Mantle transition zone beneath the central Tien Shan: Lithospheric delamination and mantle plumes

Accepted Manuscript Mantle transition zone beneath the central Tien Shan: Lithospheric delamination and mantle plumes

Grigoriy Kosarev, Sergey Oreshin, Lev Vinnik, Larissa Makeyeva PII: DOI: Reference:

S0040-1951(17)30513-9 https://doi.org/10.1016/j.tecto.2017.12.010 TECTO 127717

To appear in:

Tectonophysics

Received date: Revised date: Accepted date:

11 July 2017 15 September 2017 15 December 2017

Please cite this article as: Grigoriy Kosarev, Sergey Oreshin, Lev Vinnik, Larissa Makeyeva , Mantle transition zone beneath the central Tien Shan: Lithospheric delamination and mantle plumes. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Tecto(2017), https://doi.org/ 10.1016/j.tecto.2017.12.010

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Mantle transition zone beneath the central Tien Shan: lithospheric delamination and mantle plumes Grigoriy Kosarev, Sergey Oreshin, Lev Vinnik and Larissa Makeyeva Institute of physics of the Earth, Moscow, Russia Abstract

PT

We investigate structure of the mantle transition zone (MTZ) under the central Tien Shan in

RI

central Asia by using recordings of seismograph stations in Kyrgyzstan, Kazakhstan and adjacent

SC

northern China. We apply P-wave receiver functions techniques and evaluate the differential time between the arrivals of seismic phases that are formed by P to SV mode conversion at the

NU

410-km and 660-km seismic boundaries. The differential time is sensitive to the thickness of the MTZ and insensitive to volumetric velocity anomalies above the 410-km boundary. Under part

MA

of the southern central Tien Shan with the lowest S wave velocity in the uppermost mantle and the largest thickness of the crust, the thickness of the MTZ increases by 15-20 km relative to the

D

ambient mantle and the reference model IASP91. The increased thickness is a likely effect of

PT E

low (about -150 K) temperature. This anomaly is indicative of delamination and sinking of the mantle lithosphere. The low temperature in the MTZ might also be a relic of subduction of the

CE

oceanic lithosphere in the Paleozoic, but this scenario requires strong coupling and coherence between structures in the MTZ and in the lithosphere during plate motions in the last 300 Myr.

AC

Our data reveal a reduction of thickness of the MTZ of 10 - 15 km under the Fergana basin, in the neighborhood of the region of small-scale basaltic volcanism at the time near the CretaceousPaleogene boundary. The reduced thickness of the MTZ is the effect of a depressed 410-km discontinuity, similar to that found in many hotspots. This depression suggests a positive temperature anomaly of about 100- 150 K, consistent with the presence of a thermal mantle plume. A similar depression on the 410-km discontinuity is found underneath the Tarim basin.

1

ACCEPTED MANUSCRIPT 1. Introduction

The central Tien-Shan (Fig. 1) is the most active segment of the largest intra-continental mountain belt. The ongoing growth of the Tien Shan is a likely effect of the India-Eurasia collision (Molnar and Tapponier, 1975), though the central Tien Shan is separated from the

PT

India-Eurasia suture zone by more than 1000 km. Previous mountain-building episodes in the region of the Tien Shan took place in the Paleozoic (e.g. Windley et al., 1990), but for about

RI

100 Myr prior to the onset of the present-day orogenesis the lithosphere of the Tien-San was

SC

quiet. Tectonic activity resumed at about 25-20 Myr in the south of the study region (e.g. Sobel and Dumitru, 1997) and at 11 Myr in the north (Bullen et al., 2001). At present Tarim

NU

converges with Eurasia at about 20 mm/yr (e.g., Reigber et al., 2001).

MA

Early analyses of teleseismic travel time residuals revealed a reduced wave velocity in the upper mantle to the east of the Talas – Fergana fault (Vinnik and Saipbekova, 1984 ). This

D

result was later confirmed by travel-time tomography based on analogue recordings of local

PT E

earthquakes (Roecker et al., 1993) and interpreted as an indication of delamination of the lithosphere. Possibilities for structural studies were greatly improved by deployment of a few digital networks (Fig. 1). The biggest networks are KNET, KRNET, GHENGIZ and MANAS.

