Seismological Constraints on Fault Plane Curvature

Abstract:

The down-dip geometry of seismically active normal faults is not well known. Many examples of normal faults with down-dip curvature exist, such as listric faults revealed in cross-section or in seismic reflection data, or the exposed domes of core complexes. However, it is not understood: (1) whether curved faults fail in earthquakes, and (2) if those faults have generated earthquakes, is the curvature a primary feature of the rupture or due to later modification of the plane? Even if an event is surface-rupturing, because of the limited depth-extent over which observations can be made, it is difficult to reliably constrain the change in dip with depth (if any) and therefore the fault curvature. Despite the uncertainty in seismogenic normal fault geometries, published slip inversions most commonly use planar fault models.

We investigate the seismological constraints on normal fault geometry using a forward-modelling approach and present a seismological technique for determining down-dip geometry. We demonstrate that complexity in the shape of teleseismic body waveforms may be used to investigate the presence of down-dip fault plane curvature. We have applied this method to a catalogue of continental and oceanic normal faulting events. Synthetic models demonstrate that the shapes of SH waveforms at along-strike stations are particularly sensitive to fault plane geometry. It is therefore important to consider the azimuthal station coverage before modelling an event. We find that none of the data require significant down-dip curvature, although the modelling results for some events remain ambiguous. In some cases we can constrain that the down-dip fault geometry is within 20° of planar.