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.
9-1  9.1 Observations Bridge waterways commonly narrow or constrict natural channels, forcing water to flow through a contracted area, thereby increasing the magnitude of velocity and turbulent kinetic energy of flow passing through the waterway. If these increases cause erosion of the waterway boundaries, the contracted section may scour. Existing procedures for estimating contraction scour assume a relatively simplified situation where the contracted channel is straight in alignment, rectangular in cross section, with banks that are resistant to lateral erosion. It is also generally assumed that the bed is formed of uniform non-cohesive sediment. The resulting vertical erosion or scour in a constricted channel is commonly termed contraction scour. Live-bed and clear-water scour can occur along channel contractions. The former scour condition commonly occurs in the main channel of an alluvial river, while the latter condition is more typical for a floodplain contraction or at a relief bridge located on the floodplain. For clear-water scour, the governing principle is that the depth of scour in the contracted section corresponds to the occurrence of critical bed shear stress or mean flow velocity when scour approaches its equilibrium state. For live-bed contraction scour, the limiting condition is continuity of sediment transport between the upstream approach flow section and the contracted section. An additional consideration in the live-bed condition (generally overlooked in prior studies) is the role of differing bedform morphology in the approach and contracted channels. Existing equations are based on sediment transport theory using approaches developed over 50 years ago by Laursen (1960; live-bed contraction scour) and Laursen (1963; clear-water contraction scour). Both equations assume that the scour is due solely to the contraction effect and that local effects are negligible (i.e., that the contraction is hydraulically long), and both solve for the depth of flow (y2) in the contracted section after scour has occurred. Early work on contraction scour did not take into account additional factors that can influence scour depth that need to be considered for estimation of the actual depth of contraction scour as described in this report. The research approach for NCHRP Project 24-47, âRevised Clear-Water and Live-Bed Contraction Scour Analysisâ included the following: ⢠Fundamental re-analysis of the hydraulics of open-channel flow contractions ⢠Evaluation of existing contraction scour equations with reference to available laboratory and field data ⢠Extensive laboratory testing to develop more reliable data on clear-water and live-bed contraction scour ⢠Computational modeling to supplement the results of laboratory testing ⢠Proposing modifications to the existing contraction scour equations C H A P T E R 9 Observations, Conclusions, and Suggested Research
9-2 Revised Clear-Water and Live-Bed Contraction Scour Analysis ⢠Application of the revised equations to a typical field case of contracted flow in a bridge reach and comparison of the results with estimates obtained from the existing equations ⢠Evaluating the reliability of the existing and recommended analysis approaches using the laboratory database developed under this study This study showed that the deepest region of contraction scour (not including the entrance corners) occurred along the vena-contracta formed by flow entering a contraction. Therefore, this report focused on this region and refers to the remainder of the contraction as the long portion of the contraction. Most bridge crossings are in the region this report calls a short contraction. Though many factors contribute uncertainty to estimates of contraction scour, the uncertainty analysis of Section 8.5 was limited to comparisons of the proposed and existing methods for scour depth estimation. The other sources of uncertainty remain topics for further research. 9.2 Conclusions This section summarizes the main conclusions drawn from the flume experiments and related analyses of this study. Section 9.3 provides recommendations for further research. The open-channel contractions of interest for this study were contractions associated with rivers at bridge crossings. For most waterway crossings, the bridge is located in the immediate vicinity of the short segment of the contraction. Moreover, the contractions were taken to be superimposed on a channel that was initially uniform in width. This arrangement is representa- tive of bridge crossings on rivers having a single main channel and no floodplain. This study did not include channels with a substantial floodplain (i.e., compound channels). The main conclusions to be drawn from this study are as follows: 1. The description of open-channel hydraulics given in Chapter 2 was necessary to explain the flow characteristics of a contracted reach. Such a description was missing in all prior studies (i.e., prior studies did not adequately identify the complexity of the flow hydraulics associated with scour along contracted channels). 2. Given the initial conditions of a level bed (before scour commences), the approach flow passes through a contraction and enters a narrower channel. The resulting flow profile depends on the extent to which the contraction chokes the approach flow and forces the water level to rise at the contraction entrance, creating a backwater water surface profile extending upstream of the entrance. The magnitude of choking or water-level rise varies from negligible to substantial, depending on the geometry of the contraction and the length of the contracted channel. 