An Innovative Method For Forest Fir

An Innovative Method For Forest Fire Risk
Zoning Map Using Fuzzy Inference System And
GIS
Mohammadali Rajabi, Aliasghar Alesheikh, Alireza Chehreghan, Hassan Gazmeh
Abstract: Forest Fire causes considerable environmental damage and brings about a significant change in the ecosystem of the region. It is a humanistic and national duty to protect against fire the forests. Most of such forest fire incidents result from human nonchalance. Other causes such as thunderstorm, glass objects and etc. are also considered as triggers of such incidents, this paper mainly focuses upon the human factor. The considered parameters such as distance from the road, residential areas, river, slope, climate and type of vegetation, and GIS play a significant role in the analysis and determining the factors impacting fire incidents. GIS was used for the analysis and calculations required in regard with these parameters. Therefore the locations with fire risk are determined by a combination of experimental model, fuzzy inference system and GIS. The obtained results indicate high accuracy and good efficiency. Results discussed extensively in paper. To evaluate the proposed method, the obtained results were compared with fire incidents of past years. The comparison results indicate an improvement in predication by this method in comparison with other methods.
Keywords: Forest Fire Risk Fire Risk Zoning Map, GIS, Fuzzy Inference System, Experimental Model. —————————— ——————————
1 INTRODUCTION
Forest fire is not dependent on a specific season, but most of forest fire incidents of Golestan Province have occurred when trees have been dry and the ground has been thick with dead leaves. The fact that the region is frequented by tourists has led to widespread fire incidents in Golestan Forest. Tourists’ disregard of safety regulations such as not putting out their fire before leaving the region and etc. has led to many undesired and serious fire incidents [1]. The biggest fire incident happened in November 2010 which lasted for several weeks and burned a large area of the forest. Repeated fire incidents have destroyed the soil texture of the region which in some cases has destroyed some plant species and has generally made the plant growth problematic. Studies on the behavior of fire and its impact factors will certainly prepare us for prevention and protection against such incidents [2]. Many countries make use of satellite images for monitoring fire incidents and detecting the burning regions. Forest fires are generally categorized into two types: ground fire and crown fire.
Ground fires mostly consume the dead leaves, shrubs and small trees. This type of fire is used in several regions such as Africa for prevention of unwanted fires. In this system, the speed and direction of wind is simulated by various methods such as Cellular Automata to extinguish fire in certain areas [3]. The second type of fire, the crown fire, is considered as one of the most dangerous and detrimental types of fire[4]. Most of the fires experienced in Golestan forest have unfortunately been of crown type. Controlling this type of fire is of great difficulty and requires advanced equipment. For on-time and optimum allocation of resources and equipment during fire incident, a fire spread simulation system is required. Various factors, such as density and type of vegetation, humidity, temperature, slope, distance from road and residential area, may trigger and affect fire incidents. Such factors can be found more in lawns and bushy areas than in forest areas with oak and beech trees. Humidity, on the other hand, as one of the significant impact factors, has a greater effect on bushy areas than on forests. In case the flammable materials (e.g. the grass) are humid, the risk of fire for such materials reaches its minimum [5]. In general, fire incidents can be classified into three groups: natural fire incidents, intentional man-made fires and unintentional man-made fires caused by human carelessness [6]. The second group, intentional man-made fires, is taken into consideration in this study for preparation of the fire map. The risk potential has been prepared according to fuzzy inference method and using factors such as slope, climate, vegetation, and distance from road, river and residential areas. This potential risk map can be used for preventive operations as well as for optimum and timely allocation of resources and equipment during such incidents. More patrols and sentinels can be deployed in regions with high fire risk during fire seasons. With regard to unprotected regions with great fire potential, access roads can be built to such areas so the transfer time of resources to these regions during fire incidents reaches its minimum. According to Adab [7] can be mentioned as one of the Iranian studies carried out on this subject. They conducted the zoning of fire potential in forest areas of Mazandaran province per all seasons of 2005 as well as a 15- year period

Hasil (Bahasa Indonesia) 1:

