Aquacultural Engineering 32 (2005) 425–442

www.elsevier.com/locate/aqua-online

Characterization of water quality factors

during intensive raceway production of juvenile
Litopenaeus vannamei using limited discharge
and biosecure management tools
Jason M. Cohena,b, Tzachi M. Samochaa,b,*, Joe M. Foxb,
Ryan L. Gandya, Addison L. Lawrencec
a
Texas Agricultural Experiment Station, Shrimp Mariculture Research Facility,
4301 Waldron Rd., Corpus Christi, TX 78418, USA
b
Texas A&M University-Corpus Christi, 6300 Ocean Dr., Corpus Christi, TX 78412, USA
c
Texas Agricultural Experiment Station, Shrimp Mariculture Research Lab,
1300 Port St. Port Aransas, TX 78373, USA
Received 28 September 2004; accepted 28 September 2004

Abstract

Disease epizootics have negatively affected production and expansion of the shrimp culture
industry. This, along with environmental concerns regarding limited water resources and contam-
ination of receiving streams, has caused the industry to investigate more sustainable and biosecure
management practices. A study was conducted to evaluate the effect of limited water exchange on
water quality, growth and survival of the Pacific white shrimp Litopenaeus vannamei postlarvae (PL)
in greenhouse-enclosed raceways. Concentrations of NH4-N did not exceed 2.0 mg l
1 during this
period; whereas, NO2-N exceeded 26.4 mg l
1, indicating assimilation of primary amines by primary
productivity. Periodic removal of suspended solids by a common pressurized sand filter and injection
of oxygen into culture water resulted in high-survival rates for both raceways (97.5 and 106.0%) with
an average biomass yield of 4.29  0.06 kg m
3. Shrimp samples collected during the nursery trial
and at harvest showed no signs of bacterial or viral pathogen infections.
# 2004 Elsevier B.V. All rights reserved.

Keywords: Intensive nursery; Limited discharge; Pacific white shrimp; Litopenaeus vannamei

* Corresponding author. Tel.: +1 361 937 2268; fax: +1 253 390 6081.
E-mail address: [email protected] (T.M. Samocha).

0144-8609/$ – see front matter # 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.aquaeng.2004.09.005
426 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442

1. Introduction

Rapid expansion of shrimp aquaculture can be attributed to technological advances,

high-market demand and reduced supplies of wild stocks (Barg et al., 1999). Despite this
growth, the shrimp farming industry has experienced frequent setbacks, largely as a result
of catastrophic viral epizootics (Lightner et al., 1997). This problem has persisted into the
present and is evidenced by the recent outbreak of Taura Syndrome Virus Disease in South
Texas. To reduce potential for infection of commercial shrimp production facilities with
catastrophic disease-related viruses, an appropriate and cost-effective strategy towards
disease prevention via improved biosecurity is required.
Probably the most common pathway for pathogen introduction into an aquaculture
facility is via introduction of water (Lotz and Lightner, 1999). In traditional intensive
shrimp culture, a substantial portion of pond water is discharged daily into receiving
streams, necessitating similar replacement with source water. According to Timmons and
Losordo (1994), the production of 1.0 kg of penaeid shrimp using traditional, flow-through
criteria requires approximately 20,000 l of water, which is ultimately released into
receiving streams. In the past, acceptable water quality for pond production was
maintained largely by release of waste pollutants via continuous large-volume water
exchange (Hopkins et al., 1993; Moss et al., 1999). In many cases, no attempt was made to
reduce these loads prior to discharge. Discharge of nutrient-rich or high-oxygen demand
pond effluent into low recharge receiving streams has precipitated conflict with other
potential users (Boyd and Yoo, 1994). Effluent from production systems is typically high in
waste nutrients (e.g., nitrogen and phosphorus), total suspended solids (TSS), volatile
suspended solids (VSS), biochemical oxygen demand (BOD) and chemical oxygen
demand (COD). Because of substantial temporal and chemical variation in discharged
pond water (Preston et al., 2000) as well as receiving streams (Trott and Alongi, 2000), the
assessment of impact remains complicated.
This discharge issue has generated substantial interest in the shrimp farming industry to
achieve zero or limited water exchange in production systems. For these systems to be
effective, though, proper feed management, adequate aeration and circulation, natural
productivity and nitrogen cycling processes must be carefully managed. The typical
operational criteria impacting recirculating aquaculture systems (RAS) as well as their
basic design components have been described extensively by Timmons et al. (2001).
Several major operational issues impact these types of production systems: management of
solids, availability of nutrients to shrimp and other system biota, nutrient ratios and
processing of nitrogenous waste.
Solid waste is generated either directly or indirectly from feed added to the system and
is present as uneaten feed residuals, metabolites, manures and/or microorganisms growing
in the system (Viadero and Noblett, 2002). For traditional open shrimp production systems,
Funge-Smith and Briggs (1998) identified feed as a significant source of organic matter
(31–50%) but contributing few solids (4–8%) to the system (Funge-Smith and Briggs,
1998). Erosion of pond soil was the major source of both solids (88–93%) and organic
matter (40–60%). Because many limited water exchange production ponds and raceways
are plastic-lined and have a reduced requirement for input water, it stands to reason that
most solids and organic matter are ultimately derived from feed. According to Thakur and
 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442 427

