Seismicity of the Baikal rift system from regional network observations

Journal of Asian Earth Sciences 62 (2013) 146–161

Contents lists available at SciVerse ScienceDirect

Journal of Asian Earth Sciences journal homepage: www.elsevier.com/locate/jseaes

Seismicity of the Baikal rift system from regional network observations N.A. Radziminovich a,⇑, N.A. Gileva b, V.I. Melnikova a, M.G. Ochkovskaya a,b a b

Institute of the Earth’s Crust SB RAS, Lermontov Street 128, Irkutsk 664033, Russia Baikal Division of the Geophysical Survey of SB RAS, Lermontov Street 128, Irkutsk 664033, Russia

a r t i c l e

i n f o

Article history: Available online 7 November 2012 Keywords: Baikal rift Regional seismic network Seismicity

a b s t r a c t In the paper we report the state-of-the-art of seismicity study in the Baikal rift system and the general results obtained. At present, the regional earthquake catalog for fifty years of the permanent instrumental observations consists of over 185,000 events. The spatial distribution of the epicenters, which either gather along well-delineated belts or in discrete swarms is considered in detail for different areas of the rift system. At the same time, the hypocenters are poorly constrained making it difficult to identify the fault geometry. Clustered events like aftershock sequences or earthquake swarms are typical patterns in the region; moreover, aftershocks of M P 4.7 earthquakes make up a quarter of the whole catalog. The maximum magnitude of earthquakes recorded instrumentally is MLH7.6 for a strike-slip event in the NE part of the Baikal rift system and MLH6.8 for a normal fault earthquake in the central part of the rift system (Lake Baikal basin). Predominant movement type is normal faulting on NE striking faults with a left lateral strike-slip component on W–E planes. In conclusion, some shortcomings of the seismic network and data processing are pointed out. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction The Baikal rift system (BRS) is an intracontinental rift consisting of a number Cenozoic basins linked by topographic highs. It extends for approximately 1500 km from northern Mongolia in the southwest to southern Yakutia in the northeast (Fig. 1). The structural setting of the rift is controlled by its position between the Precambrian Siberian craton and the Sayan-Baikal and Mongolia– Okhotsk mobile belts. According to Logatchev and Florensov (1978), Baikal rifting started in the Middle Eocene with the formation of the Southern Baikal basin that is the oldest and deepest basin of the BRS. The thickness of Cenozoic sediments is up to 7 km there (Hutchinson et al., 1992). Other basins have the sedimentary layers with thickness less than 4 km. The rift basins are structurally half grabens with steep northern or north-western sides bordered by master faults. A remarkable feature of the BRS is absence of the present day active volcanism. Late Cenozoic volcanic fields are located off the rift basins and their mountain shoulders. There are some discrepancies in the crust thickness estimates. For the rift basins they vary from 34–36 km (Gao et al., 2004; Puzyrev et al., 1973; Zorin et al., 2002) to 40–42 km (Suvorov et al., 2002; ten Brink and Taylor, 2002; Nielsen and Thybo, 2009). The depth to the Moho discontinuity beneath the Siberian platform ⇑ Corresponding author. Tel.: +7 964 2160520, +7 3952 426900; fax: +7 3952 426900. E-mail address: [email protected] (N.A. Radziminovich). 1367-9120/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jseaes.2012.10.029

and the Sayan-Baikal fold belt is considered to be 38–42 km and 45–50 km correspondingly. According to that, the estimates of the amount of crust thinning beneath the BRS range from a few km to 10 km. In the work by Nielsen and Thybo (2009) no thinning has been revealed at all. Contradictory results have been obtained also for the state of the upper mantle beneath the BRS. Some studies suggested low seismic velocities such as 7.6–7.8 km/s (Puzyrev et al., 1973; Gao et al., 2003; Achauer and Masson, 2002; Brazier and Nyblade, 2003; Tiberi et al., 2003; Zhao et al., 2006), whereas others evidenced them to be normal, i.e. 8.0–8.2 km/s (ten Brink and Taylor, 2002; Nielsen and Thybo, 2009). From a geodynamic point of view, the BRS separates the Eurasian plate and Amurian microplate diverging at 3–4 mm year 1 (Calais et al., 1998; Sankov et al., 2009). Recent geodynamic interpretations imply effects from the India-Asian collision, the Pacific plate subduction and coeval movement of the Amurian plate in a SE direction (Zonenshain and Savostin, 1981; Barth and Wenzel, 2010). The most striking evidence of active tectonic processes in the BRS is a high level of seismic activity. This level is confirmed both by the known historical and instrumentally recorded earthquakes, e.g. the 1862 M7.5 Tsagan earthquake at the eastern part of the Baikal Lake and the 1957 M7.6 Muya event in the Muya basin. Formation of the regional seismic network since 1960s has allowed obtaining the epicenter map and revealing some features of seismicity. It is of special interest due to the fact that the BRS is an example of the intracontinental rifts and, therefore, is a case study for intraplate seismicity.

147

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

expansion of the network was recognized, and already in 1961 it consisted of fifteen stations operating over the whole region including its northern part. Thus, the beginning of 1960s is considered as commencement of permanent instrumental seismological observations in the Baikal region. It should be noted that in this paper we take into consideration also the large earthquakes of the 1950s because they were recorded and processed already instrumentally. Review of the history of seismological observations in the BRS for that period can be found in the papers by Golenetskii (1977, 1990). The great role of A.A. Treskov should be emphasized here. A.A. Treskov was the head of the Irkutsk seismic station during 1926– 1963 and the chairman of the Siberian and Far East Seismological Commission. Besides the network development, A.A. Treskov contributed to the location methods, earthquake energy estimates, and study of the Earth’s crust and mantle structure. He initiated the first seismogeological investigations of large earthquakes including the Mondy and Muya events, and the 1957 M8.3 Gobi-Altai earthquake in Mongolia (Florensov and Solonenko, 1963). Research and educational activity of A.A. Treskov gives grounds to believe that he is an originator of the Siberian seismological school. With time the number of stations has been increasing up to 20– 24 and sometimes it increased due to temporary local network deployment. The stations were gradually reequipped with modern three-component short-period SKM-3 seismographs and long-period SKD ones widely used in the former USSR. Recording was done on photographic paper with time marks monitored from radio signals. On the whole, the time period of 60–90s was an important milestone in the regional network configuration. The current state of the network (its ISC code is BYKL) is represented by 23 stations controlling the region shown on Fig. 1. A new stage of instrumental observations in the BRS started in 1999 as a consequence of the conversion of analogous recording into digital one (Melnikova et al., 2010). The stations are instrumented with ‘‘Baikal-10, 11’’ equipment developed by the Siberian Division of the Geophysical Survey RAS. Each station has three (NS, EW, Z) high-sensitivity short-period channels of seismometers SM-3 and

In the paper, we present a brief review of results obtained by the regional seismic network. We consider its development, spatial seismicity distribution, seismic patterns, and present-day stress field derived from earthquake focal solutions. Some of these points have been already covered in a number of papers as well as in annual reports issued by Geophysical Survey of the Russian Academy of Sciences (e.g. Solonenko and Treskov, 1960; Misharina, 1961; Golenetskii, 1977, 1990; Solonenko and Solonenko, 1987; Kochetkov et al., 1987; Melnikova and Radziminovich, 1998; Deverchere et al., 1991, 2001; Radziminovich et al., 2005, 2006a, Radziminovich, 2010; Melnikova et al., 2007, 2010 and many others). Nevertheless, revising and summarizing data over the period of instrumental observations and for the whole BRS may be useful for researchers who deal with seismotectonic and geodynamic studies in the region and for officials responsible for the hazard mitigation as well. 2. Materials 2.1. Seismic network development The history of the instrumental earthquake observations in the Baikal region began as early as 1901 when the first seismic station was installed in Irkutsk. It was equipped by an Omori–Bosh pendulum and a Miln pendulum. During the following years similar stations were deployed in several other settlements in the southern Baikal area. Since 1912 new sensors with galvanometric recording developed by B.B. Golitsyn were put into operation that resulted in significant improvement of observation quality. Further improvements were connected with timing as with the development of radio transmission, time signals provided a more stable clock reference. However, a real advance came in the end of the fifties spurred on by the occurrence of several large (M > 6.5) earthquakes, namely the 1950 MLH7.0 or Mw6.9 Mondy earthquake, 1957 MLH7.6 or Mw7.2 Muya earthquake, and 1959 MLH6.8 Middle Baikal earthquake (Fig. 1), Table 1). The barest necessity of further

Table 1 Parameters of the BRS earthquakes with M P 5.6 since 1950. #

Earthquake name

DD/MM/YYY

HH:MM

Lat., N

Lon., E

M

Magnitude type

Reference for magnitude

STK

DIP

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Mondy Buteelin Muya Nyukzha Olekma Middle Baikal Muyakan Tas-Yuryakh Kyachta

