GEOPHYSICAL RESEARCH LETTERS, VOL. 30, NO. 11, 1586, doi:10.1029/2002GL016837, 2003
Anomalous seismograms generated by an intermediate-depth earthquake: Unusual scattering sources in the upper mantle of central Japan Cheng-Horng Lin,1 Masataka Ando,2 Naoyuki Fujii,2 Koshun Yamaoka,2 Keiichi Tadokoro,2 An-Shu Jin,3 Kazusige Obara,3 and Mizuho Ishida3 Received 23 December 2002; revised 27 March 2003; accepted 31 March 2003; published 11 June 2003.
[1] Strong scattering sources were detected in the upper mantle beneath central Japan from seismic data recorded by the high-sensitivity seismograph network (Hi-net). Anomalous seismograms, consisting of many strong later phases whose amplitudes did not decay significantly with time, were observed from an intermediate-depth earthquake. A comparison of those anomalous seismograms with others generated by other deep earthquakes shows that major scattering sources may exist predominately in the upper mantle, not in the crust. Three possible candidates have been considered as the sources of scattering in the upper mantle including magma conduits, the previous collision zone and slab melting. Although seismic waves may be scattered strongly from each potential candidate, a region associate with slab melting might be more suitable for generating the anomalous seismogram due to the existence of abundant melting spots within the subducted Philippine Sea slab that is extremely young and hot beneath central INDEX TERMS: 0669 Electromagnetics: Scattering and Japan. diffraction; 7218 Seismology: Lithosphere and upper mantle; 7230 Seismology: Seismicity and seismotectonics. Citation: Lin, C.-H., M. Ando, N. Fujii, K. Yamaoka, K. Tadokoro, A.-S. Jin, K. Obara, and M. Ishida, Anomalous seismograms generated by an intermediate-depth earthquake: Unusual scattering sources in the upper mantle of central Japan, Geophys. Res. Lett., 30(11), 1586, doi:10.1029/2002GL016837, 2003.
1. Introduction [2] The scattering of seismic waves is often observed within the crust in fault zones and volcanic areas [Nishigami, 1997; Mikada et al., 1997; Chen and Long, 2000]. There are, however, few observations of strong scattering within the upper mantle. Strong scattering of seismic waves is often observed from shallow earthquakes in that their raypaths are mainly confined within the upper crust, where the heterogeneity is significantly higher than that in the upper mantle. Open cracks or fractures are one of the most general explanations for generating strong scattering at shallow depths [Chen and Long, 2000]. The most extreme case is the seismograms recorded on the moon [Dainty et al., 1974;
1
Institute of Earth Sciences, Academia Sinica, Taipei, Taiwan. Research Center for Seismology and Volcanology, Nagoya University, Nagoya, Japan. 3 National Research Institute for Earth Science and Disaster Prevention, Tsukuba, Japan. 2
Copyright 2003 by the American Geophysical Union. 0094-8276/03/2002GL016837$05.00
40
Nakamura et al., 1982], where abundant cracks in its upper crust have been produced by the impact of meteoroids over the past. In the upper mantle for either the earth or the moon, however, it is almost impossible to have open cracks because the lithostatic pressure is too high to keep cracks or fractures at a great depth. As a result, strong seismic scattering is rarely generated in the upper mantle in general. [3] It was very surprising to see that an anomalous seismogram was generated by an intermediate-depth earthquake and recorded at several Hi-net seismic stations nearby the epicenter in central Japan. The waveform was differing from typical recordings generated by a mantle event, for which the coda is significantly weaker than the direct arrivals [Lin et al., 1999]. Instead, the waveform comprises many strong later phases whose amplitudes do not decay significantly with time. Obviously, those later phases with large amplitudes were generated by the propagation of seismic waves through some complicated structures with strong scattering sources. [4] To better discern and improve the understanding of scattering sources generating anomalous seismograms reported here, the general tectonics in central Japan is first reviewed. Seismograms generated by other deep earthquakes and recorded at the Hi-net seismic stations in central Japan are then examined to explore the spatial distribution of possible scattering sources. Finally, possible explanations for the anomalous seismograms are discussed in relation to regional tectonics within the subducted Philippine Sea slab area in question.
