Analysis of flow around impermeable groynes on one side of symmetrical compound channel: An experimental study

Hassan Safi AHMED, Mohammad Mahdi HASAN, Norio TANAKA*

1. Civil Engineering Department, Faculty of Engineering, South Valley University, Qena 83521, Egypt

2. Graduate School of Science and Engineering, Saitama University, 255 Shimo-Okubo, Sakura-ku, Saitama 338-8570, Japan

3. Institute for Environmental Science and Technology, Saitama University, 255 Shimo-Okubo, Sakura-ku, Saitama 338-8570, Japan

Analysis of flow around impermeable groynes on one side of symmetrical compound channel: An experimental study

Hassan Safi AHMED1, Mohammad Mahdi HASAN2, Norio TANAKA*3

1. Civil Engineering Department, Faculty of Engineering, South Valley University, Qena 83521, Egypt

2. Graduate School of Science and Engineering, Saitama University, 255 Shimo-Okubo, Sakura-ku, Saitama 338-8570, Japan

3. Institute for Environmental Science and Technology, Saitama University, 255 Shimo-Okubo, Sakura-ku, Saitama 338-8570, Japan

This paper presents the results of an experimental study on the influences of floodplain impermeable groynes on flow structure, velocity, and water depth around the groyne(s). A wooden symmetrical compound channel was used. Groyne models with three different groyne relative lengths, 0.5, 0.75, and 1.0, were used on one floodplain with single and series arrangements. Analysis of the experimental results using the measured flow velocity and water depth values showed that flow structure, velocity, and water depth mainly depend on groyne relative length and the relative distance between series groynes. The flow velocity at the main channel centerline increased by about 40%, 60%, and 85%, and in other parts on the horizontal plane at the floodplain mid-water depth by about 75%, 125%, and 175% of its original value in cases of one-side floodplain groyne(s) with relative lengths of 0.5, 0.75, and 1.0, respectively. The effective distance between two groynes in series arrangement ranges from 3 to 4 times the groyne length. Using an impermeable groyne with a large relative length in river floodplains increases the generation of eddy and roller zones downstream of the groyne, leading to more scouring and deposition. To avoid that, the groyne relative length must be kept below half the floodplain width.

impermeable groyne; symmetrical compound channel; floodplain; flow structure

1 Introduction

Groynes are hydraulic structures that protect against bank erosion, maintain water level by deflecting flow direction, and ensure navigation safety. They can be defined as shore protection structures (usually perpendicular to the shoreline) built to trap littoral drift or retard erosion of the shore, as structures installed on the front side of the bank, or as revetment to protect the bank or the levee against erosion (Uijttewaal 2005; Yeo et al. 2005; Yeo and Kang 2008; Kadota et al. 2008; Gu and Ikeda 2008; Teraguchi et al. 2008). They can be used for flood control, land reclamation, and provision of navigable depth (Osman et al. 2001; Rajaratnam and Nwachukwu 1983; USACE 1992). In addition, their functions can be changed along with the goals of river works and the nature of the stream. Therefore, some of thegroynes’ other main functions are increasing the hydraulic conveyance of rivers, reducing risks during flooding, securing the regular depth of water and stabilizing river flow, and improving the ecological environment and scenery. Permeable stone gabion groynes have recently been proposed as countermeasures for scouring problems and also to help develop suitable ecosystems because they maintain stable flow and bed conditions near the groyne tip (Muraoka et al. 2008).

Groynes can be classified according to their functions, objects, forms, and materials. There are many types of groynes, such as the T-groyne, permeable groyne, and impermeable groyne. The impermeable groynes are generally constructed using local rocks, gravel, or gabions, whereas the permeable ones consist of rows of piles, bamboo, or timber (Teraguchi et al. 2008; Ettema and Muste 2004).

River floodplain groynes play an important role in high flood attenuation and protection, especially for rivers with large floodplains. Submerged and non-submerged impermeable groynes, transverse levees, and bridge embankments restrict flow within the channel width. Many variables are involved in studying the flow in compound channels flanked by one or two floodplains with groynes. These variables include flow approach velocityU0, water depth in the main channelHand in the floodplainh, groyne type (permeable or impermeable), shape, angle of inclinationθ, the groyne length relative to the floodplain widthLr(Lr=Lg/b, whereLgis the groyne length andbis the floodplain width), and the arrangement and the distance between groynes in series arrangement. These affect the flow structure and the features of the eddy zone upstream and/or downstream of the floodplain groyne (Ettema and Muste 2004; Francis et al. 1968). In addition, the main channel and floodplain widths become effective variables for flow in compound channels with a narrow main channel and relatively wide floodplains. The most important variables of the floodplain impermeable groyne are the groyne(s) relative lengthLr, the distance between groynes in series arrangement, and the arrangement type.

