2 0.8 9.1 4.2 Smallmouth bass (Adult) 0.9 0.0 0.1 <0.1 <0.1 <0.1 0.0 0.1

116

Scientific Name

Mean Catch/Net

2008

2009

2010

2011

2012

2013

2014

2015

Largemouth bass (YOY) 2.6 0.4 1.5 2.3 1.8 1.0 2.9 0.9 Largemouth bass (Adult) 0.0 0.0 0.0 0.1 <0.1 0.0 0.0 0.0 Black crappie (All) 0.3 0.1 0.5 0.9 0.2 0.9 0.7 0.6 Family Percidae Yellow perch (YOY) 18.3 22.4 25.7 11.8 16.5 9.2 18.0 39.1 Yellow perch (Adult) 1.9 9.4 1.9 3.8 2.0 7.0 6.9 2.6 Logperch (All) 0.1 0.4 <0.1 0.3 0.5 1.0 0.5 0.3 Tesselated darter (All) 0.1 0.6 0.1 0.1 0.1 0.5 0.3 0.2 Walleye (YOY) 0.0 0.0 0.2 <0.1 0.3 0.0 0.0 0.1 Walleye (Adult) 0.0 <0.1 0.1 0.2 0.0 0.1 0.1 0.1 Family Sciaenidae Freshwater drum (Adult) 0.0 0.0 <0.1 <0.1 0.0 0.0 0.0 0.1

Family Gobiidae Round goby 0.0 0.0 0.0 0.0 0.0 0.0 <0.1 7.6

117

Table A11b. Catches in 1/4” mesh fyke nets, Oneida Lake 2016- (n=18 sites). Scientific Name

Mean Catch/Net

2016

2017

2018

2019

Family Lepisosteidae Longnose gar (Adult) 0.0 0.0 0.0 0.0 Family Amiidae Bowfin (Adult) 0.0 0.1 0.1 0.0 Family Clupeidae Alewife 0.0 0.0 0.1 0.0 Gizzard shad (All) 0.0 0.7 6.7 0.7 Family Cyprinidae Common carp (All) 0.0 0.0 0.0 0.1 Golden shiner (All) 0.0 0.1 0.1 0.1 Emerald shiner (All) 0.2 1.8 2.3 0.1 Spottail shiner (All) 0.8 0.3 0.7 0.7 Bluntnose minnow (All) 1.1 0.3 2.1 1.6 Family Catostomidae Longnose sucker (All) 0.0 0.0 0.0 0.0 White sucker (All) 0.0 0.1 0.2 0.0 Creek chubsucker (All) 0.1 0.0 0.0 0.0 Greater redhorse (All) 0.0 0.0 0.0 0.0 Family Ictaluridae Yellow bullhead (All) 0.2 0.7 0.3 0.8 Brown bullhead (All) 0.7 0.3 1.2 0.6 Family Esocidae Chain pickerel (All) 0.1 0.1 0.1 0.1 Family Gadidae Burbot (Adult) 0.0 0.1 0.1 0.0 Family Fundulidae Banded killifish (All) 4.9 4.9 0.8 2.5 Family Percichthyidae White perch (YOY) 0.0 0.3 0.0 0.4 White perch (Adult) 0.0 0.0 0.1 0.0 Family Centrarchidae Rock bass (Adult) 2.7 2.9 4.1 4.9 Green sunfish (Adult) 0.2 0.6 0.1 0.3 Pumpkinseed (Adult) 0.9 1.6 3.4 0.9 Bluegill (Adult) 1.2 2.4 8.3 0.7 (YOY Lepomis - <75mm) 68.7 125.4 179.4 261.5 Smallmouth bass (YOY) 1.7 1.3 0.9 0.9 Smallmouth bass (Adult) 0.1 0.0 0.1 0.2 Largemouth bass (YOY) 3.4 0.0 1.1 1.9

118

Scientific Name

Mean Catch/Net

2016

2017

2018

2019

Largemouth bass (Adult) 0.0 1.3 1.1 0.0

Black crappie (All) 0.2 0.8 0.4 0.2 Family Percidae Yellow perch (YOY) 31.6 7.6 54.6 27.4 Yellow perch (Adult) 5.4 2.3 6.1 1.4 Logperch (All) 0.2 0.9 0.2 0.8 Tesselated darter (All) 0.1 0.0 0.1 0.0 Walleye (YOY) 0.1 0.1 0.3 0.1 Walleye (Adult) 0.0 0.0 0.1 0.2 Family Sciaenidae Freshwater drum (Adult) 0.0 0.2 0.0 0.0 Family Gobiidae Round goby 27.1 29.4 13.4 14.6

