Using Seismicity to Map Fractures in Hengill, SW-Iceland

Abstract

The Hengill high-temperature geothermal field in SW-Iceland has been developed and exploited since 1990. A second power plant started production in 2006 about 10 km to the southeast of the first and together they produce 420 MW electricity and 430 MW of warm water. Hence, the Hengill geothermal area is of great importance for electricity and hot water production in Iceland and understanding the dynamics of the area is of great scientific and economical interest. Hengill is located on a tectonic triple junction in Southwest Iceland connecting the Reykjanes Volcanic Zone, the Western Volcanic Zone and the South Iceland Seismic Zone (SISZ). Between 1993 and 1999, the area was subject to a significant volcano-tectonic event which caused 8 cm uplift over a period of 4 years. The uplift induced more than 90 thousand earthquakes in the Hengill region, a very high number considering the uplift rate, which highlights the intensity of the stresses that must have been present in the crust prior to the onset of the uplift. It has been suggested that the uplift was caused by a magmatic event and that the regional stress is enhanced by the loading of the SISZ prior to two M 6.5 events in 2000, possibly even causing the uplift. With time the seismicity propagated south from the uplift-source region and into the neighboring the SISZ, where two events of ML more than 5 were generated in 1998 less than 10 km from the uplift center. We use micro-earthquakes to study fault activation in a ca. 5 x 6 km big area around the uplift centre. Locations of the earthquakes are improved significantly by the use cross-correlation to refine wave arrival times and application of a double-difference algorithm. The vast majority of seismic activity is scattered but twenty-five (25) alignments consisting of 10 to 363 earthquakes, 200 to 3000 m long, could be identified and are interpreted as faults. Most of these faults are sub-vertical (within 5° from vertical) and only two dip at 62°and 65°, respectively. The relocated earthquake distribution shows that the faults are activated in agreement with the regional stress field rather than respecting the uplift geometry, suggesting that we see re-activation of existing, sub-optimally oriented faults rather than the formation of new faults. This underlines the fact that the stress caused by the uplift is small in comparison to the regional stress and functions as a mere trigger for the seismicity.