04/04/1950 06/02/1957 27/06/1957 05/01/1958 14/09/1958 29/08/1959 11/11/1962 18/01/1967 13/05/1989 25/10/1989 27/12/1991 21/08/1994 29/06/1995 13/11/1995 25/02/1999 21/03/1999 21/03/1999 16/09/2003 10/11/2005 11/12/2205 27/08/2008

18:18 20:34 00:09 11:30 14:21 17:03 11:31 05:34 03:35 20:29 09:09 15:56 23:02 08:43 18:58 16:16 16:17 11:24 19:29 15:54 01:35

51.77 50.00 56.20 56.70 56.73 52.68 55.90 56.59 50.17 57.45 50.98 56.70 51.71 56.13 51.64 55.83 55.85 56.05 57.37 57.43 51.60

101.00 105.50 116.40 121.20 121.03 106.98 113.12 120.96 105.34 118.84 98.08 118.03 102.70 114.55 104.82 110.34 110.26 111.34 120.77 120.90 104.04

6.9 6.5 7.6 6.5 6.4 6.8 5.8 7 5.7 5.7 6.3 6.0 5.8 5.9 6.0 5.9 5.9 5.6 5.8 5.7 6.3

Mw MLH MLH MLH MLH MLH MLH MLH Mw Ms Mw Mw Mw Mw Mw Mw Mw Mw Mw Mw Mw

D NC NC NC NC NC NC NC H I H H H H H H H H H H H

100 190 228 60 32 249 215 51 31

75 90 60 42 52 48 58 66 82

0 120 147 138 94 64 78 158 155

D VB VB VB VB B Db B S

246 242 84 60 249 200 223 244 90 265 104

80 48 44 82 70 54 44 60 50 45 63

8 65 40 92 88 160 94 74 85 75 47

H MR MR MR Ra M H Rc Rb H H

Busiingol Chara Tunka South Muya South Baikal Kichera I Kichera II Kumora Charuoda I Charuoda II Kultuk

RAKE

Reference for focal solution

Epicenters and origin times are from the catalog of the Baikal Division of GS SB RAS. Foci depths are not given being poor constrained for most events. Focal mechanisms in forms of the strike, dip and rake of a nodal plane are given mainly from the first motion regional solutions, if not they are from teleseismic body-wave inversion. For the 1989 October, 25 there is no focal solution. MLH is magnitude from surface waves given in Kondorskaya and Shebalin, 1982. The references for magnitude values and focal solutions are the following: H – (Harvard) Global CMT Project (www.globalcmt.org), D – Delouis et al., 2002, Db – Doser, 1991b, I – International Seismological Center ISC (www.isc.ac.uk), NC – Kondorskaya and Shebalin, 1982, VB – Vvedenskaya and Balakina, 1960, B – Balakina et al., 1972, MR – Melnikova and Radziminovich, 1998, Ra – Radziminovitch et al., 2005, Rb – Radziminovich et al., 2006b, Rc – Radziminovich et al., 2009, M – Melnikova et al., 2007, S – Solonenko et al., 1993.

148

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

SM-3KV recording ground velocities in the range 0.01–1000 lm/s and some stations include also three components of accelerometers (OSP-2M) with registration range from 50 lm/s2 to 2.5 m/s2. For ‘‘Baikal’’ stations the frequency range is 0.5–20 Hz and sampling rate is 100 s 1. Time control comes from GPS signals. The regional network has been recently implemented with seven stations of the local network of the Buryatian Geological institute in Ulan-Ude. This network coded as ASRS is located at the eastern side of Lake Baikal providing denser station coverage of the southern and central Baikal basins. Each site is equipped with the same ‘‘Baikal’’ stations. There is also an IRIS station coded TLY and located at the southern tip of Lake Baikal. Up state of the stations, getting the data, processing and storing the records are the duties of the Baikal Division of Geophysical Survey as well as warning the public administration and emergency structures. On its web-site (http://www.seis-bykl.ru) a regularly refreshed seismicity map can be found. The Institute of the Earth’s crust staff deals with any analysis and interpretation of the data.

2.2. Data acquisition and processing Most of the stations are located in hard-to-reach areas and the records are transferred to the center in Irkutsk on CD with some time delay. In case of relatively large or felt events the data are treated by the station staff immediately and the information is transferred by radio or phone connection. Event detection is done by an analyst. Routine location of hypocenters is carried out using local software based on a linearized iterative procedure of RMS travel time residual minimization. Regional average travel-time curves are used as velocity model, where Pn = 8.0 km/s, Pg = 6.15 km/s, Sn = 4.54 km/s, and Sg = 3.55 km/s (Golenetskii and Perevalova, 1988). Refracted phase Pn becomes the first arrival beyond 180 km epicentral distance. On the one hand, one-layer average velocity model is very rough; however, for purpose of routine seismological practice it is quite sufficient, at least, for epicenter determination. It should be taken into account that the territory under observation is vast, and it is

Fig. 1. Area under control of the regional seismic network. Here and further topography was built from USGS/NASA SRTM digital elevation data processed by Jarvis et al. (2008) (http://srtm.csi.cgiar.org). (1) Stations of the regional BYKL network. (2) Stations of the local ASRS network. (3) IRIS station. (4) Epicenters of M > 6.5 earthquakes for the instrumental period of observations. (5) Boundaries of areas with the reliable registration of earthquakes with K P 7 (M P 1.7) and K P 8 (M P 2.2) for the current state of the network. (6) State borders. (7) Administrative borders.

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

composed of several large-scale tectonic units such as the Precambrian Siberian craton, Cenozoic rift basins, Sayan-Baikal Early Paleozoic mobile belt and Mongolia–Okhotsk Mesozoic structures. Thus, velocity averaging is needed. In cases of special studies, relocation procedures are applied using more detailed and reliable velocity models. Relocation is also necessary to constrain earthquake depth because the rather sparse network and the average one-layer model prevent from hypocenter routine determining. Due to the geometry of the network, epicenter error ellipses are stretched across the BRS. In the regional catalog the standard error of epicenter location is mostly within 5 km. In the former Soviet Union and now in Russia the size of small and moderate earthquakes is estimated through the energy class K (K-class) defined as a common logarithm of radiated energy in Joules. This scale was introduced in the end of 1950s during detailed studies in the Garm region (western part of the Pamir, Tajikistan). For the Baikal region the K-class is determined from Rautian nomogram developed for intracontinental settings (Rautian, 1964). The general empirical relationship between magnitude and K-class is M = (K 4)/1.8 (Rautian, 1964; Rautian et al., 2007). K-class scale saturates above approximately class 16, but in fact for relatively large events (M P 5.0) the magnitude values are derived from the world seismological centers. There might be a contamination of the regional catalog by explosions. The criteria for explosion discriminating are clustering low energy class events located near settlements with mines and their daily origin time though some single events distributed over the region are not excluded to be explosions. The areas suffering from contamination by explosions are well known. These are mainly Transbaikal region, the southern part of the Siberian platform, and the Baikal-Amur railway. In case of doubt, an event is marked as ‘‘probably explosion’’ in the catalog. 2.3. Earthquake catalog completeness, Gutenberg–Richter relationship The BYKL reporting region occupies several thousand square kilometers and extends from the northern part of Mongolia to the southern Yakutia and from the East Sayan Mountains in the west to the Stanovoi belt in the east. The area of BYKL responsibility covers three administrative units: Irkutsk district, Buryatia Republic and Transbaikal region. The regional earthquake catalog consists of date and origin time, epicenter location, and energy class for over 185,000 events. Since 1987 the energy class has become to be estimated with decimal digit accuracy. The completeness of a catalog depends on the seismic network state (Fig. 2a). Development of our network during the first decade after 1960 resulted in an increasing number of recorded earthquakes. From 1969 to 1980 this number of about 2000 events per year was quite stable excepting some cases of earthquakes with numerous aftershocks. During the next 12 years the network was the most complete due to the deployment of temporary additional stations between the Upper Angara and Muya basins. In this period the average annual number of events in the catalog was 4000 and the percentage of small earthquakes recorded increased from 14% to 28%. Then, in the middle nineties the number of stations was reduced because of a bad economic situation in Russia. The situation improved in the beginning of the present century. Digital recording and GPS time control have resulted in more accurate wave arrivals picking that, in turn, led to decreasing hypocenter uncertainties and improving catalog completeness. The number of recorded events has been increased twice mainly due to significant augmentation of low magnitude earthquakes. At present, about 700–800 earthquakes per month are observed for the denoted region. During the period with analogous recording no K P 8.0 (M  2.2) events were omitted in a wide zone along the BRS. Nowadays, this zone along the BRS is an area of the reliable

