St. Louis District Using Micro Model
The Corps of Engineers in St. Louis is using a laser scanner to measure riverbed elevations in its micro river modeling system. This system relies on accurate measurements of the river bed model to determine sediment deposition and scour under simulated flow conditions. Capturing elevations with a mechanical digitizer requires an engineer to actually touch hundreds of points on the bed to capture their coordinates, which takes about 30 minutes and is subject to error because of the variability induced by a human operator positioning the probe. Also, the bed is not a hard surface, so the operator must take care not to disturb it. A laser scanner captures elevations in 10 minutes, using a laser light that moves over the surface. With this method, nothing physically touches the model. The difference in cost between the two methods has been estimated at $10,000 per study.
The micro river model was developed and is operated by the Applied River Engineering Center at St. Louis, Missouri. This is an extremely small scale, physical sediment transport model. The size of a tabletop, the model is about 5000 times smaller than the typical large modes. Cost is far less, and the miniature scale allows faster simulations, usually about three months, which reduces the cost of a typical sedimentation study to around $70,000.
Micro modeling uses synthetic materials and computer automation to simulate forces in a river created by the geology of the surrounding land, the geomorphology of the river channel, the slope of the floodplain and the annual flood cycles or hydrographs. The first step in the modeling process involves the creation of a scaled down replica of the river from survey data. The existing boundaries are constructed according to detailed, high-resolution aerial photographic maps, using polystyrene to form a model insert that represents the existing configuration of the river or stream. This insert is then placed within a hydraulic table-sized flume. The flume is custom-built with features such as rotational jacks to control the slope of the model, a water reservoir and a sediment chamber.
Five granular sizes of plastic sediment are used to represent the silt, sand, gravel and cobbles that comprise the sediment transported in the river. Steel mesh and clay are used to simulate the forces created by natural and artificial structures in the river, such as training structures, harbor facilities, islands and natural rock outcroppings. These are placed in the model in their exact scaled dimensions. The hydraulic processes of the river of stream under study are replicated by process control valves, centrifugal pumps and customized computer hardware and software. These devices allow the engineer to control the flow of water and sediment through the model, allowing the natural principles of moving water and sediment to develop duplicate bed forms of the actual river within the micro model.
When the system was first developed in 1994, topography measurements were taken with a digital micrometer. This device only allowed depth data to be collected, while x,y coordinates for each measurement had to be predetermined. After the depth was collected, each value had to be manually matched to its corresponding x,y coordinate. This process was time consuming and somewhat inaccurate. In 1996, a mechanical digitizer was introduced. Although this instrument greatly decreased the time of a study, while increasing the accuracy, there were a number of problems. One was that the tip of the digitizer had to actually touch the soft surface of the riverbed to determine the x,y,z coordinate location. If the operator wasn’t careful, the act of taking the measurement could invalidate the model. Highly trained operators were required. The method was time consuming, and there was a great deal of variability based on the skill of the operator.
When the price of laser scanners dropped, the model operators considered augmenting the mechanical digitizer with a 3D laser scanner. After evaluating the products on the market, the developers decided on the ModelMaker from Nvision in Irving, Texas. The major components of the system are a 3D laser sensor, a mechanical digitizer on which the sensor is attached, a PC and software that extracts, displays and manipulates the data.
Engineers and technicians need only minimal training to operate the ModelMaker. To determine elevations, they hold the laser sensor so that a line of laser light appears on the riverbed. The sensor is a single viewpoint laser stripe sensor. Laser stripe sensors, which are significantly faster than simple laser point sensors, work by projecting a line of laser light onto the object while a small CCD camera views the line as it appears on the surface. The sensor is attached to a FaroArm from FARP Technologys Inc. , Lake Mary, Florida. This mechanical arm moves freely about the riverbed, allowing the engineer to position the sensor easily and capture data rapidly and with high resolution.
While the mechanical digitizer used in the past provided a working sphere with a diameter of 66 inches, the laser scanner provides a diameter of about 110 inches. The micro models used by the Corps are about 76 inches long. Therefore, the mechanical digitizer had to be repositioned to cover the full length of models, while the laser scanner can cover them in just one pass. This results in significant additional time savings.
As the engineer moves the sensor over the surface of the riverbed, a dedicated interface card translates the video image of the line into more than 500 3D coordinates, allowing maximum data capture rate of 14,000 points per second. These data are combined with the Cartesian and angular coordinates generated at each position of the digitizer. The result is a dense cloud of 3D data describing the surface of the object. After water and sediment are run in the model, the scanning system collects hundreds of thousands of data points over the plastic sediment bed. The data are processed, converted to real world coordinates and rendered into bathymetric survey maps. These maps are then compared to hydrographic survey maps from the actual river or stream to determine if the model reacts similarly to the real world. Once the model has been validated, it can be used to predict qualitative changes resulting from remedial design measures. Many designs can be quickly analyzed using the micro modeling method. Another benefit of this method is that it helps engineers convey highly complex hydraulic concepts to people who don’t have engineering backgrounds, such as biologists, farmers, towboat pilots and landowners.
The micro modeling approach has been used to design many construction projects. These include the environmental restoration of side channels on the Mississippi River, improvements to sediment and flow conditions in the Mississippi and Missouri Rivers, evaluation and remediation of hazardous navigation conditions at a lock and dam, and protection of highway bridge crossings on small streams. In each situation, the bed and flow of the river reacted as predicted in the model. For projects that can’t afford a multi-million dollar analysis, the micro method delivers a fast, affordable and accurate alternative to large-scale river modeling.