Seismic attenuation beneath Kanto, Japan: evidence for high attenuation in the serpentinized subducting mantle
© Nakajima; licensee Springer. 2014
Received: 9 December 2013
Accepted: 25 March 2014
Published: 16 April 2014
The three-dimensional (3-D) P-wave attenuation (Q p -1) structure beneath Kanto, Japan, is estimated by using a large number of waveform data. Corner frequencies of earthquakes are initially calculated from spectral ratios of S-coda waves, followed by an inversion to simultaneously determine attenuation terms and frequency-dependent site amplification factors. The attenuation terms are then inverted for estimation of the 3-D Q p -1 structure. The obtained results show that seismic attenuation is highly heterogeneous, and pronounced high-attenuation areas are located in the continental lower crust and mantle of the Philippine Sea slab. Seismic activity is very low in the high-attenuation lower crust, which is most likely attributable to ductile deformation facilitated by fluids supplied from the underlying Philippine Sea slab. The high-attenuation area in the Philippine Sea slab represents the serpentinized mantle, and two M ~ 7 earthquakes are documented to have occurred along the western boundary of this area. Interplate earthquakes on the Pacific slab are absent in areas overlain by the serpentinized Philippine Sea slab, which is likely due to the low viscosity of serpentine promoting continuous ductile deformation rather than brittle failures along the plate boundary.
Seismic attenuation provides additional insights into seismogenesis and plays a crucial role in the estimation of strong surface motions, but there are few studies on three-dimensional (3-D) seismic attenuation beneath Kanto. A pioneering work by Sekiguchi (1991) imaged the subducting Philippine Sea slab as a marked low-attenuation area, and Nakamura et al. (2006) revealed the existence of a pronounced E-W trending high-attenuation area at depths of 20 to 60 km. However, we still do not have enough information on seismic attenuation compared to seismic velocity. Therefore, precisely estimating seismic attenuation is of considerable importance to understand the complicated structure and mechanical interactions between the two subducting slabs. In this study, the 3-D P-wave attenuation (Q p -1) structure is estimated by using a large number of high-quality waveform data recorded at a nationwide seismograph network in Japan. The results are then discussed with reference to seismotectonics beneath Kanto.
Corner frequency, attenuation term, and site response
The three-step approach developed by Nakajima et al. (2013) was applied to estimate 3-D Q p -1 structure, which can minimize potential tradeoffs among unknown parameters to be solved, in addition to attenuation terms (t*) along a ray path. Firstly, corner frequencies (f c ) of earthquakes were estimated from spectral ratios of S-coda waves, and then a joint inversion was performed in order to determine attenuation terms and frequency-dependent site-amplification factors. Finally, the set of t* was inverted for the 3-D Q p -1 structure. The same criteria for waveform analysis and parameter selections given in Nakajima et al. (2013) were adopted for this study, and the outline of the analysis is described in the following paragraphs.
Velocity waveforms were collected for 336 earthquakes with focal depths (H) of less than 30 km and 2,621 earthquakes with H ≥ 30 km, all of which occurred from January 2003 to October 2013. Magnitudes of these earthquakes ranged from 2.5 to 5.0. Spectral amplitudes of the transverse component were calculated for S-coda waves, with a time window of 10 s taken at twice the theoretical S-wave travel time for the 1-D seismic velocity model (Hasegawa et al. 1978). Spectral ratios were calculated at common stations for available earthquake pairs that had a magnitude difference of ≥0.5, and the spectral ratios were subsequently stacked. Then, an ω2 source model (Brune 1970) was fitted to the stacked spectral ratio in the 1- to 32-Hz frequency range, and values of f c for the earthquake pair were estimated. As a result, f c for 1,915 earthquakes were determined.