149

registration of K P 7.0 or M P 1.7 earthquakes (Fig. 1). To outline the areas of the reliable definite-class earthquake recording, the utmost registration distance for each station was estimated. A Gutenberg–Richter plot constructed for the period of digital observation (Fig. 2b) confirms that M P 1.7 earthquakes are not left out. However, it should be kept in mind that the magnitude values are from the correlation with K-class mentioned above. Thus, it is not quite correct to say about a magnitude threshold; in Russia ‘‘representative K-class’’ term is used. For this reason Gutenberg–Richter relationship in Russia is usually drawn as a recurrence plot where the plot slope has a value c varying in the approximate range from 0.45 to 0.55. The average estimates of c and b values obtained by the method of the least squares for several BRS areas during 1969–2008 are given in the Table 2. The lowest c (b) value is observed for the Tunka–Khubsugul region and the highest one is for the most active central part of the BRS. 2.4. Waveforms, earthquake intensity Waveform format used at the BRS stations is specific, which is a problem for quick international exchange and needs for converters. An example of acceleration waveform for an earthquake is shown at Fig. 3. The BRS earthquake waveforms have not still been used for seismic moment tensor determination or for other purposes. Nevertheless, an attempt to determine the dynamic parameters of the earthquakes sources has been recently made (Dobrynina, 2009). In this work the seismic moment, moment magnitude, source radius, stress drop, and amplitude of displacement were obtain for 62 earthquakes using the amplitude Fourier spectra of the S waves and the classical Brune model. We estimated in the same way the parameters for the last strong event (Mw6.3) in our region that occurred in 2008 in the southern part of Lake Baikal (named as the Kultuk earthquake in Table 1). The scalar value of 4.77  1018 N m derived from the regional records exceeds a little the value from the Global CMT catalog (3.41  1018 N m). The rupture length of 22 km is in a good agreement with the aftershock field length. The value of stress drop is 15.7  105 Pa and displacement is 0.42 m. Near field strong motion records are implemented for engineering purposes and they are also used to compare with shaking intensity estimated from macroseismic survey. The MSK-64 scale is applied for intensity estimation. The largest earthquake recorded instrumentally occurred in 1957 in the Muya basin and it had MLH7.6 (Kondorskaya and Shebalin, 1982) or Mw7.2 (Doser, 1991b). The shaking in the epicenter area was X, but fortunately there were no casualties because it was sparsely populated region. In this sense the seismic hazard is higher for the southern part of the Baikal region, which is more populated and industrially developed. Earthquakes of the Baikal basins, the East Sayan Mountains, Tunka basins and northern Mongolia constitute a threat to Irkutsk city with population more than 600,000. In the Transbaikal region the largest cities are Ulan-Ude and Chita that also experience shaking from the regional earthquakes. 3. Results 3.1. Earthquake depth Depth of BRS earthquakes is poorly constrained because of a small number of stations and of the absence of reliable velocity models for different areas of the BRS. Good-quality results are possible only in some cases suitable for precise relocation. During a relocation procedure a more detailed velocity model can be applied as well as additional temporary stations, and only the best records are involved in processing.

150

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

a

b

Fig. 2. (a) Histogram of the total number of earthquakes recorded by BYKL network as a function of time. Each bin holds 1 year of data. (b) Frequency–magnitude distribution for earthquakes recorded from 2004 to 2010 within the BRS.

Table 2 Parameters c of the recurrence plot and b values in the Gutenberg–Richter relationship for the different areas of the BRS. Magnitude values were converted from K-class using the Rautian correlation M = (K 4)/1.8. Regions

Number of earthquakes used in calculations

K-class (magnitude) range

Khubsugul, Tunka, East Sayan South and Central Baikal North Baikal, Barguzin, Muya Chara, Tokkinskaya basins

2328 7369 16,583 2718

8–13 8–14 8–14 8–12

The first special study aimed at determining hypocentral depths from additional temporary stations was made for aftershocks of the 1959 Middle Baikal earthquake (Golenetskii, 1961; Misharina, 1961), according to which the focal depth range for the majority of events was 15–22 km. The depth of the main shock of August 29, 1959, determined from the converted wave sP at remote stations was 18 ± 3 km (Golenetskii, 1961).

(2.0–5.3) (2.0–5.8) (2.0–5.8) (2.0–4.7)

c

b 0.468 ± 0.006 0.491 ± 0.011 0.529 ± 0.001 0.488 ± 0.005

0.842 ± 0.010 0.884 ± 0.019 0.954 ± 0.002 0.879 ± 0.008

Another opportunity to locate earthquakes accurately was offered by deployment of the temporary local seismic network in the region between the Upper Angara and Muya basins in 1979– 1993 during the construction of the Baikal-Amur railway. The region is distinguished by frequent earthquake swarm occurrence. The results show that the earthquake swarms are localized there at depths of 9–15 km that is shallower than the depths of the

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

151

Fig. 3. An example of acceleration waveform. This record is for the 2002, July 28 earthquake with Ms4.7 located at the Central Baikal. The epicenter distance is 96 km.

background earthquakes (Kochetkov et al., 1987; Sankov et al., 1991). The bottom of the seismically active layer for the area was supposed at 24 km. In the 1990s, hypocenter relocation was carried out with the use of the master event technique and some programs like Hypoinverse (Deverchere et al., 1993, 2001; Gileva et al., 2000). The results show that the crust has a brittle behavior down to 30 km. In several works of the last decade, velocity models of the horizontally-layered crust based on deep seismic sounding were applied to the earthquakes that occurred within the Southern and Central Baikal basins (Radziminovich et al., 2005; Suvorov and Tubanov, 2008; Arefiev et al., 2008). Most of the events were found to occur in the depth range of 10–25 km. Gathering all available data on earthquake depth for the BRS, we can conclude that the highest seismic activity in the Baikal region is mainly observed in the depth range 10–25 km (Fig. 4). It is quite possible that the percentage of the shallow events is underestimated due to some methodological problems arising in the location of the subsurface earthquakes. On the other hand, the relatively low-strength sedimentary deposits are unlikely to generate the majority of shocks. Another question regarding the lower limit of earthquake distribution is of crucial importance in the discussion about the strength properties of the lithosphere. Our viewpoint is that the bottom of the seismically active layer corresponding to the depth above which 90% earthquakes are originated is about 25 km. There is no convincing evidence of subcrustal seismicity; however, several earthquakes in the NE part of the BRS with focal depths of more than 40 km have been described in (Deverchere et al., 1991; Vertlib, 1978, 1981). Taking into account the location vertical error and the ambiguity in the crust thickness, they could occur in the lower part of the crust though Deverchere with coauthors believe that the upper mantle is seismogenic yet. A tendency for earthquakes to deepen in the NE part of the BRS is explained by the closeness of the Precambrian Aldan shield with a thicker and cooler crust (Emmerson et al., 2006; Doser and Yarwood, 1994). It is beyond the scope of this paper to consider the thermomechanical properties of the crust or lithosphere explaining the hypocenter distribution, but in brief we should note that the cut-off value of 25 km is rather high for a rift zone and it is consistent with

dominantly mafic middle and lower crust, high pore pressure and a crust cooler than expected for a rift zone. Discussions on the earthquake depth distribution as well as thermal and strength state of the lithosphere can be found elsewhere (e.g., Petit et al., 1997; Deverchere et al., 2001; Emmerson et al., 2006; Jackson et al., 2008; Petit and Deverchere, 2006). 3.2. Spatial distribution of seismicity Seismicity map generated for the instrumental observation period (Fig. 5) reveals a quite clear picture. Weak and diffuse seismicity is observed in areas of the Siberian craton adjoining to the BRS. Scattered seismicity is also observed for Transbaikal region to the east of the Baikal Lake. Within the rift basins and their close mountain frame seismicity is more concentrated and confined to the main structures and faults. Below we will consider the seismicity in different areas in more detail. 3.2.1. Southwestern part of the BRS In the East Sayan Mountains a clear strip of epicenters is distinguished along the Main Sayan fault that separates the Siberian craton and the mobile belt (Fig. 6). The highest activity of the fault is confined to its southeastern segment as far as its junction with the Okino-Zhombolok fault that penetrates in the East Sayan Mountains. The Okino-Zhombolok fault is known by a Holocene volcanic field comprising a number of hawaiitic lava flows (Ivanov et al., 2011). The strongest known earthquake of the area is the Mw6.9

Fig. 4. Composite histogram showing the distribution (in percentage) of the earthquakes in the Baikal rift zone in 10-km depth intervals.

152

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

Fig. 5. Map of the M P 2.8 seismicity from 1950 to 2008. (1) Epicenters of earthquakes within given magnitude ranges. (2) Focal mechanisms in the lower hemisphere from first motions (by black) and waveform inversion (by gray). (3) Numbers of the earthquakes with M P 5.6 corresponding to Table 1.

