SEISMICITY AND BLOCK STRUCTURE OF THE CENTRAL
AND SOUTHEAST ASIA

1YU.G. GATINSKY, 2T.V. PROKHOROVA, 1D.V. RUNDQUIST, 2G.L. VLADOVA

1Vernadsky State Geological Museum, Mokhovaya ul. 11, korp. 2, Moskva, 125009,
 
e-mail: yug@sgm.ru;  2International Institute of Earthquake Forecast and Mathematic Geophysics, Profsoyuznaya ul. 84/32, Moskva, 117997, e-mail: vladova@mitp.ru

Abstract: The intensity of seismicity and density of epicenter distribution were analyzed for Central and SE Asia east of 65° E.L. within the former Eurasian Plate. This analysis allows to distinguish some belts and local areas of the stable development of high seismicity (≥ 1 event par 5 years). They have been distinguished as along active margins and collision zones at boundaries with the Pacific, Australian and Indian plates as in some inner parts of the continent (a “triangle” of seismic activity between the Baikal Lake and both syntaxes of Himalayas). Side by side with the consistent dissipation of the seismic energy in front of the Indian Plate, a local increase of it has been established withdrawing from the plate boundary (Tien Shan, Bayanhar). Data on active faults allow to make more precise boundaries of the earlier established North Eurasian Plate and some transit zones between it and adjacent lithosphere plates, including East and Central Asian zones. Boundaries of more than 40 crust and mantle-crust blocks were corrected inside these zones, among them, the Amurian and Japan-Korean blocks. The investigation of the boundaries has been based on the earthquake mechanism solution in epicenters along them, the quantity of the seismic energy, which escape in 50-km bands on both sides of boundaries and vectors of absolute (ITRF) and relative horizontal displacements of blocks after space-geodetic data. Noticeable deformations of blocks have been revealed after vectors analysis in transit zones near their boundaries with the Pacific and Indian plates. Vectors of block horizontal displacements don’t often coincide with vectors of main plates and change together with the change of block rigidity. As a result of the analysis, a tentative electronic scheme of up-to-date block structure is elaborated for Central and SE Eurasia based on a complex of geologic-geophysical and space-geodetic data.


I. INTRODUCTION

According to the plate tectonics conception, the Earth’s lithosphere is divided into some main rigid plates, which displace over the more plastic asthenosphere. The well-known model NNR-NUVEL-1A calculated by plate movements above hot spots and by linear magnetic anomalies describes their displacement. At the same time, zones of maximal density of earthquake epicenters (≥ 1 event by 5 years) are distributed not only along the plate boundaries, but often penetrate far from them into inner parts of plates. The central and eastern parts of the Eurasian Plate are an excellent example of such epicenter distribution (Fig. 1). The analysis of seismicity makes it possible to establish an independent North Eurasian Plate and some blocks situated south and east of it, which are limited by seismoactive faults [4]. The GPS vectors of blocks often don’t coincide with vectors of the main plates.


Figure 1. Scheme of seismicity and modern block structure of Central and SE Asia. Epicenters of earthquakes with magnitudes of 8.0-8.9 - large gray, 7.0-7.9 - black, 5.0-6.9 - small gray points. One can see a “triangle” of the maximal seismicity in Central Asia northeast from the Indian Plate. Boundaries of lithosphere plates - black (dashed black - supposed), transit zones - dashed gray, blocks - gray. Arrows correspond to vectors of horizontal displacement: gray - GPS2005 in ITRF system, black - with respect to the stable Eurasian Plate. Blocks: 1) Sayan; 2) Altai; 3) West Mongolia; 4) Hangay; 5) Ebinur; 6) Junggar; 7) south Gobi; 8) Kuril - east Kamchatka; 9) Tien Shan; 10) Beishan; 11) Pamir; 12) West Tarim; 13) Quidam; 14) Qilian; 15) Jartai; 16) North Japan; 17) Afghan; 18) Punjab; 19) Himalaya; 20) North Tibet; 21) South Tibet; 22) Bayanhar; 23) Kam-Yunnan; 24) Ryukyu - Central Honshu; 25) Izu-Bonin; 26) Andaman - West Myanmar; 27) Shan; 28) North Luzon; 29) West Marianna; 30) East Marianna; 31) West Philippines; 32) East Philippines; 33) Sulawesi Sea; 34) Halmahera; 35) Caroline; 36) Mentawai; 37) North Sulawesi: 38) Banda Sea; 39) Sula “spur”; 40) Sorong; 41) Manus; 42) Timor; 43) Central Papua; 44) Bismarck Archipelago; 45) Murua (Woodlark), 46) Solomon Sea. LS - Longmen Shan Fault; Q - Qinlin Zone; T - Tanlu Fault.


