Use and performance of the Forest Fire Weather Index to model the risk of wildfire occurrence in the Alpine region

Assessing a territory’s fire proneness is fundamental when planning and undertaking effective forest protection and land management. Accurate methods to estimate the risk of fire ignition in natural environments have been proposed over the last decades and digital mapping has been used to identify critical areas. The Canadian Forest Fire Weather Index is a well-known fire danger rating index created and improved during the last 45 years by the Canadian Forest Service. The goal of this paper is twofold. Firstly, we evaluated whether the Forest Fire Weather Index is an adequate instrument to predict fire ignition in Alpine and sub-Alpine areas using quite a large dataset of meteorological and forest fire data collected in the Lombardy region (Northern Italy) between 2003 and 2011. By means of a spatial binary regression model, we demonstrated that Forest Fire Weather Index has a significant impact on the probability of fire ignition. Since this approach allows us to account for other characteristics of the territory in order to provide a more accurate estimate of the spatial wildfire dynamics at a moderately large scale, the second goal of the paper aims at creating a model to assess fire risk occurrence using the Forest Fire Weather Index and land use information. It has been found that ignition can easily occur in large forested areas whereas denser urban areas are less exposed to fire since they usually have no fuels to ignite. Nevertheless, since human activity has a direct impact on fire ignition human presence, it fosters ignition in forested areas. Finally, the model, including these spatial dimensions, has been employed to derive a probability map of fire occurrences at 1.5 km resolution, which is a fundamental instrument to develop optimal prevention and risk management policy plans for the decision maker.

[1]  F. Agterberg,et al.  Regression models for estimating mineral resources from geological map data , 1980 .

[2]  D. Stoyan,et al.  Statistical Analysis and Modelling of Spatial Point Patterns , 2008 .

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

[4]  Alan A. Ager,et al.  Wildfire risk estimation in the Mediterranean area , 2014 .

[5]  William J. Elliot,et al.  Spatial Prediction of Landslide Hazard Using Logistic Regression and ROC Analysis , 2006, Trans. GIS.

[6]  Francesco Uboldi,et al.  A spatial consistency test for surface observations from mesoscale meteorological networks , 2010 .

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

[8]  A. Menzel,et al.  Projection of fire potential to future climate scenarios in the Alpine area: some methodological considerations , 2013, Climatic Change.

[9]  A. González-Cabán,et al.  Proceedings of the second international symposium on fire economics, planning, and policy: a global view , 2008 .

[10]  R. Platt.  A Model of Exurban Land-Use Change and Wildfire Mitigation , 2006 .

[11]  Susan I. Stewart,et al.  Defining the Wildland-Urban Interface , 2007 .

[12]  Francisco Rego,et al.  Modelling the effects of distance on the probability of fire detection from lookouts , 2006 .

[13]  Fire Weather Index Implementation in the Alpine Area , 2012 .

[14]  J. H. Schuenemeyer,et al.  Generalized Linear Models (2nd ed.) , 1992 .

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

[16]  Timo Pukkala,et al.  Using multiscale spatial analysis to assess fire ignition density in Catalonia, Spain , 2011, Annals of Forest Science.

[17]  Harald Bugmann,et al.  Linking Forest Fire Regimes and Climate—A Historical Analysis in a Dry Inner Alpine Valley , 2009, Ecosystems.

[18]  I. Hernandez-Magallanes,et al.  Combining ungrouped and grouped wildfire data to estimate fire risk , 2014 .

[19]  Nikos Koutsias,et al.  FIRE OCCURRENCE PATTERNS AT LANDSCAPE LEVEL: BEYOND POSITIONAL ACCURACY OF IGNITION POINTS WITH KERNEL DENSITY ESTIMATION METHODS , 2008 .

[20]  Susan I. Stewart,et al.  The wildland-urban interface in the United States based on 125 million building locations. , 2005, Ecological applications : a publication of the Ecological Society of America.

[21]  F. Agterberg,et al.  Automatic contouring of geological maps to detect target areas for mineral exploration , 1974 .

[22]  M. Vasconcelos,et al.  Spatial Prediction of Fire Ignition Probabilities: Comparing Logistic Regression and Neural Networks , 2001 .

[23]  Fernando Bação,et al.  Characterizing and modelling the spatial patterns of wildfire ignitions in Portugal: fire initiation and resulting burned area. , 2008 .

[24]  Jeffrey A. Cardille,et al.  Occurrence of wildfire in the northern Great Lakes Region: Effects of land cover and land ownership assessed at multiple scales , 2001 .

[25]  M. Moritz,et al.  Environmental controls on the distribution of wildfire at multiple spatial scales , 2009 .

[26]  N. Fisher,et al.  Spatial logistic regression and change-of-support in Poisson point processes , 2010 .

[27]  J. Symanzik.  Statistical Analysis of Spatial Point Patterns (2nd ed.) , 2005 .