This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elseviers archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights
 Author's personal copy

w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 8 4 3 e2 8 5 0

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Use of UASB reactors for brackish aquaculture sludge

digestion under different conditions

Natella Mirzoyan, Amit Gross*

Department of Environmental Hydrology and Microbiology, Zuckerberg Institute for Water Research, Blaustein Institutes for Desert Research,
Ben-Gurion University of the Negev, Midreshet Ben-Gurion 84990, Israel

article info abstract

Article history: Treatment and disposal of high volume of salty waste production in recirculating aqua-
Received 6 November 2012 culture systems (RASs) is a major challenge and the sludge is often a source of environ-
Received in revised form mental pollution and salinization of receiving soils and water bodies. Anaerobic digestion
28 January 2013 is an efficient mean for the treatment of wastes of different origins and might serve a
Accepted 22 February 2013 useful tool for the reduction of salty aquaculture discharge load. Use of an upflow anaer-
Available online 13 March 2013 obic sludge blanket (UASB) reactor for digestion of brackish aquaculture sludge from RASs
under different C:N ratios, temperatures, and hydraulic retention times demonstrated high
Keywords: removal efficiencies of over 92% as volatile solids (VS), 98% as chemical oxygen demand
Aquaculture and 81% as total suspended solids in all reactors. Methane production topped 7.1 mL/gVS d
Brackish sludge and was limited by low C:N ratio but was not influenced by temperature fluctuations. The
Methane treated liquid effluent from all reactors was of sufficient quality for reuse in the RAS,
Salinity leading to significant water recycling and saving rates. UASB may be an attractive solution
UASB for brackish sludge management in RASs.
2013 Elsevier Ltd. All rights reserved.

1. Introduction Mirzoyan et al., 2012). As a result of intensive culture, the

While the main environmental risk associated with discharge
of low-salinity waste is pollution caused by high nutrient
loads, high-salinity waste production additionally leads to
salinization of soils and receiving water bodies when dis-
charged into the environment. Global water scarcity accom-
panied by high rates of population growth and climate change
has necessitated sustainable management of water resources,
including the reduction of saline discharge.
Inland brackish and marine water aquaculture production
is a continuously expanding source of salty waste (FAO, 2008).
Even in controlled recirculating aquaculture systems (RASs),
which produce a smaller environmental footprint and use less
water, salt and energy compared to traditional aquaculture;
between 25 and 50% of applied fish feed is estimated to
accumulate in the form of sludge (Timmons and Ebeling, 2007;
actual volume of salty solid waste produced for the growth of
100 tons of fish is comparable to that of a mid-sized town
(Hardy, 2001).
Salty waste is handled in the same manner as low-salinity
waste, e.g., after separation it is usually discharged from the
RAS into either the local sewer system or a decentralized
treatment unit (Timmons and Ebeling, 2007). However,
elevated salt levels limit sludge-management options as con-
ventional disposal methods are less efficient and land spread is
prohibited due to high sludge salinity. Disposal and discharge
of this sludge into the environment leads to salinization of
soils and receiving water bodies as a result of leaching.
There is growing interest in inland culture of aquatic
animals in RASs, salt conservation and the release of saline
effluent from these processes should be addressed to prevent
the continually increasing negative impact. Reducing

* Corresponding author. Tel.: 972 8 6596896; fax: 972 8 6596909.

