Detection and mitigation of ionospheric artifacts in the azimuth ground displacements through the SAOCOM-1 L-band SAR data exploitation

Abstract

In a scenario where an increasing number of L-band Synthetic Aperture Radar (SAR) satellite systems is expected to be launched, such as the NISAR (NASA-ISRO), PALSAR-3 (JAXA) and ROSE-L (ESA) sensors, in addition to the already operative PALSAR-2 (JAXA) and SAOCOM-1 (CONAE) systems, an important challenge is represented by the development of innovative techniques to mitigate ionospheric effects on the generated SAR images and derived products. Indeed, widely used methods for ground displacement analysis, as for instance the Differential SAR Interferometry (DInSAR), the Multi Aperture Interferometry (MAI) and the Pixel Offset Tracking (POT) techniques, can suffer for the presence of such ionospheric effects, which can have a major impact on both the phase and the amplitude of the L-band SAR data. In this regard, it is well known that the propagation delay of the microwave signal induced by the variation of the Total Electron Content (TEC) in the ionosphere is inversely proportional to the frequency of the transmitted signal, in other words, the low-frequency signals experience more delay than the higher-frequency ones.In the recent years, several methods have been proposed to estimate and mitigate ionospheric effects in the DInSAR measurements. Among them, we mention the Faraday rotation estimation, the azimuth shift and the range split-spectrum techniques. In particular, we underline that the range split-spectrum method will be exploited at system level as a solution for systematically correcting the ionospheric artifacts in the DInSAR products achieved through the NISAR mission. Indeed, the L-band NISAR sensor includes a 5-MHz sideband separated from the 20- or 40-MHz main band, allowing to mitigate the ionospheric and non-dispersive phase artifacts. However, it is also worth remembering that the mentioned range split-spectrum method fails if the ionospheric effects impact also the azimuth displacement component. The variations of the TEC along the azimuth direction causes the so-called “azimuth streaks” that reveals itself as an offset in the pixel azimuth position, thus affecting the SAR co-registration procedure and, consequently, the surface displacement measurements.The aim of this work is firstly to investigate, in presence of ionospheric effects, the performance degradation of the MAI and POT techniques. Moreover, a solution technique is introduced that capitalizes on the large L-band data archives collected over the area of interest to effectively detect and mitigate the ionospheric artifacts. To this end, the StripMap L-band SAR images acquired by the SAOCOM-1 constellation are extensively exploited. In particular, results are presented based on the analyses carried out following the seismic events occurred on  February 2023 in South-East Türkiye near the border with Syria and the Litli-Hrútur volcano eruption in Iceland, which took place on  July 2023.