Search Results (83)

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. Full article

In the Mediterranean basin, wildfires burn an average of 600,000 ha per year, causing severe ecological, economic, and social impacts. Fire behavior modeling is essential for wildfire prevention and control. Three-dimensional physics-based fire behavior models, such as Fire Dynamics Simulator (FDS), can represent heterogeneous fuels and simulate fire behavior processes with greater detail than conventional models. However, they require accurate information about species composition and 3D distribution of fuel mass and bulk density at the voxel level. Working in a Mediterranean ecosystem study area we developed a methodology based on the use of geometric and spectral features from UAS-based digital aerial photogrammetric point clouds for (i) species segmentation and classification using machine learning algorithms, (ii) generation of biomass prediction models at individual plant level, and (iii) creation of 3D fuel scenarios and modeling wildfire behavior. Field measurements were conducted on 22 circular plots with a radius of 5 m. Data from the field measurements, combined with species-specific allometric equations, were used for the evaluation of classification and prediction models. Fire behavior variables such as rate of spread, heat release rate, and mass loss rate were monitored and assessed as outputs from 20 different scenarios using FDS. The overall species classification accuracy was 80.3%, and the biomass regression R² values obtained by cross-validation were 0.77 for Pinus halepensis and 0.83 for Anthyllis cytisoides. These results are encouraging further improvement based on the integration of sensors onboard UAS, and the characterization of fuels for fire behavior modeling. These high-resolution fuel representations can be coupled with standard risk assessment tools, enabling fire managers to prioritize treatment areas and plan for resource deployment. Full article

The increasing frequency and intensity of wildfires pose serious threats to ecosystems, property, and human safety worldwide. Accurate semantic segmentation of wildfire images is essential for real-time fire monitoring, spread prediction, and disaster response. However, existing deep learning methods heavily rely on large volumes of pixel-level annotated data, which are difficult and costly to obtain in real-world wildfire scenarios due to complex environments and urgent time constraints. To address this challenge, we propose a semi-supervised wildfire image segmentation framework that enhances segmentation performance under limited annotation conditions by integrating multi-scale structural information fusion and pixel-level contrastive consistency learning. Specifically, a Lagrange Interpolation Module (LIM) is designed to construct structured interpolation representations between multi-scale feature maps during the decoding stage, enabling effective fusion of spatial details and semantic information, and improving the model’s ability to capture flame boundaries and complex textures. Meanwhile, a Pixel Contrast Consistency (PCC) mechanism is introduced to establish pixel-level semantic constraints between CutMix and Flip augmented views, guiding the model to learn consistent intra-class and discriminative inter-class feature representations, thereby reducing the reliance on large labeled datasets. Extensive experiments on two public wildfire image datasets, Flame and D-Fire, demonstrate that our method consistently outperforms other approaches under various annotation ratios. For example, with only half of the labeled data, our model achieves 5.0% and 6.4% mIoU improvements on the Flame and D-Fire datasets, respectively, compared to the baseline. This work provides technical support for efficient wildfire perception and response in practical applications. Full article

Wildfires have become increasingly frequent and intense due to climate change, posing severe threats to ecosystems, infrastructure, and human lives. As a result, accurate wildfire spread prediction is critical for effective risk mitigation, resource allocation, and decision making in disaster management. In this study, we develop a deep learning model to predict wildfire spread using remote sensing data. We propose LinU-Mamba, a model with a U-Net-based vision Mamba architecture, with light spatial attention in skip connections, and an efficient linear attention mechanism in the encoder and decoder to better capture salient fire information in the dataset. The model is trained and evaluated on the two-dimensional remote sensing dataset Next Day Wildfire Spread (NDWS), which maps fire data across the United States with fire entries, topography, vegetation, weather, drought index, and population density variables. The results demonstrate that our approach achieves superior performance compared to existing deep learning methods applied to the same dataset, while showing an efficient training time. Furthermore, we highlight the impacts of pre-training and feature selection in remote sensing, as well as the impacts of linear attention use in our model. As far as we know, LinU-Mamba is the first model based on Mamba used for wildfire spread prediction, making it a strong foundation for future research. Full article

