Secondary earthquake fractures generated by a strike-slip fault in the South Iceland seismic zone

This task was included in the PRENLAB-1 workprogramme and was not originally included within the PRENLAB-2 workprogramme. The latest developments of this topic, however, lead us to include the main results in the present report. Most earthquakes in the South Iceland seismic zone occur on N-S trending dextral strike-slip faults. The resulting rupture zones display complex en-echelon patterns of secondary structures including NNE-trending arrays of (mostly) NE-trending open fractures (O.F.) and hillocks.

  

Figure 26: Open fractures (O.F.) angles predicted from the combined effect of the main fault rupture and secondary fault (S.F.) rupture along the strike direction. The observed range and relative frequency of O.F. angles is shown shaded (from Bjarnason et al. 1993). If the friction coefficient f varies between 0.2 and 1.5 predicted O.F. angles vary within the circular sectors contoured in black. Solid lines in panel (a) represent the angle expected for O.F. belonging to dextral arrays. Dashed lines in panel (b) represents the angle expected for O.F. belonging to sinistral arrays. In both panels, longer lines indicate the predicted mixed-mode O.F. trends, while shorter lines indicate pure tensile trends. Mixed-mode O.F. are assumed to share the same style of faulting (sinistral or dextral) with the array to which they belong. The dotted lines indicate $22.5^\circ , 45^\circ , 67.5^\circ$ directions (for reference). Coloured lines refer to particular friction values (indicated). Friction increases as indicated by the red arrow.

Three spatial scales characterize the surface faulting pattern: the length of the main fault (M.F. $\sim$104 m), the arrays here interpreted as surface evidence of secondary faults ($\sim$102 m) and the individual O.F.($\sim$10 m). In order to improve our understanding of the genetic relationship between the O.F. and the M.F. we computed the stress field induced by slip on a buried M.F. using a dislocation model in a layered half-space: the fault surface is assumed to be embedded in the basement rock, topped by a softer near-surface layer. The O.F. were preliminarily considered as pure mode-I cracks opening in the near surface layer in the direction of the maximum (tensile) principal stress. Alternatively, secondary fractures were interpreted, as mixed-mode cracks, slipping at depth as shear cracks and opening near the surface due to low confining pressure. The Coulomb failure function after the earthquake (obtained summing the M.F. stress change and the lithostatic stress) suggests that secondary faulting (S.F.) can be expected to occur in response to the main rupture below the upper soft layer down to few hundreds of meter depth. The total stress change induced by the M.F. and the S.F. (of smaller scale) is shown to yield quantitative explanations of the complex geometry observed in the fault region in terms of simple frictional laws and friction coefficients very close to those measured in the lab (Figure 26).