Determining Effective Factors on Forest Fire Using Geographically Weighted Regression, a Case Study: Golestan, Iran

Determining the effective factors on fire is so important, because it helps us to identify most dangerous locations and times in forest fire. Hence, we can prepare the necessary facilities andequipments near those locations and we can prevent many of ignitions with more attention, education to people, and law enforcement. In the current study, we tried identifying some impressive factors on the fire in Golestan forest using Geographically Weighted Regression(GWR) method that is suitable for the spatial regression problem, because of its autocorrelation and non-stationarity properties. This research shows that both of the biophysical andanthropogenic factors have significant effects on forest fire in our study area. In biophysicalfactors, the aspect with R2=0.939 and in anthropogenic factors, the road distance with R2=0.866 were most important factors and combination of these two factors had significant effect, too with R2=0.893.

کلیدواژه ها:

Geographically Weighted Regression ، Forest Fire ، Human-caused fires ، Golestan Forest