Use of cumulus parameterization and explicit microphysics for climate studies over the Rio Grande basin

As part of a multidisciplinary effort to understand the hydrologic cycle at regional scales for the Rio Grande Basin, we are currently undertaking numerical simulations to improve estimates of precipitation using a mesoscale atmospheric model. We will present results of simulations of summer-storm cases. Even though summer convection seems to be driven locally within the monsoon circulation on any given day, large-scale conditions may also lead to important areas of convergence needed for convective initiation and storm development over the basin.

A coarse grid convering most of the continental U.S., parts of Canada, and parts of the Pacific ocean is required for these simulations. This required large domain limits us in terms of the finest grid resolution that is practical at the current time. We use a nested-grid domain to alleviate the afore-mentioned limitation by progressively achieving higher resolution toward the basin. In this nested-grid setup, coarse grids spawned within the domain can have resolution appropriate for cumulus parameterization schemes, while some inner grids require resolution between ~2-km and 20-km where neither explicit microphysics nor cumulus parameterization schemes are adequate. In this grid-nesting procedure, one needs to address issues related to validity and utility of cumulus and explicit microphysics parameterizations respectively. We will present a series of simulations that will illustrate how the two types of schemes may represent local convection in predicting reasonable precipitation estimates at various model resolutions.

The Regional Atmospheric Modeling System (RAMS) is used for these simulations. The National Center for Environmental Prediction (NCEP) data are used to initialize RAMS. Surface observations are used to compare and validate the model results when applicable.