Mondy earthquake, which occurred in 1950 in the western part of the Tunka basin. Other three basins, Khubsugul, Darkhad and Busiingol, striking S–N are also seismically active. Although these basins are formally regarded as a part of the BRS, this relation is disputable. At present, the highest level of activity is observed for the Busiingol basin and its surroundings. Seismicity map here is characterized by several M > 5.5 earthquakes, the largest of which was the 1991 Mw6.3 earthquake (Table 1) followed by a longlasting powerful aftershock sequence. 3.2.2. Baikal basins Morphostructurally Lake Baikal occupies three basins: Northern, Central, and Southern, separated by the Olkhon Island and the submerged Akademichesky Ridge and by the Selenga-Buguldeika accumulative submarine isthmus, respectively (Fig. 7). However, tectonically the Central and Southern basins are a single large basin which is the oldest and deepest in the BRS (Logachev, 2003). Looking at the current seismicity map one can note an uneven distribution of earthquakes within the basins. The highest density of epicenters as well as all the larger recent events is observed in the Southern and Central basins, whereas the Northern Baikal basin is characterized by a moderate activity (Figs. 5, 7 and 8). Earthquakes located at the southern tip of Lake Baikal could be connected with the Main Sayan fault since epicenters appear to trend the prolongation of this fault. Within Lake Baikal this fault is supposed to be split into several en echelon faults. Kinematics of these faults is oblique normal faulting with a left-lateral strike-slip component. On August 27, 2008 an Mw6.3 earthquake occurred here, named as Kultuk after the most damaged

settlement. The earthquake is the largest event for the Southern Baikal during the instrumental observation period and it had a serious social effect. Obruchevsky fault is a steeply dipping fault that follows the craton border and defines Baikal basin as a half-graben. Its activity is marked by an epicenter belt running along the western side of the depression. Intrabasin faults also possess a significant seismic potential as has been shown by the 1999 South Baikal Mw6.0 earthquake (Radziminovitch et al., 2005; Peacock and Douglas, 2008). It was a foreshock-aftershock sequence due to normal faulting on a NE striking plane steeply dipping towards the NW. In the latitude of the Selenga delta the western epicenter band is interrupted. In this place the Obruchevsky fault bifurcates into two branches. One branch is the Primorsky fault running on the land and another fault is located within the basin along its western side and on the eastern side of the Olkhon Island. Seismicity is mainly concentrated in this latter area. The Primorsky fault distinctly visible in the topography reveals a surprisingly low seismic activity. Another epicenter strip running along the eastern Baikal side begins at the Selenga delta. This strip is quite wide and it occupies all the space between the eastern basin side and the Olkhon Island. Here the largest historical earthquake occurred. It was the famous 1862 Tsagan M7.5 earthquake (Kondorskaya and Shebalin, 1982) that resulted in the subsidence of a 720 km2 land area and forming the Proval bay named so after the event (Proval is translated as collapse or subsidence). Solonenko and Treskov (1960) reported that NE-striking surface ruptures of the 1862 earthquake were further activated northeastwards during the 1959 Middle Baikal event with MLH6.8 (Table 1).

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

153

Fig. 6. Topography, seismicity and main faults of the southwestern part of the BRS. The earthquakes mentioned in the text are noted. MSF – Main Sayan fault, TF – Tunka fault, OZF – Okino-Zhombolok fault, IMF – Ikhorogol-Mondy fault.

3.2.3. Northeastern part of the BRS To NE of the northern Baikal Lake seismicity forms three epicenter belts pinching at the Muya basin (Fig. 5). The first belt is running along the Barguzin basin, the Ikatskii Ridge, and the incipient Tsipa-Baunt depression (Fig. 8). The second middle belt is stretched in a NE direction from the Barguzin Ridge to the North Muya Ridge and the Muyakan basin. The third epicenter belt encompasses the earthquakes of the Kichera and Upper Angara basins. In this region seismicity is confined not only to the basins but also to the mountains framing or separating them. The mountain area between Barguzin and Kichera basins and between Upper Angara and Muya basins is distinguished by a frequent occurrence of earthquake swarms.

The Kichera depression is the northern termination of the Northern Baikal basin on land. During the instrumental period the most prominent event here is the Kichera earthquakes that occurred on March, 21 1999 (Table 1). These two earthquakes occurred with the one minute interval and there are some discrepancies in their magnitude given by different agencies. According to NEIC, the moment magnitude for the first shock was Mw5.7 and for the second it was Mw5.6. Global CMT solution gives values of 5.9 for both events. As to the focal mechanisms, first motion solution obtained for the first event reveals strike-slip mechanism, whereas the waveform modeling shows normal faulting in both cases. Earthquakes of the Upper Angara basin are located mainly at its northern and eastern parts (Fig. 8). The largest event is the Kumora

154

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

Fig. 7. Seismicity and main faults of the Southern and Central Baikal basins. The earthquakes mentioned in the text are noted. MSF – Main Sayan fault, OF – Obruchevsky fault, PF – Primorsky fault, MF – Morskoy fault, ChF – Chersky fault.

2003 Mw5.6 earthquake followed by aftershocks. Spatial position of the mainshock and its focal solution reveal the continuing process of subsidence of the heterogeneous basement beneath the basin. Seismicity of the area between the Upper Angara and Muya basins named as the North Muya region was intensively studied in 1980s due to local network deployment for the Baikal-Amur railway construction (Sankov et al., 1991). This region is very complex from structural aspect due to the succession of narrow ridges and incipient basins that are the appendices of the Muya basin. Besides NE trending faults, there are a number of NW faults having structural and seismicity control among which the most known is Pereval fault controlling the Angarakan swarm (Kochetkov et al., 1987). The Muya basin is complex and it consists of several depressions. The majority of epicenters are located here along the central part and southern side (Fig. 9). At the eastern termination of the basin the largest earthquake of the BRS recorded instrumentally occurred in 1957. The magnitude of the earthquake named as Muya event was estimated to be MLH 7.6 (Kondorskaya and Shebalin, 1982) and Mw7.2 from waveform inversion (Doser, 1991b). More precisely the epicenter was located at a spur of the South Muya Ridge and the coseismic dislocations were located in the narrow embryonic Namarakit depression. According to Solonenko (1977), the length of the surface rupture was 35 km. Sinistral strike-slip displacements with a normal component on a W–E plane steeply dipping to the north resulted in a lateral movement of the depression westward of about one meter with a subsidence of 5–6 m.

3.2.4. Northeastern termination of the BRS The Chara basin, Kodar and Udokan Ridges also have a high level of seismic activity (Fig. 9). The largest earthquake recorded instrumentally here is the 1994 August, 21 Mw6.0 shock located at the southern part of the Chara basin (Table 1). It was preceded by an earthquake with Ms5.7 occurred on 26, April 1994. Both events were followed by aftershocks, and the second largest earthquake had also foreshocks, whereas before the first one a seismic quiescence had been observed for several years. The most distant and extreme basin of the BRS is the Tokkinskaya depression. Further east ‘‘rifting dies out being blocked by thick lithosphere of Aldan shield’’ (quotation from Logachev, 2003). The Tokkinskaya depression is rather small and is not clearly expressed in the seismicity (Fig. 9). Instead, to the east and south-east of it we can see two epicenter fields. The southern epicenter field has been formed by several M > 6.0 earthquakes. These are the M6.5 Olekma and M6.4 Nyukzha events which occurred in 1958, and the M7.0 1967 Tas-Yuryakh earthquake. Their focal mechanisms show normal faulting with a strike-slip component (Fig. 5, Table 1). Another seismically active area is located at the northeast termination of the Udokan Ridge to the east of the Tokkinskaya depression. Activity here was caused by a swarm of 1997 and two 2005 events with Mw5.8 and Mw5.6 followed by aftershocks. The mainshocks were also normal faults on W–E striking planes. The Olekma River running here along an S–N striking fault is assumed to be the eastern termination of the BRS.

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

155

Fig. 8. Topography, seismicity and main faults of the Northern Baikal basin, Barguzin and Upper Angara basins and their mountain surroundings. The earthquakes as well as swarms (outlined by circles) mentioned in the text are noted. NBF – North Baikal fault, BF – Barguzin fault, PF – Pereval fault, UAF – Upper Angara fault, NMF – North Muya fault, SMF – South Muya fault.

