Estimating Forest Fire Losses Using Stochastic Approach: Case Study of the Kroumiria Mountains (Northwestern Tunisia)

ABSTRACT Kroumiria Mountains (northwestern Tunisia) have experienced major fires, making them the main loss reason of Tunisian forested areas. The ability of accurately forecasting or modeling forest fire areas may significantly aid optimizing fire-fighting strategies. However, there are still limitations in the empirical study of forest fire loss estimation because the poor availability and low quality of fire data. In this study, a stochastic approach based on Markov process was developed for the prediction of burned areas, using available meteorological data sets and GIS layers related to the forest under analysis. The Self-organizing map (SOM) was initially used to classify spatiotemporal factors influencing the fire behavior. Subsequently, the SOM clusters were incorporated into a Hidden Markov Model (HMM) framework to model their corresponding burned areas. Results achieved using a database of 829 forest fires records between 1985 and 2016, showed the appropriateness of the HMM approach for the prediction of burned areas compared with a state-of-the art machine learning methods. The transition probability matrix (TPM) and the emission probability matrix (EPM) were also analyzed to further understand the spatiotemporal patterns of fire losses.

[1]  Smain El-Amine Henaoui,et al.  THE FIRE RISK OF THE PLANT GROUPINGS WITH CISTUS IN THE AREA OF TLEMCEN (WESTERN ALGERIA) , 2013 .

[2]  M. C. Llasat,et al.  Climate change impacts on wildfires in a Mediterranean environment , 2014, Climatic Change.

[3]  B. R. Ramesh,et al.  Environmental susceptibility model for predicting forest fire occurrence in the Western Ghats of India , 2012 .

[4]  Emilio Chuvieco,et al.  Lightning-caused fires in Central Spain: Development of a probability model of occurrence for two Spanish regions , 2012 .

[5]  J. Pausas,et al.  Changes in Fire and Climate in the Eastern Iberian Peninsula (Mediterranean Basin) , 2004 .

[6]  Shuying Sun,et al.  HMM-DM: identifying differentially methylated regions using a hidden Markov model , 2016, Statistical applications in genetics and molecular biology.

[7]  James K. Hane,et al.  CodingQuarry: highly accurate hidden Markov model gene prediction in fungal genomes using RNA-seq transcripts , 2015, BMC Genomics.

[8]  Michelangelo Ceci,et al.  Network regression with predictive clustering trees , 2011, Data Mining and Knowledge Discovery.

[9]  Milenov Pavel,et al.  An automated SOM clustering based on data topology , 2010, ESANN 2010.

[10]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[11]  F. Moreira,et al.  Factors affecting post-fire crown regeneration in cork oak (Quercus suber L.) trees , 2009, European Journal of Forest Research.

[12]  J V Tu,et al.  Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. , 1996, Journal of clinical epidemiology.

[13]  Alan A. Ager,et al.  Assessing the effect of a fuel break network to reduce burnt area and wildfire risk transmission , 2016 .

[14]  J. Randerson,et al.  Assessing variability and long-term trends in burned area by merging multiple satellite fire products , 2009 .

[15]  Spyros Makridakis,et al.  The M3-Competition: results, conclusions and implications , 2000 .

[16]  Carlos Guestrin,et al.  "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[17]  A. P. Dimitrakopoulos,et al.  A statistical classification of Mediterranean species based on their flammability components , 2001 .

[18]  S. Menard.  Applied Logistic Regression Analysis , 1996 .

[19]  Imad H. Elhajj,et al.  Efficient forest fire occurrence prediction for developing countries using two weather parameters , 2011, Eng. Appl. Artif. Intell..

[20]  Samuel Kaski,et al.  Self-Organized Formation of Various Invariant-Feature Filters in the Adaptive-Subspace SOM , 1997, Neural Computation.

[21]  Pierre Ailliot,et al.  Markov-switching autoregressive models for wind time series , 2012, Environ. Model. Softw..

[22]  Kadim Tasdemir,et al.  An automated SOM clustering based on data topology , 2010, ESANN.

[23]  W. Dlamini.  Application of Bayesian networks for fire risk mapping using GIS and remote sensing data , 2011 .

[24]  Jiaqiu Wang,et al.  Integrated Spatio‐temporal Data Mining for Forest Fire Prediction , 2008, Trans. GIS.

[25]  Linda Hoffmann,et al.  Multivariate isotonic regression and its algorithms , 2009 .

[26]  Krista M. Gebert,et al.  Spatially explicit forecasts of large wildland fire probability and suppression costs for California , 2011 .

[27]  Nikos Koutsias,et al.  Modelling long-term fire occurrence factors in Spain by accounting for local variations with geographically weighted regression , 2013 .

[28]  D. Xie,et al.  Prediction for Burned Area of Forest Fires Based on SVM Model , 2014 .

[29]  Yue-Hong Chou,et al.  Management of wildfires with a geographical information system , 1992, Int. J. Geogr. Inf. Sci..

