The Tibetan Plateau is the product of crustal thickening caused by collision between India and Asia. Plate tectonic reconstructions suggest continuous northward movement of the Indian plate relative to stable Eurasia at nearly 50 mm/yr for the last 50 My. The plateau is now at ~5 km elevation with steep topographic gradients across the southern and northern margins. These gradients are also associated with large lateral variations in geoid and gravity anomalies. Uplift late in the tectonic evolution of the plateau, the widespread extension, and the associated magmatism have been attributed to removal of the lower part of lithospheric mantle and its replacement by hotter and lighter asthenosphere.

Here we present a two-dimensional lithospheric thermal and density model of the present day structure and numerical modeling of the evolution of the Tibetan Plateau:
- The two-dimensional lithospheric model is along a transect from the Indian plate to Asia, crossing the Himalaya front and the Tibetan Plateau. The model is based on the assumption of local isostatic equilibrium, and is constrained by the topography, gravity and geoid anomalies and by thermal data within the crust. Our results suggest that the height of the Tibetan Plateau is compensated by thick crust in the south and by hot upper mantle to the north. The Tibetan Plateau as a whole cannot be supported isostatically only by thickened crust; a thin and hot lithosphere beneath the northern Plateau is required to explain the high topography, gravity, geoid and crustal temperatures.
- We also investigate numerically the long-term (50 My) evolution of crustal and lithospheric thickness, thermal structure, topography, and strain-rate of the Tibetan plateau through time, using a planform viscous approach. This suggests that lithospheric mantle must have been removed from beneath Tibet, and that the crust must have been warmed and weakened by an increase of radiogenic heat production at depth due to crustal thickening.

1. Influence of mantle dynamics on the topographic evolution of the Tibetan Plateau: Results from numerical modelling

(Publication: pdf)

We investigate numerically the evolution of crustal and lithospheric thickness, thermal structure, topography, and strain-rate of the Tibetan plateau through time, using the thin viscous sheet approach. We show that lithospheric mantle must have been removed from beneath Tibet to explain the present surface elevation and lack of regional surface slope. In the absence of this removal, the modelled topography reaches a maximum elevation of <4000 m (for weak rheology), or the surface slopes significantly northwards (for strong rheology). The crust must have been warmed and weakened by an increase of radiogenic heat production at depth due to crustal thickening. In the absence of this warming, viscous stresses associated with plate convergence exceed stresses produced by topography, and the present pattern of vertical thinning and east-west extension would not have developed. Continuous removal of lithosphere, by delamination or convection, does not allow sufficient crustal warming, and fails to reproduce either the present topography or the pattern of active deformation on the plateau in a reasonable time. Geologically rapid removal of the lithospheric root beneath the thickened crust of Tibet successfully explains the current elevation of the plateau, its lack of surface slope, the steep south and north margins, and the pattern of the present deformation, including vertical thinning, E-W extension, and extrusion and vertical axis rotation on the eastern margin. Our modelling suggests that this removal took place within the last 12 m.y.

Tectonic map, colour shades show elevation (1000, 3000, and 5000 m contours). ATF: Altyn Tagh Fault, MBT: Main Boundary Thrust of Himalaya, JF: Jiali Fault Zone, KF: Karakorum Fault, ITS: Indus-Tsangpo Suture, BNS: Bangong-Nujiang Suture.

An example of the evolution of the geotherm before and after the removal of lithospheric mantle. a) Time evolution of a geotherm under constant thickening and at different stages: 1) Initial, 2) after 40 m.y. of constant thickening, 3) After removal of the lithospheric mantle, 4) 2 m.y. after the lithospheric removal, resulting from conduction and constant thickening. b) Time evolution of: lithosphere and crustal thickness; Moho temperature and lithospheric mantle density; elevation and changes in gravitational potential energy (GPE(t) - GPE(t=0)).

Model results after 57 m.y. of convergence at 50 mm/yr and 7 m.y. after removal of lithospheric mantle below the 700°C isotherm where thickness exceeds 155 km. Radiogenic heat production: H=2.5 exp(-z/15) µW/m3 and H=2.5 exp(-z/55) µW/m3 if the crust is thicker than 40 km. a) Elevation predicted from the model (contours) and measured (grey pattern); b) crustal thickness predicted from the model (contours) and data (grey pattern, compiled by Bassin et al., 2000); c) lithospheric mantle thickness predicted from the model.

