Please use this identifier to cite or link to this item:
|Title:||First Steps towards a Long Term Forest Fire Risk of Europe|
|Authors:||SANTOS DE OLIVEIRA SANDRA; CAMIA Andrea; SAN-MIGUEL-AYANZ Jesus|
|Citation:||Proceedings of the VII International EARSeL Workshop - Advances on Remote Sensing and GIS applications in Forest Fire Management p. 79-83|
|Type:||Articles in periodicals and books|
|Abstract:||This work provides the current status of a research effort aimed at developing a long term fire risk map of Europe, which will be included as a component of the European Forest Fire Information System (EFFIS). The fire risk model adopted for the assessment is based on the approach that combines fire occurrence and fire outcome, thus encompassing probability of ignition, estimated fire behavior and expected consequences, and aiming to integrate physical, biological and socio-economic factors. The first step has been the enhancement of the fire occurrence data stored in the European Fire Database of EFFIS, in which recorded fire ignitions exhibit a certain degree of geo-location uncertainty. Location of fire ignition points is given in most cases as administrative district without geographical coordinates. Therefore methods to approximate density estimations of the spatial distribution of fire ignition points are needed. One of the options tested in this study is the use of land cover data to constrain the geo-location of the ignition points recorded in a given administrative district inside the boundaries of the fire spatial domain (i.e. forested and wildland areas). The point distribution is made randomly or with a weighted probability filtering, and a continuous surface is then created by kernel density methods. A second step is the analysis of potential variables affecting fire occurrence. The list of these variables is being compiled on the basis of extensive literature review and experts¿ knowledge. The final selection of the variables to be used in the model will be based on data availability and exploratory statistical analysis. To assess the significance of the predictor variables in fire occurrence, several alternative methods are being explored, among which are logistic regression and geographically weighted regression. The methodology firstly developed for the Euro-Mediterranean region will then be applied to the other European countries, making the necessary adjustments to cope with the specific conditions of each area.|
|JRC Institute:||Institute for Environment and Sustainability|
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.