4.0 Article

A Systematic Review of Applications of Machine Learning Techniques for Wildfire Management Decision Support

Related references

Note: Only part of the references are listed.
Article Engineering, Multidisciplinary

Fire Detection Methods Based on Various Color Spaces and Gaussian Mixture Models

Amr Munshi

Summary: This study proposes an intelligent fire detection method by investigating three approaches based on the RGB, YCbCr, and HSV color models. The experimental results show that these methods are efficient in detecting both indoor and outdoor fires.

ADVANCES IN SCIENCE AND TECHNOLOGY-RESEARCH JOURNAL (2021)

Article Environmental Sciences

Forecasting fire risk with machine learning and dynamic information derived from satellite vegetation index time-series

Yaron Michael et al.

Summary: Mapping fire risk is crucial for fire management and firefighting efforts. This study investigates the impact of long-term vegetation metrics on fire risk mapping, finding that they can significantly improve model performance and provide important dynamic information.

SCIENCE OF THE TOTAL ENVIRONMENT (2021)

Article Engineering, Multidisciplinary

A digital twin framework for machine learning optimization of aerial fire fighting and pilot safety

T. I. Zohdi

Summary: The work aims to model and simulate aerial drops of fire retardants in dangerous fire environments, developing a computational framework that includes a meshless discrete element and a machine learning algorithm to optimize aircraft dynamics for improved fire-retardant release effectiveness and enable digital twin technologies for real-time flight-path planning for aerial first-responders.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2021)

Article Biodiversity Conservation

Application of remote sensing and machine learning algorithms for forest fire mapping in a Mediterranean area

Meriame Mohajane et al.

Summary: This study developed five new hybrid machine learning algorithms for mapping forest fire susceptibility in the north of Morocco, achieving higher accuracy and performance in predicting forest fires. The results can provide crucial support for the management of Mediterranean forest ecosystems and can be applied to other areas.

ECOLOGICAL INDICATORS (2021)

Article Ecology

A deep learning ensemble model for wildfire susceptibility mapping

Alexandra Bjanes et al.

Summary: This study introduces a novel Ensemble model based on deep learning networks, which achieved remarkable results in predicting forest fire susceptibility and environmental risks. Different sample patch sizes were found to affect the algorithm's predictive performance significantly.

ECOLOGICAL INFORMATICS (2021)

Article Forestry

Characterizing past fire occurrence in longleaf pine ecosystems with the Mid-Infrared Burn Index and a Random Forest classifier

Wade A. Wall et al.

Summary: The study developed a model using Landsat satellite imagery and a Random Forest (RF) machine learning algorithm to identify burned areas and estimate fire history for Fort Bragg, NC from 1991 to 2019. Results showed that the model, compared to a continental U.S. dataset, was able to capture the reduced fire frequency in wetlands, providing a novel approach for estimating local fire frequency for plant species in the area.

FOREST ECOLOGY AND MANAGEMENT (2021)

Article Forestry

Forest Fire Probability Mapping in Eastern Serbia: Logistic Regression versus Random Forest Method

Slobodan Milanovic et al.

Summary: Forest fire risk has increased globally in recent decades, with Mediterranean region being traditionally the most at risk in Europe. Effective fire protection systems are necessary to prevent damage to forests and infrastructure, with forest fire probability mapping being a key tool for resource allocation and firefighting efficiency. Machine learning methods such as Random Forest have shown better predictive ability in forest fire occurrence mapping compared to traditional Logistic Regression models.

FORESTS (2021)

Article Forestry

Forest Roads and Operational Wildfire Response Planning

Matthew P. Thompson et al.

Summary: Research into forest road planning, construction, and maintenance is growing as it is identified as a benefit for wildfire management. The focus is on how road networks support fire control operations and how this information is used in pre-season assessment and planning. Real-world examples show that roads are identified as primary control features, emphasizing the importance of enhanced decision support and empirical analysis in future research.

FORESTS (2021)

Article Environmental Sciences

Exploratory Analysis of Driving Force of Wildfires in Australia: An Application of Machine Learning within Google Earth Engine

Andrea Sulova et al.

Summary: Recent studies have shown an increase in wildfires globally due to climate change. A study developed an automatized workflow using remote sensing data for generating a training dataset of fire events at a continental level to predict fire occurrences in Australia during the 2019-2020 summer season. The findings, particularly using the Random Forest algorithm, were able to identify driving factors of Australian wildfires, which can be valuable for policymakers, environmentalists, and climate change researchers.

REMOTE SENSING (2021)

Article Environmental Sciences

Wildland Fire Tree Mortality Mapping from Hyperspatial Imagery Using Machine Learning

Dale A. Hamilton et al.

