Geophysical logging

In the preparatory phase of an earthquake, stress accumulation is expected to be connected with the creation of borehole breakouts (BOs), changes in the number and size of cracks, a possible variation of the stress direction, etc. Therefore it is very important to monitor the following set of geoparameters:

[$\bullet$] P-wave travel time. [$\bullet$] electrical conductivity. [$\bullet$] stress information from borehole breakouts (orientation and size). [$\bullet$] crack density.

In the framework of this project, we had the chance to carry out repeated logging to obtain a time series of logs in the South Iceland seismic zone (SISZ). An 1100 m deep borehole (LL-03, Nefsholt) inside the zone ( $63.92^{\circ}$N, $20.41^{\circ}$W, 7 km south of the seismic station SAU) was used and provided the unique opportunity to perform measurements much nearer to earthquake sources than usual - the hypocenter depths at that location range between 6 and 9 km. Moreover, data could be obtained for a depth interval of more than 1000 m, uninfluenced by the sedimentary cover and less disturbed by surface noise.

This was achieved by repeated logging with tools as: sonic log (BCS), gamma-ray (GR), spectral gamma-ray (SGR), neutron-neutron log, dual induction/latero log (DIL), 16"- and 64"-normal resistivity log, spontaneous potential log (SP), a borehole temperature log (BHT), four-arm-dipmeter (FED), and borehole-televiewer (BHTV).

The neutron-neutron log, the 16"- and 64"-resistivity log, the SP log, and temperature logs were run with the logging equipment of OS, the rest with the Halliburton logging truck of GFZ.

Investigations on the stress field in the SISZ

Besides the repetition of logs in borehole LL-03 (Nefsholt), we performed single logging campaigns at other boreholes to check the state of the regional stress field. This is important for two reasons:

$\bullet$

From the San Andreas fault we know (Zoback et al. 1987) that fault zones may be in a low stress state between earthquakes, which gets visible through stress orientations perpendicular and not pointing at an angle of 30$^\circ$ to 45$^\circ$ to the strike-slip fault. To determine the present state of stress in the SISZ, it is important to see if there are stress components, that are not perpendicular to existing faults and thus favour earthquakes on them.

$\bullet$

The SISZ is no typical transform zone. Looking at the orientation of rift opening and the adjacent rifts, one would expect a left-lateral strike-slip zone in NW-SE direction (N103$^\circ$E) to connect the Reykjanes ridge and the eastern volcanic zone (EVZ) of Iceland. Instead earthquakes occur on en-echelon N-S striking right-lateral faults. Assuming an angle of 45$^\circ$ between the maximum horizontal compressive stress and the fault (as it is done constructing fault plane solutions) both planes are equivalent. However, from a rock mechanics point of view, expecting an angle of about 30$^\circ$ between fault and maximum horizontal principal stress, the stress orientation at N-S striking faults should be N30$^\circ$E, compared to N60$^\circ$E at an E-W striking transform.