WEST KALIMANTAN FOREST FIRE PROBABILITY MAPPING USING BINARY LOGISTIC REGRESSION

Abstract

Forest fires, which occur almost annually, are common in West Kalimantan during the dry season. There is little question that the rate of regional development will slow down as a result of the huge effects this condition has on the social, economic, and environmental domains. Naturally occurring factors are one of the many potential causes of forest fires. The goal of this research was to identify the factors that significantly influence forest fires and to produce a map showing the likelihood of forest fires occurring in various West Kalimantan cities and districts. The analytical technique that enabled us to achieve our goal was logistic regression. The existence or absence of forest fires is one of the dependent variables being used. The temperature, geography, vegetation, and human influences are the independent variables during this time. The bulk of forest fires that occurred in West Kalimantan were caused by human activity as opposed to natural causes, per the study's findings. There are several reasons why humans set off forest fires, whether on purpose or accidentally, but one of them is the distance that people can go to conduct activities inside the forest. Beyond the variables listed above, there are two other criteria that can start a forest fire: the distance from the point to the road and the distance from the point to the air. Using logistic regression, it was discovered that the variable distance between the site and the river contributed thirty percent to the likelihood of forest fires.