DEM-Assisted 3-D Reconstruction of Aletsch Glacier Displacements Using Monostatic and Bistatic Differential Interferometry

Abstract

Differential interferometry is a powerful remote sensing technique that allows radar systems to capture very small displacements from a far observational position. Most systems are monostatic, meaning that they can only capture displacement changes in the line of sight (LOS). The simplicity of monostatic systems also restricts their suitability to either displacements with a single principal motion direction or with a known direction in space. Adding a passive receiver (bistatic systems) provides an additional LOS of information, allows to reconstruct more complex vector fields, and reduces the assumptions on the displacement direction. This study presents a set of differential interferometric observations retrieved in the Aletsch glacier during a winter campaign in March 2022, using Ku-band Advanced Polarimetric Radar Interferometer (KAPRI), a Ku-band ground-based radar interferometer. We investigate the glacier motion based on the datasets corresponding to each of KAPRI's configurations and propose two independent approaches to solve the displacement field: monostatic (1 LOS) and monostatic-bistatic (2 LOS). As the monostatic and bistatic datasets were retrieved simultaneously, it is possible to do a direct comparison of the resulting displacements and the respective uncertainties. Results show that the accumulated displacement for the 09:01:30 h to 10:13:30 h at the near-range region (ROI 1) is 3.83mm in the monostatic case and 4.42mm in the monostatic-bistatic case. In the far-range region (ROI 3), the monostatic displacements result in 8.54mm while according to the bistatic approach 17.82mm. Assumptions used for the monostatic resolution of the glacier displacements hold for the flatter areas corresponding to the main glacier flow but fail in the steep terrain. Monostatic acquisitions present a high signal-to-noise ratio (SNR), which maintains a low uncertainty [below 0.5mm for all regions of interest (ROIs)] throughout the scene compared to the bistatic dataset. Monostatic-bistatic derived displacements are susceptible to the increased noise in the far range but yield a more robust (direction and magnitude) and smoother displacement field in the near-mid range. These comparisons provide valuable insights on the advantages and shortcomings of employing monostatic and bistatic configurations for 3-D displacement retrieval and serve as proxy for upcoming satellite missions that plan to use bistatic radar configurations.