In this study scientists investigate the properties of metrics derived from the availability of repeated acquisitions of Envisat Advanced SAR (ASAR) data, acquired in the moderate resolution (150 m) Wide Swath Mode (WSM), to demonstrate their usefulness in the context of land surface characterisation and assess the potential of moderate resolution SAR data to complement classification of water bodies in global land cover products with particular regard to the detection of open water bodies.
In this study scientists investigate the properties of metrics derived from the availability of repeated acquisitions of Envisat Advanced SAR (ASAR) data, acquired in the moderate resolution (150 m) Wide Swath Mode (WSM), to demonstrate their usefulness in the context of land surface characterisation and assess the potential of moderate resolution SAR data to complement classification of water bodies in global land cover products with particular regard to the detection of open water bodies.
Six study areas have been selected in Europe and Central Siberia to assess the consistency of the multi-temporal metrics and the robustness of a new water body detection approach and to come up with a mapping algorithm, capable to withstand effects of seasonal conditions on the SAR backscatter and different typologies of land cover and water bodies, terrain conditions and seasonality.
Temporal variability of the SAR backscatter (TV) and the minimum SAR backscatter (MB) in a time series of measurements spanning one year were used. The behaviour of the two SAR multi-temporal metrics was consistent at the six study areas and a simple thresholding algorithm in the feature space of TV and MB allowed an overall classification accuracy above 90% when restricting to pure water pixels; the accuracy decreased linearly when considering pixels with a water fraction. For a threshold of 50% of water fraction, the overall accuracy was approximately 80%.