The unhomogeneity of lithosphere plate margins is established by studying on seismicity and active faults. Some “transit zones” [5] or, after other researchers, “diffuse boundaries” [1, 6] are developed along these margins. Such zones divide large lithosphere plates and provide transfer and relaxation of tectonic stresses that arise between these plates during their interactions. In Central and SE Asia, they are the Central Asian, East Asian, and some smaller zones, which consist of numerous crust or crust-mantle blocks (see Fig. 1). In this paper, we’ll examined the seismicity, seismic energy and up-to-date block mobility within the transit zones. Our analysis is fulfilled by using databases on earthquake distribution, their intensity and mechanism solution [http://earthquake.usgs.gov/regional/neic], active faults [14], space-geodetic measure-ments in the ITRF system [http://itrf.ensg.ign.fr/ITRF_solutions/2005/ITRF2005.php], and some other geologic-geophysical data contained in the Electronic Geodynamic Globe, which was created in SGM RAS with participation of authors in 1995-2005 years (http://earth.jscc.ru).

II. SEISMICITY AND BLOCK DEFORMATION IN TRANSIT ZONES

The Central Asian transit zone is situated between the Indian and North Eurasian plates and coincides with a “triangle” of the maximal seismicity of Central Asia (see Fig.1). The vertex of the “triangle” is in the south end of the Baikal Lake and its base lies between both syntaxes of the Himalayas. The East Asian zone spreads all over the most active margins of Eurasia in boundaries with Pacific, Philippines, Australian and Indian plates as well as over adjacent parts of the continent. Such large blocks as Tarim, North Tibet, Amurian and Japanese-Korean ones, SE China, Indochina-Sunda and some others are included in these zones. Abnormal high seismicity, numerous active faults and widespread GPS stations characterize the examined territory. The Russian-Chinese geologic-geophysical transect GGT 21 runs across it from Altai up to Taiwan [16] and complex geophysical investigations were fulfilled in the Himalayas and Tibet by the International Project INDEPTH [9, 13].

The analysis of mechanism solutions in block boundaries has been making it possible to precise the type and direction of displacement along them and together with space-geodetic data often reveals the relative rigidity of the majority of blocks. For example, the Tien Shan Block (9 – here and further, numbers of blocks are given in brackets in Fig. 1) is limited by left-lateral wrench faults along its boundary with the North Eurasian Plate besides predominating thrusts. As concern the Tarim Block, the compression predominates in its western boundary together with right-lateral wrench faults, which are also developed in the northeastern boundary with the Beishan Block (10). Block margins undergo a deformation by the action of transpressive stress from subduction and collision zones. It is well catching by analysis of space-geodetic data. Thus, the modern horizontal movement of the Tien Shan and Tarim blocks cannot be regarded as the motion of absolutely rigid bodies, because a difference between the measured and model velocities goes beyond the error ellipse at the majority GPS stations within them [5]. In all probability, this is caused by the location of blocks in the periphery of the collision zone with intensive deformation processes between Hindustan and Eurasia.

The depth of hypocenters in the southern boundary of the Pamir Block (11) reaches the maximal value for Central Asia (160-240 km). They correspond to a north-dipping slab of the Indian lithosphere plate. The mechanism solutions show the predominance of the compression together with local left-lateral wrench faults (Fig. 2). The hypocenters are shallower (up to 40-80 km) in the northern boundary of Pamir with the North Eurasian Plate, where the compression also predominates, but with the south-dipping Benioff Zone. The analysis of seismic tomography data together with rheology modelling [10] has been showing a faster nearly vertical subduction of the Indian slab in the south and slower less sloping subduction of the Eurasian slab in the north.