Lin (2003), the major water quality-related problem associated with closed system
production of shrimp is the potential for rapid eutrophication, resulting from increasing
concentration of nutrients and organic matter over the culture period. Ultimately, this
buildup impacts carrying capacity of the culture environment (Lin, 1995) to the point that
commercial success depends upon striking a balance between waste production and
assimilation capacity (Thakur and Lin, 2003). A study conducted in zero-exchange shrimp
ponds in Belize (Burford et al., 2003) showed that despite experiencing ‘‘poor’’ water
quality in ponds (i.e., high-nutrient concentrations, high and unstable phytoplankton and
bacterial numbers), shrimp yields surpassed those achieved in conventional (i.e.,
traditional) ponds.
The traditional approach to intensive culture of shrimp has used high rates of water
exchange (e.g., flushing) and assumed that shrimp derive the majority of their nutrition
from feed (as opposed to biotic pond resources). Past and recent researchs (Rubright et al.,
1981; Hunter et al., 1987; Moriarty et al., 1987; Leber and Pruder, 1988) have indicated the
importance of benthic fauna to penaeid nutrition in ponds. Litopenaeus vannamei cultured
in RAS typically use microalgae and associated detritus as food sources (Moss et al., 1999).
The contribution of selected fractions of shrimp pond water to growth enhancement of L.
vannamei was demonstrated by Moss et al. (1992).
Another area of interest in the development of intensive minimal water exchange
systems is the biogeochemical cycling of nutrients important to overall maintenance of the
penaeid shrimp culture system. Several recent studies have addressed this cycling via
development of either management strategies or nutrient budgets (Avnimelech, 1999;
Burford and Williams, 2001; Thakur and Lin, 2003; Burford et al., 2003; Wang, 2003;
Otoshi et al., 2003; Jackson et al., 2003). Thakur and Lin (2003) showed for closed culture
of Penaeus monodon that despite the major source of nutrient input being feed (76–92% of
nitrogen and 39–67% of phosphorus), shrimp could assimilate only 23–31% and 10–13%
of these nutrients, respectively. For open intensive ponds, Funge-Smith and Briggs (1998)
found only 6–11% of carbon applied was assimilated, implying it was either incorporated
into plankton biomass, volatilized or trapped in sediments. The main source of dissolved
nitrogen in ponds is excretion from shrimp gills as ammonia; however, a substantial
amount of dissolved organic nitrogen (DON) comes from feed (dissolved primary amines)
and feces, as urea (Burford and Williams, 2001). High-density, zero exchange culture of
shrimp in plastic-lined ponds in Belize (Burford et al., 2003) resulted in high
concentrations of dissolved inorganic nitrogen (DIN; 2.29–5.56 mg l
1), DON (0.17–
10.66 mg l
1), DOC (14.2–48.1 mg l
1) and phosphate (0.07–1.17 mg l
1). With added
aeration, a high level of bacterial flocculation occurs. This results in establishment of a high
DOC/DN ratio in culture water, which ultimately promotes breakdown of organic matter,
and reduces both DIN and feed input (Avnimelech et al., 1994; Avnimelech, 1999; Browdy
et al., 2001).
There is currently strong research interest in establishing a proper ratio between carbon
and nitrogen levels in shrimp culture water in order to enhance nitrogen cycling and
promote growth of bacteria as a low-cost nutrient source. Avnimelech (1999) used glucose
to reduce inorganic nitrogen in experimental tanks containing P. monodon. He suggested
that feeds containing 30% protein could be amended by an additional 46.5% carbohydrate
in order to assimilate ammonium flux into microbial protein. This called for use of a
428 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442

20.48% CP feed having a C/N ratio of 15.75:1. Net uptake of DIN occurs only when C/N
ratio of organic matter in coastal marine ecosystems is higher than 10 (Lancelot and Billen,
1985). However, in intensive closed shrimp pond production, Burford et al. (2003) showed
that increased pond age (i.e., increased addition of carbon) had no major effect on
heterotrophic status of the pond and that due to high availability of nitrogen and
phosphorus to phytoplankton, ponds fluctuated between autotrophy and heterotrophy. They
suggested that a heterotrophic system might only be achieved if feed carbon input exceeded
that of primary production.
In order to improve production scheduling, most limited water exchange production
systems require integration with a preliminary nursery phase. Use of this phase has long
been recognized to have several benefits over direct stocking of PL in commercial shrimp
production operations (Sandifer et al., 1991). Intermediate production of juvenile shrimp in
smaller raceway impoundments at high densities has been shown to improve survival and
feeding efficiency (Samocha and Benner, 2001; Samocha et al., 2000, 2002, 2003). From
the standpoint of space utilization, it is more efficient to stock production ponds with larger,
juvenile shrimp which will also reduce production cycle duration. These systems must be
managed under biosecure criteria with simple strategies towards management of dissolved
nitrogen and solids to avoid potentially toxic and eutrophic conditions. The present study
investigated the use of phytoplankton in combination with periodic solids removal and
oxygen supplementation as management tools for the production of juvenile shrimp in
plastic-lined limited water exchange shrimp nurseries.
The objectives of this study were to (1) characterize change in various physiochemical
water quality factors during duplicate limited water exchange raceway production trials of
juvenile shrimp; (2) evaluate maintenance of biosecurity conditions; and (3) based upon
results, suggest means whereby the present research system could be modified for
commercial production.

2. Materials and methods

2.1. Experimental design

This study was conducted in two 68.5 m2 (45 m3) HDPE-lined greenhouse-enclosed
raceways (RW-1, -2) at the Texas Agricultural Experiment Station, Shrimp Mariculture
Research Facility, Corpus Christi, Texas. Physiochemical factors associated with raceway
production of juvenile shrimp were compared and characterized in side-by-side production
trials involving similar raceways stocked with the same-age shrimp and species.

2.2. Water source and treatment

Hypersaline seawater (40–45 ppt) was used to fill experimental raceways with filtered
water (500 mm) from a nearby lagoon (Laguna Madre). After adjusting the salinity to 30
ppt with municipal freshwater, water was chlorinated with a concentrated liquid bleach
solution (12.5%) to achieve an initial chlorine concentration of 10.0 mg l
1 with targeted
residual chlorine of 1.0 mg l
1 after 24 h. After dechlorination with aeration, water was
 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442 429

fertilized with urea, phosphoric acid and sodium silicate to achieve concentrations of 5.5,
0.4 and 0.7 mg l
1, respectively. Raceway water was then inoculated with the diatom,
Chaetoceros muelleri, to yield an initial algal cell density of 50,000 cells mg 1
1.

