Scour at Bridge Foundations on Rock (2012)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

6Research Approach 2.1 Overview This research project produced results on the following two related aspects of rock scour problems: 1. Geotechnical characteristics of rock masses exposed to flowing water and 2. Forces generated by the flowing water on a daily basis and accumulated over a period of years to represent the service life of a bridge. Sand-bed channels respond to flowing water in a way characterized by the threshold force required to initiate movement of the sand particles. This threshold concept clearly applies to certain rock-mass conditions and may apply more broadly to rock masses exposed to flowing water. A sketch of the familiar Hjulstrom diagram (Krumbein and Sloss, 1963; Graf, 1971) is shown in Figure 2.1 and modified to show a continuum of increasing block size to the right of “boulders” and increasing cohesion and cementation to the left of “clay.” Gravel- and boulder- size fragments (blocks) of in-place, fractured, durable bedrock that are completely detached from each other exhibit threshold-controlled behavior in a manner somewhat similar to gravel- and boulder-size fragments of stream-bed materials; in rock-bed channels, the process of block removal is called quarrying and plucking. Non-durable or erodible rock masses respond to flowing water in a gradual and cumulative manner, whether or not a threshold condition is required for the onset of erosion. The primary difference between scour of sand-bed channels and rock-bed channels is that sand-bed material tends to be re-deposited during the waning stages of a flood, whereas erosion of rock-bed material is permanent. If a scour hole in a rock-bed channel fills during the waning stages of a flood, it will be filled with sand that responds to future flood events in ways that can be predicted by procedures described in HEC 18. As noted in HEC 18 Section 12.10, the rock scour problem is determining if rock is resistant to scour. An important discovery made during the course of this research project is that the power of water in the channel flowing past the bridge is as important in rock scour as the properties of the rock mass providing the bridge foundation. An initial task was to define rock scour modes, each of which implies stream flow conditions. HEC 18 expresses the forces of flowing water in terms of velocity or hydraulic shear stress asso- ciated with peak discharge events. Peak velocity and peak hydraulic shear stress are appropriate parameters for which sand-bed-channel response can be predicted because of the threshold- controlled behavior of sand. Hydraulic forces that produce gradual effects that accumulate over time on materials not controlled by exceedance of threshold conditions cannot be expressed solely as peak velocity or peak shear stress. A single hydraulic parameter that can be accumu- lated is stream power. Stream power is the product of flow velocity and hydraulic shear stress; thus, stream power is a rich hydraulic parameter because it incorporates flow depth, hydraulic C h a p t e r 2

research approach 7 gradient, and discharge. Furthermore, stream power can be integrated over time and expressed as the cumulative stream power at a bridge site for the service life of a bridge. For erodible rock materials, this research utilizes stream power for these reasons. Prediction of scour of rock-bed channels is a rock-water interaction process that is similar to prediction of load-carrying capacity of bridge foundations as a soil-structure interaction pro- cess. The bearing capacity of a soil deposit is related directly to its geotechnical parameters, such as unit weight, shear strength, and deformation modulus, but it also is related directly to the structural footing parameters, such as dimensions, bearing depth, and stiffness. Thus, the scour resistance of a rock mass is not solely a geotechnical factor, but also depends on the hydrau- lic loading conditions. As an example to illustrate the hydraulic loading conditions, concrete, stone—and even steel—are cut routinely with high-pressure water jets (Summers, 1995). This research utilizes a probability-weighted approach to expressing flood frequency for erod- ible rock masses. This approach was first applied to sediment yield in arroyo channels in the southwestern United States by Lagasse et al. (1985) to show annualized probability of sediment yield. The area under the annualized probability curve is the average annual sediment yield. This approach is used in the rock scour research by transforming the traditional annualized prob- ability of peak discharge from a flood-frequency analysis into the scour depth associated with each of the flood events to produce a curve of annualized scour depths. The area under the curve is the average annual scour for the hydraulic conditions represented by the flood-frequency analysis results. The research objective involving time-rate of scour is provided by the average annual scour from the probability-weighted approach. The research objective involving design scour depth is provided by the product of the average annual scour and the remaining life of an existing bridge or the design life of a new bridge. 2.2 Research Plan Modifications Modifications to the research plan were made during the project regarding • Flume tests, • Rock mechanics, • Hydrology and hydraulics at bridge sites on ungaged streams, and • Benchmark materials for rock. Laboratory flume tests seemed to be an important technical component of the rock scour research, and they were included in the initial scope and budget. Panel review comments to Figure 2.1. Schematic representation of the Hjulstrom diagram modified to include rock masses.

