While C4 plants occupy a smaller area of the global land surface, they are typically more efficient water users and often adapted to warmer and dryer conditions compared to most C3 plants. Thus, accurate representation of C3 and C4 biogeography, especially in grasses where C4 photosynthesis is most common, is important for Earth system models (ESMs) to predict carbon and water fluxes. However, the spatiotemporal coverage of C3 and C4 plants predicted by or inputted into different ESMs varies widely. Here, we developed 5-arcmin maps of C3 and C4 crops and grasses over the contiguous U.S. during 2000-2019. C3 and C4 crop maps were generated by harmonizing a national inventory of harvested crop area and Crop Land Data Layer (CDL) maps. During 2000-2019, mean areas of C3 and C4 crops were estimated to be 81 and 35 million hectares (Mha), respectively. Our data show striking expansions in C4 crop area – principally due to additional corn planting after 2006, while the C3 crop area shrank during this time period, primarily associated with the abandonment of wheat and hay fields. Maps of grasses were constructed by combining the MODIS Vegetation Continuous Fields (MOD44B.051 VCF) non-tree cover product and a cross-walking table to partition shrubs from herbaceous cover. Then, the physiologically based crossover temperature model was used to identify areas where C3 or C4 grasses should predominate. The estimated C3 and C4 grass cover areas were on average 138±7 and 54±7 Mha, respectively, during the study period. Meanwhile, we observed small expansions in C4 grass areas after 2009 but relatively constant C3 grass areas. Our new data products can serve as inputs for ESMs and have the potential to improve estimates of carbon and water exchanges from C3 and C4 vegetation across the contiguous U.S.