Characterizing Topographic Influences of Bushfire Severity Using Machine Learning Models: A Case Study in a Hilly Terrain of Victoria, Australia

Topography plays a significant role in determining bushfire severity over a hilly landscape. However, complex interrelationships between topographic variables and bushfire severity are difficult to quantify using traditional statistical methods. More recently, different machine learning (ML) models are becoming popular in characterizing complex relationships between different environmental variables. Yet, few studies have specifically evaluated the suitability of ML models in predictive bushfire severity analysis. Hence, the aim of this research is twofold. First, to determine suitable ML models by assessing their performances in bushfire severity predictions using remote sensing data analytics, and second to identify and investigate topographic variables influencing bushfire severity. The results showed that random forest (RF) and gradient boosting (GB) models had their distinct advantages in predictive modeling of bushfire severity. RF model showed higher precision (86% to 100%) than GB (59% to 72%) while predicting low, moderate, and high severity classes. Whereas GB model demonstrated better recall, i.e., completeness of positive predictions (56% to 75%) than RF (49% to 61%) for those classes. Closer investigations on topographic characteristics showed a varying relationship of severity patterns across different morphological landform classes. Landforms having lower slope curvatures or with unchanging slopes were more prone to severe burning than those landforms with higher slope curvatures. Our results provide insights into how topography influences potential bushfire severity risks and recommends purpose-specific choice of ML models.

[1]  M. Cruz,et al.  Prediction of wildfire rate of spread in grasslands using machine learning methods , 2022, Environ. Model. Softw..

[2]  Weiqi Wang,et al.  Burn Severity in Canada's Mountain National Parks: Patterns, Drivers, and Predictions , 2022, Geophysical Research Letters.

[3]  Muzaffer Can Iban,et al.  Machine learning based wildfire susceptibility mapping using remotely sensed fire data and GIS: A case study of Adana and Mersin provinces, Turkey , 2022, Ecol. Informatics.

[4]  A. Rajabifard,et al.  Remote Sensing and Meteorological Data Fusion in Predicting Bushfire Severity: A Case Study from Victoria, Australia , 2022, Remote. Sens..

[5]  Anping Chen,et al.  Machine learning–based observation-constrained projections reveal elevated global socioeconomic risks from wildfire , 2022, Nature Communications.

[6]  Yi-Fan Chen,et al.  Next Day Wildfire Spread: A Machine Learning Dataset to Predict Wildfire Spreading From Remote-Sensing Data , 2021, IEEE Transactions on Geoscience and Remote Sensing.

[7]  Ross A. Bradstock,et al.  Derivation of a Bayesian fire spread model using large-scale wildfire observations , 2021, Environ. Model. Softw..

[8]  Douglas G. Woolford,et al.  Comparing calibrated statistical and machine learning methods for wildland fire occurrence prediction: a case study of human-caused fires in Lac La Biche, Alberta, Canada , 2021, International Journal of Wildland Fire.

[9]  A. Taylor,et al.  Drivers of fire severity shift as landscapes transition to an active fire regime, Klamath Mountains, USA , 2021, Ecosphere.

[10]  Sayanti Mukherjee,et al.  Impact of geophysical and anthropogenic factors on wildfire size: a spatiotemporal data-driven risk assessment approach using statistical learning , 2021, Stochastic Environmental Research and Risk Assessment.

[11]  R. Keenan,et al.  The severity and extent of the Australia 2019–20 Eucalyptus forest fires are not the legacy of forest management , 2021, Nature Ecology & Evolution.

[12]  Kai Liu,et al.  Landslide and Wildfire Susceptibility Assessment in Southeast Asia Using Ensemble Machine Learning Methods , 2021, Remote. Sens..

[13]  L. Brotóns,et al.  Towards a comprehensive look at global drivers of novel extreme wildfire events , 2021, Climatic Change.

[14]  Fabian Ewald Fassnacht,et al.  Explaining Sentinel 2-based dNBR and RdNBR variability with reference data from the bird's eye (UAS) perspective , 2021, Int. J. Appl. Earth Obs. Geoinformation.