Wildfires increasingly threaten human life, ecosystems, and infrastructure, with events like the 2025 Palisades and Eaton fires in Los Angeles County underscoring the urgent need for more advanced prediction frameworks. Existing physics-based and deep-learning models struggle to capture dynamic wildfire spread across both 2D and 3D domains, especially when incorporating real-time, multimodal geospatial data. This paper explores how generative artificial intelligence (AI) models—such as GANs, VAEs, and transformers—can serve as transformative tools for wildfire prediction and simulation. These models offer superior capabilities in managing uncertainty, integrating multimodal inputs, and generating realistic, scalable wildfire scenarios. We adopt a new paradigm that leverages large language models (LLMs) for literature synthesis, classification, and knowledge extraction, conducting a systematic review of recent studies applying generative AI to fire prediction and monitoring. We highlight how generative approaches uniquely address challenges faced by traditional simulation and deep-learning methods. Finally, we outline five key future directions for generative AI in wildfire management, including unified multimodal modeling of 2D and 3D dynamics, agentic AI systems and chatbots for decision intelligence, and real-time scenario generation on mobile devices, along with a discussion of critical challenges. Our findings advocate for a paradigm shift toward multimodal generative frameworks to support proactive, data-informed wildfire response. Full article

Accurate real-time tracking of forest fires using UAV platforms is crucial for timely early warning, reliable spread prediction, and effective autonomous suppression. Existing detection-based multi-object tracking methods face challenges in accurately associating targets and maintaining smooth tracking trajectories in complex forest environments. These difficulties stem from the highly nonlinear movement of flames relative to the observing UAV and the lack of robust fire-specific feature modeling. To address these challenges, we introduce AO-OCSORT, an association-optimized observation-centric tracking framework designed to enhance robustness in dynamic fire scenarios. AO-OCSORT builds on the YOLOX detector. To associate detection results across frames and form smooth trajectories, we propose a temporal–physical similarity metric that utilizes temporal information from the short-term motion of targets and incorporates physical flame characteristics derived from optical flow and contours. Subsequently, scene classification and low-score filtering are employed to develop a hierarchical association strategy, reducing the impact of false detections and interfering objects. Additionally, a virtual trajectory generation module is proposed, employing a kinematic model to maintain trajectory continuity during flame occlusion. Locally evaluated on the 1080P-resolution FireMOT UAV wildfire dataset, AO-OCSORT achieves a 5.4% improvement in MOTA over advanced baselines at 28.1 FPS, meeting real-time requirements. This improvement enhances the reliability of fire front localization, which is crucial for forest fire management. Furthermore, AO-OCSORT demonstrates strong generalization, achieving 41.4% MOTA on VisDrone, 80.9% on MOT17, and 92.2% MOTA on DanceTrack. Full article

Wildfires pose significant threats to ecosystems, human lives, and property worldwide. Understanding the behavior of fire spread on sloped terrain is essential for developing effective firefighting strategies and improving fire prediction models. Previous research has successfully demonstrated the accuracy of numerical tools in comparison to laboratory experiments. This study focuses on the influence of terrain slope and wind speed on wildland fire behavior using Computational Fluid Dynamics (CFD) simulations. In the first phase, the numerical model was validated for a 5 m high single Douglas Fir tree under various mesh sizes, yielding heat release and mass loss rates in close agreement with experimental data. The second phase extends the model to simulate a plantation of 66 Douglas Fir trees under varying slopes and wind conditions. The results indicate that a downward slope of 30° reduces the peak heat release rate, while an upward slope of 30° increases it, with wind speed amplifying these effects. Based on these data, a new reduced-order model is proposed to quantify the influence of slope angle on the heat release rate (HRR) in wildland fires. These findings are critical for enhancing predictive fire models and mitigating wildfire risks in complex terrains. Full article