Summary: The use of imagery from small unmanned aircraft systems (sUAS) allows for the production of accurate data on the effects of wildland fires, enabling land managers to make informed decisions. By classifying trees and calculating canopy cover from hyperspatial imagery, it is possible to assess tree mortality and burn severity indicators. Comparison with pre-fire data helps in mapping burn severity and identifying areas of different fire intensities.

REMOTE SENSING (2021)

Article Environmental Sciences

Data-Driven Wildfire Risk Prediction in Northern California

Ashima Malik et al.

Summary: Throughout the years, wildfires have caused significant economic and environmental damage in California, leading to the development of machine learning models to predict wildfire risks. Two data-driven approaches based on random forest models were proposed to predict wildfire risk near Monticello and Winters. The combined model demonstrated higher accuracy compared to ensemble models, achieving a cutting-edge 92% accuracy in predicting wildfire risks.

ATMOSPHERE (2021)

Article Environmental Sciences

Modeling Low Intensity Fires: Lessons Learned from 2012 RxCADRE

Rodman R. Linn et al.

Summary: Coupled fire-atmosphere models are increasingly used to study low-intensity fires, such as prescribed fires. This study highlights the critical need to include variable wind fields as inputs in modeling marginal fire conditions. It also emphasizes the complexities in comparing physics-based model results against observations in marginal burning conditions.

ATMOSPHERE (2021)

Article Environmental Sciences

Exploring the multidimensional effects of human activity and land cover on fire occurrence for territorial planning

Jaime Carrasco et al.

Summary: The study discusses the multidimensional effects of various drivers, such as human activity, geographic and topographic factors, and land cover, in fire risk management. It highlights the importance of informed decision making through preventive policies to minimize risk and adapt to new environmental conditions for co-existence with wildfires.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2021)

Article Environmental Sciences

Hybrid inversion of radiative transfer models based on high spatial resolution satellite reflectance data improves fractional vegetation cover retrieval in heterogeneous ecological systems after fire

Jose Manuel Fernandez-Guisuraga et al.

Summary: This study evaluated the potential of estimating FVC from high spatial resolution satellite reflectance data compared to coarser imagery in landscapes affected by a megafire in the western Mediterranean Basin. The accuracy of FVC retrieval was found to be substantially higher from WorldView-3 than from Sentinel-2, with oak forests showing more accurate retrieval compared to heathlands and broomlands. The findings highlight the effectiveness of high spatial resolution satellite data in capturing FVC ground spatial variability using a hybrid RTM retrieval method in heterogeneous burned areas.

REMOTE SENSING OF ENVIRONMENT (2021)

Article Engineering, Environmental

Minimal effect of prescribed burning on fire spread rate and intensity in savanna ecosystems

Aristides Moustakas et al.

Summary: Recent wildfires have surpassed national firefighting capabilities, sparking debates on using prescribed burns to control their spread. Machine learning analysis showed that while fire return interval and seasonality play a role in moderating fire spread rate and intensity, meteorological variables still have the most significant impact.

STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT (2021)

Article Forestry

A Deep Learning Approach to Downscale Geostationary Satellite Imagery for Decision Support in High Impact Wildfires

Nicholas F. McCarthy et al.

Summary: The study proposes a statistical downscaling scheme based on deep learning to monitor active fire progression in real time. The system relies on multi-source Remote Sensing data and has been optimized for real time operational deployment, demonstrating high spatiotemporal resolution in monitoring fire spread.

FORESTS (2021)

Article Environmental Sciences

Quantifying Forest Fire and Post-Fire Vegetation Recovery in the Daxin'anling Area of Northeastern China Using Landsat Time-Series Data and Machine Learning

Jie Qiu et al.

Summary: This study employed the VCT algorithm to map forest disturbances in the Daxing'anling area of Northeastern China, and utilized SVM, SMLR, and RF to assess the statistical relationships between vegetation recovery characteristics and various factors. The results showed that forest disturbance events obtained from VCT had high spatial accuracy, and the RF algorithm provided better prediction accuracy for post-fire vegetation recovery.

REMOTE SENSING (2021)

Article Green & Sustainable Science & Technology

Towards Mountain Fire Safety Using Fire Spread Predictive Analytics and Mountain Fire Containment in IoT Environment

Imran et al.

Summary: This paper proposes a three-fold methodology for fire safety in mountain areas, including an optimization model, an ensemble prediction model, and an Internet of Things-based task orchestration approach. The proposed method aims to minimize damage caused by mountain fires and provide timely information to safety authorities for efficient mountain fire safety management. Evaluation results show that the proposed fire safety mechanism is effective in enhancing mountain fire safety.

SUSTAINABILITY (2021)

Article Environmental Sciences

Mapping Forest Burn Extent from Hyperspatial Imagery Using Machine Learning

Dale Hamilton et al.

