2016 Volume 32 Issue 4 Pages 133-145
Forest areas in northern Thailand are endangered by wildfires. Fuel load is recognized as one of the important factors that influence wildfire occurrence and affect fire behavior. We compared the capabilities of seven vegetation indices (VIs) of Landsat satellite data in estimating leaf biomass, which is a parameter used in a leaf fuel load prediction model. The model contributes to the assessment of wildfire risk by identifying the spatial distribution of leaf fuel load to assess wildfire-prone areas across different landscapes. Significant relationships between the calculated standard leaf biomass and the seven VIs showed that a normalized difference vegetation index (NDVI) had the strongest relationship with leaf biomass. The NDVI images of normal and dry seasons (i.e., a seasonal NDVI) were used to estimate the quantities of seasonal leaf biomass and used to detect the missing leaf biomass or the leaf fuel load on the ground surface. The model of leaf fuel load prediction, based on the seasonal NDVI images, achieved accuracy of 80.43% (dipterocarp) and 71.36% (deciduous) using a statistical inference between the predicted and field-derived data. Moreover, model validation using a paired t-test indicated there was no significant difference between the means of the two data sets (p-value >0.05). Therefore, the predicted leaf fuel load derived from the developed model could be used as a substitute for estimating the actual leaf fuel load in forested areas, especially in dipterocarp and deciduous forests.
