Resurgence of spawning rainbow smelt (Osmerus mordax) in the Mohican Canyon Creek, Otsego Lake, NY INTRODUCTION

Transcription

1 Resurgence of spawning rainbow smelt (Osmerus mordax) in the Mohican Canyon Creek, Otsego Lake, NY Mary K. Mulvihill 1 and John R. Foster 2 Abstract: Rainbow smelt (Osmerus mordax), a key component of the 1980s cold-water fish fauna of Otsego Lake, was decimated in the 1990s by the introduction of alewives (Alosa pseudoharengus). With the recent collapse of the alewife population, the rainbow smelt population was expected to rebound. The goal of this study was to examine the population dynamics of spawning rainbow smelt in the Mohican Canyon Creek, for evidence of that recovery. Between 9-14 April 2016, (9:00pm-12:00am) 1,357 adult smelt were sampled using a Halltech backpack electrofisher. This 2016 study found that the smelt spawning population had increased, however, average size of spawners, spawning age and sex ratios had not return to the pre-alewife levels (1983 and 1984). INTRODUCTION Rainbow smelt (Osmerus mordax) were introduced into Otsego Lake in 1978 where they used many of the lake s tributary streams for spawning (Cornwell 2001, Cornwell 2004, 2014), including Mohican Canyon (Cornwell 2001), 3-Mile Point Creek (Cornwell 2004), 6-Mile Point Stream and Shadow Brook (Harman et al 1997). Smelt abundance was high in the 1980s and a dip net fishery developed in these tributary streams during the spawning runs (MacWatters 1983). Population studies were conducted of rainbow smelt in Mohican Canyon Creek by MacWatters (1984), Cornwell (2004), Best (2015) and Best and Foster (2016). Spawning behavior there was also documented (Cornwell 2001). Rainbow smelt was a major component of the Otsego Lake ecosystem, providing a food source to piscivorous fish species into the mid-1980s. However, following the 1986 alewife introduction (Foster 1990) and their subsequent population explosion (Foster and Gallup 1991) the rainbow smelt population declined (Harmen et al. 1997). The diet of alewife, as well as their habitat within the water column, overlaps the smelt niche (Smith 1970). Thus, smelt populations could have been negatively impacted through species competition or predation on smelt fry by alewife (Simonin et al. 2012). Walleye (Sander vitreum) were introduced into the lake in 2000 to enhance Otsego Lake s fisheries and to potentially control the overabundant alewife (Golding 2006). This led to the decline and subsequent extirpation of the alewife population (Waterfield and Cornwell, 2014). With its primary competitor gone, rainbow smelt populations were expected to resurge. In order to monitor this comeback, the spawning population in Mohican Canyon Creek was studied. The goal of the study was to characterize the population dynamics of rainbow smelt spawning in Mohican Canyon Creek. To meet this goal, sex ratio, growth, age and length distribution and population size was determined. These data were compared to data collected by MacWatters (1984) before the introduction of alewives and data collected in 2015 (Best and Foster 2016). 1 SUNY Oneonta Biol. Field Station Intern, Fisheries & Aquaculture Student, SUNY Cobleskill, Cobleskill, NY. 2 Professor & Chair, Fisheries, Wildlife & Environmental Science Department, SUNY Cobleskill, Cobleskill, NY.

2 This study was conducted at Mohican Canyon Creek (Latitude , Longitude ) near 5-Mile Point (Figure 1). Smelt in their spawning run have been studied at Mohican Canyon Creek by MacWatters (1984), Cornwell (2001 and 2004) and Best and Foster (2016). MacWatters (1984) initial study was conducted at Mohican Canyon Creek, because of the large spawning run and its location on private property away from public access. MATERIALS & METHODS This study was conducted 09, 11 and 14 April 2016, between 9:00pm and 12:00 am. A Halltech Backack Electrofisher was used to capture spawning smelt. Several electrofishing runs were conducted at 45 minute intervals with a shock time of seconds. Electrofishing started at the stream mouth and continued upstream for 26m, covering an area of 93.3m 2. Headlamps were used to conduct the sampling. Captured smelt were measured (total length), sexed, and a few scales were removed from behind the pectoral fin for aging. All sampled fish received an adipose fin clip. Smelt were kept in totes during processing and then released back into Mohican Canyon Creek. Figure 1. This study was conducted at Mohican Canyon Creek at its confluence with Otsego Lake at 5-Mile Point.

