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.
44 C H A P T E R 5 5.1 Bank Protection Pronounced bank erosion has been observed at meander bends due to pressure-driven secondary currents. These are helicoidal flow patterns, directed from the outer to the inner bank along the channel bed, that result from the local imbal- ance between the curvature-induced transverse pressure gradient and the centrifugal force along the bed. Secondary currents result in an increase in water surface elevation on the outside bank, which is known as superelevation, and redistribute stream-wise momentum within the channel cross-section. Pressure-driven secondary circulation can persist downstream of the apex, leading to undesired scour or deposition (Army Corps of Engineers, 1991). In both meandering and straight reaches, unchecked scour at the toe of banks will eventually lead to failure (Odgaard, 1988); therefore, to evaluate bank protection, scour at the bank toe was analyzed. Both sill and single-arm structures were evaluated for their ability to provide bank protection. All of the single-arm structures were successful at shifting the channel thalweg toward the center of the channel (Figure 5-1), thus providing some level of bank protection. To compare the applicability of different structure types for bank protection, the length of protected bank for each structure was quantified in terms of the upstream (Lu) and downstream (Ld) length of bank where deposition was calculated relative to the structure-free channel (see Figure 3-3). The final optimized structure arrays were evaluated for the three single-arm structure types. In general, the BW arrays provided the greatest length of bank protection. In VSL-G, these structures provided consistent bank protection, with little scour along the inner bank and a thalweg located in the middle of the channel. RVs and JHs also provided bank pro- tection in their optimal set-up but with a less consistent thal- weg, less consistent bank protection, and risk of scour to the inner bank (Figure 5-2). For sill-type structures in VSL-G, the length of bank pro- tection was very similar for all structure types. Minimal protection was provided along the outer bank (Figure 5-2) downstream of the apex. However, quasi-equilibrium scour holes downstream of the sill structures extended to the outer bank [also observed in the OSL and indoor StreamLab (ISL) experiments], threatening bank stability on both banks downstream of the structure. In the VSL-S channel, all single-arm structure arrays pro- vided outer bank protection that extended through the mean- der test section (Figure 5-3). RVs provide a slightly greater length of bank protection, and BWs and RVs both resulted in a more consistent thalweg than JHs, which created more scouring of the inner point bar. In VSL-S, CVs and the modified CVA sill-type structures provided a greater length of bank protection than the WW (Figure 5-3). As in the VSL-G, OSL, and ISL channels, the scour hole downstream of the structures extended to either bank, therefore threatening bank stability immediately downstream of sill-type structures. 5.2 Scour Hole Dimensions The depth and location of scour holes are important for (1) structure stability and (2) the secondary goal of aquatic habitat (see Figure 5-4). In the VSL-G channel, JHs met their stated goal of creating scour holes for aquatic habitat (Rosgen, 2006). RVs created smaller, less deep scour holes, and BWs created the smallest scour hole. All three sill-type structures resulted in a large, deep scour hole downstream of the structures that extended to the outer bank in VSL-G (Fig- ures 5-5 and 5-6). Similar results were seen in the VSL-S chan- nel. JHs fulfilled their intended purpose of creating a scour hole, RVs had smaller scour holes, and BWs created a relatively con- sistent channel geometry with indistinct scour holes in VSL-S. All three sill-type structures in the VSL-S channel created large scour holes; however, in the more sinuous VSL-S channel, the Comparison of In-Stream Flow Control Structure Types
45 Figure 5-2. Comparison of bank protection for the final design layout for different structure types in VSL-G. The DZ represents the quasi-equilibrium bed topography of the baseline case (with no rock structure) subtracted from the quasi-equilibrium bed topography of the case with a rock structure. Flow is from left to right. 0.4 0.2 0.0 0.2 0.4 0.6 0.8 1.0 0.2 0 0.2 0.4 0.6 0.8 1 1.2 z/ H y/B VSL-G RV JH BW 1.0 0.5 0.0 0.5 1.0 0.2 0 0.2 0.4 0.6 0.8 1 1.2 z/ H y/B VSL-S RV JH BW Figure 5-1. Cross-sectional profile just downstream of meander apex for VSL-G (top) and VSL-S (bottom) illustrating the shift of the channel thalweg with single-arm structures installed. Zero on the x-axis is at the inner bank.