3. Choking occurs when the approach flow has insufficient specific energy (energy relative to channel bed) to pass through the contraction without increasing the flow depth such that the flow becomes supercritical as it enters the contracted channel, thereby altering the flow field at the contraction entrance. Previous studies mixed choking and no-choking conditions (e.g., Dey and Raikar 2005; Gill 1981; Webby 1984). 4. The flow entering a contracted channel forms a vena-contracta (see Figure 8-2) whereby the actual width of the contracted flow is less than the geometric width of the contracted channel. The scour depth in the vena-contracta region within the contraction entrance gives the deepest scour along a contracted channel (other than the scour at the corners of the entrance). The corner scour at the contraction entrance may be influenced more by an abutment effect than by the width of the contraction, while the scour in the vena-contracta region of the short-contraction segment relates more directly to the contraction effect. 5. For the long-contraction reach, the measured depth of contraction scour along the con- tracted channel was not constant. Instead, the bed shear stress varies along the contraction
Observations, Conclusions, and Suggested Research 9-3 and typically causes the bed-surface profile to become shaped like a gently upwardly curved mound. The form of the mound was complicated by the non-uniform development of bedforms along the contracted channel. The flume experiments showed that the dimensions of the bedforms varied spatially with distance along the contracted channel. The influence of bedforms on the same order of magnitude as the scour depth in the flume complicated the measurement of scour depth and added an additional source of uncertainty to measure- ments at most locations along the contracted channel. 6. Considerations of open-channel hydraulics indicate that the depth of scour along the long- contraction segment will vary as the length of the contracted channel varies. Lengthening the contracted channel increases flow resistance due to an overall increase in shear stress along the channel bed. The analyses associated with the contraction scour equations in HEC-18 do not account for the effect of contraction length and are applied in practice to hydraulically short contractions. 7. While it is convenient to define the approach flow for a contraction in terms of the flow depth measured at or near the approach to a contracted channel (say, at a bridge-waterway crossing), it is important to recognize that a flow depth in this region is within the backwater water surface profile created by the contracted channel. None of the published datasets from contraction scour studies measured the depth of flow in the contracted section before scour occurred. To evaluate the datasets, this value had to be calculated or estimated. All but one of the published studies assumed that the depth of flow in the contracted section prior to scour was the same as the depth of flow in the approach section, thereby ignoring the importance of upstream backwater and hydraulic drawdown in the contraction. Also, most previous studies were done under clear-water conditions. 8. Upstream (approach) section hydraulic conditions influence sediment supply and vary significantly through the scour process, tending to reduce the excess shear and backwater associated with the flow contraction. HEC-18 methods as applied in practice use pre-scour rigid-bed modeling and predicted hydraulic conditions to predict equilibrium post-scour flow depths. 9. The overall trends for the depth of contraction scour in the vena-contracta region are delineated qualitatively in Figure 8-1. Conceptually, the trends indicate that when the bed shear stress causes bed-sediment transport along the approach channel bed and into the contracted channel (i.e., the shear-stress ratio exceeds one), the depth of contraction scour is not as deep as when the bed shear-stress increases without an inflow of bed sediment. In other words, the depth of scour during the live-bed condition of scour is less than it is for a clear-water condition. This finding is supported by the data trends obtained from the flume experiments for the present study. Prior studies do not consider general trends for contraction scour. The fact that in the contraction more bed sediment moves in suspension rather than as bedload also complicates the scour trends for high values of bed shear stress. Therefore, the balance assumed for live-bed scour should only apply to bedload entering the vena-contracta region, because the suspended load passes over the region of contraction scour. This statement requires further investigation (see Section 9.3). 10. A regression analysis was developed from the measured data obtained from the flume experi- ments to estimate the suggested vena-contracta correction coefficient, Kv, (see Figure 8-26). This equation agreed well with the measured flume data. Revisions to the existing clear- water and live-bed contraction scour equations based on this correction factor are suggested for use when estimating contraction scour in channels whose beds are formed of relatively uniform, non-cohesive sediment, and the width-to-depth ratio and entrance conditions are comparable with those used for this study. In general, these conditions apply to bridges classified as âhydraulically narrowâ by FHWA in the NBI. Bridges in this category constitute about one-third of the approximately 600,000 bridges in the NBI (see Sections 8.3 and 8.4).
