| Summary: | Snow is a crucial element within Arctic ecosystems that is undergoing rapid change in the context of Arctic warming. High spatial resolution synthetic aperture radar (SAR) systems have shown high potential to characterize changes in snow properties, such as depth, density, water equivalent and microstructure. However, distinguishing between the contributions of snow, ground, vegetation and moisture to the SAR signal is challenging, in particular in remote environments with sparse reference data such as the Tundra. Our study focuses on a long-term monitoring site at Trail Valley Creek, Northwest Territories, Canada where ongoing snow science related to the Canadian Space Agency Terrestrial Snow Mass Mission has provided a reference snow dataset to evaluate. We completed a time series of TerraSAR-X (TSX) and TanDEM-X (TDX) VV-VH and HH-HV acquisitions between February 2018 and February 2019 in an 11 day repeat cycle. We used the Stripmap-mode of TDX that acquires continuous scene strips with a swath width of 15 km, variable strip length and a spatial resolution of up to 2.5 m. By spatial analysis of X-Band time series, snow depth from extensive repeated winter field campaigns, vegetation height from aerial LiDAR as well as soil moisture and ground temperature from in-situ observations we explain dynamics of X-Band backscatter for different tundra landscape units during winter.
|
|---|