Filtering method for obtaining stock indices by shark species from species-combined logbook data in tuna longline fisheries

Abstract

A method of filtering logbook data to obtain estimates of catch per unit effort (CPUE) for sharks has been proposed which simultaneously addresses the issues of under-reporting and the lack of species-specific catch records in historical data. Logbook catch data from vessels with high shark reporting rates are used to represent catch rates for the abundant blue shark Prionace glauca and low reporting vessels data are used for the main commercially valuable species, the shortfin mako Isurus oxyrinchus. Logbook reporting rate filter (RRF) levels are evaluated through analytical and model-based comparisons to species-specific logbook records available since 1994 and shipboard observer data. At the high reporting rates, the ≥80% filter avoids large numbers of false zeros and provides the best fit to observer data for blue shark. At the low reporting rates, the <40% filter best matches the observer data for makos, but this filter level includes large numbers of false zeros and sharks of other species. The ≤20% filter produces a dataset that is better targeted to mako catches but considerably under-estimates CPUE relative to observer records. For these reasons, other means of estimating mako catch rates are suggested for further research.