Aeration of Water Suppliesfor Fish Culture in Flowing Water

Richard W. Soderberg
Fish Culture Program, Biology Department, Mansfield State College
Mansfield, Pennsylvania16933

ABSTRACT: An analyticalapproachto the reaerationof flowingwaterfor aquacultureispresented,

togetherwith a rational methodfor the assignment of dissolvedoxygenminima on the basisof
respiratorycharacteristics
of fish.Methodsfor calculationof expectedoxygentransfercapabilities
of
gravitydevicesand mechanicalunitsare given.

The water requirementsfor intensivefish culture more precisemethodfor determiningthe solubilityof

with respectto the oxygenconsumptionof fish have oxygen in water of any temperatureat any altitude
been widely documented(Haskell 1955; Willoughby would be by calculation. Several authors provide
1968; Piper 1970; Westers 1970; Liao 1971; Westers empiricalformulaefor the determinationof the equi-
and Pratt 1977)and themanagersof modernhatcheries libriumconcentrationof oxygenin water(Truesdaleet
generallyhavea working knowledgeof this informa- al. 1955; Whipple et al. 1969; Liao 1971) but the
tion. The extent to which water can be reused before expressiongiven by Truesdaleet al. (1955) givesthe
intolerable levels of un-ionized ammonia accumulate most widely acceptedvalues:
is similarly well understood(Westersand Pratt 1977).
Quantitative knowledge concerning the transfer of Ce = 14. 161 - 0.3943T + 0.007714T 2 - 0.0000646T 3
atmosphericoxygen to water in which the dissolved
oxygen (DO) level has been reducedbelow the equi- where Ce = equilibrium concentration of oxygen
hbrium concentrationis required for successfulfish (milligrams per liter) at a pressure of 760 mm of
culture. In this paper an analytical approach to the mercury, and T = water temperature in degrees
aeration of flowing water is offered. Celsius.
In contrastto the conditionsfor intensivehusbandry
Table 1. Equilibrium concentrations(Ce) of dissolved
of terrestrialanimals,in fish farming the supplyof the
oxygen in water at various temperatures and three
oxygencarrier, water, is limited. Therefore,the first
elevations (q[?er Truesdaleet al. 1955,'Liao 1971).
considerationin evaluatinga water supply for aqua-
culture is its volume (Haskell 1955;Willoughby 1968;
Piper 1970; Westors 1970; Liao 1971; Westersand Temperature Cc(rag/1.)
ofI)Oatelevations
aof
(ø C) 0 500 1000
Pratt 1977). The fish production that a given volume
of watercan supportdependson theamount of DO in 0 14.16 13.29 12.51
the water, the oxygenconsumptionrate of the fish, 2 13.40 12.58 11.84
4 12.70 11.92 11.23
and theefficiencywith whichfishcanextract DO from
the water. 6 12.05 11.31 10.66
8 I 1.47 10.76 10.13
10 10.92 10.25 9.65
Solubility of Oxygen in Water 12 10.43 9.78 9.22
14 9.98 9.36 8.82
Although oxygen occupiesnearly 2167c of the at- 16 9.56 8.97 8.45
mosphereby volume, it is only sparinglysoluble in 18 9.19 8.62 8.12
water. As with all dissolved gases, its solubility in 20 8.84 8.30 7.82
22 8.53 8.00 7.54
waterdecreasesasthetemperatureincreases, asatmos-
24 8.25 7.74 7.29
pheric pressuredecreases,or both. Cold water can
26 7.99 7.50 7.06
therefore hold more DO than warm water, and at sea
28 7.75 7.27 6.85
levelwater canhold more DO than at higherelevations 30 7.53 7.07 6.65
(Table 1). Most fish culturistsusesimilar tables with
correction tactors for different elevations. I believe a "Meters abo•c sea lc•cl.