2. Tectonics [5] Central Japan is located at a very complicated tectonic place where four plates have converged by two strongly active subduction systems (Figure 1) [Ishida, 1992; Seno et al., 1996; Noguchi, 2002]. One convergence system is the westward subduction of the Pacific (PAC) plate beneath both the Okhotsk (OKH) plate and the Philippine Sea (PHS) plate along the Japan and Izu-Bonin trenches, respectively. The geometry of the subducted PAC plate can be defined unambiguously from the well-recorded numerous earthquakes. The depths of the subducted PAC plate increases uniformly in a westward direction ranging from 80 km to 300 km beneath central Japan. The volcanic front associated with the subducted PAC plate can be traced on the surface through active volcanoes. The other convergence system is the northwestward subduction of the PHS plate beneath both the Eurasian (EUR) plate and OKH plate along the Suruga and Sagami troughs, respectively. - 1
40 - 2
LIN ET AL.: STRONG SCATTERING FROM THE UPPER MANTLE
Figure 1. General tectonics and locations of the Hi-net seismic stations (small squares) and two mantle earthquakes (Events A and B marked by co-circles) in central Japan. The leading edge and iso-depth contours of the subducted PHS, depicted by the Wadati-Benioff zone (modified from Noguchi [2002]), are shown by thick and thin lines, respectively. The volcanic front depicted by active volcanoes (solid triangles) and the Itoigawa-Shizuoka tectonic line (ISTL) are marked by thick and thin dashed lines, respectively. The inset map marks the location of the study area (box) in the regional tectonic map. The EUR and OKH plates in central Japan are separated by an earlier collision zone, the Itoigawa-Shizuoka Tectonic Line (ISTL) [Seno et al., 1996].
3. Seismic Data and Analyses [6] Seismic data used in this study were recorded by the National Research Institute for Earth Science and Disaster Prevention’s (NIED) high-sensitivity seismograph network (Hi-net), which comprises of more than 600 short-period borehole stations at depths ranging from 100 m to 200 m in Japan [Obara, 2002]. The seismograms with a high signal to noise ratio we are interested in this paper were generated by mantle earthquakes (Table 1) and recorded at several tens of Hi-net seismic stations in central Japan. Since mantle events took place near the geographic center of Japan, Table 1. Earthquake Parameters for Two Mantle Events With Magnitude Greater Than 4.0 in Central Japan Event
Time (yr/mo/dy/hr/mi)
Longitude (°E)
Latitude (°N)
Depth (Km)
Mag.
A B
2001/03/13/06/07 2001/08/04/22/31
137.4679 137.2237
36.0456 35.2928
264 310
4.6 4.1
Figure 2. Three-component seismograms recorded at the station KGN and generated by (a) Event A and (b) Event B. hypocenters of those events were well located by the densely distributed Hi-net seismic stations with an excellent station-coverage (Figure 1). [7] At the first glance, it is interesting to notice that anomalous seismogram from Event A was recorded at the station of KGN (Figure 2a). The coda is spindle shaped. In other words, the seismogram included many later strong phases with relatively large amplitudes. The amplitudes for most of later P- or S-waves did not decay significantly over time, and some of them were greater in amplitude than the first arrivals. In short, those seismograms were unlike general seismograms with simpler waveforms from a typical mantle earthquake (Figure 2b), but very similar to the seismograms recorded on the moon where abundant cracks in the upper crust and low attenuation material in the moon may have resulted in many strong seismic scattering as well as minimum attenuation on seismic waves [Nakamura et al., 1982]. [8] It is generally accepted that a complicated waveform, consisting of many later phases such as anomalous seismograms, is often excited by the propagation of seismic waves through abundant, strong scattering sources along the raypaths between the hypocenter and seismic station. Thus, it is not a surprise to see an anomalous seismogram if major seismic waves are propagated within the upper crust where strong heterogeneity scattering sources may have existed. In other words, a shallow earthquake might generate an analogous anomalous seismogram if major ray-paths propagate across some complicated structures such as fault zones or volcanic areas.