P-wave and noise spectral amplitudes were subsequently calculated from the vertical component with a window length of 2.56 s after and before the onset of P-wave arrival, respectively. We assumed a frequency-dependent attenuation term, t* (f), as , where t0* is the attenuation term at 1 Hz (e.g., Stachnik et al. 2004). A frequency-dependent term, α, was estimated by the method of Stachnik et al. (2004), yielding the optimal range of 0.2 to 0.3. The most relevant α for the mantle, 0.27 (e.g., Jackson et al. 2002), lies in this range, and therefore, α of 0.27 was adopted as the frequency-dependent factor. Observation equations were constructed as a set of equations for many earthquakes at one station and a joint inversion was carried out for t0*, frequency-dependent site-amplification factors, and an offset that controls the level of spectra.
The 3-D Q p -1 structure was estimated by the inversion of t0* (e.g., Rietbrock 2001). The ray-tracing technique of Zhao et al. (1992) was used to calculate ray paths and travel times for the 3-D P-wave velocity model of Nakajima et al. (2009). The Conrad and Moho discontinuities in the continental plate (Katsumata 2010), and the upper boundary of the subducting Pacific plate (Nakajima et al. 2009) were introduced as velocity discontinuities in the model space. Two 3-D grid nets were set in the model space, and a value of Q p -1 for each grid node was estimated. One net covered the crust and mantle in the continental and Philippine Sea plates, with grid nodes spaced at a horizontal interval of 0.1° and vertical intervals of 10 to 20 km down to a depth of 200 km. The other net covered the Pacific slab, where grid nodes with a horizontal interval of 0.2° were set at distances of 5, 20, and 40 km from the slab surface and orientated subparallel to it. Initial values of 0.0033 were used for Q p -1 for the crust and mantle and 0.001 for the Pacific slab.
High-attenuation areas in the lower crust of the continental plate
The high-attenuation area that is laterally extensive at a depth of 20 km can be divided into a region located around active volcanoes in the western part of the study area and a region located in non-volcanic areas. As high attenuation observed in the volcanic areas in western Kanto is comparable to that observed in volcanic areas, for example, in northeastern Japan (Nakajima et al. 2013) and Mexico (Chen and Clayton 2009), it is probably caused by high temperatures and magmatic fluids. In contrast, high attenuation in the lower crust of the non-volcanic areas cannot be due to a high-temperature effect, as this area has the lowest heat flow values (e.g., Tanaka et al. 2004) and the deepest cutoff depth (>25 km) of crustal earthquakes in the Japanese Islands (Omuralieva et al. 2012).
A key observation to interpret these non-volcanic, high-attenuation areas of the lower crust is that almost no seismicity occurs in regions characterized by high attenuation and low velocity (Figures 5 and 6). A plausible cause of the aseismic and high-attenuation lower crust is the localization of overpressured fluids with a fluid pressure to lithostatic pressure ratio of approximately 1. Such high pore pressures may no longer sustain significant regional shear stresses, which promotes aseismic deformation and enhances high attenuation. The enrichment of slab-derived silica in the lower crust (e.g., Breeding and Ague 2002) may also promote aseismic deformation. Abundant fluids, supplied from the subducting Philippine Sea slab over geological time scales due to slab metamorphism, are likely to play important roles in the promotion of a large-scale aseismic deformation in the lower crust.
The high-attenuation area observed at the corner of the mantle wedge corresponds to an area of low velocity and moderate-to-high Vp/Vs ratio, where serpentinization may occur (e.g., Kamiya and Kobayashi 2000; Matsubara et al. 2005). Assuming that high attenuation represents the serpentinized mantle, serpentinization is confined to a small region immediately below the continental Moho. Because seismic ruptures along the plate boundary can propagate to greater depths beyond the localized serpentinized mantle (e.g., Collings et al. 2012), comprehensive evaluations involving the slip deficit along the plate boundary and frictional properties of serpentine are required to conclude whether the serpentinized mantle wedge could act as a barrier to megathrust seismic ruptures.
