Документ взят из кэша поисковой машины. Адрес оригинального документа : http://dynamo.geol.msu.ru/personal/vsz/papers/Zakharov_etal_2013_eng.pdf
Дата изменения: Wed Mar 27 13:11:58 2013
Дата индексирования: Thu Feb 27 22:15:37 2014
Кодировка:

ISSN 0145 8752, Moscow University Geology Bulletin, 2013, Vol. 68, No. 1, pp. 1016. © Allerton Press, Inc., 2013. Original Russian Text © V.S. Zakharov, A.I. Karpenko, S.P. Zav'yalov, 2013, published in Vestnik Moskovskogo Universiteta. Geologiya, 2013, No. 1, pp. 1118.

Seismic Nails in Various Geodynamic Conditions
V. S. Zakharov, A. I. Karpenko, and S. P. Zav'yalov
Faculty of Geology, Moscow State University, Moscow, 119899 Russia e mail: vszakharov@yandex.ru, AlexandroKarpenko@yandex.ru, zasergey@rambler.ru
Received May 25, 2012

Abstract--Isometric in plan, near vertical earthquake clusters (seismic "nails") were identified in different regions of the world. Seismic nails are 1090 km in vertical length; most earthquakes that make them up are weak and their formation time is 1060 days. Some nails are related to strong earthquakes and volcanic erup tions. Many seismic nails are not evidently related to fault zones and other tectonic structures. The Hurst exponent (H > 0.5) indicates the persistency in the sequence of earthquake depths. Keywords: subvertical clusters of earthquake hypocenters, seismic nails, volcanoes, faults, fluids, and Hurst exponent DOI: 10.3103/S0145875213010080

INTRODUCTION V.N. Vadkovsky (1996, 2012), studied the spatial distribution of earthquake hypocenters in the Japanese Islands according to the JMA Earthquake Catalog for 19831990 and identified almost vertical short lived clusters of earthquake hypocenters that were isometric in plan called seismic "nails," which are made up largely of weak earthquake foci (M = 23). A nail is 1050 km in vertical length and is formed at the depth of 90 km at most. The nail's epicentral projection is 5 20 km in diameter. Subvertical zones with earthquake foci were also noted in other regions (Alaska and the Banda Sea). The objective of this investigation was to identify and to analyze seismic nails and similar clusters of earthquake hypocenters in various regions of the world and to study their relationship to seismic process and geodynamics. SUBVERTICAL CLUSTERS OF EARTHQUAKE FOCUSES The identification of seismofocal structures, such as seismic nails, is rarely considered in the scientific literature. A subvertical cluster of hypocenters that were related to the November 29, 1994 earthquake (MW = 5.1) was identified near Lefkada Island (Ionian Sea) (Papadimitriou et al., 2006). The nail that formed 3 weeks prior was 012 km in depth. The focal mech anisms are consistent with the direction of the observed coseismic fractures on the surface. This fact, in our opinion, is indicative of its relationship to faults. Wang and Zhao (2006) studied the distribution of aftershocks of the strong Fukuoka earthquake that occurred on March 20, 2005 (MJMA = 7.0). Hypo centers that are distributed irregularly along the fault
10

are grouped in structures such as seismic nails. The foci occur at depths of up to 20 km . Researchers sug gest a considerable role of fluids in the trigger genera tion of a main event. Padhy et al. (2011) provided data on the after shocks of the Hoto Hanto earthquake that occurred on March 25, 2007 (MW = 6.6) in Japan. The hypo centers were distributed in the form of a nail; the foci depth was up to 13 km and a possible role of fluids in earthquake generation was suggested. Kilburn (2003) described cylindrical clusters of hypocenters that are confined to volcanoes and that formed before an eruption. A cylinder is a few kilome ters in diameter, up to 57 km in depth, and has mag nitudes of 02 and 56. McNutt (1996) provided data on similar clusters that are related to the eruption of the Mount St. Helens Volcano and other. The focal mechanism implies the multiscale destruction and integration of fractures. Yukutake et al (2010) described structures like nails confined to the Hakone Volcano. Their epicentral projections were 1 km in length at most and up to 6 km in depth. Shevchenko et al. (2011) identified a column shaped near vertical cluster of earthquake foci in the center of the Garm geodynamic site (Tajikistan). Clusters of aftershock focuses that were similar in morphology and position in the earth's crust were identified for the Altai (2003), Neftegorsk (1996), Kultuk (2008), and Dagestan earthquakes. Mean while, there is no relationship of clusters to any tec tonic elements of the indicated regions; thus, their relationship to the inflow of deep fluids has been sug gested.

