Confidence intervals for trawlable abundance from stratified-random bottom trawl surveys

Abstract

A basic objective of most trawl surveys is to estimate average fish density and account for uncertainty in the estimation. This is of increasing relevance because of the desire to better reflect uncertainty about stock status in fisheries management decisions. I develop score-statistic confidence intervals based on the common stratum area-weighted average estimator for stock abundance based on catches from stratified-random bottom trawl surveys and on the assumption that variation in survey catches is negative binomial in form. These confidence intervals cover only positive values and have improved coverage accuracy compared with standard-t or bootstrap intervals that are commonly used for this type of data. Methods are applied to six case studies, and simulations based on these case studies are used to evaluate the accuracy of the confidence intervals.