|
Résumé :
|
In the European Mediterranean region, around 500 000 hectares of forest are being burnt every year, and national authorities have set up important fire fighting systems. The estimation of fuel moisture content (FMC) is the critical aspect of fire danger estimation, since vegetation is the main fuel material on a wildland fire. FMC is generally estimated from meteorological variables (Viegas et al., 1994). However it has been shown that vegetation stress can be monitored using remote sensing data, especially NOAA-AVHRR data, due to an avaibility of optical and thermal infrared bands and a daily and exhaustive coverage. Within the EC-funded MEGAFIReS project, ten different AVHRR-based indices have been tested with '96 and '97 summer data. These indices are NDVI, ST, surface minus air temperature (ST-AT), NDVI/ST (Alonso et al., 1996), SAVI (Huete, 1988), GEMI (Pinty & Verstraete, 1992), greeness and site moisture indices (Eidenshink et al., 1990) and water stress index (WSI, Vidal et al., 1994). These indices are validated against fuel moistures measurements made on three different study areas, characteristic of the Mediterranean vegetation : Les Maures (France), Cabañeros national park (Spain), and Chalkidiki peninsula (Greece). The results obtained are presented and discussed in an operational context.
|
