Abstract:

Forest fires can occur for a variety of reasons and can spread quickly. So it is a big environmental problem. 12 million hectares of land in Turkey, especially in the Aegean and Mediterranean regions, carries a risk of forest fire. Risky areas in forest fire are places where the fire can start easily and spread quickly to other areas. It is difficult to control nature. In this context, the evaluation of the criteria that cause forest fire is discussed in this study. These criteria are topography, vegetation, climate and socioeconomic, which are 4 main criteria. The sub-criteria of these criteria are height, slope, front, dry vegetation, biomass density, temperature, humidity, wind speed, precipitation, distance to settlement, land use, population density and distance to power lines. Therefore, proposing a multi-criteria decision-making (MCDM) model is invaluable to obtain convincing results, due to the high rate of forest fire. Therefore, Pythagorean Fuzzy (PF) clusters, which offer a wide evaluation scale to the decision maker, PBAHP, Analytical Network Process (AAP) and Analytical Hierarchy Process (AHP) methods, which are used in many applications in the studies in the literature, were applied. The criteria that have an impact on forest fire were prioritized using these two methods. Comparative analysis was made with the results obtained and it was shown how important the result was. Then, scenario-based sensitivity analysis was performed and situations in different conditions were observed. Thus, the fire risk status of forested areas was determined by the study. Therefore, it will contribute to the efficient allocation of available resources.

Keywords: