A data-driven approach to assess large fire size generation in Greece

Identifying factors and drivers which control large fire size generation is critical for planning fire management activities. This study attempts to determine the role of fire suppression tactics and behavior, weather, topography and landscape features on two different datasets of large fire size (500–1000 ha) and very large fire size (>1000 ha) compared to two datasets of small fire size (<50 ha) which occurred in Greece, during the period 1984–2009. In this context, we used a logistic regression (LR) analysis and two machine learning algorithms: random forest (RF) and boosting classification trees (BCT). The models comparison was based on the area under the receiver operating characteristic curve and the observed overall accuracy. The comparison indicated that RF had greater ability than LR and BCT to predict large fire generation. Results from the RF classifier algorithm showed that large fire generation mainly depended on fire suppression tactics and the prevailing weather conditions. This improved understanding of the factors which drive fire ignitions to turn into large fire sizes will provide the opportunity for land and forest managers to increase fire awareness and the development of concrete initiatives for successful fire management.

[1]  Christopher Lucas,et al.  Prediction of the probability of large fires in the Sydney region of south-eastern Australia using fire weather , 2009 .

[2]  A. P. Dimitrakopoulos,et al.  PYROSTAT -- a computer program for forest fire data inventory and analysis in Mediterranean countries , 2001, Environ. Model. Softw..

[3]  F. Moreira,et al.  Landscape--wildfire interactions in southern Europe: implications for landscape management. , 2011, Journal of environmental management.

[4]  R. Trigo,et al.  Daily synoptic conditions associated with large fire occurrence in Mediterranean France: evidence for a wind‐driven fire regime , 2017 .

[5]  Yu Chang,et al.  Land Use and Land Cover Change Analysis and Prediction in the Upper Reaches of the Minjiang River, China , 2009, Environmental management.

[6]  J. Carreiras,et al.  Fire frequency analysis in Portugal (1975-2005), using Landsat-based burnt area maps , 2012 .

[7]  S. Ventura,et al.  ENVIRONMENTAL AND SOCIAL FACTORS INFLUENCING WILDFIRES IN THE UPPER MIDWEST, UNITED STATES , 2001 .

[8]  J. L. Parra,et al.  Very high resolution interpolated climate surfaces for global land areas , 2005 .

[9]  E. Chuvieco,et al.  Estimation of dead fuel moisture content from meteorological data in Mediterranean areas. Applications in fire danger assessment , 2007 .

[10]  Heiko Balzter,et al.  Mapping regional patterns of large forest fires in Wildland-Urban Interface areas in Europe. , 2016, Journal of environmental management.

[11]  J. M. Moreno,et al.  Spatial distribution of forest fires in Sierra de Gredos (Central Spain) , 2001 .

[12]  R. Rothermel,et al.  How to Predict the Spread and Intensity of Forest and Range Fires , 2017 .

[13]  Bruce Shindler,et al.  Trust, acceptance, and citizen–agency interactions after large fires: influences on planning processes , 2010 .

[14]  Qianlai Zhuang,et al.  Modeling Large Fire Frequency and Burned Area in Canadian Terrestrial Ecosystems with Poisson Models , 2012, Environmental Modeling & Assessment.

[15]  Ulric J. Lund,et al.  Identifying geographical patterns of wildfire orientation: A watershed-based analysis , 2012 .

[16]  David J. Hand,et al.  A Simple Generalisation of the Area Under the ROC Curve for Multiple Class Classification Problems , 2001, Machine Learning.

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

[18]  Tom Fawcett,et al.  An introduction to ROC analysis , 2006, Pattern Recognit. Lett..

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

[20]  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 .

[21]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[22]  D. Hosmer,et al.  Applied Logistic Regression , 1991 .

[23]  Susan I. Stewart,et al.  Human and biophysical influences on fire occurrence in the United States. , 2013, Ecological applications : a publication of the Ecological Society of America.

[24]  J. Abatzoglou,et al.  Modeling very large-fire occurrences over the continental United States from weather and climate forcing , 2014 .

[25]  J. Seibert,et al.  On the calculation of the topographic wetness index: evaluation of different methods based on field observations , 2005 .

[26]  F. Moreira,et al.  Modeling and mapping wildfire ignition risk in Portugal , 2009 .

[27]  David R. Brillinger,et al.  Probability based models for estimation of wildfire risk , 2004 .

[28]  Zhihua Liu,et al.  Climatic and Landscape Influences on Fire Regimes from 1984 to 2010 in the Western United States , 2015, PloS one.

[29]  A. Gill,et al.  Large fires, fire effects and the fire-regime concept , 2008 .

[30]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[31]  Lluís Brotons,et al.  Predictive modelling of fire occurrences from different fire spread patterns in Mediterranean landscapes , 2015 .

[32]  Francisco Moreira,et al.  Size-dependent pattern of wildfire ignitions in Portugal: when do ignitions turn into big fires? , 2010, Landscape Ecology.

[33]  Xavier Pons,et al.  Spatial patterns of fire occurrence in Catalonia, NE, Spain , 2004, Landscape Ecology.

[34]  David L. Martell,et al.  A simulation model of the growth and suppression of large forest fires in Ontario , 2007 .

