Characterization of Eurasian Wetlands Using Microwave Remote Sensing for Ecosystem Carbon Flux Models

Wetland ecosystems are a dominant landscape feature of the northern high latitudes. Because of their effect on land-atmosphere carbon (CO2 and CH4) exchange, wetlands have a crucial role in the global carbon cycle and in the global climate system. Characterizing wetland biomes in terms of their extent and dynamics is extremely important to understanding the role of these ecosystems in the global climate. Microwave remote sensing is an effective geophysical tool for these purposes because it enables monitoring of large inaccessible areas on a temporally consistent basis regardless of atmospheric conditions or solar illumination. Here we employ multi-temporal high resolution (~100m) synthetic aperture radar (SAR) data from ALOS-PALSAR (L-band) to map wetland type and extent within sub-regions in Eurasia. We combine this information with time series inundated area estimates derived from AMSR-E, SSM/I, QuikScat and ASCAT to assess surface inundation patterns. We present details on the decision tree based classification approach used to generate the high resolution SAR based wetland maps as well as details of cross-product harmonization between fine and coarse resolution wetland/surface inundation products. Finally, we show the applicability of the wetland maps in an ecosystem carbon flux model. Portions of this work were carried out at the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.