[15]  Jamal Jokar Arsanjani,et al.  Exploratory Analysis of Driving Force of Wildfires in Australia: An Application of Machine Learning within Google Earth Engine , 2020, Remote. Sens..

[16]  Jaehyun Yoon.  Forecasting of Real GDP Growth Using Machine Learning Models: Gradient Boosting and Random Forest Approach , 2020, Computational Economics.

[17]  S. Suárez‐Seoane,et al.  Evaluation of fire severity in fire prone-ecosystems of Spain under two different environmental conditions. , 2020, Journal of environmental management.

[18]  Jagannath Aryal,et al.  Mapping Urban Tree Cover Changes Using Object-Based Convolution Neural Network (OB-CNN) , 2020, Remote. Sens..

[19]  G. Newell,et al.  Training data requirements for fire severity mapping using Landsat imagery and random forest , 2020 .

[20]  M. Schwartz,et al.  Intensified burn severity in California’s northern coastal mountains by drier climatic condition , 2020, Environmental Research Letters.

[21]  D. Roberts,et al.  Enhanced burn severity estimation using fine resolution ET and MESMA fraction images with machine learning algorithm , 2020 .

[22]  T. Danaher,et al.  A remote sensing approach to mapping fire severity in south-eastern Australia using sentinel 2 and random forest , 2020 .

[23]  Mark Crowley,et al.  A review of machine learning applications in wildfire science and management , 2020, Environmental Reviews.

[24]  D. Chicco,et al.  The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation , 2020, BMC Genomics.

[25]  B. Collins,et al.  Multi-scaled drivers of severity patterns vary across land ownerships for the 2013 Rim Fire, California , 2019, Landscape Ecology.

[26]  Gabriele Bitelli,et al.  TWI computation: a comparison of different open source GISs , 2019, Open Geospatial Data, Software and Standards.

[27]  Jong Hae Kim.  Multicollinearity and misleading statistical results , 2019, Korean journal of anesthesiology.

[28]  Ionut Cristi Nicu,et al.  GIS-based Landform Classification of Eneolithic Archaeological Sites in the Plateau-plain Transition Zone (NE Romania): Habitation Practices vs. Flood Hazard Perception , 2019, Remote. Sens..

[29]  Grant J. Williamson,et al.  Geographic Patterns of Fire Severity Following an Extreme Eucalyptus Forest Fire in Southern Australia: 2013 Forcett-Dunalley Fire , 2018, Fire.

[30]  Graeme Newell,et al.  The utility of Random Forests for wildfire severity mapping , 2018, Remote Sensing of Environment.

[31]  Hamid Reza Pourghasemi,et al.  Spatial Modelling of Gully Erosion Using GIS and R Programing: A Comparison among Three Data Mining Algorithms , 2018, Applied Sciences.

[32]  Lee Sangwoo,et al.  Structural equation model for burn severity with topographic variables and susceptible forest cover. , 2018 .

[33]  Nathaniel P. Robinson,et al.  Mean Composite Fire Severity Metrics Computed with Google Earth Engine Offer Improved Accuracy and Expanded Mapping Potential , 2018, Remote. Sens..

[34]  H. Zald,et al.  Severe fire weather and intensive forest management increase fire severity in a multi-ownership landscape. , 2018, Ecological applications : a publication of the Ecological Society of America.

[35]  S. Dobrowski,et al.  What Drives Low-Severity Fire in the Southwestern USA? , 2018 .

[36]  Tarunpreet Bhatia,et al.  GIS-based evolutionary optimized Gradient Boosted Decision Trees for forest fire susceptibility mapping , 2018, Natural Hazards.

[37]  Y. Choi,et al.  Complex Relationships of the Effects of Topographic Characteristics and Susceptible Tree Cover on Burn Severity , 2018 .

[38]  A. Taylor,et al.  Previous burns and topography limit and reinforce fire severity in a large wildfire , 2017 .

[39]  A. Dowdy.  Climatological Variability of Fire Weather in Australia , 2017 .

