SRAC Publication No. 374

Southern Regional Aquaculture Center

VI March 1995 PR

Open Channel Flow in Aquaculture

J. David Bankston, Jr.1 and Fred Eugene Baker2

Open channel flow of water has be designed and maintained to a radioactive material or fluores- been used in aquaculture produc- handle the needed flow. This fact cent dye. The dilution technique tion for many years. Distribution sheet will discuss the design of uses theoretical formulas to indi- canals, raceways and drainage open channels and measurement cate stream flow and does not ditches are some examples. Since of flow in open channels. alter the flow. A disadvantage of the beginning of civilization man this method is that it requires has been interested in flow in Methods of determining costly equipment that is not very open channels. Attempts to record flow in open channels rugged in field use. the levels on the Nile River date back to 3500 B.C. In 52 A.D. By definition, an open channel Velocity area Sextus Julius Frontinus as Water flow is flow in any channel in In this method, the flow rate is Commissioner of Rome attempted which the liquid flows with a free calculated measuring the cross- to determine the quantity of water surface. There are several ways to sectional area of the channel and delivered to each user by measur- determine flow in open channels. multiplying that area by the mean ing the cross-sectional area of each Some are discussed below. flow velocity across the area. discharge spout. The technology available today is much more Time gravimetric Hydraulic structure accurate but, for the most part, is With this method the entire In this method some type of an adaptation of these earlier con- stream flow is collected in some hydraulic structure is placed in cepts. type of container for a measured the stream. The purpose of the The flow of water in open chan- length of time. The flow rate is hydraulic structure is to produce a nels can be an effective and effi- calculated by dividing the volume relationship between the liquid cient way to move water for aqua- of water collected by the time to level (head) and the flow rate of culture. Often, the large flow rates collect it. This is the “bucket and the stream. These structures are needed in aquaculture require stop watch” technique. Most prac- normally either weirs or flumes as large, costly piping systems. Open tical considerations limit the use shown in Figure 1. Restrictions channels offer an alternative for of this method to low flow rates. caused by weirs or flumes will movement of large flows over In aquaculture uses, if the contain- alter the flow of the stream. long distance if the land slope is er is large enough, i.e., a pond of appropriate and water losses to known volume, the technique Slope-hydraulic radius may have practical application. evaporation and percolation are In this method the slope of the acceptable. Open channels should water surface, the cross-sectional Dilution area and the wetted perimeter 1 Louisiana Cooperative Extension Service With this procedure the flow rate over a length of uniform section and Sea Grant Program and is measured by determining how 2 of the channel are used to mea- Louisiana Cooperative Extension Service, much the flowing stream dilutes sure flow using a resistance equa- Louisiana State University Agricultural an added tracer solution, usually tion such as Manning’s Formula. Center. Upstream Water Surface Upstream Water Surface

B. FLUME

A. WEIR Downstream Water Surface

Downstream Water Surface

Figure 1. Hydraulic structures are either weirs or flumes.

The wetted perimeter is the length How to use the velocity-area six-tenths-depth method. The of wetted surface of the cross-sec- method two-point method is used where tional area and is an indicator of The velocity-area method is a pro- the water depth is more than 2 the efficiency of the channel shape cedure that is simple and practical feet. Here the velocity is mea- and therefore the resistance to for channels of the size normally sured at .2 and .8 of the depth and flow. The Manning Formula used in aquaculture. The basic averaged. Accuracy of this requires a knowledge of the cross- relationship is: method is high. The six-tenths- section of the channel, the slope of depth method measures velocity the channel, the water depth and Q = VA, Where at .6 the depth from the surface. a roughness factor dependent on The accuracy is not as good as the Q = Flow in cubic feet per minute the channel surface. two-point method, but adequate V = Velocity in feet per minute in shallow channels. With both 2 Practical flow measure- A = Cross-sectional area in feet methods a current meter is ment in aquaculture This is the basic equation of all required. Velocity measurements flow. Therefore, to evaluate a are recorded at equal intervals It should be understood that the channel to determine the amount across the channel and averaged. study of open channel flow is of water flowing we need to mea- The equation Q = AV is then complex. Flow is seldom uniform, sure the cross-sectional area (A) applied to determine the flow (Q). the roughness of the channel may and the average velocity (V) Accuracies of ± 20 percent should vary, turbulence is almost always across the cross-sectional area. result if care is taken with mea- present, and the cross-sectional surements. area may not be constant. Most Area (A) is determined by mea- suring the width (W) (Figure 2) Devices used to measure velocity flow measurements are estima- in open channels are known as tions at best, and empirical meth- and taking measurements of the depth at equal distances across W current meters. A number of dif- ods (methods developed by real ferent types of current meters are world experiences) are very com- and averaging them. Then W X D will give the cross-sectional area. available. Morris and Wiggert mon. (1972) reviewed these devices. Because of the length of this pub- Measuring the velocity is the diffi- The more common are Doppler lication and the practicality to cult part of this procedure since Ultrasonic meters, Turbine meters, aquaculture, more in-depth dis- the velocity varies widely across rotating element meters, electro- cussion will be presented only for the cross-section. There at least 8 magnetic probe meters and the velocity-area and the slope- procedures used to determine the Eddyshedding Vortex meters. hydraulic radius methods. mean velocity. The most common Improvements in electronics have are the two-point method and the enhanced the accuracy of these meters.