SEISMIC NAILS IN VARIOUS GEODYNAMIC CONDITIONS

11

MATERIALS AND METHODS We used the following earthquake catalogs: the National Earthquake Information CenterPreliminary Determination of Epicenters Catalog for 19732011 (NEIC PDE, http://earthquake.usgs.gov/regional/ neic/); Japanese Meteorological Agency Catalog (JMA); catalogs of the Southern California Earthquake Data Center (SCEDC, http://www.data.scec.org/) and Northern California Earthquake Data Center (NCEDC, http://quake.geo.berkeley.edu/) 1967 2010; as well as the catalog of the Geophysical Survey of the Kamchatka Branch of the Russian Academy of Sciences for 19622011 (http://data.emsd.iks.ru). As a fine seismic structure is the object of our inves tigation, it is essential to provide a high accuracy of the earthquake coordinates. The error of the coordinates attains 510 km for the PDE and Kamchatka catalogs, 1 km at most for the JMA catalog, and < 3 km for the SCEDC and NCEDC catalogs (commonly, < 1). It should be noted that due to the fact that different cat alogs use scales of different magnitudes, we indicate their type when needed. As well, data on the boundaries of tectonic plates (Bird, 2003), maps of active fault in Eurasia (Trifonov, 2004), the Quaternary Fault and Fold Database of the United States (http://earthquake.usgs.gov.hazards. qfaults/), and the map of the volcanoes of the Global Volcanism Program (GVP, http://www.volcano.si.edu/ index.cfm) were used to compare seismic structures with tectonic and geological structures. As a study of earthquake source processes was not the objective of the investigation, we did not take the source dimensions into account and regarded them as a point with specified spatial coordinates and magni tude (the spatial extension of the source, which is dependent on earthquake force and focal mechanism, is left for future studies). To identify seismic nails, special software that makes it possible to search for them in the catalog was created. The searching algorithm is based on the fol lowing criteria: a short period of cluster formation (1090 days), an isometric form in plan and small size (150 km); the minimum number of earthquakes in a nail is 100200; and a nail is two time longer vertically as horizontally. The indicated software carries out the preliminary searching for nail candidates, while final identification is performed manually. The Hurst exponent was calculated to study the formation of earthquake clusters in time. The Hurst exponent was calculated by R/S analysis (Lukk et al., 1996; Turcotte, 1997), where R is the range of time series, i.e., the difference between the highest and low est cumulative deviation from the current average (in the given time interval ) and S is the standard devia tion of the time range in the same interval. The follow ing relationship between R/S and was established for many natural processes: R/S ~ H, where H is Hurst exponent.
MOSCOW UNIVERSITY GEOLOGY BULLETIN Vol. 68

