Evaluating landscape-scale wildfire exposure in northwestern Iran

Abstract

We implemented a fine-scale fire modeling approach to assess wildfire exposure in the highly valued resources and assets (HVRAs) of Ardabil Province (18,000 km2), northwestern Iran. For this purpose, we used the minimum travel time algorithm and simulated 60,000 wildfires under wildfire season most frequent weather scenarios. Wildfire exposure was analyzed on different vegetation types and municipalities using burn probability (BP), conditional flame length (CFL), and fire size (FS) modeling outputs. Also, we obtained the fire potential index (FPI) and source–sink ratio metrics to assess wildfire transmission across the study area. The BP ranged from 0.0003 to 0.013 (mean = 0.0008) and varied substantially among and within the HVRAs of the study area. While the lowest BP values located in broadleaf forests, the highest BP values concentrated on flashy fuel areas, including cereal crops, mountain meadows, and grazed pastures. The average CFL was 0.3 m, with the highest values peaking in cereal crops and wooded pastures located on slopes. FS ranged from about 1–1700 ha, with an average value of 225 ha. Fires ignited in the northern part of the study area resulted in the most significant FS values, due to the large contiguous patches of high fuel loads. High FPI values were associated with large fire ignition areas and anthropic fire occurrence hotspots in the northern and southern parts of the study area. Cereal crops and grazed pastures behaved as relevant wildfire sources of fires exposing rural communities. The results of this study may help support the development of an improved wildfire risk management policy in the study area. The methods from this study could be replicated in neighboring areas and other cultural landscapes of the Middle East, where wildfires pose a threat to human assets and natural values.