基岩河道河流水力侵蚀模型及其应用：兼论青藏高原基岩河道研究的迫切性

STREAM-POWER INCISION MODEL AND ITS IMPLICATIONS： DISCUSSION ON THE URGENCY OF STUDYING BEDROCK CHANNEL ACROSS THE TIBETAN PLATEAU

河流作为对构造-气候相互作用最为敏感的地貌单元，记录了丰富的水系演化、构造变形以及气候变化信息。目前基于数学推导与物理模型相结合的水力侵蚀方程，将构造抬升与基岩河道水力侵蚀作用相结合，使得通过河流纵剖面形态提取基岩隆升速率的时空分布特征成为可能。现有研究与应用主要包括：1）利用方程的稳态形式获取水系陡峭系数（坡度-面积对数分布图，Chi-plot），并以此探讨构造隆升速率高低及空间分布特征；2）构建河流裂点的溯源迁移模型，定量获取裂点迁移速率；3）利用方程的线性非稳态形式，获取区域基岩隆升历史；4）获取不同流域水系河段x值，判别分水岭的迁移方向。对于方程的非线性非稳态形式，在求解及应用等方面都存在一定的不足，这也将是水力侵蚀模型今后亟需解决的关键问题。文章在简单回顾现有青藏高原基岩河道研究成果的基础上，提出开展不同时间、不同空间尺度青藏高原基岩河道相关研究的迫切性。

Owing to its sensitiveness to tectonics and climate, fluvial channel records the relevant signals. How to retrieve the related information from river profiles has long been a hot issue in the earth science. Fortunately, the stream-power river incision model, based on physical experiments and rigid mathematic theory, combines bedrock uplift and river incision inherently, making it possible to extract rock uplift history from river profiles. Under steady assumption, in which the rates of rock uplift are balanced by that of river incision, the derived slope-area analysis yields a steepness index to quantify the spatial pattern of rock uplift rates. For its heavy reality on binning and smoothing of elevation data, this method may lack the temporal and spatial resolution of tectonic forcing. Chi￣plot, an integral approach to the steady equation, makes up for such drawbacks. Integrating the contributing drainage area with upstream distance as a  value, the river profile can be transformed into a straight line. Besides, we can also recognize knickpoints and transient signals along the streams more easily. As the numerical and analytical solutions to the transient equation were put forward, we can extract the temporary and spatial information of rock uplift rates. Based on a linear assumption between local channel gradient and river incision rates, the vertical velocities of knickpoints migrating upstream were demonstrated to be controlled only by tectonic uplift rates. Then all the knickpoints can be preserved in the river profile, which resulting a full rock uplift history without any miss. However, under non-linear condition, velocities of knickpoints migration vary a lot, which may result consuming knickpoints and an incomplete uplift history. When a knickpoint passed the whole river profile, the previous tectonic information might be erased. In other words, whether we can get a full tectonic uplift history also depends on the scale of fluvial system. Fluvial system evolution not only causes vertical change in stream profiles but is accompanied by water divides migration and fluvial network reorganization. Knowing the migrating orientation of water divides is critical to understand the geological and earth surface process. For catchments with the similar elevation difference, lower  value means larger steepness index, or higher erosion rates. Field work and numerical model both have showed that water divides migrate towards areas with lower erosion rates. Hence, this feature can be a good indicator to network reorganization. Although the research and application of stream-power river incision model has got a great progress, there are also some problems. We have got the temporary pattern of rock uplift rates under linear assumption. However, spatial information from the non-linear transient equation is still confusing. Besides, we treat these problems based on an assumption of uniform lithology and climate. How to take these parameters into consideration is difficult. Throughout the paper, we give a brief introduction to the so-called stream-power river incision model, including the model theory, solutions of both steady and linear transient equation, application, and problems unsolved. We think that many of the problems will be solved in the near future. By presenting the available studies related to the Tibetan Plateau, we also highlight that systematic analysis of bedrock channels across the Tibetan Plateau is needed to understand the deformation history and landscape evolution.