Evaluation of Gill-Net Sampling for Estimating Abundance and Length Frequency of Reservoir Shad Populations

Abstract

Horizontal and vertical gill nets were used to intensively sample gizzard shad Dorosoma cepedianam and threadfin shad D. petenense in Lake Texoma, Texas–Oklahoma, in August 1991. We evaluated spatial patterns, sampling precision, mesh-size selectivity, and sample sizes required to detect differences in shad abundance. Abundance patterns differed among species, length-groups, and sampling sites. Age-0 shad occurred mainly near the surface, whereas older fish were more evenly distributed over depth. Spatial variation in abundance was more pronounced among embayments than between nearshore and offshore sites within embayments. Spatial aggregation of shad was similar to that reported for a variety of other aquatic organisms, suggesting that experimental designs for gill-net surveys should incorporate statistical approaches similar to those used in other sampling programs. With horizontal gill nets, investigators can estimate median catch rate (number per net) within 25% of the true median with 95% certainty by using 37 or fewer nets, provided that the variance of the log10-transformed catch data is 0.09 (the median observed variance) or less. At the maximum observed variance (0.55), 4–20 gill nets would be required for a t-test to detect a twofold difference between two independent density estimates at ∝ = 0.05 and power (1 – β) = 0.95. Catch data in vertical gill nets produced less precise abundance estimates, and length frequencies of shad collected in the same mesh sizes differed between horizontal and vertical nets. Estimates of mesh-size retention probabilities, based on relations between shad girth and net mesh perimeter, were used to determine mesh sizes needed for efficient sampling of a wide range of sizes of both shad species. Mesh sizes recommended for sampling shad (55–380 mm) collected at Lake Texoma ranged from 8- to 51-mm-bar measure.