3.3. Seismicity patterns Clustered seismic events like foreshock–aftershock and swarm sequences provide essential information on the seismic process. They are known to take a significant portion of regional catalogs; however, for some tasks they should be removed from them. The results of declustering, which has been performed using the method by Molchan and Dmitrieva (1992), show that aftershocks of the events with M P 4.7 (energy class K P 12.5) make up 25% of the Baikal regional catalog. Solonenko and Solonenko (1987), who had thoroughly considered aftershocks and swarms of the Baikal rift until 1982, stated that as a rule large earthquake and swarms were spatially detached. However, the longer period of observations shows that both swarms and large earthquakes do occur at the same areas. Moreover, in some cases these events can be linked together through stress accumulation, redistribution and release. There are several examples (e.g., the 1991 Busiingol earthquake and the 2005 Charuoda events) when swarms preceded and/or accompany mainshocks with aftershocks resulting in long-lasting (more than 10 years) earthquake sequences with the relatively strong events occurring there from time to time. Swarms also often appear as a

sort of foreshocks like it was for example for the 1999 South Baikal earthquake (Fig. 10). As was noted yet in (Solonenko and Solonenko, 1987), the spatial distribution of swarms over the BRS is irregular. Earthquake swarms are uncommon in its western part, namely in the Tunka basins and the East Sayan Mnt., whereas in the vicinity of the Barguzin basin and at the southern mountain bounding of the Kichera, Upper Angara and Muya basins they are more frequent. Swarms also happen within the Baikal basin; however, they are not long-lasting and not as numerous as in the NE part of BRS. Some cases are difficult to distinguish among a foreshock–mainshock–aftershock sequence and a swarm. Swarm-like sequences are sometimes presented by several ‘‘mainshocks’’ with their own aftershocks rather than by sequences with the time and energy distribution of events typical for swarms. As to the immediate foreshocks before strong earthquakes, they are (if any) commonly few (from several events to several tens of shocks) and they occur from several days to some minutes before mainshocks. A significant part of foreshocks can be often presented by aftershocks of a strong foreshock. The lack of ‘‘classical’’ foreshocks with exponentially increasing number of events for the Baikal region may be explained by lower earthquake magnitudes compared to other interplate settings.

156

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

Fig. 9. Topography, seismicity and main faults of the northeastern part of the BRS. The earthquakes mentioned in the text are noted. WUF – Western Udokan fault, TF – Temulyakit fault, KhF – Khani fault.

3.4. Focal mechanisms, stress field, and strain rate For the Baikal region there are more than 500 fault plane solutions evaluated using P-wave polarity data (Vvedenskaya and Balakina, 1960; Solonenko et al., 1993; Melnikova and Radziminovich, 1998 annual reports issued by Geophysical Survey of the Russian Academy of Sciences and others). This method allows determining the mechanisms only for K > 10.0 (M > 3.3) earthquakes in order to involve as many as possible stations. The smaller events are combined in composite solutions that results in increasing the number of the solutions up to several thousands. The composite mechanisms for the small events confirm the regularities revealed by the individually determined mechanisms and here we consider only the latter. The majority of the solutions are for the BRS earthquakes but some of them characterize the adjacent areas such as northern Mongolia and Transbaikal region. Fault plane solutions of the BRS earthquakes show the predominance of normal faulting mechanisms (72%). Around 19% of the solutions have strike–slip mechanisms and the rest are reverse faults. Extension axis demonstrates a very stable NW–SE stretching direction slightly rotating to S–N at the flank terminations of the BRS (Fig. 11). As can be seen from Fig. 12, T axis lies mainly in the sector 295–345° and its plunge is predominantly less than 15°. P axis is not so stable but it has a NE trend in orientation and it is plunging much steeply. Despite of the normal faulting prevailing, strike–slip type earthquakes play a significant role due to the geometry of the structures in the regional stress field. For example, the sublatitudinal strike of the SW and NE branches of the BRS is favorable to oblique extension. Moreover, the largest events such as the 1957 Mw7.2Muya, the 1967 Mw7.0 Tas-Yurah, the 1950 Mw6.9 Mondy, and the 1991 Mw6.3 Busiingol earthquakes that occurred at the rift

a

b

Fig. 10. (a) The cumulative number of earthquakes occurred in the vicinity of the 1999 February, 25 Mw6.0 South Baikal earthquake before the mainshock. (b) The time interval corresponding to the abrupt increased activity in more detail.

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

157

a

b

Fig. 11. Orientation of T (a) and P (b) axes in the earthquake foci over the Baikal rift system. Rectangles encompass the NE part (I), the Central part (II), and SW part (III) of the BRS for which the components of the strain rate tensor are shown in Table 3.

terminations were pure strike-slip whereas the 1959 MLH6.8 Middle Baikal earthquake was a dip-slip normal faulting event. On the other hand, the fact that such high magnitude events are of strike-slip type implies not only the effect of geometry but also a kinematic boundary condition, e.g. shear movements along the boundaries of the Amurian plate at least within the NE part in the prolongation of the Stanovoy belt (Barth and Wenzel, 2010). Accordingly, the present day stress field shows changes along the rift strike. For the central parts, including the Baikal, Barguzin basins and up to the Chara basin, pure extension is observed, whereas at the system terminations oblique extension with left-lateral motions is revealed. In the SW part of the BRS, the

transtention regime begins at the southern termination of the Baikal basin (Radziminovich et al., 2006a). Further west, in the Tunka area, the compression axis is getting more low-dipping, which leads to a diversity of focal solutions. The stress field is defined here as strike-slip or even transpressive (e.g. Delouis et al., 2002; Arjannikova et al., 2004). It is a region of transition stress field from rift extension to compression acting in the NE azimuth over the huge territory of the Central Asia including Mongolia probably due to the Indo-Asian collision. At the NE termination of the BRS sinistral movements were clearly revealed by the strong earthquakes and the strike–slip regime becomes apparent eastward from the Chara basin.

158

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

a

c

b

d

Fig. 12. Rose diagrams of the azimuth and plunge of T (a and b) and P (c and d) axes in the earthquake focal solutions.

Table 3 Components of the average seismotectonic strain rate tensor from focal solutions of the earthquakes with M P 4.5. Regions

Volume, 103 km3

RM0, 1027 dyn sm

I. NE part II. Central part III. SW part

4750 4020 4710

1.64 0.06 0.33

(Eij), 10 Exx

9

year

1

Eyy

2.52 0.11 0.39

4.84 0.12 0.47

Ezz 2.31 0.23 0.08

Exy 14.4 0.30 3.77

Eyz 9.73 0.08 0.22

Ezx 4.32 0.07 1.11

Time period used for strain rate calculation is 50 years. System of coordinates is x – to the east, y – to the north, and z –up. Minus corresponds to shortening, opposite case means elongation of the crust. Regions are outlined by boxes in the Fig. 11.

The rate of seismic deformation has been calculated from the sum of tensors of earthquake seismic moments normalized by the shear modulus, volume, and time based on the widely known formulae by Kostrov (1974). The data used in the study were quite representative because the sum M0 of events with known mechanisms was greater than 80% of the sum M0 of all earthquakes recorded instrumentally. The region was subdivided into three areas (Fig. 11) and thickness of the seismogenic layer was assumed to be equal 20 km. The results show that the largest values are observed for the horizontal shear component Exy (Table 3). The minus sign points to sinistral shear displacements along the sublatitudinal plane. As compared to the other parts of the rift, the northeastern sector is characterized by larger values of seismic deformation rate. For the South Baikal basin it is possible to compare the strain rate derived from seismological data with results of GPS measurements. The geodetic rate is estimated as 2.1e 8 year 1 (Sankov et al., 2009) while the seismotectonic value is 2.95e 9 year 1 (Radziminovich et al., 2006a) that is smaller than results of geodetic calculations by a factor of 7. In other words, observed seismicity contribution to the average total strain does not exceed 10%. As to the style and direction of strain, GPS and seismological data are in a good agreement.

A few words about the geometry of nodal planes should be added. In general, the strike of nodal planes coincides with the orientation of the main structures. The most prominent instance is the NE striking Baikal basin bounded by the faults of the same orientation. Submeridional and NW-trending planes prevail in focal mechanisms of pure and oblique reverse type. In case of strike–slip solution, left lateral movement along W–E oriented planes and right lateral slip along S–N planes are observed in agreement with the regional stress field. Dip angle of the nodal planes is typical for normal faults. Most of them are in the range 40–60°. Low-angle planes are rare. They are met in less than 15% of all solutions. In addition, in the case of dip-slip solutions with vertical and gently planes we cannot decide between them with few exceptions. Apart from first motion method, the waveform inversion technique has been applied to some Baikal M > 5.0 earthquakes. It concerns centroid moment tensor solutions routinely determined by Global CMT Project and USGS as well as some special works (e.g. Emmerson et al., 2006; Doser 1991a,b; Delouis et al., 2002; Peackock and Douglas, 2008; Bayasgalan et al., 2005). In most cases the double couple solutions obtained from different approaches are similar, but for some earthquakes the waveform modeling has revealed complex rupture processes like three subevents during the