2.3. Stocking of raceways and experimental system

Raceways were stocked (3,300 PL m
3) with certified specific-pathogen-free 8–10-day

old postlarvae (PL8–10) of the Pacific white shrimp L. vannamei purchased from a
commercial hatchery (Harlingen Shrimp Farms Ltd., Los Fresnos, TX). Each raceway was
outfitted with a recirculating water treatment system including external centrifugal pumps
(2 hp) and pressurized sand filter (265 l min
1 capacity; Aquatic Eco-Systems Inc.,
Apopka, FL). Raceways were equipped with a longitudinal center partition positioned over
a 5.1-cm diameter PVC return pipe. This pipe was outfitted with a one 5.1-cm diameter
venturi injector (Mazzei Injector Co., Bakersfield, CA) and a series of spray nozzles along
its length. The purpose of this pipe was to enhance water movement near the bottom so as to
reduce potential for sedimentation and, if required (see below) to provide oxygen enriched
water along the raceway (Sturmer et al., 1992; Samocha et al., 2002). Oxygen was provided
by pressurized cylinders and manually injected into the venturi. For standard operational
aeration, each raceway also contained eighteen 5.1-cm diameter airlift pumps, six air
diffusers (1.0-m length; Bio-WeaveTM, Aquatic Eco-Systems Inc., Apopka, FL).
Raceways were maintained under limited water exchange for the duration of the study.
Water (chlorinated with subsequent dechlorination) was intermittently added to raceways
to compensate for evaporation and increased salinity and for backwashing of pressurized
sand filters. Mean daily water exchange rate for the duration of the study was 0.022% of the
total raceway volume for both raceways. Pressurized sand filtration of raceway water was
undertaken once total suspended solids exceeded 120 mg l
1 and/or dissolved oxygen
(DO) decreased to below 75% saturation.

2.4. Water quality determinations

Dissolved oxygen, pH, salinity, and temperature were monitored daily. Ammonium-N,
nitrite-N, nitrate-N, total phosphorus, reactive phosphorus, chemical oxygen demand, 5-
day carbonaceous biochemical oxygen demand (cBOD5), total suspended solids, and
volatile suspended solids were monitored weekly (APHA et al., 1995; Table 1).

2.5. Feed management

Postlarvae were fed two inert ‘‘micronized’’ feeds: PL Redi-Reserve (Zeigler Bros.,
Gardners, PA), and Rangen crumble feed 45/10 size 0–3 (45% crude protein and 10% squid
meal; Rangen Inc., Buhl, ID). Newly hatched Artemia nauplii were provided for two days
post- stocking at a rate of 50 nauplii PL
1 day
1. Daily rations were adjusted according to
observed feed consumption and in conjunction with frequent growth monitoring. Initially,
daily feed ration was established as 14% of estimated wet shrimp biomass (as dry weight
feed) in each raceway. Daily rations were eventually reduced to about 8% during the last
week of culture.
430 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442

Table 1
Water quality indicators analyzed and monitoring method
Parameter Method
Ammonium (NH4) Oriole model EA 940 Ion analyzer and model 95-12 ammonia electrode;
standard method #4500-NH3 Da
cBOD5 5-day incubation at 20 8C; standard method #5210 Ba
COD Reactor digestion method, HACHd Method #8000b;
standard method # 5220 D
Dissolved oxygen YSIc model 610
Nitrate Cadmium reduction method; HACH method #8039;
standard method # 4500-NO3
Ea
Nitrite Diazotization method; HACH method # 85071b; method # 4500-NO2
Ba
pH YSI model 610
Reactive phosphorus Ascorbic acid method; HACH #8084b; standard method # 4500-P Ea
Salinity YSI model 610
Temperature YSI model 610
Total phosphorus Acid persulfate digestion; HACH method #8190b;
standard method #4500-P Ba
Total suspended solids Standard method # 2540 Da
Volatile suspended solids Standard method # 2540 Ea
a
APHA et al. (l995).
b
USEPA approved for reporting of water and waste water.
c
Thermo Orion, Beverly, MA.
d
HACH Company, Loveland, CO.

2.6. Substantiation of biosecurity

Every 2 weeks a random sample of 25 shrimp was collected from each raceway and
placed on ice. Samples were shipped frozen to a diagnostic lab (Texas Veterinary Medical
Diagnostic Laboratory, College Station, TX). A pooled sample taken from the 50 shrimp
was used to test for infection by Infectious Hypodermal Necrosis Virus, Taura Syndrome
Virus and White Spot Syndrome Virus using one-step polymerase chain reaction (PCR)
procedure.

2.7. Statistical analyses

Differences in mean daily and weekly water quality parameters were evaluated
between raceways via paired student’s t-tests at a significance level of P = 0.05 (SPSS,
1999).

Table 2
Harvest results from a 50-day nursery trial in raceways
Raceway FCR Mean weight (g) Yield (kg/m3) Survival (%) Water added
(% total volume/day)
#1 0.86 1.12 4.25 97.5 1.1
#2 0.98 1.01 4.33 106.0 1.1
 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442
Table 3
Summary of daily water quality data for raceway 1 and 2 during a 50-day nursery trial
A.M.
RW Temperature (C) Salinity (ppt) D.O. (mg l
1) pH
Mean Maximum Minimum Mean Maximum Minimum Mean Maximum Minimum Mean Maximum Minimum
1 28.1  1.6 30.8 24.5 29.6  0.7 30.9 27.7 5.8  1.2 8.3 2.1 7.6  0.4 8.5 6.9
2 28.4  1.3 30.5 26.4 29.7  1.2 31.2 26.3 6.2  0.7 7.6 4.4 7.6  0.3 8.3 6.7
P.M.
1 28.0  1.5 30.8 24.8 29.6  0.7 30.8 27.8 5.8  0.9 7.4 3.7 7.6  0.4 8.5 6.7
2 28.2  1.3 30.5 24.8 27.3  1.2 31.5 25.3 5.8  1.1 8.2 3.1 7.5  0.3 8.1 6.9

431
432 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442

3. Results

3.1. Growth and survival

Mean wet weight of shrimp at harvest after 50 days was 1.12 and 1.01 g for RW-1 and -
2, respectively (Table 2). Shrimp survival over this same period was 97.5 and 106.0% for
RW-1 and -2, respectively, with feed conversion ratios (FCR) values below 1.0 for both
raceways (Table 2).