119

Table A12a. Catches (#/hr) from spring shoreline electrofishing, Oneida Lake 2011-2015. Catches are means from 8 sites, with each site comprised of a 1 hour predator run and 2 15 minute all fish runs (total predator effort = 12 hours, effort for non-predators = 4 hours).

Scientific Name

Common Name

Catch/Hour

2011

2012

2014

2015

Predators Family Lepisosteidae Lepisosteus osseus Longnose gar (Adult) 1.8 2.0 2.2 3.1 Family Amiidae Amia calva Bowfin (Adult) 2.6 1.6 2.6 1.8 Family Ictaluride Ictalurus punctatus Channel catfish (All) 0.1 0.2 0.2 0.0 Family Esocidae Esox niger Chain pickerel (Age-1) 0.5 0.0 0.6 0.3 Chain pickerel (Adult) 4.0 3.7 6.6 9.3 Family Gadidae Lota lota Burbot (Adult) 3.8 1.0 0.6 0.9 Family Centrarchidae Micropterus dolomieu Smallmouth bass (Age-1) 0.9 1.0 0.2 0.2 Smallmouth bass (Adult) 2.3 3.7 3.7 4.3 Micropterus salmoides Largemouth bass (Age-1) 0.9 2.1 0.2 0.0 Largemouth bass (Adult) 9.6 8.2 12.0 9.7 Family Percidae Sander vitreus Walleye (Age-1) 4.0 1.5 3.3 3.9 Walleye (Adult) 4.7 8.0 4.8 11.7 Family Sciaenidae Aplodinotus grunniens Freshwater drum (Adult) 1.8 5.4 1.8 2.8

120

Scientific Name

Common Name

Catch/Hour

2011

2012

2014

2015

OTHER SPECIES Family Clupeidae Dorosoma cepedianum Gizzard shad (All) 9.5 10.0 3.0 1.5 Family Cyprinidae Notemigonus crysoleucas Golden shiner (All) 4.0 4.5 3.0 6.0 Notropis atherinoides Emerald shiner (All) 15.0 30.3 3.0 117.8 Notropis hudsonius Spottail shiner (All) 4.0 2.0 0.0 0.3 Pimephales notatus Bluntnose minnow (All) 1.8 0.8 0.8 0.0 Family Catostomidae Catostomus commersoni White sucker (All) 4.0 1.3 0.5 0.5 Erimyzon oblongus Creek chubsucker (All) 0.3 0.0 0.3 0.0 Family Ictaluridae Ameiurus natalis Yellow bullhead (All) 0.5 1.0 2.0 0.5 Ameiurus nubulosus Brown bullhead (All) 28.0 37.0 44.0 35.0 Family Atherinopsidea Labidesthes sicculus Brook silverside (All) 0.3 0.3 0.3 0.0 Family Fundulidae Fundulus diaphanus Banded killifish (All) 20.3 3.0 2.3 2.5 Family Percichthyidae Morone americana White perch (All) 0.8 0.3 0.8 0.5 Family Centrarchidae Ambloplites rupestris Rock bass (Age-1) 3.8 0.3 0.0 1.0 Rock bass (Adult) 6.5 5.5 15.0 8.0 Lepomis cyanellus Green sunfish (Adult) 0.8 1.3 1.8 0.3 Lepomis gibbosus Pumpkinseed (Age-1) 4.3 7.8 0.5 1.8 Pumpkinseed (adult) 62.8 22.8 42.0 35.0 Lepomis macrochirus Bluegill (Age-1) 2.8 5.5 0.3 0.0 Bluegill (Adult) 24.5 6.5 12.8 3.0 Pomoxis nigromaculatus Black crappie (Age-1) 0.3 0.0 0.0 0.0 Black crappie (Adult) 0.8 0.5 0.5 0.5 Family Percidae Perca flavescens Yellow perch (Age-1) 129.0 24.5 51.5 106.8

