Read naja_67_306.193_201.tp text version

North American Journal of Aquaculture 67:193­201, 2005 Copyright by the American Fisheries Society 2005 DOI: 10.1577/A04-058.1

[Article]

Mixed-Cell Raceway: Engineering Design Criteria, Construction, and Hydraulic Characterization

JAMES M. EBELING*

The Conservation Fund, Freshwater Institute, 1098 Turner Road, Shepherdstown, West Virginia 25443, USA

MICHAEL B. TIMMONS

Department of Biological and Environmental Engineering, College of Agriculture and Life Sciences, Cornell University, 302 Riley-Robb Hall, Ithaca, New York 14853, USA

JEREMY A. JOINER

The Conservation Fund Freshwater Institute, 1098 Turner Road, Shepherdstown, West Virginia 25443, USA

RODRIGO A. LABATUT

Department of Biological and Environmental Engineering, College of Agriculture and Life Sciences, Cornell University, 302 Riley-Robb Hall, Ithaca, New York 14853, USA Abstract.--A prototype raceway was constructed in a research greenhouse at the Conservation Fund's Freshwater Institute to study engineering design criteria and the hydraulics of a large mixed-cell raceway. The raceway measured 16.3 5.44 1.22 m and was constructed of structural lumber with a high-density polyethylene liner. The basic rationale of a mixed-cell raceway is to be able to operate it as a series of square or octagonal tanks, each having a center drain for continuous removal of solids and sludge. A series of vertical manifolds along the sidewalls directed the water through orifice discharges parallel and perpendicular to the walls to establish the desired rotary circulation. This was then combined with the concept of the ``Cornell-type'' dual-drain system, where 10­20% of the total flow into a tank is removed from a center bottom drain and 80­90% of the flow is removed from the side upper drain. Settlable wastes and sludge are then removed from the center drains and collected in a settling sump. Velocities were measured in three dimensions in representative cells at three depths (5­10 cm off the bottom, middepth, and 10 cm below the water surface) on a 0.5-m grid pattern over the raceway floor. The raceway mean rotational velocity was 5­6% of the inlet jet velocities. The results from this study can be used to further refine the engineering design criteria and demonstrate the potential for employing mixed cells as a new design concept both for production systems and for retrofits of existing raceway systems.

For years, raceways have been used for the production of salmonids and other species by federal and state agencies for stocking purposes as well as by commercial growers. Where there are large groundwater resources, such as the Thousand Springs area in Idaho, raceways are the most common rearing tank design being used to grow the majority of rainbow trout Oncorhynchus mykiss produced in the USA. One significant advantage of raceways is their better utilization of floor space and easier handling and sorting of fish as compared with traditional circular tanks. Primary disadvantages are the large volume of water required, high

* Corresponding author: [email protected]

Received January 31, 2005; accepted February 10, 2005 Published online June 15, 2005

turnover rates, and limited self-cleaning ability (Timmons et al. 1998). Raceways depend upon a high turnover rate to maintain acceptable water quality. Water enters the raceway at one end and flows through the raceway in a plug­flow manner. As a result, the best water quality is found at the head of the raceway, where the water first enters, and deteriorates steadily toward the raceway outlet. Because of low velocities through the raceway (2.0­4.0 cm/s), removal efficiency of settled solids is very poor, requiring frequent cleaning and maintenance (Timmons et al. 1998). Low velocities are a result of the hydraulic design being based on oxygen design requirements rather than on cleaning requirements. In practical terms, raceways are incapable of producing the optimum water velocities recommended for fish health, muscle tone,

193

194

EBELING ET AL.

