Characteristics of forest fires in Korea using national forest fire statistical data over a 29-year period, from 1990 to 2018, were analyzed in this study. Among the causes of forest fires, the most frequent was visitors to the mountain. Even though the proportion of forest fires from mountain visitors has been gradually decreasing over the years as compared with the other causes, it is still the most frequent in each region. However, in cities with high populations, such as Seoul, Incheon, and Busan, the number of fires caused by smoking and children was relatively large as compared with other regions. Thus, the results are influenced by the population and meteorological characteristics of each region. Kruskal-Wallis nonparametric variance analysis was performed to analyze regional differences according to causes of forest fires during the spring fire seasons. The analysis showed significant differences among the top three regions that had the highest occurrences of forest fires. The results showed that the number of forest fires from farmers was significant in Jeollanam-do and Chungcheongnam-do during the month of February, whereas the number from visitors to the mountain was significant in Gangwon-do and Chungcheongbuk-do during the month of May.

For effective management and prevention, wildfire risk prediction needs to consider the substantial impacts of climate change on wildfire patterns. This study analyzed the probability of wildfire occurrence in South Korea using the Maximum Entropy (MaxEnt) model and predicted future wildfire occurrence under shared socioeconomic pathway (SSP) climate change scenarios. The model utilized historical fire occurrence data and was trained using 12 environmental variables encompassing climate, topography, vegetation, and socioeconomic factors. Future wildfire risk was predicted under the SSP2-4.5 and SSP5-8.5 scenarios for 2041–2060 and 2081–2100. Increased average temperature and solar radiation were key drivers of elevated wildfire risk, whereas increased precipitation and relative humidity reduced this risk. Under current conditions, 367,027 ha (6.52%) within the study area were classified as high-risk based on the MaxEnt model output (p > 0.6). Under both SSP scenarios, a decline in the at-risk area was observed over time. This study provides fundamental data for wildfire management and prevention strategies in South Korea and provides quantitative evidence on the potential impact of climate-related environmental changes on wildfires.

Forest fires pose significant threats to ecosystems, economies, and human lives. However, existing forest fire risk assessments are over-reliant on field data and expert-derived indices. Here, we assessed the nationwide forest fire risk in South Korea using a dataset of 2289 and 4578 fire and non-fire events between 2020 and 2023. Twelve remote sensing-based environmental variables were exclusively derived from Google Earth Engine, including climate, vegetation, topographic, and socio-environmental factors. After removing the snow equivalent variable owing to high collinearity, we trained three machine learning models: random forest, XGBoost, and artificial neural network, and evaluated their ability to predict forest fire risks. XGBoost showed the best performance (F1 = 0.511; AUC = 0.76), followed by random forest (F1 = 0.496) and artificial neural network (F1 = 0.468). DEM, NDVI, and population density consistently ranked as the most influential predictors. Spatial prediction maps from each model revealed consistent high-risk areas with some local prediction differences. These findings demonstrate the potential of integrating cloud-based remote sensing with machine learning for large-scale, high-resolution forest fire risk modeling and have implications for early warning systems and effective fire management in vulnerable regions. Future predictions can be improved by incorporating seasonal, real-time meteorological, and human activity data.

Severe droughts and changing precipitation patterns could alter the biogeochemical properties of the soil, affecting soil carbon cycles in forest ecosystems. A throughfall exclusion (TFE) experiment was conducted in a continental climate coniferous stand in Gangwon Province, Korea, to examine the effects of excluding rainfall on total soil respiration (SR), heterotrophic soil respiration (HR), autotrophic soil respiration (AR), sapling diameter growth, and soil bacterial communities from July 2016 to October 2017. The soil water content (SWC) was significantly decreased by the exclusion of the throughfall, resulting in changes in the bacterial communities, and subsequently a decrease in HR. Although AR did not present significant differences between the control and TFE plots, the rate of sapling growth was significantly lower in the TFE plots compared with that in the control plots. An exponential function relating SR to soil temperature accounted for 0.61% and 0.82% of the variance in SR in the control and TFE plots, respectively (Q₁₀ = 2.48 and 2.86, respectively). Furthermore, a multivariate nonlinear model based on soil temperature and SWC explained 0.89% and 0.88% of the variance in SR in the control and TFE plots, respectively. When soil temperature was high, SR showed high fluctuations due to SWC variation. However, when SWC was low, we detected relatively small fluctuations in SR due to soil temperature. The results of this study show that the activity of soil microbial and root respiration during the growing season may be lower under future drought conditions.

The fuel characteristics of a forest are important factors in designing a forest fire danger rating system and must be analyzed to understand forest fire behavior. Therefore, in this study, forest fire intensity was predicted depending on the amount of fuel in the forests in the Yeongdong and Yeongseo regions of Gangwon-do from November 2020 to December 2021. The possibility of forest fire occurrence, from surface fire to crown fire, was analyzed depending on the fuel conditions. Regional differences were found in the amount of fuel and the fuel structure of the surface layer in the forest. The intensity and type of forest fire changed according to these differences. The Yeongdong area is more susceptible to crown fires and more intense of surface fire than the Yeongseo area. These results show the need to prevent early forest fires by learning the characteristics of forests in different regions and analyzing the intensity and type of forest fire predicted.

This study examines the effects of meteorological fire danger and human activity on wildfire ignition patterns in South Korea using records from 2004 to 2023. A percentile-based Fire Weather Index (FWI) classification, derived from negative binomial regression, identified critical daily fire frequency thresholds at FWI 4.39 (μ ≥ 1 fire/day) and FWI 6.84 (μ ≥ 2 fires/day). Bivariate LISA analysis revealed a spatial mismatch between resident population density and wildfire frequency: High–High (HH) clusters were concentrated in the Seoul metropolitan fringe, while Low–High (LH) clusters appeared in mountainous provinces where forest visitor ignitions and agricultural burning are the primary causes. In HH clusters, cigarette-related ignitions and structure-to-forest transitions were comparatively more frequent. Wildfire events were concentrated in age class 4–5 coniferous and broadleaf stands, and mean ignition-to-building distances in metropolitan areas frequently fell below 150 m. These findings suggest that prevention strategies should shift from uniform resident-oriented approaches toward spatially differentiated management targeting transient populations in LH areas and Wildland-Urban Interface (WUI) exposure in HH areas.