Evaporative enrichment of oxygen-18 and deuterium in lake waters on the Tibetan Plateau

Abstract

Stable isotopes (δ18O and δD) are useful tracers for investigating hydrologic and climatic variability on a variety of temporal and spatial scales. Since the early isotopic studies on mountainous glaciers in the late 1960s, a great deal of information has been generated on the isotopic composition of rainfall, snow, ice, surface waters, and lake carbonate sediments across the Tibetan Plateau. However, measurements of δ18O and δD values of lake water are scarce. Here we present a new dataset of δ18O and δD values of lake waters collected from 27 lakes across the plateau during a reconnaissance survey in summer 2009. δ18O and δD values of lake water range from −19.9 to 6.6‰ and from −153 to −16‰, respectively. The average values of δ18O and δD are −6.4 and −72‰, considerably greater than those of precipitation observed in this region. The derived Tibetan lake water line, δD = 5.2δ18O − 38.9, is significantly different from the global meteoric water line. Most of the lakes, including some freshwater lakes, contain water with negative values of d-excess (d). There is a negative correlation between d and total dissolved solids (TDS). Each of these findings indicates that evaporation-induced isotopic enrichment prevails in Tibetan lakes. Moreover, we develop an isotope modeling scheme to calculate E/P ratios for Tibetan lakes, using a combination of existing isotopic fractionation equations and the Rayleigh distillation model. We use the intersection of the local evaporation line and GMWL as a first approximation of δ18O and δD values of lake water inputs to infer an E/P ratio for each lake. Our modeling calculations reveal that although variable from lake to lake, the water budget across the plateau is positive, with an average E/P of 0.52. This is in good agreement with other observational and model data that show varying degrees of increases in lake size from satellite imagery and significant decreases in lake salinity in many lakes on the plateau over the last several decades. Together with the new isotopic dataset, the proposed modeling framework can be used to examine and quantify past changes in a lake’s hydrologic balance from the isotopic record of downcore carbonate sediments in the region.