3.2. Physiochemical factors

Tables 3 and 4 summarize daily and weekly water quality indicators, respectively,
measured in the two raceways over the 50-day study. No significant differences in
concentrations of daily and weekly water quality indicators were found between raceways.
All daily readings were within parameters considered adequate for commercial production
of juvenile marine penaeid shrimp. A low-oxygen concentration of 2.1 mg 1
1 was
measured in RW-2 near the end of study that necessitated water exchange in conjunction
with oxygen injection.
Figs. 1–3 show temporal variation in ammonium-N, NO2-N and NO3-N, respectively,
for both raceways over the culture period. Ammonium levels fluctuated substantially in
both raceways, between 0.001 and 2.030 mg l
1. An exponential increase in both NO2-N
and NO3-N was shown for both raceways, eventually increasing to 26.4 mg l
1 and greater
than 13.8 mg l
1, respectively.
Five-day carbonaceous biochemical oxygen demand ranged from minima of 5.04 and
6.52 to maxima of 29.88 and 28.60 mg l
1 in RW-1 and -2, respectively. In general, cBOD5
increased from stocking to week 5 or 6 after which it declined until harvest (Figs. 4 and 5).
At week 5, water in both raceways was pressurized sand-filtered as part of a prescribed
treatment to reduce TSS. Chemical oxygen demand in both raceways decreased from

Table 4
Summary of weekly changes in water quality indicators for raceways 1 and 2 in 50-day nursery trial
Week # Raceway # COD NH4-N NH3-N NO2-N NO3-N cBOD5 TSS VSS
(mg l
1) (mg l
1) (mg l
1) (mg l
1) (mg l
1) (mg l
1) (mg l
1) (mg l
1)
1 1 1,990 0.001 T 0.58 0.7 5.04 94.0 58.0
2 2,430 0.001 T 0.91 0.9 6.52 90.0 54.0
2 1 1,810 0.851 0.01 0.42 0.2 5.92 120.0 70.0
2 1,660 0.652 0.03 1.24 0.7 5.12 100.0 20.0
3 1 610 0.486 0.01 0.175 0.6 12.16 80.0 na
2 1,060 0.878 0.06 1.30 0.6 10.96 80.0 na
4 1 810 0.342 0.01 1.16 0.7 18.4 110.0 62.0
2 730 2.030 0.04 2.14 1.0 16.76 80.0 28.0
5 1 870 1.820 0.29 1.82 3.7 29.88 160.0 76.0
2 880 1.284 0.10 3.06 3.3 25.32 190.0 114.0
6 1 1,110 1.698 0.32 11.3 8.0 28.40 70.0 46.0
2 1,260 1.880 0.21 10.85 5.0 28.60 80.0 70.0
7 1 1,570 1.382 0.01 26.40 13.8 27.84 130.0 74.0
2 1,490 0.250 0.03 26.4 16.9 21.60 90.0 50.0
 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442 433

Fig. 1. Total ammonia nitrogen (TAN) concentration in raceways 1 and 2.

Fig. 2. Nitrite (NO2-N) concentration in raceways 1 and 2.

Fig. 3. Nitrate (NO3-N) concentration in raceways 1 and 2.

Fig. 4. Five-day carbonaceous biochemical oxygen demand (cBOD5) in raceways 1 and 2.

434 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442

Fig. 5. Chemical oxygen demand (COD) in raceways 1 and 2.

Fig. 6. Total suspended solids (TSS) in raceways 1 and 2.

Fig. 7. Volatile suspended solids (VSS) in raceways 1 and 2.

maxima at week 1 to minima at either week 3 (RW-1) or week 4 (RW-2). After these points,
COD continued to gradually increase until harvest.
Total and volatile suspended solids concentrations for RW-1 and -2 are shown for the
duration of the study in Figs. 6 and 7. In general, levels of both these factors increased until
week 5, at which time TSS levels in both raceways exceeded 120 mg l
1. Both raceways
 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442 435

were then subjected to pressurized sand filtration of culture water. This prescribed
treatment resulted in an acute decrease in TSS and moderate decrease in VSS. Within
1 week of harvest, TSS began to increase again, whereas VSS showed a mixed
response.

4. Discussion

4.1. Nitrogen transformations

There was considerable flux in ammonium-N in both raceways over the entire culture
period (Fig. 1). This could have been associated with either delayed development of
nitrification reaction 1 bacteria (e.g., Nitrosomonas sp.) or general flux in phytoplankton
concentration. Large-scale fluctuation in ammonium has been demonstrated in closed
P. monodon culture systems (Thakur and Lin, 2003) and was generally attributed to
variation in uptake by phytoplankton or nitrifying bacteria. Flux in phytoplankton
density (approximated in the present study as VSS) in closed, high-density shrimp
production systems appears to be a natural phenomenon of limited impact on growth and
survival (Burford et al., 2003). Because both NO2-N and NO3-N concentrations increased
exponentially over the study, it appears that the entire nitrification process was
substantially delayed and that assimilation by phytoplankton was primarily responsible
for changes in ammonium-N. Reaction 1 nitrification (NH3/NH4-N ! NO2-N) appears to
have started around weeks 5 and 6 of culture (Figs. 1–3). A delay in nitrification could
have been associated with rapid increase in organic carbon, shifting raceway bacterial
communities towards heterotrophy (Zhu and Chen, 2001). This is somewhat substantiated
by the general increase in cBOD5 in both raceways over the duration of the study
(Fig. 4).
Although NH4-N maxima for RW-1 and -2 were 0.32 and 0.21 mg l
1, it apparently did
not affect shrimp growth and survival in either raceway. Burford et al. (2003) demonstrated
that in closed and plastic-lined shrimp ponds operated at high-stocking densities, a TAN
level of 0.56 mg l
1 had no major impact on production. It has been the authors’ experience
that high concentrations of NH4-N are often experienced in pond cultures under limited
water exchange conditions due to increased pH associated with development of heavy algal
blooms. Daily observations of pH in both raceways indicated only slightly basic culture
conditions. As mentioned, NH4-N and NH3-N appeared to be controlled by uptake via
phytoplankton.
High NO2-N concentrations were eventually recorded in both raceways (Fig. 2). Chen
and Chin (1998) reported a 96-h LC50 for exposure of P. monodon PL to NO2-N of
45 mg l
1 and recommended a ‘‘safe’’ level of exposure of less than 4.5 mg l
1 for pond
grow-out. Despite a maximum of 26.4 mg l
1 NO2
-N shown in the present study for both
raceways, no apparent detrimental effect on either growth or survival of shrimp was
observed. Studies by Chen and Chin (1998) and Chen and Lei (1990) appear to indicate an
age-related tolerance of NO2-N. It is postulated that because NO2-N concentration
increased relative to age of shrimp (Fig. 2), the rate of change was sufficiently slow to allow
for acclimation.
436 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442

