Comparison of two methods in multi-temporal differential SAR interferometry: Application to the measurement of Mexico city subsidence

In multi-temporal InSAR processing, both the permanent scatterer (PS) and small baseline (SBAS) methods are optimized to obtain ground displacement rates with a nominal accuracy of millimetres per year. In this paper, we compare ground motion time series obtained by both approaches applied on 38 ENVISAT images from November 2002 to March 2007 to map the Mexico City subsidence. The first approach developed to extract reliable phase delay information is based on the selection of point targets, "permanent scatterers". The deformation rate is extracted from a stack of wrapped differential interferograms through a 2D linear regression. The interferograms are first unwrapped temporally at a local scale, then integrated in space. Synthetic tests allow gaining insights on possible error sources in the temporal unwrapping step. Lopez-Quiroz et al. (2009) developed a method based on Small Baseline interferograms, designed to solve the particular problems associated to Mexico City subsidence. They use a ground motion model obtained by adding some high quality interferograms in order to guide the phase unwrapping of the interferograms whose temporal baseline increases. A diagnostic for unwrapping errors is calculated from the closure of the interferometric system. Finally, the corrected and unwrapped interferograms are inversed to get the phase delay time series and to separate atmospheric phase screen from deformation. The subsidence rate maps obtained by both approaches are compared quantitatively and analysed at different steps of the PS approach. The difference is partly explained by errors in the spatial integration of subsidence rates in the PS technique and partly explained by the punctual differential subsidence between various human-made structures unseen by the SBAS approach.