CE

Seismic tomography and receiver functions (e.g., Vinnik et al., 2006; Lei and Zhao, 2007;

AC

Zhiwei et al., 2009; Zabelina et al., 2013 ) revealed a number of details of the crust and upper mantle beneath the Tien Shan. Coherence between the high-resolution models that were obtained in different studies varies but a stable pattern is starting to emerge. In the present study we analyze topography on the main (410-km and 660-km) boundaries in the MTZ by using P-wave receiver functions (PRFs). As the boundaries in the MTZ are related to phase transitions, their depths are sensitive to temperature, and the temperature is closely related to geodynamics. Previously PRFs were used for similar purposes by Chen et al. (1997) and Tian

2

ACCEPTED MANUSCRIPT et al. (2010), but our study differs in methodology and is based on a larger number of seismograph stations.

2. Data and results Our data are obtained from the records of almost 100 seismograph stations (Fig. 1). The

PT

PRFs are calculated by using the LQ coordinate system, where L is parallel to the principal

RI

motion direction of the P wave and Q is normal to L in the wave propagation plane. The L and Q components are low-pass filtered with the corner at around 6 seconds and deconvolved by the L

SC

components in time domain. The deconvolved Q components are stacked with move-out time

NU

corrections. In the stacked Q components we inspect the travel times of seismic phases P410s and P660 s (converted from P to SV at the 410-km and 660-km boundaries).

MA

The main problem of interpretation of travel times of P410s and P660s is separation of the effects of topography on the 410-km and 660-km boundaries and of volumetric velocity

D

variations in the crust and upper mantle above the 410-km boundary. We solve this problem by

PT E

calculating the differential time between arrivals of P660s and P410s. The wave paths of P660s and P410s in the crust and upper mantle are close to each other for the same seismic recording,

CE

and, as a result, the differential time is insensitive to the properties of the earth above the MTZ. To detect P660s and P410s and to map the differential time, a large number of the PRFs should

AC

be stacked. One possibility is to apply a version of CCP (Common Conversion Point) stacking: to divide the Earth’s surface into cells and to stack with appropriate move-out time corrections the PRFs, the projections of the conversion points of which fall into the same cell. The surface projections of the conversion points of P410s and P660s for the same recording are at different distances (around 1° and 2°, respectively) from the seismograph station, and the set of PRFs thus selected for the detection of P410s differs from that for P660s. Then the differential time of P660s and P410s can be affected by lateral heterogeneity in the crust and upper mantle above the MTZ. In order to avoid this, we find conversion points in the middle of the MTZ (at a depth of 3

ACCEPTED MANUSCRIPT 535 km) and stack those PRFs, the projections of the conversion points of which are located within the same cell. Then P410s and P660s for each cell are detected in the same set of PRFs and the effect of lateral heterogeneity of the earth’s medium above the 410-km discontinuity in the differential time is minimized. Lateral resolution of the resulting map of the differential time depends on the size of the

PT

cell but a high resolution of small cells can be delusive due to low signal/noise ratio in the stacked PRFs. To evaluate the optimum size we experimented with rectangular cells of varying

RI

length (SN direction) and width (EW direction). The optimum results are obtained for 2° in NS

SC

and EW (220 km and 160 km, respectively). In the south-western margin of the study area (Pamir) we combined two standard boxes into one. As the MANAS network is very dense

NU

relative to the others, we divided the MANAS stations into clusters of 4 neighboring stations and

MA

replaced each cluster by one virtual station with a reduced number of recordings. Epicenters of seismic events of sufficient magnitude in a distance range from 35° to 90° are abundant in a

D

broad back azimuth range (Fig. 2). Surface projections of the conversion points thus obtained

PT E

cover with high density the area between approximately 38°N and 44°N and between 72°E and 82°E (Fig. 3).

