Title: Modeling long-term forest fire risk at the European level: first assessment of fire probability and vulnerability
Citation: Abstracts of the VI International Conference on Forest Fire Research
Publisher: Association for the Development of Industrial Aerodynamics (ADAI)
Publication Year: 2010
JRC N°: JRC62848
URI: http://publications.jrc.ec.europa.eu/repository/handle/JRC62848
Type: Articles in periodicals and books
Abstract: Long-term forest fire risk is based on factors that remain stable for at least one fire season, giving a valuable contribution to the improvement of fire prevention systems and to the management of fire resources. This paper presents the methodology and first results for the development of a long-term forest fire risk model at the European level, based on the combination of three pillars: fire probability, fire behaviour potential and vulnerability (expected fire outcome). The methodology applied to assess fire probability and the results obtained so far for the Euro-Mediterranean region for the period 2000-2007 are described and the approach to assess vulnerability is briefly explained. Fire probability was assessed considering ignitions and the conditions for fire to spread in case an ignition occurs. Taking into account the geo-location inaccuracy associated with the fire events, fire density was used as a proxy for fire ignition and it was calculated with kernel density estimation methods. Results show that there is a high variability between and within countries, with the Northern parts of Portugal and Spain having the highest values, while Northern France and Greece have the lowest density. Portugal is the country with the highest fire density values for the whole period, 5 times higher than Spain, which has the second highest values. Temporal differences were also found: the years 2003 and 2005 showed the higher values of fire density for most of the countries, except for Greece, where the higher values were estimated for the years 2004 and 2006. To assess the factors that influence the spatial and temporal distribution of fire density, several variables were selected based on previous knowledge of fire occurrence and extensive literature review, such as topographic variables (elevation, slope, aspect and roughness), fuel types, weather data, population density, unemployment rate and agricultural data. These variables were pre-processed to be included in the model as predictors. Vulnerability is considered as the maximum impact that fires might have at the time of their occurrence onto goods and services provided by ecosystems. The limitation of data at the European level poses a serious constraint to this assessment and requires a step-wise approach that allows the incorporation of new data as it becomes available. In a first step, qualitative values are assigned to the potential soil erosion, loss of growing stock, recreation and protected areas. A method for weighting and aggregating these parameters is under development.
JRC Directorate:Sustainable Resources

Files in This Item:
There are no files associated with this item.

Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.