Metode inovatif untuk risiko kebakaran hutanZonasi peta menggunakan sistem kesimpulan kabur danGISMohammadali Rajabi, Aliasghar Alesheikh, Alireza Chehreghan Hassan GazmehAbstrak: Kebakaran hutan menyebabkan kerusakan lingkungan yang cukup dan membawa perubahan yang signifikan dalam ekosistem wilayah. Itu adalah tugas kemanusiaan dan nasional untuk melindungi terhadap kebakaran hutan. Sebagian besar hutan seperti kebakaran akibat insiden acuh manusia. Penyebab lain seperti petir, kaca objek dan dll juga dianggap sebagai pemicu kejadian tersebut, karya ini terutama berfokus pada faktor manusia. Parameter dianggap seperti jauh dari jalan, pemukiman, sungai, lereng, iklim dan tipe vegetasi, dan GIS memainkan peran penting dalam analisis dan menentukan faktor-faktor yang mempengaruhi kebakaran insiden. GIS digunakan untuk analisis dan perhitungan yang diperlukan dalam hal dengan parameter ini. Oleh karena itu lokasi dengan resiko kebakaran ditentukan oleh kombinasi dari model eksperimental, sistem kesimpulan kabur dan GIS. Hasil yang diperoleh menunjukkan akurasi tinggi dan efisiensi yang baik. Hasil yang dibahas secara luas dalam kertas. Untuk mengevaluasi metoda yang diusulkan, diperoleh hasil dibandingkan dengan insiden kebakaran tahun. Perbandingan hasil menunjukkan peningkatan predikasi dengan metode ini dibandingkan dengan metode lain.Kata kunci: Hutan resiko kebakaran api risiko zonasi peta, GIS, sistem kabur kesimpulan, Model eksperimental. —————————— ——————————1 INTRODUCTIONForest fire is not dependent on a specific season, but most of forest fire incidents of Golestan Province have occurred when trees have been dry and the ground has been thick with dead leaves. The fact that the region is frequented by tourists has led to widespread fire incidents in Golestan Forest. Tourists’ disregard of safety regulations such as not putting out their fire before leaving the region and etc. has led to many undesired and serious fire incidents [1]. The biggest fire incident happened in November 2010 which lasted for several weeks and burned a large area of the forest. Repeated fire incidents have destroyed the soil texture of the region which in some cases has destroyed some plant species and has generally made the plant growth problematic. Studies on the behavior of fire and its impact factors will certainly prepare us for prevention and protection against such incidents [2]. Many countries make use of satellite images for monitoring fire incidents and detecting the burning regions. Forest fires are generally categorized into two types: ground fire and crown fire.Ground fires mostly consume the dead leaves, shrubs and small trees. This type of fire is used in several regions such as Africa for prevention of unwanted fires. In this system, the speed and direction of wind is simulated by various methods such as Cellular Automata to extinguish fire in certain areas [3]. The second type of fire, the crown fire, is considered as one of the most dangerous and detrimental types of fire[4]. Most of the fires experienced in Golestan forest have unfortunately been of crown type. Controlling this type of fire is of great difficulty and requires advanced equipment. For on-time and optimum allocation of resources and equipment during fire incident, a fire spread simulation system is required. Various factors, such as density and type of vegetation, humidity, temperature, slope, distance from road and residential area, may trigger and affect fire incidents. Such factors can be found more in lawns and bushy areas than in forest areas with oak and beech trees. Humidity, on the other hand, as one of the significant impact factors, has a greater effect on bushy areas than on forests. In case the flammable materials (e.g. the grass) are humid, the risk of fire for such materials reaches its minimum [5]. In general, fire incidents can be classified into three groups: natural fire incidents, intentional man-made fires and unintentional man-made fires caused by human carelessness [6]. The second group, intentional man-made fires, is taken into consideration in this study for preparation of the fire map. The risk potential has been prepared according to fuzzy inference method and using factors such as slope, climate, vegetation, and distance from road, river and residential areas. This potential risk map can be used for preventive operations as well as for optimum and timely allocation of resources and equipment during such incidents. More patrols and sentinels can be deployed in regions with high fire risk during fire seasons. With regard to unprotected regions with great fire potential, access roads can be built to such areas so the transfer time of resources to these regions during fire incidents reaches its minimum. According to Adab [7] can be mentioned as one of the Iranian studies carried out on this subject. They conducted the zoning of fire potential in forest areas of Mazandaran province per all seasons of 2005 as well as a 15- year period