8 Scour at Bridge Foundations on rock the proposal included a recommendation that flume tests be eliminated. Flume studies were removed from the research scope and the budget reallocated to other tasks. The need for flume studies in rock scour research was recognized during the project and it is included in recom- mendations for future research. Rock mechanics was viewed initially as a key discipline for quantifying rock-mass response to turbulence-induced pressures for time-rate of scour. Probabilistic fracture propagation through intact rock was one of the important rock mechanics topics. Numerical simulation of thresh- old flow conditions for block lifting included in the initial research was modified to focus on bedrock conditions in which the rock blocks were completely detached from each other and of uniform, regular shape, but fundamental rock mechanics modeling of fracture propagation was eliminated from this research project. The numerical modeling in this research project con- sidered the fluctuating turbulence intensity and other hydraulic forces around bridge piers in natural channels but excluded the plunging jet conditions common in some dam spillways that have produced impressive scour holes in hard rock. As a result of the interim report meeting, the research team was directed to conduct hydro- logic and hydraulic analyses to develop guidance for use at bridge sites on gaged and ungaged streams. Traditional methods of flood frequency analysis and flow synthesis were incorporated into the rock scour approach. A probability-weighted approach was developed to represent flood frequency results in scour-relevant terms based on stream power. As a result of the interim report meeting, the research team developed a definition for “rock” as a scour-resistant material. Identification of a benchmark material resistant to scour processes for use in rapid screening of bridge sites seemed to be desirable and concrete was selected as a candidate material. The research team considered rock-like materials of lower quality than concrete for a benchmark scour-resistant material. 2.3 Research Tasks Considering the research approach discussed and outlined, the following specific tasks were completed to accomplish project objectives. These tasks incorporate panel guidance and, with some modifications, are parallel to those suggested in the original research project statement. 2.3.1 Phase I Task 1—Review the Technical Literature: Technical literature from domestic and foreign sources was reviewed for information pertaining to scour of bridge foundations on rock. The literature review attempted to identify research in progress as well as completed work. Rock and rock-like materials have a broad range of geologic settings and geotechnical characteristics; the research team believes that the definitions and usages of the term “rock” applied in the literature search produced appropriate results. Task 2—Conduct Survey of State and Federal Agencies: The research team collected infor- mation from state and federal agency personnel to determine various practices used for estimat- ing the extent and depth of bridge foundation scour in rock and to identify bridges experiencing significant scour in rock. A commercial online survey utility was used to administer a question- naire and compile the results. The questionnaire and survey results are provided in Appendix B. Task 3—Analyze Information and Propose Preliminary Methodology: A preliminary meth- odology was proposed in the interim report based on the literature review and survey results.

