Relationships between forest variables and remote sensing data in a<i>Nothofagus pumilio</i>forest

Abstract

The use of remote sensing data is an alternative approach to extrapolate forest variables measured in the field over large and inaccessible forested areas. The objective of this study was to investigate the relationship between forest variables and four remotely sensed vegetation parameters (simple ratio, SR; normalized difference vegetation index, NDVI; green normalized difference vegetation index, GNDVI; vegetation cover fraction, VCF) in a Nothofagus pumilio forest. Forest variables were measured in the Magellan Region of Chile and vegetation parameters were retrieved from an Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) image of the study area. An accuracy comparison of regression equations showed that the models that used the VCF together with the SR had the best performances of the basal area [R 2 0.72], stem biomass (R 2 0.66], and aboveground tree biomass (AGTB) [R 2 0.73]. This study concludes that the regressions developed could be successfully used to estimate AGTB and carbon storage for the Nothofagus pumilio forest in the Magellan Region.