Characteristics of snow and ice morphological features derived from multi-polarization TerraSAR-X data
Dana Floricioiu(1), H. Rott(2), T. Nagler(2), M. Heidinger(2) and M. Eineder(2)
(1) DLR-IMF, -, Oberpfaffenhofen, Germany
(2) ENVEO IT, -, Innsbruck, Austria
SAR data are an important source of information for studies of glacier and ice sheet mass balance and glacier/climate interactions. The snow accumulation on glaciers and ice sheets is a key parameter of the mass balance of high spatial variability which so far cannot be measured reliably. For the inversion of SAR data in respect to snow and properties, and for the use of these data for mass balance investigations, it is necessary to understand the relations between backscattering coefficients and the properties of the diagenetic glacier facies: dry-snow, percolation, superimposed ice, wet snow and ice.
Backscattering signatures were studied for the main facies types at several locations in Dronning Maud Land, Antarctica, on alpine glaciers (Stubai and Ötztal Alps, Austria) and in north-east Greenland. The recent SAR database includes high resolution (6.6 m) dual-pol stripmap TerraSAR-X data at HH, VV and HV polarization acquired at various incidence angles over these sites.
The Antarctic study sites are located in the dry snow zone and percolation zone. Detailed descriptions of the physical properties of snow and firn and of snow accumulation rates are available together with X-band ground-based scatterometter measurements from former campaigns. This is a good basis for the interpretation of the TerraSAR-X signatures. Backscatter analysis is also carried out along profiles at the transition between the two zones and compared with backscatter data from other satellite sensors: C-band data of Envisat ASAR and ERS AMI, and Ku-band scatterometer measurements from QuikSCAT. The study sites in north-east Greenland are located in the dry snow zone. Data on accumulation rates and vertical snow stratification are available from snow pits and ice cores. For the alpine glaciers the focus of the TerraSAR-X signature analysis is on the percolation and ice facies. The TerraSAR-X backscattering data are compared to X-SAR data at VV polarization from the shuttle campaigns SRL-1 and -2 in 1994.
Backscatter model calculations were carried out in order to assist in the interpretation of the TerraSAR-X signatures and to study the sensitivity of backscattering to snow and ice properties. A dual layer model based on the second order dense medium radiative transfer approach was applied taking into account situations corresponding to winter snow on the percolation facies and on ice. The sensitivity of X-band co- and cross-polarized backscattering to winter snow accumulation was investigated. The X-band backscatter signatures enable clear discrimination of the main facies types and provide information on snow morphology. This is relevant support information for estimating snow accumulation on glaciers and ice sheets. Previous polarimetric SAR data such as SIR-C/X-SAR already pointed out the relevance of polarimetric parameters for glacier facies classification. The multipolarization capabilities of TerraSAR-X open up new opportunities for mapping and monitoring the physical properties of snow and firn on glaciers and ice sheets.