LIN ET AL.: STRONG SCATTERING FROM THE UPPER MANTLE
[9] However, the possibility of the anomalous seismogram generated by a shallow crust earthquake was eliminated out by reliable evidence of high quality seismic data recorded by the densely distributed Hi-net stations (Figure 1). The seismic energy generated by Event A, with a magnitude of 4.6, was significantly greater than the background noise often found to be very small at borehole seismic stations. The first P- and S-arrivals can be unambiguously picked from a three-component seismogram recorded at more than 50 Hi-net stations across central Japan. As a result, the earthquake parameters (Table 1) were well determined by using reliable P- and S-wave arrival times. Although there are still uncertainties on earthquake locations, it seems improbable to mislocate a shallow earthquake as an intermediate-depth event. Therefore, the anomalous seismograms from Event A appear to be generated by an intermediate-depth earthquake and scattered at the crust and/or the upper mantle along the ray-paths. [10] To explore the possible depth-distribution of the scattering sources, we examined seismograms generated by other deep earthquakes and recorded at the same station, KGN. Unlike the anomalous seismograms generated by Event A (Figure 2a), for example, the waveforms from Event B are simpler (Figure 2b). Such a feature, like most seismograms generated by a mantle event, indicates that there was no major scattering source along seismic ray-paths between the hypocenter and station. The significant difference between seismograms generated by Events A and B strongly suggests that scattering sources were likely located in the upper mantle and not in the crust because both seismic ray-paths from the two events to the same station were almost the same in the crust and the major differences were located mainly at the upper mantle (Figure 3). [11] To further examine the possible of scattering sources near the hypocenters, we examined seismograms generated by Event A and recorded at other Hi-net seismic stations in
40 - 3
Figure 4. Vertical-component seismograms generated by Event A and recorded at stations (a) KGN, KSOH, SSW and TKGH, and (b) ANJ, EIHH, KGI, YOK and WAT. central Japan. The results show that the anomalous seismogram was absent at some but not all stations. The anomalous seismograms were only observed at some stations located to the southeast or east of the epicenter, e.g. KGN, SSW, KSOH and TKGH (Figure 4a). However, simpler seismograms were often observed at other stations. Most of typical simpler seismograms were observed at some stations south or southwest of the epicenter, e.g. ANJ, EIHH, KGI, YOK and WAT (Figure 4b). Therefore, scattering sources are not located in and around the hypocenter nor in the crust, but along a particular zone in the upper mantle (Figure 3).
Figure 3. (a) Map view of seismicity (small dots) as well as locations of 4 stations (squares) and two mantle events (circles) in central Japan. (b) A north-south cross-sectional view of seismicity and seismic stations for the shaded area in Figure 3a showing the location of possible scattering sources (an ellipse filled with dots) and ray-paths (dashed lines with small arrows) from Events A and B. The subducted slabs of both the Philippine Sea (PHS) and Pacific plates (PAC) and the Moho-discontinuity beneath central Japan are plotted.
4. Discussion [12] Based on regional tectonics information in central Japan, three possible structures in the upper mantle might be considered as the potential candidates for generating the anomalous seismograms. The first possibility is the magma conduits or channels beneath the volcanic front of the subducted PAC, because some active volcanoes are located nearby the stations in which the anomalous seismograms were observed (Figure 1). The magma conduits or channels in the upper mantle could be the probable sources for strong
40 - 4
LIN ET AL.: STRONG SCATTERING FROM THE UPPER MANTLE
scattering of seismic waves. To generate an anomalous moonquake-like seismogram, however, magma conduits or channels in the upper mantle beneath central Japan must have as many cracks as that in the moon. [13] Since strong scattering seismograms were recorded at some stations nearby the ISTL boundary (Figure 1), the second candidate for generating the anomalous seismograms might be related to the deep structures associated with a previous collision zone beneath the ISTL, a possible plate boundary between the EUR and OKH plates [Seno et al., 1996]. If the previous plate boundary still exists, some complicated structures with strong heterogeneity in the upper mantle might be responsible for the excitation of seismic waves recorded at the stations nearby the ISTL. However, as we mentioned before, it is hard to sustain many open fractures or cracks under high lithospheric pressure in the upper mantle. Furthermore, there is no reliable evidence to support the existence of previous collision structures in the upper mantle. [14] The last possible candidate for generating the anomalous seismograms can be associated with slab melting in the subducted PHS plate. It is well recognized that slab melting could have taken place when the subducted plate was young and hot [Defant and Drummond, 1990; Morris, 1995; Peacock and Wang, 1999]. Strong partial melting could occur within the subducted PHS plate beneath central Japan not only because it is extremely young (0 – 2 Ma) [Taylor et al., 1991] and therefore hot [Peacock and Wang, 1999], but also because some extra heat might be added from the ascending magma in the PAC plate subduction system. As a result, strong slab melting in a region beyond the downward extension of the PHS where the WadatiBenioff zone is absent [Morris, 1995] might produce many hot melting spots with high heterogeneity to account for the scattering sources in the upper mantle. [15] Although no reliable evidence can be established to rule out any of the alone discussed three explanations, the last one might be more appropriate than the others to better account for the scattering sources. Certainly, other possibilities should be examined as well to explore if other interpretations are possible.