Summary: In this study, a combination of mask region-based convolutional neural network (MR-CNN) and support vector machine (SVM) was used to identify trees and burned pixels in post-fire forests, resulting in higher accuracy in burn extent mapping.

REMOTE SENSING (2021)

Article Ecology

Assessing the probability of wildfire occurrences in a neotropical dry forest

Carlos Campos-Vargas et al.

Summary: This study used remote sensing techniques to assess the probability and danger of wildfires in the Guanacaste Conservation Area in northwestern Costa Rica, finding that topography, early successional vegetation, and human accessibility are key factors influencing fire occurrence, with spatial heterogeneity of fire risk observed in the landscape. The results could be valuable for the management of the conservation area.

ECOSCIENCE (2021)

Article Geography, Physical

A workflow based on Sentinel-1 SAR data and open-source algorithms for unsupervised burned area detection in Mediterranean ecosystems

Giandomenico De Luca et al.

Summary: This study investigates the capability of SAR Sentinel-1 data for burned area mapping in Mediterranean forest ecosystems using unsupervised machine learning. Various indices and filtering methods were employed to enhance image quality and contrast. The results demonstrate high accuracy and applicability of the method in different Mediterranean ecosystems.

GISCIENCE & REMOTE SENSING (2021)

Article Thermodynamics

Classification of flame extinction based on acoustic oscillations using artificial intelligence methods

Yavuz Selim Taspinar et al.

Summary: A sound wave flame extinction system was developed in the study with high success rates achieved through rule-based machine learning methods. The study accumulated a large dataset from experiments for early extinguishing of fires and performed classification analysis on the data.

CASE STUDIES IN THERMAL ENGINEERING (2021)

Article Chemistry, Analytical

Validation of ground truth fire debris classification by supervised machine learning

Michael E. Sigman et al.

Summary: The study used a set of 767 laboratory-generated fire debris samples to train machine learning classifiers, finding that linear discriminant analysis, support vector machines, and other classifiers were able to accurately classify the fire debris samples, achieving good performance in samples produced in large-scale test burns.

FORENSIC CHEMISTRY (2021)

Article Forestry

Maximum entropy-based forest fire likelihood mapping: analysing the trends, distribution, and drivers of forest fires in Sikkim Himalaya

Polash Banerjee

Summary: The recent forest fires in Brazil and Australia in 2019 serve as tragic reminders of the dangers of forest fires. Human encroachment into the wilderness and climate change are contributing to a global increase in forest fire incidents. A study in Sikkim reveals that climatic conditions and proximity to roads are major factors influencing forest fires. The development of a forest fire prediction map using GIS and machine learning can help stakeholders take informed mitigation measures.

SCANDINAVIAN JOURNAL OF FOREST RESEARCH (2021)

Article Environmental Sciences

Exploring PRISMA Scene for Fire Detection: Case Study of 2019 Bushfires in Ben Halls Gap National Park, NSW, Australia

Stefania Amici et al.

Summary: PRISMA, a hyperspectral mission by the Italian Space Agency, was launched in 2019 to measure unique spectral features of various materials. This study investigated the potential use of PRISMA sensor for active wildfire characterization by testing three detection techniques during the Australian bushfires of 2019. The results showed promising performance of the instrument in radiance, with adaptations of CO2-CIBR and HFDI producing detection rates ranging from 0.13561 to 0.81598 and 0.36171 to 0.88431 respectively.

REMOTE SENSING (2021)

Article Environmental Sciences

Uni-Temporal Multispectral Imagery for Burned Area Mapping with Deep Learning

Xikun Hu et al.

Summary: The study compares deep learning (DL) models and machine learning (ML) algorithms for mapping burned areas from satellite imagery in three wildfire sites. DL algorithms outperform ML methods in compact burned areas, while ML methods are more suitable for dispersed burn in boreal forests.

REMOTE SENSING (2021)

Article Environmental Sciences

Landslide and Wildfire Susceptibility Assessment in Southeast Asia Using Ensemble Machine Learning Methods

Qian He et al.

Summary: This study focused on landslide and wildfire susceptibility in Southeast Asia using ensemble machine learning algorithms. The results showed that Random Forest performed superiorly in modeling susceptibility, indicating its efficiency and accuracy in hazard assessment in the region.

REMOTE SENSING (2021)

Article Environmental Sciences

Machine Learning Techniques for Fine Dead Fuel Load Estimation Using Multi-Source Remote Sensing Data

Marina D'Este et al.

Summary: A model using remote sensing data and machine learning techniques to estimate fine dead fuel load was developed, with the Random Forest model showing higher predictive power compared to Support Vector Machine and Multiple Linear Regression models. LIDAR variables were found to be more important in fuel estimation than optical and radar variables, and there was a positive relationship between 1-h fuel load and the presence of the tree component.