Live fuel moisture content (LFMC) significantly influences fire activity and behavior over different spatial and temporal scales. The ability to estimate LFMC is important to improve our capability to predict when and where large wildfires may occur. Currently, there is a gap in providing reliable near-real-time LFMC estimates which can contribute to better operational decision-making. The objective of this work was to develop near-real-time LFMC estimates for operational purposes in Portugal. We modelled LFMC using Random Forests for Portugal using a large set of potential predictor variables. We validated the model and analyzed the relationships between estimated LFMC and both fire size and behavior. The model predicted LFMC with an R² of 0.78 and an RMSE of 12.82%, and combined six variables: drought code, day-of-year and satellite vegetation indices. The model predicted well the temporal LFMC variability across most of the sampling sites. A clear relationship between LFMC and fire size was observed: 98% of the wildfires larger than 500 ha occurred with LFMC lower than 100%. Further analysis showed that 90% of these wildfires occurred for dead and live fuel moisture content lower than 10% and 100%, respectively. Fast-spreading wildfires were coincident with lower LFMC conditions: 92% of fires with rate of spread larger than 1000 m/h occurred with LFMC lower than 100%. The availability of spatial and temporal LFMC information for Portugal will be relevant for better fire management decision-making and will allow a better understanding of the drivers of large wildfires. Full article

This study investigates the physical interactions and between forest fires and the atmosphere, which often lead to conditions favourable to instability and the formation of pyrocumulus (PyCu). Using the coupled atmosphere–fire spread modelling framework, WRF-SFIRE, the Portuguese October 2017 Quiaios wildfire, in association with tropical cyclone Ophelia, was simulated. Fire spread was imposed via burnt area data, and the fire’s influence on the vertical and surface atmosphere was analysed. Simulated local atmospheric conditions were influenced by warm and dry air advection near the surface, and moist air in mid to high levels, displaying an inverted “V” profile in thermodynamic diagrams. These conditions created a near-neutrally unstable atmospheric layer in the first 3000 m, associated with a low-level jet above 1000 m. Results showed that vertical wind shear tilted the plume, resulting in an intermittent, high-based, shallow pyroconvection, in a zero convective available potential energy environment (CAPE). Lifted parcels from the fire lost their buoyancy shortly after condensation, and the presence of PyCu was governed by the energy output from the fire and its updrafts. Clouds formed above the lifted condensation level (LCL) as moisture fluxes from the surface and released from combustion were lifted along the fire plume. Clouds were primarily composed of liquid water (1 g/kg) with smaller traces of ice, graupel, and snow (up to 0.15 g/kg). The representation of pyroconvective dynamics via coupled models is the cornerstone of understanding the phenomena and field applications as the computation capability increases and provides firefighters with real time extreme fire conditions or predicting ahead of time. Full article

Accurate wildfire spread modeling critically depends on the representation of wind dynamics, which vary with terrain, land cover characteristics, and height above ground. Many fire spread models are often coupled with coarse atmospheric grids that cannot explicitly resolve the vertical variation of wind near flame heights. Rothermel’s fire spread model, a widely used parameterization, relies on midflame wind speed to calculate the fire rate of spread. In coupled fire atmosphere models such as the Community Fire Behavior Model (CFBM), users are required to specify the midflame height before running a fire spread simulation. This study evaluates the use of logarithmic interpolation wind adjustment factors (WAF) for improving midflame wind speed estimates, which are critical for the Rothermel model. We compare the fixed wind height approach that is currently used in CFBM with WAF-derived winds for unsheltered and sheltered surface fire spread. For the first time in this context, these simulations are validated against satellite and ground-based observations of fire perimeters. The results show that WAF implementation improves fire perimeter predictions for both grass and canopy fires while reducing the overestimation of fire spread. Moreover, this approach solely depends on the fuel bed depth and estimation of canopy density, enhancing operational efficiency by eliminating the need for users to specify a wind height for simulations. Full article