An absolutely random (stochastic) process is char acterized by H = 0.5. H differs from 0.5 if a process has a so called memory. At H > 0.5 time relationships are stable, i.e., if increments a positive, they are the same in the future (persistency). At H < 0.5, an increase in the observed value is replaced by a decrease and vice versa (antipersistency). The analysis was carried out using the author's FraTiS software. IDENTIFIED SEISMIC NAILS The analysis made it possible to identify some structures in different regions of the world that can be referred to as seismic nails in their configuration (Fig. 1). These clusters are characterized by the following fea tures. (1) Mount St. Helens Volcano (Cascadia), 46.2° N and 122.2° W (Fig. 2). The formation time was MarchMay 1980, the focus depth was 015 km, and the position is confined to the volcanic arc of the sub duction zone. The cluster was formed before the cata strophic eruption of the Mount St. Helens Volcano started on May 18, 1980. Four events, which were dis tributed relatively regularly in time, are characterized by mb = 5. (2) Colo Volcano (Tomini Bay, Sulawesi Island), 0.3° N and 121.8° E (Fig. 3). The formation time was July 1983 and the focus depth was 2090 km. The nail was surrounded by plates with a complicated shape that are composed of convergent, divergent, and trans form zones. The position is confined to the volcanic arc of the subduction zone. The nail was formed almost simultaneously with the eruption of the Colo Volcano (Una Una), which started on July 18, 1993. It should be noted that in spite of the probable confine ment of the nail formation to the eruption, the earth quake foci occurred at great depth. We noted many events that were distributed relatively regularly in time that were characterized by mb 5. (3) Karymsky Volcano (Kamchatka), 53.9° N and 159.4° E. The formation time was January 1996, the foci depth was 060 km, and the position is confined to the volcanic arc of the Kamchatka subduction zone. This cluster, which implies the aftershocks of the strong Karymsky earthquake, occurred on January 1 1996 (ML = 6.4, MS = 7) and is most likely related to the following eruptions of the Karymsky and Academy of Science volcanoes. The nail formation was accom panied by many events with ML > 5. (4) Southern Kamchatka, 52.3° N and 157.9° E (Fig. 4). The formation time was March 1983, the foci depth was 020 km, and the position is confined to the volcanic arc of the Kamchatka subduction zone. The relationship between the nail's epicentral projection and faults is not clear. Dormant volcanoes are located nearby. The nail formation was accompanied by an earthquake with ML = 5.3, which corresponds almost to the midpoint of the nail formation interval, but occurred at some distance from the nail body. Earth quakes had commonly occurred at depths of 010 km
No. 1 2013

12 70° 60° 50° 6 40° 30° 20° 10° 2 0° 10° 110° 7 8 3 4

ZAKHAROV et al. 70° 60° 5 1 10 9 30° 20° 10° 0° 10° 180° 170° 160° 150° 140° 130° 120° 110° 50° 40°

120°

130°

140°

150°

160°

170°

Fig. 1. Seismic nails that were identified in the course of our investigation: (110) the numbers of nails in the order they were described; the lines denote plate boundaries (Bird, 2003); triangles denote volcanoes.

(a) 0 0 2 4 6 8 10 12 14 16 18 46.5 2 4 6 Depth, km 8 10 12 14

(b)

Depth, km

122.0 16 122.1 46.4 122.2 18 46.3 122.3 itude Lati 46.2 122.4 ong tude 46.1 20 L 46.0 122.5 20.03 30.03 09.04 19.04 29.04 09.05 19.05 Date

Fig. 2. Seismofocal structure related to the Mount St. Helens Volcano eruption in May 1980: (a) 3D view of nail and epicentral projections of foci; (b) sequence of variations in earthquake focal depth in time under nail formation, the triangle denotes the start of the catastrophic eruption that occurred on May 18, 1980. The black circles indicate earthquakes with mb = 5, while the white circles indicate earthquakes with mb < 5.

prior to this event. After it, the foci occurred at a depth of 020 km. We also were able to identify subvertical seismofo cal clusters in the region of the Olyutor earthquake (Kamchatka) that occurred on April 21, 2006 (MS = 7.7), but at some distance from the focus of the main

event. However, these clusters consist of a relatively small number of hypocenters and their fitting to seis mic nails requires further studies. (5) Kodiak Island (Alaska), 57.3° N and 154.3° W. The formation time was December 1999, reactivation occurred on July 2000, and the focus depth was 0
Vol. 68 No. 1 2013