REMOTE SENSING (2021)

Article Environmental Sciences

Assessing Wildfire Burn Severity and Its Relationship with Environmental Factors: A Case Study in Interior Alaska Boreal Forest

Christopher W. Smith et al.

Summary: Recent advancements in remote sensing methods and satellite imagery have the potential to greatly enhance burn severity assessments in the Alaskan boreal forest. Machine learning classifiers show promise for reliable mapping of burn severity when sufficient ground truth data is available.

REMOTE SENSING (2021)

Article Environmental Sciences

Identifying Key Drivers of Wildfires in the Contiguous US Using Machine Learning and Game Theory Interpretation

Sally S-C Wang et al.

Summary: This study utilizes a machine learning model incorporating various predictors to predict wildfire occurrences and identify key contributors to large burned areas in different regions of the United States.

EARTHS FUTURE (2021)

Article Agronomy

Land cover composition, climate, and topography drive land surface phenology in a recently burned landscape: An application of machine learning in phenological modeling

Jianmin Wang et al.

Summary: This study quantitatively assessed the contributions of land cover composition, climate, and topography on the spatial and interannual variation in LSP using a machine learning approach. The results highlighted the essential role of land cover composition in understanding the spatial and interannual variations of LSP.

AGRICULTURAL AND FOREST METEOROLOGY (2021)

Article Ecology

Machine-learning modelling of fire susceptibility in a forest-agriculture mosaic landscape of southern India

A. L. Achu et al.

Summary: This study investigates the forest fire susceptibility in the forested landscape of the Wayanad district, India, using various geospatial data, machine learning techniques, and spatial statistical tools. A weighted approach combining different MLTs shows improved accuracy in mapping forest fire susceptibility compared to individual MLTs. The results indicate that roughly one-third of the study area is highly susceptible to forest fires, with anthropogenic factors playing a role.

ECOLOGICAL INFORMATICS (2021)

Article Ecology

Mapping forest fire susceptibility using spatially explicit ensemble models based on the locally weighted learning algorithm

Tran Thi Tuyen et al.

Summary: Ensemble models combining LWL algorithm with CG, Bagging, Decorate, and Dagging techniques were used to predict forest fire susceptibility in Pu Mat National Park, Vietnam, with CG-LWL and Bagging-LWL models showing the highest training performance. These models enhance researchers' understanding of model building processes and can be applied to predict other natural hazards by considering local geo-environmental factors.

ECOLOGICAL INFORMATICS (2021)

Article Automation & Control Systems

Pixel-level automatic annotation for forest fire image

Xubing Yang et al.

Summary: The proposed automatic annotation method for forest fire images introduces supervised information through interactive convex hulls at the pixel level, allowing visually selecting irregular fire and no-fire regions. It expands the view of fire detection to multi-class classification and loosens the constraints of i.i.d hypothesis in machine learning methods. Experimental evaluations show that it achieves higher fire detection rate and lower false alarm rate compared to state-of-the-art methods.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2021)

Article Computer Science, Interdisciplinary Applications

Deep fire topology: Understanding the role of landscape spatial patterns in wildfire occurrence using artificial intelligence

Cristobal Pais et al.

Summary: This study focuses on wildfire activity globally and proposes a deep learning framework for estimating and predicting wildfire ignition risk. The research reveals the potential of landscape topology in wildfire prediction and provides important insights for landscape management planning.

ENVIRONMENTAL MODELLING & SOFTWARE (2021)

Article Biodiversity Conservation

Restoration of forest resilience to fire from old trees is possible across a large Colorado dry-forest landscape by 2060, but only under the Paris 1.5°C goal

William L. Baker

Summary: Research has shown that passive restoration of the resistance and resilience of old trees to fire is feasible, particularly by protecting old trees with an age of >= 120 years, which could restore historical levels of old-growth by ca 2060. It is contingent on global success in achieving the Paris 1.5 degrees C goal, but passive restoration may be viable elsewhere.

GLOBAL CHANGE BIOLOGY (2021)

Article Plant Sciences

Long-term empirical evidence shows post-disturbance climate controls post-fire regeneration

Jaclyn Guz et al.

Summary: This study examines the regeneration of lodgepole pine forests in high-elevation areas following high-severity fires, revealing the significant impact of pre-fire stand structure and post-fire climate conditions on regeneration. Factors such as stocking density and minimum temperature play crucial roles in the regeneration process and radial growth, while increasing patchiness in regeneration may have cascading effects on future ecological dynamics.

JOURNAL OF ECOLOGY (2021)

Article Chemistry, Analytical

Machine Learning Methods and Synthetic Data Generation to Predict Large Wildfires

Fernando-Juan Perez-Porras et al.