5. Conclusion [16] Some anomalous seismograms from an intermediatedepth earthquake are characterized by many long lasting strong later phases continuously after the first P- and Sarrivals. These anomalous seismograms were recorded at several but not all seismic stations in central Japan suggesting a localized region of strong scattering sources. A comparison of those unusual seismograms with typical seismograms generated by other deep earthquakes nearby reveals that a locally defined region of scattering sources may exist in the upper mantle beneath central Japan. Since those anomalous seismograms were recorded at a few adjacent stations between the Tokai and Kanto areas, three possible structures have been considered as candidates for the scattering sources in the upper mantle. They are the magma conduits (or channels), a previous collision zone
and the apparent results of slab melting. Although all three candidates are viable sources for strong scattering of seismic waves, the last one might be more suitable due to the possibility of abundant melting spots in the upper mantle associated with the hot and young subducted slab. [17] Acknowledgments. We would like to thank J.M. Chiu and another anonymous reviewer for valuable comments and suggestions to improve this paper. The first author would like to thank his colleagues at the Research Center for Seismology and Volcanology in Nagoya University for their kind help in this research. While in Japan, financial support was offered by the Ministry of Education, Culture, Sport, Science and Technology of Japan. This research was also partially funded by both the National Science Council and the Institute of Earth Sciences, Academia Sinica at Taipei, Taiwan.
References Chen, X., and L. T. Long, Spatial distribution of relative scattering coefficients determined from microearthquake coda, Bull. Seismol. Soc. Am., 90, 512 – 524, 2000. Dainty, A. M., M. N. Toksoz, K. R. Anderson, P. J. Pines, Y. Nakamura, and G. Latham, Seismic scattering and shallow structure of the moon in oceanus procellarum, Moon, 9, 11 – 29, 1974. Defant, M. J., and M. S. Drummond, Derivation of some modern arc magmas melting of young subducted lithosphere, Nature, 347, 662 – 665, 1990. Ishida, M., Geometry and relative motion of the Philippine Sea plate and Pacific plate beneath the Kanto-Tokai district, Japan, J. Geophys. Res., 97, 489 – 513, 1992. Lin, C. H., B. S. Huang, and R. J. Rau, Seismological evidence for a lowvelocity layer within the subducted slab of southern Taiwan, Earth Planet. Sci. Lett., 174, 231 – 240, 1999. Mikada, H., H. Watanabe, and S. Sakashita, Evidence for subsurface magma bodies beneath Izu-Oshima volcano inferred from a seismic scattering analysis and possible interpretation of the magma plumbing system of the 1986 eruptive activity, Phys. Earth Planet. Inter., 104, 257 – 269, 1997. Morris, P., Slab melting as an explanation of Quaternary volcanism and aseismicity in southwest Japan, Geology, 23, 395 – 398, 1995. Nakamura, Y., G. V. Latham, and H. J. Dorman, Apollo lunar seismic experiment: Final summary, Proc. Lunar Planet. Sci. Conf. 13th, Part 1, J. Geophys. Res., 87, suppl., A117 – A123, 1982. Nishigami, K., Spatial distribution of coda scatters in the crust around two active volcanoes and one active fault system in central Japan: Inversion analysis of coda envelope, Phys. Earth Planet. Inter., 104, 75 – 89, 1997. Noguchi, S. I., Earthquake clusters in the Kanto and Tokai subduction zones: Implications for modes of plate consumption, in Seismotectonics in Convergent Plate Boundary, edited by Y. Fujinawa and A. Yoshida, pp. 451 – 467, Terra Sci., Tokyo, 2002. Obara, K., Nonvolcanic deep tremor associated with subduction in southwest Japan, Science, 296, 1679 – 1681, 2002. Peacock, S. M., and K. Wang, Seismic consequences of warm versus cool subduction metamorphism: Examples from southwest and northeast Japan, Science, 286, 937 – 939, 1999. Seno, T., T. Skurai, and S. Stien, Can the Okhotsk plate be discriminated from the North American plate?, J. Geophys. Res., 101, 11,305 – 11,315, 1996. Taylor, B., A. Klaus, G. R. Brown, G. F. Moore, Y. Okamura, and F. Murakami, Structural development of Sumisu rift, Izu-Bonin arc, J. Geophys. Res., 96, 16,113 – 16,129, 1991.
M. Ando, N. Fujii, K. Tadokoro, and K. Yamaoka, Research Center for Seismology and Volcanology, Graduate School of Science, Nagoya University, Furo-cho, Chikusa, Nagoya, Aichi 464-8602, Japan. (ando@ seis.nagoya-u.ac.jp;
[email protected];
[email protected];
[email protected]) M. Ishida, A.-S. Jin, and K. Obara, National Research Institute for Earth Science and Disaster Prevention, TennoDai 3-1, Tsukuba, Ibaraki, Tsukuba 305-0006, Japan. (
[email protected];
[email protected]; obara@bosai. go.jp) C.-H. Lin, Institute of Earth Sciences, Academia Sinica, P.O. Box 1-55, Nankang, Taipei, Taiwan. (
[email protected])
