Hydroacoustic survey of Otsego Lake, 2004 Thomas E. Brooking 1 Mark D. Cornwell 2 INTRODUCTION Fishery managers must often rely on uncertain information regarding abundance and distribution of offshore baitfish populations such as alewife (Alosa pseudoharengus), gizzard shad (Dorosoma cepedianum) and emerald shiners (Notropis atherinoides). Schooling characteristics and patchy distribution of these species often make conventional netting gear ineffective at providing reliable density estimates. However, hydroacoustics combined with netting often provides more reliable estimates (Wanzenbock et al. 2003). In 2004, we sampled Otsego Lake (Otsego Co.) with acoustics to estimate abundance of offshore fishes. This was a cooperative project between Cornell University Biological Field Station, SUNY Cobleskill Department of Fisheries and Wildlife, and SUNY Oneonta Biological Field Station. Otsego Lake has a warm-water fishery dominated by bass, esocids, and sunfishes, while a cold-water fishery is maintained by stocking. In recent years, a walleye population has been established through stocking as well. Nearly all of these fisheries are probably strongly affected by a dense alewife population that became established in the late 1980 s. This report summarizes the results of hydroacoustic surveys of Otsego Lake in the spring and fall of 2004, comparing them to past surveys. METHODS Cornell University researchers surveyed the offshore pelagic fish communities using hydroacoustics. Small-mesh netting for alewife was done in conjunction with those surveys by SUNY Oneonta and SUNY Cobleskill staff. Density of fish targets in the acoustics was estimated along transects in the lake, and the catch in gill nets was used to identify targets and sample length, weight, and depth distribution of various species. Hydroacoustic surveys were conducted on the nights of 2 June and 9 October 2004 along 9 transects throughout the lake. Transects ran from shore to shore along an approximately east-west axis, distributed from the northern to southern ends of the lake. Data were collected with a 70 khz, split-beam Simrad EY500 echo sounder and stored directly on the hard drive of a laptop computer. The transducer was towed at a depth of 1 Visiting Researcher, summer 2004. Cornell Warmwater Fisheries Unit. Cornell University Biol. Field Station. 900 Shackelton Point Rd. Bridgeport, NY 13030. http://www.dnr.cornell.edu/fieldst/cbfs.htm 2 SUNY Oneonta Biology MA candidate. State University of NY at Cobleskill. Fisheries and Wildlife Department. Cobleskill, NY 12043.
approximately 0.5 m. Since echoes from 1 m below the transducer are not usable, this limits the survey to water depth of about 2 m and deeper. The unit was calibrated in spring and summer of 2004 and the performance checked against a standard copper sphere. Echo integration threshold for data collection was set to 80 db. Acoustics data were analyzed with Sonardata Echoview v.3.00.81 software. On the echograms for each of the 9 transects, the surface (0-2 m) and the bottom (0.5 m from the actual bottom) were removed to leave just the open water area for analysis of fish density. Fish density (#/m 2 ) was calculated using the area backscattering coefficient and mean in situ target strength calculated for targets larger than 70 db. Only targets with a target strength of 61 db and larger were considered to be fish targets, based on target strength distributions of alewife in cages (Cornell University unpublished data), so the density was multiplied by the ratio of targets over 61 db. Target strength distributions were checked so that echoes which were too small to be fish could be removed. In these cases (fall 2002, 2004 both spring and fall), the cutoff used for target strength distributions was 51 db, and the densities increased by 1.316 to account for targets between 51 and 61 db based on cage data (Cornell University unpublished data). Density of fish per square meter was then multiplied by 10,000 to provide a density of fish/ha. Average and standard deviation were calculated based on the actual number of transects done (9 in 2004). RESULTS AND DISCUSSION Acoustic fish abundance in June was estimated to be 907 fish/ha, with a 95% confidence interval of +/- 114 fish/ha based on 9 transects. Targets corresponding in size with alewife were concentrated in the upper 15 m of the water column. The percent of yearlings (based on length frequency) in the 2004 spring netting was 44%, and older adults were 56%. Average length and weight of spring yearlings was 91 mm and 5.7 g. Average length and weight of older adults was 146 mm and 23.9 g. Biomass of these alewife, estimated from the acoustic abundance and average weight in gillnets, was estimated to be 2.3 kg/ha for yearlings, and 12.1 kg/ha for older adults. Larger fish targets (-35 db and larger) occurred in the 20-45 m depths at a density of approximately 9.1 fish/ha. These were most likely salmonids, though this estimate may include some other larger fish such as cisco, whitefish, walleye, or bass, several of which are typically bottom-oriented predators but will sometimes suspend in open water when open water forage is available. This is probably a minimal estimate of predator density just in the open water column. Whether this is a representative estimate of the total predator population in the lake is unknown, as the percent of predators in the open water vs. on bottom could vary a great deal based on a whole suit of variables. Many predatory fish are near or right on bottom, thus would not be sampled effectively with acoustics. We also don't have solid data on what size target strengths to expect from what size predators. It s highly likely that the echoes from smaller predators overlap a lot with the echoes from other fish, thus we do not have a great deal of confidence in these numbers.
