The relationship between crust-lithosphere structures and seismicity on the southeastern edge of the Tibetan Plateau

Highlights

• The Moho and LAB discontinuities depth display the opposite distribution characteristics around the Sichuan Basin.

• LAB could be used as an indicator for the size and direction of lower crustal flow.

• Lower crustal earthquakes linked to both the aftershocks of normal earthquakes and the lower crustal phase transformation.

• Seismicity has a good consistency with crust-lithosphere deformation.

Abstract

The crust-lithosphere structure, especially the discontinuities such as Moho and lithosphere-asthenosphere boundary (LAB), is important in the investigation of geodynamic process implications and tectonic evolution of the lithosphere and regional seismic activity. We stacked the S receiver functions from 51 permanent broad-band stations to investigate the crust-lithosphere structures beneath the southeastern edge of the Tibetan Plateau and further discussed the mechanism of lower-crust earthquakes, the regional tectonics deformation characteristics, and the relationship between the two discontinuities and seismicity in this region. The results show that the Moho depth increases from 40 to 52 km beneath the Sichuan Basin to 56–74 km on the both western side of the Longmenshan (LMS) fault and Anninhe-Zemuhe fault. The LAB depth ranges from 130 to 170 km beneath the Sichuan Basin, and a shallow belt ranges from 100 km to 130 km around the Sichuan Basin. On the Moho depth contour map, 374 moderate earthquakes that accounted for 72.2%, corresponding to the depth range of 44–52 km and 60–68 km on the both sides of a narrow strip (longitude 102°~104°). On the LAB depth contour map, 321 earthquakes accounted for 62.15% in the depth range of 130–150 km. The variation of LAB corresponds to the increasing crustal thickness, intensive lower crustal earthquakes (hypocenters about 60 km and magnitude 4.0 ≤ Ms. ≤ 4.9) and surface abrupt elevations. It implied that Normal strong earthquakes generate the lower crustal earthquakes or aftershocks that resulted in the shear zones where fluid-induced metamorphic transformations, the eclogitization of dry granulite, and further affect the crustal thickness and deformation to control the spatial distribution of deep faults and seismicity.

Introduction

The continuous convergence and collision between the India and Eurasian plates generated the Tibetan Plateau which has taking place over the past 50–60 million years (Molnar and Tapponnier, 1975; Yin and Harrison, 2000). It has resulted in the uplift of the Tibetan Plateau and at least 1500 km of crustal shortening within the collision region, and gives rise to a huge mass of nearly 2 × 10⁷ km² with an average elevation of 4500 m (Royden et al., 2008; Chen et al., 2010). The lifting of the Tibetan Plateau has been accompanied of a small crustal shortening from the center to the southeastern margin since 4 million years ago (Klemperer, 2006). The southeastern edge of the Tibetan Plateau has a sharp rise and fall of topography with complex geological structures and extreme development of deep faults (England and Molnar, 1997; Zhang et al., 2009). Since Quaternary, especially the late Quaternary, this region has experienced deep fault activity: frequent earthquakes have occurred, and considerable loss of life and property has been caused by extremely destructive earthquakes (Wu et al., 2008; Teng et al., 2009; Liu et al., 2013).

The Moho is consensus for a definition as “the interface between the crust and the mantle, marking the base of the crust”. This discontinuity was firstly discovered when the meteorologist and seismologist Andrija Mohorovičić found a depth near 50 km jump of seismic wave speeds by careful analysis seismic waves for an earthquake near Pokupsko in 1911. It has aroused the interest of more seismologists, and they have done a lot of subsequent research work across the globe. The compressional wave velocity first exceeds 7.6 km/s at the level (Steinhart, 1967). The thick crust is interpreted in cratonic regions with magmatic modification in rift systems and Basins province (Keller, 2013). Carbonell et al. (2013) stress that the tectono-magmatic process was important for the structure of Moho, including orogenesis, orogen collapse, and underplating. The surprising conclusions that the Moho was a seismic and electrical transition zone, and its seismic and electrical properties of lower crustal and mantle rocks significantly affected by the pressure, temperature and metamorphism based on laboratory measurements (Wang et al., 2013). Ranalli and Adams (2013) suggest that compositional stratification enhances the rheology and strength of Moho, especially higher in compressional than extensional tectonic regimes.

The relation between Moho and magmatic underplating in continental lithosphere directly affects the regional geological tectonics and seismicity. The underlying geodynamic process has been debate for decades, a central issue being the physical-chemical-mechanical nature of the lower crust and the upper mantle (Chen and Yang, 2004; Jackson et al., 2008; Searle, 2013). Jackson (2002) suggested that the strong upper crust was overlying on a weak upper mantle. Mareschal and Jaupart (2013) found that there is no correlation between heat flow and crustal thickness by comparing their model of radiogenic heat production for crustal geotherms. O'Reilly and Griffin (2013) discussed the relation between the Moho and crust-mantle boundary, and proposed that a transitional crust-mantle boundary may change in depth and thickness through time due to the crust-formation process and composition of the convecting mantle. The metamorphic transformations caused by underlying hot flow has significant influence on the petrophysical properties around the Moho, including four main processes: formation of granulite facies areas, transformation of granulites to eclogites, retrogression of eclogite facies rocks to amphibolite and green schist facies rocks, and the spinel lherzolite to granet lherzolite transition (Austrheim, 2013). Mjelde et al. (2013) reviewed the knowledge focus on the lower crust eclogite and upper mantle peridotites, and confirm that Moho generally represent the top of the mantle, lower crust eclogites bodies would significantly increase the crust thickness. In southern Tibet, lower-crustal material phase transition was confirmed from the eclogitization of dry granulite based on the laboratory experiments (Shi et al., 2018).