Summary: Wildfires are increasing in frequency globally, making prediction complex. Machine learning methods for predicting large wildfires face challenges of overfitting or underfitting due to imbalanced datasets. Using synthetic data can improve prediction accuracy, offering a new approach for Decision Support Systems in wildfire management.

SENSORS (2021)

Article Forestry

21st Century Planning Techniques for Creating Fire-Resilient Forests in the American West

John Hogland et al.

Summary: Data-driven decision making is crucial for effective wildfire protection and sustainable use of natural resources. New analytical frameworks incorporating innovative concepts, field data, and expert knowledge are needed to build information and aid decision making.

FORESTS (2021)

Article Environmental Sciences

Machine Learning Estimation of Fire Arrival Time from Level-2 Active Fires Satellite Data

Angel Farguell et al.

Summary: The article introduces a machine learning-based method to estimate fire arrival time from satellite data for improved fire modeling accuracy. Through testing, the method demonstrates good agreement with airborne fire observations, enabling monitoring of wildfires in near-real-time and providing accurate fire arrival times.

REMOTE SENSING (2021)

Article Environmental Sciences

Forest Fire Risk Prediction: A Spatial Deep Neural Network-Based Framework

Mohsen Naderpour et al.

Summary: Recognition of forest fire occurrence patterns and identification of fire risk are crucial in mitigating potential threats. Machine learning techniques like deep neural networks have shown high precision in assessing forest fire susceptibility, improving adaptability and decision-making in various regions of Australia with minor localization requirements.

REMOTE SENSING (2021)

Article Computer Science, Information Systems

Wildland Fire Susceptibility Mapping Using Support Vector Regression and Adaptive Neuro-Fuzzy Inference System-Based Whale Optimization Algorithm and Simulated Annealing

A'kif Al-Fugara et al.

Summary: This study developed and compared four hybrid models to map wildland fires in Jerash Province, Jordan, using two machine learning models (SVR and ANFIS) and two meta-heuristic models (WOA and SA). The SVR-based hybrid models showed higher accuracy compared to the ANFIS-based models, indicating their potential in guiding firefighting tactics.

ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION (2021)

Article Computer Science, Information Systems

A Fully Automatic, Interpretable and Adaptive Machine Learning Approach to Map Burned Area from Remote Sensing

Daniela Stroppiana et al.

Summary: The paper presents a fully automatic algorithm approach that combines knowledge-driven and data-driven elements to map burned areas from remote sensing. It utilizes active fire points to train a fusion function and applies a Region Growing algorithm to generate the final burned area map. The fusion function, an Ordered Weighted Averaging operator, is learned through machine learning and characterized by measures of pessimism/optimism and democracy/monarchy.

ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION (2021)

Article Geosciences, Multidisciplinary

Modeling Conditions Appropriate for Wildfire in South East China - A Machine Learning Approach

Zeeshan Shirazi et al.

Summary: The study aimed to develop a machine learning workflow to monitor fire risks in Southeast China by learning from a grid file database of important environmental parameters. Different models showed high sensitivity on the test dataset, but low precision and accuracy. The model for the mixed forest land cover performed the best among all models tested.

FRONTIERS IN EARTH SCIENCE (2021)

Article Computer Science, Information Systems

Multistage Real-Time Fire Detection Using Convolutional Neural Networks and Long Short-Term Memory Networks

Manh Dung Nguyen et al.

Summary: The paper proposes a multistage fire detection method using convolutional neural networks (CNNs) and long short-term memory (LSTM) networks, combining early detection of fire candidates and 2D features extraction to determine if it is a true fire, showing competitive performance in real-world applications.

IEEE ACCESS (2021)

Article Forestry

Comparing calibrated statistical and machine learning methods for wildland fire occurrence prediction: a case study of human-caused fires in Lac La Biche, Alberta, Canadac

Nathan Phelps et al.

Summary: FOP modelling is crucial for fire management decisions and can be based on statistical modelling or machine learning, but proper calibration is needed for accuracy. Logistic generalised additive models (GAMs) were found to perform similarly to machine learning models in FOP, highlighting the importance of discussing the pros and cons of different modelling approaches with fire management practitioners for operational use.

INTERNATIONAL JOURNAL OF WILDLAND FIRE (2021)

Article Computer Science, Information Systems

Determining the Extinguishing Status of Fuel Flames With Sound Wave by Machine Learning Methods

Murat Koklu et al.

Summary: The study utilized machine learning methods to perform classification analysis on a sound wave fire-extinguishing system, with the stacking model demonstrating the highest classification accuracy at 97.06%. The other methods showed classification accuracies of 96.58% (random forest), 96.03% (artificial neural network), 94.88% (deep neural network), and 92.62% (k-NN).

