Emad Elawady1, Masanori Michiue2 and Osamu Hinokidani3
1Ph.D. student, Tottori University, Japan (E-mail: emad8@hotmail.com)
2Dr. Eng., Prof., Tottori University, Japan (E-mail: mmichiue@cv.tottor-u.ac.jp)
3Dr. Eng., Assoc. Prof., Tottori University, Japan (E-mail: hinokida@cv.tottori-u.ac.jp)
Address: 4-101 Minami, Koyama-cho, Tottori city 680-8552, Japan.
Tel.: +81-857-315284 Fax: +81-857-287899
Abstract: Analysis of scouring process around repelling (60°) submerged spur-dikes are presented based on experiments under movable bed scour. Study objectives included highlights on the movement of bed materials and local scour around spur-dikes at constant bed shear stress with uniform flow condition. The effects on scour depth of spur-dike length (b); spur-dike height (d) and flow depth (h) have been investigated with the help of experimental data. Final state of scour holes and deposit locations along the longitudinal direction are discussed herein. The longitudinal bed and water surface profiles are presented to find out the effect of bed degradation on the water surface level. Comparison between repelling and deflecting (90°) spur-dikes has also been discussed from the local scour point of view. It was found that the scour process around repelling and deflecting spur-dikes had little difference only in the shape of scour area, which depends on the inclination angle of the spur-dike in the flow direction. Both local scour and maximum scour depth observed at final state have found to be affected by the height (d) and the length (b) of the spur-dike. In case of small spur-dike length (b), local scour area was limited and no bank erosion can occur in downstream area. The data collected for both repelling and deflecting spur-dikes will be useful for the development of numerical models of scour around submerged spur-dikes.
Keywords: open channel flow, submerged spur-dikes, local scour and maximum scour depth
Knowledge of sediment transport in a river is
essential in all studies where morphological problems exist. It is therefore
necessary to study the characteristics of the sediments, as well as their modes
of transport and to link the sediment transport to the hydraulic parameters of
the river under consideration. The movement of sediment always causes scour of
the riverbed and banks. So that, training structures are required to resist the
current and to protect the channel against the changes. Spur-dikes are one of
the river training works constructed at an angle to the flow direction to push
the river towards a more suitable alignment, to protect sections of eroding bank
and they are especially useful for maintaining navigable depths in streams. The
types of spur-dikes vary depending on their action on the stream flow. They may
be classified as deflecting, repelling and attracting spur-dikes. Scour around
bridge piers, abutments and spur-dikes have been studied in detail by many
researchers in the last few decades. Contributions by Grade (1961)1)
and Gill (1972)2) are among the notable earlier studies. Recent
studies include Zaghloul (1983)3), Rajaratnam (1983)4),
Michiue (1984)5), Melville (1992)6), and Kuhnle (1999)7).
Most of the previous studies concerning local scour around non-submerged
spur-dike with angle equal 90° to the flow direction. The focus of
this study is on characterizing local scour and the maximum depth of scour
around repelling submerged spur-dike and compare it with the data collected from
the previous research8) of deflecting spur-dike under the same
conditions.
Fig. 1 Details of the experimental set-up
The purpose of this investigation was to spot on the movement of bed materials and scouring process around the spur-dikes under the following limitations:
· A single submerged spur-dike placed at the left sidewall of the flume with angle equal 60° to the flow direction was considered.
· The opening ratios (
) used were varied from 0.77 to 0.98
· The overtopping ratios (h/d) used were varied from 1.07 to 5.2
· A constant bed shear stress was considered in
all cases (
= 0.079).
· The slope of the bed surface was 1/2500, and Manning roughness was 0.014 in all cases.
(1) Experimental Set-Up
The experiments were conducted in a tilting steel flume 0.4m wide, 0.4m deep and 18.0m long. The central 8.0m of the flume is equipped with transparent walls. Flow rate in the flume was measured using a triangular weir that was calibrated prior to the experiment. One size of sand was used in the experiments reported herein with mean size 0.75mm. Acrylic sheets were used to model different spur-dikes, which locates in the middle of the sand bed area 9.0m downstream of the entrance with angle equal 60° to the flow direction. The thickness of the spur-dike was 1.5cm in all cases and the spur-dikes projected a perpendicular distance of 5.0,10.0 and 15.0cm into the channel. Bed surface profiles were measured using electric resistance bed profiler, and the measurements recorded with under flow condition. The measurement device was mounted on an instrument carriage that traveled on steel rods over the channel. The details of the experimental set-up and cross section at the nose of the spur-dike are shown in Fig. 1 and Fig. 2 respectively. The experimental conditions and the flow characteristics are summarized in Table 1 for the repelling spur-dike.
