Conclusions & Outlook
The focal objective of this study was to provide an assessment of state and dynamics of Chilean vegetation inside 4 sites along a latitudinal gradient. This was accomplished by producing extensive LULC maps using machine learning algorithms (RF & SVM) and calculating NDVI time series for a 4- year period between 2013 and 2017. The BFAST algorithm (Verbesselt et al., 2010) was applied to statistically analyze the time series catchment-wide, vegetation category specific and inside altitude belts to investigate the impact of climatic forcing. Furthermore, the performance of the recently released 12 m resolution TanDEM-X DEM in topographic correction and as an input in the classification variable framework was tested and compared to a Lidar DEM and the established SRTM DEM.