Figure 2. Mechanical solutions for South Pamir, Himalayas and Tibet.
 Black - plate boundaries, gray - block boundaries.


More differentiated picture of seismicity can be seen in the Himalayas and Tibet. Local thrusts characterize the boundary of the Indian Plate and the Punjab Block (18) besides predominating left-lateral wrench-faults. The compression prevails inside the Himalayas and at the boundary of this block with the Indian Plate, but at the same time right-lateral wrench faults dominate along boundaries of North (20) and South Tibet (21). The mechanical solutions have been showing the local extension in central parts of Tibet within narrow and short submeridional zones, which correspond to up-to-date rifts connected with “crawling off” the mountain ensemble in the latitude direction due to collision processes. GPS vectors with respect to the stable Eurasian Plate have been confirming Tibet crawling off. They form a characteristic divergence with a western deflection (10º NE - 345º NW) near the western syntax in NW Tibet and Tien Shan and eastern deflection up to 50-70º NE and nearer the eastern syntax in SE Tibet, Quidam and Sichuan (see Fig. 1). Such divergence and Tibet crawling off are perhaps connected with the move aside of the crust materials in front of the Indian indenter [11], including a possible influence of stress from the relatively rigid Tarim Block. Some geologists explain the vectors divergence by a slab tear model, in which the Indian lithosphere splitted into two slabs: a northward-moving slab subducting steeply beneath the western sector of the Tibet Plateau, and a northeastward-moving slab subducting at a low angle beneath the eastern sector of the Plateau and the Three Rivers region [15].

A GPS vectors’ analysis has been permitting also to make more precise the boundaries of different tectonic elements, for instance between Amurian and Japanese-Korean blocks. Coinciding model velocities in the majority of GPS stations of earlier distinguished Amurian, Ordos and north China blocks proves a small displacement between these blocks [5]. So, they were amalgamated in the single Amurian Block withdrawing its southern boundary along the Qinlin Zone (see Fig. 1). This supposition concerns only the up-to-date geokinematic field, since some Holocene displacements took place along faults, dividing above-mentioned blocks. By the way, an active zone between former Amurian and North China blocks with a velocity of extension of about 2.4 mm/yr [7] can be a local intrablock deformation. The weak modern activity inside the single Amurian Block is possibly connected with its position on the western wing of the large right-lateral Tanlu fault, which divides this block and the Japanese-Korean one.

Some other smaller transit zones have been established in East and SE Asia. The North Pacific Zone is situated between Pacific and North American plates and includes Okhotsk, Kuril - East Kamchatka and more northern blocks. The Marianna Zone divides the Pacific and Philippines plates; the Melanesian Zone is situated between the Pacific and Australian plates. The examination of these zones is out of frame of this paper.

III. SEISMIC ENERGY RELEASED WITHIN TRANSIT ZONES

The block boundaries are often represented in transit zones not only by single faults, but also by relatively wide interblock zones. They are characterized by an intensive shattering of rocks together with releasing a significant quantity of seismic energy, and so, can be regarded as seismically dangerous. The depth of hypocenters within them is mainly of 20-40 km that proves the non-dip penetration of these zones in the lithosphere. Much rarely, it can reach 80-240 km (Pamir). One can see in such zones a certain analog of transit zones between main plates that reflects a fractal structure of the continental lithosphere.

The total volume of seismic energy within the Central Asian transit zone is diminished away from the northern boundary of the Indian Plate. The diminishing rate is correlated with the decrease of the deviation modules of experimental vectors from the vectors calculated by the NNR-NUVEL-1A model [5]. So, on the whole a dissipation of energy is directly proportional to the block mobility decrease. But, sometimes the maximal quantity of energy releases in inner parts of the transit zone at the distance 500-1500 km from the plate boundary. The most active interblock zones limited the following blocks: Pamir (11), Tien Shan (9), and Bayanhar (22). A volume of the seismic energy released along each of them reaches 5·1022 erg, while along other boundaries it doesn’t exceed 3·1019 - 2·1022 erg. Making this calculation we took 50-km bands on both sides of boundaries.

