Comment on ``Seismological evidence for the existence of anisotropic zone in the metastable wedge inside the subducting Izu-Bonin slab'' by Takashi Iidaka and Kazushige Obara

George Helffrich

Earth Sciences Department, University of Bristol, Bristol, England
In press, Geophys. Res. Lett.: Received April 8, 1998; accepted July 21, 1998.

      Nobody fully understands the causes of deep earthquakes. They appear to be related to the depths where phase transformations occur in slabs [Kirby, 1987; Green and Burnley, 1989; Frohlich, 1989; Kirby et al., 1991; Helffrich and Brodholt, 1991; Green et al., 1992; Furukawa, 1994; Rubie and Ross, 1994; Green and Houston, 1995; Bina, 1996; 1997], and it is possible to make fault-like structures in experiments when olivine is deformed in a metastable state prior to a phase transformation [Green and Burnley, 1989]. These observations constitute the evidence for the variously-called "transformational faulting" (TF) or "anticrack" (AC) theory of deep earthquakes. [Green and Burnley, 1990; Kirby et al., 1991].

      This theory holds that oriented lenses of transforming or transformed material form in the body that ultimately faults. Establishing the presence of lenses in subducted slabs would convince many of the theory's validity. Thus the paper by Iidaka and Obara [1997] tackles an important problem in the study of deep earthquakes, and the authors approach it in a creative way. The authors claim that their results indicate anisotropy in the region between 340 and 390 km depth in the Izu-Bonin subduction zone. They cite observations of a difference in shear wave splitting times from events in the upper and lower parts of a double seismic zone as proof.

      Intriguing as these observations are, they are problematic to interpret on three accounts:

• The information presented is insufficient to assess whether anisotropy in the region between the earthquakes is required by the observations.
• The effect of more than one anisotropic region on the results isn't addressed.
• The required uniformity of stress to achieve the lens alignment that generates anisotropy is absent.
  

Localizing Anisotropy Within a Region

      A shear wave is split by an anisotropic medium when the wave's polarization direction is not parallel to a symmetry axis of the medium [Babuska and Cara, 1991]. The amount of splitting depends on the distance travelled, the wave vector in the medium, and the degree of anisotropy. To assess whether the splitting attributed to a region is actually developed in it, one needs to show that: 1) identical initial S polarizations from above and below the region yield different amounts of splitting; and 2) the splitting delay time linearly depends on the depth into the presumed anisotropic region, techniques established by earlier studies [Kaneshima and Silver, 1995; Gledhill and Gubbins, 1995; Fouch and Fischer, 1996].

      The initial S wave polarization is crucial to the demonstration of anisotropy. There may be uniform anisotropy outside the double seismic zone that the groups respond differently to if the source polarization of the events in the upper group is uniformly different to the lower group's. For example, the upper group's polarization might parallel a medium symmetry axis in an anisotropic mantle wedge underlying an anisotropic crust. This group would be insensitive to splitting in the mantle wedge, but be split during the crustal transit. If differently polarized, the lower group's S waves would however be split successively by the mantle wedge and the crust, possibly leading to longer apparent delay times [Silver and Savage, 1994]. Without focal mechanism information to establish the initial S polarization, this scenario is impossible to reject.

      The second effect expected for an anisotropic region is increasing delay times for increasing path lengths in the region. This is not evident in the observations (Figure 1) [Iidaka and Obara, 1997], which show unexpectedly shorter delay times with increasing depth.

Figure 1. Observed delay time dependence with depth, reported by Iidaka and Obara [1997]. Upper and Lower refer to positions in the Izu-Bonin double seismic zone. t is the time lag between the fast and slow polarization direction arrivals. The uncertainties represent the reported one- observational errors and the depth uncertainties are ±10 km customarily attributed to hypocenters reported by the International Seismic Centre (cited by the authors as their source). In contrast to a trend of increasing delay times with depth expected for shear waves passing through an anisotropic region, the trend has the opposite sense. It is, however, unresolvable given the observational uncertainties.

      Finally, source dimension effects could give rise to a difference in the apparent splitting in the two event groups. If one group ruptures through the whole distance separating the two, this could yield differences in splitting simply because the source polarization isn't yet established. By providing the earthquake magnitude or moment, the authors could easily dismiss this improbable effect, but the information is not provided.

Multiple anisotropic regions

      Generally, the crust is anisotropic. Average crustal splitting delay times are 0.2 s [Kaneshima, 1990; Silver and Chan, 1991; Barruol and Mainprice, 1993], which is close to the delay time attributed to the wedge region here. The potential for it affecting the deeper signal should be clear because the delay found in the study is about 0.20 s (Figure 1). The crustal anisotropy may modify the splitting differently for the two event groups if their polarizations differ before they enter the crust. Small differences in initial polarization may have large effects on apparent delay times due to the phase contribution of the two layers [Silver and Savage, 1994]. Ruling out crustal effects at small delay times demands that the incoming S wave polarization be considered.