The large-scalegroyne system has been introduced in some rivers with large floodplains, suchas the Arakawa River, Japan, in order to attenuate the flood peak for higher safety of downstream areas, but extensive and quantitative analyses have not yet been performed. Therefore, the main purposes of the present research are (1) to study and verify experimentally the effects of the large-scale floodplain impermeable groynes on the flow structure, velocity, and water depth; and (2) to evaluate the advantages and disadvantages of using floodplain groynes in flood attenuation and protection works.

2 Experimental apparatus and case studies

The experiments were conducted in a water re-circulating flume at Saitama University in Japan. The flume is 0.50 m in depth, 0.50 m in width, and 15.0 m in length. The workingsection of the flume is the middle one with a length of 13.0 m, starting from a point 1.0 m downstream of the inlet to a point 1.0 m upstream of the outlet, as shown in Fig. 1(a). At the downstream end of the flume, the tail-water depth was controlled by a vertical sluice tailgate.

As shown in Fig. 1(b), the rectangular flume section was converted into a wooden symmetrical compound channel section with a main channel widthBof 0.1 m and a total water depthHof 0.24 m, and two symmetrical floodplains with a widthbof 0.2 m (the floodplain relative widthb/B= 2.0). The roughness coefficients were kept constant and equal in both the main channel and floodplains. The flow was steady flow with a dischargeQof 0.015 m3/s, a floodplain water depthhof 0.08 m, and a Froude number of 0.26, while the Reynolds number was always sufficiently high to guarantee fully turbulent flow. The discharge was measured with an electromagnetic flow meter (model: FD-UH100H, Keyence Corporation; maximum measurable discharge: 0.033 m3/s; accuracy: –0.02% to 0.25%). As shown in Fig. 1(b), the longitudinal and transverse flow velocitiesU(m/s) andV(m/s) were measured at nine points on the horizontal plane HP with an interval of 0.05 m and at five points on the vertical plane VP (points marked with black circles), respectively. The velocities were measured with an electromagnetic velocity meter (type of main amplifier: VM-2000; type of sensor: VMT2-200-04P, Kenek Company, Ltd.). The sensor was 15.0 mm in length and 4.0 mm in diameter, and the measurement point was located at the middle height of the sensor, with 20.0 s measuring time and 50 Hz as sampling frequencies. At each point, the mean velocity value in the longitudinal and transverse directions,UandV, respectively, were obtained by averaging the measured velocity values, then the mean resultant velocity value ( (U2+V2)0.5) and directions were obtained. The longitudinal velocityUprofiles on both the horizontal and vertical planes, HP and VP, respectively, were measured at several locations upstream and downstream of the main groyne Gr1. The water surface elevation was measured at both the main channel and floodplains centerlines with and without the groyne models at the same locations as the velocity profiles by means of three point gauges (with an accuracy up to 0.1 mm), fixed and mounted on a movable sliding carriage.

The experiments were conducted using models of straight impermeable groynes fixed perpendicularly to the main channel centerline and to the longitudinal flow direction. The groynes were made of wood plates with a thickness of 0.01 m with three different relative lengthsLr, 0.5, 0.75, and 1.0. The arrangements of groynes in case studies are shown in Table 1. In the case of a series of groynes in one floodplain, the downstream groynes were fixed at a distance of 8.5 m upstream of the flume outlet to decrease the backwater effects.

Fig. 1 Experimental setup

Table 1 Arrangement of groynes and experimental conditions

3 Results and discussion

In this section, the results and analysis of the experimental program conducted to better understand the effects of river floodplain groynes on compound channel water flow structure, velocity, and water depth are presented. Symbols in figures of velocity profiles are defined as follows: Gr1 is the downstream groyne (the main groyne) and Gr2 is the upstream groyne in series arrangement; D25 means that the profile is located 25 cm downstream of the main groyne Gr1; U20 means that the profile location is 20 cm upstream the groyne (or the groyne group); while in the series arrangement, ND means 5 cm upstream of Gr1; NM is located at the mid-distance between Gr1 and Gr2 (atx=ds/2 upstream Gr1); and NU means 5 cm downstream of Gr2. The downstream groyne (Gr1) was considered the zero distance (x= 0), and the distance downstream of Gr1 was considered positive while the distance upstream of Gr1 was negative as shown in Fig. 1.

