Selectivity Evaluation for Two Experimental Gill‐Net Configurations Used to Sample Lake Erie Walleyes

Abstract

We used length frequencies of captured walleyes Sander vitreus to indirectly estimate and compare selectivity between two experimental gill‐net configurations used to sample fish in Lake Erie: (1) a multifilament configuration currently used by the Ohio Department of Natural Resources (ODNR) with stretched‐measure mesh sizes ranging from 51 to 127 mm and a constant filament diameter (0.37 mm); and (2) a monofilament configuration with mesh sizes ranging from 38 to 178 mm and varying filament diameter (range = 0.20–0.33 mm). Paired sampling with the two configurations revealed that the catch of walleyes smaller than 250 mm and larger than 600 mm was greater in the monofilament configuration than in the multifilament configuration, but the catch of 250–600‐mm fish was greater in the multifilament configuration. Binormal selectivity functions yielded the best fit to observed walleye catches for both gill‐net configurations based on model deviances. Incorporation of deviation terms in the binormal selectivity functions (i.e., to relax the assumption of geometric similarity) further improved the fit to observed catches. The final fitted selectivity functions produced results similar to those from the length‐based catch comparisons: the monofilament configuration had greater selectivity for small and large walleyes and the multifilament configuration had greater selectivity for mid‐sized walleyes. Computer simulations that incorporated the fitted binormal selectivity functions indicated that both nets were likely to result in some bias in age composition estimates and that the degree of bias would ultimately be determined by the underlying condition, mortality rate, and growth rate of the Lake Erie walleye population. Before the ODNR switches its survey gear, additional comparisons of the different gill‐net configurations, such as fishing the net pairs across a greater range of depths and at more locations in the lake, should be conducted to maintain congruence in the fishery‐independent survey time series. Received January 5, 2011; accepted May 13, 2011

Size Selectivity of Multifilament Gill Nets for Sampling Alligator Gar: Modeling the Effects on Population Metrics

“…This approach is seldom used because it requires either sampling a known population (Hamley and Regier 1973;Myers and Hoenig 1997;Buckmeier and Schlechte 2009) or using a gear with known biases for comparison (Rollefsen 1953;Holt 1963). Because the abundance and size structure of the population we were sampling were unknown we used the indirect method for estimating size selectivity (Hansen et al 1997;Millar and Fryer 1999;Vandergoot et al 2011). We used Millar's R code (Millar and Holst 1997;Millar and Fryer 1999;Millar 2010;R Development Core Team 2014) to fit five, commonly used selectivity functions: the normal with fixed and proportional variability (a.k.a., normal location and normal scale, respectively), the lognormal, the binormal, and the bilognormal.…”

“…In addition, it is often unclear how the size biases will affect other metrics of interest for management, as many are based on age (e.g., growth, survival). Accurate estimation of population vital rates from fish caught in gill nets requires either a representative sample of the population or the size selectivity of the nets be known to allow estimates to be corrected for bias (Ricker 1975;Vandergoot et al 2011). It is important that biologists estimate the size and age structure of fish stocks accurately to better understand their population status and dynamics (Isley and Grabowski 2007).…”

Seal exclusion devices in cod pots prevent seal bycatch and affect their catchability of cod

“…The type 1 error term for all statistical tests was set at 0.5. The methodology and model is described in Holst and Revill (2009) and the method has subsequently been used in studies with both trawls and passive gears (Kynoch et al, 2011;Vandergoot et al, 2011). The GLMM models proportions of the total catches of cod in the experimental pots relative to the control pots (i.e.…”

Gill net selectivity of yellow perch

“…The literature describing gill net selectivity of some percids is well established (Hamley & Regier 1973;Jensen 1986;Grant et al 2004;Irwin et al 2008;Vandergoot et al 2011). For example, a study of walleye Sander vitreus suggests that incorporating selectivity functions into estimates of age composition would reduce bias (Vandergoot et al 2011).…”

“…For example, a study of walleye Sander vitreus suggests that incorporating selectivity functions into estimates of age composition would reduce bias (Vandergoot et al 2011). Further, walleye have been shown to have positively skewed selectivity curves and up to 40% of the ideal-sized fish that encounter a given mesh size would escape (Grant et al 2004).…”