IEEE ACCESS (2021)

Article Geosciences, Multidisciplinary

Spatial Assessment of Wildfires Susceptibility in Santa Cruz (Bolivia) Using Random Forest

Marcela Bustillo Sanchez et al.

Summary: The study investigated wildfires in Bolivia, specifically in the department of Santa Cruz, utilizing datasets and digital spatial data to develop a model based on Random Forest for predicting wildfires susceptibility. Results demonstrated that flooded savanna and shrub or herbaceous cover are the top factors in predicting wildfires.

GEOSCIENCES (2021)

Article Chemistry, Analytical

Classification of ground-truth fire debris samples using artificial neural networks

Nicholas A. Thurn et al.

Summary: This study assesses the classification performance of neural networks on fire debris samples, selecting optimal models and decision thresholds, demonstrating their potential in fire debris evidence evaluation.

FORENSIC CHEMISTRY (2021)

Article Meteorology & Atmospheric Sciences

Using machine learning to predict fire-ignition occurrences from lightning forecasts

Ruth Coughlan et al.

Summary: Lightning-caused wildfires are a significant contributor to burned areas, with fuel moisture and ignition probability having a clear connection but poorly understood mechanisms. A machine learning approach was used to develop a predictive model for wildfire ignition, showing promising results with ensemble methods exhibiting an accuracy of 78%. The super-learner developed from this study is planned to enhance information connected to fire management in an operational context.

METEOROLOGICAL APPLICATIONS (2021)

Article Environmental Sciences

Temperature-based fire frequency analysis using machine learning: A case of Changsha, China

Zhisheng Xu et al.

Summary: This study analyzed fire data and weather data in Changsha from 2011 to 2017, using machine learning methods to predict fire frequency under climate warming scenarios. The random forest regression model was found to be the most suitable for this prediction. The study also projected an increase in fire frequency in Changsha in the next 50 years.

CLIMATE RISK MANAGEMENT (2021)

Article Computer Science, Information Systems

Learning Representation of Secondary Effects for Fire-Flake Animation

Myungjin Choi et al.

Summary: This study introduces a data-driven neural network-based fire-flake simulation model that trains a neural network using precomputed fire simulation data to generate natural fire flakes. The model enhances performance while maintaining visual quality, being faster and easier to implement compared to traditional methods, saving time for artists and developers.

IEEE ACCESS (2021)

Review Engineering, Multidisciplinary

A Survey of Machine Learning Algorithms Based Forest Fires Prediction and Detection Systems

Faroudja Abid

Summary: Forest fires are a major environmental concern globally, causing economic, ecological damage, and loss of human lives. The trend now is towards integrating artificial intelligence into fire prediction and detection systems. Research focuses on machine learning algorithms and factors influencing fire occurrence and risk.

FIRE TECHNOLOGY (2021)

Article Forestry

Managing Fragmented Fire-Threatened Landscapes with Spatial Externalities

Christopher J. Lauer et al.

FOREST SCIENCE (2020)

Article Computer Science, Information Systems

Machine learning based approach for multimedia surveillance during fire emergencies

Faisal Saeed et al.

MULTIMEDIA TOOLS AND APPLICATIONS (2020)

Article Computer Science, Information Systems

Convolutional neural network based early fire detection

Faisal Saeed et al.

MULTIMEDIA TOOLS AND APPLICATIONS (2020)

Article Environmental Sciences

Wildfire management in Mediterranean-type regions: paradigm change needed

Francisco Moreira et al.

ENVIRONMENTAL RESEARCH LETTERS (2020)

Article Environmental Sciences

Sentinel-1 observation frequency significantly increases burnt area detectability in tropical SE Asia

Joao M. B. Carreiras et al.

ENVIRONMENTAL RESEARCH LETTERS (2020)

Article Computer Science, Artificial Intelligence

Real-time detection of wildfire risk caused by powerline vegetation faults using advanced machine learning techniques

Jun Ma et al.

ADVANCED ENGINEERING INFORMATICS (2020)

Article Computer Science, Artificial Intelligence

A wildfire warning system applied to the state of Acre in the Brazilian Amazon

I. D. B. Silva et al.

APPLIED SOFT COMPUTING (2020)

Article Thermodynamics

Learning-based prediction of wildfire spread with real-time rate of spread measurement

Chunjie Zhai et al.

COMBUSTION AND FLAME (2020)

Article Engineering, Multidisciplinary

A machine-learning framework for rapid adaptive digital-twin based fire-propagation simulation in complex environments

T. I. Zohdi

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2020)

Article Environmental Sciences

A remote sensing approach to mapping fire severity in south-eastern Australia using sentinel 2 and random forest

Rebecca Gibson et al.

