Gorgan University Of Agricultural Sciences and Natural Resources Journal of Wood and Forest Science and Technology 2322-2077 25 2 2018 08 23 Determination of the effective criteria of forest fire occurrence by using GIS and ANN (Case study: Golestan province) Determination of the effective criteria of forest fire occurrence by using GIS and ANN (Case study: Golestan province) 110 136 4197 10.22069/jwfst.2018.5611.1339 FA Sajad Alimahmoodisarab Jahangir Feghhi Safargholi Khaje Journal Article 2014 05 23 Forest fire management has important role in fire prevention and effects. The aim of this study was applied of Geographical Information System(GIS) and Artificial Neural Network (ANN) to determine forest fire criteria in Golestan province. The influence of each parameter on fire ignition was determined by collecting of 37 sample from burned area and 37 sample from not burned area. For formation network between criteria and fire occurrence used of Multilayer perceptron (MLP) with Hyperbolic Pattern Algorithms. the results shown raining and distance from road had must influence on forest fire ignition. Validation test showed that the best network obtained in run 4 and epoch 450 with 0.0038 Final Mean Square Error (FMSE) in training steps. Also, about 95 percent of area burned and 84 of unburned areas has been properly classified. Finally, forest fire hazard maps was obtained based on each criteria weights. Result showed this network with 2 hidden layer and 12 neuron in each of them has best accuracy, and correlation coefficient (R) was 0.80. Forest fire management has important role in fire prevention and effects. The aim of this study was applied of Geographical Information System(GIS) and Artificial Neural Network (ANN) to determine forest fire criteria in Golestan province. The influence of each parameter on fire ignition was determined by collecting of 37 sample from burned area and 37 sample from not burned area. For formation network between criteria and fire occurrence used of Multilayer perceptron (MLP) with Hyperbolic Pattern Algorithms. the results shown raining and distance from road had must influence on forest fire ignition. Validation test showed that the best network obtained in run 4 and epoch 450 with 0.0038 Final Mean Square Error (FMSE) in training steps. Also, about 95 percent of area burned and 84 of unburned areas has been properly classified. Finally, forest fire hazard maps was obtained based on each criteria weights. Result showed this network with 2 hidden layer and 12 neuron in each of them has best accuracy, and correlation coefficient (R) was 0.80. https://jwfst.gau.ac.ir/article_4197_55490ea1055fdaf77d45c1aad573ee35.pdf