Brought on by anthropogenic actions, accelerated soil erosion inflicts extreme changes in terrestrial and aquatic ecosystems. These field-scale (30 m) changes have neither been fully surveyed in the present, nor predicted for a probable future. Water-driven soil erosion (i.e., sheet and rill erosion) rates across the contiguous United States were estimated for the present, and then predicted for the future using three alternative Shared Socioeconomic Pathway and Representative Concentration Pathway (SSP-RCP) scenarios (2.6, 4.5, and 8.5) of the Coupled Model Intercomparison Project Phase 6 (CMIP6). The G2 erosion model which is integrated with Machine Learning (ML) and Remote Sensing (RS) techniques were used to estimate soil erosion based on gauge observations of long-term precipitation, and climate and land use land cover (LULC) scenarios. The baseline model (2020) estimated soil erosion rates of 2.32 Mg ha−1 yr−1 under current conservation agriculture practices (CPs). Maintaining current CPs, future scenarios predict an 8 % to 21 % increase in soil erosion under different combinations of SSP-RCP climate and LULC change scenarios. The findings of this study can help policy makers for future conservation planning on maintaining soil fertility, mitigating environmental impacts, and promoting food security. This reprository provide the soil erosion maps for united states in present and future.