Wildfire activity is intensifying under climate change, particularly in temperate East Asia where human-driven ignitions interact with extreme fire-weather conditions. This study examines wildfire risk during the March 2025 large wildfire event in Korea by applying explainable machine-learning models to assess ignition-prone environments and their spatial relationship with socio-environmental features relevant to exposure and management. CatBoost and LightGBM models were used to estimate wildfire susceptibility based on climatic, topographic, vegetation, and anthropogenic predictors, with SHAP analysis employed to interpret variable contributions. Both models showed strong predictive performance (CatBoost AUC = 0.910; LightGBM AUC = 0.907). Temperature, relative humidity, and wind speed emerged as the dominant climatic drivers, with ignition probability increasing under hot (>25 °C), dry (6 m s⁻¹) conditions. Anthropogenic factors—including proximity to graves, mountain trails, forest roads, and contiguous coniferous stands (≥30 ha)—were consistently associated with elevated ignition likelihood, reflecting the role of human accessibility within pine-dominated landscapes. The socio-environmental overlay analysis further indicated that high-susceptibility zones were spatially aligned with arboreta, private commercial forests, and campsites, highlighting areas where ignition-prone environments coincide with active human use and forest management. These results suggest that wildfire risk in Korea is shaped by the spatial concurrence of climatic extremes, fuel continuity, and socio-environmental exposure. By situating explainable susceptibility modeling within an event-conditioned risk perspective, this study provides practical insights for identifying Wildfire Priority Management Areas (WPMAs) and supporting risk-informed prevention, preparedness, and spatial decision-making under ongoing climate change.

As wildfires intensify globally due to climate change, accurate wildfire danger forecasting systems have become essential for effective disaster management and early warning. Fuel Moisture Content (FMC), defined as the ratio of water mass to dry fuel mass, plays a critical role in determining ignition probability and fire spread dynamics. This study conducts a comparative analysis of five major national wildfire danger rating systems: the National Fire Danger Rating System (NFDRS, USA), Canadian Forest Fire Danger Rating System (CFFDRS), European Forest Fire Information System (EFFIS), Australian Fire Danger Rating System (AFDRS), and the Korean Forest Fire Danger Rating System (KFDRS). Using a multi-criteria comparative framework, the systems were evaluated based on fuel classification structure, input variables, modeling approach, and spatiotemporal prediction resolution. The results reveal substantial disparities in spatial resolution (100 m to district-level), temporal resolution (hourly vs. daily), and fuel moisture modeling approaches (physics-based, index-based, and hybrid systems). Specifically, NFDRS and AFDRS provide high-frequency forecasting with hourly temporal resolution, operating at spatial resolutions of 1 km and 100 m, respectively, and incorporating dynamic fuel moisture modeling. In contrast, CFFDRS and KFDRS primarily rely on daily index-based predictions. Furthermore, while many global systems increasingly leverage remote sensing and machine learning for real-time FMC estimation, South Korea’s KFDRS remains predominantly empirical and weather-driven. The analysis identifies critical limitations in the KFDRS, including coarse spatial resolution (district-level), limited integration of Live Fuel Moisture Content (LFMC) modeling, and the lack of AI-augmented hybrid approaches. Accordingly, this study proposes a phased three-stage policy roadmap (2026–2035), emphasizing sensor-network expansion, AI–physics fusion modeling, and high-resolution (10 m) FMC mapping to enhance forecasting accuracy in complex terrains. These findings provide strategic insights for improving wildfire risk management and supporting the transition from reactive response to predictive wildfire forecasting under increasing climate variability.

자원과학연구(Resources Science Research)

Recent climate change-driven warming and prolonged drought have reduced fuel moisture in forests, increasing the frequency and severity of wildfires. In particular, crown fires have become more frequent, highlighting the importance of quantifying crown fuel characteristics. However, studies on the crown fuel biomass of Pinus densiflora in South Korea remain limited. This study aimed to quantify crown fuel biomass and develop a diameter at breast height (DBH)-based allometric model across major P. densiflora distribution areas. A total of 67 sample trees were destructively sampled across seven regions. Crown fuel components were classified into needles and branch diameter classes, and their dry biomass was measured. A log-transformed allometric model was developed, and its reliability was evaluated using correction factors, confidence intervals, and prediction intervals. Branches accounted for approximately 78% of total crown fuel, while needles accounted for approximately 22%. The model demonstrated high explanatory power, with $R_{adj}^2$ ranging from 0.783 to 0.945. Combustible fuels (≤1.0 cm) accounted for approximately 45% of total crown fuel. The developed model can be used as an input for wildfire spread prediction and provides a basis for crown fire risk assessment and fuel mapping.

Military–civilian interface zones (MCIZs) adjacent to the Korean Demilitarized Zone (DMZ) represent complex wildfire environments shaped by restricted access, intensive military activities, and adjacent civilian land use. This study develops a spatially explicit wildfire ignition risk assessment framework for the DMZ and Civilian Control Zone (CCZ) in Paju, South Korea, employing Random Forest (RF), Generalized Additive Models (GAM), and Geographically Weighted Regression (GWR) in a complementary analytical design. A dataset of 318 wildfire ignition events (2001–2024), including 78 associated with military activities, was analyzed. The RF model achieved high predictive accuracy (AUC = 0.81), identifying proximity to military training zones, relative humidity, wind speed, and proximity to built infrastructure as dominant ignition drivers. GAM revealed narrow nonlinear thresholds—relative humidity at 13.8%–14.0% and wind speed at 13.5–14.0 m/s—corresponding to peak ignition probabilities. GWR demonstrated pronounced spatial heterogeneity, with military proximity exerting a stronger influence in the eastern and northern sectors, while the meteorological effects varied geographically. Based on these outputs, a unified analytical framework was established in which RF-derived ignition probabilities were interpreted alongside GAM- and GWR-based explanatory layers to provide spatially explicit ignition susceptibility assessments without numerical map fusion. The proposed approach provides a scientifically rigorous and operationally applicable method for quantifying ignition risk in politically sensitive, access-restricted landscapes, offering valuable insights for adaptive wildfire prevention and spatially informed governance of transboundary fire risk.