The Regional Atmospheric Modeling System (RAMS) at Colorado State University has been utilized for regional climate model simulations of the North American Monsoon System for the summer seasons of 1988 (U.S. drought), 1993 (Mid-west floods), and 1997 (strong El Nino). These control simulations agree well with large-scale reanalysis data with respect to the mid and upper-level height and wind fields as well as low-level moisture and circulation patterns. The model is fully capable of simulating burst and break periods of the monsoon as well as the low-level jets in the northern Gulf of California and Gulf of Mexico. The timing and location of modeled precipitation associated with surge periods is also consistent with daily rain gauge data over the U.S. and Mexico.

Among the simulated seasons, the most distinct variability that influences the timing and intensity of the monsoon seasons is the strength and positioning of the monsoon ridge and Pacific coast trough. The upper-level synoptic conditions strongly regulate the moisture and circulation patterns near the surface as well as the large scale vertical lift and subsidence fields. Specific positioning and strengthening of the monsoon ridge is necessary for an optimally strong and persistent monsoon season.

While the upper level conditions largely dictate the monsoon response, the surface conditions can feed back onto the larger scale and alter the mid to upper level geopotential heights and winds. Sensitivity tests with respect to SST and soil moisture were performed for each monsoon season (June-August) mentioned above. The SSTs and soil moisture were independently varied over the whole domain as well as locally over the core monsoon region so as to assess the potential for large and local scale responses depending upon the size of the imposed surface anomalies.

Generally, large (small) scale changes to the surface conditions resulted in large (small) scale changes in the surface fluxes, which subsequently, altered the seasonal precipitation patterns and magnitudes over Mexico and the U.S. While the local scale changes produced the greatest surface flux and precipitation response close to the anomaly, there were weaker responses well displaced from the anomaly. Changes to the mid-level height field were present for all tests but were rather small when compared to those experienced with the passage of a baroclinic system. Weak anomalies in the height field were still significant enough to partially contribute to variations in vertical motion and precipitation over the U.S.