The same interblock zones are characterized by a maximal specific energy by 1 km of their length (> 4.5·1019 erg) and by a maximal deviation of GPS vectors from average vector values on the main plates. The eastern boundary of the Bayanhar Block coincides with a global lineament of 102-103º E.L. [4], which passes here along the Longmen Shan fault. A steep step corresponds to it in the crust and whole lithosphere with their thickness diminishing to the east (Fig. 3). It is worthy to note that a difference in the lithosphere mantle and crust interrelation is approximately assumed in both sides of the lineament. There is strong coupling between crust and mantle to the west beneath Tibet, but a complete decoupling between crust and mantle east beneath the Yunnan crust. The dynamic model of the mantle shows that the Yunnan and Indochina crust is moving southward with respect to the mantle at rates as high as ~30 mm/yr, while the mantle is displaced northeastward. And beneath Tibet both are moving northward [2, 12].

The total volume of the energy released along Bayanhar interblock zone (6.358-6.376·1023 erg) is only in 2.5 lesser than the energy of one of the most active North Japan subduction zone (15.332·1023 erg) and nearly equal to the total energy along the northern boundary of the Indian Plate ( 6.096·1023 erg). At the same time, it is by order greater than the energy of relatively weakly active subduction zones, for example, south Ryukyu (7.913·1022 erg). Therefore, the most active interblock zones of Central Asia differ from subduction and collision zones mainly by the depth of their penetration in the lithosphere and underlying upper mantle and are rather near to them by the realizing seismic energy volume.

Besides above-mentioned high-energy interblock zones it is necessary to name some others, the specific energy of which comes to 1.0-4.5·1019 erg. They limited Sayan (1), South Tibet (21), northern side of the Amurian Block and also can be regarded as potentially dangerous. It is worth to mention that in considered parts of Central and SE Asia, the majority of interblock zones goes in Russia, China, Myanmar, Viet Nam and other countries through regions of dense population, widely developed infrastructure and large mineral deposits. It emphasizes the applied significance of examination of problems connected with the geodynamics of these zones.


Figure 3. The southeastern part of the transect GGT 21 after the work [16] with some additions: 1) The upper part, and 2) The lower part of the upper crust; 3) Middle crust: 4) The upper, and 5) The lower part of the lower crust; 6) Thermal lithosphere (LID); 7) Electromagnet lithosphere; 8) Low velocity layers in the crust. Small figures on the transect profile correspond to velocity values of S-waves. The horizontal scale is reduced to one to seventh of the vertical scale.


IV. DISCUSSION AND CONCLUSIONS


Geologic-geophysical data from different regions show that, the boundaries of the large lithosphere plates nowhere are simple contacts and almost always are complex multilevel zones. Geophysical soundings including the dip seismic tomography allow to suppose a wide diapason of the depth of block mechanical soles in transit zones. Such blocks as Junggar (6), Tien Shan (9) and some others have undoubtedly the crust nature and are not distinguished on the lithosphere level as it can be seen in the southeastern part of the transect GGT 21 (see Fig. 3). Just so, 95 % of earthquake hypocenters lie in Tien Shan within upper 20 km of the crust. At the same time Tarim, Quidam (13), Bayanhar (22) and especially SE China Block have distinct thick lithosphere roots. The same roots (170-200 km) are also established for central parts of the Amurian Block [8]. It is worthy to note that just these blocks are characterized by the higher rigidity and a relatively weak deformation.

That would be the lithosphere substance in transit zones is on different levels at specific quasi-plastic conditions. As results of magnetotelluric sounding in the Himalayas and Tibet show, some hypothetical partly molten horizons are situated in the depth 20-25 km [9, 13]. According to seismic tomography data low velocity layers are established in some sectors under Tarim and Bayanhar at the depth 20-30 km and under Junggar and Tien Shan at 35-45 km [16]. Under Quidam, the same layers are situated on both mentioned levels (see Fig. 3). They also most likely correspond to partly molten horizons.

The examination of transit zones in Central and SE Asia shows abnormally high release of the seismic energy within relatively narrow interblock zones withdrawn from active subduction and collision zones. We suppose that it can be connected with a different rigidity and a specific dip structure of some blocks, which result in their non-equivalent reaction at tectonic strains. Blocks with the thickest lithosphere roots are the most rigid and weakly deformed (SE China, Amurian Block). This problem requires the further examination as well as a closer definition of levels, on which the block mechanical soles are settled down in transit zones.