FIGURE 1.--Mixed-cell raceway layout and flow pattern. See text for details.

and respiration (Timmons et al. 1998). Through the use of lower exchange rates and lower velocities, raceways are severely limited because of the unavailability of large quantities of high-quality water, increased concern about their environmental impacts on receiving waters, and difficulties presented when treating large flows from effluent discharge. One solution to this problem is to convert raceways into a series of counter-rotating mixed cells (Watten et al. 2000). The concept of a mixedcell raceway was first proposed by Watten et al. (2000) to eliminate metabolite concentration gradients, increase current velocities, and improve solids scour at low water exchange rates. Watten et al. (2000) modified a standard raceway section 2.4 m wide by creating a series of 14.5 m long 2.4 m. A six counter-rotating cells, each 2.4 m series of vertical pipe sections with jet ports were installed in the corners of each cell, and water was directed parallel to the sidewalls to create rotary circulation. Water was withdrawn from centrally located floor drains. Mixed-cell raceways can be managed as either partial-reuse systems (Summerfelt et al. 2004) or as intensive recirculation systems, allowing for a substantial increase in production. Design Concept The mixed-cell raceway acts as a series of hydraulically separated round tanks. The basic design concept of the mixed-cell raceway (Watten et al. 2000) is to operate it as a series of adjacent counter-rotating square or octagonal tanks, each having a center drain for the continuous removal of solids and sludge (Figure 1). Early research on mixedcell raceways by Watten et al. (2000) examined

their use in existing raceway retrofits at federal and state hatcheries that was reflected in the overall small systems size of 22.7 m3. In contrast, this work started with the design of a small, 108-m3 commercial production system. In addition, Timmons et al. (1998) recommended tank diameterto-depth ratios from 5:1 to 10:1 for good selfcleaning capability compared with 3:7 for Watten et al. 2000 versus 5:5 in this study. A prototype raceway was constructed in one bay of an existing greenhouse at the Conservation Fund's Freshwater Institute. Approximate dimensions of the raceway were 16.3 5.44 1.22 m, which created three mixed cells. Each cell received water from four vertical manifolds (down legs) that extended to the raceway floor and were located in the corners of each cell and at the intersection between adjacent cells (Figure 2); four of the manifolds supplied water to two cells concurrently. Water was pumped through several orifice discharges (jet ports) from each of the down leg pipes to establish rotary circulation in the cell; adjacent cells rotated in opposite directions. Each cell had a bottom drain located at the center that was connected to a drain line, which discharged solids and sludge to a settling sump. A small fraction of the total recirculated flow (e.g., 10­20%) was withdrawn from this sump and returned to the raceway, creating a ``Cornell-type'' dual-drain system (Timmons et al. 1998). Engineering design for an aquaculture system often starts with the water exchange rate required for the production tank, usually defined by the dissolved oxygen requirements based on loading densities or some other limiting parameter. For this design analysis, a water exchange rate of 1.5 ex-

MIXED-CELL RACEWAY DESIGN

195

vative approach assumed a proportionality constant of 10% for this initial design; a desired cell rotational velocity of 20 cm/s was selected for the cleaning velocity. Using the previously assumed values, vo was estimated to be about 200 cm/s. The relationship between orifice velocity and hydraulic head pressure is shown in the following equation (Timmons et al. 1998): vo C(2gh)½, where vo is the discharge velocity from the orifice, C is an empirically determined constant for a square-edged orifice (0.6), g is 9.81 m/s2, and h is the head in meters. Using the equation, the required pressure head to achieve the required discharge velocity was approximately 0.58 m gauge. It needs to be recognized that each of the vertical manifold down legs in the corners provided orifice jets to only a single cell, while the four center down legs supplied orifice jets to two adjacent cells (Figure 1). We accomplished this by either directing two jets in opposite directions parallel to the sidewalls or by directing a single set of jets across the raceways and perpendicular to the raceway wall at a point that separates the two mixed cells. As a result, the center manifold down legs were required to discharge twice the flow of the corner manifold down legs. In one case, this was accomplished with twice the number of orifices as the corner manifold orifices; in the second case, this was accomplished by increasing the orifice diameter to provide twice the flow as the corner manifold orifices. The center down legs that discharged in two directions parallel to the sidewalls had 14 orifices (7 for each direction). The corner manifolds and the remaining two down legs that discharged perpendicular to the raceway walls had 7 orifices. For trial 1, a discharge of approximately 34.7 and 68 L/min for the two sets of manifolds, respectively, were required to achieve 1.5 tank water exchanges per hour, or a flow rate of 172 m3/s. Using the orifice equation and the continuity equation (flow velocity the cross-sectional area of the orifice) and solving for the diameter of the orifices yielded an orifice size of 20 mm for the four corner manifolds and for the two manifolds that discharged parallel to the sidewalls in opposite directions. The two manifolds that discharged perpendicular to the cell walls required a 28-mm orifice when the intention was to provide double the discharge rate from the orifices. In trial 2, it was decided to keep all orifices the same size (15 mm). Every attempt

