An Integrated GNSS-SAR Approach to Improve Ground Deformation Analysis in the Field of Geo-energies: activities planned within the SMILE project

The SMILE network, part of the Marie Sklodowska-Curie doctoral network funded by the Horizon Europe (2021-2027) program, undertakes an innovative role in addressing geo-energy project challenges while promoting net-zero greenhouse gas emissions in line with global sustainability goals.

A segment of the SMILE network is dedicated to addressing the issue of ground deformation analysis. This research aims to develop a software tool that capitalizes on Global Navigation Satellite System (GNSS) data, coupled with Synthetic Aperture Radar (SAR) inputs and ground modeling information. This abstract outlines the methodological framework for the anticipated software development.

In the initial phase, we aim to integrate the high temporal resolution GNSS data with the high spatial resolution of SAR data. The goal of this process is to combine the advantages of both data types while minimizing their limitations. SAR provides extensive spatial detail but has limited temporal frequency and directional sensitivity. On the other hand, GNSS data provides comprehensive three-dimensional vectors with high temporal frequency, but spatially limited to the points where GNSS stations are located.

Merging the displacement measurements from GNSS and SAR requires temporal synchronization and the reconciliation of their different displacement vectors: GNSS captures vertical, east, and north components, and SAR measures in the line-of-sight direction. The optimal joint operation for this task is proposed through a Kalman filter. Due to the complexity of building a joint filter, the proposed method seeks to first analyze by considering the projected displacements only in the vertical direction. In this case, the measure in the line-of-sight of the SAR satellite will be projected in the vertical direction.

The next phase will focus on an innovative approach to enhance the covariance matrix within the Kalman filter algorithm. Instead of using a homogeneous and isotropic constant covariance in time, this enhancement strategy will harness observed data as input. Primarily, it will smooth the covariance evolution in time, exploiting past observations; this may improve the stability of the outcomes. The method under development proposes to improve the covariance modeling further by enabling the consideration of anisotropic and non-homogeneous scenarios. Finally, the proposed method aims to integrate the monitoring data into Thermo-Hydro-Mechanical (THM) modeling.

The proposed expansion is expected to bring significant advancements in ground deformation analysis, improving its resolution and precision. The tool will integrate GNSS and SAR datasets into a comprehensive ground deformation analysis suitable for geomatics applications within geo-energy projects.