Using artificial intelligence to save coral reefs

q u i 0G01UI1MQhg unsplash

Coral reefs are some of the most diverse ecosystems in the world, sometimes called the “rainforests of the sea.” Coral reefs are a vital part of marine life, act as guards against forces of nature and are a key source of livelihood for millions. Unfortunately, these diverse habitats are rapidly degrading. Here’s how advanced technologies such as Artificial Intelligence (AI) are being used to save coral reefs.

Coral reefs: rainforests of the sea

Coral reefs are natural protectors for coastlines against storms and erosion. Healthy coral can help prevent loss of life and property from waves, floods and storms by absorbing 97% of the energy. Tropical coral reefs cover a mere 0.1% of the ocean, but an estimated 25% of marine life depends on these diverse ecosystems.

According to the Economics of Ecosystems and Biodiversity (TEEB) study, “approximately 850 million people live within 100 km of and derive some benefits from coral reefs, with at least 275 million depending directly on reefs for livelihoods and sustenance.” In the U.S., the total economic value of coral reef services including fisheries, tourism, and coastal protection—is over $3.4 billion each year.

Coral reefs are endangered. According to the United Nations Environment Programme, between 25 and 50% of the world’s live coral has been lost in the last 30 years. Coral reefs face many threats—natural (such as diseases, predators, and storms) as well as from human activities. Some of them involve physical damage from coastal development, dredging, quarrying, and boat anchors. Additionally, there is a cascading effect due to activities such as overfishing and coral harvesting which alter the structures of marine life. The United States Environmental Protection Agency lists the “increased ocean temperatures and changing ocean chemistry (ocean acidification)” as global threats to coral reef ecosystems. The aggregate effect of these factors together is placing pressure on 75% of coral reefs globally.

This calls for timely measures to adequately protect coral reefs from further damage.

AI to the rescue?

A crucial part of the effort towards protecting coral reefs include mapping, monitoring, analyzing, and restoring them. The process of collecting data for reef monitoring has traditionally been done by divers by capturing video footage and pictures. However, many times, divers have to interfere with their surroundings and it unintentionally affects the results of the survey. In addition, there are practical restrictions such as time spent under water. Overall, there are many gaps which exist in these approaches, be it in terms of in mapping coral reefs or real-time monitoring. Some of these gaps can be plugged in by deploying AI and machine learning.

An AI-powered solution by Accenture, Intel (INTC), and the Sulubaaï Environmental Foundation—to monitor, characterize, and analyze coral reef resiliency dubbed as project CORaiL, was announced in April 2020. The teams from Accenture, Intel, and Sulubaaï implemented an artificial concrete reef—called a Sulu-Reef Prosthesis (SRP)—to provide support for unstable coral fragments underwater. In addition to the fragments of living coral which were planted on the SRP, intelligent underwater video cameras, equipped with the Accenture Applied Intelligence Video Analytics Services Platform (VASP), were planted to detect and photograph fish in the reef surrounding the Pangatalan Island in the Philippines.

VASP uses AI, powered by Intel Xeon, Intel FPGA Programmable Acceleration Cards, and Intel Movidius VPU to count and classify the marine life. The data collected is structured using analytics enabling researchers to make data-driven decisions that will help the reef progress. The teams are now working on infrared cameras, which enable nighttime video capture to create a complete picture of the coral ecosystem.

In October 2020, the researchers from the Arizona State University Center for Global Discovery and Conservation Science have generated a global coral reef extent map using a single methodology capable of predicting the location of shallow coral reefs with nearly 90% accuracy. The researchers at ASU have further developed a new airborne mapping approach. “By combining laser-guided imaging spectroscopy and artificial intelligence, the new approach reveals unprecedented views of coral reefs below the ocean surface. The maps show where live corals persist as well as areas of degraded reef,” according to the study.

Overall, many research papers see AI and machine learning applications as a promising approach to eliminating the bottleneck in data processing for coral reef monitoring. The use of these technologies can enhance the speed of processes at a much-reduced form, and dramatically improve how coral reefs are measured and monitored worldwide.

Prableen Bajpai