ASSESSING GRASSLAND MOISTURE AND BIOMASS IN TASMANIA - THE APPLICATION OF REMOTE-SENSING AND EMPIRICAL-MODELS FOR A CLOUDY ENVIRONMENT

The operational feasibility of NOAA/AVHRR data and two semi-empirical moisture models were evaluated in the grasslands of southeastern Tasma nia (Australia) during the 1988/89 fire season. A limited ground-truth ing experiment compared the grassland dry biomass, soil moisture and f uel moisture with the satellite derived NDVI and the Soil Dryness Inde x (SDI) and the Grassland Curing Index (GCI). The NDVI gave good resul ts for fuel moisture content (FMC) and soil moisture content (SMC) but unreliable image availability precludes the use of NDVI as a stand al one system for fire managers. The SDI and GCI also performed well in p redicting SMC and FMC. Very good results were obtained when the NDVI a nd the GCI were combined. These results suggest the combination of dat a will provide both the accuracy and the continuity of information nee ded for operational use by fire managers. The methods used here could be cheaply and quickly repeated for use in other similar fire prone an d cloudy environments.