Anticipating Illegal Maritime Activities from Anomalous Multiscale Fleet Behaviors

Abstract: Illegal fishing is prevalent throughout the world and heavily impacts the health of our oceans, the sustainability and profitability of fisheries, and even acts to destabilize geopolitical relations. To achieve the United Nations' Sustainable Development Goal of "Life Below Water", our ability to detect and predict illegal fishing must improve. Recent advances have been made through the use of vessel location data, however, most analyses to date focus on anomalous spatial behaviors of vessels one at a time. To improve predictions, we develop a method inspired by complex systems theory to monitor the anomalous multi-scale behavior of whole fleets as they respond to nearby illegal activities. Specifically, we analyze changes in the multiscale geospatial organization of fishing fleets operating on the Patagonia Shelf, an important fishing region with chronic exposure to illegal fishing. We show that legally operating (and visible) vessels respond anomalously to nearby illegal activities (by vessels that are difficult to detect). Indeed, precursor behaviors are identified, suggesting a path towards pre-empting illegal activities. This approach offers a promising step towards a global system for detecting, predicting and deterring illegal activities at sea in near real-time. Doing so will be a big step forward to achieving sustainable life underwater.