Characterizing and Correcting Phase Biases in Short-Term, Multilooked Interferograms

Abstract

Interferometric Synthetic Aperture Radar (InSAR) is widely used to measure deformation of the Earth’s surface over large areas and long time periods. A common strategy to overcome coherence loss in long-term interferograms is to use multiple multilooked shorter interferograms, which can cover the same time period but maintain coherence. However, it has recently been shown that using this strategy can introduce a bias (also referred to as “fading signals”) in the interferometric phase, particularly over vegetated areas. We isolate the signature of the phase bias by constructing daisy chain sums of short-term interferograms covering identical 1-year time periods, but using interferograms of different time spans. This shows that the shorter interferograms are more affected by this phenomenon and that different ground cover types are affected differently. We, propose a method for correcting the phase bias, based on the assumption that the bias in an interferogram is linearly related to the sum of the bias in shorter interferograms spanning the same time. We tested the algorithm over a study area in western Turkey by comparing average velocities against results from a phase linking approach that has been shown to be rather insensitive to the phase bias. Our corrected velocities agree well with those from phase linking approach. Our approach can be applied to global compilations of short-term interferograms and offer the possibility of accurate long-term velocities without a requirement for coherence in long-term interferograms.

Nearly Three Centuries of Lava Flow Subsidence at Timanfaya, Lanzarote

“…For this area the difference between the 3 and 11 network average velocities was 6.2 mm/yr in the ascending and 5.1 mm/yr in the descending. This is comparable with corrections of up to 27.4 mm/yr in cropland areas in Turkey, attributed to phase bias (Maghsoudi et al., 2022).…”

“…For other land cover types, where there is good coherence as in Lanzarote, we present an option to reduce the impact of the phase bias by creating networks of longer period interferograms (averaging more than 1 month). This is an alternative to correcting for the bias empirically (Maghsoudi et al., 2022). This bias was not apparent in the previously published ERS InSAR data (González & Fernández, 2011) as longer period interferograms formed the basis of their analysis.…”

“…These fading signals have been attributed to the result of soil moisture (De Zan & Gomba, 2018) that causes a breakdown in the phase consistency in triplets of interferograms (De Zan et al., 2014) which occurs as a result of the multi‐looking and spatial filtering that is commonly implemented in interferogram processing strategies (Michaelides et al., 2019). It particularly affects short period interferograms (Ansari et al., 2021; Maghsoudi et al., 2022) of weeks to a few months, and whilst in any individual interferogram the error is small compared to other noise terms such as from the atmosphere, cumulatively it propagates from the network time series causing a systematic bias error that varies spatially. This bias affects both look directions in a similar fashion (usually a negative bias).…”

Use of InSAR data for measuring land subsidence induced by groundwater withdrawal and climate change in Ardabil Plain, Iran

“…(3) Although we try to remove the unwrapping errors in LiCSBAS, some unwrapping errors might still exist and cause some unreal signals such as the uplift signal; (4) It can be the effect of phase bias due to existence of short-interval interferograms. The size of the cumulative loop closure phases can be significantly reduced as the length of the short-interval interferograms in the loop increases 61,62 . This can lead to the observation of an additional uplift in the area.…”