The number of stacked PRFs in most of the cells is on the order of several hundred. The

CE

largest number is close to 1800, and we use only one cell with the number of conversion points

AC

less than 100. These numbers are sufficient for a robust detection of P660s and P410s (Fig. 4). The accuracy of the resulting estimates of the differential time is typically ±0.2 s. The stack for each cell is usually based on the data from the seismograph stations which are spread over almost the entire central Tien Shan (Fig. 5). The delay times of P410s and P660s are summarized in Table 1. For most cells the resulting residuals of the differential time with respect to the IASP91 value (23.9 s) are on the order of a fraction of a second (Fig. 6). Large residuals (more than 1.0 s) are obtained for three cells: (40° - 42°N, 76° - 78°E, +1.5 s), (40° - 42°N, 72° - 74°E, -1.1 s) and (38° - 40°N, 80° - 82°E, -1.5 s). The resulting anomalies of thickness of the MTZ are 4

ACCEPTED MANUSCRIPT +15 km, -11 km, and -15 km, respectively. These estimates are obtained on the assumption of standard wave velocities in the MTZ. In addition to the robust estimates of thickness of the MTZ we obtained rough estimates of topography of the 410-km and 660-km boundaries. The anomalies of travel time of either P410s or P660s depend mainly on topography on the discontinuity and volumetric velocity anomalies

PT

in the crust and upper mantle above the 410-km discontinuity. The Ps converted phases propagate within the crust and the uppermost mantle only in the nearest vicinity of the

RI

seismograph stations. These stations are outside the related box (Fig. 5) and spread in a region,

SC

which is comparable in size with the station network. Therefore the residual that is accumulated in the Earth above the 410-km discontinuity and observed in a certain box may likely be close to

NU

the average residual for the station network. By using this logic we assume that outside the three

MA

strongly anomalous boxes the anomalies of travel times of P410s and P660s (Fig. 7 and Fig. 8) are dominated by the effect of volumetric velocity anomalies in the crust and upper mantle above the MTZ. We evaluate this effect as the average residual of travel times of either P410s or P660s

PT E

D

in ‘normal’ boxes. The estimates thus obtained are +0.5±0.3 s for P410s and +0.7 ±0.4 s for P660s. Both estimates are close and for the further calculations we adopt +0.6±0.3 s. After the removal of the anomalies of wave propagation above the MTZ, the travel time

CE

residuals in the south-central Tien Shan are -0.6±0.3 s and +0.9±0.3 s for P410s and P660s,

AC

respectively. The related anomalies of depth are -6±3 km and +9±3 km for the 410-km and 660km boundaries. For the anomalous box in the Fergana basin the residuals are +1.2±0.3 s and +0.1±0.3 s for P410s and P660s, respectively. The related anomalies of depth are +12±3 km and +1±3 km for the 410-km and 660-km boundaries, respectively. For the anomalous box in Tarim the anomalies are +1.3±0.3 s and -0.2±0.3 s, for P410s and P660s, respectively. The related anomalies of depth are +13±3 km and -2±3 km for the 410-km and 660-km boundaries, respectively. To conclude, under the south-central Tien Shan the 410-km boundary is uplifted and the 660-km boundary is depressed by several kilometers. As a result the width of the MTZ 5

ACCEPTED MANUSCRIPT increases by about 15 km. Under Fergana and Tarim the 410-km boundary is depressed by 12-13 km, whereas the depth to the 660-km boundary is close to the standard value. 3. Discussion and conclusions We have found three pronounced anomalies in the MTZ: 1 - under the south-central Tien Shan, 2 – Fergana and 3-Tarim.

