ACOUSTICS AS A TOOL FOR THE ASSESSMENT OF GREAT-LAKES FORAGE FISHES

Sharp reductions in forage fish populations in Lake Michigan have raised concerns about the continued ability of the forage stocks to support large populations Of lake trout Salvelinus namaycush and other salmonid predators. There was a need for a more comprehensive and accurate assessment of forage fish abundance and distribution to evaluate these concerns. In response. cooperative diel surveys of the Lake Michigan forage species were conducted in late summer 1987 and spring 1989 with acoustics, midwater and bottom trawls. In late summer 1987, the acoustic estimate (night) of the midwater standing stock of alewives Alosa pseudoharangus, rainbow smelt Osmerus mordax, and bloaters Coregonus hoyi was 366 900 +/- 73 600 (95% CI) t. In comparison, bottom-trawl catches estimated the stocks at 235 500 t (day) and only 23 500 t (night). In spring 1989, the acoustic estimate was 435 000 +/- 110 000 t and the bottom-trawl estimates were 269 300 t (day) and 25 400 t (night). The composition of the stock (by weight) was similar both years; bloaters dominated the catches with alewives and smelt represented in roughly equal proportions. The target strength-length (cm) relationship, TS = 18.2 log10(L)-67.5, was determined for rainbow smelt, based on midwater trawling data. As most of the fish caught in midwater were rainbow smelt, the equation was nol representative of the mixed-species population so the average backscattering cross-section per unit weight was used to scale the echo-integrator for biomass computations. Data from the acoustic surveys appeared to provide a more accurate estimate of the size of the mixed-species populations in Lake Michigan than did the trawl catches.