A Multiscale Method to Mitigate Topographically Correlated Delays and Orbital Errors for InSAR Measurements
Vajedian, Sanaz1; Motagh, Mahdi2; Nilfouroushan, Faramarz3
1Tehran University, IRAN, ISLAMIC REPUBLIC OF; 2GFZ, GERMANY; 3Uppsala University, SWEDEN
Interferometric Synthetic Aperture Radar (InSAR) has become a geodetic method of choice to assess ground deformation associated with seismic and volcano activity. InSAR observations suffer from error sources such as inaccurate orbital information and in particular from variation in the refractive index of the atmosphere, which can compromise the accuracy of repeat-pass interferometry.
Because of spatial variation in vertical stratification of the lower part of atmosphere (troposphere), interferometric phase component is often correlated with elevation, mainly in mountanous area. Previous studies have used a linear relationship between phase and topography to reduce atmospheric effects. This assumption is only valid for simplified cases , where only static tropospheric delay prevails. However, there are alway other factors in an InSAR measuement that can affect the phase, including deformation due to tectonic/non-tectonic processes and effects of orbital errors and atmospheric delay due to turbulent atmospheric circulation which affect the simplifed assumption of statif atmospheric delay.
In this study we apply a multiscale algorithm for mitigating topographically correlated delays and orbital errors in InSAR measurements in mountainous regions. We decompose both SAR interferograms and topography to independent components to discriminate between various components that contribute to interferometric phase. This band passed decompsition enables us to determine bands wherein there are good correlation between phase an topography, which we can use to determine static troposphere delay. Other factors including orbit and techtonic displacement components could be identified in other decomposed bands.We have tested our approach by analysis of the Envisat and ALOS radar data during the period 2003–2008 on Damavand volcano, Iran. A validation has been carried out by comparing the estimated time series with time series calculated from continous GPS stations in the region. Our InSAR results are in good agreement with GPS observations and indicate that atmospheric and orbital correction using our proposed method could imrove the accuracy of deformation measurements.