PT

In the south-central Tien Shan the 410-km boundary is uplifted and the 660-km boundary

RI

is depressed by several kilometers. We evaluated the related temperature anomalies by using Clapeyron slopes of the related phase transitions. For rough calculations we assume 3.0 MPa/K

SC

for the 410-km discontinuity and -2.0 MPa/K for the 660-km discontinuity (Bina and Helffrich,

NU

1994 ). These slopes result in 10 K/km for the 410-km discontinuity and -16 K/km for the 660km discontinuity. Both, the uplift at a depth around 410 km and subsidence at a depth of 660 km

MA

in the south-central Tien Shan are indicative of anomalously low temperature in the MTZ, with the anomaly in a range from about -50 K to -150 K. The inferred temperature anomalies in the

D

MTZ can be used to evaluate the related velocity anomalies by using the approximate expression

PT E

(Karato, 1993):

δln Vs/δT = - 1.35 *10-4 K-1

CE

where T is temperature, and assuming δln Vs/ δln Vp = 1.7. For δT = -100 K the anomalies of Vs and Vp are 1.4% and 0.8%, respectively. Seismic tomography (Zhiwei et al., 2009) reveals at

AC

the same location a high-velocity body with the anomaly of Vp of 1% in the depth interval from 300 km to 600 km. This anomaly is clearly compatible with our estimate for δT = -100 K. A qualitatively similar high-velocity body is present in the tomographic model by Lei and Zhao (2007). The velocity anomalies would be doubled by a temperature anomaly of -200 K. In our interpretation of the anomalies of the differential time we neglected effects of possible volumetric velocity anomalies within the MTZ. This effect can be taken into account in the south-central Tien Shan. If similar (1%) anomalies are present in Vp and Vs, our estimate of the anomaly of thickness of the MTZ of 15 km is downward biased by 2-3 km, as well as the 6

ACCEPTED MANUSCRIPT depth to the 660-km boundary. The bias can be slightly larger if the anomaly of Vs is more than 1%. Moreover, the estimate of thickness of the MTZ can be downward biased if the actual diameter of the anomalous body is small relative to the size of the rectangle (220 x 160 km) in our calculations. This possibility is suggested by the tomographic model (Zhiwei et al., 2009). The quantitative estimates of the bias in the thickness of the MTZ are somewhat uncertain but

PT

qualitatively they are robust and the 660-km boundary may in reality be a few kilometers deeper than is obtained from the simplified calculations. The temperature anomaly at a depth around

RI

660 km may reach -200 K.

SC

We note that the residual of the differential time of +1.5 s corresponds very closely to the lowest S-wave velocity in the uppermost mantle (Fig. 9) and to the largest thickness of the crust

NU

(Fig. 10) as found by Vinnik et al. (2006). The anomalously low temperature in the MTZ, if

MA

taken together with the low S wave velocity in the upper mantle and the thick crust, is indicative of delamination of the mantle lithosphere (e.g., Houseman et al., 1981). In particular, the MTZ

D

can be cooled by the detached and sinking mantle lithosphere (Chen et al., 1997). The low

PT E

velocity in the upper mantle can be an effect of the replacement of the detached lithosphere by a hotter rock from the deeper mantle. If the detachment is coeval with the onset of orogenesis at 25-20 Myr (e.g., Sobel and Dumitru, 1997), the detached mantle lithosphere might reach the base

CE

of the MTZ with a speed of ~3 cm/yr. The low temperature in the MTZ might also be a relic of

AC

Paleozoic subduction of the oceanic lithosphere, but this scenario requires strong coupling and coherence between structures in the MTZ and at the Earth’s surface in spite of plate motions during the last 300 Myr. This looks not very likely but cannot be excluded (Jordan and Paulson, 2013; Vinnik et al., 2017). In the literature there are descriptions of small-scale basaltic volcanism in the Tien Shan of the age from 72 to 60 Myr (e.g., Sobel and Arnaud, 2000; Simonov et al., 2008). As there are no indications of rifting at that time, the magmatic activity is usually attributed to a mantle plume. The depression on the 410-km discontinuity under the Fergana basin is located in the 7

ACCEPTED MANUSCRIPT neighborhood of the region of the basaltic eruptions (Fig. 6). The corresponding positive temperature anomaly at a depth of 410 km is 100 - 150 K. Practically no topography is found at a depth of 660 km. The similar structures are often found in hot spots which are interpreted as manifestations of mantle plumes (e.g., Du et al., 2006). The lack of topography at a depth of 660 km may be explained by a superposition of two phase transition with opposite Clapeyron

PT

slopes in the hot mantle (Hirose, 2002). Note that most of the basalt flows are in the ‘normal’ block in the west-central Tien Shan, not in the anomalous south-central block. This difference

RI

is accompanied by a pronounced difference in age between the basalt flows and the uplift of the

SC

south-central Tien Shan.