Arid and semiarid ecosystems face significant water scarcity due to high evaporation rates exceeding precipitation. This study examines temporal variations in water relations of two woody species, Vachellia schaffneri (S. Watson) Seigler & Ebinger, and Prosopis laevigata (Humb. & Bonpl. ex Willd.) M.C. Johnst, and one epiphyte, Tillandsia recurvata (L.) L. (Bromeliaceae), to assess their drought tolerance and water storage capacity. We hypothesized that species with greater water storage capacity would exhibit lower drought tolerance due to reduced osmotic adjustments, whereas species with lower storage capacity would maintain turgor through osmotic regulation and cell wall rigidity. Predawn and midday water potentials (Ψ_pd, Ψ_md) were measured, and pressure–volume (P–V) curves were used to derive parameters such as saturated water content (SWC), osmotic potential (π_o), turgor loss point (Ψ_TLP), relative water content at Ψ_TLP (RWC_TLP), bulk modulus of elasticity (ε), and full turgor capacitance (C_FT). Significant correlations were found between C_FT and Ψ_TLP (positive), π_o (positive), and ε (negative). P. laevigata and T. recurvata exhibited higher water storage capacities (41.46 and 26.45 MPa⁻¹, respectively) but had a lower ability to maintain cell turgor under drought conditions. In contrast, V. schaffneri exhibited the lowest water storage capacity (11.88 MPa⁻¹) but demonstrated the highest ability to maintain cell turgor (Ψ_TLP = −1.31 MPa) and superior osmotic adjustments (π_o = −0.59 MPa). Both V. schaffneri and P. laevigata exhibited rigid cell walls, whereas T. recurvata displayed greater elasticity in its cell structures. The lowest moisture content in V. schaffneri suggests increased flammability and fire spread potential. Future studies should focus on live fuel moisture content across more species, explore seasonal variations in hydraulic traits, and integrate these physiological parameters into fire risk models to enhance wildfire prediction and management. Full article

The increasing frequency and intensity of wildfires highlight the need to develop more efficient tools for firefighting and management, particularly in the field of wildfire spread prediction. Classical wildfire spread models have relied on mathematical and empirical approaches, which have trouble capturing the complexity of fire dynamics and suffer from poor flexibility and static assumptions. The emergence of machine learning (ML) and, more specifically, deep learning (DL) has introduced new techniques that significantly enhance prediction accuracy. ML models, such as support vector machines and ensemble models, use tabular data points to identify patterns and predict fire behavior. However, these models often struggle with the dynamic nature of wildfires. In contrast, DL approaches, such as convolutional neural networks (CNNs) and convolutional recurrent networks (CRNs), excel at handling the spatiotemporal complexities of wildfire data. CNNs are particularly effective at analyzing spatial data from satellite imagery, while CRNs are suited for both spatial and sequential data, making them highly performant in predicting fire behavior. This paper presents a systematic review of recent ML and DL techniques developed for wildfire spread prediction, detailing the commonly used datasets, the improvements achieved, and the limitations of current methods. It also outlines future research directions to address these challenges, emphasizing the potential for DL to play an important role in wildfire management and mitigation strategies. Full article