9-4 Revised Clear-Water and Live-Bed Contraction Scour Analysis 11. The widely used HEC-18 equations for estimating the depth of contraction scour were adjusted to include a coefficient for estimating the minimum width of the vena-contracta. The adjustments were made to the HEC-18 equations used for the clear-water and live- bed conditions of contraction scour and are given in Section 8.3.3. The adjusted equations yielded results that agree well with the measured depths of contraction scour in the vena-contracta region (Section 8.3.6). 12. The adjustments were also applied to the NCHRP Project 24-20 form of the clear-water and live-bed contraction scour equations used as a base for estimating abutment scour (Ettema et al. 2010 and Arneson et al. 2012). These equations are given in Section 8.3.4. Again, the adjusted equations yielded results that agree well with the measured depths of contraction scour in the vena-contracta region (Section 8.3.6). 13. This study evaluated the reliability of the existing and proposed contraction scour equations (Section 8.5). The results indicate that the existing HEC-18 and NCHRP Project 24-20 equations, using upstream pre-scour hydraulic conditions as input, produce reasonable if slightly underpredicted estimates of equilibrium post-scour contraction scour flow depth. 14. The NCHRP Project 24-20 contraction scour equations produce less variability and are more conservative than the equations derived directly from HEC-18. 15. The application of the Kv factor developed for this study significantly increases the statistical reliability of both the HEC-18 and NCHRP Project 24-20 equations. This Kv factor applica- tion is suggested for hydraulically narrow single-span bridges with wingwall abutments (see Conclusion 10). Application of the Kv concept is also limited to streams with relatively uniform, non-cohesive bed material. Further research will be required to generalize this studyâs findings. 16. In the interim, for bridges that do not meet the criteria of Conclusion 15, the application of the existing best-practice modeling methods (see, for example, Robinson et al. 2019) and the NCHRP Project 24-20 contraction scour equations are recommended. 17. Using pre-scour contracted-section flow depth, yo, as the vertical (zero) datum to evaluate the post-scour change in bed elevation resulted in significant variability and increased under- prediction of equilibrium scour. This study suggests use of the downstream (exit) flow depth, y3, as a candidate for evaluating the change in bed elevation relative to the equilib- rium post-scour flow depth predicted by HEC-18 and NCHRP Project 24-20 contraction scour methods. This approach is promising but could not be fully evaluated with the current studyâs experimental design and scope (see Section 9.3). 18. Using theoretical normal-depth hydraulics to predict post-scour flow depths and scour resulted in substantial underprediction of observed scour and is not recommended. 19. The present study considered the use of field data, as well as laboratory data for the revised analysis of the contraction scour equations. It was concluded that there is a lack of reliable field data for comparison with either the existing or revised contraction scour equations. The preponderance of the available field data include the influence of abutments and, in some cases, the presence of bridge piers. Additional considerations included (1) available reports regarding field data are not supported by a description of the hydraulics of the contracted reaches involved; (2) at bridge sites in the field, contraction scour can involve both vertical and lateral erosion of a channel, which complicates the interpretation of the data; and (3) temporal considerations add uncertainty regarding the relationship between measurements and equilibrium or ultimate scour conditions at a field site. 20. A general conclusion drawn from this study is that the inherently complex nature of con- traction scour (involving non-uniform and unsteady flow on a mobile, non-cohesive bed) leads to approximate estimates of depth of contraction scour. However, reliability consider- ations (Conclusions 13 through 15) provide a guide to the variability of contraction scour estimates and suggest methods for improved evaluation techniques for contraction scour of direct use to the practitioner.
Observations, Conclusions, and Suggested Research 9-5 9.3 Suggestions for Further Research The results of this updated and revised analysis of the hydraulics of contraction scour lead to a set of recommendations regarding topics for further investigation: 1. The values of the vena-contracta coefficient, Kv, should be determined for different entrance shapes, notably shapes typical of bridge abutments (e.g., spill-through or vertical wall abutments), which typically form the entrance to bridge-waterway crossings. Additional experimental studies are needed to estimate Kv values because 1D and 2D numerical models do not accurately predict flow separation regions at the entrance to the contraction. 2. Additional research is needed to refine the application of the vena-contracta coefficient, Kv, and generalize its application. Calibrated CFD modeling would support such research. 3. Additional research needs to be conducted for beds consisting of non-uniform and coarser sediments. Such beds may develop armor layers. The present flume experiments using a relatively uniform 0.26-mm sand did not evaluate how sensitive contraction scour (and corner scour) might be to armoring effects. 4. Experiments should be conducted for ranges of shear-stress ratio greater than those used in the present study. These experiments are needed to confirm or modify the conceptual trends evaluated in this study. 5. The influence on contraction scour exerted by the presence of bridge piers should be investi- gated. Piers would modify the vena-contracta region and generate additional, and presently unaccounted for, turbulence in the region. 6. Research is needed on the relative influence of suspended sediment load versus bedload in the contracted reach. 7. Bed scour resistance and geotechnical issues at bridge contractions should be investigated. The findings of this study should be generalized to accommodate scour estimation in cohesive materials and erodible bedrock. 8. Additional laboratory or computational investigations would be required to evaluate the influence of entrance conditions and angle of attack (skew) at bridge contractions. 9. Additional laboratory or computational investigations are needed to evaluate the suitability of using a downstream flow depth as the datum for calculating contraction scour from post- scour flow depths. 10. Further research on the application of force decay concepts (introduced in Section 8.2.6) would improve our understanding of scour processes at bridges with widely graded or cohesive bed material, and for bridges that are founded on soil layers with variable erod- ibility (Items 3 and 7). Existing scour prediction methods, including those developed under this study, may not be applicable to these more complex conditions.