E-mail addresses: natella.mirzoyan@weizmann.ac.il (N. Mirzoyan), amgross@bgu.ac.il, amgross@exchange.bgu.ac.il (A. Gross).
0043-1354/$ e see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.watres.2013.02.050
 Author's personal copy

2844 w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 8 4 3 e2 8 5 0

discharge is probably the most feasible approach to address-

ing this problem. Therefore, feasible treatment methods for
sludge produced in RASs (specifically from brackish opera-
tions) have been identified as crucial and have been the sub-
ject of several recent reports (Gebauer, 2004; Gebauer and
Eikebrokk, 2006; Sharrer et al., 2007; Mirzoyan et al., 2008,
2012; Tal et al., 2009).
The use of anaerobic digestion (AD) is a relatively new
approach to aquaculture sludge treatment. Anaerobic diges-
tion is biological degradation of organic matter by microbes
under anaerobic conditions. This is a natural process, carried
out by facultative and obligatory anaerobic bacteria. There are
four main stages in the AD process: hydrolysis, fermentation,
acetogenesis and methanogenesis and under ideal conditions,
the ultimate product of this process is biogas composed of
methane and carbon dioxide with small levels of hydrogen
sulfide and ammonia (Appels et al., 2008; Appels et al., 2011).
While the bulk of the literature relates to freshwater
aquaculture sludge digestion (Kugelman and Van Gorder, 1991;
Lanari and Franci, 1998) and, to some extent, marine sludge
digestion (Arbiv and van Rijn, 1995; McDermott et al., 2001; Klas
et al., 2006; Tal et al., 2009), there has been only limited
research into brackish aquaculture sludge (BAS) digestion Fig. 1 e Schematic representation of UASB reactor.
(Gebauer, 2004; Gebauer and Eikebrokk, 2006; Mirzoyan et al.,
2008). Moreover, whereas marine and freshwater environ-
ments are relatively well-defined, the definition of brackish environmental footprint with respect to, for example, nutrient
water comprises a wide range of water qualities and compo- and salt discharge.
sitions, because the source of the water consists of wells in However, there is still very little quantitative information
different locations. Treating BAS may therefore be challenging on BAS treatment by UASB reactors. Moreover, there are gaps
and caution should be exercised when comparing studies. in our understanding of the effects of different operational
Gebauer (2004) and Gebauer and Eikebrokk (2006) were conditions, such as hydraulic retention time (HRT), tempera-
among the first to demonstrate the potential of anaerobic ture and C:N ratio, on the efficiency of aquaculture sludge
digestion of BAS using a continuously stirred tank reactor. digestion in any type of digestion system used to date. The
Despite its high efficiency, the system required significant aims of the current study were to investigate: a) the effect of
financial investment to cover the heating, stirring and con- different conditions on BAS digestion efficiency in UASB re-
struction costs. Mirzoyan et al. (2008) characterized BAS from actors; b) the quality of the treated effluent and its potential
several RASs and suggested the use of an upflow anaerobic reuse in the RAS; c) the rates, quality and quantity of biogas
sludge blanket (UASB) reactor as a potentially suitable method production under these conditions, for its potential use as an
for BAS digestion. energy source in fish farms.
Upflow anaerobic sludge blanket is one of the most
commonly used anaerobic sludge digesters (Appels et al.,
2008). The reactor consists of a tank, part of which is filled 2. Materials and methods
with an anaerobic sludge blanket composed of active bacterial
species (Fig. 1). The sludge in the bottom of the tank flows up Brackish aquaculture sludge digestion and methane produc-
and through the anaerobic sludge blanket. An inverted cone tion were analyzed in four UASB reactors (R1eR4) located in a
settler at the top of the digester allows efficient solideliquid temperature-controlled greenhouse and operated under
separation. The sludge is immobilized by the formation of various environmental conditions (Table 1). Reactors R1 and
large microbial aggregates that form distinct granules with a
high volatile suspended solids content and specific microbial
activity. Upflow anaerobic sludge blanket reactors perform
best when the suspended solids concentration is less than 3% Table 1 e Operational conditions of UASB reactors
treating brackish aquaculture sludge.
and when it has good granulation capability (Appels et al.,
2008; Marchaim, 1992). Reactor Hydraulic Temperature C:N ratio Volume (L)
retention
Some of the major advantages of the UASB reactor are its
time (d)
simplicity, low operational costs and high efficiency
(Marchaim, 1992), which make this method a more attractive R1 8 Ambient 15.4 8
option for fish farmers. Moreover, successful application of R2 8 30  C 15.4 8
R3 6 Ambient Cotton 6
this method may result in enhanced water and salt recycling
addition
as well as biogas production, which may further improve RAS
R4 6 Ambient 15.4 6
operation, lessen its operational costs and reduce its
 Author's personal copy
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 8 4 3 e2 8 5 0
R2 were 18 cm in diameter and 45 cm high with an operating
volume of about 8 L. Reactors R3 and R4 were 13 cm in
diameter and 48 cm high with an operating volume of about
6 L. Reactors R1 and R2 were operated for 335 d, reactors R3
and R4 for 308 d.
All reactors were inoculated with 500 mL of sludge from