The exponential NO2-N concentrations in both raceways towards termination of

the study appeared to indicate delayed or limited oxidation of nitrite to nitrate (Reaction 2).
A decrease in NH4-N in both raceways was shown in week 7, which possibly indicates
initiation of nitrification Reaction 1. This could have coincided with a reduction in
phytoplankton concentration in both raceways associated with solids management
(i.e., removal by pressurized sand filtration once TSS had exceeded 120 mg l
1). A
substantial portion of phytoplankton could have been removed from the raceways,
improving conditions for nitrification of dissolved inorganic nitrogen by chemoautotrophic
bacteria.
Another explanation for nitrite-N accumulation in limited-discharge aquatic systems
could be oxygen limitation. Of the chemoautotrophic bacteria responsible for nitrification,
the Nitrobacter-like species have been shown to be most sensitive to low oxygen (Komaros
and Lyberatos, 1998). Thus, under sub-optimal DO conditions, an increase in NO2-N could
have occurred. This phenomenon can result in reversal of nitrification (NO3-N ! NO2-N).
Also, because this reaction develops more slowly than Reaction 1 (NH4+ ! NO2
), a
general lag effect could have been shown (duration of the study was only 5 weeks). Low-
oxygen concentrations were observed periodically in both raceways during the course of
the study.
In some systems, ammonium oxidation can lead to NO2
and NO3
accumulation in the
culture system (Krom et al., 1995). Van Rijn and Rivera (1990) showed nitrite accumulated
in trickling filters at NH4+ removal rate > 0.3 mg NH4+-N l
1. Nitrite accumulation occurs
as a succession in the developing nitrification system where a lag period develops between
the conversions of NH4+ and NO2
, and NO2
to NO3
(a common sequence in
hypertrophic systems; Krom et al., 1995). Persistent accumulation of NO2
, beyond this
simple progression, could be explained by the following mechanisms: firstly, presence of
low DO zones in the system (van Rijn and Rivera, 1990). These zones, found where organic
matter accumulates at the bottom of culture system components, providing a substrate for
bacteria that reduce NO3
to NO2
(Arbiv and van Rijn, 1995). This reaction is often
referred to as passive denitrification. With the presence of extensive aeration and turbulent
mixing, it is not likely that this was the cause. A second factor could be slow development
of the nitrifying community related to oxidation of nitrite-N to nitrate-N.
It has been found that the second step in the nitrification process, NO2
reduction, is
inhibited by the presence of nitrates (Komaros et al., 1996). Furthermore, this reaction
appears to be more sensitive to oxygen (Komaros and Lyberatos, 1998). Since the second
step is the slowest step, NO2
-N could accumulate in significant concentrations.
Ultimately, the most efficient means of controlling nitrite accumulation is to assure its
conversion via Reaction 2 of nitrification by maintenance of a completely aerobic
environment. This could be accomplished by continuous and increased aeration or
increased saturation with the use of pure oxygen gas (i.e., oxygen injection). The authors
postulate that physiochemical control of the system described in this study should focus
primarily on assurance of an aerobic environment and, secondly, limitation of nitrite in
water. Although high concentration of nitrite can be lethal to shrimp, the short-term
exposure to nitrite concentrations found in this study (26.4 mg l
1) should be considered of
secondary importance due to its apparent limited influence on mortality and health of the
juvenile shrimp.
 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442 437

It is likely that overall control of nitrification was affected by a combination of nitrifying

bacteria, heterotrophic bacteria and phytoplankton. Of these three agents, heterotrophic
bacteria and phytoplankton appear more probable due to lack of substrate for colonization
of nitrifying bacteria. The potential for NO2-N toxicity in similar systems could perhaps be
avoided through the use of pre-conditioned biofilters in the treatment process.
In the present study, very little sedimentation of particulate matter (a possible substrate
for denitrification) was observed. This lack of substrate, in combination with the
observation that NO3
-N also accumulated in both raceways appears to substantiate a
general lack of denitrification and support the hypothesis of low, but not anoxic, DO
conditions. Confirmation of the contribution of nitrification in overall system DO demand
is warranted for future studies.

4.2. Evaluation of oxygen demand-related factors

Carbonaceous oxygen-demanding materials (measured as cBOD5 and COD) appeared

to initially show a gradual increase with subsequent stabilization until termination of the
study (Figs. 4 and 5, respectively). These increases are to be expected in semi-closed
systems in which only a portion of suspended material is periodically removed. Gradual
increases in cBOD5 and COD were likely an indirect result of increased feed input
associated with increasing shrimp biomass. Temporal increases in cBOD5 could have been
associated with increased availability of substrates for microbial growth (e.g., NH3/NH4+
via deamination of uneaten feed, feces). cBOD5 values for both raceways appeared to
stabilize in weeks 6 and 7, after pressurize sand filtration during week 5. A similar trend
was shown for TSS, VSS and NH4+-N, but not for either NO2
-N or NO3
-N. This
stabilization appears to have coincided with increased frequency of sand filtration,
indicating that a substantial portion of cBOD5 could have been removed by mechanical
means. Concentrations of cBOD5 in excess of 29 mg l
1 are approximately three times
higher than levels approved for discharge for commercial shrimp farms in Texas (Samocha
and Lawrence, 1997; Samocha et al., 2004).
Chemical oxygen demand in both raceways showed a decrease mid-way through the
production trial (Fig. 5). Levels eventually increased by week 7 to those slightly below that
of week 2. Apparently, periodic pressurized sand filtration was able to reduce COD levels.
However TSS levels continued to increase throughout the study as loading rates associated
with increasing shrimp biomass were increased. Increases in COD above those capable of
being alleviated by mechanical filtration were eventually encountered. COD levels shown
in both raceways were similar to levels reported for outdoor ponds operated with no
discharge (Samocha et al., 2002). Natural waters seldom show COD concentrations in
excess of 100–150 mg l
1 (Boyd, 1990). Raceways in the present study showed COD
maxima of 1,990 and 2,430 mg l
1, substantially in excess of that shown for typical
aquaculture scenarios. For example, channel catfish ponds fed at a rate of 224 kg ha
1
showed a minimum COD level of 165 mg l
1 (Cole and Boyd, 1986). One approach to
management of COD in recirculating aquaculture systems could be the use of a peripheral
sequencing batch reactor (SBR) to reduce nitrogen and carbon in raceway sludge. This
device can be operated either aerobically or anaerobically for removal of nitrogen and
chemical oxygen demand, respectively, and has proven effective in laboratory-scale trials
438 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442

