Integrating high-resolution static data into WRF for real fire simulations

The development of fire models coupled within WRF brings about the capability to simulate real fires using many of the same tools originating from numerical weather prediction. The WRF preprocessor is capable of integrating real data from a variety of sources into a WRF simulation; however, it was not designed for processing the kind of data needed for fire. The standard global surface dataset, while sufficient for mesoscale weather forecasting, is too low resolution for a realistic fire simulation. In addition, WRF provides no mechanism for importing the geospatial vector data used to store fire perimeters. We present a case study simulating the Witch fire that burned just outside San Diego in 2007. Using tools developed for WRF-Fire and its extension hosted at openwfm.org, we describe the integration of data from public sources into the WRF workflow creating a realistic simulation of this event. This procedure provides a template for future research analyzing past fires as well as predicting the spread of ongoing events.