an experimental UASB reactor that had been operating for
120 d with BAS (Mirzoyan et al., 2008). Reactors were batch-
fed once a day with 1000 mL of fresh BAS (0.4% total sus-
pended solids (TSS) from the R.A.M. commercial brackish
water RAS (located 50 km south of Beer Sheva, Israel, 30 58
N, 34 43 E). Thus, an average HRT of 8 d for reactors R1 and
R2 and 6 d for reactors R3 and R4 was provided. Reactors R1,
R3 and R4 were operated at ambient temperature, whereas
reactor R2 was held at a fairly constant 30  C using an
aquarium heater with a thermostat. The average ambient
temperature fluctuated between 11 and 27  C in winter, 12
and 30  C in spring, 21 and 30  C in summer and 17 and 28  C
in autumn, based on onsite continuous temperature moni-
toring by data-logging thermometer (Cole Parmer). Reactors
R1, R2 and R4 were operated at an average C:N ratio of 15.4
(natural ratio for raw sludge in the current study). The C:N
ratio in reactor R3 was elevated by the addition of 100 g
cotton wool (low-soluble carbon source). Every 4 months, the
cotton was removed from the reactor, washed and weighed
for a determination of mass loss, and 100 g of fresh cotton
wool was added.
To check operational stability, at least twice a month,
250 mL sludge was collected from each reactor and charac-
terized. Briefly, sludge samples (raw and digested) were
analyzed for pH, electrical conductivity (EC) and redox po-
tential (ORP) by laboratory meters. Total suspended solids,
volatile solids (VS), chemical oxygen demand (COD), 5-day
biological oxygen demand (BOD5) and total alkalinity (TA)
were analyzed according to standard methods (APHA, 1998).
More details on the methods used can be found in Mirzoyan
et al. (2008).
The difference in sludge parameters between the feed
sludge and digested sludge after 2 weeks was designated
reduction and was not representative of the overall reactor
removal rates, as the reactor was not mixed and sludge was
sampled only from the sludge bed. However, this estimation is
an important measure of the stability of operation.
The sludge volume in the UASB reactor was monitored
periodically as an indication of sludge-mass accumulation or
reduction. No solids were discharged other than for sampling.
A sludge mass balance was conducted at the end of the
experiment by collecting the remaining sludge and subjecting
it to volume and weight determinations. Sludge removal was
calculated in terms of TSS, COD and VS for the entire period of
operation.
Following the introduction of raw sludge at the bottom of
the UASB reactors, a similar volume of treated effluent was
released from the top sampling port of the reactor. The ef-
fluents were collected periodically, 20 times during the
experiment, and analyzed for pH, EC, total ammonia nitrogen
(TAN), TSS, BOD5, turbidity and NO
2 by standard methods
(APHA, 1998). Analysis of NO 3 was conducted by the second-
derivative method (Ferree and Shannon, 2001) and H2S was
assessed using Aquaquant 1.14416 test kits (Merck).
2845
The biogas produced in each reactor was volumetrically
collected and analyzed for the presence of methane by gas
chromatography in a Varian GC 3800 using a thermal con-
ductivity detector and a Hayesep Q packed column (Varian).
The differences between raw and treated sludge quality
were compared by repeated-measurement analysis of vari-
ance (RM-ANOVA) followed by post-hoc test when needed
using the SigmaStat 3.1 package (SPSS, 1997).
3. Results
3.1. Raw sludge characteristics
To assess the UASB reactors potential for BAS digestion and
methane production under different conditions, the compo-
sition of the raw sludge was analyzed for parameters that
most affect sludge digestion (Table 2). Our findings were in
agreement with previous studies on aquaculture sludge
characteristics (Gebauer, 2004; Mirzoyan et al., 2012) and
except for a relatively low C:N ratio, the raw BAS was found
suitable for digestion in the UASB reactors.
3.2. Sludge digestion
Digestion of BAS was monitored during the reactors opera-
tion to determine reactor performance under different con-
ditions, and the measured parameters are given in Table 3.
The pH of the UASB reactors varied between 6.9 and 7.7 on
average, with a maximum of 8.1. Irrespective of the operating
conditions, all four reactors operated under stable EC and
anaerobic conditions. The ORP was highest in reactor R2 (high
temperature treatment) and lowest in reactor R4 (low HRT
treatment).
Reduction in BOD5 and VS was checked biweekly throughout
operation to determine the stability of the digestion process
over time. Similar reductions (P < 0.05) were recorded in all
four reactors (Table 3, Figs. 2 and 3). The range of VS reduction
was in agreement with others findings (Lanari and Franci,
1998; Gebauer and Eikebrokk, 2006). While VS content of
feed sludge was fairly constant, the VS content of digested
sludge decreased over time, reaching its lowest level by the
end of experiment in all four reactors, regardless of opera-
tional conditions. BOD5 reduction, however, fluctuated over
time corresponding to the fluctuating BOD5 content of feed
Table 2 e Characteristics of raw brackish aquaculture
sludge from a recirculating aquaculture farm (data are
average standard error).
Parameters
pH 7.5  0.1
Electrical conductivity (mS/cm) 4.5  0.1
Redox potential (mV) 117  30
C:N 15.4  2.8
5-day biochemical oxygen demand (g/kg) 108  13
Volatile solids (g/kg) 642  21
Total solids (g/L) 3.8  0.2
Salinity (g/L) 2.6  0.1
 Author's personal copy

2846 w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 8 4 3 e2 8 5 0

Table 3 e Performance of UASB reactors digesting

brackish aquaculture sludge. Reactors were operated
under the conditions detailed in Table 1.
Parameters R1 R2 R3 R4

BOD5 (DW) reduction (%) 28  10 40  7 42  18 31  9

VS reduction (%) 22  4 29  3 27  9 20  6
COD removal (%) 99.6 99.3 97.7 98.4
Total suspended 91.9 88.7 80.5 91.5
solids removal (%)
VS removal (%) 97.8 97.7 92.0 94.8
Biogas production 11.1 14.6 15.5 6.0
rate (mL/gVS d)
Biogas production 4.2 5.5 5.9 2.3
rate (mL/gCOD d)
Methane concentration 15  0.8 4  0.0 53  9.2 2  2.5
(% in biogas)
SMP rate (mL/gVS d) 1.4 0.5 7.1 0.1
Fig. 3 e Concentration of 5 d biochemical oxygen demand
SMP rate (mL/gCOD d) 0.71 0.24 3.55 0.04 of sludge over time in UASB reactors operating under
pH 7.1  0.1 7.7  0.1 6.9  0.0 7.2  0.1 different environmental conditions (operational conditions
Electrical conductivity 4.6  0.1 4.8  0.1 4.7  0.1 5.0  0.1 are specified in Table 1).
(mS/cm)
Redox potential (mV) 240  18 232  23 291  7 304  7
Total alkalinity 457  23 489  28 646  34 455  47
(mg/L as CaCO3) cotton). Overall, a high efficiency of sludge digestion was
VFA (mg/L) <5 <5 <5 <5
observed in all four reactors.
BOD5: 5-day biochemical oxygen demand; DW: dry weight; VS:
volatile solids; COD: chemical oxygen demand; SMP: Specific 3.3. Effluent quality
methane production.

