Citizen scientists asked to identify clouds in satellite data for climate research

Written by Copernical Team Friday, 23 May 2025 12:57

font size decrease font size increase font size
Print
Email

London, UK (SPX) May 22, 2025

CloudCatcher, a citizen science program developed by STFC RAL Space, is inviting the public to help improve the accuracy of cloud detection in satellite imagery. The project plays a vital role in enhancing how scientists measure Earth's surface temperatures from space by verifying when clouds obstruct the satellite's view. This latest version of CloudCatcher builds on the success of a 2020

Citizen scientists asked to identify clouds in satellite data for climate research
by Sophie Jenkins
London, UK (SPX) May 22, 2025

CloudCatcher, a citizen science program developed by STFC RAL Space, is inviting the public to help improve the accuracy of cloud detection in satellite imagery. The project plays a vital role in enhancing how scientists measure Earth's surface temperatures from space by verifying when clouds obstruct the satellite's view.

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.

Research Report:CloudCatcher: citizen science can be used to help to crowd source a satellite cloud validation dataset