Mitigating tropospheric and ionospheric uncertainties in InSAR Time Series Analysis: the 5.8Mw Earthquake in the Eastern Cordillera (Central Andes), Northwestern Argentina

Detecting, characterizing and monitoring ground deformation are relevant tasks for natural risk assessments. The Synthetic Aperture Radar-SAR Interferometry (InSAR) technique stands out as a widely applied method to survey ground movements due its ability to resolve small-magnitude displacements. However, uncertainties associated with atmospheric delays prevent the detection of very small deformation signals that may be overprinted by noise. Common corrections applied either to single interferograms or time series analysis do not completely mitigate the influence of delayed signals, but they allow to distinguish between causes of delay , i.e. tropospheric or ionospheric.

In this study, we explore options to minimize the impact of any delay signal in time series analysis by using subsets of available SAR scenes. We analyzed 5 years, from 2018 to 2022, of Sentinel 1 A/B data in ascending (Track 76) and descending (Track 10) orbits. The scenes cover the Eastern Cordillera in the Northwestern Argentina, the easternmost range of the Central Andes. For each date, atmospheric delays are estimated using modern processing techniques: (i) tropospheric delays are investigated from global atmospheric models and, (ii) ionospheric delays are estimated from Split Range-Spectrum technique. Then we carefully remove noisy scenes and perform InSAR time series analysis. We evaluate our method by comparing displacements from the 5.8Mw earthquake that occurred on 29-Nov-2020 with an epicenter near Quebrada de Humahuaca. The analysis is expanded to the time-motion history retrieved from landslides in this area, which also serves to study the relationship between displacements rates and the earthquake. Finally, we explore how the quality of InSAR pairs precipitates into coherence, errors of phase unwrapping, and estimation of topographic residuals. Our results suggest that image quality assessment and subsequent SAR-scene removal is an effective tool for improving the quality of the time series.