Seismicity and structure of the 85°E volcanic complex at the ultraslow spreading Gakkel ridge from local earthquake tomography

• Published in: Geophysical Journal International
• Main Authors: Korger, Edith, Schlindwein, Vera
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: WILEY-BLACKWELL PUBLISHING 2013
• Subjects: Arctic Ocean; Sea ice
• Online Access: https://epic.awi.de/id/eprint/33347/; https://hdl.handle.net/10013/epic.42998

Accretion mechanisms at ultraslow spreading ridges are still only poorly understood due to difficult survey conditions for seismic experiments at these ridges. Melts gets focused in distinct magmatic centres, with thin crust in between. At the 85°E/85°N volcanic complex at Gakkel ridge, Arctic Ocean, where a recent spreading episode has been observed, perennial sea ice cover challenges traditional investigations of seismic structure and microseismicity. We used an unusual survey set-up to gather seismological data during 16 days in July 2007, using seismometer arrays mounted on ice floes. Despite only 12 stations, the drift of the ice floes over the survey area resulted in a multitude of crossing rays in the rift valley. The data included 303 microearthquakes of which 248 events could be confidently located. We compiled a 1D velocity model by localizing a subset of these earthquakes with a suite of randomly created velocity models. In this model, the Moho is placed at 7 km depth below seafloor, inferring a thick, basaltic crust. Using 124 events which were recorded by at least two arrays, we inverted for P-wave velocity structure in a local earthquake tomography. Resolution tests indicate reliable results in the central rift valley, illuminating the thermal structure underneath the Asgard volcanic chain in the aftermath of its most recent spreading episode. Our results show microearthquake activity down to 16 km beneath seafloor, inferring a cold lithosphere. Most hypocentres cluster at the centre of the rift valley at the site the Asgard volcanic chain. This may mean that existing thermal models for this class of ridges have to be refined. An area of decreased seismic velocities crosses the rift valley at this location and microearthquake activity is located at its eastern fringe where the velocity gradient is highest. We therefore speculate that the reduced velocities may be caused by warm intruded material and that the observed seismicity predominatly reflects the relaxation of thermal stresses following ...