Detecting illegal maritime activities from anomalous multiscale fleet behaviours

Abstract

To achieve sustainable use of our oceans globally, our ability to detect and even predict illegal maritime activities must improve. The challenge is that most vessels that commit illegal acts will not transmit their location and are in effect unobservable to enforcement agencies. To overcome this challenge, we have developed a method for inferring the location and timing of illegal maritime activities inspired by complex systems theory: by monitoring the multiscale spatial behaviour of those vessels that we can observe, we are able to monitor for anomalous spatial patterns that could be associated with the presence of illegal activities (committed by an unobserved vessel). These spatial anomalies are represented as distortions in the ‘shape’ or multiscale spatial organization of fleets, which can be quantified using methods from information theory. To explore this approach, we developed a spatially explicit agent‐based model of fishing vessels, where we examined the utility of several multiscale spatial anomaly indexes. We then examined the performance of these indexes in a real‐world test‐case, with a focus on illegal events that occurred on the Patagonia Shelf. Our results suggest that this method offers insight into the location and timing of illegal activities at sea, and consequently offers a promising tool for helping to improve the security and sustainability of our oceans.

Document Details

Document Type
Pub Defense Publication
Publication Date
May 03, 2022
Source ID
10.1111/faf.12669

Entities

People

  • A. John Woodill
  • J R Watson

Organizations

  • Defense Advanced Research Projects Agency
  • National Aeronautics and Space Administration
  • Oregon State University

Tags

Fields of Study

  • Environmental science

Readers

  • Distributed Systems and Data Platform Development
  • Maritime Security/Maritime Homeland Security
  • Theoretical Analysis.

Technology Areas

  • AI & ML