You must have a javascript-enabled browser and javacript and stylesheets must be enabled to use some of the functions on this site.


Ground deformation analysis in Paris using a multi-interferogram Least-Squares approach

Stephane Le Mouélic(1) and Benoît Deffontaines(2)

(1) BRGM, 117 Av de Luminy, BP167, 13276 Marseille, France
(2) Ecole Pratique des Hautes Etudes, 15 Bd de la mer, 35800 Dinard, France


The ERS satellites have provided an unprecedented archive over Europe since 1992. This archive is a valuable tool to investigate any issue related to ground stability (or unstability), both from natural or anthropogenic origins. The cross comparison of single interferograms can reveal large deformation events with a centimetre scale amplitude. However, it is hampered by the ambiguity with atmospheric artefacts and by coherence loss. The combination of a comprehensive set of interferometric combinations can be used to perform a more efficient detection and quantification of the most subtle deformation features. In this presentation, we will show the results of a least-squares approach performed on a set of 30 ERS images covering the period from 1992 to 2000. From these 30 images, we compute automatically the whole set of interferograms having a baseline lower than 200 m. Orbital and topographic fringes are removed automatically using a recursive analysis based on orbital parameters and FFT computation. An adaptive filtering has been used to increase the signal to noise ratio of the interferograms. We then derived a mean velocity map of Paris during the 1992-2000 period by using a least-squares adjustment of 90 unwrapped interferograms. Using a high number of interferograms allows a subsequent reduction of the atmospheric noise level by introducing an averaging effect. It also improves the mean coherence level over sparsely vegetated areas. Another Least-squares adjustment is used on raw interferograms in order to retrieve the deformation screen for each date, following the approach proposed by Usai et al. (1999, 2003). We then use a temporal filtering on a pixel by pixel basis to reduce the atmospheric noise level of the chronologically ordered sequence of phase screens. As a result, deformation time series can be obtained, highlighting the dynamic evolution of the most interesting areas. We will also show that an automated detection of time non-linear deformation events can be obtained by computing the standard deviation of each phase screen previously corrected for a synthetic phase component corresponding to the average deformation rate. . This test shows that the only wide scale non-linear event on Paris is found in the St-Lazare area (and to a lesser extent in the Grand Palais area). A specific study of the St-Lazare deformation event and the cross comparison with independent data sets such as piezometric levels and optical levelling will be the subject of another presentation (Fruneau et al., this issue). The Least Squares approach reveals unambiguous ground deformation events as low as 1 mm/yr (for example around the Butte Montmartre, were unstable areas are observed on top of abandoned underground gypsum quarries). Finally, a comparison of the results issued from both the Least-squares approach and the Permanent Scatterers approach (80 ERS-1/2 images, time span 1992-2000) will be performed in order to evaluate their consistency. The Least-squares approach could be used on densely urbanised areas to investigate the overall ground stability. Conversely, when a high spatial resolution is required (e.g. to monitor two opposite sides of a large building), or whenever the area to be investigated is densely vegetated (but with a PS density still greater than 5 PS/km2), the Permanent Scatterers provides a valuable alternative, since it does not foresee any spatial low-pass filtering of the interferograms.


Workshop presentation

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry