Bina, C. R., and E. A. Okal, Investigating the depth of the deepest deep earthquakes, Abstracts of the Sixth Symposium of Study of the Earth's Deep Interior (SEDI), Tours, France, 119, 1998.
How precisely deep are the deepest deep earthquakes? What controls the cessation of seismicity at the bottom of the transition zone? Do maximum depths vary with region and how? Could the lower mantle entertain some level of undetected seismicity?
Through a detailed analysis of seismicity at the base of the transition zone, we obtain an updated value of 690 km (with a precision of 10 km) for the maximum reliable depth of confirmed seismicity. We classify deep subduction zones into three groups: those whose seismicity does not reach beyond 620 km (Solomon, Argentina); those whose seismicity appears to terminate around 650 to 660 km (Sangihe, Banda Sea, Java, Kuriles, Bolivia, Peru-Brazil, Marianas; and isolated shocks in Colombia-Northern Peru and Spain); and those whose seismicity extends to 685-690 km (Tonga-Fiji and the Vityaz cluster). We note that the maximum depth of seismicity does not seem to correlate simply with the thermal parameter of subduction, as introduced by Kostoglodov or Kirby. Rather, we suggest that the depth extent of seismicity is controlled by the second mantle transition, responsible for the "660-km" seismic discontinuity, and which is deflected to greater depths in cold slabs than in warmer ones. We note that the transition marks the depth below which thermal perturbation of phase transitions no longer generates buoyancy anomalies and their large attendant down-dip compressive stresses, and below which strain energy generated by other mechanisms may not accumulate to seismogenic levels due to superplastic weakness in fine-grained materials. Thermodynamic models of the deflection of the discontinuity are consistent with the observed 30 km increase in the maximum depth of seismicity between the uniquely fast-sinking Tonga slab and most other subduction zones.
We then address the question of the possible existence of undetected seismicity in the lower mantle below 690 km. We extend to that region Okal and Kirby's modeling of moment-frequency relations for deep earthquakes, and conclude that the failure to observe even one earthquake below 690 km in the roughly 40 years since the advent of adequate seismological networks requires that any seismic activity in the deep mantle, if present, must be at least three orders of magnitude weaker than that observed in the transition zone.Copyright © 1998 SEDI