A Raster-based forest fire regime model

Considered both the natural (e.g. climate) and anthropogenic factors which contribute to regional fire dynamics, a new fire regime model is proposed. It consists of four components: fuel module, fire occurrence module, and fire behavior module, and fire effect module. We derive the fuel load data from ecosystem model's carbon and nitrogen pools and fix the fuel properties (e.g. depth, size and compactness) based on other related studies. The fire occurrence is related to the fire danger index, ignition sources (e.g. lightning and human-caused ignition), and fuel load. The fire danger index is an aggregative indicator which can be calculated based on regional temperature and precipitation data. As for the ignition sources, we hypothesize that human-caused ignitions vary with human population density, while the lightning ignitions are distributed evenly. With the spatially explicit fuel load, wind speed data, population density data, and climate data, the model has been tested in forest ecosystem of China.