In the context of global climate warming, climate change is subtly reshaping the patterns of wildfires. Therefore, it is particularly urgent to conduct in-depth research on climate change, wildfires, and their management strategies. This study relies on detailed fire point data from 2001 to 2020, skillfully incorporating a spatial autocorrelation analysis to uncover the mysteries of spatial heterogeneity, while comprehensively considering the influences of multiple factors such as climate, terrain, vegetation, and socioeconomic conditions. To simulate fire conditions under future climates, we adopted the BCC-CSM2-MR climate model, presetting temperature and precipitation data for two scenarios: a sustainable low-development path and a high-conventional-development path. The core findings of the study include the following: (i) In terms of spatial heterogeneity exploration, global autocorrelation analysis reveals a striking pattern: within the southern forest region, 63 cities exhibiting a low–low correlation are tightly clustered in provinces such as Hubei, Anhui, and Zhejiang, while 48 cities with a high–high correlation are primarily distributed in Guangxi and Guangdong. Local autocorrelation analysis further refines this observation, indicating that 24 high–high correlated cities are highly concentrated in specific areas, 14 low–low correlated cities are located in Hainan, and there are only 3 sparsely distributed cities with a low–high correlation. (ii) During the model construction and validation process, this study innovatively adopted the LR-RF-SVM ensemble model, which demonstrated exceptional performance indicators: an accuracy of 91.97%, an AUC value of 97.09%, an F1 score of 92.13%, a precision of 90.75%, and a recall rate of 93.55%. These figures, significantly outperforming those of the single models SVM and RF, strongly validate the superiority of the ensemble learning approach. (iii) Regarding predictions under future climate scenarios, the research findings indicate that, compared to the current fire situation in southern forest areas, the spatial distribution of wildfires will exhibit a noticeable expansion trend. High-risk regions will not only encompass multiple cities in Hunan, Hubei, southern Anhui, all of Jiangxi, and Zhejiang but will also extend northward into southern forest areas that were previously considered low-risk, suggesting a gradual northward spread of fire risk. Notably, despite the relatively lower fire risk in some areas of Fujian Province under the SS585 scenario, overall, the probability of wildfires occurring in 2090 is slightly higher than that in 2030, further highlighting the persistent intensification of forest fire risk due to climate change. Full article

We describe enhancements to an operational forecast system based on the Weather Research and Forecasting (WRF) model for the prediction of high-impact weather events affecting power utilities, particularly conditions conducive to wildfires. The system was developed for Pacific Gas and Electric Corporation (PG&E) to forecast conditions in Northern and Central California for critical decision-making such as proactively de-energizing selected circuits within the power grid. WRF forecasts are routinely produced on a 2 km grid, and the results are used as input to wildfire fuel moisture, fire probability, wildfire spread, and outage probability models. This forecast system produces skillful real-time forecasts while achieving an optimal blend of model resolution and ensemble size appropriate for today’s computational resources afforded to utilities. Numerous experiments were performed with different model settings, grid spacing, and ensemble configuration to develop an operational forecast system optimized for skill and cost. Dry biases were reduced by leveraging a new irrigation scheme, while wind skill was improved through a novel approach involving the selection of Global Ensemble Forecast System (GEFS) members used to drive WRF. We hope that findings in this study can help other utilities (especially those with similar weather impacts) improve their own forecast system. Full article

The frequency of extreme wildfire events (EWEs) is expected to increase due to climate change, leading to higher levels of atmospheric pollutants being released into the air, which could cause significant short-term impacts on human health (both for the population and firefighters) and on visibility. This study aims to gain a better understanding of the effects of EWEs’ smoke on air quality, its short-term impacts on human health, and how it reduces visibility by applying a modelling system to the Portuguese EWEs of October 2017. The Weather Research and Forecasting Model was combined with a semi-empirical fire spread algorithm (WRF-SFIRE) to simulate particulate matter smoke dispersion and assess its impacts based on up-to-date numerical approaches. Hourly simulated particulate matter values were compared to hourly monitored values, and the WRF-SFIRE system demonstrated accuracy consistent with previous studies, with a correlation coefficient ranging from 0.30 to 0.76 and an RMSE varying between 215 µg/m³ and 418 µg/m³. The estimated daily particle concentration levels exceeded the European air quality limit value, indicating a potential strong impact on human health. Health indicators related to exposure to particles were estimated, and their spatial distribution showed that the highest number of hospital admissions (>300) during the EWE, which occurred downwind of the fire perimeters, were due to the combined effect of high smoke pollution levels and population density. Visibility reached its worst level at night, when dispersion conditions were poorest, with the entire central and northern regions registering poor visibility levels (with a visual range of less than 2 km). This study emphasises the use of numerical models to predict, with high spatial and temporal resolutions, the population that may be exposed to dangerous levels of air pollution caused by ongoing wildfires. It offers valuable information to the public, civil protection agencies, and health organisations to assist in lessening the impact of wildfires on society. Full article