Author : Jesse Rutledge Barker
Publisher :
ISBN 13 : 9781267398031
Total Pages : pages
Book Rating : 4.3/5 (98 download)
Book Synopsis Rapid, Abundant Velocity Observation to Validate Million-element 2D Hydrodynamic Models by : Jesse Rutledge Barker
Download or read book Rapid, Abundant Velocity Observation to Validate Million-element 2D Hydrodynamic Models written by Jesse Rutledge Barker and published by . This book was released on 2012 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: Two-dimensional (depth-averaged) hydrodynamic models have existed for decades and are used to study hydrogeomorphic processes, evaluate stage-dependent physical habitat patterns, and design river-rehabilitation projects. Rapid computer and coding advances are revolutionizing the size and detail of 2D models. Meanwhile, advances in topographic mapping and environmental informatics are providing the data inputs to drive large, detailed simulations. Million-element computational meshes are readily operational. With simulations of this size and detail, the primary challenge has shifted to finding rapid and inexpensive means for testing model predictions against observations. Standard methods for collecting velocity data include boat-mounted ADCP and point-based sensors on boats or wading rods. These methods are labor intensive and often limited to a narrow flow range. Also, they generate small datasets at a few cross-sections, which is inadequate to characterize the statistical structure of the relation between predictions and observations. Drawing on the long-standing oceanographic method of using drogues to track water currents, previous studies have demonstrated the potential of small dGPS units to obtain surface velocity in rivers. However, dGPS is too inaccurate to test 2D models. Also, there is financial risk in losing drogues in rough currents. In this study, a real time kinematic (RTK) GPS unit was mounted onto a manned whitewater kayak. The boater positioned himself into the current and used floating debris to maintain a speed and heading consistent with the ambient surface flow field. RTK GPS measurements were taken every 5 seconds. From these positions, near-surface 2D velocity vectors were obtained. The method was tested over ~7.2 km of the lower Yuba River in California in flows ranging from ~15-140 m3/s, yielding 5816 observations. To compare surface velocity magnitude against the 2D model-predicted depth-averaged value, kayak-based surface values were scaled down by a published value and three optimized constants, which had no negative effect on regression analysis. The R2 value for speed was 0.79 by this method, compared with 0.57 based on 199 points from traditional point measurements. The R2 value for velocity direction was 0.80. Although it is not ideal to rely on observed surface velocity to evaluate depth-average velocity predictions, all available velocity-measurement methods have a suite of assumptions and complications. Using this method, the availability of 10-100x more data was so beneficial that the outcome was among the highest 2D-model performance outcomes reported in the available literature.