Prog. Fish-Cult.44(2), April 1982 89

 The most accurateway to correctfor pressureis to
measure the barometric pressurewhen the oxygen
solubility is to be calculated.The pressurecorrection
factor then is P/760, where P: measuredbarometric
pressurein millimetersof mercury. If a barometeris
not available,elevationcan be usedas an approxima-
tion of barometricpressure.Liao (1971) providesthe
following pressure correction factor: 760/(760 +
E/32.8), where E: elevation in feet above sea level
(sincethisis an empiricalformula, the original English
units are retained).
Water containingDO at theequilibriumconcentra-
tion is saturatedwith oxygen. The amount of DO in
water is almostuniversallyexpressedin concentration
units (milligrams per liter), but for fish respiration
problemsit is more usefullyexpressedas pressurein
millimetersof mercury,whichisequivalentto express-
ing the amount of DO as percentof saturation.Since
standardatmosphericpressureis 760 mm Hg and the
atmospherecontains20.946% oxygenby volume, the
tension of oxygen in air at standard conditions is
159.2 mm Hg, in accordancewith Dalton's Law. The
oxygentensionin air at anybarometricpressureisthat
pressure(in millimetersof mercury)multipliedby the
decimal fraction 0.20946. To calculate the oxygen
tensionin water, the percentof saturation(measured
DO concentration divided by the calculated equi-
librium concentration)is multiplied by the tensionof
oxygenin air at the site.Ifa barometerisnot available
the averagebarometric pressurecan adequatelybe
estimated if the elevation is known. Thus, P/760:
760/(760 + E/32.8), where P: averageatmospheric
pressurein millimetersof mercuryand E = elevation
•n feet above sea level.
Oxygen Requirementsof Fish
In respiration fish blood picks up oxygen and
releasescarbondioxideat thegillsand picksup carbon
dioxide and releases oxygen at the tissues. The
efficiencyat which the blood combineswith oxygen
and carbon dioxide at different tensions therefore
determines the reaction of the fish to reduced DO
concentrationsin the water. Brungs (1971), in a
carefully controlled long-term study, found that the
growth of fathead minnows (Pimephalespromelas)
was reduced at all DO concentrations below saturation.
Andrews et al. (1973) reported that channelcatfish
(Ictaluruspunctatus)ate lessand grewmore slowlyat
60% saturation than at 100% saturation. The DO
tensionin fish culture systemscannot be kept at the
saturationlevelbecausefish respirationcontinuously
depletesthe oxygenpresentin the water. Minimum
oxygentensionsacceptablefor adequategrowth and
health of the fish should be defined as aquaculture
facility designcriteria.
Fish in warm water can tolerate lower concentrations
90
of DO than can fish in cold water because the tension
of a givenconcentrationof oxygenbecomesgreateras
the solubilitydecreases.This relation hasbeenverified
by field observationsfrom the literature. Smith and
Piper (1975), Piper (1970), Willoughby (1968), and
Westersand Pratt (1977) suggestedthat aquaculture
facilities for trout be designedso that the fish are
exposedto a minimumDO concentrationof 5.0 mg/L.
Bussand Miller (1971) called for aeration at trout
hatcherieswhen the DO concentrationwas predicted
to fall below 5.0-7.0 mg/L. Burrows and Combs
(1968) reportedthat salmongrowthwasreducedwhen
the DO concentrationfell below 6.0 mg/L. Warm-
waterfishare reportedto be able to surviveprolonged
exposuresas low as 1.0 mg/L, but in the range of
1.0-5.0 mg/L growth is reduced (H. S. Swingle,
Auburn University,unpublisheddata). Carlsonet al.
(1980) found that at 25øC the growth of channel
catfish was reduced at a constant DO exposureof
3.5 mg/Lbut notat 5.1 mg/L.
Piper (1970) reporteda water temperatureof 10øC
and an elevation of 1,500 m above sea level for his
station. At this site a minimum DO of 5.0 mg/L is
probablya reasonable designcriterionfor troutculture.