(Fontenot, 2004). Eventually, dissolved oxygen levels in both raceways decreased to levels
requiring supplementation of ambient air with pure oxygen. This, in combination with
increasingly frequent pressurized sand filtration, enabled raceways to maintain appropriate
DO concentrations for normal growth and survival of juvenile shrimp (Tables 2 and 4).

4.3. Evaluation of TSS and VSS

Declines in TSS and VSS (Figs. 6 and 7) after week 5 appeared to have been related to
periodic removal of suspended particulate matter via pressurized sand filtration. Filtration
removed 56 and 58% of TSS and 39% of VSS in RW-1 and -2, respectively. Because VSS
comprised about 56% of TSS, the potential for filtration to influence phytoplankton levels
and other organic constituents of the raceways is clearly demonstrated. If management of
NH3/NH4 is to largely be accomplished via assimilation by phytoplankton, and nitrifying
bacterial population in the water column substantial care should be given to establishment
of appropriate levels of TSS and VSS. Filtration of culture water at 120 mg l
1 TSS proved
useful in the present study in transforming raceways from autotrophy to a mixed trophic
situation in which both heterotrophy (manifested as increasing cBOD5) and chemoauto-
trophy were promoted.
The benefit of suspended organic material in shrimp culture systems has been
previously demonstrated by several authors (Hunter et al., 1987; Leber and Pruder, 1988;
Moss and Pruder, 1995; Otoshi et al., 2003). Moss et al. (1992) showed that in the presence
of water-column solids between 0.5 and 5.0 mm, shrimp growth rates increased 53% over
that attained in clear well water. In a subsequent study, Moss and Pruder (1995)
demonstrated rapid growth of L. vannamei in microcosm tanks containing a mean
particulate organic carbon (POC) concentration in excess of 6.98 mg l
1.

4.4. Carbon:nitrogen ratio

Although total carbon:nitrogen ratio (C:N) of culture water was not determined in the
present study, some inferences regarding this management factor can be made using feed
nitrogen content and cBOD5. Because shrimp are relatively poor assimilators of protein-
nitrogen (Muthuwani and Lin, 1996; Funge-Smith and Briggs, 1998; Thakur and Lin,
2003; Burford et al., 2003; Jackson et al., 2003) use of commercial feeds in high-density
intensive culture systems, especially those in which water exchange is limited and ponds/
raceways are lined, is questionable. Thakur and Lin (2003) estimated shrimp feed
accounted for 76–92% of nitrogen input in closed intensive culture systems. If, according
to these authors, only 23–31% of this nitrogen is assimilated as shrimp tissue, the rest is
available as primary amines for either growth of microorganisms (e.g., phytoplankton and
heterotrophic bacteria) or potentially transformed to NO2-N by chemoautotrophs. The low
FCR (<1.0:1) reported for the present study indicates that a high proportion of feed
nitrogen was probably incorporated into bacterial tissue via assimilation. This bacteria, in
turn, was consumed by shrimp reducing FCR. Increased raceway heterotrophy was
primarily indicated by a gradual increase in raceway cBOD5 to elevated levels (maxima of
29.88 and 28.60 mg l
1 in RW-1 and -2, respectively).
 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442 439

4.5. Oxygen supplementation

The use of oxygen to support culture of shrimp was required in the final week of culture,
once biomass density had exceeded 3.0 kg m
3. In order to reduce operational cost and
simplify overall system management, a reduction in biomass density (i.e., partial harvest)
could be considered. The objective of this strategy would be to maintain the culture system
at or near its carrying capacity to achieve an appropriate-sized juvenile for stocking into
ponds. The decision to harvest could be based upon some limiting factor such as oxygen. In
order to better evaluate this option, the cost:benefit of oxygen supplementation under the
present culture conditions would have to be determined.

4.6. Biosecurity and animal health

All samples sent for disease diagnosis were found free of pathogenic viruses.
Furthermore, no pathogenic bacterial infections were detected in the shrimp harvested
from either raceway. These results suggest that the use of viral-pathogen-free PL along
with water preparation and management practices were suitable to prevent viral/bacterial
pathogen disease outbreaks in this study.

5. Conclusions

This study showed that the nursery system described in this trial was adequate for high-
yield production of viral-pathogen-free juveniles L. vannamei with limited water exchange
under biosecure conditions using readily available commercial feeds, simple and relatively
simple management techniques. Design criteria for improvement of limited water exchange
production systems using phytoplankton for assimilation of dissolved nitrogen should
consider maintenance of adequate C:N to promote nitrification. Excessive C:N could be
regulated by either simple removal of organic material (e.g., TSS) by pressurized sand
filtration, foam fractionation or possibly by reduction of feed protein-nitrogen content.

Acknowledgements

This research was funded in part by a grant from the United States Department of
Commerce Marine Shrimp Farming Program CSREES Grants No. 2002-38808-01345,
Project R-9005 of the Texas Agricultural Experiment Station, Texas A&M University
System.

