Research papersControls on the stable isotopes in precipitation and surface waters across the southeastern Tibetan Plateau

•A large amount of precipitation occurs during the spring period.•The moisture source in the spring period was investigated.•Characteristic isotope-elevation relationships for surface waters were documented.•Isotopic signature of water reflects both topography and atmospheric circulation pattern.

Constraining temporal and spatial variability in water stable isotopes (δ18O and δD) is requested for interpreting proxy records of paleoclimate/paleoaltimetry. The southeastern Tibetan Plateau (TP) receives large amounts of precipitation in both summer (JJAS) and spring (MAM) and this makes it different from most other parts of the TP where annual precipitation concentrates only in summer. However, our knowledge of controls on precipitation and surface runoff generation in this region is still far from sufficient. In this study, the δ18O and δD of precipitation and stream waters across the southeastern TP were analyzed to investigate moisture sources and empirical isotope-elevation relationships. Herein, seasonal precipitation patterns, moisture trajectories and precipitation isotopes suggest this region is seasonally dominated by the monsoon in summer and the southerlies (from the Bay of Bengal) or a mix of southerlies and westerlies in spring. Spatially, vertical variations in precipitation seasonality exert profound influences on isotopic variability for stream waters. Larger contributions of spring precipitation (with higher δ18O and d-excess (d-excess = δD-8δ18O) compared to summer precipitation) vs. summer precipitation in the surface runoff generation at lower elevations account for the uncommon altitudinal decrease in streamwater d-excess. Such a cause also contributes to the slightly greater vertical lapse rates of streamwater δ18O (−0.28 to −0.48‰/100 m) relative to the Himalayan front. In addition, although a robust δ18O-elevation relationship is demonstrated based upon our measured and other published data on a broad spatial scale (over a 5200 m elevation range), this relationship is found to deviate from the empirical/theoretical pattern in the Himalayan front, which is also caused by the substantial spring precipitation in the southeastern TP. It is suggested that long-term changes in δ18O or δD of paleowater in this region actually represent both changes in past regional elevations and contributions of summer vs. spring atmospheric circulations.