REMOTE SENSING OF ENVIRONMENT (2020)

Article Public, Environmental & Occupational Health

Big Data and Predictive Analytics in Fire Risk Using Weather Data

Puneet Agarwal et al.

RISK ANALYSIS (2020)

Article Multidisciplinary Sciences

Comparisons of Diverse Machine Learning Approaches for Wildfire Susceptibility Mapping

Khalil Gholamnia et al.

SYMMETRY-BASEL (2020)

Article Environmental Sciences

Evidence-based mapping of the wildland-urban interface to better identify human communities threatened by wildfires

Alejandro Miranda et al.

ENVIRONMENTAL RESEARCH LETTERS (2020)

Article Environmental Sciences

Observed changes in fire patterns and possible drivers over Central Africa

Yan Jiang et al.

ENVIRONMENTAL RESEARCH LETTERS (2020)

Article Environmental Sciences

Intensified burn severity in California's northern coastal mountains by drier climatic condition

Yuhan Huang et al.

ENVIRONMENTAL RESEARCH LETTERS (2020)

Article Environmental Sciences

Topological data analysis reveals parameters with prognostic skill for extreme wildfire size

Rebecca Bendick et al.

ENVIRONMENTAL RESEARCH LETTERS (2020)

Review Environmental Sciences

A review of machine learning applications in wildfire science and management

Piyush Jain et al.

ENVIRONMENTAL REVIEWS (2020)

Article Engineering, Civil

Using machine learning in physics-based simulation of fire

B. Y. Lattimer et al.

FIRE SAFETY JOURNAL (2020)

Article Multidisciplinary Sciences

Performance Evaluation of Machine Learning Methods for Forest Fire Modeling and Prediction

Binh Thai Pham et al.

SYMMETRY-BASEL (2020)

Article Chemistry, Multidisciplinary

Fire Risk Assessment Models Using Statistical Machine Learning and Optimized Risk Indexing

Myoung-Young Choi et al.

APPLIED SCIENCES-BASEL (2020)

Article Environmental Sciences

Assessing the biophysical and social drivers of burned area distribution at the local scale

Sandra Oliveira et al.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2020)

Article Agronomy

Rapid recovery of net ecosystem production in a semi-arid woodland after a wildfire

Qiaoqi Sun et al.

AGRICULTURAL AND FOREST METEOROLOGY (2020)

Article Computer Science, Information Systems

A smart approach for fire prediction under uncertain conditions using machine learning

Richa Sharma et al.

MULTIMEDIA TOOLS AND APPLICATIONS (2020)

Article Environmental Sciences

Training data requirements for fire severity mapping using Landsat imagery and random forest

Luke Collins et al.

REMOTE SENSING OF ENVIRONMENT (2020)

Article Multidisciplinary Sciences

Quantifying the drivers and predictability of seasonal changes in African fire

Yan Yu et al.

NATURE COMMUNICATIONS (2020)

Article Multidisciplinary Sciences

A machine learning framework for multi-hazards modeling and mapping in a mountainous area

Saleh Yousefi et al.

SCIENTIFIC REPORTS (2020)

Article Environmental Sciences

Evaluation of fire severity in fire prone-ecosystems of Spain under two different environmental conditions

Paula Garcia-Llamas et al.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2020)

Article Computer Science, Information Systems

Assessment of Interventions in Fuel Management Zones Using Remote Sensing

Ricardo Afonso et al.

ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION (2020)

Article Chemistry, Analytical

Adaptive Quasi-Unsupervised Detection of Smoke Plume by LiDAR

Riccardo Rossi et al.

SENSORS (2020)

Article Chemistry, Multidisciplinary

Developing a New Hourly Forest Fire Risk Index Based on Catboost in South Korea

Yoojin Kang et al.

APPLIED SCIENCES-BASEL (2020)

Article Computer Science, Artificial Intelligence

Enhancing wildfire spread modelling by building a gridded fuel moisture content product with machine learning

Tyler C. McCandless et al.

MACHINE LEARNING-SCIENCE AND TECHNOLOGY (2020)

Article Environmental Sciences

Testing the utility of the blue spectral region in burned area mapping: Insights from savanna wildfires

Kudzai Shaun Mpakairi et al.

REMOTE SENSING APPLICATIONS-SOCIETY AND ENVIRONMENT (2020)

Article Engineering, Multidisciplinary

Estimation of Burned Areas In Forest Fires Using Artificial Neural Networks

M. Hanefi Calp et al.

INGENIERIA SOLIDARIA (2020)

Article Environmental Studies

Estimating the probability of wildfire occurrence in Mediterranean landscapes using Artificial Neural Networks

Mario Elia et al.

