Research Article

Revised paleoaltimetry data show low Tibetan Plateau elevation during the Eocene

See allHide authors and affiliations

Science  01 Mar 2019:
Vol. 363, Issue 6430, eaaq1436
DOI: 10.1126/science.aaq1436

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

Ancient height of the Tibetan Plateau

The elevation of the Tibetan Plateau has a major impact on climate, affecting the monsoons and regional weather patterns. Although some isotope proxies have suggested a roughly equivalent height for the plateau as far back as the Eocene (∼40 million years ago), other lines of evidence suggest a lower elevation in the distant past. Botsyun et al. used a model to show that several previously overlooked factors contribute to the isotopic record from the Eocene (see the Perspective by van Hinsbergen and Boschman). The results harmonize the isotopic record with other proxies and argue for a Tibetan Plateau that was about 1000 meters lower than it is today.

Science, this issue p. eaaq1436; see also p. 928

Structured Abstract


The uplift history of the Tibetan Plateau (TP) is critical for understanding the evolution of the Asian monsoons and the geodynamic forces involved in collisional orogens. The early topographic history of the TP is uncertain, and the timing of the initiation of uplift remains controversial. The majority of studies that find evidence for an old plateau (as early as the Eocene, ~40 million years ago) rely on stable isotope paleoaltimetry. This method is based on observations and models that show depletion in heavy oxygen isotopes in rainfall during orographic ascent. Ancient carbonates of the TP record past rainfall isotopes; when available, their isotopic signature can be compared with isotopic lapse rates in order to estimate at what elevation they grew in the past, but the use of this method in deep time has many uncertainties.


Applying stable-isotope paleoaltimetry in greenhouse climates makes the implicit assumption that the factors that control atmospheric distillation and rainfall oxygen isotopic composition (δ18Ow) have remained constant over millions of years. However, the impact of past climate change on δ18Ow values is unclear. In particular, for the Eocene, higher atmospheric CO2 concentration (Pco2) and markedly different Asian paleogeography, including a wide and shallow Paratethys Sea in central China and a latitudinal shift of the southern Tibet margin ~10° to the south, have been hypothesized to modify the Asian climate and regional δ18Ow values. In addition, the carbonate formation temperature is often unknown, increasing the uncertainty in the reconstructed δ18Ow.


We ran climate simulations with Eocene boundary conditions and varying TP elevation. We accounted for changing Pco2, land surface albedo, orbital variation, and sea-surface temperatures, which potentially cause shifts in δ18Ow, by changing the relative contribution of different air masses and the local hydrological cycle—the evaporation-to-precipitation ratio and the fractioning between convective and large-scale rainfalls. In our simulations, the south-shifted location of the entire Indian foreland induces strong convection over the southern flank of the TP and a radically different pattern of water recycling compared with that of present day. Our simulations reproduce monsoonlike seasonal precipitation over the Indian foreland, with summer convective rainfall reaching the TP. An intense anticyclonic circulation during summer months induces widespread aridity on the northern part of the Plateau. This peculiar atmospheric circulation, together with intensified water recycling and multiple moisture sources, results in a reversed isotopic lapse rate across the southern flank of the TP, with the most negative δ18Ow over northern India and increased δ18Ow northward. On the basis of Eocene experiments with varied boundary conditions, we argue that this is a robust feature of Eocene climate over the region. Furthermore, this pattern is the opposite of the present-day δ18Ow over the Himalayas, which decreases with elevation, driven by orographic rainout following a Rayleigh distillation process. Last, using our simulated temperatures and δ18Ow, we derived virtual carbonates δ18O for different elevation scenarios and compared them with the geological record. Statistical analysis shows that a low TP topography during the Eocene is the scenario that provides the best match between model and data.


Our simulations indicate that standard stable isotope paleoaltimetry methods are not applicable in Eocene Asia because of a combination of increased convective precipitation, mixture of air masses of different origin, and widespread aridity. In the Eocene, the presence of a reversed isotopic lapse rate precludes use of any of the previously developed δ18Ow-elevation relationships for estimating the Eocene elevation of the TP. Rather, a model-data comparison on the carbonates δ18O suggests that the TP reached only low to moderate (<3000 m) elevations during the Eocene, reconciling oxygen isotope data with other proxies of elevation and with geodynamic models that propose a recent (Neogene) uplift. More generally, we suggest that using climate models in conjunction with stable isotope data from the geological archives provides a powerful tool to incorporate climatic changes into the analysis of paleoelevations.

Simulated stable oxygen isotope composition of summer precipitation for the Eocene (42 million years ago) Asia.

The combination of changes in water recycling and the source of air masses induces the reversal of isotopic lapse rate. Summer winds trajectories highlight intense westerlies and easterlies at each border of the TP. Circles indicate localities of paleoaltimetry sites used in this study.


Paleotopographic reconstructions of the Tibetan Plateau based on stable isotope paleoaltimetry methods conclude that most of the Plateau’s current elevation was already reached by the Eocene, ~40 million years ago. However, changes in atmospheric and hydrological dynamics affect oxygen stable isotopes in precipitation and may thus bias such reconstructions. We used an isotope-equipped general circulation model to assess the influence of changing Eocene paleogeography and climate on paleoelevation estimates. Our simulations indicate that stable isotope paleoaltimetry methods are not applicable in Eocene Asia because of a combination of increased convective precipitation, mixture of air masses, and widespread aridity. Rather, a model-data comparison suggests that the Tibetan Plateau only reached low to moderate (less than 3000 meters) elevations during the Eocene, reconciling oxygen isotope data with other proxies.

View Full Text