Dehydration earthquakes in subducting slabs

Most earthquakes occur at relatively shallow depths (less than 50 km) because at greater depths rock becomes too warm and plastic to undergo brittle deformation.  Earthquakes at subduction zones extend to slightly greater depths, partly because the subducting oceanic lithosphere is relatively cool.  Even on subduction zones, however, earthquakes beyond 50 km depth are difficult to explain.

Earth Scientists from University College London are now suggesting that some of the deeper earthquakes that occur on both the top and bottom surfaces of subducting slabs, are a result of the rapid release of water from dehydration of the mineral serpentine, rather than by brittle deformation of the rock. (Dobson et al., 2002).

Serpentine [Mg₃Si₂O₅(OH)₄] is primarily formed through the hydration of olivine and pyroxene,  minerals which are abundant in sea-floor crust.  Much of the serpentinization takes place near to the spreading ridges, where the rocks are still hot, and and there is well-developed convection of sea water through the new crust.  As described by Kerrick (2003), the slowly-spreading Atlantic lithosphere is consistently more strongly serpentinized than the more rapidly spreading Pacific lithosphere.  Kerrick speculates that some serpentinization may also take place near to the "outer rise", where oceanic lithosphere bends prior to being subducted – and that this process could generate additional serpentine in the Pacific lithosphere.

Dobson et al. have not actually observed water-generated earthquakes in subduction zones, but they have subjected serpentinite to subduction-like pressures and temperatures in the laboratory.  They have observed “acoustic emissions” at serpentine dehydration temperatures, and they propose that under real conditions such events could present as seismic events.  Obara (2002) believes that he has observed flowing water in the deep part of a subduction zone, and Dobson et al. suggest that the pre-cursor tremors recognized by Obara are the same as the dehydration acoustic emissions observed in the laboratory.

References

Dobson, D., Meredith, P. and Boon, S., 2002, Simulation of subduction zone seismicity by dehydration of serpentine, Science, V. 298, p. 1407-09. (Nov. 2002)

Kerrick, D., 2002, Serpentine seduction, Science, V. 298, p. 1344-45. (Nov. 2002)

Obara, K., 2002, Nonvolcanic deep tremor associated with subduction in southwest Japan, Science, V. 296, p. 1679-81 (May 2002)  (see a summary of the Obara paper)

Steven Earle, Geology Department, Malaspina University College, 2002. Return to Earth Science News