Risk-Based Approach for Bridge Scour Prediction

Current practice for the prediction of scour depth at bridge piers and abutments uses empirical equations developed primarily from laboratory-scale studies, supplemented by limited data from field measurements. Equations for contraction scour (both clear-water and live-bed conditions) are based on an approach that combines both empirical and deterministic relationships. Additionally, the statistical analysis that was performed on the data collected from the laboratory studies and was used to create these relationships employs various statistical approaches that possibly provide more conservative results than necessary. When you also take into account the uncertainty associated with the development of key parameters used in the empirical relationships, the room for error is significant. In contrast, because of numerous advantages, bridge structural engineers, and more recently geotechnical engineers, have adopted Load and Resistance Factor Design (LRFD) which is a probabilistic approach to design. LRFD considers a probabilistic approach and allows for the possibility of assessing the level of risk associated with a given design. There is a need for the bridge scour engineer to have the option of performing scour calculations using probabilistic methods so that risk can be more appropriately assessed and the option of something other than the most conservative design considered. Current practice for determining the total scour prism at a bridge crossing involves the calculation of various scour components (e.g., pier scour, abutment scour, contraction scour, and long-term channel changes). Using the principle of superposition, the components are considered additive and the scour prism then is drawn as a single line for each frequency flood event (e.g., 50-year, 100-year and 500-year flood events). This approach does not provide an indication of the uncertainty involved in the computation of any of the additive components. Uncertainties in hydrologic and hydraulic models and the resulting uncertainty of relevant inputs (e.g., design discharge, flow duration, velocity, depth, flow direction, etc.) to the scour calculations will have a significant influence on scour prediction. To develop an overall estimate of confidence in the estimated scour magnitude, one must examine the level of confidence associated with the results of the hydrologic analysis (design discharges, flow duration, etc.), the level of confidence associated with the hydraulic analysis (depths, velocities, flow direction, etc.), and the level of confidence associated with the scour estimates (pier, abutment, contraction, long-term channel changes, etc.). Scour reliability analysis involves quantification of the uncertainties in each of these steps and then combines them in such a way that the overall estimate of confidence is known for the final prediction of scour. For the hydrologic analysis component, the desired end product could result in a probability density function (PDF) of the peak discharge. This can be done by examination of the flood flow frequency curve developed from gage records. If no gage records are available and regional regression equations are used, levels of confidence based on the results of the statistical analysis used to develop the regression equations can be used. If a single or lumped-parameter hydrologic model is used, important parameters could be identified, a PDF developed for these parameters, and a Monte Carlo simulation of these parameters could be performed to obtain the PDF of the peak discharge. The same can be performed for the hydraulic model except that the PDFs of the relevant hydraulic parameters would be developed using Monte Carlo simulations. Current practice provides an estimate of scour based on the hydrologic and hydraulic conditions associated with a specified design event (a 100- or 500-year flood, for example). The scour equations are generally understood to be conservative in nature, and have been developed as "envelope" curves for use in design. The research objective is to develop a methodology that can be used in calculating bridge scour so that the scour estimate can be linked to a probability; for example, there is a 95.0% probability that the maximum scour will be 8.3 feet or less over the life of the bridge. To achieve this objective, at a minimum the following tasks must be performed: (1) Review of existing knowledge: Some work along these lines has already been done in the area of hydrologic and hydraulic analysis. Relating the uncertainty associated with the hydrologic and hydraulic analysis to the uncertainty associated with the scour estimation techniques needs to be performed. Other disciplines where risk and reliability approaches are being integrated into engineering design also should be explored and documented by the research team (2) Identify uncertainties: This task will consist of identifying and evaluating the parameters associated with each of the various components (hydrology, hydraulics, and scour). (3) Formulate the methodology<em>: </em>This task will consist of combining the uncertainty associated with each of the various components (hydrology, hydraulics, and scour) into a procedure to use for scour prediction. The results of this task will ultimately lead to a probabilistic method to compute and evaluate bridge scour that will be consistent with LRFD approaches used by structural and geotechnical engineers. (4) Proof of concept: This task will consist of validating the methodology against data sets where variability in measured scour has been quantified. The new methodology must be demonstrated to be consistent with probabilistic approaches currently used by bridge structural and geotechnical engineers. >Final Report: The final report will be written in two parts. The first part will document the research performed to arrive at the methodology. The second part will be written in the form of a manual that provides design guidelines for practitioners in the field of bridge scour calculation. Currently scour estimates at bridge foundations use the best available technology, but are still roundly criticized as being overly conservative. The most common complaint is that the equations that were developed under laboratory conditions don't fit conditions at the site. Often this results in deeper foundations than necessary which leads to more costly bridge designs, which can stress already overloaded state department of Transportation budgets for bridge replacement and repair. Bridge designers and engineers are in need of a tool to make cost versus reliability tradeoff decisions with respect to scour and foundation design. A reliability-based design procedure for estimating scour at bridges will provide a consistent methodology for making decisions on design scour depth based on calculated risk instead of estimates which can be overly conservative. The pay-off is a scour estimate that will be more reliable in that it will be tied to a selected level of reliability that can be effectively communicated to the public. This type of approach will help alleviate over-conservatism in bridge design inconsistent with accepted target risk levels.<div></div>


  • English


  • Status: Completed
  • Contract Numbers:

    Project 24-34

  • Sponsor Organizations:

    Federal Highway Administration

    1200 New Jersey Avenue, SE
    Washington, DC  United States  20590

    American Association of State Highway and Transportation Officials (AASHTO)

    444 North Capitol Street, NW
    Washington, DC  United States  20001

    National Cooperative Highway Research Program

    Transportation Research Board
    500 Fifth Street, NW
    Washington, DC  United States  20001
  • Project Managers:

    Dekelbab, Waseem

  • Performing Organizations:

    Ayres Associates, Incorporated

  • Principal Investigators:

    Lagasse, Peter

  • Start Date: 20100421
  • Expected Completion Date: 0
  • Actual Completion Date: 20130628
  • Source Data: RiP Project 22374

Subject/Index Terms

Filing Info

  • Accession Number: 01464179
  • Record Type: Research project
  • Source Agency: National Cooperative Highway Research Program
  • Contract Numbers: Project 24-34
  • Files: TRB, RiP, USDOT
  • Created Date: Jan 3 2013 2:37PM