Modeling the Impacts of Climate Change, Landuse Change, and Human Population Dynamics on Water Availability and Demands in the Southeastern U.S.

Abstract

Projected changes in population, land use and land cover, and climate could negatively impact the regional water resources in the southeastern U.S. The objective of this study is to develop a method to fully budget annual water availability for water supply (Precipitation – Evapotranspiration (ET) + Groundwater supply + Return Flow) and water use from thermoelectric, irrigation, domestic, industry, livestock, mining, and commercial uses. We used a generalized annual ET model that estimates water loss as a function of potential ET, annual precipitation, land cover type, and topography. Both the groundwater supply and return flow rates were derived from USGS historical databases. Water uses for the domestic and irrigation sectors for the future were projected using empirical models derived from historical USGS databases. The Water Supply Stress Index (WSSI) as the ratio of water demand and supply was developed to evaluate water stress conditions. The Water Supply Stress Index Ratio (WSSIR) was developed to quantify the impact of future changes in climate, land use, and population individually or in combination. Modeling results from two Global Circulation Models (GCMs) (Hadley and CGC1), one land use change model, and one population change model were integrated to project future water supply and use over the next 25 years. All model runs were performed at the 8-digit USGS Hydrologic Unit Code (HUC) level across the 13 southern states. We found that population increase will greatly increase water use in metropolitan areas, but overall its impact on total water use may not be large. Predicted future land use changes (i.e., urbanization) will have little positive effect on the water supply-water use relationship. In contrast to population and land use change, climate change that has large uncertainty have the most pronounced effect on regional water supply and demand.