New Publication: Using AI to Improve Marine Mammal Conservation

Philip T. Patton

January 14, 2025

Marine Mammal Research Program PhD Student Philip T. Patton has published an article in Conservation Biology with co-authors about the use of AI to improve marine mammal conservation.

Understanding animal populations is essential for conservation, but traditional methods like manual photo identification, that is, identifying the same animal in different images, can be costly and time-consuming. Our recent paper explores how artificial intelligence (AI) can streamline this process, making population assessments faster and more cost-effective. By using machine learning algorithms to analyze thousands of images of animals from the field, we can identify individual animals more quickly, reducing the need for labor-intensive human verification. This ultimately makes population assessments timelier, and more relevant  for wildlife managers and policymakers. 

AI helps us efficiently identify animals, which is essential to effective conservation decision-making. Image courtesy of Jochen Zaeschmar

However, automation comes with challenges. While AI can significantly lower costs, small errors in identification can impact population estimates. Our study tested different strategies for balancing automation with accuracy, ensuring that conservation efforts remain reliable. We found that for species with distinct markings, AI-driven identification performs well, but careful human oversight is still needed in cases where AI struggles with image quality or subtle differences between individuals.

Timely population assessments are critical for policymakers, who must make challenge decisions as marine mammals face complex threats like climate change.

By optimizing these AI-driven approaches, conservation organizations can allocate resources more efficiently, making it easier to monitor endangered species and track changes in their populations over time. This research is a step toward integrating cutting-edge technology with wildlife management, helping conservationists make data-driven decisions that will protect biodiversity for future generations.

Please see here for a link to the full text of the article: https://onlinelibrary.wiley.com/share/author/NESYNAJ6FJB7CICMQXAR?target=10.1111/cobi.14436