Arctic Retrogressive Thaw Slumps (ARTS): digitisations of pan-Arctic retrogressive thaw slumps, 1985-2021

The Arctic permafrost is undergoing abrupt thawing which positively feedback to climate warming. Retrogressive thaw slumps (RTS) are one of the most disruptive and rapid abrupt thawing events yet not well studied, primarily because the lack of pan-Arctic mapping and automated mapping with deep learning methods requires a large amount of manual digitisations. This data set contains manually digitised polygons of RTS across the Arctic sourced from 25 standalone data sets focusing on different regions of the Arctic (e.g. Siberia of Russia, Northwest territory of Canada, Svalbard of Norway etc). The source data sets were compiled into a uniform data set and added important metadata to provide information about each digitisation. Unique identifiers generated by UUID5 were used to index individual RTS instances. Due to the highly dynamic nature of RTS, frequent changes and additions to the data set are expected, therefore we used Git to track the changes made to the data set in the future and hosted the live version on GitHub. The data set can be used in various of Arctic-related studies, including generating input images for deep-learning model training. Cite the paper: Yang, Y., Rodenhizer, H., Rogers, B.M. et al. A Collaborative and Scalable Geospatial Data Set for Arctic Retrogressive Thaw Slumps with Data Standards. Sci Data 12, 18 (2025). https://doi.org/10.1038/s41597-025-04372-7