During the entire operation of the reactors, 332 L of treated

sludge. Interestingly, there were no differences in the perfor- effluent was released from reactors R1 and R2, and 305 L from
mance of the different reactors with respect to BOD5 and VS reactors R3 and R4. Quality parameters of the treated UASB
reduction. effluent are given in Table 4. The effluent pH was close to
Sludge removal was determined at the end of the experi- neutral and its salinity was similar to that of the source water.
ment as the difference between the cumulative mass of TSS, Of all the parameters tested, turbidity and concentrations of
VS and COD introduced throughout the study and the total sulfide, nitrite and TAN were high for direct use in aquaculture.
mass of these parameters in the reactors at the end of the
study. Chemical oxygen demand removal in all four reactors
3.4. Biogas production
(Table 3) was higher than in previous reports (Kugelman and
Van Gorder, 1991; McDermott et al., 2001; Gebauer and
Biogas volume, produced in UASB reactors fluctuated over
Eikebrokk, 2006). Volatile solids removal was high, reaching
time and on an average was similar in all treatments except
up to 98%, and TSS removal in reactors R1, R2 and R4 was
slightly higher than in reactor R3 (operating with added

Table 4 e Effluent quality (data are average standard

error) of UASB reactors digesting brackish aquaculture
sludge. Reactors were operated under the conditions
detailed in Table 1.
Parameters R1 R2 R3 R4

