Evaluation of Alternate Analytical Procedures and Variance Components for Strait of Georgia Creel Census Catch and Effort Data

Résumé du livre

A representative creel survey for the Strait of Georgia from Victoria to Campbell River has been conducted since 1980 and provides statistics on catch & effort based on data from angler interviews & aerial overflights that count the number of fishing boats. This report analyzes the utility of several probability models to describe the error structure in the survey's catch per effort (CPE) data. The within-day variation in CPE data is compared with across-day variation within a month to comment on the allocation of interview effort within survey strata. A Monte Carlo analysis is used to define the relationship between variance in average CPE within strata and interview sampling effort. The report evaluates the utility of estimating fishing activity proportion at the time of overflight by fitting a beta distribution model to all activity data within a stratum to account for the time dependence among observations. Variance estimates derived from this approach are compared with those estimated with those derived from other approaches. Daily activity patterns across statistical areas and over time are examined to evaluate the possibility of using activity patterns from adequately sampled strata to increase the precision of poorly sampled strata within a Bayesian framework. Finally, the report examines temporal trends in the proportion of angling effort targeted at salmon to evaluate the possibility of revising the current trip-based method for quantifying effort.