开放水面蒸发同位素富集经典模型在太湖的直接检验与发展

Stable isotopes of water provide unique constraints for ecological, meteorological and hydrological research. The theory of isotopic fractionation in evaporation process is the fundamental principle underpinning these applications. The first mathematic equations calculating the isotopic evaporation of open water body were derived by Craig and Gordon (1965) and are widely used by ecologists, hydrologists and meteorologists. There are two kinds of uncertainties with the Craig-Gordon model: firstly, scientists usually assume that the water undergoing evaporation is well mixed, that is, no isotopic gradient exists between the water surface and the bulk water when applying the model; secondly, parameterization of the kinetic fractionation factor was established from laboratory experiments, which may not be applicable to natural conditions. So far, the model has never been evaluated based on measurements over open water bodies and under field conditions, due to the limitation of observation technology. We will conduct in-situ and continuous measurement with high temporal resolution (hourly) on the isotopic concentration of evaporative water vapor over Lake Taihu based on the Off-Axis Integrated Cavity Output Spectroscopy, and calculate isotopic evaporation on longer temporal resolution (monthly to yearly) based on the isotopic mass balance equations. We will evaluate and improve the Craig-Gordon model against the measurement and calculation results. Using the improved Craig-Gordon model, we will calculate the water budget of Lake Taihu and redo the partition of global terrestrial evapotranspiration. The project will improve the Craig-Gordon model, and provide data and theory support to the research based on the model.

稳定同位素方法能为生态学、气象学和水文学等领域的研究提供独特的信息，而蒸发过程中的同位素分馏理论是稳定同位素研究的基础。Craig和Gordon于1965年首次提出了开放水面蒸发同位素含量的计算公式，并得到了广泛应用。该模型理论上存在两点不确定性：即水体充分混合的假设和动力学分馏系数的参数化方案，这会造成相关研究的较大误差。但是，由于观测技术的限制，该模型一直没有得到自然条件下开放水体直接观测数据的检验。本项目拟采用离轴积分腔输出光谱技术在太湖上原位、连续和高频次地（小时尺度）观测湖泊蒸发的同位素含量，并采用同位素质量守恒方程计算月到年尺度上湖泊蒸发的同位素含量；全面检验该模型模拟各种时间尺度上自然开放水面蒸发同位素分馏的能力，并且提出改进方案。这将推进该模型的发展，并为相关研究提供数据支持和理论依据。

水的氢氧稳定同位素被广泛用于研究地球科学问题。而这些应用的前提是开放水面蒸发的同位素组分是可以采用环境要素数据进行计算的。在此，长期存在争论的问题是蒸发过程中动力学分馏（或扩散传输）究竟起多大作用。传统湖泊研究大多采取实验室条件下风洞水槽实验得到的经验值，即H218O和HDO的动力学分馏系数分别为14.2‰和12.5‰，但是全球气候模式中计算海洋蒸发的动力学分馏系数远远小于上述数值，分别约为6.2‰和5.8‰。本项目选取大型湖泊太湖（面积2400 km2）为研究对象，基于光谱技术对大气水汽的HDO和H218O组分开展原位观测，基于通量梯度法首次实现了开放水面蒸发的HDO和H218O组分的长期连续观测，然后，对Craig-Gordon模型中不确定性最大的动力学分馏系数进行检验。研究结果表明传统湖泊水文学研究中采用动力学分馏系数明显偏高，而自然条件下湖泊水面蒸发过程中的动力学分馏效应要弱很多，海洋研究采用的数值更为适用。在用稳定同位素质量守恒法计算湖泊蒸发时，如果采用海洋算法，则与涡度相关观测结果一致，如果采用湖泊算法，则湖泊蒸发被高估72‰。此外，本项目也进一步评估了较弱的动力学分馏效应对区域水汽循环、全球蒸散组分拆分、洋面过量氘-湿度关系和全球降水过量氘模拟结果的影响。研究结果表明采用较弱的动力学分馏系数，北美洲五大湖区水汽再循环的效果与气象模型研究结果更为接近。在全球蒸散组分拆分研究中，如果采用较弱的动力学分馏系数，则全球蒸腾占蒸散的比值从原有结果的84%±9%降低到76%±19%，与非同位素方法的更为符合。此外，采用较弱的动力学分馏系数，开放洋面过量氘与湿度之间的关系和全球气候模型中对降水过量氘的模拟更为合理。但是，本研究的结果是否适用于小型湖泊，这一点还不确定，有待于进一步的野外观测研究。