Spatial variability of stable isotopes in river water over the Tibetan Plateau

Abstract Estimating isotopic changes of river water on the Tibetan Plateau (TP) remains highly uncertain due to variable water sources related with seasonality of atmospheric circulations and local kinetic fractionation. The in‐depth understanding of such changes is crucial for accurate palaeoaltimetry reconstruction and for river budget evaluation which related with surviving of one‐fifth of the world's population under accelerated warming. Here, we present a comprehensive river water isotopic composition dataset (δ 18 O, δD and d‐excess, 1852 samples) over the TP to systematically describe spatial distribution and explore possible controlling factors of isotopic data in theses rivers. Results showed that δ 18 O in river water range from −20.8‰ to −5.2‰ (latitudinally, 0.92‰ per degree), enriching northwardly in the seven exorheic river basins. The d‐excess values range from −19.7‰ to 20.5‰ with the pronounced low values in endorheic rivers basins. We identified that moisture sources and atmospheric transports leave significant signals in isotopic compositions of river water in the Upper Brahmaputra (UB), Upper Mekong (UM) and Upper Yangtze (UYA) basins, and local processes modify the isotopic signals mainly in the Upper Salween (US), Upper Indus (UI), Upper Yellow (UYE) and Endorheic Rivers (ER) basins. The δ 18 O‐elevation relationships are only established in the UB, UI and UYE with lapse rates of −4.4‰/km, −3.4‰/km and −3.5‰/km, respectively. We suggest that as changes of moisture sources and lapse rates in different basins, it is cautious to apply δ 18 O‐elevation lapse rates for palaeoaltimetry reconstruction. Our dataset will benefit regional studies of hydrological, biogeochemistry, palaeoaltimetry reconstruction and palaeoclimate.