[30]  A. Murat Ozbayoglu,et al.  Estimation of the Burned Area in Forest Fires Using Computational Intelligence Techniques , 2012, Complex Adaptive Systems.

[31]  Aarti Singh,et al.  PREDICTING BURNED AREAS OF FOREST FIRES : AN ARTIFICIAL INTELLIGENCE APPROACH , 2017 .

[32]  P. Cortez,et al.  A data mining approach to predict forest fires using meteorological data , 2007 .

[33]  F. Mouillot,et al.  Integrating MODIS images in a water budget model for dynamic functioning and drought simulation of a Mediterranean forest in Tunisia , 2012 .

[34]  F. J. Lozano,et al.  A multi-scale approach for modeling fire occurrence probability using satellite data and classification trees: A case study in a mountainous Mediterranean region , 2008 .

[35]  Giuseppe Amatulli,et al.  Assessing long‐term fire risk at local scale by means of decision tree technique , 2006 .

[36]  O. Sass,et al.  Using multi variate data mining techniques for estimating fire susceptibility of Tyrolean forests , 2014 .

[37]  Terrence J. Sejnowski,et al.  Unsupervised Learning , 2018, Encyclopedia of GIS.

[38]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[39]  Nilton Cesar Fiedler,et al.  Applying GIS to develop a model for forest fire risk: A case study in Espírito Santo, Brazil. , 2016, Journal of environmental management.

[40]  Hammadi Achour,et al.  Modelling the spatial distribution of snake species in northwestern Tunisia using maximum entropy (Maxent) and Geographic Information System (GIS) , 2017, Journal of Forestry Research.

[41]  S. Fotheringham,et al.  Modeling the spatial variation of the explanatory factors of human-caused wildfires in Spain using geographically weighted logistic regression , 2014 .

[42]  Zhiqiang Ma,et al.  Hadoop-based ARIMA Algorithm and its Application in Weather Forecast , 2013 .

[43]  Ku Ruhana Ku-Mahamud,et al.  Pattern Extraction and Rule Generation of Forest Fire using Sliding Window Technique , 2009, Comput. Inf. Sci..

[44]  K. Malarz,et al.  Are Forest Fires Predictable , 2002 .

[45]  Robert Sabourin,et al.  Leave-One-Out-Training and Leave-One-Out-Testing Hidden Markov Models for a Handwritten Numeral Recognizer: The Implications of a Single Classifier and Multiple Classifications , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[46]  Youssef Safi,et al.  Prediction of forest fires using Artificial neural networks , 2013 .

[47]  Hayley J. Fowler,et al.  An assessment of changes in seasonal and annual extreme rainfall in the UK between 1961 and 2009 , 2013 .

[48]  Esa Alhoniemi,et al.  Clustering of the self-organizing map , 2000, IEEE Trans. Neural Networks Learn. Syst..

[49]  Young-Seuk Park,et al.  Hazard rating of pine trees from a forest insect pest using artificial neural networks , 2006 .

[50]  Douglas G. Woolford,et al.  Erlangian Approximations for the Transient Analysis of a Fluid Queue Model for Forest Fire Perimeter , 2009, INFOR Inf. Syst. Oper. Res..

[51]  V. Caselles,et al.  Application of artificial neural networks and logistic regression to the prediction of forest fire danger in Galicia using MODIS data , 2012 .

[52]  A. Derridj,et al.  Analysis of forest fires causes and their motivations in northern Algeria: the Delphi method , 2013 .

[53]  Paola Ovando,et al.  Accounting for single and aggregated forest incomes: Application to public cork oak forests in Jerez (Spain) and Iteimia (Tunisia) , 2008 .

[54]  Wfp,et al.  The State of Food Insecurity in the World , 2011 .

[55]  J. Tao,et al.  Analysis of forest fires in Northeast China from 2003 to 2011 , 2013 .

[56]  Joseph Bockhorst,et al.  Auto-Regressive HMM Inference with Incomplete Data for Short-Horizon Wind Forecasting , 2010, NIPS.

[57]  E. Chuvieco,et al.  Human-caused wildfire risk rating for prevention planning in Spain. , 2009, Journal of environmental management.

[58]  Yoram J. Kaufman,et al.  An Enhanced Contextual Fire Detection Algorithm for MODIS , 2003 .

[59]  Arne Eigenfeldt,et al.  A Methodology for the Comparison of Melodic Generation Models Using Meta-Melo , 2013, ISMIR.

[60]  Anthony Widjaja,et al.  Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond , 2003, IEEE Transactions on Neural Networks.

[61]  Soohee Han,et al.  Forest fire risk mapping of Kolli Hills, India, considering subjectivity and inconsistency issues , 2012, Natural Hazards.

[62]  J. Greenberg,et al.  Spatial variability in wildfire probability across the western United States , 2012 .

[63]  Zohre Sadat Pourtaghi,et al.  Investigation of general indicators influencing on forest fire and its susceptibility modeling using different data mining techniques , 2016 .