Evaporation dominates the loss of plateau lake in Southwest China using water isotope balance assessment

The water balance budget in remote plateau lakes provides a fundamental information on the local climate-hydrological pattern. However, integrated investigation on the waters entering the lake, especially groundwater, was limited. To assess the current climate stress on Yunnan-Guizhou Plateau lakes, we collected rivers, groundwater, lake, and precipitation with varying isotopic compositions in the Chenghai Lake basin over four separate campaigns during a hydrological year. The wide and enriched variation of isotope composition in rivers, groundwater, and lake indicate that they have undergone distinct evaporations, which further reveal the recharging and mixing processes. Based on the similar isotopic signals between rivers and precipitation, rivers can serve as proxies for precipitation. Groundwater was primarily replenished by high mountain precipitation duo to the stable isotopic values in aquifers. Even through mass water in lake was able to smooth out some variability, the considerable isotopic variation of lake during the four collections suggested the influence of meteorological conditions. According to the assessment of isotope balance model, lake evaporation accounts for almost 65 % of the total inflow for one year, which partially explains the climate stress on the lake level. As the most sensitive variables, changes in relative humidity (h) and isotope composition of atmospheric moisture (δA) resulted in remarkable variations in E/I ratios and the constructed water isotope framework. These results shed light on the capacity of evaporation relative to lake input and provide interpretations on local paleoclimate and predicted-climate construction. [Display omitted] •Groundwater was primarily replenished by high mountain precipitation because of the stable isotope composition.•More than two-thirds of total incoming water in Chenghai Lake is consumed by evaporation.•Relative humidity and isotope composition of atmospheric moisture are the most sensitive variables to isotope model.