ABSTRACT Friedrich-Schiller-University Jena Institute for Geography Department of Earth Observation Grietgasse 6, 07743 Jena www.eo.uni-jena.de STUDY AREA LAND COVER AND INUNDATION DYNAMICS L-BAND BACKSCATTER AND VEG HEIGHT L-BAND BASED INUNDATION MAP PRELIMINARY RESULTS 1. Sentinel-1 can detect flooded grasslands, but signal saturation over flooded forests 2. PALSAR-2 can detect flooded forests, but limited data access; low repetition rates (44 d) 3. PALSAR-2 backscatter sensitive to trees up to 10 m, but signal saturation for tall trees Next steps: Sentinel-1 time series for wetland dynamic mapping PALSAR-2 ScanSAR time series for forest inundation mapping Interferometry (TanDEM-X and PALSAR-2) for vegetation height estimation CHARACTERIZATION OF AMAZON FLOODPLAIN FOREST AND ITS INUNDATION DYNAMICS USING PALSAR-2 AND SENTINEL-1 Mikhail Urbazaev 1 , Laura Hess 2 , Antje Uhde 1 , Felix Cremer 1 , Christiane Schmullius 1 1 Friedrich-Schiller-University Jena, 2 University of Califronia Santa Barbara In this study we present first results on the potential of ALOS-2 PALSAR-2 and Sentinel-1 data to support the Earth observation based assessment of 1) land cover and inundation dynamics mapping, and 2) vegetation structure mapping, in floodplain forests of the Brazilian Amazon. The first results have been produced in the frame of the currently funded BONDS project. The BONDS project (Balancing biOdiversity conservatioN with Development in Amazon wetlandS; a six-country consortium funded 2019-2022 by Belmont BioDiversa) seeaks to understand how ecosystem structure and function variables (vegetation cover, inundation period, hydrologic connectivity, water quality), fisheries management regimes, and projected climate change affect biodiversity and provisioning services of floodplain forests, fish, and phytoplankton. PALSAR-1 based inundation map for Juruá site The method will be adopted to PALSAR-2 ScanSAR data with higher repetition rate Curuai (S1 VV RGB) Janauacá (S1 VV RGB) Laeticia (S1 VV RGB) Juruá All RGB composites are based on multi-temporal metrics (R: percentile 95, G: median, B: percentile 5) S1 HH RGB PALSAR-2 HH RGB Dry season Wet season