One anticipated impact of global increase in CO2 is a shift in patterns of precipitation and temperature. This shift may affect water resources, and, in turn, agricultural productivity (Gleick,1989). General Circulation Models (GCMs) have been used to study changes in global climatology resulting from increased concentrations of greenhouse gases. However, GCMs do not have sufficient spatial or temporal resolution to meet the demands of growing regional-scale research efforts. As a result, numerous techniques have been employed to scale down GCM output to regional scale (Georgi and Mearns,1991). In this study, an investigation of potential impacts of altered climate on the water resources of the Upper Brazos watershed in Texas has been performed.
Observations of daily temperature, precipitation, and other relevant variables were obtained from the National Climatic Data Center for the study area. Streamflow data were obtained for the Brazos river at Seymour, Texas, from the U.S. Geological Survey. 500hPa data were acquired from the National Center for Atmospheric Research gridded NMC archive for the period 1946 - 1994. Output from 30 year runs of the CSIRO Mark 2 GCM (Commonwealth Scientific and Industrial Research Organisation - Australia) were obtained for 1xCO2,and 2xCO2 scenarios.
Circulation pattern types were catalogued from the NCAR NMC gridded 500hPa database for the western two-thirds of the U.S. using a Principal Components Analysis (PCA) coupled to a k-means procedure (Matyasovsky, et al.,1993). Each day of record was assigned to a pattern type, and a departure of heights from the category mean was calculated. It was then possible to make a correlation between height anomalies and surface climate anomalies.
Daily outout from the two scenarios of the GCM were sorted into the same pattern types found from the climate pattern analysis. Based on the correlations previously found, projections of temperature and precipitation, under stressed CO2 conditions were made. These projections were then input to the watershed hydrological model.
The watershed was modeled with the USGS Modular Modeling System. Marginal shifts of the meteorological input altered the water resources for the Upper Brazos. As a consequence, this rich agricultural area would require careful reallocation of occasionally uncertain water supplies.