NASA is calling on citizen scientists to help identify and classify the world's corals by playing a virtual diving game, so that experts can better understand how they are evolving and how they might be best preserved for the future.
What you need is the NASA NeMO-Net game for iOS and iPadOS (macOS and Android coming soon). It'll take you on a series of virtual dives in the ocean, and your job is to identify the coral that you come across. The computer-generated underwater environments are based on data collected over the last few years by NASA's Ames Research Center in California. The team there has been using fluid-lensing cameras to map out the ocean in greater detail than ever before.
Originally developed to give astronomers on the ground a view of the stars that's free from atmospheric distortion, these cameras can also be applied to map the seafloor through the distortions of water. Mounted on drones or aircraft to carry out their mapping, these devices will eventually be fixed to satellites to carry on their monitoring.
However, as advanced as they are, these cameras don't reveal the whole pictures about the corals sitting under the waves – and that's where you come in. By touring around and identifying the types of coral you see in the game, and exactly where they are, you can help join some of the dots in the data that NASA has collected.
"NeMO-Net leverages the most powerful force on this planet: not a fancy camera or a supercomputer, but people," says Ved Chirayath Chirayath, from the research center. "Anyone, even a first grader, can play this game and sort through these data to help us map one of the most beautiful forms of life we know of."
The game is educational as well, teaching you about the types of coral in the world's oceans and about the conservation efforts to protect and restore them. You can earn badges as you go along to keep you interested in more dives.
Everything that gets submitted by users will then be processed by the Pleiades supercomputer at the Ames Research Center, training it how to recognize different types of coral based on the raw data. As with any other neural network, as it gets trained it should improve over time, so it'll eventually be able to spot coral types on its own, even with lower quality data. The more people that sign up to use NeMO-Net – which stands for Neural Multi-Modal Observation and Training Network – the smarter the system will become.
With coral reefs under threat from warming waters, increasing ocean acidification and human pollution, every little helps in keeping these ecosystems – and the marine life that rely on them – flourishing in the future.
Source: NASA
