新疆天山降水稳定同位素的时空特征与影响因素

SPATIO-TEMPORAL CHARACTERISTICS AND INFLUENCING FACTORS OF STABLE ISOTOPES IN PRECIPITATION ACROSS THE CHINESE TIANSHAN MOUNTAINS

降水是水文循环的重要环节，研究降水中氢氧稳定同位素有助于深入理解水循环过程与陆一气相互作用。天山是亚洲中部干旱区典型的湿岛，也是新疆乃至亚洲中部水体同位素研究的关键区域。为了明确新疆天山降水稳定同位素的时空特征与影响因素，本文根据近年来建立的降水同位素监测网络，介绍了该区域降水同位素的基本情况以及水汽再循环、云下蒸发、水汽路径和天气系统等因素对降水同位素的影响。总体而言，降水中的6D和δ18O呈南高北低和夏高冬低的趋势，但是过量氘（d）却没有如此明显的时空特点，区域大气水线的斜率和截距均低于全球大气水线；各采样点降水同位素比率与气温存在显著的正相关关系，但各地降雪（或降雨）样品与所有样品之间的同位素温度梯度相对大小并没有一致的规律；利用降水同位素温度效应以及气温和降水量格网化数据，再结合实测资料订正即可得到降水同位素景观图。同位素三元混合模型可用于评估绿洲区水汽再循环，绿洲区植物蒸腾水汽对降水的贡献率普遍大于地表蒸发水汽的贡献率；基于改进的Stewart模型，云下二次蒸发对南坡降水同位素的影响较北坡大，对夏半年的影响较冬半年大，云下蒸发效应订正后降水同位素的区域一致性更加明显；与其他天气系统相比，中亚低涡控制下的降水中重同位素更为贫化，结合大气比湿订正的拉格朗日模型发现，天山南坡和北坡降水同位素差异亦与不同的水汽传输路径有关。

The stable hydrogen and oxygen isotopes in precipitation are useful to understand the hydrological processes and land-atmosphere interaction. In arid Central Asia with scarce precipitation and high evaporation capacity, the Tianshan Mountains is usually considered as a typical wet island. The total length of the east-west mountain ranges is approximately 2500km, and the eastern portion （ca.1700km）is located in Northwest China＇s Xinjiang. Although the precipitation isotopes were discontinuously measured at some sites in the Tianshan Mountains, a systemic measurement of stable isotopes in precipitation was absent for the entire mountains in past decades. To investigate the spatio-temporal pattern and regime of isotopes in precipitation in this region, an intensive and integrated network was established across the Chinese Tianshan Mountains. More than 20 stations are selected to collect event-based precipitation samples, covering typical landscapes including deserts, oases and mountains in arid Central Asia. Generally, the values of δD and δ18O in precipitation are higher in south and lower in north, and the heavy isotopes are enriched in summer months and depleted in winter months. However, the spatial pattern and seasonal variation for deuterium excess （d） are not coherent as those for hydrogen and oxygen isotopes. The local meteoric water lines （LMWL） can be derived from event-based and amount-weighted monthly isotopic data, respectively, and the slope and intercept in both lines are lower than those in global meteoric water line （GMWL）. The isotopic compositions in precipitation significantly correlate with air temperature for each sampling site. The linear gradients between isotopic composition and air temperature for snow （or rain） samples are not always larger or less than those for all precipitation samples at different sites. Based on the temperature effect and fine meteorological grid products （daily mean air temperature and precipitation amount） , the isoscape can be derived using the in-situ measurements. In a three-component isotopic mixing model, the precipitating moisture is considered as a mixture of advection, transpiration and surface evaporation moistures. The mixing model can be applied to assess the moisture recycling in the typical oases, and the proportional contribution of transpiration flux to local precipitation is always larger than that of surface evaporation flux for each oasis. Based on a modified Stewart model, the isotopic variation in raindrops from cloud base to ground can be calculated for each event. The influence of below-cloud evaporation on precipitation isotopes on the southern slope is larger than that on the northern slope, and the impact in summer months is more significant than that in winter months. The spatial coherence of precipitation isotopes is improved if the isotopic compositions are corrected based on below-cloud evaporation. Compared with other synoptic systems in the study region, Central Asian vortex usually corresponds to depleted heavy isotopes in precipitation. Using a Lagrangian model adjusted with specific humidity along the backward trajectory, the difference in stable isotopes on the northern and southern slopes may be also related to the various moisture paths.