How to use the slope hydraulic radius method The Manning Equation in its vari- ous forms is the most used empir- ical method for estimating the flow in open channels. This equa- Figure 2. Typical cross-section of any open channel. tion can also be used to design

2 open channels. The Manning Equation can be written as fol- lows: 2 1 3 2 1.49 R x S x A Q = A X V = n Q = Flow in cubic feet per second R = Hydraulic Radius = A/P in ft. Figure 3. Measuring water flow in an open ditch. A = Cross-sectional area of the 2 channel in ft. As an example, suppose we want From the discussion and the P = Wetted perimeter of the to know how much water is flow- example it should be evident that channel in ft. ing in an open ditch with the fol- in the design of an open channel lowing trapezoidal cross-sectional that is to be nearly straight and S = Slope of the channel - shape (Figure 3). uniform, the Manning Equation dimensionless The ditch bottom is known to can be very useful. By considering n = The roughness coefficient - various cross-sectional shapes and dimensionless drop 1 foot in 400 feet of nearly straight uniform cross-section. depths of flow with known If Q is desired in gal/min, the The channel is vegetated and not slopes, needed flow rates for equation becomes well kept. aquacultural applications can be 2 1 3 R x S 2 x A n = (Table 1) is estimated as .030 Q = 669 elevation difference 1 n S = = = .0025 For the Manning formula to be section length 400 applied the channel should be nearly uniform in slope, cross-sec- From Table 3 tion and roughness, and free of A = (b + zh) h = [4 + 3 (1) ] 1 = 7 outside water sources, turns and (b + zh) h [4 + 3 (1)] 1 7 rapids. The straight section should R = = = = .68 be at least 200 feet long (longer if b + 2h 1 + z2 4 + 2 (1) 1 + (3)2 = 10.32 possible). 2 1 (R) 3 2 It can be seen that the average Q = 669 x S x A velocity and flow vary inversely n with the roughness of the channel 2 1 669 (.68) 3 x (.0025) 2 x 7 669 (.77) (.05) (7) (n). Therefore, the higher the n Q = = value the lower the flow, all other .030 .030 factors being equal. Table 1 gives n values for several channel types. Q = 6,010 gal/min. R is defined as the hydraulic radius, the cross-sectional area divided by the wetted perimeter. This seems to be a large amount planned. Remember, the calculat- This is dependent on the slope of water flowing. Is the velocity ed values are only as good as the and depth of water flowing. Table too high for the sandy loam soil? input values. Variances in the 2 provides geometric elements for Table 4 shows that the maximum shape, slope and straightness will channels of 4 cross-sectional allowable mean velocity to mini- cause the procedure to be less slopes. The most common shape mize erosion = 160 ft/min. The accurate and less useful. for earthen open channels is the average velocity in our example trapezoid. This shape allows for channel is: varying side slopes which can 3 3 minimize erosion. Table 3 shows Q = 6,010 gal/min = 803 ft /min (7.48 gal = 1 ft ) Q 803 permissible side slopes depending V = = = 115 ft/min on soil types. This information is A7 useful when designing an open channel. Therefore, the velocity will not The slope of the channel (S) is a cause excessive erosion. measure of the channel bottom elevation difference at the ends of the uniform section divided by the length of the section.

