Water is critical for supporting population and industry and for maintaining agriculture and energy supplies that are traded across regions. Uncertain future climate and socioeconomic changes will have significant impacts on water resource availability. However, characterizing future water availability is difficult given the complex and uncertain drivers of future water supply and demand. We address this challenge by characterizing global variability in runoff at the river basin scale and by leveraging integrated modeling to understand how this variability propagates across energy, water, and agricultural systems. We apply stochastic watershed modeling techniques to generate a large ensemble of alternative future time series of annual runoff for 235 river basins globally. We run these alternative hydrologic futures through the Global Change Analysis Model, an integrated human-earth system model that represents the interactions between the socioeconomic, climate, land, energy, and water sectors. We explore the impacts of varying runoff availability on agricultural productivity, biomass production, and water withdrawals. We find that impacts of hydrologic variability vary in magnitude across basins and the metrics considered. Impacts are observed most strongly for the water sector, particularly in water scarce basins. Most basins are able to absorb variability in runoff in the agricultural and energy sectors by shifting their crop distributions and relying more on groundwater, when available. This work contributes a novel exploration of how future hydrologic variability propagates across environmental systems and spatial scales. Our findings can provide valuable insights into the extent and impacts of uncertain water availability in the future, which could inform future regional resource management analyses and discussions.