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

1957 Muya earthquake (Doser, 1991b) or a two-stage rupture of the 1999 South Baikal event (Peackock and Douglas, 2008).

4. Discussion In this section we dwell on some peculiarities deduced from above. Earthquakes are spaced unevenly within the basins, and yet it evidences in no way the seismic potential of the different basin parts. For example, in the Tunka basins the current seismicity is located in the western part, whereas the eastern part is an area of seismic gap. However, the 1995 Mw5.8 Tunka earthquake occurred exactly in this area. It was accompanied neither by foreshocks nor by aftershocks. Another instance is the Northern Baikal which looks less active in comparison with the Southern and Central basins. At the same time, along the western side of the Northern Baikal basin several recent seismic deformations have been found, so its seismic potential should not be underestimated. The low level of seismic activity is now observed also along the largest Main Sayan and Primorsky faults. It arises some questions challenging for seismotectonics. Do mature and well developed faults have a low frictional resistance and slip occurs mainly by creep, whereas rare large earthquake on them result from rupturing of large asperities? Or is this a result of strain partitioning between border and intrabasin faults? Does such a short-term picture reflect the fault behavior (stage of the seismic cycle, fault interaction, rheological state and so on) rather than the long-term seismicity pattern? Anyway, it should be kept in mind that we see only the snapshot of the seismicity pattern that again points out the uncertainty of using short-term seismic observations (including historical catalog in our case) for seismic hazard assessment. A fact of irregular distribution of earthquake swarms over the BRS is another question to be explained. It is remarkable that in the BRS swarms occur mostly in the area of the huge granite Angaro-Vitim batholit that occupies a territory to the east of the Northern Baikal. Gravity data show that the outcropped granite massifs merge into one intrusive body at depth forming a pluton of 750 km length and 250 km width with thickness varying from 2 to 30 km and having the average thickness of 10 km (Turutanov, 2011). It should be noted that many cases of other intraplate earthquake swarms are also connected with granitic plutons, including such a known area of swarm occurrence as the Vogtland/NW-Bohemia (Hofmann et al., 2003). One of the possible trigger mechanisms of swarm occurrence is fluid flow processes, and even being not related with active volcanism swarms often occur in areas of enhanced hydrothermal activity. The Barguzin region and adjoining territory is characterized by numerous mineral and hot springs. Thus, fracturing in the upper part of granitic massifs (we noted above that the BRS swarms in the considered area are shallower comparing the background earthquakes) and hydrothermal activity could promote swarm type seismicity. A generating mechanism of swarms as deglaciation flexure has been suggested for the Northern Norway (Hicks et al., 2000). Although glacial dimensions of the Barguzin region and the Fennoscandian shield are incommensurable, postglacial rebound as an additional factor for more frequent occurrence of swarms in the Barguzin Ridge might be speculatively assumed. The largest Pleistocene glaciation of the Barguzin Ridge has been reconstructed from moraines and geomorphology evidence. The thickness of the glacier was estimated up to 700 m (Levi et al., 1998). Unfortunately, this area has not been involved in GPS campaign, so the only data on the vertical movement rate came from geodetic measurements conducted 20 years ago. The results of the

159

measurements given in (Levi et al., 1998) let the authors suppose postglacial uplift to contribute to the rate, which reaches 8–10 mm/year. Fewer occurrences of swarms in the SW part of the BRS point out that strain realizes here by movements along the favorable oriented main faults rather than fracturing. The lower b-value also argues for that. Moreover, the stress field of the crust in this area is characterized by the significant influence of the horizontal compression in NE direction that gives transpression stress regime, and it is known that strength of the crust under compression is higher than under extension. Comparison of the strain rate derived from seismological and GPS data showed much higher values for the latter. A similar ratio of seismic to geodetic rates that is less than one has been found in other regions, e.g. Kamchatka (Zobin, 1987), Tan-Shan (Sycheva et al., 2005), United States (Ward, 1998), Iran (Masson et al., 2005). It is still a question whether such a ratio is due to an actually large inelastic component of strain or it is caused by any methodological problems when comparing GPS and seismic data. Ward (1998) argued that a significant reason for the observed ratio is duration and completeness of earthquake catalogs that fails to reflect the long-term situation. The largest earthquake recorded instrumentally in the BRS was the 1957 Muya event with MLH7.6 or Mw7.2. This magnitude value is quite high for intracontinental setting though much lower than several M P 8 earthquakes of Mongolia which occurred in the last century. All great Mongolian earthquakes, namely the 1905 Ms7.8 and Ms8.2 Bolnai events, the 1931 Ms8.0 Fuyun event and the 1957 Ms8.1 Gobi-Altai earthquake were caused by movement along large strike-slip faults accommodating strain transfer from the Indo-Asia collision. The same setting could be advocated for some M7.6–8 paleoearthquakes revealed on the shear Main Sayan fault zone (Chipizubov and Smekalin, 1999). The Muya earthquake is also a strike–slip event which occurred in somewhat different setting. The NE segment of the BRS where the earthquake occurred is considered to be a part of the northern boundary of the Amurian microplate. This boundary is under strike–slip stress regime with an ENE–WSW orientation of the maximum horizontal compressional stress (Barth and Wenzel, 2010). Normal fault earthquakes connected with basin subsidence also show large magnitudes value, more precisely M7.5 from the historical catalog (the 1862 Tsagan event) and M6.8 (the 1958 Middle Baikal event) recorded instrumentally. Such seismic potential implies an important seismic hazard for several administrative districts around Lake Baikal. The southern part of the East Siberia should be a region of special attention in this sense because it is the most populated area and a number of dangerous factories and constructions are located there.

5. Conclusion The BYKL earthquake catalog consists of over 185,000 events. Change of the recording system from analogous to digital mode at the close of the last century improved the detection capability of the seismic network and allowed involving in the catalog the M > 1.7 earthquakes. At the same time, there are current system shortcomings. No real-time system and preliminary automated locations are working in the region. A quick preliminary catalog is available nearly on-the-fly at the web site but only for earthquakes with K P 9.5 (M P 3), which is inconvenient for quick aftershock study, for example. The final catalog where locations are greatly improved appears with a delay of several years. Modern advances in waveform analysis have not been fully used. Undoubtedly, all the advantages of digital recording will be taken in the near future.

160

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161

Nevertheless, observations of the regional seismic network for 50 years have allowed identifying the main features of the BRS seismicity. The epicenter map obtained shows that seismicity is mainly concentrated within the rift basins and adjoining mountain areas forming epicenter bands or strips that strike mainly in accordance with the main structures. As to the hypocenter resolution, we have had only a general picture of the active structures but this is not enough for detecting the fault geometry. The earthquake frequency-size distribution has an average bvalue of 0.9. By now the maximum magnitude of earthquakes recorded by the regional network is MLH7.6 for a strike–slip event in the NE part of the BRS and MLH6.8 for a normal fault earthquake of the Baikal basin. The seismicity patterns are formed by both independent random events and interrelated earthquakes. Aftershocks follow the majority of M > 5 events, whereas foreshocks occur quite seldom. Swarms are typical patterns in the BRS and on some cases they can be connected with the strong events. Independent swarms occur more often in the central and NE parts of the BRS evidencing higher level of fracturing of the crust there. Predominant movement type is normal faulting on NE striking faults with a left lateral strike–slip component on W–E planes. The central part of the BRS is under steady extension acting in the NW–SE direction, whereas transtension or strike–slip stress regime has been revealed at the system terminations. Moreover, the SW part (Tunka basins and East Sayan) is thought to be under transpression conditions with a compression axis lying in the NE azimuth due to the Indo-Asian collision. In conclusion it should be noted that upgrade of the network is needed that would imply increasing the number of stations, installing broadband stations, developing a strong motion network, and real-time data transfer. It will result in a better understanding of regional seismotectonics and more efficient hazard prediction. Besides, developing of the seismic network is an indispensable condition for planning and response of civic and emergency services. Acknowledgments The regional earthquake catalog used in the paper has been compiled by the Baikal Division of the Geophysical Survey of SB RAS. We thank anonymous reviewers and Jacques Deverchere as a Guest Editor for their constructive comments and suggestions that improved this manuscript significantly. This work was supported by a grant of the Russian Foundation for Basic Research (12-05-91161-GFEN_a). References Achauer, U., Masson, F., 2002. Seismic tomography of continental rifts revisited: from relative to absolute heterogeneities. Tectonophysics 358, 17–37. Arefiev, S.S., Bykova, V.V., Gileva, N.A., Masalsky, O.K., Matveev, I.V., Matveeva, N.V., Melnikova, V.I., Chechelnitsky, V.V., 2008. Preliminary results of epicentral observations of the Kultuk earthquake August 27. Vopr. Inzh. Seismol. 35 (4), 5– 15 (in Russian). Arjannikova, A., Larroque, S., Ritz, J.-F., Déverchère, J., Stephan, J.-F., Arjannikov, S., San’kov, V., 2004. Geometry and kinematics of recent deformation in the Mondy-Tunka area (south-westernmost Baikal rift zone, Mongolia-Siberia). Terra Nova 16 (5), 265–272. Balakina, L.M., Vvedenskaya, A.V., Golubeva, N.V., Misharina, L.A., Shirokova, E.I., 1972. Elastic Stress Field in the Earth’s Crust and Earthquake Focal Mechanism. Nauka, Moscow (in Russian). Barth, A., Wenzel, F., 2010. New constraints on the intraplate stress field of the Amurian plate deduced from light earthquake focal mechanisms. Tectonophysics 482, 160–169. http://dx.doi.org/10.1016/j.tecto.2009.01.029. Bayasgalan, A., Jackson, J., McKenzie, D., 2005. Lithosphere rheology and active tectonics in Mongolia: relations between earthquake source parameters, gravity and GPS measurements. Geophysical Journal International 163, 1151–1179. Brazier, R.A., Nyblade, A.A., 2003. Upper mantle P velocity structure beneath the Baikal Rift from modeling regional seismic data. Geophysical Research Letters 30 (4), 1153. http://dx.doi.org/10.1029/2002GL016115.

