Evaluating IPCC and Other Scenarios of Future Climate for Regional-to-Local Water Resource Applications in the US

English Abstract:

Assessments of projected climate change impacts to water resources at regional-to-local scales typically use a combination of numerical models for evaluating the various differences across projected scenarios and impacts. However, the sets of approaches for projecting regional climate and its possible effects - statistical or dynamical models, e.g., to generate the regional-to-local climate and weather - together with the hydrologic models used to transfer changes in these model outputs into changed hydrologic processes most often include only one or a few models, and thisinhibits their depiction of the uncertainty space around the projected climate and impacts. Additionally, inter-model differences are often not constrained to define the uncertainties in the simulations consistently. And more generally, the community performing water-resource impacts assessments (among most others) has focused very intently on characterizing uncertainty across climate projections produced by multiple climate models running several emission scenarios, but have done little work to understand the interacting uncertainties in the driving climate and hydrologic models.

Water-resource managers face a linked series of questions about using these models related to their decision-making for climate-changed futures, including: 1) how do different climate downscaling approaches affect the projection of relevant hydrometeorological variables? 2) how can observational datasets best be used to drive analyses of downscaled climate and hydrology? and 3) how many/which hydrologic model/s can be used, and how can they best be configured and calibrated for these applications?

This talk will summarize some of the methods and results coming from the US Army Corps of Engineers' work with partners to help provide fuller answers to those questions. Two salient points are: 1) different methods for producing gridded meteorological fields can have very different effects on projected hydrologic outcomes, with uncertainties as large as the climate change signal; 2) many popular statistical downscaling methods produce hydroclimate representations with too much drizzle, wrong extreme events representations, and improper spatial scaling characteristics relevant to hydrology.