Rémy Dominique, Falvey M., Bonvalot Sylvain, Chlieh Mohamed, Gabalda Germinal, Froger J.L., Legrand D. (2011). Variability of atmospheric precipitable water in northern Chile : impacts on interpretation of InSAR data for earthquake modeling. Journal of South American Earth Sciences, 31 (2-3), p. 214-226. ISSN 0895-9811.
Titre du document
Variability of atmospheric precipitable water in northern Chile : impacts on interpretation of InSAR data for earthquake modeling
Rémy Dominique, Falvey M., Bonvalot Sylvain, Chlieh Mohamed, Gabalda Germinal, Froger J.L., Legrand D.
Source
Journal of South American Earth Sciences, 2011,
31 (2-3), p. 214-226 ISSN 0895-9811
The use of Synthetic Aperture Radar interferometry (InSAR) in northern Chile, one of the most seismically active regions in the world, is of great importance. InSAR enables geodesists not only to accurately measure Earth's motions but also to improve fault slip map resolution and our knowledge of the time evolution of the earthquake cycle processes. Fault slip mapping is critical to better understand the mechanical behavior of seismogenic zones and has fundamental implications for assessing hazards associated with megathrust earthquakes. However, numerous sources of errors can significantly affect the accuracy of the geophysical parameters deduced by InSAR. Among them, atmospheric phase delays caused by changes in the distribution of water vapor can lead to biased model parameter estimates and/or to difficulties in interpreting deformation events captured with InSAR. The hyper-arid climate of northern Chile might suggest that differential delays are of a minor importance for the application of InSAR techniques. Based on GPS, Moderate Resolution Imaging Spectroradiometer (MODIS) data our analysis shows that differential phase delays have typical amplitudes of about 20 mm and may exceptionally exceed 100 mm and then may impact the inferences of fault slip for even a Mw 8 earthquakes at 10% level. In this work, procedures for mitigating atmospheric effects in InSAR data using simultaneous MODIS time series are evaluated. We show that atmospheric filtering combined with stacking methods are particularly well suited to minimize atmospheric contamination in InSAR imaging and significantly reduce the impact of atmospheric delay on the determination of fundamental earthquake parameters.
