In-DMU: Modeling Uncertainty in Interferometric SAR-Based Deformation Monitoring

Quantifying the precision of Interferometric Synthetic Aperture Radar (InSAR) deformation monitoring is a fundamental aspect of ensuring its reliability in operational applications. Traditional uncertainty estimation methods, which are based on phase noise and error propagation laws, tend to overestimate monitoring precision. In this study, we systematically investigate the combined effects of coherence, deformation magnitude, SAR wavelength, and spatial resolution on monitoring uncertainty through multi-parameter-controlled experiments. Based on this, a wavelength-coherence integrated function model, referred to as the InSAR Deformation Monitoring Uncertainty model (In-DMU), was developed to estimate the precision of the deformation measurements. To rigorously define applicability of the model, a Pettitt test based on sliding window standard deviation (Pt-SWStd) is used to detect change points in the error distribution, thus establishing critical gradient thresholds of 6.433 mm/m (L-band), 1.407 mm/m (C-band), and 0.456 mm/m (X-band). Beyond these thresholds, phase unwrapping constraints lead to the retrieval breakdown regime, limiting reliable deformation estimation. Furthermore, the influence of multi-looking processing on the In-DMU model was systematically quantified. To assess practical applicability, the In-DMU model was validated across diverse observational scenarios using data from five representative SAR satellites: ALOS-1 (L-band), ALOS-2 (L-band), Radarsat-2 (C-band), Sentinel-1 (C-band), and TerraSAR-X (X-band). The In-DMU model provides a universal tool for priori estimation of InSAR monitoring precision, offering valuable guidance for SAR data selection and optimization of multi-looking configurations.