ACKNOWLEDGEMENTS

This work is fulfilled with the assistance of the Earth Sciences Department RAS (Program No. 6 “Geodynamics and deformation mechanism of lithosphere”) and RFBR (grant No. 06-05-64866). Authors are grateful to Prof. Cao Dinh Trieu, Dr. T.V. Romanyuk, Dr. V.A. Sankov, and Prof. G.A. Sobolev for kind useful remarks and advices during the preparation of the paper.

REFERENCES

1. Bird P., 2003. An updated digital model of plate boundaries. Geochem., Geophys., Geosystems, 4/3: 1027; doi: 10.1029/2001GC000252.

2. Flesch L.M., Holt W.E., Silver P.G. et al., 2005. Constraining the extent of crust-mantle coupling in Central Asia using GPS, geologic, and shear wave splitting data. Earth and Planet. Sci. Letters, 238: 248-268.

3. Gatinsky Yu.G., Rundquist D.V., 2004. Geodynamics of Eurasia: Plate tectonics and block tectonics. Geotectonics, 38/1: 1-16.

4. Gatinsky Yu.G., Rundquist D.V., Cherkasov S.V., 2004. Trans-Eurasian divider: Structural and metallogenic evidences. 32-IGC Abst., 1/136-13: 620. Florence.

5. Gatinsky Yu.G., Rundquist D.V., Tyupkin Yu.G., 2005. Block structure and kinematics of Eastern and Central Asia from GPS Data. Geotectonics, 39/5: 333-348.

6. Gordon R.G., 1998. The plate tectonic approximation: Plate nonrigidity, diffuse plate boundaries and global plate motions. Ann. Rev. Earth and Planet. Sci., 26: 615-642.

7. Jin S., Park P.-H., Zhu W., 2007. Micro-plate tectonics and kinematics in Northeast Asia inferred from a dense set of GPS observations. Earth and Planet. Sci. Letters, 257/3-4: 486-496.

8. Levi. K.G., 2008. Map of neotectonic motions. Irkutsk Institute of the Earth Crust, Siberian Branch of RAS (in Russian).

9. Li S., Unsworth M.J., Booker J.R. et al., 2003. Partial melt or aqueous fluid in the mid-crust of Southern Tibet: Constraints from INDEPTH magnetotelluric data. Geophys. J. Intern., 153/2: 289-304.

10. Negredo A.M., Replumaz A., Villaseñor A., Guillot S., 2007. Modelling the evolution of continental subduction processes in the Pamir-HinduKush region. Earth and Planet. Sci. Letters,  259/1-2: 212-225.

11. Shen Z., Zhao C., Yin A. et al., 2000. Contemporary crustal deformation in East Asia constrained by Global Positioning System measurements. J. Geophys. Res.  105/B3: 5721-5734.

12. Sol S., Meltze A., Bürgmann R. et al., 2007. Geodynamics of the southeastern Tibetan Plateau from seismic anisotropy and geodesy. Geology, 35/6: 563-566; doi: 10.1130/G23408A.1.

13. Solon K.D., Jones A.G., Nelson K.D. et al., 2005. Structure of the crust in the vicinity of the Banggong-Nujiang suture in central Tibet from INDEPTH magneto-telluric data. J. Geophys. Res., 110/B10: 102; doi:10.1029/2003JB002405.

14. Trifonov V.G., Soboleva O.V., Trifonov R.V., and Vostrikov G.A., 2002. Recent geodynamics of the Alpine-Himalayan Collision Belt. Trans. Geol. Inst. RAS, 541, 225 pp. (in Russian).

15. Xiao L., Wang C., Pirajno F., 2007. Is the underthrust Indian lithosphere split beneath the Tibetan Plateau? Intern. Geol. Rev., 49: 90-98.

16. Yuan X., Egorov A.S., and GEMOC, 2000. A short introduction to Global Geoscience Transect 21: Arctic Ocean - Eurasia - Pacific Ocean. Sci. Press, 32 pp.