Early Climate Impact Assessments Targeted 2025—How Did They Perform for Global Water Resources?

ABSTRACT

Climate impact studies from the 1990s produced projections for global water resources extending to 2025. We evaluate the first two global water resource assessments and the IPCC's 2001 statements on future water resources against subsequent observations. Temperature projections were broadly consistent with what has since been observed. The studies projected a small increase in global precipitation, with increases at high latitudes and decreases in the subtropics; this pattern is consistent with present‐day model projections, though robust observed changes have yet to emerge. Runoff projections for 2025 are hard to assess given limited data and multi‐year variability, but warming effects on snowmelt‐dominated streamflow are consistent with observations. Population projections matched (8.23 billion observed versus 8.01–8.06 billion projected), whereas water demand projections overestimated the increase in water withdrawals (+13% observed versus +19%–52% projected). The IPCC's 2001 projections of impacts on water resources were consistent with its 2021/2022 assessment. These early studies proved influential in research and policy, initiating a proliferation of global water modeling and climate impact assessments. We discuss how the scope of such assessments has evolved. Despite improvements, challenges remain, including uncertainty in runoff and water use data, alongside scope for better water stress indicators and use of earth observation. There is also potential to review early impact projections at sub‐global scales and for other sectors. Progress on water security has been checkered, and improved modeling alone has not been sufficient to drive political change. Meeting the scale and urgency of the challenge will require going beyond model refinement, attending to how risks are framed and identifying other levers to stimulate political will and investment.