Combining bottom trawl and acoustic data to improve survey derived abundance estimates of semipelagic species
MetadataShow full item record
Abundances of semipelagic fishes are often estimated using acoustic or bottom trawl (BT) surveys, both of which sample a fraction of the water column. Acoustic instruments are effective at sampling the water column, but they have a near-bottom acoustic dead zone (ADZ), where fish near the seafloor cannot be detected. Bottom trawl surveys cannot account for fish that are located above the effective fishing height (EFH) of the trawl. In this dissertation I develop methods for combining BT and acoustic data to improve abundance estimates of semipelagic species. Semipelagic walleye pollock (Gadus chalcogrammus) was chosen as a case study because they are a dominant species with important commercial and ecological roles in the North Pacific. A model combining a subset of acoustic and BT data was developed to estimate ADZ correction and BT efficiency parameters. Fitting this model to the data provided estimates of the catchability ratio between BT and acoustics, the EFH of the BT, and the density dependent efficiency of the BT. Estimates of experimentally-derived ADZ correction and BT efficiency parameters were then used to develop a model predicting BT efficiency as a function of BT catch rate. It was found that BT efficiency decreased with increasing bottom trawl catches resulting in hyperstability of the abundance index derived from BT survey. Density-dependent BT efficiency resulted in spatially and temporarily variable bias in survey CPUE and biased population age structure derived from survey data. Logistic regression models were developed to predict the availability (qa) of pollock to both acoustic and BT gears using environmental predictors and fish length. Findings indicated that on average, availability of pollock in the EBS to the BT was larger than to the acoustics. Availability to both gears depended mostly on bottom depth, light conditions, and fish length, and to a lesser extent on sediment size. Availability to the acoustic gear also depended on surface temperature. A method was developed for combining pollock abundance estimates from BT and acoustic surveys using estimates of efficiency and availability to the BT and acoustic gears. Coefficients of variation (CV) obtained for combined estimates were generally lower than those obtained from either BT, or acoustic surveys. Although this work specifically addresses the assessment of walleye pollock in the EBS it has general applicability in the assessment of other semipelagic species. Methods presented in this dissertation can be applied to other semipelagic species to obtain estimates of ADZ correction or BT efficiency parameters. Similarly, methods for estimating density dependence of the BT, availability to the BT and acoustic gears, and combining BT and acoustic abundance estimates can be applied to other species.
- Fisheries