fw

d neural networks were applied to develop models for predicting biomass Accurate estimation of forest biomass is required for et al., 2004; Patenaude et al., in press). Forest inventories and remote sensing are the two methods used for estimating C provides a good source for statistics of state and change in numbers reported by independent means. In their paper, Rosenqvist et al. (2003) have presented an overview of the ontext of the Kyoto Remote Sensing of Environment 9greenhouse gas inventories, terrestrial carbon (C) accounting and climate change modeling studies. Forests are important components in the global carbon budget since they play a key role in the terrestrial C cycle. Boreal forests are of particular interest because, among all the biomes, they may undergo the greatest climatically induced changes during the 21st century (Bonan et al., 1992; Myneni et al., 1997)