Fig. 3 Measurements locations
(2) Experimental procedures
Before the beginning of the experiments, the flume was examined to adjust the required slope that gives the uniform flow conditions far upstream. The suitable longitudinal surface slope was found to be 1/2500. A single spur-dike was fixed to the left sidewall of the flume in the appointed position. A 15.0 & 18.0cm thick bed material was laid into the flume and leveled to make the sand bed surface parallel to the channel bottom. Before each run the tailgate was adjusted to a suitable level to give a considered flow depth, which keep the shear velocity constant in all cases with different discharges. The runs started by slowly allowing the water to flow over the horizontal bed until it reaches to the height of the temporary gate. The discharge valve was slowly adjusted to give a supply of the flow required for each run. Then a temporary gate was opened slowly and completely, so that there was no effect in the flow and the bed surface level, and almost no scour happened before achieving the desired water depth; thereafter the timer was switched on.
The bed readings at the nose of the spur-dike were taken at various time intervals to get the rate of scour. The run was continued until the bed reading at the nose of the spur-dike changed so slowly with time nearly up to 300min. After recording the maximum scour depth, a temporary gate inserted again so that the scour pattern was not disturbed. Measurements of scour profiles were taken at close intervals around the spur-dike and at less-closer spacing far upstream and downstream of the flume as shown in Fig. 3. After completing one run, the flume then was prepared for another run with same procedures. 5 experimental cases with 23 runs were conducted having different initial arrangements.
Fig.
4 Scour pattern around spur-dikes with different
dike heights
(1) Local scour contours around spur-dikes
Typical scour contours around spur-dikes with different heights; lengths and different flow depths are shown in Fig.4 and Fig.5 respectively. Elevations are in cm, contour interval is 1.0cm, and elevation of initial bed surface was 0.0 and elevations less than –2.0cm are shaded.
In Fig.4, only the height of the spur-dike (d) above the original bed was variable
and all other parameters were constant (b, h and
). It is found that, at d = 2.5cm the
observed maximum scour depth was small as well the scouring area too, which
locates almost in upstream area within the spur-dike length. At d = 5.0cm the observed maximum scour was
two times the scour in previous case and the scouring area was also larger in
both upstream and downstream area. Also at d
= 7.5cm the scour area was larger not only upstream but also downstream close to
the flume sidewall, which leads to erosion can occur to the stream bank under
this conditions. It was further noticed that, the width of the scour hole (L) in
front of the spur-dike had consistent correlation with the height of the
spur-dike (d) as well the maximum
scour depth (S).
In Fig.5, in horizontal shapes the length of the
spur-dike (b) only was variable and in
vertical shapes the variable was the flow depth (h) only and the remaining parameters were constant. To study the
effect of the length of the spur-dike (b),
it is found that the maximum scour depth and the total scouring area have a
consistent correlation with the length of the spur-dike. The minimum scour depth
was observed in run 1-6, which has the largest value of the opening ratio (
).
Fig. 5 Scour pattern around spur-dikes
with different lengths and different flow depths
Looking vertically at Fig.5 (a, b and c) reaches to comparison between three cases with different flow depths (h). In all cases the scour depth slightly decrease with the increase of the flow depth. Also it is found that there is large scour area downstream in case 3 at the middle of the flume and reaches to both sidewalls, that downstream scour area was produced because of the action and the length of the repelling spur-dike and it did not observed in case of deflecting spur-dike8).
(2) Final state of scour and deposit
Fig.6 shows the final state of scour holes and deposit locations along the longitudinal direction at 1.0cm from the nose of the spur-dike in each run. In the figure, axis x is the longitudinal distance from the spur-dike towards downstream, and the spur-dike locates at X = 0.0cm.
Fig. 6 Longitudinal profiles of the scour holes
Upstream the spur-dike, scour holes were very steep within short distance. Scour holes started within 15.0 to 40.0cm upstream and the line of scour slope in almost all runs was approximately parallel with angle equal to the angle of repose of the sand used in the channel bed (30°).
Downstream the spur-dike, scour holes were elongated with shallower slopes in most of the runs, and with steep slope in run 5-1 only. Scour holes finished within 30.0 to70.0cm downstream in all runs except in run 3-4 there was no deposit area observed up to 100.0cm downstream. The maximum deposit recorded was about 2.0cm above original bed level and it locates at X = 80.0 to 100.0cm.