121

Scientific Name

Common Name

Catch/Hour

2011

2012

2014

2015

Percina caprodes

Yellow perch (Adult) Logperch (All)

90.5 25.0

19.3 26.8

98.0 8.0

46.3 14.3

Etheostoma olmstedi Tesselated darter (All) 1.8 1.0 0.0 2.3 Family Gobiidae Neogobius melanostomus Round goby 0.0 0.0 0.0 1.3

122

Table A12b. Catches (#/hr) from spring shoreline electrofishing, Oneida Lake 2017-. Catches are means from 8 sites, with each site comprised of a 1 hour predator run and 2 15 minute all fish runs (total predator effort = 12 hours, effort for non-predators = 4 hours).

Common Name

Catch/Hour

2017

2018

2019*

Predators Family Lepisosteidae Longnose gar (Adult) 3.4 3.8 Family Amiidae Bowfin (Adult) 2.9 3.1 Family Ictaluride Channel catfish (All) 0.0 0.1 Family Esocidae Chain pickerel (Age-1) 0.1 0.0 Chain pickerel (Adult) 3.7 4.8 Family Gadidae Burbot (Adult) 0.4 0.1 Family Centrarchidae Smallmouth bass (Age-1) 3.3 0.1 0.3 Smallmouth bass (Adult) 3.8 1.6 2.0 Largemouth bass (Age-1) 3.5 0.6 0.0 Largemouth bass (Adult) 18.9 17.0 7.6 Family Percidae Walleye (Age-1) 5.6 0.9 Walleye (Adult) 9.2 13.0 16.1 Family Sciaenidae Freshwater drum (Adult) 2.7 3.8

*Only black bass collected in 2019.

123

Common Name

Catch/Hour

2017

2018

2019

Family Clupeidae Gizzard shad (All) 7.5 1.8 Family Cyprinidae Golden shiner (All) 3.0 2.3 Emerald shiner (All) 3.3 20.5 Spottail shiner (All) 0.8 0.0 Bluntnose minnow (All) 0.0 0.3 Family Catostomidae White sucker (All) 1.3 2.0 Creek chubsucker (All) 0.0 0.0 Family Ictaluridae Yellow bullhead (All) 0.5 0.8 Brown bullhead (All) 20.0 14.6 Family Atherinopsidea Brook silverside (All) 0.8 0.5 Family Fundulidae Banded killifish (All) 6.3 2.3 Family Percichthyidae White perch (All) 0.3 1.5 Family Centrarchidae Rock bass (Age-1) 8.8 0.8 Rock bass (Adult) 15.5 5.3 Green sunfish (Adult) 0.5 0.8 Pumpkinseed (Adult) 34.3 7.8 Bluegill (Adult) 23.8 10.5 Age-1 13.3 3.3 Black crappie (Age-1) 0.3 0.0 Black crappie (Adult) 0.0 0.5 Family Percidae Yellow perch (Age-1) Yellow perch (Adult) Logperch (All)

64.3 79.5 6.8

29.8 34.0 8.8

Tesselated darter (All) 0,0 0.3

124

Family Gobiidae Round goby 10.8 1.5

125

Table A13. Open water angling effort (boat hours) since 2002 as determined by tower counts, Oneida Lake.

Year Month May June July August September TOTAL

2002 12,773 21,132 24,983 19,156 15,465 93,509 2003 15,675 24,041 33,281 28,375 20,859 122,231 2004 22,230 37,240 34,681 32,012 17,925 144,088 2005 30,738 35,344 38,622 29,799 21,564 156,069 2006 25,004 41,381 63,308 30,230 19,807 179,730 2007 30,942 40,203 41,183 35,748 26,844 174,921

2010 49,180 40,749 43,819 48,552 26,179 208,479 2011 58,774 41,997 52,025 38,090 23,774 214,660 2012 53,554 49,933 56,295 35,629 18,159 213,570 2013 42,479 59,037 62,224 35,169 19,480 218,389 2014 43,253 57,078 55,955 40,951 20,312 217,548 2015 50,372 51,784 58,005 48,020 24,957 233,139 2016 32,828 50,517 52,422 194,366 2017 23,396 49,789 54,560 184,731 2018 32,079 36,615 43,775 29,127 21,343 168,5561

2019 27,511 48,730 44,538 176,364

1Includes October effort.