FIGURE 2.--Vertical manifold pipe section (down leg) that extends to the mixed-cell raceway floor and is located in the corners of each cell and between cells.

changes/h was chosen for all trials, which would correspond to a moderate fish biomass density (e.g., 30­50 kg/m3). As constructed, the mixedcell raceway had a volume of 108 m3 and a water depth of 1.22 m. At 1.5 water exchanges/h, the required total flow rate was 172 m3/s. This included a withdrawal rate from the center drains of 25 m3/s or 15% of the total flow (i.e., a ``Cornelltype'' dual-drain system). The rotational velocity in the cells can be controlled by the design of the orifice discharges, either by increasing the flow rate, the discharge velocity, or the total number of orifices. Research has shown that water rotational velocity in round tanks is fractionally proportional to the discharge velocity from the orifices in the water inlet structure (Paul et al. 1991). Timmons et al. (1998) reported that this proportionality constant between tank rotational velocities is generally 15­20% of the inlet velocity, although this was applied to round tanks with smaller diameters. Watten et al. (2000) reported that the grand mean of current measurements corresponded to 3.7% of the inlet jet velocities and was calculated based on orifice diameter and head. This was substantially smaller than the value reported for round tanks and was attributed to the increased drag associated with forced distortion of circulating cells within the raceway's rectangular boundaries (Watten et al. 2000). Water velocities greater than 15­30 cm/s are required to drive settable solids to the center of a tank or cell (Timmons et al. 1998). Our conser-

196

EBELING ET AL.

FIGURE 3.--Cross section of the mixed-cell raceway showing construction details, pipe manifolds, vertical manifolds, and drain details.

was made to ensure uniformity among the orifice jets, but some difference in discharge velocities among orifices, even with the same hydraulic pressure, could be expected as a result of small burrs or imperfections in the orifice jets. Methods Construction details.--The mixed-cell raceway was constructed as an aboveground tank with a width of one greenhouse bay. This resulted in a 5.44 1.22 raceway with dimensions of 16.3 2.44 m) were prem. Sidewall modules (1.22 6-in construction studs (1 in fabricated of 2-in 2.54 cm) spaced 0.61 m apart and covered with 12-mm plywood sheeting. These sidewall modules were supported on a 6-in 6-in treated wood beam ``foundation'' and connected together with 12-mm lag bolts. In addition, a top plate was added to the sidewall modules to provide additional rigidity (see Figure 3). Normally, such a raceway would be constructed below grade, allowing backfill material to provide structural support to the walls. To provide this structural support, a series of polypropylene-impregnated wire ropes were run across the top and below the raceway at five equally spaced intervals along the sidewalls. In addition, a single cable was strung the length of the raceway on the top at the center and two cables were strung lengthwise below the insulated floor. These cables were secured into the sidewall top plates and into the 6-in 6-in treated wood beam foundation with 18-mm-diameter eyebolts and forged galvanized steel hook-and-eye turnbuckles to allow adjustments and tightening. By experience, lighter-gauge

