Earth and Atmospheric Sciences, Department of
Date of this Version
2007
Citation
JOURNAL OF CLIMATE VOLUME 20 4702-4716
DOI: 10.1175/JCLI4270.1
Abstract
Previous studies have identified several major causes for summer rainfall variations over the southwest United States, for example, land memory (i.e., relationships between antecedent winter season precipitation and snow cover anomalies and subsequent summer rainfall anomalies over the southwest United States; these anomalies are likely most important in the northwest United States, although antecedent anomalies in the southwest United States also may be important in determining summer rainfall variations) and sea surface temperature (SST) anomalies in the North Pacific. Atmospheric responses to these “boundary forces” interact with moisture flows from the Gulf of Mexico and from the Gulf of California to influence the rainfall in the Southwest. The land memory and the SST effects were further found to be “naturally separated,” in the sense that they each played a dominant role influencing the monsoon rainfall variation during different periods of the last century. This separation was also manifested by different dominant low-level moisture transport anomalies in those periods. Several new questions have arisen from these findings: How have the land memory and the SST effects been “separated,” so as to affect the monsoon rainfall variations during different periods, or “regimes”? And, what are the corresponding changes of low-level flows, and hence moisture transports into the southwest United States that help achieve the land memory or the SST effects on the rainfall variations during these different regimes? These questions, and related issues, are addressed using a numerical model of regional climate. The model was used to simulate 14 individual warm seasons (April–October) in each of the postulated regimes. Analyses of the simulation results showed systematic and significant changes in atmospheric circulation anomalies between the two regimes. In the early regime (1961–90), when the land memory effect was strong, the average geopotential height was lower and storm activity was more intense over the central and western United States than in the more recent regime (from 1990 on), indicating reduced eddy energy and momentum exchanges between high and low latitudes in the western United States. The effects of these changes on the monsoon rainfall were achieved by very different low-level flow and moisture transport anomalies. In the earlier regime, low-level flow and moisture transport anomalies in the southwest United States were primarily due to easterlies and southeasterlies into the Southwest for its wet monsoon conditions, with reversed anomalies for dry conditions. In the recent regime, these anomalies changed, with primarily southerlies and south- westerlies from the Gulf of California into the Southwest during its wet monsoon conditions, and reversed flow anomalies for dry conditions. These changes indicate that different physical processes, including those responsible for the planetary-scale atmospheric circulation, led to monsoon rainfall variations during each of these regimes.
Comments
2007 American Meteorological Society