Midwest Fish & Wildlife Conference 2019

AUTHORS: Nicholas B. Kludt, South Dakota State University; Mark J. Fincel, South Dakota Game, Fish & Parks; Brian D.S. Graeb, South Dakota State University

ABSTRACT: Rainbow Smelt Osmerus mordax and Cisco Coregonus artedi are the primary coldwater forage species in Lake Oahe, South Dakota. Understanding the dynamics of these species is an important aspect of Walleye Sander vitreus and Chinook Salmon Oncorhynchus tshawytscha management. As these coldwater species are pelagic and heterogeneously distributed throughout the stratified reservoir zone, they have been historically surveyed using hydroacoustics. Hydroacoustics offers the ability to efficiently survey large areas, but can be time consuming. We compared the traditional cross-sectional transects (2.5 ± 0.8 km) with an abbreviated longitudinal transect (0.5 ± 0.0 km) survey, using a paired design replicated over three months and two years (n=97). We then analyzed the observed target densities of Rainbow Smelt and Lake Herring using a mean square error (MSE) approach. Observed densities were highly correlated for both Rainbow Smelt (r = 0.91) and Cisco (r = 0.94). Decomposing MSE revealed random error components of 67.3% and 99.7% of Rainbow Smelt and Cisco, respectively, indicating no systemic differences between the paired estimates. In either case, estimates were statistically comparable to a 1:1 line with a zero intercept, indicating high observational agreement. These results show no discernible difference between survey designs. While travel time between sites remains constant, the difference between longitudinal (6 min) and cross sectional (30 min) transect scanning times equates to an 80% time savings (1 hr, 42 mins vs. 8 hr, 38 mins). We therefore recommend the adoption of the longitudinal design for future standardized sampling of Lake Oahe coldwater stocks.