Dr.
Shahane De Costa
Open University of Sri Lanka, 43/9, Issipathana Av., Anderson Rd, Dehiwela, Sri Lanka
E-mail: scosta@sti.lk, Tel / Fax - ++ 941 826100
S. A. Nandadeva
Lanka Mineral Sands Ltd, 167, Sri Vipulasena Mw, Col. 10, Sri Lanka.
Abstract: A wave tank and a wave generator was developed in stages in the laboratory to obtain different types of waves, wave breakers and currents. Tests on movement of sand were carried out using actual beach sand comprising heavy minerals of a wider range of grain sizes under pure waves, currents and a combination of waves and currents. It was found that the transportation and deposition of sand uniformly over a stretch of artificial beach is caused neither by pure waves nor by pure currents but by a combination of waves and currents.
Keywords: wave action, pure currents, deposition of minerals
There is a huge deposit of heavy mineral sands such as ilmenite, rutile, zircon, monazite & garnet on the beach at Pulmoddai in the North-East coast of Sri Lanka (Fig - 1). No such deposit is found elsewhere in the country. The beach sands are of different grain sizes ranging from 63 to 300 microns and the minerals are fully liberated into individual minerals. Each year, it is observed that large quantities of heavy minerals are brought from the sea by waves and deposited on the beach during monsoon season.
Observations of cyclic events during a period of one year shows that only during the monsoon season that heavy mineral bearing sands moves to the beach in large quantities. Even in the monsoon period, sand does not move to the beach during days in which waves are breaking in the sea away from the beach. In the off-monsoon season where waves are not violent, the sand that has already got deposited on the beach go through a process of separation and sorting, making the beach of very high grade heavy minerals exceeding a heavy minerals grade of 85%.
Note : Ilmenite and rutile are black in colour and during this period the beach becomes a black sand beach.
For this purpose a Wave Tank out of 06mm steel plates 3660mm (L) x 1830mm (W) x 760mm (D) in size was erected. This was fitted with a wave plate, driven by a fixed rpm electric motor, four interchangeable driving wheels arrangements to obtain four different rpm at wave plate, and a wave plate driving disc with five hole positions to obtain five different wave amplitudes was originally fabricated. However, with this tank waves could not be generated. Instead, water was splashing at the center of the tank and spilling over the tank forming wave resonance. Therefore the tank length was extended by another 1830mm., and tests were repeated. However it was observed that there was no improvement. Tank length was further extended by another 2440mm., with a slope in the form of the beach and tests were repeated. Waves were then formed. However, different shapes of waves could not be obtained at the selected four rpm positions and at the five wave amplitudes. The motor was then replaced with a variable speed electric motor. Wave tank sides (front and rear) were fitted with wave absorbers (sponge sheets and wire mesh), and the two longitudinal sides were cut and fitted with transparent Perspex sheets. Beach-end side was welded with a launder connecting it to the side of the tank with a pipe. (Fig. 02 & 03.) Waves of various shape were then formed by varying the motor rpm and wave plate amplitude.
(1) Keeping the water depth in the tank and wave plate amplitude fixed and varying the rpm.
(2) For the same water depth, the above test was repeated for five different amplitude positions (ie. Five different hole positions in the wave plate driving disc)
The results of the above tests indicated that the wave height increases as the rpm of the wave generator is increased and reaches a peak and then decreases even though the rpm increases. The same is true for different wave generator amplitude positions. However, the wave height increases with the increase of wave generator amplitude. Plunging waves were formed at higher amplitudes of the wave generator.
Bottom of the wave tank was covered with sand up to a thickness of 115mm - 150mm using beach sand and the beach side (inclined plate) was also covered with beach sand to form an artificial beach. The same experiments done above were repeated with sand bed. Sand motion was observed through the Perspex sheet windows.
