Gravity anomalies of the crust and upper mantle for central and South Asia

Abstract

Studying the density of both the crust and mantle is one of the topical problems in modern geophysics. Gravity modeling in combination with seismic tomography is an important tool for detecting density inhomogeneities in the crust and mantle, which can cause stresses and thus significantly impact the regional tectonics [Pogorelov, Bara-nov, 2010], especially in zones wherein continental margins actively interact with subducting oceanic plates and the entire depth of the tectonosphere is subject to stresses. Associated processes lead to considerable horizontal and ver-tical stresses that often cause catastrophic events on a global scale. The challenge of studying the global tectonic pro-cesses in the Earth's tectonosphere can be addressed by gravity modeling in combination with seismic surveying. Data from previous studies. I.L. Nersesov et al. [1975] pioneered in calculating the spatial pattern of mantle den-sity inhomogeneities in Central Asia. Although the accuracy of their estimations was not high due to the limited data-base, their study yielded significant results considering the structure of the crust. Numerous subsequent geophysical projects have researched the crust to a level sufficient to develop regional models, that can give quite adequate infor-mation on the depths of external and internal boundaries of the crust and suggest the distribution patterns of seismic velocities and density values. With reference to such data, mantle density inhomogeneities can be studied with higher accuracy. This paper reports on the estimations of gravity anomalies in the crust and upper mantle in Central and South Asia. The study region represents the full range of crust thicknesses and ages, as well a variety of crust formation ty-pes [Christensen, Mooney, 1995]. We used the 3D gravity modeling software package 3SGravity developed by Senachin [2015a, 2015b] that considers the spherical shape of the Earth's surface, and estimated gravitional anomalies using Baranov's digital model of the crust, AsCrust [Baranov, 2010]. The study area includes the Alpine-Himalayan folded belt, the triple junction of rift zones in North Africa, and the marginal seas of Southeast Asia, which are framed by deep troughs with associated volcanic belts. Its relief ranges from the highest mountains in Himalayas to deepest troughs in Indonesia. In this region, the collision of the Indian and Asian plates causes thrusting at the Asian plate margin which results in thickening of the continental crust [Oreshin et al., 2011]. This process may be accompanied by the separation of the crustal layer of the Indian lithosphe-ric plate from its mantle 'cushion', i.e. delamination, the mechanism of which is not fully understood [Jiménez-Munt et al., 2008; Krystopowicz, Currie, 2013; Ueda et al., 2012] (Fig. 1). AsCrust, the digital model of the Earth's crust: depth to Moho map. A large volume of new data on reflection, refraction and surface waves from earthquakes and explosions was analyzed and integrated into the AsCrust model (1×1° grid). Ten digital maps were constructed: Moho depth, the upper, middle and lower crustal layers, as well as Vp velocities and densities in these layers [Baranov, 2010]. In our study, we calculated gravitational anomalies from the values of thicknesses and density of crustal layers at each point of the grid. The density in the layers was calculated from longitudinal wave velocities using the formula described in [Brocher, 2005] (Fig. 2). The algorithm for gravity anomaly calculations. Modeling the gravity of large regional objects needs to take into account the curvature of the Earth's surface. Algorithms for calculating the gravity field from bodies bounded by spherical surfaces are proposed in [e.g. Kosygin et al., 1996; Starostenko et al., 1986; Strakhov et al., 1989; Jones et al., 2010; Li et al., 2011; Schmidt et al., 2007]. In this study, we used an algorithm based on equations for direct calcula-tions of the gravity effect, which can be obtained for specific points located on the pole of the sphere. Such equations considerably simplify the algorithm, but require constant recalculation of the coordinate system for each calculation point, which complicates the task (Fig. 3). Source data, and methods of gravity anomaly calculations. Our computational model includes seven layers: an water layer, three sedimental layers (depths of boundaries, and density values of the sedimental layers) from the model described in [Laske, Masters, 1997], and three crustal layers (depths of boundaries, and density values of the crust, which were estimated from velocities Vp) from the AsCrust model [Baranov, 2010], considering the territory covered by the model. For the surrounding regions, data on the structure and properties of the