3.1 Single groyne

In this part of the study, a single groyne on one floodplain was used. As shown in Figs. 2 and 3, the value ofLraffected the flow, especially downstream of the groyne, while a small distance was affected upstream. Downstream of the groyne, a re-circulating flow region was generated. AsLrincreased, the center of the eddy zone moved toward the groyne. As shown in Figs. 2 and 4 (a), the main channel maximum longitudinal velocityUmaxwas in the lower zone of the vertical plane VP, while the minimum velocity shifted upward to the surface. The values of the maximum relative velocity (U/U0)maxwere 1.40, 1.60, and 1.85 with changes to theLrof 0.5, 0.75, and 1.0, respectively. As shown in Figs. 4(b) and (c), the flow moved toward the main channel and the opposite floodplain, and the flow longitudinal velocityUon the opposite floodplain increased by 75%, 125%, and 175% whenLrvalues were 0.5, 0.75, and 1.0, respectively. The value of the negative velocity at the downstream side of the groyne was more than –55% of the approach velocity at the same streamline whenLr= 1.0.

Fig. 2Uprofiles on VP for single groyne on one floodplain

Fig. 3 Velocity contour maps ofUon HP for single groyne on one floodplain

Fig. 4 Values of (U/U0)maxand (U/U0)minon HP and VP for single groyne on one floodplain

Fig. 5 shows the changes of flowing water depth on both floodplains upstream and downstream of the groyne (change = (depth without groyne – depth with groyne)/depth without groyne) × 100%). The water depth on the left floodplain, where the groyne is located, was more affected than that on the right side. On the groyne side, the percentages of increasing upstream depth were 0.8%, 3.6%, and 6.3%, and the decreasing percentages downstream were 2.5%, 5.5%, and 6.7%, whenLrvalues were 0.5, 0.75, and 1.0, respectively. Fig. 6 shows the relationship between the lengthLsand widthWsof the flow separation zone normalized by the groyne length for the same three cases. AsLrincreased, the relative lengthLsr(whereLsr=Ls/Lg) of the separation zone clearly decreased, while its relative widthWsr(whereWsr=Ws/Lg) slightly decreased. The average value of the relative widthWsrfor the three cases can be taken to be 1.65. Also, the eddy size increased withLr.

Fig. 5 Changes of water depth on left (LFP) and right (RFP) floodplain in case of single groyne on one floodplain

Fig. 6 Relationship betweenWsrandLsrof flow separation zone in case of single groyne on one floodplain

3.2 Two symmetrical groynes on one side

In this part of the study, two identical groynes with relative lengthsLrof 0.75 and 1.0 were located on one floodplain and arranged in two lines with various relative distancesdbetween them (as illustrated in Table 1). As shown in Figs. 7 through 14, bothLranddaffected the flow structure and the cross-sectional active area, where the flow patterns moved toward the main channel and the opposite floodplain. As shown in Figs. 8(a), 9, 11(b), and 12(c), the values of (U/U0)maxon the VP decreased asdincreased untild= 4, while (U/U0)maxincreased withdwhendwas larger than 4. Figs. 8(b), 8(c), 12(a) and 12(b) show the values of (U/U0)maxand (U/U0)minoccurring on the HP along the flume, respectively. The longitudinal flow velocityUin the main channel and floodplains, upstream and downstream of the groynes, depends on bothLrandd. For groynes withLr= 0.75, the effective distancedsis equal to 3 to 4 times the groyne length. In addition, as shown in Figs. 10 and 13, bothWsrandLsrincreased withLr, andWsrvalues in the area between groynes decreased asdincreased. The separation zone of the upstream groynes group is not affected byd. Each of the two groynes in the group may stand alone as a single groyne ifdincreases significantly. Moreover, ifdis less than or equal to 2, the groyne group works as one groyne (one block) where no single separation occurs between the two groynes (Fig. 7). From the results shown in Fig. 14, the water surface level of the upstream groyne group depends more onLrand less ond. The location of the lowest point of the water surface in the distance between the two groynes moves toward the upstream one asdincreases. Downstream of the groyne group, the water surface increases again and resumes its normal value at a distance 10 to 12 timesLg. This increase starts right away from upstream of the main groyne Gr1.

Fig. 7 Velocity contour maps ofUon HP for two groynes on one floodplain whenLr= 0.75

Fig. 8 Values of (U/U0)maxand (U/U0)minon HP and VP for two groynes on one floodplain whenLr= 0.75

Fig. 9Uprofiles on VP in case of two groynes on one floodplain whenLr= 0.75

Fig. 10 Relationship betweenWsrandLsrof flow separation zone in case of two groynes on one floodplain whenLr= 0.75

Fig .11Uprofiles in case of two groynes on one floodplain whenLr= 1.0

Fig. 12 Values of (U/U0)maxand (U/U0)minon HP and VP whenLr= 1.0

Fig. 13 Relationship betweenWsrandLsrof flow separation zone in case of two groynes on one floodplain whenLr=1.0

Fig. 14 Changes of water depth on LFP and RFP in case of two groynes on one side of floodplain

4 Conclusions

The findings from the analysis of the experimental results of this study may have apractical value, especially regarding the safety of the floodplain bed and bank stability. The following main conclusions can be drawn:

(1) Flow structure, velocity, and water depth mainly depend on the groyne type, relative lengthLr, and relative distance between two groynes in series arrangements.

