The mechanism for uplift of the Tibetan Plateau and the mechanism for maintaining the high gravitational potential of the plateau have been a focus of discussion for decades in the earth science community around the world. One theory based on gravity data, topographic data, and flexural modeling proposes that the Tibetan Plateau is partially supported by the lithosphere of the surrounding lowlands, including the Indian plate, the Tarim Basin, and the Sichuan Basin, in the form of regional compensation. We present two-dimensional (2-D) transects of the lowlands to the highlands around the plateau to review the lateral compensation and weakening of the Indian plate when it is subducting beneath the Tibetan Plateau in the south, the possible support of the Tarim lithosphere while it is underplating the northern edge of the Tibetan Plateau, and the possible lateral compensation style beneath the eastern plateau associated with the western foredeep of the Sichuan Basin. Lateral variation of lithosphere strength is traditionally calculated based on the coherence theory in the spectral domain. This theory requires the study areas to be rectangular and the elastic thickness within each rectangular box to be constant. However, a rectangular geometry is hardly consistent with any natural shape of a tectonic unit. In this paper, we calculate lateral variation in lithosphere strength consistent with the natural shapes of tectonic units, such as the Himalayas, Western Kun Lun Shan, Central Tibetan Plateau, Qilian Shan, Tarim Basin, Tian Shan, and so on, based on three-dimensional (3-D) flexural modeling in the spatial domain. Flexural modeling from gravity data around the Tibetan Plateau is mainly based on the way in which the Moho deflects from the deformed and weakened lithosphere in convergent plate boundaries. However, flexural modeling hardly differentiates among the variations in architecture of deformation in the upper crust, the lower crust, and the upper mantle lithosphere. In order to further understand the present architecture and the evolution of Tibetan lithosphere, an integrated study using gravity data, surface faulting, structural modeling, seismic imaging, deep seismic sounding, and earthquake events is also discussed. Rheological modeling using a simple geothermal structure at various locations over the Tibetan Plateau reveals that the base of the strong upper crust of Tibet is at 30 to 35 km depth. The tendency for the strong upper crust to flow on the weak, ductile lower crust (or middle and lower crusts) depends on the development of the thickness of the weak lower crust. In other words, the weakness depends on the thickness difference between the base of the upper crust and the Moho depth. The larger the difference is, the easier it is for the upper crust to flow relative to the strong upper mantle lithosphere. Therefore, improvements in Moho depth calculations based on 3-D spatial flexural modeling will help to clarify the mobility of the lower-crustal flow of the Tibetan plateau.

Our inability to determine the growth history and growth mechanism of the Tibetan plateau may be attributed in part to the lack of detailed geologic information over remote and often inaccessible central Tibet, where the Eastern Kunlun Range and Qaidam basin stand out as the dominant tectonic features. The contact between the two exhibits the largest and most extensive topographic relief exceeding 2 km across their boundary inside the plateau. Thus, determining their structural relationship has important implications for unraveling the formation mechanism of the whole Tibetan plateau. To address this issue, we conducted field mapping and analysis of subsurface and satellite data across the Qimen Tagh Mountains, part of the western segment of the Eastern Kunlun Range, and the Yousha Shan uplift in southwestern Qaidam basin. Our work suggests that the western Eastern Kunlun Range is dominated by south-directed thrusts that carry the low-elevation Qaidam basin over the high-elevation Eastern Kunlun Range. Cenozoic contraction in the region initiated in the late Oligocene and early Miocene (28–24 Ma) and has accommodated at least 48% upper-crustal shortening (i.e., ∼150 km shortening) since that time. In order to explain both the high elevation of the Eastern Kunlun Range and the dominant south-directed thrusts across the range, we propose that the Cenozoic uplift of the range has been accommodated by large-scale wedge tectonics that simultaneously absorb southward subduction of Qaidam lower crust below and southward obduction of Qaidam upper crust above the Eastern Kunlun crust. In the context of this model, the amount of Qaidam lower-crust subduction should be equal to or larger than the amount of shortening across Qaidam upper crust in the north-tapering thrust wedge system, thus implying at least 150 km of Qaidam lower-crust subduction since the late Oligocene and early Miocene. The late Oligocene initiation of contraction along the southern margin of Qaidam basin is significantly younger than that for the northern basin margin in the Paleocene to early Eocene between 65 and 50 Ma. This temporal pattern of deformation indicates that the construction of the Tibetan plateau is not a simple process of northward migration of its northern deformation fronts. Instead, significant shortening has occurred in the plateau interior after the plateau margins were firmly established at or close to their current positions. If Cenozoic crustal deformation across the Eastern Kunlun and Qaidam regions was accommodated by pure-shear deformation, our observed >48% upper-crustal shortening strain is sufficient to explain the current elevation and crustal thickness of the region. However, if the deformation between the upper and lower crust was decoupled during the Cenozoic Indo-Asian collision, lower-crustal flow or thermal events in the mantle could be additional causes of plateau uplift across the Eastern Kunlun Range and Qaidam basin.