3 Per cent Spawners RESULTS This study was based on 1,357 spawning smelt collected from Mohican Canyon Creek on 9-14 April Water temperature during spawning runs ranged from 4.3 C to 6.3 C; water turbidity was NTU, ph ranged from 6.4 to 7.7 and dissolved oxygen ranged from 13.2 to 15.5 mg/l (Table 1). Table 1. Water parameters in Mohican Canyon Creek during April 2016 sampling. Date Air Temp Water Temp ph Conductivity Dissolved Oxygen Turbidity ( C) ( C) (us/cm) (mg/l) (NTU) 4/9/ /11/ /14/ Sex Ratio In Mohican Canyon Creek, female smelt were significantly less abundant than male smelt in 2016, 2015 and 1983 (Chi square test, P<.001, Figure 2). Of 424 smelt sexed in 2016, only 16% of smelt captured were females. Similarly, in 2015 only 7% were females, while in 1983 females made up 32% of the spawning population and 47% in Male Female Year Figure 2. Per cent of males and females spawning in Mohican Canyon Creek in 1983, 1984, 2015 and 2016.

4 Number of Spawning Smelt Males were present in the stream throughout the evening sampling period ( hrs.; Table 2). Female smelt were sampled in each run, however, their numbers increased throughout each night. This was different than the pre-alewife study, showing the peak time females entered the stream was around 2230 hrs. (MacWatters 1983). Table 2. The number of male and female rainbow smelt captured along a 26m spawning reach of Mohican Canyon Creek, Date Run Time Males Females Size of Spawning Smelt Spawning smelt were primarily between 121mm and 140mm (Figure 3). Spawning females (mean length 133mm) were significantly larger than spawning males (mean length 129 mm; T-test, P <.03) Females Males Total Length (mm) Figure 3. Length frequency distribution of spawning rainbow smelt captured in Mohican Canyon Creek in April 2016.

5 Per Cent Spawning Age In 2016, spawning smelt were dominated by one year old fish, which made up 55% of the spawning population (Figure 4). In pre-alewife years (1983 and 1984), 2-year-old smelt dominated the spawning population (62% and 88%). The percentage of 3-year-old smelt in 2016 was 13%, which was greater than the percentages captured in all other studies (6% in 1983; 0.3% in 1984; 5% in 2015). There were no smelt in any of the studies that were over 3+ years of age. The age frequency distribution of rainbow smelt spawning in Mohican Canyon Creek was significantly different between all 4 years studied (Chi square test, P<.001) Age of Spawning Population Figure 4. Age frequency distribution of rainbow smelt spawning in Mohican Canyon Creek in 1983, and Growth On average, the one year old smelt sampled in 2016 were the same size as the smelt captured in 1983 and 1984 (Figure 5). However, the size of the 2-year old smelt captured in 2016 was 10-13mm smaller than in 1983 and The size of the three year old smelt were 20mm below the average length of the smelt captured in all other studies.

6 Total Lenght (mm) Age1 Age 2 Age Year Figure 5. Size at age for rainbow smelt spawning in Mohican Canyon Creek. Spawning Population Size Estimations of the population size of spawning smelt presents problems. Marked males were much more likely to be recaptured than marked females (Chi square Test, P <.001). The percent of marked males recaptured was 11.3%, while the percent of marked females recaptured was only 4.3%. Therefore, the estimation of the population size of spawning smelt was likely to be biased because males remain on the spawning ground longer than females. Further, since smelt freely move into and out of the spawning area, the population being sampled is not closed. While the population estimate may not be completely accurate, it may be useful for future comparisons. Two methods were used to estimate spawning smelt population in Mohican Canyon Creek in The Schnabel method (Ricker 1975) estimates the population at 5,709, while Chapman (Ricker 1975) estimates the population to be 9,315. The number of spawning smelt captured in 2016 was considerably higher than the number of smelt captured in In 2015, 152 smelt were captured in 9 runs giving an average number captured per run at 17. In 2016, 1357 smelt were captured in 6 runs giving an average number captured per run at 226. Thus, the catch per unit effort, as well as the density of spawning smelt was 13.3 times higher in 2016 compared to 2015.

