Locating and measuring marine aquaculture production from space: a computer vision approach in the French Mediterranean

TL;DR

This paper presents a scalable computer vision method to identify and measure marine aquaculture cages from satellite imagery, improving monitoring and data accuracy for the industry in the French Mediterranean.

Contribution

It introduces a novel, adaptable approach using remote sensing and machine learning to accurately locate and quantify aquaculture production at large scales.

Findings

Identified 4,010 cages in French Mediterranean from 2000-2021.

Demonstrated cost-effective, scalable monitoring of aquaculture from satellite imagery.

Provided a framework to estimate production and quantify uncertainty.

Abstract

Aquaculture production -- the cultivation of aquatic plants and animals -- has grown rapidly since the 1990s, but sparse, self-reported and aggregate production data limits the effective understanding and monitoring of the industry's trends and potential risks. Building on a manual survey of aquaculture production from remote sensing imagery, we train a computer vision model to identify marine aquaculture cages from aerial and satellite imagery, and generate a spatially explicit dataset of finfish production locations in the French Mediterranean from 2000-2021 that includes 4,010 cages (69m2 average cage area). We demonstrate the value of our method as an easily adaptable, cost-effective approach that can improve the speed and reliability of aquaculture surveys, and enables downstream analyses relevant to researchers and regulators. We illustrate its use to compute independent estimates of production, and develop a flexible framework to quantify uncertainty in these estimates. Overall, our study presents an efficient, scalable and highly adaptable method for monitoring aquaculture production from remote sensing imagery.