The power of artificial intelligence has been harnessed to enable monitoring of coral reef health from space. The remote sensing and analysis system leverages advances in computer vision techniques, deep learning and satellite imaging technology to quickly and accurately identify and measure reef halos.

The presence of these ring-like patterns of bare sand that occur around coral patch reefs is readily visible from satellite images. Ring halos are considered important indicators of the health and vitality of coral reefs, and their measurement has been a time-consuming process — until now.

Scientists from the University of Hawaii and Planet Labs (California) have developed a set of deep learning algorithms capable of taking into account the diversity of ring halo patterns globally and identify and measure halos with surprising accuracy on a global scale. Research published in Remote Sensing of Environment explains how a deep learning framework was trained and tested through the use of Planet SkySat images to detect halo presence globally and measure their size.

A freely accessible web app is now under development to enable conservationists and resource managers to monitor reef health remotely, rapidly and cost-effectively, using satellite or drone imagery.