ENVIRONMENTAL IMPACT ASSESSMENT REVIEW (2020)

Article Environmental Sciences

Unitemporal approach to fire severity mapping using multispectral synthetic databases and Random Forests

Raquel Montorio et al.

REMOTE SENSING OF ENVIRONMENT (2020)

Article Computer Science, Information Systems

A Deep Learning Framework for Detection of Targets in Thermal Images to Improve Firefighting

Manish Bhattarai et al.

IEEE ACCESS (2020)

Article Computer Science, Information Systems

Deep Learning-Based Automated Forest Health Diagnosis From Aerial Images

Chia-Yen Chiang et al.

IEEE ACCESS (2020)

Article Computer Science, Information Systems

Optimal Placement and Intelligent Smoke Detection Algorithm for Wildfire-Monitoring Cameras

Jie Shi et al.

IEEE ACCESS (2020)

Article Computer Science, Artificial Intelligence

Forest fire forecasting using ensemble learning approaches

Ying Xie et al.

NEURAL COMPUTING & APPLICATIONS (2019)

Article Engineering, Civil

Predictive modeling of wildfires: A new dataset and machine learning approach

Younes Oulad Sayad et al.

FIRE SAFETY JOURNAL (2019)

Article Engineering, Multidisciplinary

Wildland Fire Spread Modeling Using Convolutional Neural Networks

Jonathan L. Hodges et al.

FIRE TECHNOLOGY (2019)

Article Environmental Sciences

Modeling initial attack success of wildfire suppression in Catalonia, Spain

Marcos Rodrigues et al.

SCIENCE OF THE TOTAL ENVIRONMENT (2019)

Article Environmental Sciences

Spatial Distribution of Forest Fire Emissions: A Case Study in Three Mexican Ecoregions

Maria Isabel Cruz-Lopez et al.

REMOTE SENSING (2019)

Article Environmental Sciences

Machine learning models accurately predict ozone exposure during wildfire events

Gregory L. Watson et al.

ENVIRONMENTAL POLLUTION (2019)

Article Environmental Sciences

Machine Learning-Based Integration of High-Resolution Wildfire Smoke Simulations and Observations for Regional Health Impact Assessment

Yufei Zou et al.

INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH (2019)

Article Geosciences, Multidisciplinary

Forest Fire Susceptibility Modeling Using a Convolutional Neural Network for Yunnan Province of China

Guoli Zhang et al.

INTERNATIONAL JOURNAL OF DISASTER RISK SCIENCE (2019)

Article Engineering, Industrial

Forest fire induced Natech risk assessment: A survey of geospatial technologies

Mohsen Naderpour et al.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2019)

Article Environmental Sciences

Burned area detection and mapping using Sentinel-1 backscatter coefficient and thermal anomalies

Miguel A. Belenguer-Plomer et al.

REMOTE SENSING OF ENVIRONMENT (2019)

Article Geosciences, Multidisciplinary

Assessment of post-wildfire debris flow occurrence using classifier tree

Priscilla Addison et al.

GEOMATICS NATURAL HAZARDS & RISK (2019)

Article Engineering, Geological

Object based burned area mapping with random forest algorithm

Resul Comert et al.

INTERNATIONAL JOURNAL OF ENGINEERING AND GEOSCIENCES (2019)

Article Computer Science, Artificial Intelligence

Feature selection in machine learning: A new perspective

Jie Cai et al.

NEUROCOMPUTING (2018)

Article Green & Sustainable Science & Technology

Modeling Post-Fire Mortality in Pure and Mixed Forest Stands in Portugal-A Forest Planning-Oriented Model

Brigite Botequim et al.

SUSTAINABILITY (2017)

Article Green & Sustainable Science & Technology

Addressing Wildfire Risk in Forest Management Planning with Multiple Criteria Decision Making Methods

Susete Marques et al.

SUSTAINABILITY (2017)

Article Forestry

Characterization of wildfires in Portugal

S. Marques et al.

EUROPEAN JOURNAL OF FOREST RESEARCH (2011)

Article Forestry

Integrated spatial fire and forest management planning

Mauricio A. Acuna et al.

CANADIAN JOURNAL OF FOREST RESEARCH (2010)

Article Ecology

A working guide to boosted regression trees

J. Elith et al.

JOURNAL OF ANIMAL ECOLOGY (2008)

Article Biotechnology & Applied Microbiology

What is a hidden Markov model?

SR Eddy

NATURE BIOTECHNOLOGY (2004)

Article Environmental Sciences

Analysis of the prescribed burning practice in the pine forest of northwestern Portugal

P Fernandes et al.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2004)

Article Computer Science, Artificial Intelligence

Random forests

L Breiman

MACHINE LEARNING (2001)