Accuracy and precision of hydroacoustic estimates of Gizzard Shad abundance using horizontal beaming

Abstract

Prey-density data are often used when making management decisions for piscivorous fish species (i.e. stocking rates). Gizzard Shad (Dorosoma cepedianum) are an important prey species in lakes and reservoirs throughout much of the United States. Currently, gill nets are the most common gear used to collect Gizzard Shad data used for deriving population characteristics, but this gear can be time and labor intensive, lacks precision, and may lack accuracy. Horizontally-oriented echosounders may be a better alternative, but accuracy and precision must be measured to determine if this sampling technique produces reliable data. We released Gizzard Shad into a net pen (15-m long × 15-m wide × 4.5-m deep with 6.35-mm square mesh) to produce several different densities of fish. Mean densities were estimated using five passes with a Simrad® EK60 120 kHz split-beam echosounder. Density estimates were acquired by echo-counting and echo-integration. Mean density estimates were then compared to known densities using a linear mixed-effects model and relative standard error (RSE) was calculated for each trial from the five sampling passes. Both echo-counting and echo-integration had slopes not significantly different from one (t = 0.82, d.f. = 13, P = 0.33; t = 3.58, d.f. = 13, P = 0.55) but intercepts that were significantly greater than zero (t = 2.89, d.f. = 88, P < 0.01; t = 3.53, d.f. = 88, P < 0.01) indicating horizontally-oriented echosounders can accurately detect changes in Gizzard Shad density, but may overestimate actual density, particularly when small. Horizontally-oriented echosounders accurately estimated relative Gizzard Shad density with good sampling precision (i.e., when imaging the same aggregation of fish), indicating data collected with this method would be reliable when making management decisions.