Acoustic fish abundance in October was estimated to be 2,418 fish/ha, with a 95% confidence interval of +/- 1,026 fish/ha based on 9 transects (Table 1). Targets corresponding in size with alewife were concentrated between 10-20 m deep in the water column (Figure 1). Unfortunately, fall gillnetting from the surface to 14 m deep produced no alewife. Since no alewife were caught in the nets, biomass of alewife can only be estimated using the catch in past gillnets. From 2000 to 2003, the proportion of YOY to adult alewife in fall gillnetting averaged 18.0%, with average weights of YOY and adults averaging 3.14 g and 18.71 g, respectively. Assuming these averages hold true for 2004, the biomass of yoy and adult alewife in fall of 2004 was estimated to be 1.4 and 37.1 kg/ha, respectively. If all fish in 2004 were assumed to be adult, the maximum biomass in fall of 2004 would be estimated to be 45.2 kg/ha. Larger fish targets (-35 db and larger) occurred in the 20-40 m depths (Figure 1) at a density of approximately 3.9/ha (Table 2) with the same uncertainties described above. Table 1 provides a comparison with past acoustic surveys in fall. Fall surveys were done in 1996, 1997, and 2000-2004. All surveys were re-analyzed with the Echoview software using the same methods outlined above, except for Fall 2000 which was done with a different acoustics setup (Biosonics). While there have been some discrepancies between the different acoustics units used, the 95% confidence intervals generally overlap in surveys done with both units at the same time (Mason and Schaner 2001). Abundance of alewife was moderate in 1996 (about 5,200/ha), lower in the 1997 and 2000 surveys (about 1,400 2,000/ha), increased to around 10,000/ha in 2001 and 2002, and decreased considerably over the last 3 years to about 2,000/ha (Figure 2). Date Alew (#/ha) # transects stdev 95%CI 9/16/1996 5170 7 1434 1063 10/12/1997 2053 9 798 521 10/1/2000 (Biosonics) 1382 8 925 774 10/13/2001 8562 9 3811 2490 10/1/2002 10901 16 4886 2394 10/10/2003 3851 16 2901 1421 10/9/2004 2418 9 1571 1026 Table 1. Otsego Lake fall alewife density from acoustics surveys. Date Predators (#/ha) N stdev 95%CI 9/16/1996 8 39 10 3 10/12/1997 4 37 6.4 2.07 10/13/2001 35 41 37 11 10/1/2002 12 59 14.5 3.7 10/10/2003 1 91 2.9 0.6 10/9/2004 4 35 5.6 3.6 Table 2. Estimated abundance of predator-size echoes from acoustics.
Alewife have decreased substantially over the past 2 years. This brings up the question of whether the alewife decrease seen from 2002 to 2004 (and resultant increases in cladaceran zooplankton densities and their mean size; Albright et al. 2005) can be attributed to predation by walleye, which have been stocked throughout that time period. The 2004 alewife densities are similar to the 1997 and 2000 fall surveys. The alewife density found in Fall 2004 (about 2,400/ha) is well within the range of densities observed in the Finger Lakes (1,500-4,000/ha, Cornell University, unpublished data) so alewife are not necessarily "low" in Otsego Lake. The densities observed in Otsego Lake were similar to lake-wide mean densities observed in Lake Michigan (95% CI = 1,169-2,545 fish/ha) in 2004, and both Otsego Lake and Lake Michigan exhibited a peak in fall mean alewife density in 2002 (Warner et al. 2005). It seems likely there was a large year class or two in 2001-2002. Alewife dynamics are often driven by strong year classes that dominate the population for several years, with strong year classes occurring in years of both high and low adult abundance (Madenjian et al. 2005). Has predation by walleye and other predators just cropped this back to "normal" densities, seen in the earlier years of the time series? Predation may have helped reduce numbers, but how low they will be reduced, and whether it is just until the next big year class of alewife comes along, remains to be seen. Alewife densities in fall 2004 were much higher in Otsego Lake than in Lake Huron (<1 fish/ha), where a collapse of the alewife population occurred between 1997 and 2004 (Warner et al. 2005). Further monitoring of the alewife population and more intensive analysis of spring and fall hydroacoustic data for alewife should provide researchers with additional insight into the effects that walleye and salmonid predators may have on the Otsego Lake alewife population. Adult alewife YOY alewife Predators Figure 1. Depth distribution of echoes of different target strengths, 10/9/2004.
14000 12000 Alewife density (#/ha) 10000 8000 6000 4000 2000 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 Year Figure 2. Fall alewife density in Otsego Lake, 1996-2004, with 95% CI. REFERENCES Albright, M., L. Hingula and R. Hamway. 2005. A survey of Otsego Lake s zooplankton community, summer 2004. In 37 th Ann. Rept. (2004). SUNY Oneonta Biol. Fld. Sta., SUNY Oneonta. Mason, D. M., and T. Schaner. 2001. Final report to the Great Lakes Fisheries Commission for the acoustics inter-calibration exercise in 1999. Madenjian, C. P., T. O. Hook, E. S. Rutherford, D. M. Mason, T. E. Croley II, E. B. Szalai, and J. R. Bence. 2005. Recruitment variability of alewives in Lake Michigan. Transactions of the American Fisheries Society 134: 218-230. Wanzenbock, J., T. Mehner, M. Schulz, H. Gassner, and I. J. Winfield. 2003. Quality assurance of hydroacoustic surveys: the repeatability of fish-abundance and biomass estimates in lakes within and between hydroacoustic systems. Ices Journal of Marine Science 60:486-492.
Warner, D.M., R. M. Claramunt, C. S. Faul, and T. P. O Brien. 2005. Status of Pelagic Prey Fish in Lake Michigan, 2001-2004. A report to the Great Lakes Fishery Commission, Lake Michigan Committee, Ypsilanti, MI. March 22, 2005. Warner, D.M., T. P. O Brien, C. S. Faul, and. R. G. Stickel. 2005. Status of Pelagic Prey Fish in Lake Huron in 1997 and 2004. A report to the Great Lakes Fishery Commission, Lake Michigan Committee, Ypsilanti, MI. March 22, 2005.