Fluid dynamics across the interface in gravel-bed rivers; quantification and numerical modelling of flow in the hyporheic zone

水在天然河流中的流动方式是环境中最复杂的建模和预测现象之一。对于河床表面之下的水流(在一个被称为“高位带”的区域)，在鹅卵石和构成河底的石头之间的空间之间，情况更是如此。从鹅卵石之间的微小空间获得水流速度的测量迄今被证明是一个无法解决的问题，这一事实阻碍了对这些流动进行准确建模的努力。但这为什么要让科学家们感到担忧呢？首先，河流生态学家现在认识到，浅水区是各种物种的重要栖息地。从海床上方流入地下的水流方式在很大程度上决定了向这个栖息地提供多少氧气和营养物质。其次，渔业管理者早就明白，鲑鱼卵在河床上孵化的可能性将取决于向产卵的粘稠沉积物持续供应含氧水。第三，河流系统中的污染物(如重金属)通常会附着在被称为胶体的微观颗粒上，这种颗粒往往沿着水流的路径流动。因此，理解水流在河床中的运动方式将对建立污染物行为大有裨益。因此，有一系列非常重要的环境问题需要对河床中的水如何运动进行详细的预测，但没有办法准确地测量或模拟这一点。这项提案采用开创性的新方法，力求迎接这一挑战。第一个任务是精确测量床层内的流动，这一重大问题将使用一种新的微型PIV(粒子成像测速)技术来克服。该系统借鉴了为医疗应用开发的技术，使用了可以放置在实验河床中的小型内窥镜数码相机。通过在水流中播撒微小的反射颗粒，并提供激光的高强度照明，内窥镜相机可以记录它们如何在河床卵石之间的小缝隙中移动。使用一种特殊的处理器，这些数字图像可以转换成数字数据，准确地记录水流如何穿过河床，然后进入河床。这样的测量以前从来都不可能。该项目的第二阶段是利用这一独特的数据集带来的新认识，开发和测试一个3D数值模型，该模型可以准确地预测河床表面上下的水流情况。这将使用一个特别修改的计算流体动力学(CFD)模型来实现。这样的模型代表了最先进的技术，然而地下水流的问题被证明太有问题，以至于它们不能在这样的环境中应用。然而，我们的团队已经设计出一种方法，通过这种方法，鹅卵石可以被“抹掉”，并预测它们周围和鹅卵石之间相邻缝隙中的水流。该项目中将使用的测量和建模方法的进步代表着真正的突破，将揭开从最具挑战性的自然环境之一获得有用数据的内在问题。同时，开发一种可广泛应用的数值模式将确保这种新的理解能够被应用和调整，以应对各种现实世界的环境挑战。

The way in which water flows within a natural river is one of the most complex phenomenon to model and predict accurately in the environment. This is even more so for the flow that occur just beneath the surface of the river bed (in a region termed the 'hyporheic' zone), between the spaces of pebbles and stones that make up the bottom of a river. Efforts to accurately model these flows have been hampered by the fact that obtaining measurements of water velocity from the tiny spaces between pebbles has so far proved an irresolvable problem. But why should this worry scientists? Firstly, stream ecologists now recognise that the hyporheic zone is an important habitat for a diverse range of species. The way flow from above the bed makes its way into the subsurface largely dictates how much oxygen and nutrients are supplied to this habitat. Secondly, fisheries managers have long understood that the probability of salmon eggs laid in river beds hatching will be dependent on a continuous supply of oxygenated water to the grevelly sediments in which they are laid. Thirdly, pollutants in river systems (such as heavy metals) often become attached to microscopic particles called colloids, which tend to follow flow pathways. An understanding of how flow moves within a river bed will thus go a long way to establishing pollutant behaviour. There are thus a broad range of highly important environmental issues that require detailed predictions of how water moves within a river bed, yet there is no way of measuring or modelling this accurately. Using pioneering new approaches this proposal seeks to meet this challenge. The first task is to accurately measure flow within the bed, this significant problem will be overcome using a new micro-PIV (particle imaging velocimetry) technique. This system borrows technology developed for medical applications by employing a small endoscopic digital camera which can be placed within an experimental river bed. By seeding the flow with tiny reflective particles, and providing high intensity illumination from a laser, the endoscopic camera can record how they move within the small gaps found between pebbles in the river bed. Using a special processor, these digital images can be turned into numerical data that accurately records how flow moves across and then into the river bed. Such measurements have never been possible before. The second phase of the project is to use the new understanding made possible by this unique dataset to develop and test a 3-D numerical model that can precisely predict how water will flow above and below the surface of a river bed. This will be achieved using a specially modified computational fluid dynamics (CFD) model. Such models represent the state-of-the-art, yet the issue of subsurface flow has proved too problematic for them to be applied in such environments. However, our team has devised a method whereby the pebbles can be 'blanked out' and the flow predicted around them and into adjacent gaps between pebbles. The advances in measurement and modelling approach that will be used in this project represent real breakthroughs that will unlock the inherent problem of gaining useful data from one of the most challenging of natural environments. Meanwhile, the development of a numerical model that can be widely applied will ensure that this new understanding can be applied and adapted to meet a variety of real world environmental challenges.