New constraints on the late Quaternary landscape evolution of the eastern Tibetan Plateau from 10Be and 26Al in-situ cosmogenic nuclides

Abstract

Based on Terrestrial Cosmogenic Nuclide (TCN) constraints from depth profiles of one granitic regolith from Wumingshan and five fluvial terraces from Xianshuihe and Zagunao rivers, we discuss the timing of the last deglaciation, the landscape-scale denudation and fluvial incision rates across the eastern Tibetan Plateau, in relation to previous work. We present a three-dimensional-graph visualization approach and corresponding constraints to better assess the feasibility and applicability of cosmogenic nuclides depth-profile dating. The exposure age (older than 19.4 ka) of the Wumingshan regolith corresponds to the retreat of the palaeo-Daocheng ice cap, which covered the Yidun terrane during the Last Glacial Maximum (LGM). Most basin-wide denudation rate data in the eastern Tibetan Plateau are lower than 130 mm/ka (47%, n = 90), which is consistent with the Wumingshan regolith denudation rate (lower than 52.8 mm/ka), and thus indicate that the landscape-scale denudation has been stabilized after the last deglaciation. Considering the reduction of integrated bulk density due to the accumulation of lower-density loess, we estimate mean exposure ages of Xianshuihe and Zagunao river terraces of 4.0 ± 0.7 ka, 5.9 ± 0.3 ka, 13.4 ± 2.0 ka, and 16.6 ± 1.4 ka. The observed increase in incision rate from 0.39 mm/yr over long timescale (∼600 ka) to 5.88 mm/yr over the last 15 ka at the Xianshuihe river site is probably due to the transition from glacial to interglacial climatic conditions. The fewer abandoned terraces along the Zagunao river after the Heinrich event 1 (H1) indicates that the climate change during the latest glacial-interglacial transition impacted less the landscape evolution in a relatively lower elevation area. Comparison with previously reported fluvial incision rates elsewhere across the eastern margin of the Tibetan Plateau indicates that incision rates are mainly influenced by abrupt climate change or intensified summer monsoon since the early Holocene, but probably controlled by regional tectonic uplift or fluvial headward retreat as the timescale increases. Overall, we propose a synthetic pattern of landscape evolution mainly dominated by long-term tectonic uplift together with fluvial headward erosion, episodically influenced by climatic change throughout the late Quaternary.