Variation in Bluegill Catch Rates and Total Length Distributions among Four Sampling Gears Used in Two Wisconsin Lakes Dominated by Small Fish

Abstract Many Bluegill Lepomis macrochirus populations are dominated by fish ≤125 mm total length (TL) that may be underrepresented when using standard sampling gears. To identify efficient sampling methods for these populations, we compared catch per unit effort (CPUE) and TL frequency distributions of Bluegill captured in cloverleaf traps, boat electrofishing, mini-fyke nets, and beach seine hauls from two northern Wisconsin lakes supporting populations dominated by fish ≤125 mm TL. Mean Bluegill CPUE ranged from 41 (SE = 11) fish per cloverleaf trap lift to 16 (SE = 8) fish per beach seine haul. Cloverleaf traps generally captured smaller Bluegill relative to other gears and were the only gear to consistently capture Bluegill ≤80 mm TL. Conversely, boat electrofishing captured the widest TL range of Bluegill, and fish ≥80 mm TL composed a greater proportion of catch (37%) relative to other gears. With few exceptions, the effort required to detect 10% or 25% changes in Bluegill CPUE was >100 units of effort regardless of lake, sampling gear, or month. Furthermore, there was no consistency between lakes or months in terms of which sampling gear required the fewest number of samples to detect a 50% change in CPUE. Estimated units of effort needed to detect 10% or 25% changes in mean Bluegill TL were ≤16 for all sampling gears on the lake with consistently higher CPUE (i.e., more fish to measure per unit). In the lake with lower CPUE, cloverleaf traps consistently required less effort to detect changes in mean TL. We note that comparing sample size requirements among gears is not straightforward because gears are sampling differing segments of the Bluegill population. Our study emphasizes the importance of evaluating gear biases and sampling efficiency so that fisheries managers can develop suitable sampling protocols.