High-attenuation mantle of the subducting Philippine Sea slab
Small repeating earthquakes along the upper surface of the Pacific slab (white stars in Figure 7) (Uchida et al. 2009) occur in areas that are overlain by low-attenuation and high-velocity mantle of the Philippine Sea slab and are absent below the serpentinized slab mantle (Figure 7). The results of deformation experiments performed on serpentine show that its viscosity is significantly lower than that of the major non-metamorphic mantle-forming minerals; thus brittle failures are unlikely to occur (e.g., Moore et al. 1997; Hilairet et al. 2007). Observations in this study are consistent with the interpretation made by Hilairet et al. (2007) in that interplate earthquakes do not occur when the plate boundary is in contact with serpentinized mantle.
Two M ~ 7 earthquakes in the Philippine Sea slab (red stars in Figure 7) are interpreted to have occurred as a result of right-lateral deformation along the western boundary of the serpentinized mantle (Nakajima and Hasegawa 2010). The western boundary of the observed high-attenuation mantle coincides with that of the serpentinized mantle (Figure 7a), suggesting that these two earthquakes occurred at the sharp boundary between the high-attenuation mantle to the east and the low-attenuation mantle to the west. Although it is inconclusive as to whether the western boundary of the serpentinized mantle corresponds to the thermal limit of the stability field of serpentine or whether it is controlled by the distribution of fluids, frictional properties and pore fluid pressures may change abruptly across the boundary. These observations highlight the importance of heterogeneity in the subducting mantle for the facilitation of intermediate-depth earthquakes.
The lower continental crust above the Philippine Sea slab shows high attenuation and is almost aseismic. Ductile deformation is likely to be dominant, associated with overpressured fluids supplied from the underlying Philippine Sea slab.
The corner of the mantle wedge in the continental plate may be serpentinized, but the extent of serpentinization is limited to a small region below the Moho.
The serpentinized mantle of the Philippine Sea slab shows high attenuation as well as low velocity, which strongly suggests that serpentinization enhances attenuation in the seismic frequency band.
The attenuation model obtained in this study will provide a practical constraint on the evaluation of realistic strong motions at the surface, which is crucial for the assessment and management of seismic hazards in the Tokyo metropolitan area.
A. Hasegawa, I. Katayama, and K. Hirauchi are thanked for comments and discussions, and N. Uchida is thanked for providing data relating to small repeating earthquakes. Constructive and careful reviews by two anonymous reviewers improved the manuscript. This work used waveform data recorded at a nationwide seismograph network and arrival time data in the unified catalogue of the Japan Meteorological Agency. This work was supported by the Ministry of Education, Culture, Sports, Science, and Technology of Japan, under its Observation and Research Program for the Prediction of Earthquakes and Volcanic Eruptions, and by JSPS KAKENHI Grant Number 24740300.
- Breeding CM, Ague JJ: Slab-derived fluids and quartz-vein formation in an accretionary prism. Otago Schist, New Zealand, Geology 2002, 30: 499–502.Google Scholar
- Brune JN: Tectonic stress and spectra of seismic shear waves from earthquakes. J Geophys Res 1970, 75: 4997–5009. 10.1029/JB075i026p04997View ArticleGoogle Scholar
- Chen T, Clayton RW: Seismic attenuation structure in central Mexico: image of a focused high-attenuation zone in the mantle wedge. J Geophys Res 2009, 114: B07304. doi:10.1029/2008JB005964Google Scholar
