A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science, School of Engineering and Technology

Thesis (M.Sc.) - Asian Institute of Technology, 2005

Process Based Models (PBMs) can be driven with information derived from satellite sensors. In particular, Absorbed photosynthetically active radiation (APAR) was estimated from global solar radiation, derived from an established empirical relationship based on average maximum and minimum temperatures, and from a linear relation with the satellite-derived normalised difference vegetation index (NDVI) which represents the photosynthetic capacity of all vegetation and is often correlated with the fraction of PAR absorbed FPAR. 3-PGS is a production model driven by remote sensing data, running at monthly time steps and have many advantages and reduce fields work that take time and laborious. The using should be considered in each parameter that are site specific and type of satellite sensor. In this study, has been use d LANDSAT ETM+ to derived NDVI. For 3PG model output showed the good relationship between measured (groundbased measurements) and simulated (modeled) values by Coefficient of determination (R) more than 0.7 and The root mean square errors (RMSE) is less than other models for all parameters (Net Primary production ;NPP ,Aboveground Biomass ; Wabv , Mean Annual Increment ;MAI and Stand Volume ;SV ). For 3PGS model output showed the not that good relationship between measured (ground-based measurements) and simulated (modeled) values by Coefficient of determination (R) less than 0.5 in Eucalyptus plot and The root mean square errors (RMSE) is more than 3PG model for all parameters. Because of low NDVI value that is effected by soil reflectance. Acacia mangium (plot A1) showed highly correlated between measured (ground-based measurements) and simulated (modeled) values; more than Eucalyptus camaldulensis (plot E2 to E6), for Wabv and NPP A were in near 1:1 agreement with Coefficient of determination (R) more than 0.8. Remote sensing data are utilized in these types of models providing 4 alternatives strategies (1) To estimate input variables (2) To test and validate predictions (3) To update or adjust ecological models (4) To apply ecological models to understand RS respond.