(3) Longitudinal bed and water surface profiles
Longitudinal profiles of bed and water surface at the nose of the spur-dikes for runs 1-2, 2-2 and 3-2 are shown in Fig.7 (a, b and c) respectively for 300 minutes. In these runs the uniform flow depth was 6.0cm and the spur-dike locates at X = 0.0cm along the x-axis with height (h = 2.5cm) above the original bed.
It can be seen from Fig.7-a that, the bed surface upstream rises and start to degraded from X = –14.0cm up to X = 40.0cm downstream and the water surface does not influence with the changes happened in the bed. In next run (Fig.7-b), there was no raise in bed surface upstream and the scour start at X = –24.0cm and continue up to X = 70.0cm downstream. That was because of the longer spur-dike, and there was a decrease in water surface observed in some areas at the location of the spur-dike and downstream as a result of decrease in bed surface. In last run (Fig.7-c), the scour area was very big and the changes in water surface was very limited, which means that there was no relation between the changes happens in both water and bed surfaces.
(a) Run 1-2
(b) Run 2-2
(c) Run 2-2
Fig. 7 Longitudinal profiles of bed and water surface
(4) Effect of opening ratio (
) and (b/h)
Fig.8 shows the relation between (S/d) with the opening ratio (
) for repelling and deflecting spur-dikes. It is clear that the results of both
cases are almost the same. For all runs of small spur-dike heights (d = 2.5cm) the relationship could be
present by straight line of an adverse slope. While for runs with higher
spur-dike (d = 5.0 and 7.5cm) the
changes in (S/d) ratios were very
small at different values of opening ratio. That means-opening ratio cannot be a
guide to predict S/d ratio, and there
are two different phases of scouring process regarding to the height of the
spur-dike and its ratios with other parameters.
Fig.9 shows the relation between S/d and b/h for all cases of this study. Except for few runs, it is found that the scour ratio increases with increase of the ratio (b/h) and the relationship can be represented by a set of positive curves. The variance observed in case 1 and case 3 could be occur because of the actual values of the effective bed shear stress near to the spur-dike.
Fig.
8 Variation of (S/d) with (a)
Fig.
9 Variation of (S/d) with (b/h)
In this paper, experimental studies were conducted to investigate the characteristics of scour around repelling (60°) submerged spur-dikes. This investigation has shown that, the scour process around repelling and deflecting spur-dikes had little difference only in the shape of scour area, which depends on the inclination angle of the spur-dike in the flow direction. Both local scour and maximum scour depth observed at final state have found to be affected by the height (d) and the length (b) of the spur-dike, so that these parameters should be considered on the design of the submerged spur-dikes. In case of small spur-dike length (b), local scour area was limited and no bank erosion can occur in downstream area. The data collected for both repelling and deflecting spur-dikes will be useful for the development of numerical models of scour around submerged spur-dikes.
References
[1] Grade, R. J., ASCE, M., Subramanya, K. and Nambudripad, K. D.: Study of scour around spur-dikes, Journal of the Hydraulics Division, ASCE, Vol. 87, No. HY6, pp. 23-37, 1961.
[2] Gill, M. A. and ASCE, M.: Erosion of sand beds around spur dikes, Journal of the Hydraulics Division, ASCE, Vol. 98, No. HY9, pp.1587-1602, 1972.
[3] Zaghloul, N. A.: Local scour around spur-dikes, Journal of Hydrology, 60, pp. 123-140, 1983.
[4] Rajaratnam, N. and Nwachukwu, B. A.: Erosion near groyne-like structures, Journal of Hydraulic Research, IAHR, Vol. 21, No. 4, pp. 277-287, 1983.
[5] Michiue, M., Suzuki, K. and Hinokidani, O.: Formation of low-water bed by spur-dikes in alluvial channels, Fourth Congress-Asian and Pacific Division, IAHR, pp. 685-698, 1984.
[6] Melville, B. W.: Local scour at bridge abutments, Jour. of Hydraulic Engrg., ASCE, Vol. 118, No. 4, pp. 615-631, 1992.
[7] Kuhnle, R. A., Alonso, C.V. and Shields, F. D.: Geometry of scour
holes associated with 90° spur
dikes, Journal of Hydraulic Engrg., ASCE, Vol. 125, No. 9, pp. 972-978, 1999.
[8] Elawady, E., Michiue, M. and Hinokidani, O.: Movable bed scour around submerged spur-dikes, Submitted to Annual Journal of Hydraulic Engrg., JSCE, Vol. 45, October 2, 2000.