A similar structure (the depression of 13 km on the 410-km boundary and flat 660-km

NU

boundary) is present under part of Tarim, but apparently without recent basaltic eruptions.

MA

However, the upper mantle model of Tarim at depths of 200-300 km contains a layer of low seismic velocity (Lei and Zhao, 2007). Lateral extent of this anomaly is large relative to the

D

depressed region on the 410- km boundary, in agreement with the expected mushroom shape of

PT E

the plume.

In summary, the seismic observations support the idea of lithospheric delamination in the

CE

central region of the Tien Shan and small-scale plumes in peripheral regions. Acknowledgment

AC

This study was partially supported by the RFBR grant 15-05-04938. The seismograms were obtained from the IRIS DMC. The authors appreciate comments from an anonymous reviewer. References Bina, C.R. , Helffrich, G., 1994. Phase transition Clapeyron slopes and transition zone seismic discontinuity topography. Journ. of Geoph. Res., 99(B8), 15853-15860.

8

ACCEPTED MANUSCRIPT Bullen, M.E., Burbank, D.W., Garver, J.I. and Abdrakhmatov, K.Y., 2001. Late Cenozoic tectonic evolution of the northwestern Tien Shan: New age estimates for the initiation of mountain building. Geologic Society of America Bulletin, 113(12), 1544-1549. Chen, Y.H., Roecker, S.W., Kosarev, G.L., 1997. Elevation of the 410 km discontinuity beneath the central Tien Shan: Evidence for a detached lithospheric root. Geophysical Res. Letters,

PT

24(12), 1531- 1534.

Du, Z., Vinnik, L.P. and Foulger, G.R., 2006. Evidence from P-to-S mantle converted waves for

RI

a flat “660-km” discontinuity beneath Iceland. Earth and Planetary Science Letters, 241(1-2),

SC

271-280.

NU

Jordan, T.H. and Paulson, E.M., 2013. Convergence depths of tectonic regions from an ensemble of global tomographic models. Journal of Geophysical Research: Solid Earth, 118(8), 4196-

MA

4225.

Hirose, K., 2002. Phase transitions in pyrolitic mantle around 670‐km depth: Implications for

D

upwelling of plumes from the lower mantle. Journal of Geophysical Research: Solid Earth,

PT E

107(B4). ), DOI: 10.1029/2001JB000597.

CE

Houseman, G.A., McKenzie, D.P. and Molnar, P., 1981. Convective instability of a thickened boundary layer and its relevance for the thermal evolution of continental convergent belts.

AC

Journal of Geophysical Research: Solid Earth, 86(B7), 6115-6132. Karato, S., 1993. Importance of anelasticity in the interpretation of seismic tomography. Geophysical Research Letters, 20(15), 1623-1626. Lei, J. and Zhao, D., 2007. Teleseismic P-wave tomography and the upper mantle structure of the central Tien Shan orogenic belt. Physics of the Earth and Planetary Interiors, 162(3-4), 165185. Molnar, P and Tapponnier, P., 1975. Cenozoic tectonics of Asia: effects of a continental collision. Science, 189(4201), 419-426. 9

ACCEPTED MANUSCRIPT Reigber, C., Michel, G.W., Galas, R., Angermann, D., Klotz, J., Chen, J.Y., Papschev, A., Arslanov, R., Tzurkov, V.E. and Ishanov, M.C., 2001. New space geodetic constraints on the distribution of deformation in Central Asia. Earth and Planetary Science Letters, 191(1-2), 157165. Roecker, S.W., Sabitova, T.M., Vinnik, L.P., Burmakov, Y.A., Golvanov, M.I., Mamatkanova,

PT

R. and Munirova, L., 1993. Three‐dimensional elastic wave velocity structure of the western and

RI

central Tien Shan. Journal of Geophysical Research: Solid Earth, 98(B9), 15779-15795.

