عنوان مقاله [English]
نویسندگان [English]چکیده [English]
Fire is a phenomenon that causes many losses in forests and natural resources. For fire prevention and reduction of its losses, it is necessary to recognize high potential area using affective parameters. In this research fire risk modeling has been done for Golestan province using multi-criteria analysis (MCA). Static and dynamic parameters were divided in different classes and weight of each class was determined using expert and data knowledge. Then index map of each parameter was produced using derived weight and static/dynamic fire risk map through combination of parameters with AHP derived weight. Finally, fire risk is produced through combination of static and dynamic fire risk map. According to the result, the proposed method recognized 27.88 percent of study area as danger region that 91.8 percent of fires occurred in that region in June 2005. The proposed model recognized 17.5 percent for September 2005 and 30.6 percent for June 2004 of study area as danger region which 66.7 and 89.44 percent of fires occurred in those areas, respectively. Also accuracy criterion has been improved through AHP method and has been increased from 2.14 to 2.43. The result shows that MCA and satellite data have more capability in recognition of high potential area for fire occurrence.