Meso-scale spectral model simulations over the San Jacinto Mountain region

The Meso-scale Spectral Model (MSM), a non-hydrostatic version of the Regional Spectral Model (RSM), will be used to demonstrate the capability of the model in simulating near surface meteorological variables at 1-km grid intervals. Accurate weather data input is critically important in simulating the advance of the surface fire front in the event of a wildland fire. Previously only meteorological models with coarse resolution, 5-km at best, or diagnostic models under heavy dynamical assumptions were applied in fire spread modeling studies. These previous efforts were hampered by the uncertainties of the weather data used in the fire-spread simulation. The uncertainties usually arise either from the coarse spatial scale of the meteorological model or from the over-simplified assumptions of the model.

For this study, we selected a study area over the San Jacinto mountain region, where the Esperanza wildfire in October 2006 consumed 16137 hectares of chaparral and desert scrub vegetation in mountain terrain roughly 50 km east of Riverside, CA. Five US Forest Service fire fighters tragically perished in the fire suppression effort, primarily due to the unexpected intensified wind shift prior to the burn-over incident. This Santa Ana wind driven wildfire over complex terrain therefore provides a unique opportunity for model validation before an operational fire-spread predictive tool can be implemented. To facilitate the validation effort, an array of 6 Remote Automatic Weather Stations (RAWSs) were deployed around the hills and valleys of the fire site covering a study area of 10 km by 10 km. The experimental hourly observation started in October 2008. Data have been automatically transmitted to the data collection center via satellite. Simulations of MSM during the Esperanza fire event and during October 2008 will be validated first against the nearby first order surface weather stations and the existing two RAWSs. Data compiled from the 6 experimental RAWSs will then be added to the available analysis to validate simulations during October 2008.