SC

Simonov, V.A., Mikolaichuk, A.V., Rasskazov, S.V. and Kovyazin, S.V., 2008. Cretaceous-

Geology and Geophysics, 49(7), 520-533.

NU

Paleogene within-plate magmatism in Central Asia: data from the Tien Shan basalts. Russian

MA

Sobel, E.R. and Dumitru, T.A., 1997. Thrusting and exhumation around the margins of the western Tarim basin during the India-Asia collision. Journ. of Geophys. Res, 102(B3), 5043-

D

5063.

PT E

Sobel, E.R. and Arnaud, N., 2000. Cretaceous–Paleogene basaltic rocks of the Tuyon basin, NW China and the Kyrgyz Tian Shan: the trace of a small plume. Lithos, 50(1-3), 191-215.

CE

Tian, X., Zhao, D., Zhang, H., Tian, Y. and Zhang, Z., 2010. Mantle transition zone topography

AC

and structure beneath the central Tien Shan orogenic belt. Journal of Geophysical Research: Solid Earth, 115(B10), doi:10.1029/2008JB006229. Vinnik, L.P. and Saipbekova, A.M., 1984. Structure of the lithosphere and the asthenosphere of the Tien Shan. Annales Geophysicae, 2(6), 621-626. Vinnik, L.P., Aleshin, I.M., Kaban, M.K., Kiselev, S.G., Kosarev, G.L., Oreshin, S.I. and Reigber, C., 2006. Crust and mantle of the Tien Shan from data of the receiver function tomography. Izvestiya Physics of the Solid Earth, 42(8), 639-651.

10

ACCEPTED MANUSCRIPT Vinnik, L.P., Oreshin, S.I. and Makeyeva, L.I., 2017. Siberian traps: Hypotheses and seismology data. Izvestiya, Physics of the Solid Earth, 53(3), 332-340. Zabelina, I.V., Koulakov, I.Y. and Buslov, M.M., 2013. Deep mechanisms in the Kyrgyz Tien Shan orogen (from results of seismic tomography). Russian Geology and Geophysics, 54(7), 695-706.

PT

Zhiwei, L., Roecker, S., Zhihai, L., Bin, W., Haitao, W., Schelochkov, G. and Bragin, V., 2009.

RI

Tomographic image of the crust and upper mantle beneath the western Tien Shan from the MANAS broadband deployment: Possible evidence for lithospheric delamination.

SC

Tectonophysics, 477(1-2), 49-57.

NU

Windley, B.F., Allen, M.B., Zhang, C., Zhao, Z-Y., Wang, G-R., 1990. Paleozoic accretion and

AC

CE

PT E

D

MA

Cenozoic redeformation of the Chinese Tien Shan Range, central Asia. Geology, 18(2), 128-131.

11

ACCEPTED MANUSCRIPT

44 – 46

AC

CE

PT E

D

MA

Table 1. Delay times of P410s and P660s

PT

tP660s (s) 67.9 69.3 68.1 68.3 68.6 69.0 69.4 68.8 68.5 68.8 68.8 68.6 68.6 68.1 68.3

RI

42 – 44

tP410s (s) 44.1 44.9 44.4 45.9 45.8 44.5 44.0 44.6 44.3 44.6 44.8 44.7 44.1 44.3 44.7

SC

40 - 42

Longitudes (°) 72 - 76 76 – 78 78 - 80 80 – 82 72 – 74 74 – 76 76 - 78 78 - 80 80 - 82 72 - 74 74 - 76 76 - 78 78 - 80 80 - 82 74 - 76

NU

Latitudes (°) 38 – 40

12

tP660s - tP410s (s) 23.8 24.4 23.7 22.4 22.8 24.5 25.4 24.2 24.2 24.2 24.0 23.9 24.5 23.8 23.6

ACCEPTED MANUSCRIPT Highlights ■ Detached lithosphere in the transition zone of the central Tien Shan. ■ Evidence of thermal mantle plume beneath the Fergana Basin.

AC

CE

PT E

D

MA

NU

SC

RI

PT

■ Evidence of thermal mantle plume beneath part of Tarim.

13

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10