LIHU‘E — If AI ever rises up, at least the reefs will be safe.
Researchers at the University of Hawai‘i at Manoa have developed new artificial intelligence technology to aid in the conservation of coral reefs off Hawai‘i’s coasts and beyond.
By developing a series of deep learning algorithms, coral ecologists in the UH Manoa School of Ocean and Earth Science and Technology are now able to identify and measure reef halos from space.
Also known as “grazing halos” or “sand halos,” these geographical features consist of ring-like patterns of bare sand surrounding coral patch reefs. The “halos” form when fish and invertebrates leave the reef to forage for seagrass, and researchers can measure changes in their size and presence to study changes in the reefs themselves.
“Reef halos may be important indicators of the health and vitality of coral reefs, but until now, their measurement and tracking has been a challenging and time-consuming process,” said Simone Franceschini, lead author of the study and postdoctoral research fellow at SOEST’s Hawai‘i Institute of Marine Biology.
“However, with this new method, we can accurately identify and measure reef halos on a global scale in a tiny fraction of the time it would take a human being to accomplish the same task.”
Coral reefs are one of the most diverse ecosystems on the planet. Despite occupying less than 1 percent of Earth’s ocean floor, at least 25 percent of all marine species call reefs home, including fish, crustaceans and mollusks. Additionally, the reefs serve several direct benefits to humans, from tourism to fishing and shoreline protection, with a global economic value estimated to be in the trillions of dollars.
However, factors such as climate change, overfishing and pollution have decimated the planet’s reefs in recent decades. In 2020, UH Manoa scientists estimated between 70 and 90 percent of all coral reefs will disappear within 20 years.
Without knowing changes in the ecosystem, researchers are unable to appropriately react and adapt as environmental issues arise. Ecologists have increasingly relied on image analysis and satellite imaging technology in recent years to aid in the process, with many researchers embracing AI technology to increase the efficiency of their work.
“Reef halos are sometimes very clear in satellite imagery, with distinct edges and high contrast with background vegetation, but sometimes they are quite faint and hard to distinguish, even by a highly trained observer,” Franceschini said.
“In the end, our team was able to develop a set of algorithms capable of taking into account the diversity of these patterns globally and measure halos with surprising accuracy. It is hugely satisfying for us to now have built something that can accurately identify more than 90 percent of halos in some parts of the world,” she said.
In the near future, the team is aiming to develop a free web app that can allow conservation practitioners, scientists and resource managers to remotely, quickly and inexpensively monitor aspects of reef health through the use of satellite or drone imagery.
“This breakthrough is a key step in scaling up, in both space and time, our ability to monitor and quantify aspects of coral reef ecosystem health,” said Elizabeth Madin, an associate research professor at the Hawai‘i Institute of Marine Biology and the study’s senior author.
“By providing a more efficient and effective way to measure coral patch reefs and their surrounding halos, this new method paves the way for the development of a global-scale reef conservation and monitoring tool based on the phenomenon of reef halos.”
•••
Jackson Healy, reporter, can be reached at 808-647-4966 or jhealy@thegardenisland.com.