 |
| Titel |
Asymmetric slip of normal faults resulting from a series of dike intrusions constrained by InSAR (Afar, Ethiopia) |
| VerfasserIn |
R. Grandin, A. Socquet, E. Jacques, G. King, A. Nercessian, S. Tait, A. Lemarchand, A. Delorme |
| Konferenz |
EGU General Assembly 2009
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 11 (2009) |
| Datensatznummer |
250024610
|
|
|
|
|
|
| Zusammenfassung |
| An exceptional rifting episode is currently under way in Afar (Ethiopia). Available geodetic
data covers not only the main rifting event of September 2005, but also a series
of at least 10 smaller events in 2006-2008. These are characterised by rapid dike
intrusion/inflation at depth (within hours), and motion on two sets of conjugate normal fault
systems above the dike, inducing local subsidence of the ground surface. Geodetic
data show that the main dike intrusion was ~60Â km long, and induced an ~3-6Â m
offset of the surface due to slip on two roughly parallel fault zones that bound a
~2-4Â km wide unfaulted zone. For later events, surface ruptures are observed to
occur over a length of ~10-20Â km along-strike, with slip reaching ~1Â m in a few
cases.
We identify three important problems from the study of these faults. These issues seem to
discard a series of hypotheses that constitute the basis of previous models of dike-induced
faulting. (1) InSAR data encompassing each rifting event allows to retrieve the horizontal (H)
and vertical (V) component of surface offset across each of the two conjugate fault systems,
assuming deformation is cylindrical. The H/V ratios can be used to infer an “equivalent” fault
dip at depth assuming pure dip-slip behaviour. The resulting calculation suggests dip values
in the range of ~30-50-, much shallower than values of ~60-80- deduced from field
observations and fault mechanisms in similar tectonic contexts. (2) A systematic
asymmetry of fault motion occurs on each side of the subsiding graben floor. We
show that while vertical throws seem to be approximately balanced, horizontal
dilation is significantly higher on one side of the rift than on the other. (3) Seismicity
accompanying dike emplacement is always low with regard to the amount of static
deformation detected by geodetic techniques: the maximum magnitude reaches
only Mw=5.6 for the main event, and Mw=4.5-4.8 for the largest subsequent rifting
events.
We performed 2D elastic modeling using a Boundary Elements Method in order to test a
series of geometric and kinematic configuration of the dike and overlying faults.
Combining these observations with fault mapping and topographic analysis, we
suggest that we are assisting to the birth of a new set of normal faults in response to
the emplacement of dikes at depth. This hypothesis is based on previous models
that have suggested that the early stage of fault development is characterised by
the formation of Mode I cracks, inducing a dilatancy of the incipient shear zone,
prior to the initiation of a through-going rupture surface. We further suggest that, in
the case studied here, the presence of a preferred direction of weakness resulting
from the pre-existence of a fault fabric may enhance the dilatancy effect. Also,
the importance of magmatism in assisting this process should not be forgotten, as
well as the low confining pressure within the depth range considered in this study. |
|
|
|
|
|
|