To test this hypothesis, we selected the heavy monsoonal rains in Arizona (AZ) in early July 1999 as a case study for our numerical experiments. The non-hydrostatic NCAR/Penn State Mesoscale Model Version 5 (MM5) was used in this simulation of the monsoon onset in AZ. The MM5 accommodates four-dimensional data assimilation, multi-nested domains and other physical parameterizations, which allow us to stimulate the general features of the boundary layer during the monsoon period. Our work focuses on two-way nested simulations, including the GOC, exploring the idea that the SSTs in the N. GOC exceeding about 29 C are needed for heavy rainfall and the favorable changes in the boundary layer and the circulation that lead to it. Although SSTs are normally fixed (from observations), a do loop for incrementing SSTs is implemented into one of the MM5 decks, such that SSTs in different regions of the GOC could be modified. MM5 simulations were performed for the period of July 5-8, 1999, beginning and ending at 00 UTC. The empirical study revealed a rapid increase in rainfall rate over Arizona - New Mexico when Northern GOC SSTs exceeded 29 C.
Similarly, MM5 modeling results show a rapid increase in precipitable water over the northern GOC when SSTs increase from 29 C to 30 C. This is due to a dialation of the marine boundary layer, apparently resulting from buoyancy driven updrafts prior to and around sunrise, where buoyancy is derived from warmer SSTs and higher mixing ratios of water vapor. When northern GOC SSTs were 29 C or cooler, predicted updrafts over AZ reached the 500 mb level, but reached 200 mb with a doubling in updraft velocity when SSTs were 30 C. Both the tropospheric water content over the GOC and the circulation changed dramatically over AZ when the northern GOC SST increased from 29 to 30 C.
Supplementary URL: