Producing biochar from forest fuels represents a sustainable approach to mitigating wildfires, though its potential for addressing heavy metal contamination remains poorly understood. This study employed slow pyrolysis with restricted oxygen to produce biochar from the surface fuels of four typical tree species in North Asia (Larix gmelinii, Pinus sylvestris var. mongolica, Quercus mongolica, and Fraxinus mandshurica) at varying temperatures (300–700°C). Adsorption kinetics and isotherms for Cd(II), Cr(VI), Pb(II), and Zn(II) ions were systematically investigated. Characterization was performed using Fourier Transform Infrared Spectroscopy (FT-IR) and X-ray Photoelectron Spectroscopy (XPS). Machine learning (Random Forest RF) and Structural Equation Modeling (SEM) were innovatively applied for in-depth analysis. Results indicate that Broadleaf tree biochar (Quercus mongolica, Fraxinus mandshurica) demonstrates superior adsorption for Pb (II) and Cd (II) (81 mg/g and 39 mg/g, respectively), while conifer biochar shows enhanced adsorption for Cr (VI) (11.84 mg/g). Mechanistic analysis indicates that adsorption of these four metal ions primarily follows monolayer adsorption accompanied by chemical interactions, including: chelation by oxygen-containing functional groups, M-π electron interactions, and CO32− precipitation. SHapley Additive exPlanations (SHAP) (based on models with R2 > 0.94) and Partial Least Squares Structural Equation Modeling (PLS-SEM) revealed that Electrochemical Properties (Ep) and specific surface area (SSA) are key physicochemical properties influencing adsorption efficiency, directly affecting biochar's ability to adsorb heavy metal ions. Collectively, these findings demonstrate that biochar produced from forest surface fuels not only offers a strategic solution for wildfire management, but also effectively remediates heavy metal contamination, thereby contributing to sustainable ecosystem development.

The Three-North Shelterbelt Program (TNSP) region, a key ecological barrier in China, has recently faced more frequent wildfires, while comprehensive evaluations of its fire risk are still lacking. Here, we integrate multisource remote sensing and socioeconomic datasets representing climatic, vegetative and human-activity factors and employ five machine learning classifiers - Random Forest (RF), Gradient-Boosting Decision Tree (GBDT), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Logistic Regression (LR) - together with explainable-AI methods to model and analyze regional fire susceptibility levels. Model comparisons show that tree-based ensemble models generally outperform linear models, with the LightGBM achieving the highest performance (cross-validated AUC > 0.98); the small differences among the ensemble models suggest that the data quality and feature dimensionality factors have greater effects on performance than the selected algorithm does. Seasonal fire hazard maps reveal a clear bimodal pattern, with primary peaks in the spring and autumn, a secondary peak during the summer, and minimal activity in the winter. High-risk areas are concentrated in the northeastern forest zone, the Inner Mongolian grasslands, the North China Plain and the Loess Plateau-with spatial patterns that closely match historical fire records. Independent validations conducted using three active-fire datasets (MODIS C6.1, J1 VIIRS C2 and SUOMI VIIRS C2) for 2021-2024 produce average AUC values of 0.8576, 0.8160 and 0.8174, respectively, supporting the predictive ability of the map across multiple fire products. The results of a driver analysis indicate that vegetation (based on the normalized difference vegetation index and leaf area index) and meteorological factors (based on the fire weather index) dominate in the spring and autumn (with autumn amplified by drought); summer risk is more strongly regulated by human activity (human footprint); and winter risk is correlated most strongly with temperature, drought indicators (drought code) and socioeconomic variables (gross domestic product). SHAP-based explainability tests further reveal that driver effects change gradually within typical value ranges but intensify sharply once certain extreme thresholds are exceeded. This study presents a scalable, robust framework for fire hazard prediction and interpretation that can support sustainable management and risk mitigation strategies for the TNSP region and similar large-scale ecological projects.

Lightning is a major natural ignition source of wildfires across forest, grassland, and cropland ecosystems. Accurate prediction of lightning-ignited fire occurrence remains challenging due to uncertainties in spatiotemporal alignment caused by vegetation-dependent smoldering delays and the difficulty of representing heterogeneous fuel conditions in mixed-vegetation regions. This study proposes a semi-automated lightning-fire alignment framework that integrates land cover information and historical fire records to improve spatiotemporal matching across different vegetation types and to reduce misclassification from human-induced fires in agricultural areas. To better characterize fuel conditions, two feature-level vegetation fusion parameters-total vegetation cover and leaf area index weight-are introduced and combined with hourly meteorological variables and lightning characteristics to develop a tuned random forest prediction model. The framework is applied at a regional scale in the Greater Khingan Mountains and southwestern forest regions of China, with predictions conducted at an event-based temporal scale using hourly inputs. The vegetation-fused model achieves an AUC of 0.93, outperforming models without vegetation fusion. Analysis of model outputs indicates that hourly maximum temperature, leaf area index weight, precipitation, and wind speed are key factors influencing lightning-ignited fire occurrence. This study demonstrates the value of semi-automated alignment and vegetation feature fusion for improving lightning-ignited fire prediction in heterogeneous landscapes, supporting regional wildfire risk assessment and potential early-warning applications.

This study integrates multi-source driving factors from 2005 to 2024, including meteorological elements, fire danger indices, vegetation, topography, and human activities, to comparatively analyze the performance of four machine learning models—RF, LightGBM, XGBoost, and DNN—in the daily scale prediction of forest and grassland fires in Sichuan Province. A Nested Cross-Validation framework, using year as the grouping variable, was employed to evaluate model robustness, and the SHAP method was introduced to quantify the driving mechanisms of the factors. The results indicate that LightGBM is the overall optimal model, with its ROC AUC reaching 0.9193 in nested cross-validation and 0.9411 in independent temporal testing, demonstrating superior predictive capability and cross-year generalization performance. SHAP analysis reveals a hierarchical structure of fire drivers: land type constitutes the a priori physical constraint for fire occurrence, meteorological and drought indicators dominate the differentiation of the risk gradient, while topography and human activities serve as spatial modulators. In an independent sample validation on 9 December 2024, high-fire-risk grids showed high consistency with the distribution of actual fire points. For transmission line application scenarios, a risk distribution map constructed by coupling fire risk values with wind speed thresholds successfully identified actual fire point areas, indicating that this framework can provide a scientific basis for forest and grassland fire prevention and for power grid enterprises to conduct precise early warning and regionalized control.

Forest fires are usually caused by lightning or by human-beings, which are called lightning-caused fire or human-caused fires. Fire cause identification is crucial for fire case studies and subsequent forest fire prevention and control. a large amount of historical forest fire records are classified as with “unknown fire cause,” which hinders effective fire management. To address this issue, we collected eight categories of 27 variables related to meteorology, vegetation, topography, and human activities in Heilongjiang Province, China, and such variables will be used for fire cause identification. We analyzed the peak values of high-incidence variables for each fire cause and examined the spatial clustering of different fire causes using spatial analysis techniques. Eight machine learning algorithms were evaluated via the entropy weight-TOPSIS method to identify the optimal model for fire cause prediction, which was then applied to predict unknown fire causes. The results indicate that lightning-caused fires occur predominantly in the Greater Khingan Mountains, smoking-related fires cluster near the border of Heihe and Hegang, and agricultural/forestry fires are distributed across multiple fire-prone regions in Heilongjiang. The entropy weight-TOPSIS analysis identified the Multilayer Perceptron (MLP) algorithm as the optimal model, with a comprehensive score of 0.91. Key drivers for MLP included DME, elevation, AWCE, and vegetation type, all with SHAP values exceeding 0.15. The MLP model demonstrated high prediction accuracy in major fire-prone areas such as the Greater Khingan Mountains, addressing regional fire cause prediction challenges. Furthermore, for approximately 52.4% of unknown fires, MLP provided two probable causes. and the sum of the predicted probabilities of these two fire causes exceeds 80%. This study proposes a comprehensive framework for forest fire cause prediction, capable of identifying causes for historically unknown fires and assisting in future fire cause determination. The approach provides a scientific basis for fire management, enhances prevention efficiency, and offers critical insights for forest fire risk mitigation.

Climate change is exacerbating global forest fire regimes, creating an urgent demand for sophisticated predictive modeling tools. Focusing on the climatically sensitive and ecologically critical Daxing'anling Mountains in China, this study develops a hybrid deep learning framework termed CNN-ATT-SYNERGY. This model integrates Convolutional Neural Networks (CNN) with attention mechanisms to fuse multi-source environmental covariates, encompassing meteorological conditions, topographic features, lightning activity, anthropogenic disturbances, socioeconomic factors and historical fire records, thereby enabling robust forest fire occurrence assessment. Trained on a comprehensive dataset involving 3,368 fire incidents across 20 forestry bureaus, the proposed model delivers state-of-the-art performance, with an overall accuracy of 81.22% and an AUC of 87.97%, surpassing conventional benchmark models including Support Vector Machine (SVM) and Random Forest. In addition to its superior predictive capacity, the model effectively identifies dynamic spatiotemporal fire risk patterns: fire risks cluster in eastern regions during spring and autumn, migrate westward in summer, and localized high-risk zones are also detected in winter. Accordingly, the CNN-ATT-SYNERGY framework serves as a seasonally adaptive and scalable decision-support instrument for forest fire management, providing a refined technical solution for boreal forest regions grappling with intensifying wildfire risks.

This study treated typical forest surface fuels in China's Three-North Region with Armillaria mellea to explore a green, economical fire risk reduction approach. Using multiple analytical methods (ANOVA, correlation, PCA, cluster analysis, exponential degradation modeling, random forest, and SEM), the results show a pronounced time-dependent cumulative effect. Within 120 days, larch lignin loss reached 24.68%; birch's comprehensive combustion index dropped from 46.31 & times; 10(-7) to 17.08 & times; 10(-7), and its peak ignition point advanced by 15-30 days. During the efficient degradation period (R-2 > 0.7), mixed fuel achieved the best Olson model fit (R-2 = 0.90004). Thermogravimetric analysis revealed a distinct pyrolysis plateau in birch near 420 degrees C, while larch's lignin exothermic peak collapsed and shifted forward. PCA with cluster analysis categorized birch combustibility into three stages, transitioning to the lowest level at day 105, whereas larch remained in a low-combustibility stage for longer. SEM indicated that fungal-induced lignin enrichment reduces combustibility, especially in early degradation. In summary, Armillaria mellea not only reduces fuel load but also alters chemical composition and thermophysical properties to mitigate fire risk, confirming its potential as a biological fire retardant and providing a basis for sustainable microbial fire prevention. Future research should integrate field trials and multi-species synergy.

Near-surface wind field simulation in complex mountainous terrain is essential for predicting wildfire behavior and supporting fire risk management. WindNinja, a widely used diagnostic wind downscaling model, is strongly dependent on its initial input data; however, systematic evaluations of its input sensitivity and simulation accuracy remain limited. In this study, a representative canyon area was selected as the study site. WindNinja was driven by three types of input data: local meteorological station observations, national meteorological station observations, and ERA5-Land reanalysis data. Two indices—the Wind Forcing Intensity (WFI) index and the Thermal Forcing Intensity (TFI) index—were constructed to classify weather-forcing scenarios and evaluate simulation accuracy under different conditions. The results show that differences in the statistical characteristics of the initial wind sources produce pronounced sensitivity in WindNinja simulations. Simulations driven by local meteorological observations generally overestimate wind speed, whereas ERA5-Land-driven simulations systematically underestimate wind speed, with national-station results falling between these two cases. Simulation accuracy varies with terrain position: wind direction errors dominate in valleys, whereas wind speed errors dominate on ridges and hilltops. Weather background conditions significantly influence simulation accuracy. Wind forcing intensity dominates the magnitude and dispersion of simulation errors, while strong thermal forcing leads to an overall decline in simulation accuracy and stability. These findings highlight the sensitivity of WindNinja to initial wind sources and weather background conditions in complex terrain and provide guidance for its application and uncertainty control in wildfire behavior modeling.

To identify the major drivers of lightning-caused fire and to develop predictive models in the Greater Khingan Mountains, we integrated multi-source data from 2021 to 2024, including lightning records, forest fire records, and multi-source environmental data. We identified candidate lightning events using a spatiotemporal mat-ching approach, and addressed the imbalance problem of sample type using the EasyEnsemble method to construct the model dataset. We then selected predictor variables through correlation analysis, collinearity diagnosis, and feature importance analysis, developed multiple statistical and machine learning models, and interpreted the final model by SHAP (SHapley Additive exPlanations). The results showed that a total of 103 lightning events associated with lightning-caused fires were identified. The average distance between lightning strikes and fire ignition points was 0.84 km, with a mean holdover time of 1.4 days. 83.5% of fires were detected within three days after lightning occurrence. The predictive model identified the duff moisture code, drought code, and hourly fire weather index as the primary predictors, with temperature and wind speed as the secondary predictors. The random forest model showed the best overall performance, with an AUC of 0.7959, sensitivity of 0.7229, and specificity of 0.7752. The optimal classification thresholds were mainly concentrated between 0.45 and 0.55. The occurrence of lightning-caused fire was jointly influenced by fire weather, meteorological, topographic, and lightning-related factors, exhibiting significant nonlinear and threshold effects.

A large amount of fuel has accumulated in the forests of Northeast China. Scientific regulation of surface fuel is crucial for reducing forest fire risk. To explore the effectiveness of different fuel regulation treatments, we set up six regulation treatments in Pinus sylvestris var. mongolica plantations in the Daxing'anling Mountains: low, medium, and high strength (different degrees of mowing, shrub clearing, pruning, and clearing dead surface fuel), surface clearing, fuel load enhancement, and shrub clearing, with untreated stands as the control. The effects of different treatments on physicochemical properties, fire behavior, and pyrolysis characteristics of dead surface fuel were examined. The results showed that high-strength treatment significantly increased ash content, ignition point, and water content of the fuel by 58.0%, 4.3 ℃, and 60.9%, respectively, while crude fat content, load, flame height, and maximum combustion temperature significantly decreased by 7.5%, 74.0%, 43.8%, and 100.8 ℃. In the surface clearing treatment, water content of the fuel significantly increased, while crude fat content, load, and rate of spread significantly decreased. In the fuel load enhancement treatment, crude fat content, load, and rate of spread of the fuel significantly increased. In the shrub clearing treatment, crude fat and ash content of the fuel significantly increased, while the rate of spread significantly decreased. In the medium-strength treatment, the ignition point of the fuel significantly increased, while the flame height significantly decreased. In the low-strength treatment, the spread rate of the fuel significantly increased. Compared with the control, the holocellulose degradation temperature range of each treatment increased significantly. After shrub clearing, the peak and average weight loss rates of holocellulose, the exothermic peak area, the percentage of lignin loss, and the total consumption decreased significantly, while the differences in other treatments were not significant. The principal component analysis comprehensive flammability ranking showed that the flammability from high to low was: fuel load enhancement treatment > medium-strength treatment > shrub clearing treatment > untreated > low-strength treatment > surface clearing treatment > high-strength treatment. In summary, the flammability of dead surface fuel in Pinus sylvestris var. mongolica forests significantly decreased under high-strength treatment, which could be used as a fuel regulation treatment for short-term fire prevention.

The Daxing’anling Mountains, as a climate-sensitive region, are experiencing forest fires that threaten the area’s ecological security. Nevertheless, most of the existing fire prediction models are stationary. They do not have an all-embracing scheme for simultaneously managing fire ignition causes, dynamic fire scenarios and spatial targeting. Hence, the development of an accurate and efficient forest fire forecasting system is vital. This study establishes a prediction framework that integrates long-term survey data with multi-source remote sensing, incorporating spatiotemporal clustering, spatial autocorrelation and an optimised ensemble of LR–RF–SVM–GBDT algorithms. Among the 3368 recorded fire incidents, lightning-ignited fires accounted for 51.19%, making lightning storms the predominant cause of ignition. While the frequency of lightning-induced fires increased significantly (1.24 per year, p < 0.05), the total burned area remained relatively stable. The proposed framework outperformed individual models by achieving higher predictive metrics (accuracy = 0.89, AUC = 0.94, F1 = 0.89) and providing robust support for operational early warning and real-time management. The projections for future climate, based on the SSP126 and SSP585 scenarios, depict a notable geographical shift in fire-prone areas. Besides the traditionally known eastern areas of Xiaogenhe and Chabanhe, which are expected to see an increase in fire occurrences, new high-fire-risk areas are expected to emerge in the central–western regions, such as Huzhong and Wuyuan. Quantitative findings reveal that the divergence in forest fire probabilities between the high-emission SSP585 and SSP126 scenarios will increase over time. The expected increase ranges from 0.29% in the 2030s to 0.92% in the 2050s, then rises to 4.48% in the 2070s and reaches 6.48% by the 2090s. These figures highlight the urgency of implementing fire management practices that are not only adaptive but also specific to particular areas. The scenario-based forecasts represent a proactive approach to assisting forest fire governance under climate change, providing a basis for future decisions as quantitative evidence.