research approach 9 The preliminary methodology focused on screening to identify modes of rock scour so that only relevant modes were evaluated for further quantification. Task 4—Interim Report and Updated Phase II Work Plan: The research team prepared and submitted an interim report documenting the information developed in Tasks 1 through 3. The interim report contained a detailed discussion of the Task 3 rock scour modes and summarized the Task 1 literature review and Task 2 survey results. The research team met with the NCHRP 24-29 panel members to discuss the interim report and the focus of the Phase II Work Plan. The Phase II Work Plan was finalized after the interim report meeting was completed. 2.3.2 Phase II Task 5—Investigate Bridge Sites: Five bridge sites were visited by the research team. Few candidate bridge sites were identified in the Task 2 survey, but three candidate bridge sites were identified from the literature (Interstate Highway 90 over Schoharie Creek, New York; Interstate Highway 10 over Chipola River, Florida; and State Route 22 over Mill Creek, Oregon). A fourth bridge site was suggested by one of the panel members during the interim report meeting (State Route 262 over Montezuma Creek, Utah). The fifth bridge site was suggested by a participant enrolled in a National Highway Institute course on bridge scour being taught by the co-principal investigator of this research project (State Route 273 over Sacramento River, California). Each of the five bridge sites visited by the research team was located in a state represented by a research project panel member. Task 6—Conduct Laboratory, Field, and/or Modeling Studies: Laboratory, field, and mod- eling studies were conducted as part of the rock scour research. The goal of the laboratory, field, and modeling studies was to use conventional equipment and methods to the extent practical. Testing in a specialized device in Gainesville, Florida, was provided by the Florida Department of Transportation. The Utah Department of Transportation and the California Department of Transportation provided field and laboratory data for use by the research team. The New York State Department of Transportation and Oregon Department of Transportation provided drill rigs and crews for drilling borings specifically for the rock scour research project. Most of the laboratory testing conducted for this project was done on hand-samples using standard (ASTM) procedures. Modifications to a key test procedure made by researchers at Oregon State Univer- sity (Dickenson and Baillie, 1999) were used in this project. Modeling studies were performed to develop hydrology and hydraulic parameters, simulate quarrying and plucking of durable rock blocks, and characterize rock structure from bore hole and scan line data. Task 7—Develop Methodology for Determining Time-Rate of Scour and Scour Depth: The research team developed a methodology for screening bridge sites to determine modes of rock scour, including procedures for potentially dismissing two processes from further consideration (dissolution of soluble rock and cavitation). A methodology for determining scour depth was developed for two modes of scour (quarrying and plucking of durable rock blocks and progres- sive wear of degradable rock), whereas a methodology for determining time-rate of scour was developed for one mode of scour (progressive wear of degradable rock). The methodology for evaluating scour of degradable rock is based on cumulative stream power; otherwise the meth- odologies are consistent with the procedures described in HEC 18. Task 8—Develop Design and Construction Guidelines: The research team reviewed and evaluated the design and construction guidelines for rock foundations that are included in HEC 18 (Section 2.2 General Design Procedure, Steps 7.b and 7.c) and considered practices used by several departments of transportation and the U.S. Army Corps of Engineers. It appears that existing guidelines are generally applicable and that the findings of the rock scour research proj- ect may not require significant modification. Characterization of the geology and geotechnical

10 Scour at Bridge Foundations on rock conditions at bridge sites utilizes conventional procedures, but laboratory testing of samples of degradable rock uses a recommended procedure that differs from an ASTM standard. Task 9—Submit Final Report: The research team submitted a final report that documents the entire research effort and is published as NCHRP Report 717: Scour at Bridge Foundations on Rock. A summary was prepared and included to outline the research and describe key findings and recommendations. 2.4 Report Organization Findings from this research are available as follows: NCHRP Report 717 that contains • Findings from the review of literature and bridge site visit, • An overview of laboratory testing results, • Interpretation and appraisal of findings and results, • Guidelines for characterizing bridge sites with rock foundations, • A methodology for predicting time-rate of scour and design scour depth at bridge piers, • Conclusions and recommendations, and • Suggested areas for future research. Web-based appendixes as follows: • A bibliography from literature review, • A survey questionnaire and complete results, • Results of parametric numerical analyses of quarrying and plucking, • An overview of headcut erodibility index and erodibility index methods, • Descriptions of the bridge sites, • Definition of discontinuities from bore holes, and • Estimation of block size from field measurements.