气候变化与人类活动已经实质影响到全球各大洲大洋的自然系统，也给诸多国家带来了能源、资源和生态问题。气候变化对于水资源时空分布严重不均、农业用水需求极大、地区生态问题突出的中国而言，将会加剧水资源短缺等问题，亟需开展生态水文综合研究。流域生态水文模型作为研究生态水文的重要工具之一，能够对当地生态水文过程机理进行适当描述，对大量生态水文变量进行模拟，从整体和细节上刻画流域生态水文情势。黄土高原作为黄河流域流经的重要区域，气候变化明显、人类活动频繁、下垫面改造剧烈，适合作为典型流域进行流域生态水文响应研究。本文选取黄河中游、黄土高原上的无定河流域作为典型流域，于2011年6月在流域内部建立了生态水文观测站点，进行了气象、涡度相关、土壤剖面等一系列生态水文变量的观测。在此基础上，结合气象水文站点及遥感影像资料，本文对榆林站、无定河流域以及各个子流域的历史气候因子、叶面积指数、河道径流之变化进行了特性和趋势分析。本文结合国际上广泛应用的陆面过程模型CLM与分布式水文模型GBHM分别在垂直方向生态水文过程刻画与水平方向拓扑关系描述的优势，通过代码级别的紧密耦合方案，开发出了新的流域生态水文耦合模型CLM-GBHM。借助并行计算平台，CLM-GBHM模型在多种情景下均能够成功运行。本文在典型流域站点尺度、流域尺度的驱动与下垫面资料基础上，利用经过率定的CLM-GBHM分别构建了站点尺度模型与流域尺度模型。通过模型长期模拟结果与实测数据的对比，以及与相同尺度下陆面过程模型CLM的结果比较，本文基于多尺度多过程对模型进行了综合评价。基于无定河流域下垫面剧烈变化的现状，本文设置了三个情景，利用CLM-GBHM分别模拟了纯气候变化、考虑动态植被影响以及实际植被输入情形下流域的径流变化，进而分析了植被动态对流域土壤水、蒸散发、径流等水文过程的影响。此外，使用归因分析方法，本文计算了气候变化、动态植被以及人类活动等其他因素对径流变化的贡献，并与基于水热耦合平衡理论的方法进行了对比。
Climate change and human activities have affected both the natural and the artificial systems across all continents and oceans for many centuries, bringing a lot of energy crisis and eco-hydrological problems to many countries all over the world. With regard to China, a country suffers from excessively high proportion of agricultural water demand and severe uneveness of water resource distribution both spatially and temporally, these substantial influences exacerbate the water shortage, leading to a growing demand of comprehensive eco-hydrological researches. As an important and effective research technique, the eco-hydrological model is able to consider local eco-hydrological processes in detail, simulate a large number of eco-hydrological variables, and describe the eco-hydrological regime in certain basin.Recently, studies have shown that a significant decrease occurred in river discharge in the middle reach of the Yellow River Basin in the past five decades, which leads to an intensified water resources crisis in this region. Reasons to this phenomenon is generally considered as the interactions between climate forcing, regional land use/cover change and vegetation growth, which eventually determined by climate variability and local human activities. In this paper, a semi-arid basin called Wudinghe River Basin (WRB), which located in the middle reach of the Yellow River and north of the Loess Plateau with a drainage area of 28,706 km2, was selected as a typical watershed for our research. An eco-hydrological observation site was built in June, 2011 and lots of experiments and observations were conducted.Lots of land surface models have been developed and widely used for climate change and hydrology studies. These models usually provide comprehensive information for understanding the vertical eco-hydrological processes within a grid. However, the simplification (or neglect) of river network and river routine in the models may lead to large uncertainty in studying the river discharge-vegetation interactions. On the contrary, distributed hydrological models mainly specialize the description of watershed and the parameterization of river routine, while usually simplify the vertical eco-hydrological processes. In this study, we aim at developing a new eco-hydrological model which combines the advantages of both types of model, in order to explore the interactions among river discharge, vegetation, and climate in water-limited watersheds. To fill these gaps, this study chose the Community Land Model version 4 (CLM4) and the Geomorphology Based Hydrological Model version 2 (GBHM2), and replaced the runoff generation and flow routing schemes of CLM4 by the schemes used in GBHM2. After a comprehensive calibration and validation, the model was applied for simulation of the eco-hydrological changes in the past five decades.Based on the fact that the underlayer of WRB has changed dramatically, three simulation cases were designed, reflecting different scenarios: only climate change was taken account, climate change and vegetation dynamics were both considered, and all factors impacting underlayer were taken account by using accutual vegetation condition. Comparison of results from three cases was made to illustrate the influence of vegetation dynamics on surface soil moisture, evapotranspiration and river discharge, and attribution analysis was also used to show the contribution of climate change, vegetation dynamics and human activities to river discharge change.
