NEW HIGH-RESOLUTION DSM/DTM DATA FOR THE AMAZON - PRELIMINARY ANALYSIS WITH SRTM, ASTER GDEM
The analysis showed that ASTER GDEM presents a high level of noise and artefacts from the automatic image processing chain, with low correlation to the morphology depicted in the other DEMs. RAM Digital Surface Models (i.e., canopy height) have a good correlation with SRTM, although with higher elevation due the use of X-band Radar, which does not penetrates the forest canopy. RAM Digital Terrain Models exhibits the topography under the forest allowing the identification of morphological features that could be hidden under the vegetation.
Future studies should be carried out to determine, for instance, the level of detail of DTM-derived drainage networks as well as to evaluate the noise of 5m-resolution DTMs and possible filtering or smoothing procedures.
This work was supported by CNPq grant 306294/2012-5 and by a collaborative Dimensions of Biodiversity-BIOTA grant supported by FAPESP (2012/50260-6), NSF and NASA.