This latest version of CloudCatcher builds on the success of a 2020 prototype, which engaged over 1,600 volunteers. That earlier project proved that non-experts could reliably identify clouds, with results published in the journal Weather. The updated platform includes enhanced tools to detect thin clouds and to confirm the clarity of images previously marked as cloud-free, using false-colour imaging.
Clouds significantly affect satellite temperature readings by obscuring the Earth's surface. To address this, satellites apply a 'cloud mask' to identify which pixels are cloud-covered. However, these masks are not flawless. CloudCatcher aims to highlight inaccuracies, helping researchers determine where these automated systems fall short.
Accessible via the Zooniverse platform, CloudCatcher presents users with satellite images from the Sea and Land Surface Temperature Radiometer (SLSTR), onboard the European Space Agency's Sentinel-3 satellites. Volunteers classify small image sections based on cloud presence. Designed for ease of use, the tool requires no technical expertise and is open to families, schools, and individuals alike.
The collected data will support the refinement of cloud masks used by SLSTR, aiding global efforts in climate monitoring and weather forecasting. Although more testing is needed to ensure minimal cloud interference, the contributions of CloudCatcher volunteers mark an important first step toward more reliable Earth observation from orbit.