Date of Award

Fall 2016

Document Type


Degree Name

Doctor of Philosophy in Civil Engineering - (Ph.D.)


Biomedical Engineering

First Advisor

John R. Schuring

Second Advisor

Robert Dresnack

Third Advisor

Fadi A. Karaa

Fourth Advisor

Yuan Ding

Fifth Advisor

Frank L. Golon


Bridge scour, which is the erosion of soil and rock around bridge abutments and piers, is the principal cause of bridge failure in the United States and around the world. Previous investigations of scour have focused mostly on fine sediments such as sand, silt, and small gravel, because such alluvium underlies a majority of bridges. The erosion resistance of coarser sediments has received limited attention to date, even though they dominate many small to medium-size rivers in the northern tiers of the United States, Europe, and Asia. This study focuses on the scour behavior of extremely coarse particles (ECP), namely cobbles and boulders.

A main objective of this research is to develop a relationship between critical (entrainment) velocity and grain size for sediment particles in the size range of 5 to 50 cm (2 to 20 in). This is accomplished by performing a limit analysis of sediments that exist within the stream beds of actual bridges. A basic premise is that the residual sediments reflect the maximum historic flow and velocity that has occurred over the life of the bridge.

Thirty-five bridges in Northern New Jersey are initially screened for the study, and 12 bridges are selected for final analysis. Field visits are made to characterize the grain size distribution of the ECP sediments present at each site. Due to the extreme coarseness of the sediments, nontraditional methods are employed such as optical granulometry and statistical pebble counts. To assure geologic and hydrologic diversity for the data set, the sites span three of New Jersey’s physiographic provinces: Highlands, Valley and Ridge, and Piedmont.

Hydraulic analyses are used to estimate the maximum velocity that the bridge has experienced during its lifetime. These are based on the maximum discharges measured at USGS gaging stations or computed with USGS StreamStats software. Final limit velocities range from 245 to 549 cm/sec (8 to 18 ft. /sec.).

The limit analysis is performed by regression of the particle size and velocity data. This yields a nonlinear, exponential relationship between critical velocity and median particle size. The variance fraction associated with the data set is 0.642, indicating a reasonable fit. The limit analysis results are also compared with traditional sediment transport relationships, including Newton’s Law and Hjulstrøm’s envelope.

Several applications of the limit analysis relationship are proposed and explored. The first is a method to assess the general scour risk for bridges underlain by ECP sediments. First, the median grain size of the bed sediments is measured. The corresponding limit velocity is then computed and compared with the desired scour design storm, e.g., Q100. If the design scour velocity exceeds the limit velocity, then the bridge is considered to have a higher scour risk.

The limit analysis curve is also used to extend the useful range of the standard scour design equations, including the HEC-18 critical velocity and USACE EM 1601 riprap relations. Extrapolation of the limit results generally produces lower and more conservative critical velocities within the ECP size range than did the standard relations. Another application provides adjustment coefficients, which are useful for rapid photographic measurement of sediments (size and gradation) using WipFrag.