Weak middle crust beneath central Tibet: constraints from shoreline deformation around Siling Co

Whether the deep crust beneath the Tibetan Plateau is weak enough to flow on geologic timescales remains a central point of debate. Here we attempt to constrain the effective elastic thickness (Te) by exploiting the flexural deformation of highstand shorelines around the Siling Co, central Tibet, in response to climatically-induced lake recession. Extensive flights of well-preserved paleo-shorelines are distributed around the lake, and extend up to 60 m above present day lake level. In this study, we examined the highstand shoreline (~4594 m a.s.l.) in an effort to ascertain whether it is deflected. This highstand shoreline is characterized by obvious constructional features (beach ridges, benches, spits, bars and cuspates) that continuously connect to wave-cut scarps which define a clear geomorphic boundary between an older landscape characterized by dissected alluvial channels/gullies and a lower one characterized by younger, recessional shorelines. The age of relict shorelines was determined by OSL (optically stimulated luminescence) and of cosmogenic 36Cl depth profiles. The OSL ages for four samples from the highstand complex range from 9.3 ka to 4.3 ka, suggesting a relatively stable lake level during this time and the timescale of the lake unloading of the Siling Co, ~ 10 ka. Flights of shorelines below this highstand show ages younger than 3.3 ka, while degraded, relict shorelines highstand are significantly older; OSL ages range from ~ 18 - 43 ka and two 36Cl depth profile yield ages of 113 ka and 178 ka. Comparing observed shoreline deflections with models of elastic deformation in response to removal of a 3D load, we find that the effective elastic thickness of central Tibetan crust is relatively thin (Te = 10 - 15 km), suggesting that most of the mechanical strength of the crust resides in the upper crust. The timescale of lake recession (~ 10 ka) implies strain rates on the order of ~ 1e-16 1/s, which, when combined with existing constraints on the thermal and seismic velocity structure of the crust in central Tibet, allow us to place bounds on the range of probable viscosity beneath central Tibet. Assuming a simple, two-layer model with a viscous layer beneath the elastic upper crust implies viscosity on the order of ~ 1e19 Pa s. A more comprehensive consideration of possible strength profiles consistent with available seismic and thermal data is consistent with a significant reduction in effective viscosity below depths of ~ 20 - 40 km. In this treatment, viscosities above this level range from 1e20 - 1e22 Pa s, while viscosities at depth range from 1e18 - 1e21 Pa s. Collectively, our findings of thin elastic upper crust and a reduction in viscosity at depth suggests that the middle and lower crust beneath central Tibet is likely relatively weak at timescales of 10 - 100 ka. However, a more definitive determination awaits a refined understanding of lake loading history.