At a given water depth and frequency and as the amplitude of the wave generator increased and reached a particular amplitude sand grains started to move to and fro,. When the test was continued at this initial motion setting, sand ripples were formed. The ripples were symmetrical and stable. The ripple height and the length were measured. The ripple height increased as the frequency (wave generator rpm) increased. The sand ripples formed were almost stationary and symmetrical. The shape and position of sand ripples were plotted on a tracing paper fixed on to the Perspex plate on the longitudinal side of the wave tank every hour. The ripples so formed did not change with the time provided that the settings (water depth, amplitude and frequency) were not altered. Ripple height and length, wave height, rpm, water depth etc. were recorded. After eight hours, sand was allowed to settle, and the tank was very slowly de-watered. The ripple pattern inside the tank was roughly plotted. Sand samples in three locations were taken and any change in the original position of the sand bed was studied. These tests with water waves, indicated that there was no noticeable shifting of the sand on to the beach.
As a current generation arrangement, two square openings were made on the two longitudinal sides of the wave tank 4570mm from the wave generating side and covered with boil boxes. 3"/2" slurry pump driven by a 15 HP, 1440 rpm, 3 phase Motor with the suction side connected to the bottom of one boil box and the delivery side connected to the top of the other boil box was installed. Water was delivered across the tank using four types of gates as follows without waves and water velocities across the tank were measured using a capillary tube.
(1) Using the tapered nozzle with a square/oval shape outlet - Large pits (cavities) were formed in the sand bed at the delivery and suction ends. No uniform flow (currents) was formed.
(2) Using the vertical flap arrangement - Slight improvement, but still pits were formed.
(3) Using square openings - Vast improvement in uniform flow across the tank and small pits.
(4) Honeycomb type openings - Flow of water across the tank was very uniform. Negligible cavities at the suction and delivery ends were formed. Therefore this gate was used for the creation of currents in the rest of the experiments.
Wave tank bottom was laid with a homogenous mixture of beach sand (known heavy mineral sand percentage and grain size) up to a thickness of 115-150 mm. An inclined beach was also made using same sand. Fresh water was slowly pumped into the tank through a hole in the bottom of the tank using a gate valve control arrangement without disturbing the sand bed. Tank was filled upto a predetermined height. The wave plate amplitude was set at the lowest (Hole No.01) and the wave generator was started. The generator rpm was increased very slowly from zero until the "initial motion" of sand grains on the top surface of the sand bed took place. The rpm was then further increased slightly and allowed to run continuously for eight hours. Ripple pattern on the sand bed was recorded every hour. Frequency, wave height, ripple length, ripple height, water depth etc. were recorded. This experiment was repeated the following day with the introduction of currents (starting the pump) and the motion of the sand grains on the sand bed was observed through the Perspex window. (Fig.4). The ripple patterns were also observed. A sample pattern is given in Fig.5.
A typical data set with some important features are given below.
Hole No. 5 : Wave generator stroke = 23.5 cm
Average water depth = 27.5 cm
Wave period Wave height Movement of Ripple Crest in
(Sec) (cm) intervals of 2 Hrs. in cm
0 - 2 Hrs 2-4 Hrs 4-6 Hrs.
----------------------------------------------------------------------------------------
2.31 15 * 3.0 2.2
2.00 10 * 5.0 1.0
1.50 15 * 3.6 3.5
1.43 29 * 6.5 0.5
1.30 18 * 3.0 3.5
----------------------------------------------------------------------------------------
l Fully developed ripples were formed with in the first two hours.
With pure water waves sand ripples are formed gradually and become stable in height and shape. The ripples that were stable and stationery under pure water waves move after the introduction of water currents across the tank at an angle (not perpendicular) to the direction of water waves. A close look at the sand ripples show that the shifting of sand ripples with time takes place as several steps of “vortices formation and disappearance” as shown in Fig.4. This “vortex formation and disappearance” starts only after the introduction of currents. In pure sinusoidal waves regular “vortex formation and disappearance” does not take place. Introduction of currents causes the wave to be asymmetrical which in turn causes the systematic repetition of “vortices formation and disappearance” there by moving the sand grains towards the beach.
