Impact of multi–source DEM accuracy on integrated InSAR time–series processing for 3D deformation decomposition in the Pasuruan fault zone
摘要
The main challenge in analyzing fault activity using geodetic methods is estimating accurate 3D deformation components. This study integrates multiple geodetic datasets including LOS InSAR, azimuth offset tracking, and GPS Ina–CORS observations to generate reliable horizontal and vertical displacement fields. To ensure accurate LOS InSAR deformation estimation, we evaluate the suitability of three DEMs (SRTM–1, Copernicus, DEMNAS) for topographic phase correction and assess their vertical accuracy using field based RTK measurements. Additionally, PS–InSAR and SBAS are combined to achieve high pixel density and temporal coherence in deformation mapping. The methodological innovation of this study lies in the development of an extended 3D decomposition model that, unlike existing approaches, jointly incorporates PS–InSAR, SBAS, azimuth offset tracking, and GPS horizontal velocity components into a unified matrix solution. This study adopts an extended 3D decomposition that jointly incorporates PS–InSAR, SBAS, GPS horizontal velocities, and where quality permits azimuth offset tracking as a low–weight auxiliary constraint. Given the limited azimuthal sensitivity of Sentinel–1 for interseismic signals, azimuth offsets are down–weighted in the WLS inversion and evaluated via ablation to ensure they do not degrade the north–south component. By applying this framework to the Pasuruan Fault Zone in East Java, Indonesia, the resulting 3D deformation fields (