Error propagation for velocity and shear stress prediction using 2D models for environmental management.

Gregory B. Pasternack, Andrew T. Gilbert, Joseph Wheaton, Evan M. Buckland

Research output: Contribution to journalArticlepeer-review

78 Citations (SciVal)

Abstract

Resource managers, scientists, government regulators, and stakeholders are considering sophisticated numerical models for managing complex environmental problems. In this study, observations from a river-rehabilitation experiment involving gravel augmentation and spawning habitat enhancement were used to assess sources and magnitudes of error in depth, velocity, and shear velocity predictions made at the 1-m scale with a commercial two-dimensional (depth-averaged) model. Error in 2D model depth prediction averaged 21%. This error was attributable to topographic survey resolution, which at 1 pt per 1.14 m2, was inadequate to resolve small humps and depressions influencing point measurements. Error in 2D model velocity prediction averaged 29%. More than half of this error was attributable to depth prediction error. Despite depth and velocity error, 56% of tested 2D model predictions of shear velocity were within the 95% confidence limit of the best field-based estimation method. Ninety percent of the error in shear velocity prediction was explained by velocity prediction error. Multiple field-based estimates of shear velocity differed by up to 160%, so the lower error for the 2D model’s predictions suggests such models are at least as accurate as field measurement. 2D models enable detailed, spatially distributed estimates compared to the small number measurable in a field campaign of comparable cost. They also can be used for design evaluation. Although such numerical models are limited to channel types adhering to model assumptions and yield predictions only accurate to not, vert, similar20–30%, they can provide a useful tool for river-rehabilitation design and assessment, including spatially diverse habitat heterogeneity as well as for pre- and post-project appraisal.
Original languageEnglish
Pages (from-to)227-241
Number of pages15
JournalJournal of Hydrology
Volume328
Issue number1-2
Early online date17 Feb 2006
DOIs
Publication statusPublished - 30 Aug 2006

Keywords

  • Shear stress
  • 2D models
  • River restoration
  • Hydraulics

Fingerprint

Dive into the research topics of 'Error propagation for velocity and shear stress prediction using 2D models for environmental management.'. Together they form a unique fingerprint.

Cite this