Updating HEC-18 Pier Scour Equations for Noncohesive Soils

Chapter 4. Conclusions and Recommendations

A dataset of 594 pier scour observations from 2 laboratory and 3 field studies was compiled. The dataset served as the testing ground for evaluating potential enhancements to the pier scour tools for noncohesive soils in HEC-18. In the current (fifth) edition of HEC-18, there are two primary equations for pier scour in noncohesive soils. The first is the general equation applicable to most situations including clear water and live bed conditions. The second is a coarse bed material equation recommended only for use under clear water conditions with coarse bed materials. Coarse bed materials are defined as those where the D₅₀ is greater than or equal to 0.79 inches (20.1 mm) and the gradation coefficient, σ, is greater than or equal to 1.5.

The original objective of this research was to determine if the coarse bed materials equation described in figure 1 could be used for conditions beyond those under which it is currently limited. A framework for evaluating the pier scour equations was developed using qualitative and quantitative tools. Qualitative tools included visual inspection of comparative graphs. Quantitative tools included the RI as a measure of equation reliability and the RRMSE as a measure of equation accuracy.

The HN/GC equation was adjusted to a target RI of 2.0. It was evaluated by partitioning the dataset into two groups based on the key conditions, including the HEC-18 coarse bed criteria, clear water versus live bed transport conditions, gradation, and median grain size. The equation performed reasonably consistently in all partitioned datasets; this result leads to the conclusion that it can be used for a broader set of conditions.

A subset of the dataset included pier group scour observations. The equation performed better for single piers but offered a basis for predicting local scour at pier groups. The HN/GC equation tended to provide predictions higher than desired for pier groups and could be adjusted downward.

Considering all of these findings, the HN/GC equation in figure 19 is recommended for a broad range of conditions. It is further recommended that the application be generally limited to the following conditions:

• Clear water or live bed conditions (V₁/V_c,50 < 5.2).
• Sands, gravels, and cobbles (0.0079 inches (0.21 mm) < D₅₀ < 5.0 inches (127 mm)).
• Gradation coefficients (σ) less than 7.5.
• Fr less than 1.7.
• Single piers.

Designers may choose to extend these limits after site-specific evaluation of the risks and alternatives. The evaluation of pier groups in the previous chapter may be a starting point for considering the application of the equation to pier groups.

In addition to these primary conclusions and recommendations, this study of pier scour equations revealed the following issues, which warrant further investigation:

• Trends in errors did not exist for most variables involved with scour prediction. The one exception was that for both equations (HEC-18 general scour and HN/GC), trends in error were observed with increases in the ratio of approach depth to median grain size (y₁/D₅₀). These trends should be further investigated and possibly eliminated.
  
  
• Refinements either to the HEC-18 method for analyzing pier groups or an adjustment to the recommended equation should be provided to improve scour predictions for pier groups.
  
  
• The RI was employed as a quantitative measure of reliability indicating the likelihood of overpredicting actual scour. A target value of 2.0 was selected for this study based on an assessment of the reliability of current practice. Further evaluation of the appropriate reliability level would inform trade-offs between conservativism and costs.
  
  
• Investigation of other quantitative means of estimating reliability and accuracy could improve the development and testing of scour estimation methods.