- Collings R, Lange D, Rietbrock A, Tilmann F, Natawidjaja D, Suwargadi B, Miller M, Saul J: Structure and seismogenic properties of the Mentawai segment of the Sumatra subduction zone revealed by local earthquake travel time tomography. J Geophys Res 2012, 117: B01312. doi:10.1029/2011JB008469Google Scholar
- Faul UH, Jackson I: The seismological signature of temperature and grain size variations in the upper mantle. Earth Planet Sci Lett 2005, 234: 119–134. 10.1016/j.epsl.2005.02.008View ArticleGoogle Scholar
- Hasegawa A, Umino N, Takagi A: Double-planed structure of the deep seismic zone in the northeastern Japan arc. Tectonophysics 1978, 47: 43–58. 10.1016/0040-1951(78)90150-6View ArticleGoogle Scholar
- Hilairet N, Reynard B, Wang Y, Daniel I, Merkel S, Nishiyama N, Petitgirard S: High-pressure creep of serpentine, interseismic deformation, and initiation of subduction. Science 2007, 318: 1910–1913. 10.1126/science.1148494View ArticleGoogle Scholar
- Hirose F, Nakajima J, Hasegawa A: Three-dimensional velocity structure and configuration of the Philippine Sea slab beneath Kanto district, central Japan, estimated by double-difference tomography (in Japanese with English abstract). J Seismo Soc Jpn 2008, 60: 123–138.Google Scholar
- Hori S: Seismic activity associated with the subducting motion of the Philippine Sea plate beneath the Kanto district. Japan, Tectonophysics 2006, 417: 85–100. 10.1016/j.tecto.2005.08.027View ArticleGoogle Scholar
- Iwamori H: Deep subduction of H2O ad deflection of volcanic chain towards backarc near triple junction due to lower temperature. Earth Planet Sci Lett 2000, 181: 41–46. 10.1016/S0012-821X(00)00180-1View ArticleGoogle Scholar
- Jackson I, Fitz Gerald JD, Faul UH, Tan BH: Grain-size sensitive seismic wave attenuation in polycrystalline olivine. J Geophys Res 2002, 107(B12):2360. doi:10.1029/2001JB001225View ArticleGoogle Scholar
- Kamimura A, Kasahara J, Shinohara M, Hino R, Shiobara H, Fujie G, Kanazawa T: Crustal structure study at the Izu-Bonin subduction zone around 31°N: implications of serpentinized materials along the subduction plate boundary. Phys Earth Planet Inter 2002, 132: 105–129. 10.1016/S0031-9201(02)00047-XView ArticleGoogle Scholar
- Kamiya S, Kobayashi Y: Seismological evidence for the existence of serpentinized wedge mantle. Geophys Res Lett 2000, 27: 819–822. 10.1029/1999GL011080View ArticleGoogle Scholar
- Katsumata A: Depth of the Moho discontinuity beneath the Japanese islands estimated by traveltime analysis. J Geophys Res 2010, 115: B04303. doi:10.1029/2008JB005864Google Scholar
- Kimura H, Kasahara K, Igarashi T, Hirata N: Repeating earthquake activities associated with the Philippine Sea plate subduction in the Kanto district, central Japan: a new plate configuration revealed by interpolate aseismic slips. Tectonophysics 2006, 417: 101–118. 10.1016/j.tecto.2005.06.013View ArticleGoogle Scholar
- Matsubara M, Hayashi H, Obara K, Kasahara K: Low-velocity oceanic crust at the top of the Philippine Sea and Pacific plates beneath the Kanto region, central Japan, imaged by seismic tomography. J Geophys Res 2005, 110: B12304. doi:10.1029/2005JB003673View ArticleGoogle Scholar
- Mizukami T, Wallis SR: Structural and petrological constraints on the tectonic evolution of the garnet-lherzolite facies Higashi-akaishi peridotite body. Sanbagawa belt, SW Japan, Tectonics 2005, 24: TC6012. doi:10.1029/2004TC001733Google Scholar
- Moore DE, Lockner DA, Shengli M, Summers R, Byerlee JD: Strength of serpentinite gouges at elevated temperatures. J Geophys Res 1997, 102: 14787–14801. 10.1029/97JB00995View ArticleGoogle Scholar