Calais, E., Lesne, O., De´verche‘re, J., San’kov, V., Lukhnev, A., Miroshnitchenko, A., Buddo V., Levi, K., Zalutzky, V., Bashkuev, Y., 1998. Crustal deformation in the Baikal rift from GPS measurements. Geophysical Research Letters 25 (21), 4003–4006. Chipizubov, A.V., Smekalin, O.P., 1999. Paleoseismodislocations and related paleoearthquakes along the major Sayan fault zone. Russian Geology and Geophysics 40 (6), 936–947. Delouis, B., Deverchere, J., Melnikova, V., Radziminovitch, N., Loncke, L., Larroque, C., Ritz, J.F., San‘kov, v, 2002. A reappraisal of the 1950 (Mw 6.9) Mondy earthquake, Siberia, and its relationship to the strain pattern at the southwestern end of the Baikal rift zone. Terra Nova 14, 491–500. Deverchere, J., Houdry, F., Diament, M., Solonenko, N., Solonenko, A., 1991. Evidence for a seismogenic upper mantle and lower crust in the Baikal rift. Geophysical Research Letters 18 (6), 1099–1102. Deverchere, J., Houdry, F., Solonenko, N.V., Solonenko, A.V., San’kov, A.V., 1993. Seismicity, active faults and stress field of the North Muya region, Baikal rift: New insights on the rheology of extended continental lithosphere. Journal of Geophysical Research 98, 19895–19912. Deverchere, J., Petit, C., Gileva, N., Radziminovitch, N., Melnikova, V., Sankov, V., 2001. Depth distribution of earthquakes in the Baikal rift system and its implications for the rheology of the lithosphere. Geophysical Journal International 146, 714–730. Dobrynina, A.A., 2009. Source parameters of the earthquakes of the Baikal Rift system. Izvestiya, Physics of the Solid Earth 45 (12), 1093–1109. Doser, D.I., 1991a. Faulting within the western Baikal rift as characterized by earthquake studies. Tectonophysics 196, 87–107. Doser, D.I., 1991b. Faulting within the eastern Baikal rift as characterized by earthquake studies. Tectonophysics 196, 109–139. Doser, D., Yarwood, D., 1994. Deep crustal earthquakes associated with continental rifts. Tectonophysics 229, 123–131. Emmerson, B., Jackson, J., McKenzie, D., Priestley, K., 2006. Seismicity, structure and rheology of the lithosphere in the Lake Baikal region. Geophysical Journal International 167, 1233–1272. http://dx.doi.org/10.1111/j.1365246X.2006.03075.x. Florensov, N.A., Solonenko, V.P. (Eds.), (1963). The Gobi-Altai Earthquake. Moscow: Akademiya Nauk, USSR (in Russian; English translation, 1965, US Department of Commerce, Washington, DC), 424pp. Gao, S.S., Liu, K.H., Davis, P.M., Slack, P.D., Zorin, Y.A., Mordvinova, V.V., Kozhevnikov, V.M., 2003. Evidence for small-scale mantle convection in the upper mantle beneath the Baikal rift zone. Journal of Geophysical Research 108 (B4), 2194. http://dx.doi.org/10.1029/2002JB002039. Gao, S.S., Liu, K.H., Chen, C., 2004. Significant crustal thinning beneath the Baikal Rift Zone: new constraints from receiver functions analysis. Geophysical Research Letters 31, L20610. http://dx.doi.org/10.1029/2004GL020813. Gileva, N.A., Melnikova, V.I., Radziminovich, N.A., Deverchbre, J., 2000. Location of earthquakes and average velocity parameters of the crust in some areas of the Baikal region. Russian Geology and Geophysics 5, 609–615. Golenetskii, S.I., 1961. Determinations of the crust thickness from observations of waves reflected from the crust base and source depths of aftershocks of the Middle_Baikal earthquakes of August 29, 1959. Russian Geology and Geophysics 2, 111–116. Golenetskii, S.I., 1977. Seismicity of the Baikal region: history of its studying and some results. In: Gorshkov, G.P. (Eds.), Seismicity and Seismogeology of the East Siberia, Moscow, Nauka, 1977, pp. 3–42 (in Russian). Golenetskii, S.I., Perevalova, G.I., 1988. Computer methods for determination of the earthquake hypocenters from the observations at the seismic stations of local networks (under conditions of their limited number in the Baikal Zone). In: Pavlov, O.V., Borovik, N.S. (Eds.), Investigations on the Search for Earthquake Precursors in Siberia. Nauka, Novosibirsk, pp. 87–99. Golenetskii, S.I., 1990. Problem of the Study of Baikal Rift Seismicity. In: Geodynamics of Intracontinental Mountainous Regions, Nauka, Novosibirsk, pp. 228–235 (in Russian). Hicks, E.C., Bungum, H., Lindholm, C.D., 2000. Seismic activity, inferred crustal stresses and seismotectonics in the Rana region, Northern Norway. Quaternary Science Reviews 19, 1423–1436. Hofmann, Y., Jahr, T., Jentzsch,, g., 2003. Three-dimensional gravimetric modelling to detect the deep structure of the region Vogtland/NW-Bohemia. Journal of Geodynamics 35, 209–220. Hutchinson, D.R., Golmshtok, A.J., Zonenshain, L.P., Moore, T.C., Scholtz, C.A., Klitgord, K.D., 1992. Depositional and tectonic framework of the rift basins of Lake Baikal from multichannel seismic data. Geology 20, 589–592. Ivanov, A.V., Arzhannikov, S.G., Demonterova, E.I., Arzhannikova, A.V., Orlova, L.A., 2011. Jom-Bolok Holocene volcanic field in the East Sayan Mts., Siberia, Russia: structure, style of eruptions, magma compositions, and radiocarbon dating. Bulletin of Volcanology 73, 1279–1294. http://dx.doi.org/10.1007/s00445-0110485-9. Jackson, J.A., McKenzie, D., Priestley, K., Emmerson, B., 2008. New views on the structure and rheology of the lithosphere. Journal of the Geological Society, London 165, 453–465. Jarvis, A., Reuter, H.I., Nelson, A., Guevara, E., 2008. Hole-Filled Seamless SRTM Data V4 International Centre for Tropical Agriculture (CIAT). . Kochetkov, V.M., Borovik, N.S., Misharina, L.A., Solonenko, A.V., Anikanova, G.V., Solonenko, N.V., Melnikova, V.I., Gileva, N.A., 1987. In: Pavlov, O.V. (Ed.), Angarakan Swarm of Earthquakes in the Baikal Rift Zone, Nauka, Novosibirsk (in Russian).