46 scour hole shape varied (See Figures 5-5 and 5-7). Standard CVs created the smallest scour hole, the addition of the step in the modified (A-shape) CVA created a wider scour hole, and the WW had the longest and widest scour hole. Deep scour holes adjacent to the structures are a cause for concern as one of the primary modes of failure of rock structures is the undermining and subsequent shifting of individual rocks. 5.3 Flanking Potential Flanking was determined to be a major failure mechanism of in-stream flow control structures (see Appendix B for sur- vey results and Appendix C for field case studies). Flanking is the circumvention of flow behind the structure. The risk of flanking was evaluated for each structure type by com- paring the ratio of deposition upstream of each structure (compared to the average bed elevation) to flow depth (H) for each structure type. These ratios are shown in Table 5-1. BWs showed the least risk of flanking as there was little to no deposition upstream (Figure 5-8). JHs and sill structures had a moderate risk of flanking, with ratios in the range of 0.3 to 0.4, and RVs showed the greatest risk of flanking, with ratios ranging up to 0.8. Particular care should be used in keying in RV structures to ensure that the structures remain intact. 5.4 Hydrodynamic Forces on Rock Structures In this section are presented calculated force distributions along the sill and single-arm rock structures for VSL-G and VSL-S. For all of the cases, the Reynolds numbers are over 106, which implies that the major force over any immersed rock can be attributed to form drag, which is calculated using the following relationship (Raudkivi, 1967): i i i i= Ï1 2 2F C A ud where F is the force on the rocks, r is density of water, A is the cross-sectional area of the single rock, u is the local velocity magnitude, and Cd is the drag coefficient. The drag force was calculated for the top row of rocks in each structure using the local velocity magnitude immediately upstream of the cen- ter mass of each rock. The rocks nearest the surface are sub- ject to the highest flow velocities and are, therefore, the most likely to be dislodged by flow. For VSL-G and VSL-S with a Reynolds number over 106, Cd is almost constant and roughly equal to 0.2 (Raudkivi, 1967). The calculated forces in this study were non-dimensionalized by the drag force due to the mean velocity in each channel, Fo, on a representative rock Figure 5-3. Comparison of bank protection for the final design layout for different structure types in VSL-S. The DZ represents the quasi-equilibrium bed topography of the baseline case (with no rock structure) subtracted from the quasi-equilibrium bed topography of the case with a rock structure. Flow is from left to right.
47 Figure 5-4. Comparison of scour holes for the final design layout for different single-arm structure types in VSL-G (left) and VSL-S (right).
48 -0.90 -0.70 -0.50 -0.30 -0.10 0.10 0.30 0.50 0 50 100 150 200 250 300 De pt h (m ) Downstream Distance (m) CV CV_ds CVA CVA_ds WW WW_ds max scour (no structures) Figure 5-6. Depth distribution illustrating scour hole length along the channel centerline for sill structures in VSL-G. Solid black line represents the maximum depth in VSL-G with no structures. Figure 5-5. Comparison of scour holes for the final design layout for different sill structure types in VSL-G ( left) and VSL-S (right).
49 hydrodynamic forces exerted by the turbulent flow. For the rocks in a BW, a roughly uniform distribution can be seen. In Figures 5-11 and 5-12, the non-dimensionalized force distribution along the rock sill structures, including CV, CVA, and WW, are shown for VSL-G and VSL-S, respectively. The force distribution was calculated along single sill structures located at the downstream end of the straight reach with an angle of 30° (for all of the cases). In Figures 5-11 and 5-12, the force over the key rocks of CVA is significantly higher at the middle of the key (the dashed red line). The solid red line also shows that the force distribution of CVA has two peaks at the points linked to the key. The individual rocks that are located in the middle of the WW and U-shaped CV will experience slightly higher force magnitude than those along the bank (see green and blue lines in Figures 5-11 and 5-12). 5.5 Backwater and Water Surface Elevation Impacts Since flow-training structures act as obstructions to flow and change the total channel roughness, it is expected that structures will affect the water surface elevation, particularly at flows at or near bankfull. For large floods, the impact of structures will diminish as they become a smaller portion of the total flow depth. The total effect on water surface will be dependent on relative submergence of structures as well as structure geometry and layout. Lower-profile structures, such as bendway weirs, have less influence on water surface elevation than sloping structures that reach the bankfull flow elevation such as rock vanes. This was demonstrated with two sets of simulations computing the turbulent flow and free- surface evolution over the quasi-equilibrium bed geometry in the VSL-S. The VSL3D water surface simulations were carried -1.35 -0.85 -0.35 0.15 0.65 0 50 100 150 200 250 300 350 De pt h (m ) Downstream Distance (m) CV CV_ds CVA CVA_ds WW WW_ds Max scour (no structures) Figure 5-7. Depth distribution illustrating scour hole length along the channel centerline for sill structures in VSL-S. Solid black line represents the maximum depth in VSL-S with no structures. Structure Deposition/(H) Susceptibility to Flanking RV 0.1-0.8 high JH 0.3-0.4 