5-day biochemical 146  15 130  20 153  30 94  17

oxygen demand
(mg/L)
Sulfide (mg/L) 2.2  0.4 4.0  0.9 5.6  0.9 2.4  1.0
Turbidity (NTU) 124  19 85  11 201  13 140  14
Total suspended 201  25 185  36 123  23 94  7
solids (mg/L)
pH 7.6  0.0 7.7  0.1 7.2  0.0 7.5  0.0
Electrical conductivity 4.8  0.1 5.1  0.1 4.9  0.1 5.0  0.9
(mS/cm)
NO2eN (mg/L) 0.3  0.2 0.2  0.1 0.4  0.3 0.7  0.4
Total ammonia 17.7  1.4 19.9  1.3 11.7  1.8 17.1  1.1
Fig. 2 e Concentration of volatile solids over time in UASB
nitrogen (mg/L)
reactors operating under different environmental
NO3eN (mg/L) 0.2  0.1 0.3  0.1 0.5  0.1 0.5  0.2
conditions (operational conditions are specified in Table 1).
 Author's personal copy
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 8 4 3 e2 8 5 0
for treatment R4 ( p < 0.05) (Fig. 4). The highest biogas pro-
duction rate was recorded in reactors R2 and R3, averaging
15 mL/gVS d, while the lowest was observed in reactor R4
(Table 3). Despite the higher biogas production in reactor R2,
the specific methane production (SMP) rate was 0.5 mL/gVS d,
markedly lower than in reactors R1 and R3 (Table 3), due to
lower methane (gas of interest) concentration in biogas pro-
duced in R2. The highest methane concentration was
observed in reactor R3, averaging 53%. Reactor R4 produced an
average of only 2% methane in biogas, resulting in SMP rate of
0.1 mL/gVS d. The biogas production was also reported in
terms of COD as in many other reports. As expected, a pattern
similar to that of the VS was exhibited (Table 3).
4. Discussion
4.1. Sludge quality and digestion
The feed sludge was of suitable quality for anaerobic digestion
and methane production, except for its low C:N ratio.
Regardless of this low ratio, high sludge-digestion efficiencies
were observed in all UASB reactors with no operational fail-
ures. This was reflected in part by the operation stability,
including fairly stable and favorable pH and redox potentials
during the entire experimental period. The pH stability was a
result of both the high alkalinity of the sludge (Table 3)
(Ahring, 2003; Marchaim, 1992) and the acclimatization period
provided prior to initiating the experiment (Angelidaki et al.,
2003). The latter was also responsible for stable and high
BOD5 and VS reduction rates without salt inhibition of the
process (Omil et al., 1996). Given the measured removal rates
and assuming a full-scale UASB reactor at a farm such as
R.A.M., annual production of TSS will be 1.7 tons in reactors R1
and R4, 2.4 tons in reactor R2 and 4.1 tons in reactor R3,
instead of the currently produced 21.9 tons. A lower volume of
salty sludge would simplify its further handling and man-
agement toward reduced environmental impact, specifically
in terms of reductions in nutrient and salt discharge.
Fig. 4 e Volume of biogas production over time in UASB
reactors operating under different environmental
conditions (operational conditions are specified in Table 1).
2847
4.2. Effect of environmental conditions on AD of BAS
Currently, UASB is one of the most commonly used anaerobic
sludge digesters (Ahring, 2003), because its operation is asso-
ciated with low risks of operational failures and environ-
mental pollution (e.g. odors, noises, aerosols, insects or
worms) and low economical requirements (e.g. land and en-
ergy usage, constructional costs). UASB reactors perform best
when the suspended solids concentrations is less than 3%
(Marchaim, 1992), which is the case of BAS from RAS, and high
volumes of gas can be produced. Factors that affect the AD of
sludge in UASB include wastewater characteristics such as
temperature, C:N ratio and HRT (Ahring, 2003).
4.2.1. Effect of temperature stability
Temperature is one of the main environmental factors influ-
encing AD and even relatively small fluctuations seem to
affect UASB efficiency until adaptation occurs (Angelidaki
et al., 2003; Alvarez and Liden, 2008). Heating the reactors to
maintain stable temperatures requires financial inputs, and in
warmer climates (such as in Israel) the AD in fluctuating (and
still warm) temperatures can provide a feasible alternative.
Indeed, the digestion efficiency in terms of COD, TSS and VS
was similar for reactors R1 (operating under ambient tem-
peratures) and R2 (operating under a fairly constant temper-
ature of 30  C, with only rare, slight fluctuations), suggesting
no additional heat requirements if AD is used for the sole goal
of solid waste removal. Interestingly, higher biogas produc-
tion was observed in reactor R2, but due to its lower methane
concentration, the SMP was lower than in reactor R1. Alvarez
and Liden (2008) reported that in a temperature range between
11 and 25  C, forced cyclic temperature fluctuations result in a
higher rate of volumetric biogas production and higher
methane yield than in a digester operating at the isothermic
temperature of 18  C. Moreover, a single temperature drop
from 35 to 18  C reduced biogas production rate but, unex-
pectedly, increased methane production as well (Alvarez and
Liden, 2008). Considering the ambient temperature case
(reactor R1) in the current study as a cyclic temperature fluc-
tuation similar to Alvarez and Lidens (2008) study, the higher
methane production in R1 compared to R2 may be a result of
the lower production rates of CO2 (the other component of
biogas) at lower temperature. This in turn is an outcome of
reduced hydrolysis of complex organics and, therefore,
decreased acidogenesis (Masse et al., 2003), supported by
microbiological studies of archaeal/methanogenic commu-
nity structure in R1 reactor (Mirzoyan, 2009).
It was not possible to compare general gas production
since the R2 temperature was far higher than the ambient
temperature range for most of the experiment.
4.2.2. Effect of C:N ratio
The C:N ratio affects both the efficiency and pathway of
organic matter degradation in UASB reactor. The C:N ratio of
BAS in the current study was 15.4, lower than the recom-
mended optimal 21 to 32 for anaerobic digestion (Marchaim,
1992; Angelidaki et al., 2003). To increase the C:N ratio, cot-
ton wool was added to the reactor R3. Cotton has low solubi-
lity and is a slowly degradable cellulose-based carbon source
which is digested by the microbes (Soares et al., 2000). Its
 Author's personal copy
2848 w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 8 4 3 e2 8 5 0
efficiency as an alternative carbon source has been success-
fully established in anaerobic denitrification reactors (Soares
et al., 2000), but its effectiveness as a supplemental carbon
source for methanogenesis in UASB reactors has never been
studied. The main advantage of using cotton is its on mi-
crobial demand availability rather than being dissolved
directly into the reactor, thus minimizing the chance of
inaccurate dosing. This dictates higher carbon-use efficiency
as well as a lower environmental footprint. Decreased envi-
ronmental footprint is additionally maintained by skipping
the use of electrical power necessary for successful mixing
and dosing operations practiced with other commonly used
carbon sources. Moreover, cotton is a significantly cheaper
carbon source than conventionally used sources such as al-
cohols and citrate.
In the current study, the elevated C:N ratio did not increase
the efficiency of solids removal in the UASB reactors, instead,
biogas production and methane concentration were positively
correlated with cotton addition and were higher in reactor R3
(operated under a high C:N ratio) than in reactor R4. The
average daily digestion of cotton was 0.055 g/L, suggesting that
each 1 g added cotton resulted in the production of an addi-
tional 58 mL of biogas, of which 31 mL was methane. The
observed quantity of biogas/methane was lower than the
theoretical yield expected from the digestion of 1 g cotton,
most probably because of carbon loss due to assimilation by
the biomass, and aggregate formation in the reactor
(OSullivan et al., 2005), leading to inaccurate estimation of the
cotton mass remaining in the reactor. Moreover, cotton has
been reported to absorb gasses in limited amounts when used
in reactors (Soares et al., 2000), and the incomplete mass bal-
ance might therefore be a result of observed biogas yields that
were lower than what was actually occurring in reactor R3.
However, it should be stressed that the methane content in
biogas is not supposed to be affected by any of these inac-
curacies and as expected, carbon addition increased methane
content in biogas to 53% on average. This is a result of higher
volatile fatty acid (VFA) production (indicated by lower alka-
linity in this reactor compared to R4) and VFA consumption by
aceticlastic methanogens (supported by microbiological
studies of archaeal/methanogenic community structure in a
reactor; Mirzoyan, 2009). Thus, addition of a carbon source to
the reactor can be an efficient means of increasing biogas
production by altering methane production pathways toward
aceticlastic methanogenesis.
4.2.3. Effect of HRT
Hydraulic retention time influences the biochemical and
physical properties of sludge, by influencing the balance be-
tween hydrolysis of different polymers during primary
digestion and conversion of acid products into biogas (Ahring,
2003). Specifically, the occurrence of methanogenesis, as well
as the microbial composition in reactors (assuming a suitable
substrate), are mainly a function of HRT and temperature
(Halalsheh et al., 2005). Increasing HRT has major economic
implications as it requires a larger infrastructure. A study on
activated sludge denitrification reactors using fish sludge as
an intrinsic electron donor suggested that a HRT of less than
10 d is optimal for the denitrification process and prevents
significant washout of bacteria from the reactor (Klas et al.,