Seismicity is the most indicator of mechanical state of Earth's interior, especially at the level of crust-mantle discontinuities. The earthquakes could occur over a wide range of depths, including the entire crust and the uppermost mantle (Wang et al., 2013). The seismicity distribution in southern Tibet is more or less continuous from surface to as deep as 100 km (De la Torre et al., 2007; Priestley et al., 2008). The crucial issue is how can earthquakes occur in the lower crust and upper mantle, it is an enigma question regarding the mechanism of the earthquakes and the strength of material. Austrheim, (2013) proposed earthquakes and tremors result in fluid flow and high-pressure metamorphism in the zone between a metastable dry and strong lower crust and the upper mantle, and earthquakes and tremors may record ongoing metamorphic transitions. Jamtveit et al. (2018) found that the normal earthquakes in the seismogenic zone produce stress pulse that drives aftershocks in the lower crust. A cloud of aftershocks generate the shear fracture zones that provide the infiltration pathways for fluids in the upper mantle for the India-Tibet collision region. Meanwhile, fluid-induced metamorphic transformations formed ecogites and amphibolites in shear zones, breccias along fractures. So fluid-induced metamorphism is directly associated with the dynamics of the lithosphere and Seismicity. The hypocentres of early aftershocks are twice that of the usual seismicity (Ben-Zion and Lyakhovsky, 2006). Shi et al. (2018) suggested that lower crust earthquakes about 50-80 km in Southern Tibet are linked to eclogitization of dry metastable granulite.

Sub-Moho seismicity induced the phase transition of lower crustal materials, rock melting and subsequent solidification. Thus, a strengthening process further affects the spatial distribution conditions of crust-mantle discontinuities and shear fracture zones, and then even controls the faults and seismicity above the crust-mantle transition belts. In summary, the depth and topography of the Moho could reflect the brittle upper-crust deformation characteristics. The size and direction of underlying hot asthenosphere flow could be estimated from the lithosphere-asthenosphere boundary (LAB) depth variation (Sol et al., 2007; Kind et al., 2012). The two discontinuities lateral conditions has important significance to further study the regional geodynamic evolution, deformation characteristics, geological tectonics and seismicity.

The crust is dragged by the hot plastic fluid in the uppermost mantle, and the lithosphere is formed as only a weak interface between the lower crust and upper mantle lid above the asthenosphere in seismic earth models. Tomography and surface wave dispersion method could be used to detect the crust-mantle structure at a large scale and low resolution. Zhang et al. (2007) obtained the crust, lithosphere and asthenosphere discontinuities by using surface wave dispersion with group velocity dispersion inversion for the Qinghai-Tibet Plateau and its adjacent areas, but the scale was too large. Li et al. (2010) found that the movement direction of the Sichuan-Yunnan block was blocked by the thick and hard lithosphere below the Sichuan Basin when using ambient noise Rayleigh wave phase velocity tomography. However, both horizontal and vertical resolutions are low for surface wave tomography to detect lithosphere (Hu et al., 2008; Li et al., 2008). Body wave tomography has relatively high resolution but is not sensitive to the vertical velocity, and the vertical resolution is limited (McKenzie and Priestley, 2008; Priestley and Tilmann, 2009).

Numerical simulation is applicable for linear and nonlinear problems to handle complex physical problems with the advantages of low cost, rapid calculation and clear physical concepts. Wang et al. (2007) established a three-dimensional finite element model to discuss the dynamic mechanism in the Sichuan-Yunnan region by simulating the current crustal movement and stress distribution, but this method has too many assumptions, the reliability of the results is rather difficult to verify. In the last decades, the P-wave receiver function has been used to calculate a relatively accurate crust thickness, but the depth of the lithosphere is not convincing (Langston, 1977; Lawrence and Shearer, 2006). Yang et al. (2009) calculated the crustal thickness and S-wave velocity in western Sichuan using the P-wave receiver function, and the authors found that the region is a tectonic environment for large earthquakes in the future.

With respect to the P-wave receiver function, the Ps converted wave is often submerged in the multiple reflected waves from shallow discontinuities. Consequently, the Ps converted wave signal is very weak, or even false, and difficult to identify for deeper discontinuities (Li et al., 2000; Kumar et al., 2005). The depth of lithosphere interface was accurately calculated by the S-wave receiver function, and it could eliminate the disadvantages of the P-wave receiver function for lithosphere detection (Yuan et al., 2006; Hu et al., 2015). By contrast, Sp converted waves will generate as the S-wave propagates through discontinuities within the crust-mantle transition zone. Due to the fact that the S-wave velocity is faster than P-wave velocity, Sp converted waves reach the station prior to the S-waves (Kind et al., 2012; Yang et al., 2017). Thus, all other multiple reflected waves from shallow discontinuities followed the S-wave. The Sp converted wave is relatively independent without any interference from the reflected waves. The S-wave receiver function is the most effective way to calculate the discontinuities within the crust-mantle.

In this study, we use S receiver functions from 51 permanent broad-band seismic stations to estimate the Moho and LAB on the southeastern edge of the Tibetan Plateau. A total of 518 moderate earthquakes between 1970 and 2017 were collected from the Chinese State Seismological Scientific Data Sharing Center. Finally, the relationship between crust-lithosphere structures and spatial distribution characteristics of earthquakes was discussed. We suggested that the LAB could be used as an indicator for the size and direction of lower crustal flow, Lower crustal earthquakes may be linked to both the aftershocks of normal earthquakes and the lower crustal transformation, and the regional seismicity have a good consistency with crust-lithosphere deformation.