medium BW 0 low Sill (CV, WW) 0.3 medium Table 5-1. Relative susceptibility to flanking for in-stream flow control structures based on the ratio of deposition upstream of each structure to the flow depth (H). size (D100) for each structure (see Table 3-1 for rock sizes used in VSL3D simulations). The mean velocities for the bankfull flow condition were 1.33 m/s and 1.48 m/s for VSL-G and VSL-S, respectively. The representative area used to calculate Fo, using the force equation earlier in this paragraph, was defined as A = p/4(D100)2. To calculate the force distribution along each rock structure, the 3-D flow data along the struc- ture length were averaged to calculate the force distribution. The steady-state calculated hydrodynamic results (coincides with the quasi-equilibrium bed bathymetry) were employed to calculate the drag forces on rock structures. Figures 5-9 and 5-10 show the non-dimensionalized force distribution along a single RV, JH, and BW located at the apex in VSL-G and VSL-S, respectively. The hydrodynamic forces over the rocks in the hook part of a JH structure are drastically bigger than the forces on the rocks in the vane portion of the JH structure or the forces on RV or BW structures (Figures 5-9 and 5-10). Because JHs are constructed with gaps between the rocks in the hook, the forces on these individual rocks will be critical to the structureâs stability. Figures 5-9 and 5-10 also show that the rocks located near the RV tip bear most of the
50 Figure 5-8. Deposition (relative to average bed elevation) upstream of single-arm structures demonstrating risk of flanking in VSL-G ( left) and VSL-S (right). 0 1 2 3 4 5 6 0 0.2 0.4 0.6 0.8 1 VSL_G_BW50 VSL_G_RV30 VSL_G_JH30 L/Ls p o f s tr uc tu re st re am b an k F/ F o vane hook Figure 5-9. Steady-state force distribution over rocks along different structures in VSL-G. Forces are non-dimensionalized using the drag force on a single rock due to the mean flow velocity, Fo. The vane and hook sections are the hook and vane parts associated with the J-hook vane. Distance, L, is non-dimensionalized by the total length of each structure, Ls.
51 0 1 2 3 4 5 6 0 0.2 0.4 0.6 0.8 1 VSL_S_BW50 VSL_S_RV20 VSL_S_JH20 L/Ls p of st ru ct ur e st re am ba nk F/ F o hookvane Figure 5-10. Steady-state force distribution over rocks along different structures in VSL-S. Forces are non-dimensionalized using the drag force on a single rock due to the mean flow velocity, Fo. The vane and hook sections are the hook and vane parts associated with the J-hook vane. Distance, L, is non-dimensionalized by the total length of each structure, Ls. 0 2 4 6 8 10 0 0.2 0.4 0.6 0.8 1 VSL_G_CV30 VSL_G_WW30 VSL_G_CVA30 VSL_G_CVA30_Key L/Ls le ba nk rig ht b an k F/ F o Figure 5-11. Steady-state force distribution over rocks along different structures including cross vanes and a W-weir in VSL-G. The dashed line is the key part of the A-shaped cross vane. Forces are non-dimensionalized using the drag force on a single rock due to the mean flow velocity, Fo. Distance, L, is non-dimensionalized by the total length of each structure, Ls. out with three RVs and three BWs (Figures 5-13 and 5-14). RVs contributed to a maximum difference in water surface elevation of 7.4% (difference in water surface elevation between computed superelevation with structures and com- puted superelevation without structures divided by the mean flow depth). For BWs, the maximum water surface elevation difference was 5.9%. A similar comparison was done between all single-arm structures using the OSL water surface measurements (for more information about OSL experiments, see Appen- dix D). Similar to the VSL-S results, the impact of structures on water surface elevation was local, with backwater just upstream of the structures and a decrease in water surface elevation just downstream of the structures (Figure 5-15).
52 0 2 4 6 8 10 0 0.2 0.4 0.6 0.8 1 VSL_S_CV30 VSL_S_WW30 VSL_S_CVA30 VSL_S_CVA30_Key L/Ls le ba nk rig ht b an k F/ F o Figure 5-12. Steady-state force distribution over rocks along different structures including cross vanes and a W-weir in VSL-S. The dashed line is the key part of the A-shaped cross vane. Forces are non-dimensionalized using the drag force on a single rock due to the mean flow velocity, Fo. Distance, L, is non-dimensionalized by the total length of each structure, Ls. Figure 5-13. Simulation results of water surface elevation for three RVs in VSL-S. Flow direction is from top to bottom. The red zones show the backwater upstream of each rock vane structure. The maximum percent difference was calculated as the water surface with structures minus the baseline water surface normalized by the mean flow depth. Maximum differences were greatest for JHs (16%), then RVs (11%), and finally BWs (5%). While the influence of single-arm structures on water surface elevation is local, for channel-spanning sill structures such as CVs and WWs, the influence of struc- ture configuration on water surface elevation is expected to be significant and highly dependent on the sill elevation. For more information on the relationships between stage- discharge relationships and sill structure configuration, see Thornton et al. (2011).
Figure 5-14. Simulation results of water surface elevation for three bendway weirs in VSL-S. Flow direction is from top to bottom. The red zones show the backwater upstream of each bendway weir structure. Figure 5-15. Water surface elevation for (a) baseline and percent difference in water surface elevation (normalized by average depth), (b) three RVs, (c) three JH vanes, and (d) three bendway weirs in the OSL.
