Developing a spatial model for urban development planning in forest fire-prone areas

Forest fires are one of the critical issues with significant environmental and economic consequences. This study aims to comprehensively analyze the impacts of urbanization in areas sensitive to forest fires. The study area includes the districts of Karaburun, Çeşme, Urla, and Seferihisar in İzmir Province. Maps of the study area were generated for the years 1990, 2006, and 2018 using the CORINE Land Cover (CLC) dataset, classified according to Level 1 CORINE Land Cover classes. The accuracy of the classified maps was assessed. Following the verification of classification accuracy, past and present data were analyzed to simulate future urban growth trends. The 2050 projections indicate a 76.72% increase in settlement areas, a 4.97% increase in agricultural areas, and an 8.44% decrease in forest areas compared to 2018. The priorities of factors influencing fire risk were determined through Analytic Hierarchy Process (AHP) analysis. The importance weights of the identified risk factors were found as follows: Anthropogenic factors 44% (distance to power lines 13%, distance to roads 11%, distance to agricultural areas 11%, distance to settlements 9%), Climate 30% (relative humidity 12%, temperature 9%, wind speed 9%), Vegetation 21% (tree species 10%, canopy closure 6%, stand development stage 5%), and Topography 5% (aspect 3%, slope 1%, elevation 1%). Based on these importance weights, a comprehensive fire risk map was developed. According to the map, 9.28% (16,549 ha) of the study area is classified as low risk, 51.93% (92,625 ha) as moderate risk, 38.58% (68,818 ha) as high risk, and 0.21% (378 ha) as very high risk. The ROC curve method was used in the verification stage of the fire risk map, and the AUC value was found to be 71%. New urban areas projected to transition from agriculture and forest to settlement on the 2050 projection map were integrated with the fire risk map generated using the AHP method. According to the analysis results, urban development is concentrated in medium-risk areas at a rate of 45.38% and high-risk areas at a rate of 54.16%. Limited urban development is projected at a rate of 0.46% in low-risk areas and 0.01% in very high-risk areas. This study highlights the significance of developing a spatial model to analyze the effects of urbanization in areas sensitive to forest fire risk and to identify high-risk zones for better urban planning. The findings contribute to the formulation of sustainable and risk-focused urban growth strategies.