Border zones between countries are among the world regions with large spatial variability in wildfires. However, our understanding of fire regimes in these areas, particularly quantitative assessments of cross-border fire risk, remains limited. This study examines fire regimes and their driving factors along China's extensive land borders, which encompass the world's longest border and the second-highest number of neighboring countries. Utilizing Moderate Resolution Imaging Spectroradiomete satellite products, we analyzed fire occurrence (fire count), intensity (fire radiative power), and impact (burned area) within 50-km buffer zones on both sides of the border, their relationships with vegetation, and the statistical differences across borders. The results indicate that approximately 10 % of China's land border areas face threats from cross-border fires, primarily originating from Russia, Mongolia, and Laos. The wildfire metrics in neighboring countries-notably the fire count and cumulative FRP in Russia, Myanmar, and Laos and the total burned area in Mongolia and Russia-were significantly greater than those in China. However, the differences in vegetation between inside and outside the borders were not significant. The results suggest that the fire disparities were primarily driven by divergent fire management policies and resource allocation. We argue that effective mitigation of transboundary fire risk requires an integrated approach that combines regional governance, targeted prevention measures, and strengthened international cooperation. Quantifying policy-driven disparities in fire regimes provides a critical baseline for coordinated management in global border hotspots.

Wildfires critically affect ecosystems, carbon cycles, and public health. COVID-19 restrictions provided a unique opportunity to study human activity’s role in wildfire regimes. This study presents a comprehensive evaluation of pandemic-induced wildfire regime changes across global fire-prone regions. Using MODIS data (2010–2022), we analyzed fire patterns during the pandemic (2020–2022) against pre-pandemic baselines. Key findings include: (a) A 22% global decline in wildfire hotspots during 2020–2022 compared to 2015–2019, with the most pronounced reduction occurring in 2022; (b) Contrasting regional trends: reduced fire activity in tropical zones versus intensified burning in boreal regions; (c) Stark national disparities, exemplified by Russia’s net increase of 59,990 hotspots versus Australia’s decrease of 60,380 in 2020; (d) Seasonal shifts characterized by December declines linked to mobility restrictions, while northern summer fires persisted due to climate-driven factors. Notably, although climatic factors predominantly govern fire regimes in northern latitudes, anthropogenic ignition sources such as agricultural burning and accidental fires substantially contribute to both fire incidence and associated emissions. The pandemic period demonstrated that while human activity restrictions reduced ignition sources in tropical regions, fire activity in boreal ecosystems during these years exhibited persistent correlations with climatic variables, reinforcing climate’s pivotal—though not exclusive—role in shaping fire regimes. This underscores the need for integrated wildfire management strategies that address both human and climatic factors through regionally tailored approaches. Future research should explore long-term shifts and adaptive management frameworks.

This study investigates the latency of lightning-caused fires in the boreal coniferous forests of the Greater Khingan Mountains, employing advanced machine learning techniques to analyze the relationship between meteorological factors, lightning characteristics, and fire ignition and smoldering processes. Using the Random Forest Model (RFM) combined with Recursive Feature Elimination with Cross-Validation (RFECV) and SHapley Additive exPlanations (SHAP), the study identifies key factors influencing fire latency. Two methods, Min distance and Min latency, were used to determine ignition lightning, with the Min distance method proving more reliable. The results show that lightning-caused fires cluster spatially and peak temporally between May and July, aligning with lightning activity. The Fine Fuel Moisture Code (FFMC) and precipitation were identified as the most influential factors. This study underscores the importance of fuel moisture and weather conditions in determining latency of lightning-caused fire, offering valuable insights for enhancing early warning systems. Despite limitations in data resolution and the exclusion of topographic factors, this study advances our understanding of lightning-fire latency mechanisms and provides a foundation for more effective wildfire management strategies under climate change.

Smoke plume dynamics involve various smoke processes and mechanics in the atmosphere and provide the scientific foundation for the development of tools to simulate and predict smoke and its environmental and human impacts. The increasing occurrence of wildfires and the demands for more extensive application of prescribed fires in the U.S. have posed great challenges and immediate actions for advancing smoke plume dynamics and improving smoke predictions and impact assessments to mitigate smoke impacts. Numerous efforts have been made recently to address these needs and challenges. This paper synthesizes advances in smoke plume dynamics research mainly conducted in the U.S. in the recent decade, identifies gaps, and suggests future research needs. The main advances include smoke data collections from comprehensive field campaigns, new satellite products, improved understanding of smoke plume properties and chemistry, structure and evolution, evaluation and improvement of smoke modeling and prediction systems, the development of coupled smoke models, and applications of machine-learning techniques. The major remaining gaps are the lack of comprehensive simultaneous measurements of smoke, fuels, fire, and atmospheric interactions during wildfires, high-resolution coupled modeling systems of these components, and real-time smoke prediction capacity. The findings from this synthesis study are expected to support smoke research and management to meet various challenges under increasing wildland fires and impacts.

