Sensitivity of Soil Moisture to Doubling of Carbon Dioxide in Climate Model Experiments. Part II: The Asian Monsoon Region

Abstract

This is a sequel to a comparative study of the performance of five climate models over North America; here we investigate their performance over the region most directly under the influence of the Asian monsoon (taken to be equator to 60°N, 60° to 140°E). Emphasis is on soil moisture, since it is a difficult parameter to calculate and therefore provides a severe test of the climate models. Furthermore, soil moisture is a quantity of considerable concern to agriculture and the management of ecological systems generally. The five models compared were those of GFDL, GISS, NCAR, OSU and UKMO. While there were many significant differences in detail between the soil moisture distributions produced by the various control runs for winter and for summer, and also between the model climates and the observed climate of the region, the largest scale features were generally simulated quite well. These features are a considerably drier average climate in the northern part of our region compared to the southern part, with the season of highest soil moisture occurring in the mid-to-late winter in the northern part and the mid-to-late summer in the southern part. We then compared the results of five climate model sensitivity experiments in which the carbon dioxide content of the atmosphere was doubled, resulting in a warmer average temperature for the Earth. In a majority of the five cases there was apparently an intensification of the seasonally changing monster circulation, since in the southern part (most of India and Southeast Asia) there was a change toward drier winters and wetter summers. In the northern part of the region the reverse was generally true, with the magnitude of the summertime drying trend being most pronounced in the area around Manchuria. These changes are in rough agreement with evidence for the modified rainfall regimes existing during the Altithermal Period, when it was warmer in summer. These conclusions must, of course, be taken with great caution until they can be checked with improved climate models.