A rational basisfor assigningDO minima to warm-
water aquaculturesystemscould thus be the degreeof
oxygentensionin the water supply. For example,the
average oxygen tension of water at 10øC, 1,500 m
abovesealeveland a DO concentrationof 5.0 mg/Lis
72.8 mm Hg, correspondingto a concentrationof
3.44 mg/L at a water temperatureof 30øC at standard
pressure.This analysiscompareswith that of Downey
and Klontz (1981), who recommendeda minimum DO
tensionof 90 mm Hg.
When fish respirationhas reducedthe DO tension
to about 70-90 mm Hg, water can no longer be used
for fishculturebecauseat thispressurethefishcannot
efficientlyextract the oxygenpresentin thewater. The
water must be reconditionedby aeration (i.e., more
oxygenmustbe dissolvedinto the water)if it is to be of
further usefor fish production.
The extent to which water can be reconditioned with
aerationasthe only treatmentmeasuredependson the
accumulation of un-ionized ammonia. The number of
water usespermissiblebefore un-ionized ammonia
reachesunacceptablyhigh levels dependson water
temperature,feedingrate, protein contentof the diet,
and especiallyon the pH of the water.
Aeration Technology
Gravity Devices
Transfer of oxygen into water is a three-stage
processin which gaseousoxygenis transferredto the
surfacefilm, diffuses through the surfacefilm, and
Prog.Fish-Cult.44(2),April 1982
finally moves into the liquid bulk by convection
(Wheaton 1977). The rate of oxygen transferdepends
on the surfacearea over which transferoccurs.Oxygen
transferin aquaculturesystems wherewaterisquiescent
or moving in laminar flow is too slow to be an
•mportantsourceof DO for fishrespirationunlessthe
area of air-water interfaceis significantlyincreasedby
artificial turbulenceor agitation. Aeration of water
streamsusedfor aquaculturecan be accomplishedby
gravity where the energy releasedwhen water loses
altitude is used to increase the area of air-water
•nterfaceor by mechanicaldevicesthat spray water
into theair or pumpair into thewater(Wheaton1977).
The mostlogicalmeansof improvingoxygenregimes Fig. 1. Diagrams of gravity aerators. A, Simple weir
of cultured fish is by gravity fall of water between (Haskell et al. 1960; Chesnessand Stephens 1971),' B,
splashboard(Haskell et al. 1960; Chesnessand Stephens
production units, which can be provided by the
1971),'C, inclinedcorrugatedsheet(Chesness and Stephens
topographyat thefacility. The extentto whichwateris 1971),'D, inclinedcorrugatedsheetwith holes(Chesnessand
reaeratedby gravity is a fundamental concern for Stephens1971),'E, lattice aerator (Chesnessand Stephens
practicalfish culture in flowing water. 1971);and F, cascadeaerator (Tebbutt 1972).
Haskell et al. (1960) compared aeration by water
passageovera simpleweir (Fig. l A) with that obtained
by flow overa splashboard(Fig. 1B) that broke up the
water fall part way down, and that from flows over Table2. Selecteddata on measuredefficienciesof some
various screensand slat arrangements at the dam. gravity aerators over various distancesof waterfall.