materials and less reinforcing were found to be inadequate in preventing wall movement. The floor of the raceway was covered with 5 cm of fine sand and graded to provide a 2% slope to the three center drains. Walls were insulated with foam insulation board that measured 2.54 cm 1.22 m 2.44 m to minimize heat loss through the sidewalls. The outside perimeter of the floor was covered with foam insulation board that mea1.22 m 2.44 m and the center sured 5.0 cm strip of the floor was covered with expanded polystyrene foam insulation board that measured 2.54 1.22 m 2.44 m. The raceway was lined cm with a 20-mL high-density polyethylene raceway liner from Permalon (Reef Industries, Inc., Houston, Texas). A 15.24-cm drain line with three outlet drains (tee fittings) centered in each of the three cells was buried along the longitudinal axis of the raceway. A standard flange socket fitting was modified by boring out the center to allow for either a standpipe or a screened inlet and was installed on each of the three tee fitting outlets. A concentric ring of PVC sheet materials was used to secure the liner to the flat surface of the flange to provide a watertight seal at each drain. To provide for uniform bottom withdrawal from each of the three cells, a 5-cm orifice plate was installed in each flange to create a restricted outlet flow. This was designed based on a desired 5­15% bottom-withdrawal rate and a 30-cm head across the orifice to equalize flow rates. A 0.6-m-diameter circular plate was installed approximately 2.5 cm above the orifice to force water to move parallel with the bottom

MIXED-CELL RACEWAY DESIGN

197

into the outlet drain lines. The three drains discharged into a fiberglass sump tank measuring 1.83 1.83 1.83 m. The sump tank had both a standpipe for water level control and a drain line to flush the system. The sump tank was intended to fulfill several roles: solids management (by acting as a solids settling basin), water level set point with the standpipe, and harvesting basin (by flushing the production raceway through a screened harvesting cage). Eight 0.75-kW pumps were installed along the outside walls on platforms and discharged into a 10-cm schedule 40 PVC manifold that encircled the entire raceway. These pumps effectively simulated a dual-drain system, where most of the recycled water is removed from the top sides of a tank and only a small fraction is removed from the center drains. Two of the pumps were located on the sump collection tank, while the remaining six were placed at equal distances along the length of the raceway's outside walls. The inlet of the suction lines with check valves were located approximately 75 cm from the floor of the raceway. The pump discharges were connected to the manifold with a 5.08-cm flexible PVC hose that had a bronze gate valve to control flow rate. Each hydraulically separated cell created within the raceway measured approximately 5.44 5.44 m as determined by the down leg placements. Four 5.08-cm vertical manifold down legs were placed in the four corners of the raceway and four 7.62cm vertical manifold down legs constructed of schedule 40 PVC were located along the sidewalls, dividing the raceway into three equal cells. Two of the 7.62-cm vertical manifold down legs had 14 orifices that discharged in two directions parallel to the sidewalls (7 for each direction), and the other two vertical manifold discharge down legs had 7 orifices each that discharged perpendicular to the raceway walls. Starting 5.0 cm from the bottom, the orifices were spaced 15.24 cm apart Orifice openings were constructed by welding a 2.54 cm, which allowed threaded bushing 3.18 insertion of a 2.54-cm threaded plug. This allowed for easy modification of the orifice sizes and plugging of unused orifice openings. A clear section of rigid tubing was attached at the top of the down leg to act as a pieziometer tube manometer to measure vertical manifold backpressure. Analysis.--A SonTek Argonaut acoustic Doppler velocimeter (Yellow Springs Instruments, Yellow Springs, Ohio) was used to measure speed and direction within the hydraulically separated cells. The SonTek velocimeter is a single-point, three-

