Overview

Assessing wildfire potential, and fighting existing wildfires, requires accurate rapidly updated information about the local wind, humidity, precipitation, and temperature. Unfortunately, this localized weather information can often not be accurately derived from the weather forecasts by national operational centers because of their coarse–resolution model grids. Thus, analyzing and forecasting weather and climate in wildfire areas can greatly benefit from high–resolution weather models. The NCAR RAL Real–Time Four Dimensional Data Assimilation (RTFDDA) and forecasting system can be launched on demand, with a high–resolution model focusing on an area of special interest, for example the environment of a wildfire.

The core technology of RTFDDA is a Four–Dimensional Data Assimilation (FDDA) method, where observations are dynamically combined into a full–physics mesoscale model (WRF/MM5) to generate accurate real–time analyses and short–term forecasts on multi–scale domains. The high–resolution grids of the model are designed to resolve fine–scale terrain, land use, coastlines, and urban forcing.

On June 8, 2002, a sudden and rapidly developing forest fire started in Hayman, southwest of Denver, Colorado. The Hayman Fire was the largest fire in Colorado. It caused nearly $40 million in damages, burned 133 homes and forced the evacuation of 5,340 persons. Within a few hours after the fire started, RAL launched an RTFDDA model and provided high–resolution, rapidly updated weather information to support the management of fire–fighting operations.

The domains of the RTFDDA model set up to support fighting Hayman fire on June 9, 2002.