Model results after 57 m.y. of convergence at 50 mm/yr and 7 m.y. after removal of lithospheric mantle below the 700°C isotherm where thickness exceeds 155 km. a) Horizontal velocity field (arrows), vertical strain rate (color pattern, positive for thickening and negative for thinning) and crustal thickness (contours every 10 km). Grey outline is the region of the plots on Figs. b and c. b) Horizontal principal stress directions (arrows for extension and bars for compression), tectonic regime (color pattern) and elevation (contours every 0.5 km). Note that this plot is a zoom on the plateau (grey outline Fig.a). c) Vorticity (color pattern in rad/s), vertical axis finite rotations (circled numbers in degrees) and elevation (contours in km). This plot is also a zoom on the plateau (grey outline Fig.a).

2. Lithosphere structure underneath the Tibetan Plateau inferred from elevation, gravity and geoid anomalies

(Publication: pdf)

The steep topographic gradients are also related to large lateral variations in the geoid and gravity anomalies. In a SSW to NNE cross section, the Bouguer gravity anomaly decreases over a distance of 500 km from about 0 mGal in the India plate to ~ -500 mGal on the Tibetan Plateau. The geoid anomaly also presents steep gradients on both the Himalayan front and the northern margin, reaching values between 20-30 m on the plateau, suggesting a pronounced thinning of the lithospheric mantle. Uplift late in the tectonic evolution of the plateau, the widespread extension, and the associated magmatism have been attributed to convective removal of the lower part of lithospheric mantle and its replacement by hotter and lighter asthenosphere. Here we present a two-dimensional lithospheric thermal and density model along a transect from the Indian plate to Asia, crossing the Himalaya front and the Tibetan Plateau. The model is based on the assumption of local isostatic equilibrium, and is constrained by the topography, gravity and geoid anomalies and by thermal data within the crust. Our results suggest that the height of the Tibetan Plateau is compensated by thick crust in the south and by hot upper mantle to the north. The Tibetan Plateau as a whole cannot be supported isostatically only by thickened crust; a thin and hot lithosphere beneath the northern plateau is required to explain the high topography, gravity, geoid and crustal temperatures. The lithosphere reaches a maximum depth of ~260 km beneath the southern Plateau, and thins abruptly northward to ~100 km under the central and northern Plateau. The lithosphere depth increases again beneath the Qaidam basin and the Qilian Shan to ~160 km.

Data. a) topography; b) geoid; c) Bouguer gravity calculated from global free air with the 3D topographic correction [Fullea et al., submitted]; d) surface heat flow measurements from the global data set and from the Golmud-Ejn transect between Qaidam and Beishan basin. Grey line shows the position of the modelled profile. Note that d) is at a larger scale than the other figures.

Modelling results. The top three graphs (a-c) show data and results predicted from the model: a) Bouguer gravity anomaly, b) geoid and c) elevation. The data are represented by dots with bars, corresponding to the standard deviation within a range of 50 km to each side of the profile. Continuous blue lines are the modelled values. d) Lithospheric model with a vertical exaggeration of 2.4. The dots show earthquake hypocenters projected onto the model from a strip 150 km wide each side of the model..

Cartoon of the resulting profile, lithosphere scale. These results suggest that the coldest part of the mantle underlies southern Tibet and that the hottest underlies north central Tibet. The present average topography of the High Himalaya and Tibetan Plateau is supported by a non-uniform crustal and lithospheric structure. The High Himalaya and the southern Tibetan Plateau consist of thick crust and thick upper mantle whereas the northern Tibetan Plateau is underlain by a thinner crust and thin lithospheric mantle.

3. Physical modeling of the evolution of Tibet related to the present-day lithosphere structure

work in progress

We present two different numerical modeling techniques used to study the possible evolution of the plateau, the lithosphere-root removal and the delamination.
We study the delamination mechanism in a vertical cross section over the last 10 my. This work uses new algorithms of thermo-mechanical modelling, developed in MATLAB code, able to study the temporal evolution of the delamination. We model the evolution of a thickened orogenic lithosphere, which brings about a Rayleigh-Taylor gravitational instability. The motion equation and the coupled thermal equation are solved, under the extended Boussinesq Approximation, applying finite difference techniques. With this model we are able to reproduce the lateral propagation of continental lithosphere delamination. We discuss the effects of the initial geometry and the stratification of density and viscosity required to give the present Tibetan mantle structure.
We also investigate numerically the long-term (50 my) evolution of crustal and lithospheric thickness, thermal structure, topography, and strain-rate of the Tibetan plateau through time, using a planform viscous approach. This suggests that lithospheric mantle must have been removed from beneath Tibet, and that the crust must have been warmed and weakened by an increase of radiogenic heat production at depth due to crustal thickening.

Poster (pdf)