N.A. Radziminovich et al. / Journal of Asian Earth Sciences 62 (2013) 146–161 Kondorskaya, N., Shebalin, N. (Eds.), 1982. New Catalogue of Strong Earthquakes in the USSR. Washington, DC: World Data Centre A, US Department of Commerce. Kostrov, V.V., 1974. Seismic moment, energy earthquake, and the seismic flow of rocks. Izvestiya Academy of Sciences of the USSR, Physics of the Solid Earth 10, 23–44. Levi, K.G., Mats, V.D., Kusner, Yu.S., Kirillov, P.G., Alakshin, A.M., Tolstov, S.V., Osipov, E.Yu., Efimova, I.M., Buck, S., 1998. Postglacial Tectonics in the Baikal Rift. Russian Journal of Earth Sciences 1 (1) . Logatchev, N.A., Florensov, N.A., 1978. The Baikal system of rift valleys. Tectonophysics 45, 1–13. Logachev, N.A., 2003. History and geodynamics of the Baikal rift. Russian Geology and Geophysics 44 (5), 391–406. Masson, F., Chery, J., Hatzfeld, D., Martinod, J., Vernant, P., Tavakoli, F., GhaforyAshtiani, M., 2005. Seismic versus aseismic deformation in Iran inferred from earthquakes and geodetic data. Geophysical Journal International 160, 217– 226. http://dx.doi.org/10.1111/j.1365-246X.2004.02465.x. Melnikova, V.I., Radziminovich, N.A., 1998. Focal mechanisms of the earthquakes of the Baikal region for 1991–1996. Russian Geology and Geophysics 39 (11), 1598–1607. Melnikova, V.I., Radziminovich, N.A., Gileva, N.A., Chipizubov, A.V., Dobrynina, A.A., 2007. Activation of rifting processes in the Northern Cis-Baikal region: a case study of the Kichera earthquake sequence of 1999. Izvestiya, Physics of the Solid Earth 43 (11), 905–921. Melnikova, V.I., Gileva, N.A., Radziminovich, N.A., Masal’skii, O.K., Chechel’nitskii, V.V., 2010. Seismicity of the Baikal Rift Zone for the digital recording period of earthquake observation (2001–2006). Seismic Instruments 46 (2), 138–151. Misharina, L.A., 1961. Aftershocks of the middle Baikal earthquake of August 29, 1959. Russian Geology and Geophysics 2, 105–110 (in Russian). Molchan, G., Dmitrieva, O., 1992. Aftershock identification: methods and new approaches. Geophysical Journal International 109, 501–516. Nielsen, C., Thybo, H., 2009. No Moho uplift below the Baikal Rift Zone: evidence from a seismic refraction profile across southern Lake Baikal. Journal of Geophysical Research 114, B08306. http://dx.doi.org/10.1029/2008JB005828. Peacock, S., Douglas, A., 2008. Two-stage rupture of the 1999Mw 6.0 Southern Lake Baikal Earthquake. Geophysical Journal International 172, 302–310. http:// dx.doi.org/10.1111/j.1365-246X.2007.03629.x. Petit, C., Burov, E.B., Deverchere, J., 1997. On the structure and mechanical behavior of the extending lithosphere in the Baikal rift from gravity modelling. Earth and Planetary Science Letters 149, 29–42. Petit, C., Deverchere, J., 2006. Structure and evolution of the Baikal rift: a synthesis. Geochemistry Geophysics Geosystems 7, Q11016. http://dx.doi.org/10.1029/ 2006GC001265. Puzyrev, N.N., Mandelbaum, M.M., Krylov, S.V., Mishenkin, B.P., Krupskaya, G.V., Petrick, G.V., 1973. Deep seismic investigation in the Baikal Rift Zone. Tectonophysics 20, 85–95. http://dx.doi.org/10.1016/0040- 1951(73)90098-X. Radziminovitch, N., Déverchère, J., Melnikova, V., San‘kov, V., Giljova, N., 2005. The 1999 Mw 6.0 earthquake sequence in the Southern Baikal rift, Asia, and its seismotectonic implications. Geophysical Journal International 161, 387–400. http://dx.doi.org/10.1111/j.1365-246X.2005.02604.x. Radziminovich, N.A., Melnikova, V.I., San‘kov, V.A., Levi, K.G., 2006a. Seismicity and seismotectonic deformations of the crust in the Southern Baikal basin. Izvestiya, Physics of the Solid Earth 42 (11), 904–920. Radziminovich, N.A., Imaeva, L.P., Melnikova, V.I., Imaev, V.S., 2006b. Seismotectonic position of the Oldongso swarm and the 2005 aftershock sequence at the NE flank of the Baikal rift zone. In: Proceedings of the Symposium: Geodynamic Evolution of the Lithosphere of the Central Asia Mobile Belt (from the Ocean to the Continent), vol. 2 (4). Institute of the Earth‘s Crust, Irkutsk, pp. 87–90 (in Russian). Radziminovich, N.A., Gileva, N.A., Radziminovich, Ya.B., Kustova, M.G., Chechelnitskii, V.V., Melnikova, V.I., 2009. The 2003 September 16 Kumora Earthquake with Mw = 5.6, KR = 14.3 and I0 = 7. Earthquakes of the Northern Eurasia in 2003, Obninsk, Geophysical Survey of RAS, pp. 293–309 (in Russian). Radziminovich, N.A., 2010. Focal depths of earthquakes in the Baikal region: a review. Izvestiya, Physics of the Solid Earth 46 (3), 216–229. Rautian, T.G., 1964. On the determination of energy of earthquakes at distances up to 3000 km. Trudy Inst. Fiz. Zemli Akad. Nauk SSSR 199, 88–93 (in Russian).

161

Rautian, T.G., Khalturin, V.I., Fujita, K., Mackey, K.G., Kendall, A.D., 2007. Origins and methodology of the russian energy K-class system and its relationship to magnitude scales. Seismological Research Letters 78 (6), 579–590. Sankov, V.A., Dneprovskii, Yu.I., Kovalenko, S.N., Bornyakov, S.A., Gileva, N.A., Gorbunova, N.G., 1991. Faults and Seismicity of the Severomuiskii Geodynamic Research Area. In: Sherman, S.I. (Ed.), Nauka, Novosibirsk (in Russian). Sankov, V.A., Lukhnev, A.V., Miroshnichenko, A.I., Ashurkov, S.V., Byzov, L.M., Dembelov, M.G., Calais, E., Deverchere, J., 2009. Extension in the Baikal rift: present-day kinematics of passive rifting. Doklady Earth Sciences 425 (2), 205– 209. Solonenko, V.P. (Ed.), 1977. Seismic Zoning of the East Siberia and its Geology and Geophysics Base. Nauka, Novosibirsk (in Russian). Solonenko, V.P., Treskov, A.A., 1960. The Middle Baikal Earthquake on August 29, 1959. Irkutsk, Institute of the Earth’s crust (in Russian). Solonenko, N.V., Solonenko, A.V., 1987. Aftershock Sequences and Earthquake Swarms in the Baikal Rift Zone. Nauka, Novosibirsk (in Russian). Solonenko, A.V., Solonenko, N.V., Melnikova, V.I., Kozmin, B.M., Kuchai, O.A., Sukhanova, S.S., 1993. Strains and displacements in earthquake foci of Siberia and Mongolia. In: Seismicity and Seismic Zoning of Northern Eurasia, vol. 1, Moscow, pp. 113–122 (in Russian). Suvorov, V.D., Mishenkina, Z.M., Petric, G.V., Sheludko, I.F., Seleznev, V.S., Solovyov, V.M., 2002. Structure of the crust in the Baikal Rift Zone and adjacent areas from Deep Seismic Sounding data. Tectonophysics 351, 61–74. http://dx.doi.org/ 10.1016/S0040-1951(02)00125-7. Suvorov, D.V., Tubanov, Ts.A., 2008. Distribution of sources of close earthquakes in the crust beneath the Central Baikal. Russian Geology and Geophysics 49 (8), 805–818. Sycheva, N.A., Yunga, S.L., Bogomolov, L.M., Mukhamadeeva, V.A., 2005. Determination of seismotectonic crustal strains in the North Tien Shan using focal mechanisms from data of the KNET digital seismic network. Izvestiya, Physics of the Solid Earth 41, 916–930. Ten Brink, U.S., Taylor, M.H., 2002. Crustal structure of central Lake Baikal: Insights into intracontinental rifting. Journal of Geophysical Research 107 (B7), 2132. http://dx.doi.org/107.10.1029/2001JB000300. Tiberi, C., Diament, M., De´verche‘re, J., Petit-Mariani, C., Mikhailov, V., Tikhotsky, S., Achauer, U., 2003. Deep structure of the Baikal rift zone revealed by joint inversion of gravity and seismology. Journal of Geophysical Research 108 (B3), 2133. http://dx.doi.org/10.1029/2002JB001880. Turutanov, E.Kh., 2011. The Angara-Vitim batholite: data on morphology and sizes based on gravimetric data. Doklady Earth Sciences 440 (2), 1464–1466. http:// dx.doi.org/10.1134/S1028334X11100254. Vertlib, M.B., 1978. Determination of earthquake source depths in the Baikal Region. Russian Geology and Geophysics 9, 141–146. Vertlib, M.B., 1981. Determination of the Depths of Earthquake Source by the Group Method in Some Areas of the Baikal Region. In: Seismic Investigations in Eastern Siberia, Nauka, Moscow, pp. 82–88 (in Russian). Vvedenskaya, A.V., Balakina, L.M., 1960. Procedure and results of determination of stresses acting in earthquakes in Pribaikalia and Mongolia. Byulleten Soveta po Seismologii 10, 73–84 (in Russian). Ward, S., 1998. On the consistency of earthquake moment rates, geological fault data, and space geodetic strain: the United States. Geophysical Journal International 134, 172–186. Zhao D., Lei J., Inoue T., Yamada, Gao, S., 2006. Deep structure and origin of the Baikal rift zone. Earth and Planetary Science Letters 243, 681–691. http:// dx.doi.org/10.1016/j.epsl.2006.01.033. Zobin, V.M., 1987. Earthquake focal mechanism and seismotectonic strain of the Kamchatka-Komandor region for 1964–1982. Vulcanologia i Seismologia 6, 78– 92 (in Russian). Zonenshain, L.P., Savostin, L.A., 1981. Geodynamics of the Baikal rift zone and plate tectonics of Asia. Tectonophysics 76, 1–45. Zorin, Y.A., Mordvinova, V.V., Turutanov, E.K., Belichenko, B.G., Artemyev, A.A., Kosarev, G.L., Gao, S.S., 2002. Low seismic velocity layers in the Earth’s crust beneath eastern Siberia (Russia) and central Mongolia: Receiver function data and their possible geological implication. Tectonophysics 359, 307–327. http:// dx.doi.org/10.1016/S0040-1951(02)00531-0.