Estimating fishing effort in small-scale fisheries using GPS tracking data and random forests

During the last decade spatial patterns of industrial fisheries have been increasingly characterized using tracking technologies and machine learning analytical algorithms. In contrast, for small-scale fisheries, fishers’ behaviour estimating mapping fishing effort has only anecdotally explored. Following a comparative approach, we conducted boat survey in reef fishery Madagascar investigated performance random forest algorithm speed threshold effort. We monitored movements sample 31 traditional sailing boats at around 45 s time interval small GPS trackers. A total 306 daily tracks were recorded among five gear types (beach seine, mosquito trawl net, gillnet, handline, speargun). To ground-truth location data, was simultaneously by single on-board observer 49 tracks. Typical, gear-specific track observed. Overall, model found to be most reliable, generic, complex method processing detecting spatially-explicit events regardless type. Predictions mean per trip showed that both methods reached from 89.4% 97.0% accuracy across types. Our findings combined with observation would improve reliability indicators contribute more efficient management. Selection appropriate data is dependent on local use, indicators, available expertise.