A regional model for forest fire hazard zonation in forests of Dorud city (Case Study: Babahar region)

One of the initial actions in the management of forest lands and reduce the losses caused by this phenomenon is identifying areas prone to fire risk. Determine the probability of a fire hazard is effective in the prevention and management of fires. In this study, fire hazard zonation map in Doroud city was prepared and examined using statistical modeling. For the modeling, multivariate regression analysis in the SPSS software was used. 63 samples include burned and non burned lands were detected. Burned area as the dependent variable and the factors affecting fire occurrence consists of annual rainfall, the proximity (roads and residential areas), elevation, slope, percentage of vegetation cover, etc. were considered as independent variables. The quantitative values of above parameters were estimated using available maps in the Geographical Information System (GIS) and Remote Sensing (RS) environment as well as field studies. Regression analysis performed for the proposed model and validation of the model's performance was carried out. The amount of vegetation cover and rainfall were the most important factors affecting fire incidences. Finally fire hazard zonation map was prepared by applying validated model in GIS environment. The research results substantiate the effectiveness of the proposed model for fire hazard assessment.