MOSCOW UNIVERSITY GEOLOGY BULLETIN

SEISMIC NAILS IN VARIOUS GEODYNAMIC CONDITIONS (a) 0 0 20 Depth, km Depth, km 40 60 80
2 1 Latit 0 ude 1

13

(b)

20 40 60 80 100 15.07

12 124 12 122. 2.5 .5 11 120.5 1.5 5 tude 9.5 gi Lon

19.07

23.07

27.07

31.07

04.08 Date

Fig. 3. The seismofocal structure that is related to the Colo Volcano (Tomini Bay, Sulawesi) eruption in May 1980: (a) 3D view of the nail, epicentral projections of foci, position of volcanoes (triangles), and coast line contours (black lines); (b) sequence of variations in earthquake foci depth in time during nail formation; a triangle denotes the start of the eruption that occurred on July 18, 1983. The black circles indicate earthquakes with mb 5, while the white circles indicate earthquakes with mb < 5.

(a) 10 0 10 20
Depth, km

(b)

0 10 Depth, km
15 82.6 15 158.4 8.6 Lat 82.4 15 158.0 8.2 itud 82.2 7 1 e e 15 57.6 .8 itud 82.0 7.4 ong L

20 30 40

30 40 82.8

50 60 05.03 15.03 25.04 04.04 14.04 24.05 04.05 Date

Fig. 4. A seismic nail in Southern Kamchatka that formed in March 1983: (a) a 3D view of the nail, epicentral projections, posi tion of volcanoes (triangles), coast line contours (black lines), and active faults (grey lines); (b) sequence of variations in earth quake foci depth in time under the nail formation. The black circles indicate earthquakes with ML = 5.3, while the white circles indicate earthquakes with ML < 5.

70 km. The nail was located in the accretionary prism of the Aleutian subduction zone in front of the volcanic arc. Earthquakes with mb = 6.8 (66 km in depth) and then with mb = 6.5 (40 km in depth) occurred at the beginning of the first activation, while an earthquake with mb = 6.3 (43 km in depth) occurred as the reactivation started. Hence, the formation of this nail is most likely related to the indicated strong earthquakes. (6) Hokkaido Island, 43.5° N and 143.0° E. The formation time was JanuaryMarch 1989, the foci
MOSCOW UNIVERSITY GEOLOGY BULLETIN Vol. 68

depth was 040 km, the position was the volcanic arc of the subduction zone; dormant volcanoes occur nearby. The epicenters are not evidently related to faults. The nail body consisted of weak earthquakes with MJMA < 5 (Vadkovsky, 2012). (7) South of the Honshu Island (in front of the Nankai Channel), 33.2° N and 138.6° E (Fig. 5). The formation time was SeptemberOctober 1990 and the foci depth was 3080 km. The epicentral field was located near the sea mount. The initial formation was
No. 1 2013

14

ZAKHAROV et al.

0 10 20 30 40 50 60 70 80 90

October

September October

137.8 138.6 Lati tude139.4

93.8 93.4 de 93.0 gitu on L 140.2 92.6

94.2

0 2 4 6 8 1992 2001 10 12 14 16 18 .2 36 .1 36 0 La 36. 9 titu de 35. 35.8

1984 1998

Depth, km

Depth, km

July, september 1995

1 17. 5 1 117. 6 17. 1 7e 117 17. d 8 gitu .9 7 118 Lon .0 35.

Fig. 5. Seismic nails south of Honshu Island that formed in SeptemberOctober, 1990: epicentral projections, posi tion of volcanoes (triangles), and plate boundaries (grey lines). The black circles indicate earthquakes with MJMA 6, while the white circles indicate earthquakes with MJMA < 6.