dimensional Doppler current meter that measures water velocity via the Doppler shift in frequency of sound from a moving object; in this case, small particulate matter in the water current. A transmitter generates a short pulse of sound from an acoustic transmitter that travels along the beam axis. As it passes through the sampling volume, sound is reflected in all directions and received by three acoustic receivers, which convert the reflected signals into Cartesian (x, y, and z) velocities (both speed and direction). The sampling volume is approximately a cylinder 0.6 cm in diameter and 0.9 cm long. The system samples 10 times a second with short pings of sound, and the user can specify averaging intervals to produce a mean velocity profile. The SonTek probe assembly was mounted on a rigid aluminum beam supported above the width of the raceway, which allowed the probe to be moved across the raceway width in a repeatable fashion. The probe was lowered into the raceway to specified depths of 10 cm from the bottom, middepth, and 10 cm below the water surface along a 0.5-m horizontal square grid measuring system. Water depth was 1.22 m in trial 1 and 1.0 m in trial 2. Measurements were taken for a 1-min averaging interval at each of the grid points and the values were averaged and used to plot the results. The data collected from the SonTek system were downloaded into Microsoft Excel (Microsoft Corporation, Redmond, Washington) for processing, and contour graphing of the velocities was created with SigmaPlot (SYSTAT Software, Inc., Chicago, Illinois). Results Trial 1 Figure 4 shows the contour plot for cell 3 (end cell) at velocity intervals of 5 cm/s. The pressure head was approximately 0.58 m gauge. In this case, only half of the cell was profiled to save time and resources and because there was no physical reason why symmetry should not prevail. Figure 4 shows that for cell 3, relatively high scouring velocities are seen at the outside perimeter of the cell and significantly lower velocities are seen near the center of the cell. Velocity profiles were also determined for cell 2. A weighted mean velocity was determined for each cell by multiplying the average mean velocities in an annular ring 0.5 m wide times the fraction of the total surface area of each ring. The velocity profiles shown in Figure 5 were created

198

EBELING ET AL.

FIGURE 4.--Contour plot of mixed-cell raceway water velocities (cm/s) at 5-cm/s intervals for half of cell 3 with orifice diameters of 20 mm and a pressure head of 0.58 m. Note that the top, bottom, and right-hand sides of the contour plot are solid walls.

three weighted mean velocities for middle cell 2, (15.1, 12.4, 10.7 cm/s, respectively), which supports the assumption that there is no significant difference in the velocity profiles between the end and middle cells. The higher bottom velocities at the center are partially a result of the high turbulence around the drain and obey the law of conservation of flow. Velocities in the z-direction, however, were usually very small; that is, measurements were just above the drain values of 2.2­ 2.8 cm/s. As can be seen from this data, the water velocities obtained were less than the design value of 20 cm/s, although they should work well in a stocked raceway, where the fish would help maintain the solids in suspension and move the solids to the central drain. The measured mean velocities were from 6% to 8% of the inlet jet velocities, in contrast to the assumed proportionality constant of 10% that was used. The observed proportionality values are closer to the value of 3.7% that was reported by Watten et al. (2000) for a shallower raceway. Trial 2 Based on the results of the first trial, the proportionality constant between the cell rotational velocity and the inlet velocity was estimated to be closer to 7%. Thus, the design orifice discharge velocity was increased to 300 cm/s, again designing for a cell rotational velocity of approximately

by averaging the velocities at each grid point in an annular ring 0.5 m wide starting at the center at each of the three depths. Figure 5 shows the almost linear velocity profile as a function of distance from the center drain. There was not as wide a variation in the top-, middle-, and bottomweighted mean velocities for cell 3 (11.5, 10.6, and 11.3 cm/s, respectively) compared with the

FIGURE 5.--Average water velocities in annular rings 0.5 m wide in a mixed-cell raceway that start at the center of cell 3 and have orifice diameters of 20 mm and a pressure head of 0.58 m.

MIXED-CELL RACEWAY DESIGN

199

FIGURE 6.--Contour plot of mixed-cell raceway water velocities at 5-cm/s intervals for cell 3 with orifice diameters of 15 mm and a pressure head of 1.35 m. Note that the top, bottom, and right-hand sides of the contour plot are solid walls.

