Predicting fisheries bycatch: A case study and field test for pilot whales in a pelagic longline fishery. Issue 6 (12th March 2019)

• Record Type: Journal Article
• Authors: Thorne, Lesley H.; Baird, Robin W.; Webster, Daniel L.; Stepanuk, Julia E.; Read, Andrew J.
• Editors: Embling, Clare
• Is Part Of: Diversity & distributions. Volume 25:Issue 6(2019)
• Page Start: 909
• Page End: 923
• Publication Date: 2019-03-12
• Subjects: bycatch -- depredation -- dynamic management -- fisheries management -- pilot whale -- species distribution model; Biodiversity -- Periodicals; Biodiversity conservation -- Periodicals; 577
• DOI: 10.1111/ddi.12912
• Languages: English
• ISSNs: 1366-9516

Abstract

Abstract: Aim: Fisheries bycatch is a major threat to populations of protected species such as marine mammals, seabirds and sea turtles, and static management approaches are often unsuccessful in mitigating bycatch of these highly mobile species. Combining species distribution models (SDMs) with oceanographic data has been proposed as a means of predicting when and where bycatch is likely to occur. However, studies assessing whether SDMs can accurately predict fisheries bycatch using independent data are lacking. Assessing model performance using independent data is necessary to test whether a model is generalizable, and this is particularly important for models with management applications. Here, we use short‐finned pilot whale ( Globicephala macrorhynchus ) bycatch in a pelagic longline fishery as a case study to inform efforts to mitigate fisheries bycatch. Location: Offshore waters, north‐east United States. Methods: We integrated telemetry and oceanographic data using mixed‐effects generalized additive models to predict pilot whale occurrence and assessed model performance using k‐folds cross‐validation. We then evaluated the model's ability to predict pilot whale bycatch using data from independent on‐board observers. Results: The model performed well, and predictions were strongly and significantly correlated with observed rates of bycatch in space and time. Temperature and proximity to mesoscale oceanographic features (thermal fronts and sea level anomalies) were important predictors of pilot whale occurrence, and as a result, spatial predictions of the risk of bycatch varied through time. Main conclusions: Our findings demonstrate that SDMs can be used to accurately predict times and places with a high risk of bycatch, and illustrate that models using dynamic oceanographic variables can identify smaller, more specific focal management regions than static management approaches. Combining SDMs with near real‐time or forecasted environmental conditions could provide a promising tool for decreasing bycatch and will be valuable in developing adaptive management strategies to mitigate fisheries bycatch of protected species.