Chesness and Stephens( 1971)evaluatedseveraldevices
for increasing oxygen transfer over a gravity fall, Device and distance
•ncludinga splashboard,an inclinedsheetof corrugated of waterdrop Efficiency
roofing material (Fig. I C), a similar corrugated sheet (cm) (%)
pierced with holes (Fig. I D), and an open stairstep Simple weir
devicereferredto asa lattice (Fig. I E). Tebbutt (1972) 22.9 a 6.2
studied aeration down closed stairsteparrangements 30.5b 9.3
of various heightsthat he called cascades(Fig. 1F). 61.0b 12.4
The following equation (Downing and Truesdale
1955) can be used to evaluate and compare gravity Inclinedcorrugated
sheet
b
aeration devices: 30.5 25.3
61.0 43.0
actual increase in DO
E=100 x
Inclinedcorrugatedsheet
possibleincreasein DO with holesb
30.5 30.1
or E: 100 x (Cb-Ca)/(Ce-Ca), where E: efficiency, 61.0 50.1
Cb: DO below the device, Ca: DO above the device,
and Ce - equilibriumconcentration(all in milligrams Splashboard
per liter). Selected data on measured efficienciesof 22.9 a 14.1
somegravity aeratorsover variousdistancesof water 30.5b 24.1
fall are presentedin Table 2. 61.0b 38.1
Practical application of this information requires
Latticeb
rearrangementof the efficiencyequation to solvefor
30.5 34.0
the expected DO concentration below an aeration 61.0 56.2
deviceof a known efficiency:
Cascade c
Cb- E(Ce-Ca) +Ca 25.0 23.0
100 50.0 33.4
75.0 41.2
100.0 52.4
The use of this equation is best illustrated by
example.Supposethereisa 30.5-cmdrop betweentwo Haskell et al. (1960).
ponds,with a simpleweir separatingthem. If thewater Chesness
andStephens
(1971).
temperatureis 10øCand the elevationis 183 m above Tebbutt (1972).

Prog. Fish-Cult. 44(2), April 1982 91

sea level, by using the equation of Truesdaleet al.
(1955)and Liao's( 1971) pressure correctionfactor,we
find that the solubility of DO will be
•2e- 14.161- 0.3943(10) + 0.007714(10)2
760
- 0.0000646(10)3 x
760 + 600/32.8
or Ce: 10.67 mg/L. If thefishloadingin theupstream
pondissuchthattheDO concentration isdepressed to
5.0 mg/L, Ca will be5.0. Weknowthata 30.5-cmfall
overa simpleweir is 9.3%efficient(Table 2). Using
theequationof DowningandTruesdale (1955),wecan
determine the DO below the weir (Cb) as follows:
Cb: 9.3(10.67
- 5.0)+ 5.0
100
or Cb = 5.53 mg/L. However, if a lattice structure
(Chesnessand Stephens1971)is constructedbetween
the ponds,the efficiencyrating of the 30.5-cmfall will
rise to 34%, so
Cb= 34(10.67
- 5.0)+ 5.0
100
or Cb = 6.93 mg/L. Althoughthe estimatedbenefitof
this gravity aeration device is considerable,most
aquaculture sitesrequiremechanical aerationto realize
the full productionpotentialof their water supplies.
Mechanical Devices
Mechanicalunits that agitatethe water surfaceare
normally usedin flowing water aquaculturesystems
because of their convenience and ease of installation.
Aeratorsare evaluatedand comparedon the basisof
their ability to transfer oxygen to water. Tests are
conductedunder standardconditionsof 760 mm Hg
pressure,
20øC temperature,
andzeroDO in thewater
to be aerated. The amount of oxygen added to the
waterin a givenamountof time undera certainpower
level is measured.The kilogramsof oxygenper shaft
kilowatt per hour (kg/kWoh) is givenby the aerator
manufactureras a measureof its efficiencyand can be
usedto compareunits.Actualoxygentransferdepends
on the oxygenconcentrationgradient:assaturationis
approached an increasing
amountof powerisrequired
perunit of DO transferred.Reaerationabove95%of
saturation can seldom be justified on a cost basis
(Mayo 1979). Westersand Pratt (1977) list 90% of
saturationasa reasonabledesigncriterionfor reaerated
water. Sinceaquaculturesystemsoperateat the rela-
tively high DO minima of 3-7 mg/L, actualtransfer
rates will be less than those determined under standard
conditions; thus these figures may not be used as
designcriteria.
92
Surface aerators are generally rated to transfer
1.9-2.3 kg/kWoh under standard conditions (Eck-
enfelder 1970). Whipple et al. (1969) found that
mechanical aerators in polluted rivers generally
provided oxygen transfer rates of 0.61 kg/kWoh or
less,but their testwater supplywashigherin oxygen
demand than is usual for aquaculture effluents.