20 cm/s. Using the orifice equation and solving for the required inlet manifold pressure yielded a value of 1.35 m. For trial 2, the depth of the water in the systems was reduced to 1 m. The same exchange rate (1.5 exchanges/h) was used as the previous trial as well as the same center drain's withdrawal percentage (15%). Using the orifice equation and the continuity equation and solving for the diameter of the orifices yields an orifice size of 15.0 mm. For trial 2, all orifice jets were made the same size.

A contour plot for cell 3 (end cell) at velocity intervals of 5 cm/s is shown in Figure 6 for 15mm discharge orifices and a pressure head of 1.35 m. Similar to the first trial (Figure 4), relatively high scouring velocities are seen at the outside perimeter of the cell and significantly lower velocities are seen near the center of the cell. The velocity profiles as a function of distance from the center drain are shown in Figure 7, which again shows that the velocity increased linearly from the center drain out. A weighted mean velocity measurement was determined for cell 3 for the top, middepth, and bottom planes (18.0, 15.6, and 14.1 cm/s, respectively). As can be seen from these data, the water velocities obtained were approximately 30% less than the design value of 20 cm/ s. The measured mean velocities were 5.8% of the inlet jet velocities, close to the 6% obtained in trial 1. Note that even though trial 1 and trial 2 had different approaches as to how to design the down legs that emitted orifice flows perpendicular to the tank walls, the mean velocities as a fraction of orifice flow velocity were almost identical. Discussion From an engineering perspective, the most significant design relationship is how the mean cell rotational velocity is related to the primary control variable, which is the orifice velocity. The weighted average water velocities obtained at the bottom in trial 1 (11.3 cm/s for cell 3 and 10.7 cm/s for cell 2) were significantly less than the design value of 20 cm/s, although they should work well in a

FIGURE 7.--Average water velocities in annular rings 0.5 m wide in a mixed-cell raceway that start at the center of cell 3 and have orifice diameters of 15 mm and a pressure head of 1.35 m.

200

EBELING ET AL.

stocked raceway because the fish would help maintain the solids in suspension and move them to the central drain. The measured weighted average water velocities at the bottom were from 5% to 6% of the inlet jet velocities, in contrast to the assumed proportionality constant of 10% that was used. The observed proportionality values are closer to the value of 3.7% that was reported by Watten et al. (2000) for a shallower raceway. The weighted average water velocities obtained at the bottom in trial 2 (14.1 cm/s for cell 3) were also significantly less than the design value of 20 cm/s. The measured weighted average velocities at the bottom were 7% of the inlet jet velocities or close to the 6% obtained in trial 1. Note that even though trial 1 and trial 2 had different approaches as to how to design the down legs that emitted orifice flows perpendicular to the tank walls, the mean velocities as a fraction of orifice flow velocity were almost identical. Using a weighted average based on mean velocities as a function of annular ring area suggests a larger value than Watten et al. (2000) observed but significantly less than is usually reported in the literature for circular tanks (3.8% in Watten et al. 2000 versus 15.0% in the literature). Based on this work, a practical design value appears to be between 6.0% and 8.0%. Davidson and Summerfelt (2004) have also done considerable work to characterize the water velocity profiles within two full-scale production tanks at the Conservation Fund's Freshwater Institute: 10-m3 and 150-m3 circular ``Cornell-type'' dual-drain tanks. The two systems consisted of three nursery tanks 3.7 m in diameter and a single grow-out tank 9.2 m in diameter. The culture tank diameter-to-water depth ratios were 4.8:1.0 to 4.0:1.0, respectively, compared with the mixed-cell diameter depth ratio of 5.5:1.0. Operating depth for the two tank systems was 0.76 m and 2.3 m and hydraulic exchange rates were four exchanges/h and two exchanges/h, compared with 1.5 exchanges/h in the mixed-cell raceway. Average inlet velocities for both of these systems were 206 cm/s based on either flow rate and orifice size or head pressure. The observed water rotational velocity at the perimeter of the culture tanks was 15­20 cm/s (nursery) and 30­37 cm/s (grow out), compared with 23.5 cm/s for the mixed-cell raceway. Davidson and Summerfelt (2004) reported that a rotational period of 1.3­1.7 min produced optimal water velocities for both exercising fish and flushing solids from the tank. For this mixed-cell raceway design, the cell diameter was 5.44 m (or