References

APHA, American Water Works Association, and Water Pollution Control Federation, 1995. Standard methods for
the examination of water and wastewater, 19th ed. American Public Health Association, Washington, DC,
USA.
Arbiv, R., van Rijn, J., 1995. Performance of a treatment system for inorganic nitrogen removal in intensive
aquaculture systems. Aquacult. Eng. 14, 189–203.
440 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442

Avnimelech, Y., Kochva, M., Diab, S., 1994. Development of controlled intensive aquaculture systems with a
limited water exchange and adjusted carbon to nitrogen ratio. Bamidgeh 46, 119–131.
Avnimelech, Y., 1999. Carbon/nitrogen ratio as a control element in aquaculture systems. Aquaculture 176, 227–
235.
Barg, U., Subasinghe, R., Willmann, R., Rana, K., Martinez, M., 1999. Towards Sustainable Shrimp Culture
Development: Implementing the FAO Code of Conduct for Responsible Fisheries. Fisheries Department. Food
and Agriculture Organization of the United Nations (FAO), Rome, Italy, pp. 3–6.
Boyd, C.E., 1990. Water Quality in Ponds for Aquaculture. Birmingham Publishing Company, Birmingham,
Alabama, pp. 274–282.
Boyd, C.E., Yoo, K.H., 1994. Hydrology and Water Supply for Pond Aquaculture. Chapman and Hall, New York,
pp. 439–449.
Browdy, C.L., Bratvold, D., Stokes, A.D., McIntosh, R.P., 2001. Perspectives on the application of closed
shrimp culture systems. In: Browdy, C.L., Jory, D.E. (Eds.), The New Wave, Proceedings of the Special
Session on Sustainable Shrimp Culture, Aquaculture, 2001, Baton Rouge, USA. The World Aquaculture
Society, pp. 20–34.
Burford, M.A., Williams, K.C., 2001. The fate of nitrogenous waste from shrimp feeding. Aquaculture 198, 79–
93.
Burford, M.A., Thompson, P.J., McIntosh, R.P., Bauman, R.H., Pearson, D.C., 2003. Nutrient and microbial
dynamics in high-intensity, zero-exchange shrimp ponds in Belize. Aquaculture 219, 393–411.
Chen, J.C., Chin, T.C., 1998. Joint action of ammonia and nitrite on tiger prawn, Penaeus monodon, postlarvae. J.
World Aquacult. Soc. 19, 143–148.
Chen, J.C., Lei, S.C., 1990. Toxicity of ammonia and nitrite to Penaeus monodon juveniles. J. World Maricult.
Soc. 21 (4), 300–306.
Cole, B.A., Boyd, C.E., 1986. Feeding rate, water quality, and channel catfish production in ponds. Progressive
Fish Culturist 81, 25–29.
Fontenot, Q., 2004. Biological treatment of shrimp aquaculture wastewater. US Mar. Shrimp Farming Ind. Briefs
10 (2), 6.
Funge-Smith, S., Briggs, M.R.P., 1998. Nutrient budgets in intensive shrimp ponds: implications for sustainability.
Aquaculture 164, 117–133.
Hopkins, J.S., Hamilton II, R.D., Sandifer, P.A., Browdy, C.L., Stokes, A.D., 1993. Effects of water exchange rate
on production, water quality, effluent characteristics and nitrogen budgets of intensive shrimp ponds. J. World
Aquacult. Soc. 24, 304–320.
Hunter, B., Pruder, G., Wyban, J., 1987. Biochemical composition of pond biota, shrimp ingesta, and relative
growth of Penaeus vannamei in earthen ponds. J. World Aquacult. Soc. 8, 162–174.
Jackson, C., Preston, N., Thompson, P.J., Burford, M., 2003. Nitrogen budget and effluent nitrogen components at
an intensive shrimp farm. Aquaculture 218, 397–411.
Komaros, M., Lyberatos, G., 1998. Kinetic modeling of Pseudomonas denitrificans growth and denitrification
under aerobic, anoxic and transient operating conditions. Water Res. 32, 1912–1922.
Komaros, M., Zafiri, C., Lyberatos, G., 1996. Kinetics of denitrification by Pseudomonas denitrificans under
growth conditions limited by carbon and/or nitrate or nitrite. Water Environ. Res. 68, 934–945.
Krom, M.D., Ellner, S., van Rijn, J., Neori, A., 1995. Nitrogen and phosphorus cycling and transformations
in a prototype ‘non-polluting’ integrated mariculture system, Eilat, Israel. Mar. Ecol. Prog. Ser. 118,
25–36.
Lancelot, C., Billen, G., 1985. Carbon-nitrogen relationships in nutrient metabolism of coastal marine ecosystems.
In: Jannasch, H.W., Williams, J.J.L. (Eds.), Handbook of Methods in Microbial Ecology. Lewis Publishers,
Boca Raton, USA, pp. 319–325.
Leber, K.M., Pruder, G.D., 1988. Using experimental microcosms in shrimp research: the growth enhancing effect
of shrimp pond water. J. World Aquacult. Soc. 19 (4), 197–203.
Lightner, D.V., Redman, R.M., Poulos, B.T., Nunan, L.M., Mart, J.L., Hasson, K.W., 1997. Status of the major
virus diseases of concern to the shrimp fanning industries of the Americas: known distributions, hosts and
available detection methods. In: Alston, D.E., Green, B.W., Clifford, H.C. (Eds.), Proceedings of the Fourth
Symposium of Aquaculture in Central America: Focusing on Shrimp and Tilapia, 22–24 April, Tegucigalpa,
Honduras, Asociación Nacional de Aquicultores de Honduras, pp. 36–48.
 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442 441