[35]  T. Curt,et al.  Spatiotemporal patterns of changes in fire regime and climate: defining the pyroclimates of south-eastern France (Mediterranean Basin) , 2015, Climatic Change.

[36]  J. W. Wagtendonk,et al.  Fire Frequency, Area Burned, and Severity: A Quantitative Approach to Defining a Normal Fire Year , 2011 .

[37]  K. Hirsch,et al.  Large forest fires in Canada, 1959–1997 , 2002 .

[38]  Iván Torres,et al.  Fire Severity in a Large Fire in a Pinus pinaster Forest is Highly Predictable from Burning Conditions, Stand Structure, and Topography , 2014, Ecosystems.

[39]  Futao Guo,et al.  What drives forest fire in Fujian, China? Evidence from logistic regression and Random Forests , 2016 .

[40]  Zhihua Liu,et al.  Identifying the Threshold of Dominant Controls on Fire Spread in a Boreal Forest Landscape of Northeast China , 2013, PloS one.

[41]  Nairanjana Dasgupta,et al.  Land Cover Type and Fire in Portugal: Do Fires Burn Land Cover Selectively? , 2005, Landscape Ecology.

[42]  Rodney X. Sturdivant,et al.  Applied Logistic Regression: Hosmer/Applied Logistic Regression , 2005 .

[43]  Wenhui Wang,et al.  Modeling Anthropogenic Fire Occurrence in the Boreal Forest of China Using Logistic Regression and Random Forests , 2016 .

[44]  J. Dupuy,et al.  Size of wildfires in the Euro-Mediterranean region: observations and theoretical analysis , 2015 .

[45]  Marielle Jappiot,et al.  What causes large fires in Southern France , 2013 .

[46]  Thomas Curt,et al.  Wildfire frequency varies with the size and shape of fuel types in southeastern France: implications for environmental management. , 2013, Journal of environmental management.

[47]  P. Fernandes,et al.  Changes in wildfire severity from maritime pine woodland to contiguous forest types in the mountains of northwestern Portugal , 2010 .

[48]  J. Abatzoglou,et al.  Multi‐scalar influence of weather and climate on very large‐fires in the Eastern United States , 2015 .

[49]  E. Chuvieco,et al.  Fire regime changes and major driving forces in Spain from 1968 to 2010 , 2014 .

[50]  P. Bermudez,et al.  Spatial and temporal extremes of wildfire sizes in Portugal (1984–2004) , 2009 .

[51]  P. Fernandes,et al.  The role of fire-suppression force in limiting the spread of extremely large forest fires in Portugal , 2016, European Journal of Forest Research.

[52]  John E. Deeming,et al.  The National Fire-Danger Rating System , 2018 .

[53]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[54]  James M. Dyer.  Assessing topographic patterns in moisture use and stress using a water balance approach , 2009, Landscape Ecology.

[55]  N. Guiomar,et al.  Bottom-Up Variables Govern Large-Fire Size in Portugal , 2016, Ecosystems.

[56]  David L. Martell,et al.  The impact of fire suppression, vegetation, and weather on the area burned by lightning-caused forest fires in Ontario , 2008 .

[57]  Isaac C. Grenfell,et al.  The relationship of large fire occurrence with drought and fire danger indices in the western USA, 1984-2008: the role of temporal scale , 2013 .

[58]  D. Ruppert.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2004 .

[59]  Francesco Neri,et al.  Analysis of Helicopter Activities in Forest Fire-Fighting , 2014 .

[60]  José M. C. Pereira,et al.  Synoptic patterns associated with large summer forest fires in Portugal , 2005 .

[61]  Jennifer A. Miller,et al.  Mapping Species Distributions: Spatial Inference and Prediction , 2010 .

[62]  I. Mitsopoulos,et al.  Statistical analysis of the fire environment of large forest fires (>1000 ha) in Greece , 2011 .

[63]  Carol Miller,et al.  Contributions of Ignitions, Fuels, and Weather to the Spatial Patterns of Burn Probability of a Boreal Landscape , 2011, Ecosystems.

[64]  E. Johnson,et al.  The Relative Importance of Fuels and Weather on Fire Behavior in Subalpine Forests , 1995 .

[65]  Lasse Loepfe,et al.  Feedbacks between fuel reduction and landscape homogenisation determine fire regimes in three Mediterranean areas , 2010 .

[66]  Juan de la Riva,et al.  An insight into machine-learning algorithms to model human-caused wildfire occurrence , 2014, Environ. Model. Softw..

[67]  Steven E. Franklin,et al.  Multi-scale object-based image analysis and feature selection of multi-sensor earth observation imagery using random forests , 2012 .

[68]  B. McCune,et al.  Analysis of Ecological Communities , 2002 .

[69]  C. E. Van Wagner,et al.  Development and structure of the Canadian Forest Fire Weather Index System , 1987 .

[70]  J. San-Miguel-Ayanz,et al.  Analysis of large fires in European Mediterranean landscapes: Lessons learned and perspectives , 2013 .

[71]  Robert Mavsar,et al.  Analysis of factors influencing deployment of fire suppression resources in Spain using artificial neural networks , 2016 .