西北地区大气降水δ^18O的特征及水汽来源

Characteristics of δ^(18)O in Precipitation over Northwest China and Its Water Vapor Sources

根据2005年各月在中国大气降水同位素观测网(CHNIP)位于西北地区的阜康、策勒、临泽、海北、沙坡头、长武和安塞观测站点收集的降水样品,对其中的同位素的组分进行测定,分析了西北地区大气降水中!18O的时空分布特征。所建立的局地大气降水线方程"D=7.05#18O-2.17,反应了西北地区独特的局地气候特点。降水$18O的温度效应显著,而降水量效应只在夏季(6-8月)间存在。%18O的空间分布特征可以很好地反映西北地区的大气环流背景。应用瑞利分馏模型及动力分馏模型对阜康-安塞沿线降水&18O的定量模拟结果,揭示了西北地区降水水汽的分馏主要以动力分馏为主,雨滴在降落过程中历经了一定的二次蒸发过程,其降水水汽中也混入一定量的由局地再蒸发的水汽。此外,利用西北地区在全球大气降水同位素观测网络(GNIP)中的乌鲁木齐、和田、张掖、兰州、银川和西安6个站点的长时间序列的’18O与降水量、温度等气候因子建立的多元线性回归关系可以对降水(18O进行定量估算;基于乌鲁木齐站点12年的)18O资料对该地区的温度拟合,为历史气候的定量恢复提供了依据。

In order to reveal the characteristics and climatic controls on the stable isotope composition in precipitation over Northwest China, seven stations have been selected from the Chinese Network of Isotopes in Precipitation （CHNIP）. During the year 2005, monthly precipitation samples have been collected and analyzed for the composition of δ^18O. The established local meteoric water line δD = 7.05δ^18O - 2.17, based on the 50 monthly samples obtained, could be treated as isotope input functions across the region. The deviations of the slope and the intercept from the Global Meteoric Water Line indicated the specific regional meteorological conditions over Northwest China. The monthly δ^18O values were characterized by a strong correlation with the surface air temperature and a weak correlation with precipitation amount. However, if only the δ values during summer period have been taken into account, the amount effect visualized. Spatial distributions of δ^18O have properly portrayed the atmospheric circulation background in each month over Northwest China. The quantitative simulation of δ^18O, which involved a Rayleigh fractionation model and a kinetic fractional model, demonstrated that the kinetic fractionation was the dominating function of condensation of raindrops. Furthermore, the raindrops suffered re-evaporation during the falling process, and the precipitation vapor has been mixed with a quantity of local recycled water vapor. A multiple linear regression equation and a δ^18O-T relation were gained by using meteorological parameters to evaluate physical controls on the long-term values. The established δ^18O-T relation, which has been based on the present-day precipitation, was the first step of quantitatively reconstructing the paleo-climate.