Restoring Streams and River Through Cutting-edge Simulation-based Science: The Virtual StreamLab
Seokkoo Kang and Fotis Sotiropoulos
St. Anthony Falls Laboratory
Department of Civil Engineering - University of Minnesota
Lay-language version of "Coherent-Structure Resolving Simulations of Turbulence in Natural Streams with the Curvilinear Immersed-Boundary Method"
A staggering 44 percent of the 3.5 million miles of our rivers and streams is degraded due to sedimentation and excess nutrients resulting from unsustainable urban development and land use practices. The degradation of the Nation’s waterways has led to the decline of water quality over entire watersheds, rendered streams unhealthy for recreation and public contact, and had serious negative impact on aquatic life.
Figure 1: The St. Anthony Falls Laboratory Outdoor StreamLab.
To remedy such potentially disastrous consequences for the environment, the number of stream and river restoration projects around the nation has risen dramatically with an estimated amount of over $1 billion being spent in the US on such projects every year since 1990. To restore a stream, engineers strategically install man-made structures (piles of large rocks, concrete blocks, tree trunks, constructed riffles, etc.) to change the manner in which water flows and thus to alter its interaction with the stream banks, the sediments, and the biota in the stream. If installed properly such structures could enhance water quality, stabilize the stream banks, control flooding, and improve fish habitat.
The development of effective and reliable stream restoration strategies, however, is far from trivial. The underlying physical processes governing the interactions of turbulence in the water column with the sediments and the fauna and flora in the stream are very complex and not fully understood. As a result, the current practice of stream-restoration is more of an art than a science and most restoration projects either fail after a short period of time or do not perform as expected.
Figure 2: The measured natural bathymetry of the Outdoor StreamLab used to construct a virtual model of the stream that resolves directly the details of small rocks and other stream-bed irregularities.
Researchers at the St. Anthony Falls Laboratory (SAFL) and the National Center for Earth-Surface Dynamics (NCED), an NSF science and technology center headquartered at SAFL, are developing a science-based approach to stream-restoration. Central to this inherently interdisciplinary approach is a set of powerful experimental and computational tools that can be used to understand how water flows and interacts with sediments and biota in natural streams. The SAFL Outdoor StreamLab (OSL) is a unique field-scale natural stream within which physical experiments can be carried out in a controlled laboratory environment and at real-life conditions (Figure 1). The OSL is supplemented by its virtual counter-part, the Virtual StreamLab (VSL), which is a high-resolution computational fluid dynamics (CFD) computer code for simulating turbulent flow in natural streams, including stream-bed erosion processes and the interaction of the flow with fish biota (Figures 2-4).
Figure 3: Virtual StreamLab. Regions of instantaneous high velocity are visualized at the water surface by plotting contours of velocity magnitude (see also Video 1).
Existing computer models of river flows oversimplify reality. They cannot accurately account for much of the geometric complexity of real-life waterways (complex bank shapes, nature and man-made structures, boulders, rocks, gravel, and other small-scale bed features) at the resolution needed to provide a realistic model of natural turbulent flows. The Virtual StreamLab is the first computer model of its kind that can simulate turbulence in natural streams at scales that are necessary to understand how water flow impacts the stability of the stream banks and affects the biota within the stream. The computer model employs sophisticated numerical algorithms that can handle the arbitrarily complex geometry of natural waterways, features advanced turbulence models, and is designed to take advantage of the latest advances in massively parallel super-computers.
In the APS DFD 2009 meeting, our group will report the first ever simulations of turbulent flow in a real-life stream under conditions that are representative of what actually occurs in nature. The simulated virtual stream is a realistic model of the real-life meandering stream currently installed in the SAFL Outdoor StreamLab. The geometry of the OSL streambed, consisting of fine sand, rocks, and small-scale gravel, was digitally mapped using a high resolution laser scanner (Figure 2). The resulting bathymetry was then fed into the Virtual StreamLab computer model whose resolution in space and time is sufficiently fine to model turbulent eddies shed by order-of-centimeter sized gravel within the stream (Figures 3 and 4). The computer model unveiled for the first time the physics of natural water flows at an unprecedented level of detail and realism.
Figure 4: Virtual StreamLab. Instantaneous high velocity streaks near the stream bed revealing the flow around individual small rocks. See also Video 2.
The ability to simulate water flow over natural topography with such a great degree of realism provides scientists and engineers with the physical insights they need to come up with effective and sustainable stream and river restoration strategies. The computer model will be used to conduct virtual experiments in cyber-space to optimize design and placement of man-made, stream-restoration structures to effectively suppress erosion and stabilize stream-banks. Coupled with physical experiments, the model can also be used to understand how turbulence impacts the ability of fish to swim, feed, and spawn in the stream and help restore aquatic habitats in degraded waterways.