Delineation of Woody Cover by using Full and Dual Polarimetric L- and C-Band Datasets in the Kruger National Park, SA
Urbazaev, Mikhail1; Schmullius, Christiane1; Thiel, Christian1; Mathieu, Renaud2; Naidoo, Laven2; Levick, Shaun3; Smit, Izak4; Asner, Gregory5; Leblon, Brigitte6
1Friedrich-Schiller-University Jena, GERMANY; 2Council for Scientific and Industrial Research, Pretoria, SOUTH AFRICA; 3Max Planck Institute for Biogeochemistry, Jena, GERMANY; 4South African National Parks, Skukuza, SOUTH AFRICA; 5Carnegie Institution of Science, Stanford, UNITED STATES; 6University of New Brunswick, CANADA
The information on spatial distribution of woody vegetation is a crucial key to understand the spatio-temporal dynamics of savanna ecosystems. The woody vegetation affects important ecosystem processes, i.a. carbon and water cycles, energy fluxes and fire regimes. In addition, represents the woody cover an important basis for the management decisions and conservation in savanna.
In this study the mapping of fractional woody cover by using full, dual and single polarimetric L- (ALOS PALSAR) and C-Band (RADARSAT-2) imageries is investigated. From the both sensors different SAR-parameters, e.g. multipolarized intensities, polarimetric decompositions and interferometric coherence, are extracted and compared with woody cover obtained from the 1-m airborne LiDAR data. The SAR-data were acquired at different seasonal savanna cycles between 2007 and 2010. The LiDAR survey took place in April-May 2008 by CAO LiDAR sensor (Carnegie Airborne Observatory, USA) with an operational frequency of 50 kHz.
The use of fully polarimetric multifrequency SAR data should help on better understanding of scattering mechanisms from various vegetation types (shrubs, trees). An overall aim of the study is the analysis of capabilities and limitations of SAR data for woody cover mapping and the investigation of the potential synergic use of LiDAR and SAR systems. Furthermore it should be studied the influence of seasonality for mapping the woody vegetation. Finally the results of the study should provide a basis to develop an operational monitoring program based on SAR data over the whole area of the Kruger National Park.