Soderberg(1980) reportedan averagetransferrate of
0.83 kg/kWoh in static-watertrout pondswhere the
fish were heavily fed and aeration began when DO
tensions reached 70 mm Hg. Aeration of flowing
water should be more efficient becauseprocessed
water is continually being replacedfrom upstream
rather than being recirculated around the unit, as
might occur in static water conditions. To estimate
aeration requirements,we may use a conservative
value such as 0.6 kg/kWoh or we may estimate
oxygen transfer by using the formula presentedby
Whippleet al. (1969):
RT = RS
(Cet - Ca)(1.025x 20)(0.85)
Ce20
where RT = actual oxygen transfer, RS = oxygen
transferunderstandard
conditions,
Cet = equilibrium
concentration of DO at aeration site, Ca = DO above
aerator, Ce20= equilibriumconcentrationof DO
understandardconditions,and T = watertemperature
in Celsiusdegrees.
The following example illustratesthe use of this
formula. Supposean aerator is rated by the manu-
facturer
totransfer
2.0kg02/kWoh
under
standa•'d
conditions.The actual oxygentransferat a sitewhere
the watertemperatureis 10øC, barometricpressureis
725 mm Hg, and the DO in the waterto be aeratedis
5.0 mg/L, may be estimatedas follows:
RS = 2.0 kg/kWøh
Cet = 14.161- 0.3943(10) + 0.007714(10)2
- 0.0000646(10)3 x 725/760 = 10.42mg/L
Ca = 5.0 mg/L
Ce20= 14.161- 0.3943(10) + 0.007714(20)2
- 0.0000646 (20)3 =8.84 mg/L
T= 10øC
RT = 2.0
(10.42- 5.0)(1.025m-2ø)(0.85)
8.84
and, thus, RT = 0.81 kg/kWoh.
When the actual oxygentransferrate (RT) hasbeen
estimated,an estimateof the aerationcapabilityof a
particularunit and set of conditionscan readilybe
obtained.For the aboveexample,supposethat a 1.0-
kW unit will be usedin a water flow of 4,000 L/min.
The DO concentration below the aerator (Cb) is
calculated as follows:
Prog. Fish-Cult. 44(2), April 1982
 Cb = Ca +
0.80 kg x
106mg x 1.0 kW
Carlson, A. R., J. Blocher, and L. J. Herman. 1980. Growth
kWoh kg and survivalof channelcatfishand yellowperchexposed
to lowered constantand diurnally fluctuating dissolved
min h oxygenconcentrations.Prog. Fish-Cult. 42:73-78.
x
4,000 L 60 min Chesness,J. L., and J. L. Stephens.1971.A model studyof
gravityflow cascadeaeratorsfor catfishracewaysystems.
Trans. Am. Soc. Agric. Eng. 14:1167-1169, 1174.
Cb: 5.0 + 3.38 - 8.38 mg/L
Downey, P. C., and G. W. Klontz. 1981. Aquaculture
techn!ques:
Oxygen(PO2)requirement
for troutquality.
A more practicalapplicationof this knowledgemight Idaho Water and Energy ResourcesResearchInstitute,
be to sizean aeratorfor a particularjob. Supposethat Universityof Idaho, Moscow.42 pp.
for the sameexamplea unit that will return the DO to Downing, A. L., and G. A. Truesdale. 1955. Some factors
90% of saturationis desired.The amount of oxygen affecting rate of solution of oxygen in water. J. Appl.
Chem. 5:570-581.
needed would be
Eckenfelder,W. W., Jr. 1970.Oxygentransferand aeration.
Pages1-12 in W. W. Eckenfelder,ed. Manual of treatment
[(0.9) (10.42) - (5.0)]mg/L processes,Vol. I. Water ResourcesManagement Series,
EnvironmentalSciencesServiceCorporation,Stamford,
x
kg
--x--x
4,000 L ---
60 min --
1.05kg Conn.