7 DISCUSSION Data collected in this study show some evidence of a rebound of the rainbow smelt population in Mohican Canyon Creek in Smelt density and catch per unit effort increased 13.3 times over that of 2015 (Best and Foster 2016). However, other measures of smelt population dynamics were more equivocal. The mean size of the smelt sampled in 2016 (130mm) was 5mm larger than in More importantly it was 5 mm larger than the mean length (125mm) when alewives were abundant (Cornwell 2004). So, while the 2016 smelt spawning size is increasing in the absence of alewives, it is still considerably smaller than the 149mm mean size of pre-alewife smelt (MacWatters 1984). In pre-alewife years, 2-year olds dominated the spawning population (MacWatters 1984). The spawning age of smelt captured in 2015 and 2016 was dominated by 1-year olds. However, in 2016 the proportion of smelt age-2 and age-3 increased, possibly indicating a change in age structure toward the pre-alewife years. Growth to age-1in 2016, was faster than in 2015 and the same as in the pre-alewife years (1983 and 1984). However, growth to age-2 and age-3 was far below the pre-alewife years (MacWatters 1984). The smelt spawning population was mainly males in 2015 and 2016, while pre-alewife sex ratio was more evenly distributed (MacWatters 1984). However, in 2016 the proportion of females increased over that of 2015, which again may indicate movement toward the pre-alewife sex distribution. The 2015 and 2016 post-alewife studies were based on backpack electrofishing surveys and produced comparable data. However, it is difficult to compare these studies to previous studies which used dip-nets (MacWatters 1983, 1984) and seines (Cornwell 2001, 2004). A further complication is that rainbow smelt populations can have nearly a 10-fold variance over a decade (Rupp 1968). Therefore, to gain a better understanding of the changes in the smelt population in post-alewife years, spawning population should continue to be monitored at Mohican Canyon Creek. REFERENCES Best, M.J Status of rainbow smelt (Osmerus mordax) in the Mohican Canyon tributary, May In 47 th Ann. Rept. (2014). SUNY Oneonta Biol. Fld Sta., SUNY Oneonta. Best, M.J. and J.R. Foster Characterization of spawning rainbow smelt (Osmerus mordax) in the Mohican Canyon Creek, Otsego Lake, NY. In 48 th Ann. Rept. (2015). SUNY Oneonta Biol. Fld Sta., SUNY Oneonta.

8 Cornwell, M. D Evidence of rainbowsmelt (Osmerus mordax) spawning in Mohican Canyon, a tributary of Otsego Lake. In 34 th Ann. Rept. (2001). SUNY Oneonta Biol. Fld. Sta., SUNY Oneonta. Cornwell, M. D Characterization of rainbow smelt (Osmerus mordax) spawning in Mohican Canyon, In 37 th Ann. Rept. (2004). SUNY Oneonta Biol. Fld. Sta., SUNY Oneonta. Crowder, L.B Alewife, rainbow smelt, and native fishes in Lake Michigan: competition or predation? Environmental Biology of Fishes. 5: Foster, J. R Introduction of the alewife (Alosa pseudoharengus) into Otsego Lake. In 22 th Ann. Rept. (1990). SUNY Oneonta Biol. Fld. Sta., SUNY Oneonta. Foster, J.R., and S. Gallup Irruption of the alewife population of Otsego Lake. In 23 rd Ann. Rept., SUNY Oneonta Bio. Fld. Sta., SUNY Oneonta. pp Harman, W.N., L.P. Sohacki, M.F. Albright and D.L. Rosen The state of Otsego Lake Occ. Pap. #30. pp McWatters R.C The fishes of Otsego Lake (2 nd ed.) Occ. Paper #15 SUNY Oneonta Bio. Field Station, SUNY Oneonta. McWatters R.C The age, growth and food habits of the rainbow smelt, Osmerus mordax (Mitchill) in Otsego Lake, New York. In 16 th Ann. Rept. (1983). SUNY Oneonta Biol. Fld. Sta., SUNY Oneonta. Rupp, R.S Life history and ecology of the smelt (Osmerus mordax) in inland waters of Maine. Maine Dept, of Fish & Game. 36 pp. Ricker, W.E Computation and interpretation of biological statistics of fish populations. Fisheries Research Board of Canada, Bulletin 191. Simonin, P.W., D.L. Parrish, L.G. Rudstam, P.J. Sullivan and B. Pientka Native rainbow smelt and nonnative alewife distribution related to temperature and light gradients in Lake Champlain. J. of Great Lakes Research Vol.38 (Supplement 1); Smith, S.H Species interactions of the Alewife in the Great Lakes. Trans. Am. Fish. Soc. 99(4): Waterfield, H.A. and M.D.C. Cornwell Hydroacoustic survey of Otsego Lake s pelagic fish community, spring In 46 th Ann. Rept. (2013). SUNY Oneonta Biol. Fld. Sta., SUNY Oneonta.