A machine learning species distribution model for the critically endangered east Pacific leatherback turtle

The Eastern Pacific leatherback turtle population (Dermochelys coriacea) has dramatically declined in recent years, being bycatch from coastal and pelagic fisheries one of the major causes. In this study, we created a machine learning species distribution model trained with fisheries observations and remotely sensed environmental data. Through a highly collaborative international participatory approach, we obtained leatherback observation data from multiple fisheries that operated in the eastern Pacific Ocean between 1995 and 2020. A daily predictive process was applied to predict leatherback habitat suitability (probability of occurrence) for the study period as a function of dynamic and static environmental covariates. This model serves as the basis for dynamic ocean management and Ecological Risk Assessment from which outputs can inform managers and stakeholders as to appropriate management action that can reduce leatherback turtle bycatch while providing a modeling framework for analyzing fisheries observations for other data-limited vulnerable populations and species.