106mg min h h Haskell, D.C. 1955.Weight of fish per cubicfoot of water in
hatcherytroughsand ponds.Prog. Fish-Cult. 17:117-118.
, R. O. Davies, and J. Reckahn. 1960. Factors in
/he size of the unit required (in kilowatts of shaft
hatcherypond design.N.Y. Fish Game J. 7:113-129.
power) would be Liao, P. B. 1971. Water requirementsof salmonids. Prog.
Fish-Cult. 33:210-215.
1.05 kg
--.x--
kWoh - 1.3kW Mayo, R. D. 1979.A technicaland economicreviewof the
h 081 kg useof reconditionedwater in aquaculture. Pages508-520
in T. V. R. Pillay and W. A. Dill, eds. Advances in
Thefinalstepin evaluating
anaeration
program
is aquaculture. FAO technical conferenceon aquaculture,
to comparethe cost of purchasingand operatingthe Kyoto, Japan.
equipment with the potential value of the increased Piper, R. G. 1970. Know the propercarryingcapacitiesof
fish production anticipated. The relation between your farm. Am. FishesU.S. Trout News 15:4-6.
shaft power and brake power is given by the motor Smith, C. E., and R. G. Piper. 1975.Lesionsassociatedwith
chronic exposureto ammonia. Pages497-514 in W. E.
manufactureras the efficiency,and thiscan be usedto
Ribelin and G. Migaki, eds. The pathology of fishes.
estimateenergy consumption.For example, a 1.30- Universityof WisconsinPress,Madison.
kW motor that is 75% efficient would draw 1.73 kW Soderberg,R. W. 1980.Aeration to intensifytrout culturein
of energy. If run continuouslyon electricalpower that static water ponds. M. S. thesis, Auburn University,
was purchasedfor $0.10/kWoh, the unit would cost Auburn, Ala. 35 pp.
$124.80per month to operate. Tebbutt, T. H. Y. 1972. Some studies on reaeration in
cascades. Water Res. 6:297-304.
Truesdale,G. A., A. L. Downing, and G. F. Lowden. 1955.
Acknowledgment The solubility of oxygen in pure water and sea-water.
J. Appl. Chem. 5:53-62.
Preparation of manuscript was aided in part by Westers,H. 1970.Carryingcapacityof salmonidhatcheries.
Hatch Project Alabama No. 497, Auburn University Prog. Fish-Cult. 32:43-46.
Agricultural Experiment Station. , and K. M. Pratt. 1977. Rational design of
hatcheries for intensive salmonid culture, based on
metaboliccharacteristics.Prog. Fish-Cult. 39:157-165.
References
Wheaton, F. W. 1977. Aquacultural engineering. John.
Wiley & Sons,New York. 708 pp.
Andrews, J. W., T. Murai, and G. Gibbons. 1973. The
Whipple, W., Jr., J. V. Hunter, B. Davidson,F. Dittman,
influenceof dissolvedoxygen on the growth of channel
catfish. Trans. Am. Fish. Soc. 102:835-838. and S. Yu. 1969. Instream aeration of polluted rivers.
Water ResourcesRes. Inst., Rutgers University, New
Brungs,W. A. 1971.Chroniceffectsof low dissolvedoxygen
Brunswick,N.J. 196 pp.
concentrationson fathead minnows (Pimephalesprome-
Willoughby, H. 1968. A method for calculating carrying
las). J.Fish. Res. Board Can. 28:1119-1123.
capacitiesof hatchery troughs and ponds. Prog. Fish-
Burrows, R. E., and B. D. Combs. 1968. Controlled Cult. 30:173-174.
environmentsfor salmonpropagation.Prog. Fish-Cult.
30:123-136.
Buss, K., and E. R. Miller. 1971. Considerations for
conventionaltrout hatchery designand constructionin
Pennsylvania.Prog. Fish-Cult. 33:86-94. Accepted12 November1981

Prog. Fish-Cult. 44(2), April 1982 93