HR: 0800h
AN: H51E-08    [Abstracts]
TI: Spatial and Temporal Analysis of Mexico City Subsidence by Means of Interferometric Techniques
AU: * Lopez-Quiroz, P
EM: lquiroz@enst.fr
AF: TSI Département, ENST, 46 rue Barrault, Paris Cedex 13, F-75634, France, Metropolitan
AU: Tupin, F
EM: florence.tupin@enst.fr
AF: TSI Département, ENST, 46 rue Barrault, Paris Cedex 13, F-75634, France, Metropolitan
AU: Briole, P
EM: briole@ipgp.jussieu.fr
AF: Laboratoire de Géologie, ENS, CNRS, 24 rue Lhomond, Paris, 75005, France, Metropolitan
AU: Doin, M
EM: doin@mailhost.geologie.ens.fr
AF: Laboratoire de Géologie, ENS, CNRS, 24 rue Lhomond, Paris, 75005, France, Metropolitan
AU: Nicolas, J
EM: nicolas@enst.fr
AF: TSI Département, ENST, 46 rue Barrault, Paris Cedex 13, F-75634, France, Metropolitan
AB: In Mexico city, water over-consumption leads to subsidence. Before the Spanish conquest, the southern part of the Mexico Valley, an endoreic basin surrounded by mountains, was filled by a large lake. Flooding problems oblige conquerors to dry the lakes, which by now have almost completely disappeared and have been replaced by buildings. The simplified hydrogeologic structure of Mexico Valley includes a superficial 50 to 300 m thick lacustrine aquitard overlying a thicker aquifer made of alluvial deposits. The aquitard layer plays a crucial role in the subsidence process due to the very high compressibility of its clay deposits separated by a less compressible sand layer where the biggest buildings are anchored. The aquifer over-exploitation leads to a depression of its piezometric level, inducing water downwards flow in the clays, yielding compaction and subsidence (Rivera, 1990). In order to quantitatively link subsidence to water pumping, the Mexico city subsidence needs to be mapped and analyzed through space and time. It will help identify possible variations related with seasonal recharge, anchored and non anchored buildings, old and new pumping areas with varying clay compressibility through time due to consolidation (Rivera, 1990, Ortega-Guerrero et al., 1999). Radar interferometry (InSAR, Interferometric Synthetic Aperture Radar) has been successfully applied to map subsidence caused by water pumping (e. g., Amelung et al., 2000). It uses two repeated SAR acquisitions to obtain distance measurements. After geometrical corrections, the interferometric phase contains deformation information as well as residual orbital and topographic errors and atmospheric delays. A previous work using levelling, interferometry and GPS techniques over Mexico city showed that the location of the maximum subsidence rates (about 400 mm/yr) has changed and moved from the downtown area to the east of the city over a 50 years interval (Cabral-Cano et al., 2006). In this previous study, no displacement time series could be obtained from interferometry due to difficulties in the unwrapping process. We present a work based on interferometry to measure subsidence evolution spatially and temporally using the whole 54 ERS1&2/ENVISAT images set and covering 11 years (1995-2006) of ground motion. The two main obstacles are temporal decorrelation and phase unwrapping. We test a 'traditional' interferometric method (Cavalié et al., 2006) and a Persistent Scatterer method (Hooper, 2006). To maximize coherence and facilitate unwrapping, the first method uses high coherent interferograms issued from short time span image pairs. Interferograms are corrected from a layered atmospheric phase screen and from residual orbital and topographical errors. We also derive a method to unwrap 6 months to one year interferograms. Corrected interferograms are inverted to obtain deformation time series and mitigate atmospheric artifacts. The second method uses all available images to construct interferograms with respect to a common master. It performs a correction of geometrical effects and uses amplitude and phase to select pixels not affected by decorrelation, thus carrying reliable phase information. We test the 3-D (space-time) unwrapping algorithm of Hooper to recover radar propagation delays and finally separate subsidence from atmospheric artifacts. Time series obtained from the application of both methods are analyzed, compared and discussed.
DE: 6924 Interferometry (1207, 1209, 1242)
SC: Hydrology [H]
MN: 2007 Joint Assembly