A decision tree algorithm was employed to refine Cropland Data Layer (CDL) using spatial and temporal information. The Refined Cropland Data Layer (R-CDL) could be used as an alternative to researchers as it provides more accurate cropland information. Annual RCDL maps were produced for the contiguous United States from the year 2017 to 2021. To explore the data online and access the web-based services, please visit the project homepage: https://cloud.csiss.gmu.edu/icrop/ To read our full paper on RCDL, please visit: https://www.nature.com/articles/s41597-022-01169-w Cite this article: Lin, L., Di, L., Zhang, C. et al. Validation and refinement of cropland data layer using a spatial-temporal decision tree algorithm. Sci Data 9, 63 (2022). https://doi.org/10.1038/s41597-022-01169-w Note: 2020 RCDL was reproduced with the Re-released CDL on February 1, 2022. More information could be found at: https://www.nass.usda.gov/Research_and_Science/Cropland/SARS1a.php 2021 RCDL was released with the official release of 2021 CDL on February 14, 2022 2022 RCDL was released on March 24, 2022

{"references": ["Lin, L., Di, L., Zhang, C.\u00a0et al.\u00a0Validation and refinement of cropland data layer using a spatial-temporal decision tree algorithm.\u00a0Sci Data\u00a09,\u00a063 (2022). https://doi.org/10.1038/s41597-022-01169-w"]}