(1) Symmetrical oscillatory water waves cannot transport sand. They can only move the sand grains to and fro, loosen the top layer of the sand bed, form ripples and bring the lighter and fine grained sand to suspension.
(2) Pure water currents cannot transport sand evenly over the entire length of a beach continuously and systematically. It can only transport the suspended material irregularly and indirectly.
(3) Neither pure water waves nor pure water currents alone can transport sand over a entire length of a beach evenly, continuously, regularly and systematically.
(4) Currents do not loosen the sand bed and keep the sand grains in suspension as in the case of water waves.
(5) Currents make the symmetrical waves and ripples asymmetrical and therefore, the ripples formed by water waves cannot get stabilized due to asymmetry.
(6) As a result of the asymmetry, the ripples undergo a continuous cycle process of formation, deformation and formation thereby moving the sand forward in the direction of waves.
(7) Bulk of the sand moves forward in the direction of water waves as a result of the asymmetry created by the action of currents on waves at the sand bed. Small quantities of light density and coarse grained sand and fine grained denser sand under suspension are moved by the currents.
(8) Concentration of heavy minerals in the sand mixture thus is transported on to the beach by the surging water moving up, perpendicular to the beach and travelling back to the sea carrying lighter sand particles with it leaving the heavy sand particles on the beach.
(9) The water waves on the surface should be strong enough to create ripples at the bed. If the water depth above the sand bed is large, water waves may not be strong enough to make an influence at the sand bed and hence no ripples are formed. In this situation water currents can not move sand from the bed. (This situation has been observed at the sea of Pulmoddai. The ships that are calling at Pulmoddai are anchored in the open sea at a distance of 1 1/2 Km from the beach. The items that have fallen from the ships accidentally while loading, have been recovered in the same location one year after by divers. It was interesting to note that even the items such as beer cans, bottles etc. had not moved away from the original place even after one year even though there existed strong waves).
(10) The water wave/current combination moves the top layer of the sand bed hence the thickness of the bed becomes smaller and smaller with the time. This means the depth of the water up to the bed gradually increases. Therefore, the water waves have also to be gradually increased in size to make the initial motion. If this does not happen, the transportation of sand does not take place after sometime as no ripples are formed. In nature, after several years seawater waves may not transport sand to the beach in the same quantities that are transported today. In fact we have already observed from the production records that the quantities that were deposited annually on the beach at Pulmoddai 20 years ago are not being deposited today even though there is a large quantity of heavy minerals in the sea.
References
[1] Bijker.E.W., Hijum.E and Vellinga.P., (1976) : Sand Transport by Waves, Journal of Coastal Eng., pp. 1149-1167.
[2] Birkemeier.W.A., (1984) : Time Scales of Near-shore Profile Changes, Journal of Coastal Eng., pp. 1507-1521.
[3] Dally.W.R. and Dean.R.G., (1984) : Suspended Sediment Transport and Beach Profile Evolution, Journal of Waterways, Port, Coastal and Ocean Eng., Vol.110, No.1 pp. 15 - 33.
[4] Davies. A.G. and Wilkinson.R.H., (1978) : Sediment Motion Caused by Surface Water Waves, Journal of Coastal Eng., pp. 1577-1595.
[5] Einstein.H.A., (1972) : Sediment Transport by Wave Action, Journal of Coastal Eng., pp. 933 - 952.
[6] Rance.P.J. and Warren.N.F., (1968) : The Threshold of Movement of Coarse Material in Oscillatory Flow, Journal of Coastal Eng., pp. 487-491.
Fig. 1 Location of Heavy mineral deposits
Fig. 2 Wave tank, wave generator and water pump arrangement
Fig. 3 Wave disc (driving wheel)
Fig. 4 Path of sand grains and shifting of ripples
Fig. 5 Change of position and ripple pattern with time