Monitoring of tropical forest cover and biomass by combining VHR image sample and wall to wall optical and SAR data

Methods for forest resources inventory for tropical region were developed and tested in Lao PDR. The key element of the study was application of a sample of satellite data with better than one meter ground resolution. Using a hierarchical approach, information from of sample of very high resolution (VHR) image data was combined with lower resolution satellite data of AVNIR and ALOS/PalSAR that were available wall to wall. Theoretical study with the GlobCover data showed that the same accuracy could be achieved in forest area estimation with 1 % stratified sample of VHR data as with 6.25 % non-stratified sample. Good agreement between the visually interpreted VHR plots and wall to wall maps was achieved in land cover classification. In the biomass estimation a value of R2=0.6 was achieved with ground plot test data.