Applications of bistatic Ku-band radar in snow-covered environments

Abstract

Bistatic radar imaging is a remote sensing method which employs a spatially separated radio wave transmitter and receiver, in order to construct a two-dimensional image of the reflective properties of objects within a certain area. Compared to the more common monostatic radar systems (which use a co-located transmitter and receiver), bistatic radar systems are considered to be specialized tools which are more suitable for certain specific purposes, at cost of higher complexity. The special-purpose character and higher complexity of bistatic systems cause a low availability of such systems, and thus also of bistatic radar datasets. This is an obstacle for performing studies which require the use of bistatic systems. In the Earth Observation domain, such studies may be aiming, e.g., to explore non-reciprocal scattering processes which do not occur in the monostatic regime, or to investigate phenomena with specific bistatic signatures. This dissertation makes use of KAPRI, a ground-based Ku-band polarimetric radar interferometer based on the Gamma Portable Radar Interferometer (GPRI). KAPRI was specifically modified by the manufacturer to allow full-polarimetric bistatic radar acquisitions, and can thus be used for studies of the bistatic scattering processes occurring at Ku-band in a variety of environments. The Ku-band frequency of KAPRI makes the study of glacial and snow-covered environments particularly attractive, due to the relatively short but non-zero penetration depth into snow and ice, and due to high interferometric sensitivity to small displacements. In the first part of this thesis, the bistatic operation mode of KAPRI is developed and validated. In the latter two parts, bistatic KAPRI is used to investigate the bistatic scattering properties of snow and ice-covered environments, using two different approaches. The first part of this thesis focuses on development of the bistatic operation mode of KAPRI, and the associated data processing and polarimetric calibration procedures. A bistatic signal model was developed, which accounts for the offset between the internal oscillators of the two devices forming the bistatic configuration. This offset was compensated through the use of a synchronization link which transmits part of the pulse directly between the two devices. Processing procedures which allow coregistration of bistatic and monostatic datasets were developed through analysis of the elliptical acquisition geometry. The challenge of bistatic polarimetric calibration was resolved through development of a custom active calibration device usable in arbitrary geometries. The associated novel calibration method was compared with the established monostatic procedure, thus validating the novel method for bistatic use. The second investigation employs KAPRI to study the bistatic scattering properties of snow cover on top of the Great Aletsch Glacier in Switzerland. Two multi-modal time series datasets encompassing full-polarimetric, interferometric, monostatic and bistatic acquisitions were acquired, one in August 2021, and one in March 2022. Analysis of the data revealed considerable differences in polarimetric scattering between the two seasons, caused by the yearly cycle of changing structure of the snow cover. Particular attention was given to polarimetric phase differences, which exhibit a diametrally different response between the two seasons. The results indicate that the co-polar phase difference (CPD) exhibits a smooth, predictable spatial behaviour in summer when the snow cover is firn-like. In winter it exhibits rapid variation and phase-wrapping, thus complicating the use of CPD inversion methods to retrieve snow property information. Analysis of bistatic polarimetric data also revealed the presence of non-reciprocal scattering processes, which manifested itself in the non-zero value of the phase difference between the two cross-polarized polarimetric channels, HV and VH. The temporal coherence of the scene was analyzed and revealed the decorrelation timescale of the snow cover to be between 4-12 hours. This constrains the maximal allowable revisit time for repeat-pass radar methods at Ku-band. The third investigation of this theses focuses on a specific phenomenon, the coherent backscatter opposition effect (CBOE). We performed the first full bistatic characterization of this effect in the Earth’s cryosphere with a terrestrial sensor (KAPRI) at Ku-band, and with a spaceborne sensor (TanDEM-X) at X-band. The results revealed that the CBOE occurs in terrestrial snow at radio wavelengths, and is detectable at Ku-band in relatively thin seasonal snow layers with thickness of several meters. At X-band the effect was detected in deep firn areas of the Great Aletsch Glacier, indicating the need for a thicker snow layer in order to detect the effect at X-band. Through application of a CBOE scattering model, we were able to relate the angular width and height of the observed enhancement peaks to the scattering and absorption mean free paths of the radio waves within the snow layer. This showcased a possible pathway towards snow parameter estimation through bistatic radar observations of the CBOE.