[40]  M. Clarke,et al.  Interacting effects of fire severity, time since fire and topography on vegetation structure after wildfire , 2017 .

[41]  Anthony L. Westerling,et al.  Climate drives inter-annual variability in probability of high severity fire occurrence in the western United States , 2017 .

[42]  Carl N. Skinner,et al.  Factors influencing fire severity under moderate burning conditions in the Klamath Mountains, northern California, USA , 2017 .

[43]  Richard M. Teeuw,et al.  GIS-based landform classification of Bronze Age archaeological sites on Crete Island , 2017, PloS one.

[44]  Carol Miller,et al.  Weather, fuels, and topography impede wildland fire spread in western US landscapes , 2016 .

[45]  M. Storey,et al.  The role of weather, past fire and topography in crown fire occurrence in eastern Australia , 2016 .

[46]  Mariana Belgiu,et al.  Random forest in remote sensing: A review of applications and future directions , 2016 .

[47]  Gary J. Sheridan,et al.  Quantifying the effects of topographic aspect on water content and temperature in fine surface fuel , 2015 .

[48]  Robert J. McGaughey,et al.  Mixed severity fire effects within the Rim fire: Relative importance of local climate, fire weather, topography, and forest structure , 2015 .

[49]  Michael Olusegun Akinwande,et al.  Variance Inflation Factor: As a Condition for the Inclusion of Suppressor Variable(s) in Regression Analysis , 2015 .

[50]  Jiaxing Zu,et al.  Quantifying influences and relative importance of fire weather, topography, and vegetation on fire size and fire severity in a Chinese boreal forest landscape , 2015 .

[51]  T. Esque,et al.  Topography and climate are more important drivers of long‐term, post‐fire vegetation assembly than time‐since‐fire in the Sonoran Desert, US , 2015 .

[52]  Michelle Coppoletta,et al.  Post-fire vegetation and fuel development influences fire severity patterns in reburns. , 2015, Ecological applications : a publication of the Ecological Society of America.

[53]  Chengquan Huang,et al.  Use of Vegetation Change Tracker and Support Vector Machine to Map Disturbance Types in Greater Yellowstone Ecosystems in a 1984–2010 Landsat Time Series , 2015, IEEE Geoscience and Remote Sensing Letters.

[54]  Jagannath Aryal,et al.  Near-Real Time Detection of Beetle Infestation in Pine Forests Using MODIS Data , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[55]  R. Bradstock,et al.  Vegetation, terrain and fire history shape the impact of extreme weather on fire severity and ecosystem response , 2014 .

[56]  Jagannath Aryal,et al.  A Statistical Framework for Near-Real Time Detection of Beetle Infestation in Pine Forests Using MODIS Data , 2014, IEEE Geoscience and Remote Sensing Letters.

[57]  Carol Miller,et al.  A New Metric for Quantifying Burn Severity: The Relativized Burn Ratio , 2014, Remote. Sens..

[58]  R. Bradstock,et al.  Can precipitation influence landscape controls on wildfire severity? A case study within temperate eucalypt forests of south-eastern Australia , 2014 .

[59]  Jaan Praks,et al.  Flood Mapping With TerraSAR-X in Forested Regions in Estonia , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[60]  Gilles Louppe,et al.  Understanding variable importances in forests of randomized trees , 2013, NIPS.

[61]  Philippe De Maeyer,et al.  Application of the topographic position index to heterogeneous landscapes , 2013 .

[62]  Anne-Laure Boulesteix,et al.  Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics , 2012, WIREs Data Mining Knowl. Discov..

[63]  Ross A. Bradstock,et al.  A comparison of bushfire fuel hazard assessors and assessment methods in dry sclerophyll forest near Sydney, Australia , 2012 .

[64]  Qian Du,et al.  Multi-Modal Change Detection, Application to the Detection of Flooded Areas: Outcome of the 2009–2010 Data Fusion Contest , 2012, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[65]  Charles H. Luce,et al.  Both topography and climate affected forest and woodland burn severity in two regions of the western US, 1984 to 2006 , 2011 .

