Dynamics of Rifting in Afar – DoRA

There are only a few places in the world that offer the opportunity to study the dynamics of divergent plate boundaries on land. The Afar Depression is one of them. We propose two complementary volcano-tectonic rift studies: the Wal'is Dabbahu Rift (WD, Ethiopia) and the Asal-Ghoubbet Rift (AG, Djibouti) corresponding to the inland propagation of the Red Sea and of the Aden Ridge, respectively. In Northern Afar, the WD Rift is currently undergoing a major rifting episode. This event started in September 2005 with a seismic swarm that lasted 20 days. InSAR data revealed the injection of a 65 km-long mega-dyke that opened by up to 8 m, the slip of numerous normal faults and opening of fissures, and a rhyolitic eruption. Similarly, the AG Rift underwent a smaller episode of rifting in 1978 associated with the intrusion of a 2 m wide dyke into the crust. Since then, a large catalog of geodetic data that includes recent InSAR time series reveals the importance of non-steady deformation controlling the rift dynamics, deformation that is also shown in other volcanic areas in Iceland and Hawaii. Whereas deep, slow and aseismic fault slips are gaining the attention of a large number of scientists working mainly on subduction zones, we propose the study of similar transients in great detail in these active rifts, where the crustal layer is thin. Our goal is to gain an understanding of such volcano-tectonic segments on several time scales, including the dyking period itself and the post-event period. Essentially, the study of the behavior of the AG Rift during its whole post-rifting period offers an image at t+30 years of the WD segment, while keeping in mind important structural and scale differences. To gain this understanding, this proposal is based on 4 tasks. First, we propose the building of a complete and accurate set of geodetic data, including InSAR and field measurements (continuous GPS, GPS) covering the period under study. With a narrow temporal sample window, we plan to precisely describe the aseismic slip affecting the normal faults of these rifts and the periods of sudden slip and/or slip acceleration, but also to measure the deformation associated with probable future dyke intrusion in the WD Rift. Second, we aim to constrain the origin of these displacements and their relation to mass transfers within the crust. Series of gravity measurements will be initiated or pursued in both rifts. The combination of both gravity and geodetic data will help to discriminate density variations and vertical displacements. Third, the recording of seismic activity is essential to constrain the relative importance of seismic and aseismic deformation. This will also help to evaluate the thickness of the seismogenic layer. Together with structural data collected during a seismic survey in the AG Rift (complementary project), the results of these three initial tasks will offer crucial constraints on modeling to shed light on the rifting dynamics, which will correspond to a fourth subsequent task. We will then conduct three kinds of mechanical analysis of the observed displacements, taking into account the knowledge of the structure to test the relative influences of the rheology, the fault/dyke geometry and fluids on the rupture mechanics. With a thermo-mechanical approach, we investigate the respective role of the processes controlling the current 3D deformation field of these rifts, including viscous relaxation, dyke intrusion/inflation and aseismic slip. These models will be completed by Stress Coulomb analysis in order to examine the interactions between dyke intrusions within the elastic crust and the creation and development of surface faulting. Finally, we plan to compare the dynamics and morphology of the slip distribution profile along normal faults with rupture of small-scale transparent disordered materials and with numerical models of faults to investigate the role asperities on the fault dynamics. Our multidisciplinary approach should provide important new insights into the dynamics of rifting along divergent plate boundaries, and ultimately, into other geodynamical contexts affected by aseismic fault slip transients.