- Nakajima J, Hasegawa A: Subduction of the Philippine Sea plate beneath southwestern Japan: slab geometry and its relationship arc magmatism. J Geophys Res 2007, 112: B08306. doi:10.1029/2006JB004770Google Scholar
- Nakajima J, Hasegawa A: Cause of M 7 intraslab earthquakes beneath the Tokyo metropolitan area: possible evidence for a vertical slab tear at the easternmost portion of the Philippine Sea slab. J Geophys Res 2010, 115: B04301. doi:10.1029/2009JB006863Google Scholar
- Nakajima J, Hirose F, Hasegawa A: Seismotectonics beneath the Tokyo metropolitan area: effect of slab-slab contact and overlap on seismicity. J Geophys Res 2009, 114: B08309. doi:10.1029/2008JB006101Google Scholar
- Nakajima J, Hasegawa A, Hirose F: Stress regime in the Philippine Sea slab beneath Kanto, Japan. Geophys Res Lett 2011, 38: L16318. doi:10.1029/2011GL048754View ArticleGoogle Scholar
- Nakajima J, Hada S, Hayami E, Uchida N, Hasegawa A, Yoshioka S, Matsuzawa T, Umino N: Seismic attenuation beneath northeastern Japan: constraints on mantle dynamics and arc magmatism. J Geophys Res Solid Earth 2013., 118: doi:10.1002/2013JB010388Google Scholar
- Nakamura R, Satake K, Toda S, Uetake T, Kamiya S: Three-dimensional attenuation (Qs) structure beneath the Kanto district, Japan, as inferred from strong motion records. Geophys Res Lett 2006, 33: L21304. doi:10.1029/2006GL027352View ArticleGoogle Scholar
- Omuralieva AM, Hasegawa A, Matsuzawa T, Nakajima J, Okada T: Lateral variation of the cutoff depth of shallow earthquakes beneath the Japanese Islands and its implications for seismogenesis. Tectonophysics 2012, 518–521: 93–105.View ArticleGoogle Scholar
- Rietbrock A: P wave attenuation structure in the fault area of the 1995 Kobe earthquake. J Geophys Res 2001, 106: 4141–4154. 10.1029/2000JB900234View ArticleGoogle Scholar
- Sato H, Hirata N, Koketsu K, Okaya D, Abe S, Kobayashi R, Matsubara M, Iwasaki T, Ito T, Ikawa T, Kawanaka T, Kasahara K, Harder S: Earthquake source fault beneath Tokyo. Science 2005, 309: 462–464. 10.1126/science.1110489View ArticleGoogle Scholar
- Sekiguchi S: Three-dimensional Q structure beneath the Kanto–Tokai district. Japan, Tectonophysics 1991, 195: 83–104. 10.1016/0040-1951(91)90145-IView ArticleGoogle Scholar
- Stachnik JC, Abers GA, Christensen DH: Seismic attenuation and mantle wedge temperatures in the Alaska subduction zone. J Geophys Res 2004, 109: B10304. doi:10.1029/2004JB003018View ArticleGoogle Scholar
- Tanaka A, Yamano M, Yano Y, Sasada M: Geothermal gradient and heat flow data in and around Japan (I): appraisal of heat flow from geothermal gradient data. Earth Planets Space 2004, 56: 1191–1194.View ArticleGoogle Scholar
- Uchida N, Nakajima J, Hasegawa A, Matsuzawa T: What controls interplate coupling? evidence for abrupt change in coupling across a border between two overlying plates in the NE Japan subduction zone. Earth Planet Sci Lett 2009, 283: 111–121. 10.1016/j.epsl.2009.04.003View ArticleGoogle Scholar
- Uchida N, Matsuzawa T, Nakajima J, Hasegawa A: Subduction of a wedge-shaped Philippine Sea plate beneath Kanto, central Japan, estimated from converted waves and small repeating earthquakes. J Geophys Res 2010, 115: B07309. doi:10.1029/2009JB006962Google Scholar
- Wu F, Okaya D, Sato H, Hirata N: Interaction between two subducting plates under Tokyo and its possible effects on seismic hazards. Geophys Res Lett 2007, 34: L18301. doi:10.1029/2007GL030763View ArticleGoogle Scholar
- Zhao D, Hasegawa A, Horiuchi S: Tomographic imaging of P and S wave velocity structure beneath northeastern Japan. J Geophys Res 1992, 97: 19909–19928. 10.1029/92JB00603View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.