Estimating angler effort and catch from a winter recreational fishery using a novel Bayesian methodology to integrate multiple sources of creel survey data

Abstract

The range of survey methodologies for measuring daily activity, catch and harvest (i.e. creel surveys) of recreational anglers, is increasing with the advent of new technologies and improvements in remote sensing. Individual creel survey types frequently give different insights into a fishery due to their unique sources of methodological bias and coverage, which creates a problem for resource managers since markedly different estimates of important fishery metrics can result. We demonstrate a joint estimation approach using a Bayesian statistical framework that can bring together multiple survey types to derive a single estimate for important metrics. This framework is applied to data collected from a relatively large winter fishery and integrates three traditional creel methodologies (i.e. roving, access and aerial counts), each with very different sources of bias, to derive a common estimate of angler effort. Models integrating two survey types are found to be have lower uncertainty in their estimates. Further, reductions in effort for any one survey type is found to be buffered by the joint estimation approach, such that resource managers will likely find benefits in using more than one survey methodology in an integrated fashion to monitor a fishery.