Rapid changes in the electrical state of the 1999 Izmit earthquake rupture zone

Crustal fluids exist near fault zones, but their relation to the processes that generate earthquakes, including slow-slip events, is unclear. Fault-zone fluids are characterized by low electrical resistivity. Here we investigate the time-dependent crustal resistivity in the rupture area of the 1999 M_w 7.6 Izmit earthquake using electromagnetic data acquired at four sites before and after the earthquake. Most estimates of apparent resistivity in the frequency range of 0.05 to 2.0 Hz show abrupt co-seismic decreases on the order of tens of per cent. Data acquired at two sites 1 month after the Izmit earthquake indicate that the resistivity had already returned to pre-seismic levels. We interpret such changes as the pressure-induced transition between isolated and interconnected fluids. Some data show pre-seismic changes and this suggests that the transition is associated with foreshocks and slow-slip events before large earthquakes.

Low-resistivity crustal fluids occur near fault zones, but their relation to earthquake generation is unclear. Here, electromagnetic data from the Izmit earthquake reflects the pressure-induced transition between isolated and interconnected fluids that is linked to foreshocks before large earthquakes.

Electrical conductivity images of active and fossil fault zones

We compare recent magnetotelluric investigations of four large fault systems: (i) the actively deforming, ocean-continent interplate San Andreas Fault (SAF); (ii) the actively deforming, continent-continent interplate Dead Sea Transform (DST); (iii) the currently inactive, trench-linked intraplate West Fault (WF) in northern Chile; and (iv) the Waterberg Fault/Omaruru Lineament (WF/OL) in Namibia, a fossilized intraplate shear zone formed during early Proterozoic continental collision. These fault zones show both similarities and marked differences in their electrical subsurface structure. The central segment of the SAF is characterized by a zone of high conductivity extending to a depth of several kilometres and attributed to fluids within a highly fractured damage zone. The WF exhibits a less pronounced but similar fault-zone conductor (FZC) that can be explained by meteoric waters entering the fault zone. The DST appears different as it shows a distinct lack of a FZC and seems to act primarily as an impermeable barrier to cross-fault fluid transport. Differences in the electrical structure of these faults within the upper crust may be linked to the degree of deformation localization within the fault zone. At the DST, with no observable fault-zone conductor, strain may have been localized for a considerable time span along a narrow, metre-scale damage zone with a sustained strength difference between the shear plane and the surrounding host rock. In the case of the SAF, a positive correlation of conductance and fault activity is observed, with more active fault segments associated with wider, deeper and more conductive fault-zone anomalies. Fault-zone conductors, however, do not uniquely identify specific architectural or hydrological units of a fault. A more comprehensive whole-fault picture for the brittle crust can be developed in combination with seismicity and structural information. Giving a window into lower-crustal shear zones, the fossil WF/OL in Namibia is imaged as a subvertical, 14 km-deep, 10 km-wide zone of high and anisotropic conductivity. The present level of exhumation suggests that the WF/OL penetrated the entire crust as a relatively narrow shear zone. Contrary to the fluid-driven conductivity anomalies of active faults, the anomaly here is attributed to graphitic enrichment along former shear planes. Once created, graphite is stable over very long time spans and thus fault/shear zones may remain conductive long after activity ceases.