Annual maize yield maps for the US Corn Belt for 1999-2018, aggregated to 10 km resolution. Yield maps are produced with the SCYM methodology at 30 m resolution using Landsat satellite data. The original high resolution yield maps are aggregated to a 10 km grid resolution to preserve privacy. Dataset creation is described in Deines et al., 2020 in Remote Sensing of Environment (open access). The 2008-2018 map dataset described therein was extended back to 1999 for Lobell et al. 2020 in Nature Food (read-only accessible version). Maps for 2008-2018 use the USDA's Cropland Data Layers to identify maize pixels; maps for 1999-2007 use the Corn-Soy Data Layer produced by Wang et al. 2000 (data + manuscript are open access). Metadata Yield values are provided in metric tons/hectare. No Data value = -999 (grid cells outside of the study area + cells with <1% coverage of SCYM 30 m yield maps) Map projection: EPSG:5070, CONUS Albers Equal Area Preferred citation(s): Deines, J.M., R. Patel, S. Lang, W. Dado, & D.B. Lobell. 2020. A million kernels of truth: Insights in to scalable satellite maize yield mapping and yield gap analysis from an extensive ground dataset in the US Corn Belt. Remote Sensing of Environment. https://doi.org/10.1016/j.rse.2020.112174 D.B. Lobell, J.M. Deines, & S. Di Tomasso. 2020. Changes in the drought sensitivity of U.S. maize yields. Nature Food 1:729-735. https://doi.org/10.1038/s43016-020-00165-w