Efficiency and selectivity of gill nets for assessing fish community composition of large rivers

To evaluate the efficiency and selectivity of gill netting for assessing fish biodiversity in the upper Ohio River system, we compared the efficiency of five gill-net types for sampling large-bodied fishes (adult total length greater than 250 mm) during fall 2001 and spring and fall 2002. Mesh sizes ranged from 3.8 cm to 14 cm (bar measure). We set the gill nets in selected pools of the Allegheny, Monongahela, and Ohio rivers 186 times over three seasons for a total of 1,644 net-hours. Nets were attached to a variety of structures, including trees and rootwads, bridge pylons, lock and darn chambers, and channel marker buoys. Nets were fished from late evening to first light, and all fish captured were identified, enumerated, and released. A total of 823 individuals representing 30 species or hybrids were captured. All net types captured common carp Cyprinus carpio and flathead catfish Pylodictis olivaris, but we captured a significantly greater diversity of fishes in graded-mesh gill nets with small mesh (38 in in length with variable-bar mesh and 15.2 to in length with 3.8-cm-bar mesh). When adjusted for length, smaller-bar mesh nets (3.8-cm-bar mesh) were more efficient for capturing target species than graded-mesh nets. To maximize species richness, 200-225 h of effort were necessary to characterize target fish communities of large bodied riverine species.