a perimeter of 21.8 m) with an average velocity of about 23.5 cm/s, suggesting a rotational period of about 1.5 min. In addition, Davidson and Summerfelt (2004) recommended a water discharge rate through the bottom center drains of at least 5­6 L/min for every 1 m2 of tank plan area to provide for rapid flushing of settleable solids and to maintain a self-cleaning ``Cornell-type'' dualdrain tank. Again, for the mixed-cell raceway with individual cells being 5.44 m in diameter, or having a 29.6 m2 cross-sectional area, the Davidson and Summerfelt (2004) criteria for cleaning would imply a flow rate of 162 L · min 1 · cell 1 and for the three cells 486 L/min, which is higher than the 340 L/min used in this initial design. However, the mixed-cell raceway average velocity in trial 2 for the bottom plane near the floor was 14.1 cm/s or about 30% less than the design velocity of 20 cm/ s. Thus, if the Davidson and Summerfelt (2004) criteria are applied by increasing the 14.1 cm/s to 20 cm/s, then the required flow for the mixed cells would increase from 340 L/min to 482.0 L/min. From this analysis, it is clear that it is the fractional ratio between orifice velocity and average tank velocity that is the key design parameter. As found in this study, this ratio is between 6% and 8% and can be used for design purposes to achieve the desired average velocities for the mixed-cell raceway. Conclusions This study shows that the mixed-cell raceway has several inherent advantages over circular tanks, including ease of sorting, grading, and handling fish as well as optimization of floor space. Results of this study showed excellent bottom velocities for scouring solids, moving them toward the center drains in each cell of the prototype mixed-cell raceway. Further work will focus on examining the relationship between the raceway exchange rate and cell rotational velocity, solids removal efficiency of the system, and management studies on how best to use the sump tank for solids collection and as a solids settling basin. Acknowledgments This work was supported by the U.S. Department of Agriculture, Agricultural Research Service, under Cooperative Agreement 59-1930-1130, and Magnolia Shrimp, LLC, in Atlanta, Georgia. References

Davidson, J., and S. T. Summerfelt. 2004. Solids flushing, mixing, and water velocity profiles with large

MIXED-CELL RACEWAY DESIGN

201

(10-m3 and 150-m3) circular ``Cornell-type'' dualdrain tanks. Aquacultural Engineering 32:245­271. Paul, T. C., S. K. Sayal, V. S. Sakhuja, and G. S. Dhillon. 1991. Vortex-settling basin design considerations. Journal of Hydraulic Engineering 117:172­189. Summerfelt, S. T., J. T. Davidson, T. Walkrop, S. Tsukuda, and J. Bebak-Williams. 2004. A partial reuse system for coldwater aquaculture. Aquacultural Engineering 31:157­181.

Timmons, M. B., S. T. Summerfelt, and B. J. Vinci. 1998. Review of circular tank technology and management. Aquacultural Engineering 18:51­69. Watten, B. J., D. C. Honeyfield, and M. F. Schwartz. 2000. Hydraulic characteristics of a rectangular mixed-cell rearing unit. Aquacultural Engineering 24:59­73.

Information

naja_67_306.193_201.tp

9 pages

Find more like this

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

1253920


You might also be interested in

BETA
naja_67_306.193_201.tp