New publication in Nature Communications introduces GPSat, a tool that helps process constantly changing satellite data, more quickly and efficiently than older methods. This tool can help scientists better monitor changes in sea ice over time and help improve predictions for sea-level changes.
GPSat can produce detailed maps of Arctic sea ice, filling in any gaps in the satellite data and can produce data more than 500 times faster than older methods while maintaining accuracy (demonstrating less than 4 mm difference on the derived freeboards on average).
The paper, ‘Scalable interpolation of satellite altimetry data with probabilistic machine learning’ authored by William Gregory (Princeton) and involving CPOM co-authors Isobel Lawrence (ESA), Carmen Nab (UCL) and Michel Tsamados (UCL), was published on 28 August 2024 in Nature Communications.
https://www.nature.com/articles/s41467-024-51900-x#citeas
Author information:
Gregory, W., MacEachern, R., Takao, S. et al. Scalable interpolation of satellite altimetry data with probabilistic machine learning. Nat Commun 15, 7453 (2024). https://doi.org/10.1038/s41467-024-51900-x