[66]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[67]  Habshah Midi,et al.  Collinearity diagnostics of binary logistic regression model , 2010 .

[68]  R. Bradstock,et al.  Effects of weather, fuel and terrain on fire severity in topographically diverse landscapes of south-eastern Australia , 2010, Landscape Ecology.

[69]  R. Bradstock.  A biogeographic model of fire regimes in Australia: current and future implications , 2010 .

[70]  J. Sharples.  An overview of mountain meteorological effects relevant to fire behaviour and bushfire risk. , 2009 .

[71]  Penelope Morgan,et al.  A predictive model of burn severity based on 20-year satellite-inferred burn severity data in a large southwestern US wilderness area , 2009 .

[72]  Myoung-Soo Won,et al.  Relationship between landscape structure and burn severity at the landscape and class levels in Samchuck, South Korea , 2009 .

[73]  J. Keeley.  Fire intensity, fire severity and burn severity: a brief review and suggested usage , 2009 .

[74]  M. Hill,et al.  Slope, aspect and climate: Spatially explicit and implicit models of topographic microclimate in chalk grassland , 2008 .

[75]  Li Bai,et al.  Evaluating the supplier cooperative design ability using a novel support vector machine algorithm , 2008, 2008 12th International Conference on Computer Supported Cooperative Work in Design.

[76]  R. Kavanagh,et al.  Patchiness of prescribed burns in dry sclerophyll eucalypt forests in South-eastern Australia , 2007 .

[77]  Jay D. Miller,et al.  Quantifying burn severity in a heterogeneous landscape with a relative version of the delta Normalized Burn Ratio (dNBR) , 2007 .

[78]  Judith Winterkamp,et al.  Coupled influences of topography and wind on wildland fire behaviour , 2007 .

[79]  T. Sisk,et al.  Mapping the probability of large fire occurrence in northern Arizona, USA , 2006, Landscape Ecology.

[80]  Christof Bigler,et al.  MULTIPLE DISTURBANCE INTERACTIONS AND DROUGHT INFLUENCE FIRE SEVERITY IN ROCKY MOUNTAIN SUBALPINE FORESTS , 2005 .

[81]  Mark Noonan,et al.  The post-fire measurement of fire severity and intensity in the Christmas 2001 Sydney wildfires , 2004 .

[82]  Václav Hlavác,et al.  Multi-class support vector machine , 2002, Object recognition supported by user interaction for service robots.

[83]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[84]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[85]  Jean Carletta,et al.  Assessing Agreement on Classification Tasks: The Kappa Statistic , 1996, CL.

[86]  T. G. Freeman,et al.  Calculating catchment area with divergent flow based on a regular grid , 1991 .

[87]  G. Cary,et al.  What determines variation in remotely sensed fire severity? Consideration of remote sensing limitations and confounding factors , 2022, International Journal of Wildland Fire.

[88]  Binbin He,et al.  Application of Landsat ETM+ and OLI Data for Foliage Fuel Load Monitoring Using Radiative Transfer Model and Machine Learning Method , 2021, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[89]  Lee Ju Hyoung.  Prediction of Large-Scale Wildfires With the Canopy Stress Index Derived from Soil Moisture Active Passive , 2021, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[90]  G. McCarthy,et al.  Mapping fire severity and fire extent in forest in Victoria for ecological and fuel outcomes , 2017 .

[91]  Alistair M. S. Smith,et al.  Vegetation, topography and daily weather influenced burn severity in central Idaho and western Montana forests , 2015 .

[92]  Jürgen Böhner,et al.  Land-Surface Parameters Specific to Topo-Climatology , 2009 .

[93]  Carl H. Key,et al.  Landscape Assessment ( LA ) Sampling and Analysis Methods , 2004 .

[94]  Andrew D. Weiss.  Topographic position and landforms analysis , 2001 .

[95]  C. Thorne,et al.  Quantitative analysis of land surface topography , 1987 .