crust were taken from the CRUST 2.0 model [Bassin et al., 2000] and interpolated to the 1×1° grid. Thus, data with the resolution of 1×1° were used to describe the sediments and the crust, and data with the resolution of 0.1×0.1° characterized the water layer (batimetry). Model GGM01 based on satellite observation data of the GRACE project (http://www.csr.utexas.edu) simulated the Earth's gravity field and was used to calculate anomalies in 'free' air across the entire surface of our model, which took into account the correction for the elevation of an observation object. The gravity field ranges from -250 to +260 mGal. The zone of collision of the Indian and Asian plates is marked by narrow parallel anomalies of different signs, reaching 200 mGal and more. The southwestern zone with negative anomalies corresponds, apparently, to the boun-dary of the junction zone of the two plates, wherein the Indian plate subducts underneath the Asian plate, as de-scribed in [e.g. He et al., 2010; Oreshin et al., 2011]. The gravity field of the study area quite clearly shows that Tibet is separated from the Tarim plate neigbouring it in the northeast. This separation is marked by a negative anomaly to -150 mGal, the boundaries of which are outlined by narrow zones of positive anomalies. The southern Caspian Sea is also characterized by a negative anomaly to -150 mGal, while Tien Shan is marked by a narrow band of positive anomalies up to 110 mGal. In most of the study area, the field is close to normal and varies within a few dozens of mil-ligals. Moderately positive gravity (within 40÷80 mGal) is typical of the rest of the Alpine-Himalayan folded belt. A slight positive gravity field is revealed in the marginal seas of Southeast Asia, wherein there are two narrow zones of high-amplitude anomalies of different signs (up to 200 mGal), which are generated by isostatically uncompensated systems of island arcs and trenches (Fig. 6). The gravity effect of the Earth's crust estimated for Asia shows the presence of major anomalies varying in the range of 940 mGal (from -380 to +560 mGal). The maximum positive anomaly is located in the vicinity of the African triple junction of the rift zones, wherein the anomaly reaches a positive maximum of about +560 mGal. Positive ano-malies are also revealed in the Tarim Basin (+130 mGal), Southeastern China (+100 mGal), the Iranian plateau (+180 mGal), and back-arc subduction zones of the Indian and Pacific plates (+290 mGal). Large negative anomalies corre-spond to the Caspian and Black Seas (-380 mGal), Himalayas (-280 mGal), and eastern Tibet (-330 mGal). The East-ern Mediterranean is characterized by a negative anomaly (-310 mGal). The eastern Arabian Peninsula and the Mesopotamian lowlands are characterized by negative anomalies up to -220 mGal. The map of calculated crustal gravity anomalies also shows submarine ridges (+280 mGal) that trend from south to north and seem to trace 'hot spots' that burn through the lithospheric plate (Fig. 7). Gravitational anomalies in the mantle were calculated by subtracting the gravity effect of the crust from the ob-served gravity field. The anomalies range from -570 to +350 mGal, which is about twice the range of variations of this field. This directly indicates the presence of large density variations in the lithospheric mantle, which should compen-sate for the anomalous crustal masses. The largest positive mantle density inhomogeneities in the study region are revealed in the narrow band of the Himalayas (+330 mGal) and Eastern Tibet (+350 mGal). In the Caspian and Black Seas, the anomalies reach +250 and +300 mGal, respectively. The Eastern Mediterranean is characterized by a positive anomaly up to +280 mGal. The eastern Arabian Peninsula and the Mesopotamian lowlands are characterized by posi-tive anomalies of up to +220 mGal. Negative anomalies are revealed in the Tarim Basin (-190 mGal), over submarine ridges in the Indian Ocean (-340 mGal), in Southeastern China (-120 mGal), the central Hindustan (-80 mGal), the Hindu Kush and Karakoram (-150 mGal). Subduction zones of the Indian and Pacific plates are also characterized by negative anomalies of up to -250 mGal. The triple junction zone (Red Sea, Gulf of Aden, the African Rift) in the north-eastern African continent is the region of maximum negative anomalies in the mantle wherein gravity values are re-duced to -570 mGal (Fig. 8). Results and conclusion. By applying the 3SGravity software package and the AsCrust digital model, we revealed the spatial pattern of gravitational anomalies in the crust and mantle in Central and South Asia, which gives more pre-cise information about the variations in density with depth in the study area. Our estimations show a significant varia-tions of mantle gravity anomalies, several times larger than the changes in the observed anomalies.