Comparison of three sampling methods for small-bodied fish in lentic nearshore and open water habitats.

We performed a preliminary evaluation of a mobile sampling platform with adjustable push net and live box (Platform) against two common methods for sampling small-bodied fish (i.e., 10-100 mm) in two distinct lentic habitats. Nearshore (NS) littoral habitat was sampled by Platform and beach seine, and open water (OW) pelagic habitat by Platform and Kodiak trawl. Our goal was to evaluate the Platform's ability to describe fish assemblage structure across habitat types in contrast to common techniques restricted to single habitat types that are less comparable due to gear-specific bias. Platform sample speed had a significant positive effect on recapture efficiency of both nearly neutrally buoyant objects and marked fish. Marked fish recapture efficiencies were similar for Platform in NS and OW, indicating similar efficiency across habitat types. Platform capture efficiency was similar to beach seine and greater than Kodiak trawl. With similar sampling time, the Platform collected more individuals and taxa in NS relative to beach seine and in OW relative to Kodiak trawl. Greater taxa detection by the Platform suggests that it may be effective at detecting species that are numerically rare in specific habitats when compared to these methods. Fish CPUE was significantly greater NS regardless of technique. However, by using the Platform, there is greater confidence that this difference was reliable and not a gear selectivity artifact. Overall, this preliminary study demonstrates the Platform's potential to collect standardized data across NS and OW habitats, track ontogenetic habitat shifts, and detect differences in small-bodied fish taxa richness, relative abundance, and density between NS and OW habitats. Continued experimentation beyond a single reservoir and fish size range is required before consensus can be established regarding the utility of this new push net design.