3 Table 1. Channel condition and values of the roughness coefficient n.

Channel Condition Value of n

Exceptionally smooth, straight surfaces; enameled or glazed coating; glass; lucite; brass 0.009

Good wood, metal, or concrete surfaces with some curvature, very small projections, slight moss or algae growth or gravel deposition; shot concrete surfaced with troweled mortar 0.014

Rough brick; medium quality cut stone surface; wood with algae or moss growth; rough concrete; riveted steel 0.015

Well-built earth channels covered with thick, uniform silt deposits; metal flumes with excessive curvature, large projections, accumulated debris 0.018

Smooth, well-packed earth; rough stone walls; channels excavated in solid, soft rock; little curving channels in solid loess, gravel or clay, with silt deposits, free from growth, in average condition; deteriorating, uneven metal flume with curvatures and debris; very large canals in good condition 0.020

Small, manmade earth channels in well-kept condition; straight natural streams with rather clean, uniform bottom without pools and flow barriers, cavings and scours of the banks 0.025

Ditches; below average manmade channels with scattered cobbles in bed 0.028

Well-maintained large floodway; unkept artificial channels with scours, slides, considerable aquatic growth; natural stream with good alignment and fairly constant cross-section 0.030

Permanent alluvial rivers with moderate changes in cross-section, average stage; slightly curing intermittent streams in very good condition 0.033

Small, deteriorated artificial channels, half choked with aquatic growth, winding river with clean bed, but with pools and shallows 0.035

Irregularly curving permanent alluvial stream with smooth bed; straight, natural channels with uneven bottom, sand bars, dunes, few rocks and underwater ditches; lower section of mountainous streams with well-developed channel with sediment deposits; intermittent streams in good condition; rather deteriorated artificial channels, with moss and reeds, rocks, scours and slides. 0.040

Artificial earth channels partially obstructed with debris, roots, and weeds; irregularly meandering rivers with partly grown-in or rocky bed; developed flood plains with high grass and bushes. 0.067

4 Table 2. Channel section geometry and associated parameters.

Table 3. Allowable side slopes (run over rise) for trapezoidal channels in various soils. Type of Soil Z/1 Light sand, wet clay 3:1 Wet sand 2.5:1 Loose earth, loose sandy loam 2:1 Ordinary earth, soft clay, sandy loam, gravelly loam or loam 1.5:1 Ordinary gravel 1.25:1 Stiff earth or clay, soft moorum 1:1 Tough hard pan, alluvial soil, firm gravel, hard compact earth, hard moorum 0.5:1 Soft rock 0.25:1

5 Table 4. Allowable mean velocities to protect against erosion or scour in channels for various soils and materials. Description V in m/sec V in ft/min Soft clay or very fine clay 0.2 40 Very fine or very light pure sand 0.3 60 Very light loose sand or silt 0.4 80 Coarse sand or light sandy soil 0.5 100 Average sandy soil and good loam 0.7 140 Sandy loam 0.8 160 Average loam or alluvial soil 0.9 180 Firm loam, clay loam 1.0 200 Firm gravel or clay 1.1 220 Stiff clay soil; ordinary gravel soil, or clay and gravel 1.4 280 Grass 1.2 240 Coarse gravel, cobbles, shale 1.8 360 Conglomerates, cemented gravel, soft slate, tough hardpan, soft sedimentary rock 1.8-2.5 360-500 Soft rock 1.4-2.5 280-500 Hard rock 3.0-4.6 600-920 Very hard rock or cement concreted (1:2:4 minimum) 4.6-7.6 920-1,520

References Rouvari, J. Hydraulic Formulas Morris, H. M. and Wiggert, J. M. Used in Designing Fish Farms, Applied Hydraulics in Chow, V. T. Open-Channel Inland Aquaculture Engineering. New York, John Hydraulics. New York, Engineering, Aquaculture Wiley and Sons, 1972, 629 pp. McGraw-Hill Book Co., Inc., Development and 1959. 680 pp. Coordination Programme, Grant, D. M. ISCO Open Channel United Nations Development Flow Measurement Handbook Programme, Food and Agri- 3rd Edition, 1992, ISCO culture Organization of the Environmental Division, United Nations, Rome, 1984. Lincoln, NE.

The work in this publication was supported in part by the Southern Regional Aquaculture Center under USDA/CSRS Grant No. 92-38500-7100.

6