Lin, C.K., 1995. Progression of intensive marine shrimp culture in Thailand. In: Browdy, C., Hopkins, J.S. (Eds.),
Swimming Through Troubled Waters. World Aquaculture Society, Baton Rouge, Louisiana, USA, pp.
13–22.
Lotz, J.M., Lightner, D.V., 1999. Shrimp biosecurity: pathogens and pathogen exclusion. In: Bullis, R.A., Pruder,
G.D. (Eds.), Controlled and Biosecure Production Systems, Preliminary Proceedings of a Special Integration
of Shrimp and Chicken Models, 27–30 April. Sydney, Australia. World Aquaculture Society, pp. 70–72.
Moriarty, D.J.W., Cook, H.L., Hassan, R., Thanabal, M., 1987. Primary production and meiofauna in some
penaeid prawn aquaculture ponds at Gelang Patah. Malay Agric. J. 54, 37–51.
Moss, S.A., Pruder, G.D., Samocha, T.M., 1999. Environmental management and control: controlled ecosystem
and biosecure shrimp growout systems. In: Bullis, R.A., Pruder, G.D. (Eds.), Controlled and Biosecure
Production Systems, Preliminary Proceedings of a Special Integration of Shrimp and Chicken Models, 27–30
April. Sydney, Australia. World Aquaculture Society, pp. 87–91.
Moss, S.M., Pruder, G.D., Leber, K.M., Wyban, J.A., 1992. The relative enhancement of Penaeus vannamei
growth by selected fractions of shrimp pond water. Aquaculture 101, 229–239.
Moss, S.M., Pruder, G.D., 1995. Characterization of organic particles associated with rapid growth in juvenile
white shrimp, Penaeus vannamei Boone, reared under intensive culture conditions. J. Exp. Mar. Biol. Ecol.
187 (2), 175–191.
Muthuwani, V., Lin, C.K., 1996. Water quality and nutrient budget in intensive shrimp culture ponds. In:
Proceedings of World Mariculture Society Meeting, Bangkok, p. 270.
Otoshi, C., Arce, S.M., Moss, S.M., 2003. Growth and reproductive performance of broodstock shrimp reared in a
biosecure recirculating aquaculture system versus a flow-through pond. Aquacult. Eng. 29, 93–107.
Preston, N., Jackson, C., Thompson, P., Austin, M., Burform, M., 2000. Prawn farm effluent: composition, origin
and treatment. Fisheries Research and Development Corporation Final Report 95/162. FRDC, Canberra,
Australia.
Rubright, J.S., Harrell, J.L., Holcomb, H.W., Parker, J., 1981. Responses of planktonic and benthic communities to
fertilizer and feed applications in shrimp mariculture ponds. J. World Maricult. Soc. 12, 281–299.
Samocha, T.M., Benner, R., 2001. Use of intensive nursery systems in commercial shrimp production operation in
Texas. Ind. Briefs 7 (1), 1–3.
Samocha, T.M., Lawrence, A.L., 1997. Shrimp farms effluent water, environmental impact and potential treatment
methods, In: Keller, B.J., Park, P.K., McVey, J.P., Takayanagi, K., Hosoya, K. (Eds.), Water Effluent and
Quality, With Special Emphasis on Finfish and Shrimp Aquaculture. Proceedings of the 24th US–Japan
Aquaculture Panel Symposium, 8–10 October 1995. Corpus Christi, Texas. UJNR Technical Report No. 24,
Texas A&M University Sea Grant College Program TAMU-SG-97-102, pp. 33–57.
Samocha, T.M., Blacher, T., Cordova, J., De Wind, A., 2000. Raceway nursery production increases shrimp
survival and yields in Ecuador. Global Aquacult. Advocate 3 (6), 66–68.
Samocha, T.M., Hamper, L., Emberson, C.R., Davis, D.A., Mclntosh, D., Lawrence, A.L., Van Wyk, P.M., 2002.
Review of some recent developments in sustainable shrimp farming practices in Texas, Arizona and Florida. J.
Appl. Aquacult. 12 (1), 1–42.
Samocha, T.M., Gandy, R.L., McMahon, D.Z., Mogollon, M., Smiley, R.A., Blacher, T.S., de Wind, A., Figueras,
E., Velasco, M., 2003. The role of shrimp nursery systems to improve production efficiency of shrimp farms.
In: Jory, D.E. (Ed.), Responsible Aquaculture for a Secure Future. Proceedings of a Special Session on Shrimp
Farming, Baton Rouge, Louisiana 70803, USA. World Aquaculture 2003. The World Aquaculture Society, pp.
170–195.
Samocha, T.M., Lopez, I.M., Jones, E.R., Jackson, S., Lawrence, A.L., Sharma, V.K., 2004. Characterization of
shrimp farms’ effluent water and the effect on receiving water. Aquacult. Res. 35, 321–339.
Sandifer, P.A., Stokes, A.D., Hopkins, J.S., 1991. Further intensification of pond shrimp culture in South Carolina.
In: Sandifer, P.A. (Ed.), Shrimp culture in North America and the Caribbean. Advances in World Aquaculture
4. The World Aquaculture Society, Baton Rouge, LA, pp. 84–95.
SPSS, 1999. Statistical Procedures for the Social Sciences, version 10, SPSS Inc. Headquarters, 233 S. Wacker
Drive, 11th floor, Chicago, Illinois 60606.
Sturmer, L.L., Samocha, T.M., Lawrence, A.L., 1992. Intensification of penaeid nursery systems. In: Fast, A.W.,
Lester, L.G. (Eds.), Culture of Marine Shrimp: Principles and Practices. Elsevier Science Publishing
Company, Amsterdam, The Netherlands, pp. 323–346.
442 J.M. Cohen et al. / Aquacultural Engineering 32 (2005) 425–442

Thakur, D.P., Lin, C.K., 2003. Water quality and nutrient budget in closed shrimp (Penaeus monodon) culture
systems. Aquacult. Eng. 27, 159–176.
Timmons, M.B., Losordo, T.M., 1994. Aquaculture Water Reuse Systems: Engineering Design and Management.
Elsevier, Amsterdam.
Timmons, M.B., Ebeling, J.B., Wheaton, F.W., Summerfelt, S.T., Vinci, B.J., 2001. Recirculating Aquaculture
Systems, USDA-CSREES NRAC Publication no. 01-002, pp. 650.
Trott, L.A., Alongi, D.M., 2000. The impact of shrimp pond effluent on water quality and phytoplankton biomass
in a tropical mangrove estuary. Mar. Poll. Bull. 40, 947–951.
van Rijn, J., Rivera, G., 1990. Aerobic and anaerobic biofiltration in an aquaculture unit-nitrite accumulation as a
result of nitrification and denitrification. Aquacult. Eng. 9, 217–234.
Viadero, R.C., Noblett, J.A., 2002. Membrane filtration for removal of fine solids from aquaculture process water.
Aquacult. Eng. 26 (3), 151–169.
Wang, J.K., 2003. Conceptual design of a microalgae-based recirculating oyster and shrimp system. Aquacult.
Eng. 28, 37–46.
Zhu, S., Chen, S., 2001. Effects of organic carbon on nitrification rate in fixed film biofilters. Aquacult. Eng. 25, 1–
11.