Coral Reef Health: Beauty's Place in Health Assessments

What is beauty? Can science quantify it? Can scientists provide numbers for what makes coral reefs attractive? One study recently did so, in order to assess coral-reef health.

That is, researchers from the areas of biology, mathematics and art history, from institutions including San Diego State University (SDSU), the Getty Research Institute and the Scripps Institution of Oceanography, have put together a tool that aims to computationally measure coral reefs' aesthetic appearance. They say that their findings show that visual cues generated randomly from photographic images can help in assessing beauty and health among coral reefs worldwide, according to a release.

The study involved bringing together and working with a list of 109 visual features-relative size, color, distribution of objects in an image, texture, color intensity, and others. All of these are things that can assess aesthetic appeal, noted the release.

Then the team generated a computer program that assesses elements of these images and can analyze at least 2,000 (random) photo images of coral reefs worldwide. For each of those reef ecosystems, the program turned out an aesthetic score, according to the release.

They published their findings recently in the journal PeerJ. As it turns out, there's a close alliance between healthy reef ecosystems--as evaluated by reef scientists--and those that scored well for aesthetic points in randomly generated photos.

"Our results suggest that our perception of aesthetics is well-aligned with healthy, thriving ecosystems," said Andreas Haas, at SDSU and lead author of the study, in the release.

Generating a number-driven idea of beauty was useful, the team said in their report, because it helps to provide a visual standard for reef health. "By quantifying aesthetic features of coral reef systems, this method provides a cost effective tool that also targets one of the most important socioeconomic values of coral reefs -- their natural beauty," Haas said in the release.

You can learn more about the team's project at their website, here.