Spatial distribution and controlling factors of stable isotopes in meteoric waters on the Tibetan Plateau: Implications for paleoelevation reconstruction

Debates persist about the interpretations of stable isotope based proxies for the surface uplift of the central-northern Tibetan Plateau. These disputes arise from the uncertain relationship between elevation and the δ18O values of meteoric waters, based on modern patterns of isotopes in precipitation and surface waters. We present a large river water data set (1,340 samples) covering most parts of the Tibetan Plateau to characterize the spatial variability and controlling factors of their isotopic compositions. Compared with the amount-weighted mean annual oxygen isotopic values of precipitation, we conclude that river water is a good substitute for isotopic studies of precipitation in the high flat (e.g., elevation >3,300 m) interior of the Tibetan Plateau in the mean annual timescale. We construct, for the first time based on field data, contour maps of isotopic variations of meteoric waters (δ18O, δD and d-excess) on the Tibetan Plateau. In the marginal mountainous regions of the Plateau, especially the southern through eastern margins, the δ18O and δD values of river waters decrease with increasing mean catchment elevation, which can be modeled as a Rayleigh distillation process. However, in the interior of the Plateau, northward increasing trends in δ18O and δD values are pronounced and present robust linear relations; d-excess values are lower than the marginal regions and exhibit distinct contrasts between the eastern (8‰-12‰) and western (<8‰) Plateau. We suggest that these isotopic features of river waters in the interior of the Tibetan Plateau result from the combined effects of: 1) mixing of different moisture sources transported by the South Asian monsoon and Westerly winds; 2) contribution of moisture from recycled surface water; and 3) sub-cloud evaporation. We further provide a sub-cloud evaporation modified Rayleigh distillation and mixing model to simulate the isotopic variations in the western Plateau. Results of this work suggest that stable isotope-based paleoaltimetry studies are reliable in the southern through eastern Plateau margins; towards the central-northern Plateau, this method cannot be applied without additional constraints and/or large uncertainties.