Forest fires can cause enormous damage, such as deforestation and environmental pollution, even with a single occurrence. It takes a lot of effort and long time to restore areas damaged by wildfires. Therefore, it is crucial to know the forest fire risk of a region to appropriately prepare and respond to such disastrous events. The purpose of this study is to develop an hourly forest fire risk index (HFRI) with 1 km spatial resolution using accessibility, fuel, time, and weather factors based on Catboost machine learning over South Korea. HFRI was calculated through an ensemble model that combined an integrated model using all factors and a meteorological model using weather factors only. To confirm the generalized performance of the proposed model, all forest fires that occurred from 2014 to 2019 were validated using the receiver operating characteristic (ROC) curves and the area under the ROC curve (AUC) values through one-year-out cross-validation. The AUC value of HFRI ensemble model was 0.8434, higher than the meteorological model. HFRI was compared with the modified version of Fine Fuel Moisture Code (FFMC) used in the Canadian Forest Fire Danger Rating Systems and Daily Weather Index (DWI), South Korea’s current forest fire risk index. When compared to DWI and the revised FFMC, HFRI enabled a more spatially detailed and seasonally stable forest fire risk simulation. In addition, the feature contribution to the forest fire risk prediction was analyzed through the Shapley Additive exPlanations (SHAP) value of Catboost. The contributing variables were in the order of relative humidity, elevation, road density, and population density. It was confirmed that the accessibility factors played very important roles in forest fire risk modeling where most forest fires were caused by anthropogenic factors. The interaction between the variables was also examined.

A probabilistic estimation model for forest biomass using unmanned aerial vehicle (UAV) photography was developed. We utilized a machine-learning-based object detection algorithm, a mask region-based convolutional neural network (Mask R-CNN), to detect trees in aerial photographs. Subsequently, Bayesian regression was used to calibrate the model based on an allometric model using the estimated crown diameter (CD) obtained from aerial photographs and analyzed the diameter at breast height (DBH) data acquired through terrestrial laser scanning. The F1 score of the Mask R-CNN for individual tree detection was 0.927. Moreover, CD estimation using the Mask R-CNN was acceptable (rRMSE = 10.17%). Accordingly, the probabilistic DBH estimation model was successfully calibrated using Bayesian regression. A predictive distribution accurately predicted the validation data, with 98.6% and 56.7% of the data being within the 95% and 50% prediction intervals, respectively. Furthermore, the estimated uncertainty of the probabilistic model was more practical and reliable compared to traditional ordinary least squares (OLS). Our model can be applied to estimate forest biomass at the individual tree level. Particularly, the probabilistic approach of this study provides a benefit for risk assessments. Additionally, since the workflow is not interfered by the tree canopy, it can effectively estimate forest biomass in dense canopy conditions.

A prediction model was developed for the wildfire fuel load of Korean red pine (Pinus densiflora) stands with susceptibility to forest fire based on the forest-growth model. Furthermore, a time-series analysis was performed on the variation in forest-fire fuel load according to forest management. National Forest Inventory stand data of 1434 plots for P. densiflora stands were used, and the final forest-fire fuel load prediction model was developed using the Weibull function and mortality model. The fit index of the diameter distribution model ranged from 0.58 (0th percentile) to 0.96 (50th percentile), and that of the mortality model was 0.68. The prediction of the stand growth variation after 20 years based on the growth data of managed and unmanaged stands indicated a mean stand density of 1518 trees per ha for unmanaged stands, and 885 trees per ha for managed stands. Regarding the variation in the available canopy fuel load distribution, the predicted annual increase was approximately 0.7 ton/ha for unmanaged stands and approximately 0.5 ton/ha for managed stands. These findings will contribute to setting fuel management criteria to prevent forest fire spread while providing the quantitative data of the characteristics of stand growth variation and the predicted wildfire fuel load.

Tree detection and fuel amount and distribution estimation are crucial for the investigation and risk assessment of wildfires. The demand for risk assessment is increasing due to the escalating severity of wildfires. A quick and cost-effective method is required to mitigate foreseeable disasters. In this study, a method for tree detection and fuel amount and distribution prediction using aerial images was proposed for a low-cost and efficient acquisition of fuel information. Three-dimensional (3D) fuel information (height) from light detection and ranging (LiDAR) was matched to two-dimensional (2D) fuel information (crown width) from aerial photographs to establish a statistical prediction model in northeastern South Korea. Quantile regression for 0.05, 0.5, and 0.95 quantiles was performed. Subsequently, an allometric tree model was used to predict the diameter at the breast height. The performance of the prediction model was validated using physically measured data by laser distance meter triangulation and direct measurement from a field survey. The predicted quantile, 0.5, was adequately matched to the measured quantile, 0.5, and most of the measured values lied within the predicted quantiles, 0.05 and 0.95. Therefore, in the developed prediction model, only 2D images were required to predict a few of the 3D fuel details. The proposed method can significantly reduce the cost and duration of data acquisition for the investigation and risk assessment of wildfires.

While geostationary satellites can provide continuous near-real-time observations, their low spatial resolution makes it difficult to detect small wildfires. Conversely, polar-orbiting satellites are capable of observing small wildfires at high spatial resolution, but can operate only within restricted observation periods. To improve wildfire spread prediction accuracy using polar-orbiting satellite observations, this paper proposes a novel methodology to accurately reproduce wildfire perimeters observed at specific time points by these satellites. The approach employs a wildfire spread simulator combined with a rate of spread (ROS) adjustment factor. The proposed algorithm derives ROS adjustment factors for each fuel model based on differential evolution, achieving up to a 0.4 increase in the Sørensen index when reproducing wildfire perimeter data at given observation times. Incorporating these factors into simulator-based predictions allows comprehensive consideration of external factors affecting wildfire propagation, which have not been sufficiently accounted for in previous methods. Moreover, considering the frequent occurrence of small wildfires in Korea, this study establishes a mapping between major species of trees in Korea and corresponding Fire Behavior Fuel Models (FBFMs). This serves as an example of appropriately matching major species of trees to FBFMs for wildfire spread prediction in countries where FBFMs have not been previously applied. The methodology’s effectiveness is demonstrated using wildfire perimeter data from polar-orbiting satellite observations and ignition points of recent wildfires in Korea. The proposed algorithm is expected to significantly enhance wildfire response by swiftly providing critical information for accurate wildfire spread prediction. This will facilitate prompt and precise countermeasures for small wildfires independent of external conditions such as weather.