Context Information for Understanding Forest Fire Using Evolutionary Computation L. Usero 3, 4 , A. Arroyo 2 , and J. Calvo 1 1 Dpto. de Organizaci´on y Estructura de la informaci´on, Universidad Polit´ ecnica de Madrid, Spain 2 Dpto. de Sistemas Inteligentes Aplicados Universidad Polit´ ecnica de Madrid, Spain 3 Dpto. Ciencias de la Computaci´on Universidad de Alcal´a, Spain 4 Center for Spatial Technologies and Remote Sensing, U. California. One Shields Ave. 95616-8617 Davis, CA. USA aarroyo@eui.upm.es, luis.usero@uah.es, jcalvo@tdi.eui.upm.es Abstract. One of the major forces for understanding forest fire risk and behavior is the fire fuel. Fire risk and behavior depend on the fuel properties such as moisture content. Context information on vegetation water content is vital for understanding the processes involved in initi- ation and propagation of forest fires. In that sense, a novel method was tested to estimate vegetation canopy water content (CWC) from simu- lated MODIS satellite data. An inversion of a radiative transfer model called Forest Light Interaction-Model (FLIM) from performed using evo- lutionary computation. CWC is critical, among other applications, in wildfire risk assessment since a decrease in CWC causes higher proba- bility to have wildfire occurrence. Simulations were carried out with the FLIM model for a wide range of forest canopy characteristics and CWC values. A 50 subsample of the simulations was used for the training pro- cess and 50 for the validation providing a RMSE=0.74 and r2=0.62. Further research is needed to apply this method on real MODIS images. Keywords: Genetic Programing, Vegetation Water Content, Forest Fire Understanding. 1 Introduction Detecting the water content (Cw) is useful to monitor vegetation stress even forest fire. Context information gathered by remote sensing is vital to understand the forest fire risk and behavior. So it is significant to use of remote sensing to measure spectral properties of leaves can provide an indirect structural canopy variables estimation in order to obtain a comprehensive spatial and temporal distribution. Vegetation canopy water content (CWC) is the weight of the water per leaf area unit and per ground area unit. CWC retrieval results critical for several environmental applications including wildfire risk [2]. Fires front advances when J. Mira and J.R. ´ Alvarez (Eds.): IWINAC 2007, Part II, LNCS 4528, pp. 271–276, 2007. c  Springer-Verlag Berlin Heidelberg 2007