par- 1er décembre 2016 ( dernière mise à jour : 23 octobre 2017 )
I defended my Ph.D. at ISTerre November 22nd, 2016.
For more informations about my current research, please visit my research webpage :
PhD Title : Le couplage intersismique et les glissements lents vus par la géodésie spatiale : Applications à la subduction Mexicaine et à la faille décrochante Nord Anatolienne.
Supervisors : Cécile Lasserre, François Renard, Andrea Walpersdorf
The development of space geodesy during the last decades has led to the discovery of the slip modes diversity on faults. In this thesis, we focus on aseismic slip on both subduc- tion zones and continental strike slip faults. We study the subduction zone of Mexico, from Guerrero to Oaxaca regions, where large Slow Slip Events have been recorded, as well as the Ismetpasa creeping fault segment of the North Anatolian Fault in Turkey. The map of interplate coupling (in between large slow slip events) estimated between the Cocos and North America plates in Mexico shows a relatively high coupling coefficient, laterally homo- geneous at the base of the seismogenic zone. Strong lateral coupling variations are notable within the seismogenic zone. A high coupling pattern in Oaxaca is located in the same area as the 1978 seismic rupture and has accumulated 5 cm of slip deficit in 9 years. Those lateral coupling variations are correlated with trench-coast distances. We show that critical state areas are located at the transitions between low and high coupling zones. These observa- tions suggest a persistency of the coupling patterns over geological time scales. We propose a mechanism of deformation during the coseismic phase, related to the existence of fric- tional transitions on the subduction plane, that allows to reconcile decadal observations of interseismic velocities with the long term building of the coastal morphology. The interseis- mic velocities are generally considered to be constant over a few years. However, refined temporal analysis show the richness of their temporal dynamics, with evidences of slow slip events of various magnitudes and durations. We follow two different approaches to detect and characterize small amplitude slow slip events in our two study areas. (i) The dense net- work of interferograms with short repeat times acquired by the CosmoSky-Med constella- tion along the Ismetpasa fault segment allows to detect a month-long transient slip event. During the 9 other months of the analysis, no clear aseismic slip signal has been observed. With a Mw 5.2 - 5.5, this transient event has released an energy equivalent to the one that would be released during 1.5 - 2 years at the average creep rate estimated by previous stu- dies. This discovery forces us to reconsider the mechanical model in place for this segment. (ii) The development of a correlation methodology between synthetic transient slip events and post-processed GPS time series allows to detect transient events with amplitude close to the GPS background noise. Applied to a synthetic dataset over the Mexican subduction zone, we show that this method is able to precisely detect and characterize Mw>6 events. Applied to real data between February 2005 and May 2007 in the Guerrero gap area, our new GPS-matched filter allows to detect 15 new events. Those events are temporally correlated with bursts of tremors and LFEs activity and are surrounding the Mw 7.5 2006 SSE. These de- tections enable to better characterise the slow slip events scaling law and shed a new light on the spatial interaction of slow slip events at the base of the seismogenic zone.
Dans la même rubrique :