Capture-Recapture Models with Heterogeneous Temporary Emigration

We propose a novel approach for modeling capture-recapture (CR) data on open populations that exhibit temporary emigration, while also accounting individual heterogeneity to allow differences in visit patterns and capture probabilities between individuals. Our combines changepoint processes—fitted using an adaptive approach—for inferring visits, with Bayesian mixture modeling—fitted nonparametric identifying clusters of individuals similar or probabilities. The proposed method is extremely flexible as it can be applied any CR dataset not reliant upon specialized sampling schemes, such Pollock’s robust design. fit the new model motivating salmon anglers collected annually at Gaula river Norway. results when analyzing from 2017, 2018, 2019 seasons reveal two anglers—consistent across years—with substantially different patterns. Most are allocated “occasional visitors” cluster, making infrequent shorter visits mean total length stay around seven days, whereas there exists small cluster “super visitors,” regular longer 30 days season. estimate probability catching whilst more than three times higher obtained does account giving us better understanding impact fishing river. Finally, we discuss effect COVID-19 pandemic angling population by 2020 Supplementary materials this article available online.