Fig. 6. Seismic nails in the Inyo district (California) formed in 1984, 1992, 1995, 1998, and 2001: epicentral projections, position of volcanoes (triangles), and Quater nary faults (grey lines). The black circles indicate earth quakes with ML = 5, while the white circles indicate earth quakes with ML < 5.

accompanied by an earthquake with MJMA = 6.6 (60 km in depth), which was followed by an earthquake with MJMA = 6.0 (42 km in depth) and other events with MJMA < 5. This cluster is a manifestation of the after shock process of the indicated strong earthquake. (8) South of Honshu Island (behind the volcanic arc of the Izu Bonin subduction zone), 33.7° N and 139.4° E (Fig. 5). The formation time was October 1990 and the foci depth was 035 km. The nail con sisted of weak earthquakes with MJMA 4.2. According to the results of a comparison with the Japanese tectonic scheme (Sagiya et al., 2000), the lat ter two seismofocal structures could be related to faults. (9) Inyo region (California), 35.7°36.2° N and 117.5°118° W (Fig. 6). Four seismic nails with a small foci depth and horizontal size of 2 km at most were identified: August 1984 (07 km in foci depth); February 1992 and reactivation on July 2001 (0 9 km); July 1995 and reactivation in September (0 16 km); all earthquakes of the nails were weak; on March 1998 (011 km), a nail was formed as an after shock of ML = 5. The confinement of the nails to a few local faults is not evident. (10) Long Valley caldera (California) 37.637.8° N and 118.8119° W (Fig. 6). The formation of four seismic nails with a foci depth of 010 km was accom panied by reactivation in 1984, 1990, 1991, and 1996 1998. All earthquakes were weak (ML = 4.8 at most). These clusters occur inside the caldera and are most likely related to hydrothermal activity. However, there are several faults in the caldera and thus the genesis question is still open.

THE FORMATION OF A SEISMIC NAIL IN TIME The main and common property of all the identi fied seismic nails is the short time of their formation (1060 days). The formation of some nails has been shown to be initiated by strong (M > 5) earthquakes and/or volcanic eruptions. In other cases, relatively strong earthquakes occur in the course of nail forma tion rather than at the beginning and likely represent the seismofocal structure development stage. About half of the considered subvertical focal structures con sist largely of weak earthquakes. Figure 7a demonstrates variations in the number of earthquakes per day (N) in MarchApril 1983 during nail formation in Southern Kamchatka (Fig. 4). Anal ysis of this distribution failed to identify an exponen tial decrease in the number of earthquakes in time. Seismic nail formation is an extinction process, but it is likely different from the extinction of strong earth quake aftershocks. We analyzed the time sequences of variations in the foci depth Z(t) for all the studied seismofocal struc tures and failed to distinguish any trends of events in time. However, analysis using the Hurst exponent makes it possible to establish some patterns. Figure 7b demonstrates the Hurst exponent value (H) that was calculated for the depth sequence of hypocenters that formed the above described nail in Southern Kam chatka in 1983; H = 0.68. Such an analysis was carried out for all the studied seismic nails. The H values were established to exceed 0.5 for analogous sequences, which occurred in the range of 0.570.69. In this case, H error does not exceed 0.05 and the correlation coef
Vol. 68 No. 1 2013

MOSCOW UNIVERSITY GEOLOGY BULLETIN

SEISMIC NAILS IN VARIOUS GEODYNAMIC CONDITIONS Number of earthquakes 50 45 40 35 30 25 20 15 10 5 0 07.03 15.03 23.03 31.03 08.04 16.04 21.04 11.03 19.03 27.03 04.04 12.04 20.04 28.04 0 0.5 1.0 1.5 2.0 2.5 log (t) 1.0 0.5 log (R/S) 2.0 1.5 (b)

15

(a)

log (R/S) = 0.68 log (t) + 0.04 R = 0.997

Fig. 7. Seismic nail formation in time in Southern Kamchatka, 1983: (a) variations in the number of earthquakes during nail for mation; (b) calculation of the Hurst exponent for the depth sequence of hypocenters that make up a nail, H = 0.68.