(2) Using an impermeable groyne with a large relative length on a river floodplain generates flow eddies and separation zones downstream of the groyne and in the upper region of the main channel. This may lead to floodplain erosion.

(3) The velocity in the main channel upper region decreases, while, in the middle and lower regions of the main channel, it increases. The velocity on the other floodplain also increases.

(4) In cases of a single groyne whenLr= 0.5, 0.75, and 1.0, the negative velocities reached –20%, –30%, and –55% of the original velocity, respectively. Those negative velocities were substituted by increasing the flow velocity in the main channel and on the opposite floodplain. The increase could reach 1.4, 1.6, and 1.85 times the original velocity in the main channel, and 1.75, 2.25, 2.75 times the original ones on the other floodplain.

(5) The effective distance between two symmetrical groynes on one side of the floodplain is from 3 to 4 times the groyne length.

(6) Finally, using impermeable groynes with a large relative length in a river floodplain increases the risks during large flood events. River levee and embankment failures can occur if the protection works against the scouring process are weak, and the river can easily change its course and centerline. To mitigate those effects, the groyne length should be less than half the floodplain width.

Acknowledgements

The authors acknowledge Dr. Junji Yagisawa for his kindly help during the preparation of the experiments. The first author acknowledges his appreciation of the Egyptian Government for its financial support.

Ettema, R. and Muste, M. 2004. Scale effects in flume experiments on flow around a spur dike in flatbed channel.Journal of Hydraulic Engineering, 130(7), 635-646. [doi:10.1061/(ASCE)0733-9429(2004) 130:7(635)]

Francis, J. R. D., Pattanick, A. B., and Wearne, S. H. 1968. Observations of flow patterns around some simplified groyne structures in channels.Proceedings of the Institution of Civil Engineers, 829-846. London. [doi:10.1680/iicep.1968.7821]

Gu, Z. and Ikeda, S. 2008. Experimental study of open channel flow with groins.Proceedings of 16th IAHR-APD Congress and 3rd Symposium of IAHR-ISHS, 1951-1956. Beijing: Tsinghua University Press.

Kadota, A., Kojima, E., Shinya, K., and Suzuki, K. 2008. Instantaneous-advective structures of large scale coherent vortices around a single groyne.Proceedings of 16th IAHR-APD Congress and 3rd Symposium of IAHR-ISHS, 1137-1142. Beijing: Tsinghua University Press.

Muraoka, H., Fushimi, T., Kadota, A., and Suzuki, K. 2008. Experimental study on changes of bed configuration caused by a permeable groyne of stone gabion.Proceedings of 16th IAHR-APD Congressand 3rd Symposium of IAHR-ISHS, 1072-1077. Beijing: Tsinghua University Press.

Osman, M. A, Salih, A. M., and Ebrahim, A. A. 2001. Flow pattern around groynes.Sudan Engineering Society Journal, 47(39), 29-36.

Rajaratnam, N., and Nwachukwu, B. A. 1983. Flow near groin-like structures.Journal of Hydraulic Engineering, 109(3), 463-480. [doi:10.1061/(ASCE)0733-9429(1983)109:3(463)]

Teraguchi, H., Nakagawa, H., Muto, Y., Baba, Y., and Zhang, H. 2008. Effects of groins on the flow and bed deformation in non-submerged conditions.Annuals of Disaster Prevention Research Instituteof Kyoto University, 51B, 625-631.

Uijttewaal, W. S. J. 2005. Effects of groyne layout on the flow in groyne fields: Laboratory experiments.Journal of Hydraulic Engineering, 131(9), 782-791. [doi:10.1061/(ASCE)0733-9429(2005)131:9(782)]

U. S. Army Corps of Engineers (USACE). 1992.Coastal Groins and Nearshore Breakwaters. Washington, D. C.: U. S. Army Corps of Engineers.

Yeo, H. K., Kang, J. G., and Kim, S. J. 2005. An Experimental study on tip velocity and downstream recirculation zone of single groynes of permeability change.KSCE Journal of Civil Engineering, 9(1), 29-38. [doi:10.1007/BF02829094]

Yeo, H. K., and Kang, J. G. 2008. Flow analysis around a submerged groyne.Proceedings of 16th IAHR-APD Congress and 3rd Symposium of IAHR-ISHS,1762-1766. Beijing: Tsinghua University Press.

*Corresponding author (e-mail:tanaka01@mail.saitama-u.ac.jp)

Received Jul. 15, 2009; accepted Dec. 10, 2009