There was an error in the original publication [...]

Fire radiative power (FRP) is a key indicator for evaluating the intensity of wildfires, unlike traditional real-time fire lines or combustion areas that only provide binary information, and its accurate prediction is more important for firefighting actions and environmental pollution assessment. To this end, we used a combination of data from geostationary satellites and polar orbit satellites to correct the FRP data. Incorporating various factors that affect wildfire spread, such as meteorological conditions, topography, vegetation indexes, and population density, we constructed a comprehensive California wildfire spread dataset, covering information since 2017. Then, we established a deep learning framework that integrates various modules to analyze multimodal data for the accurate prediction of FRP imagery. We investigated the impact of input sequence length and loss function design on model predictive performance, leading to subsequent model optimization. Furthermore, our model has demonstrated acceptable performance in transfer learning and multistep prediction, emphasizing its application value in wildfire prediction and management. It can provide more detailed information about wildfire spread, showcasing the powerful capability of deep learning to process multimodal data and its potential in the emerging field of real-time FRP prediction.

Forest fires are a frequent and destructive phenomenon in Southwestern China, posing significant threats to ecological systems and human lives and property. In response to the growing need for effective forest fire prevention, this study introduces an innovative method for predicting and assessing forest fire risk. By integrating multi-source data, including optical and microwave remote sensing, meteorological, topographic, and human activity data, the approach enhances the sensitivity of risk models to vegetation water content and other critical factors. The vegetation water content is derived from both Vegetation Optical Depth and optical remote sensing data, allowing for a more accurate assessment of changes in vegetation moisture that influence fire risk. A time series prediction model, incorporating attention mechanisms, is used to assess the probability of fire occurrence. Additionally, the method includes fire spread simulations based on Cellular Automaton and Monte Carlo approaches to evaluate potential burn areas. This combined approach can provide a comprehensive fire risk assessment using the probability of both fire occurrence and potential fire spread. Experimental results show that the integration of microwave data and attention mechanisms improves prediction accuracy by 2.8%. This method offers valuable insights for forest fire management, aiding in targeted prevention strategies and resource allocation.

Due to natural factors and influences from neighboring countries,wild fires frequently occur in China's border areas.To quantify the activities of wild fires in border areas,we analyzed the regime of wild fires within a 2 km buffer zone on both sides of China's land borders based on MODIS fire spot data,including fire types,fre-quency,seasonality,and spatial distribution.Between 2001 and 2022,a total of 25918 vegetation fires occurred in China's border regions,with forests,cropland,and grasslands accounting for 42.0％,30.4％,and 14.4％of the fire incidents,respectively.Forest fires were most common in broadleaved forests.Cropland fires mainly resulted from traditional farming practices and the lack of fire prevention awareness among border residents,which often caused fires to spread to nearby forests,leading to forest fires.Among grassland fires,meadow steppe posed the highest risk,and grassland fires in forest-grassland ecotones were likely to trigger forest fires.There were significant differences in fire types and seasonal distribution across regions.In the northeastern border region,grassland fires,deciduous broadleaved forest fires,and cropland fires were predominant,with spring and autumn being the primary seasons for fire occurrences,especially in April and October.In the southwestern border region,evergreen broad-leaved forest fires and cropland fires were predominant,with spring and winter being peak periods for fires,espe-cially in March and December.In the northwestern border region,grassland fires and cropland fires were predomi-nant,with more vegetation fires occurring in summer and autumn,peaking in September.Within a 2 km range on both sides of the border,the number of fire spots outside the country far exceeded those within,particularly in the border areas of Inner Mongolia,Jilin,Yunnan,and Guangxi,increasing the risk of cross-border fires in these regions.Fire spots showed significant clustering,with major clusters found in the border region of Xishuangbanna Dai Autonomous Prefecture in Yunnan,Hulunbuir City in Inner Mongolia,Huma County and Jiamusi City in Hei-longjiang,and Hunchun City in Jilin.Different fire prevention strategies should be developed based on the charac-teristics of vegetation fires in different border regions,targeting vegetation types,seasonal periods,and clustering areas prone to fires,to implement effective vegetation fire prevention and control measures in border areas.