ficient of the linear regression is at least 0.98 (see the table). As mentioned above, these data are indicative of a stable trend in the studied sequences. RESULTS AND DISCUSSION We analyzed worldwide (NEIC PDE), Japanese (JMA), Californian (SCEDC, NCEDC), and Kam chatka earthquake catalogs and we identified near
Results of the investigation of seismic nails Location St. Helens Volcano Colo Volcano (Sulawesi) Karymsky Volcano Kodiak Island Honshu Island Inyo region Southern Kamchatka Hokkaido Island Honshu Island Inyo region Formation 0305.1980 07.1983 M
max

vertical earthquake clusters that were isometric in plan (seismic nails) in various regions of the world (the Pacific Sea coast of North America, Kamchatka, Jap anese Islands, and the Sulawesi Island). The nails are 1050 km (occasionally, up to 90 km) in vertical extension, and most of their earthquakes are weak (M = 4.55 at most). All discovered seismic nails are characterized by a short formation time (1060 days).

Depth

H

Additional data

Tectonic position Volcanic arc Volcanic arc Volcanic arc In front of a volcanic arc In front of a channel

Long Valley caldera

Related to volcano eruptions 5.0 015 0.63 Before eruption 5.8 2090 0.57 After eruption Follow strong earthquakes 01.1996 6.4 060 0.64 After eruption 12.1999, 07.2000 6.8 070 0.63 0910.1990 6.6 3080 0.62 Faults 03.1998 5.0 011 0.67 Contain strong earthquakes 03.1983 5.3 020 0.68 Not related to strong earthquakes 0103.1989 4.4 040 0.69 10.1990 4.2 035 0.69 Faults 08.1984 2.5 07 0.58 Faults 02.1992 4.0 09 0.69 07.1995 3.5 016 0.60 09.1995 4.0 016 0.62 07.2001 3.9 09 0.67 1984, 1990, 1991, 4.8 010 Hydrothermal activity 19961998
Vol. 68 No. 1 2013

Volcanic arc Volcanic arc Behind a volcanic arc

MOSCOW UNIVERSITY GEOLOGY BULLETIN

16

ZAKHAROV et al.

Some of these seismofocal structures are related to volcanic eruptions (they precede or follow them) and hydrothermal activity. The formation of some nails can be initiated by strong earthquakes (M > 5), while other nails contain such earthquakes in their body. Some seismic nails are not clearly related to fault zones and other tectonic faults. Their formation is likely initiated by fluid flow. The table contains the data on all the studied seis mofocal structures, the position and formation time of seismic nails, and their quantitative characteristics: foci depth, maximum magnitude (Mmax), and Hurst exponent (H). The nails are classified by their confine ment to volcanic eruptions and strong earthquakes; additional data on their tectonic positions are reported as well. CONCLUSIONS Vadkovsky (2012) analyzed the Markov properties of the foci depth sequences in seismic nails and made the conclusion that they were formed over the entire depth interval without preferable growth upwards or downwards. Our investigation allows us to update these data. Trends were not identified in the men tioned sequences, but the obtained data on the Hurst exponent are indicative of persistency in the order depth sequence of hypocenters that make up a nail. This sequence, being nonrandom, has a memory. This memory manifests itself in the fact that the current condition is controlled not only by current control variables, but also by previous characteristics. The memory of the natural processes (the Hurst effect) is explained by slow relaxation. Memory occurs when the reaction of a system to outer impact is not momen tary, but extended in time (Lukk et al., 1996). This specific feature can be of use in the study of seismic nail formation mechanisms. The geometry and formation rate of seismic nails suggest a possible role of fluids in the development process (Vadkovsky, 1996, 2012; Shevchenko et al., 2011). In this case, the memory effect can be caused by the relaxation of vis cous liquid in a porous environment, which is charac teristic for most tasks of unstable and nonlinear filtra tion. An alternative explanation implies the relaxation of accumulated tension in a fault (faults) as a series of weak earthquakes. However, surface manifestations of these faults can be absent, especially in the case of weak earthquakes that make up seismic nails. We hope that further detailed analysis of reliable catalogs, such as the JMA and California catalogs, will allow us to acquire additional data on the properties and nature of seismic nails. ACKNOWLEDGEMENTS We thank the World Data Center for Solid Earth Physics for providing us with the Californian and Jap