Stress Orientation

      Focal mechanisms give point estimates of the in-situ stress orientations at depth, an advantage over the stress orientations inferred from plate kinematics which the authors used to predict anticrack orientations. Zhou [1990] studied the P- and T-axis orientation of earthquakes in the Western Pacific, including Izu-Bonin. Zhou [1990] found a mixture of downdip P and T axes here, but with downdip T dominant in the southern part of the study area, as well as to its north. These observations contradict both the orientation and the uniformity of the maximum compressive stress claimed by the authors: where P axis orientation is uniform, it is orthogonal to the orientation inferred here, and in the study area it does not even appear to be uniform. Thus there is no particular support for the splitting observations from this imperfectly known deep stress orientation.

Conclusions

      The omission of polarization information, neglect of the effects of two-layer splitting, and the heterogeneity of stress in the Izu-Bonin slab weaken the authors' claim that the coolest part of the slab, where metastable olivine is likely to exist, is anisotropic due to aligned lenses of transformed material. Perhaps the most valuable contribution is the insight that splitting differences can be used to interrogate the metastable wedge's structure. Further attempts along this line of inquiry might be a promising basis for proving the TF/AC model.

References

Babuska, V., and M. Cara, Seismic anisotropy in the Earth, 217+viipp, Kluwer, Dordrecht, the Netherlands, 1991.
Barruol, G., and D. Mainprice, A quantitative evaluation of the contribution of crustal rocks to the shear wave splitting of teleseismic SKS measurements, Phys. Earth Planet. Inter., 78, 281-300, 1993.
Bina C.R., Phase-transition buoyancy contributions to stresses in subducting lithosphere, Geophys. Res. Lett., 23, 3563-3566, 1996.
Bina CR, Patterns of deep seismicity reflect buoyancy stresses due to phase transitions, Geophys. Res. Lett., 24, 3301-3304, 1997.
Fouch, M., and K. Fischer, Mantle anisotropy beneath northwest Pacific subduction zones, J. Geophys. Res., 101, 15,987-16,002, 1996.
Frohlich, C., The nature of deep-focus earthquakes, Annu. Rev. Earth Planet. Sci., 17, 227-254, 1989.
Furukawa, Y., Two types of deep seismicity in subducting slabs, Geophys. Res. Lett., 21, 1181-1184, 1994.
Gledhill K, and Gubbins D, SKS splitting and the seismic anisotropy of the mantle beneath the Hikurangi subduction zone, New Zealand, Phys. Earth Planet. Inter., 95, 227-236, 1996.
Green H.W., and Houston H., The mechanics of deep earthquakes, Annu. Rev. Earth Planet. Sci., 23, 169-213, 1995.
Green, H. W., and P. C. Burnley, A new self-organizing mechanism for deep-focus earthquakes, Nature, 341, 733-737, 1989.
Green, H.W., C.H. Scholz, T.N. Tingle, T.E. Young, and T.A. Koczynski, Acoustic emissions produced by anticrack faulting during the olivine-spinel transformation, Geophys. Res. Lett., 19, 789-792, 1992.
Helffrich, G., and J. Brodholt, The relationship of deep seismicity to the thermal age of subducted lithosphere, Nature, 353, 252-255, 1991.
Iidaka, T., and K. Obara, Seismological evidence for the existence of anisotropic zone in the metastable wedge inside the subducting lithosphere, Geophys. Res. Lett., 24, 3305-3308, 1997.
Kaneshima, S., Origin of crustal anisotropy: Shear wave splitting studies in Japan, J. Geophys. Res., 95, 11,121-11,133, 1990.
Kaneshima, S., and P. G. Silver, Anisotropic loci in the mantle beneath central Peru, Phys. Earth Planet. Inter., 88, 257-272, 1995.
Kirby, S. H., Localized polymorphic phase transformations in high-pressure faults and application to the physical mechanism of deep earthquakes, J. Geophys. Res., 92, 13,789-13,800, 1987.
Kirby, S. H., W. B. Durham, and L. A. Stern, Phase transformations, anomalous faulting, and deep earthquakes: A new perspective, Science, 252, 216-225, 1991.
Rubie D., and C. Ross, Kinetics of the olivine-spinel transformation in subducting lithosphere - experimental constraints and implications for deep slab processes, Phys. Earth Planet. Inter., 86, 223-241, 1995.
Silver, P. G., and W. W. Chan, Shear wave splitting and subcontinental mantle deformation, J. Geophys. Res., 96, 16,429-16,454, 1991.
Zhou, H. W., Observations on earthquake stress axes and seismic morphology of deep slabs, Geophys. J. Int., 103, 377-401, 1990.