Effects of dreissenids on monitoring and management of fisheries in western Lake Erie

Abstract

Water clarity increased in nearshore areas of western Lake Erie by the early-1990s mainly as a result of the filtering activities of dreissenid mussels (Dreissena spp.), which invaded in the mid-1980s. We hypothesized that increased water clarity would result in greater trawl avoidance and thus reduced ability to capture fish in bottom trawls during daytime compared to nighttime. We examined this hypothesis by summarizing three analyses on fish data collected in western Lake Erie. First, we used a two-tiered modeling approach on the ration (R) of catch per hour (CPH) of age-0 yellow perch (Perca flavencens Mitchell) at night to CPH during daytime in 1961-2005. The best a priori and a posteriori models indicated a shift to higher CPH at night (R > 1) between 1990 and 1991, which corresponded to 3 years after the dreissenid invasion and when water clarity noticeably increased at nearshore sites. Secondly, we examined effects of nighttime sampling on estimates of abundance of age-2 and older yellow perch, which form the basis for recommended allowable harvest (RAH). When data from night sampling were included in models that predict abundance of age-2 yellow perch from indices of abundance of age-0 and age-1 yellow perch, predicted abundance was lower and model precision, as measured by r-squared, was higher compared to models that excluded data collected at night. Furthermore, the use of only CPH data collected at night typically resulted in lower estimates of abundance and more precise models compared to models that included CPH data collected during both daytime and nighttime. Thirdly, we used presence/absence data from paired bottom trawl samples to calculate an index of capture probability (or catchability) to determine if our ability to capture the four most common benthic species in western Lake Erie was affected by dreissenid-caused increased water clarity. Three species of fish(white perch, Morone americana Gmelin; yellow perch; and trout-perch, Percopsis omiscomaycus Walbaum) had lower mean daytime catchability than nighttime catchability after dreissenids became established, which supported the hypothesis of greater trawl avoidance during daytime following establishment of dreissenids. Results from freshwater drum (Aplodinotus grunniens Rafinesque) were opposite those of the other three species, which may be a result of behavioral shifts due to freshwater drum feeding on dreissenids mussels. Collectively, these three studies suggest that dreissenids indirectly affected our ability to assess fish populations, which further affects estimates of fish densities and relationships between indices of abundance and true abundance.