| There is a growing consensus that a statistically significant, but relatively small, increase in global temperatures is evident from observational data analysis (Karl 1996). Simple physical and thermodynamic based reasoning suggests that a rise in temperature is likely to enhance the hydrologic cycle. Under a warmer climate scenario, the relative humidity of the air is likely to remain relatively unchanged. The laws of atmospheric physics then imply that tropospheric specific humidity would increase exponentially as temperature rises. Increased surface temperatures would also cause a mismatch in the surface energy budget that would, at least locally, be balanced by an increase in evaporation. A simultaneous increase in evaporation and the water holding capacity of the atmosphere suggests a global enhancement of precipitation, but with significant local and regional scale variability.; In this study, we present a systematic investigation of the spatial and temporal trends and variations in precipitation over the United States during the 20th century. Our results suggest that precipitation and streamflow have increased significantly across the United States during this period, with the largest and most spatially coherent changes taking place in the fall. This large increase in the national average precipitation is due to a large increase in the frequency of rainfall that is accompanied by a decrease in the average intensity. We provide further evidence that fall is the only season with large changes in hydroclimate in the United States by linking increases in the low flows, which typically occur in the fall, to increases in fall precipitation. A multivariate spectral analysis of fall precipitation suggests that the large national increase in precipitation is due to the presence of spatially correlated variations at decadal and secular scales across the interior of the country that increased dramatically after 1940. The observed low frequency variability can be linked to variations in sea surface temperatures in the Pacific and Indian Oceans. The results of a series of synthetic experiments suggest that the large spatially coherent decadal variability observed during fall can be linked to decadal changes in the frequency of days with precipitation. |