anese catalogs. We also thank N.A. Sergeeva, M.G. Lomize, N.V. Koronovsky, and D.A. Simonov for discussions and constructive critical remarks. REFERENCES
Bird, P., An Updated Digital Model of Plates Boundaries, Geochem. Geophys. Geosyst., 2003, vol. 4, no. 3, p. 1027. doi: 10.1029/2001GC000252 Kilburn, C.R.J., Multiscale Fracturing As a Key to Fore casting Volcanic Eruptions, J. Volcanol. Geotherm. Res., 2003, vol. 125, pp. 271289. Lukk, A.A., Deshcherevskii, A.V., Sidorin, A.Ya., and Sidorin, I.A., Variatsii geofizicheskikh polei kak proyav lenie determinirovannogo khaosa vo fraktal'noi srede (Variations in Geophysical Fields as a Manifestation of Determined Chaos in Fractal Medium), Moscow: OIFZ RAN, 1996. McNutt, S.R., Seismic Monitoring and Eruption Forecast ing of Volcanoes: a Review of the State of the Art and Case Histories, in Monitoring and Mitigation of Volcano Hazards, Berlin: Springer, 1996, pp. 99146. Miyazaki, S. and Tada, T., Continuous GPS Array and Present Day Crustal Deformation of Japan, Pure App. Geophys., 2000, vol. 157, pp. 23032322. Padhy, S., Mishra, O.P., Zhao, D., and Wei, W., Crustal Heterogeneity in the 2007 Noto Hanto Earthquake Area and Its Geodynamical Implications, Tectonophys ics, 2011, vol. 509, pp. 5568. Papadimitriou, P., Kaviris, G., and Makropoulos, K., The Mw = 6.3 2003 Lefkada Earthquake (Greece) and Induced Stress Transfer Changes, Tectonophysics, 2006, vol. 423, pp. 7382. Shevchenko, V.I., Aref'ev, S.S., and Lukk, A.A. Subvertical Clusters of Earthquake Hypocenters Unrelated to the Tectonic Structure of the Earth's Crust Izv., Phys. Solid Earth, 2011, vol. 47, no. 4, pp. 276298. Trifonov, V.G., Active Faults in Eurasia: General Remarks, Tectonophysics, 2004, vol. 380, nos 34, pp. 123130. Turcotte, D.L., Fractals and Chaos in Geology and Geophys ics, Cambridge: Cambridge Univ., 1997, 2nd ed. Vadkovskii, V.N., Nature and Mechanism of "Seismic Nails", Tez. dokl. Lomonosovskie chteniya (Abstr. Read ings in Memory of M.V. Lomonosov), Moscow, 1996, pp. 6364. Vadkovskii, V.N., "Seismic Nails": Subvertical Clusters of Earthquake Hypocenters, Vestn. ONZ RAN, 2012, vol. 4, no. Z1001. Wang, Z. and Zhao, D., Seismic Evidence for the Influence of Fluids on the 2005 West Off Fukuoka Prefecture Earthquake in Southwest Japan, Phys. Earth Planet. Inter., 2006, vol. 155, pp. 313324. Yukutake, Y. Tanada, T., Honda, R., Harada, M., Ito, H., and Yoshida, A., Fine Fracture Structures in the Geo thermal Region of Hakone Volcano, Revealed by Well Resolvecl Earthquake Hypocenters and Focal Mecha nisms, Tectonophysics, 2010, vol. 489, pp. 104118.
Vol. 68 No. 1 2013

MOSCOW UNIVERSITY GEOLOGY BULLETIN