2006). Based on this preliminary information, the effect of
HRT on BAS digestion was checked in reactors R1 and R4,
operating with HRTs of 8 and 6 d. While no differences in
digestion efficiencies were observed, the volumetric biogas
and methane production in reactor R1 was higher than in
reactor R4 (Table 3). The low HRT might cause washout of
methanogenic species, in accordance with the conclusions
drawn by Halalsheh et al. (2005) and Pind et al. (2003). How-
ever, a HRT of 6 d was sufficient for the goal of reducing sludge
volume and strength, and is therefore recommended for
economic reasons.
4.3. Effluent quality
Of all the parameters tested, only turbidity, sulfide, nitrite and
TAN concentrations of the treated effluent were high for
direct use in aquaculture. The introduction of sulfide into the
ponds is of particular concern, as it is extremely toxic for
aquatic organisms at concentrations ranging from 0.01 to
0.05 mg/L (Boyd and Tucker, 1998). However, taking into
consideration that the effluent is diluted by four orders of
magnitude (104) when recycled back to the ponds, the final
sulfide concentration in the culture pond will be lower than
the suggested safe level of 0.002 mg/L (Boyd and Tucker, 1998).
The dilution alone would also be sufficient to decrease nitrite
and TAN concentrations, as well as turbidity.
Moreover, if the effluent can be recycled into the nitrifica-
tion reactor, oxidation and nitrification are expected to
remove the excess TAN, nitrite and sulfide. This approach was
confirmed in a separate study on a zero discharge RAS in
which UASB effluent was recycled into the RAS nitrification
unit for over a year (data not shown). The higher volume of
reclaimed salty discharge (12.5e16.7% of water volume daily)
in the current conditions presents two distinct advantages: a
reduction of salt released to the environment and a decrease
in water usage by the farmer, leading to both environmental
and economic benefits.
4.4. Effect of salinity on BAS digestion in UASB reactors
One of the main concerns in anaerobic digestion of brackish/
marine aquaculture sludge is the negative effect of high salt
concentration on the digestion process (Kugelman and Van
Gorder, 1991; Gebauer, 2004; Gebauer and Eikebrokk, 2006).
Salt toxicity has been found to be predominantly determined
by the elevated concentrations of cations, such as Na, Ca2
and Mg2, and anions such as sulfates, nitrates, nitrites, and
ammonia (Chen et al., 2008).
In the current study, the nitrates and nitrites were virtually
absent from the BAS (Table 4), and high sulfate levels were
mitigated by high COD concentrations (Mirzoyan et al., 2008),
suggesting no inhibitory effect of these compounds on
anaerobic digestion.
The inhibitory effects of both sodium and ammonia
(a result of protein breakdown) are attributed to elevated
VFA concentrations in anaerobic reactors. However, in the
current study, although present, both the highest sodium
and the highest ammonia concentrations in the BAS were
lower than inhibitory concentrations reported elsewhere
(Kugelman and Van Gorder, 1991; Gebauer, 2004; Gebauer
 Author's personal copy
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 8 4 3 e2 8 5 0
and Eikebrokk, 2006). In addition, no VFA accumulation was
observed (Table 3).
The only compounds of concern for anaerobic digestion in
the current study were Ca2 and Mg2, which were present in
the sludge at levels that might be inhibitory for granulation
(Mirzoyan et al., 2008). Lower granulation rates in the UASB
reactor can lead to lower biogas-production rates. Interest-
ingly, the granulation in the UASB reactor was not reduced.
This may be explained by the complex antagonistic effects of
different cations in the reactors and/or acclimatization of
microbial groups to the ionic composition. Overall high
removal rates in UASB reactors suggest no operational failures
as a result of elevated salt content in aquaculture sludge. The
digestion of brackish/marine sludge in different anaerobic
systems was reported to be successful when the microbial
community was given time to acclimate (Omil et al., 1996), due
to the presence of microbial species that are tolerant to high
salt concentrations.
4.5. Biogas production
The biogas/methane production in the UASB reactors was
influenced by environmental conditions. A high C:N ratio
increased methane-production rates, whereas low HRT
and temperature stability decreased them. The methane-
production rates in the current study were lower than previ-
ously reported values for BAS (Gebauer, 2004; Gebauer and
Eikebrokk, 2006; Mirzoyan et al., 2008). Interestingly, a previ-
ous study (Mirzoyan et al., 2008) examining the digestion of
brackish sludge from the same source and by the same UASB
reactor demonstrated higher biogas/methane production
rates, comparable with reports of Gebauer (2004) and Gebauer
and Eikebrokk (2006).
This lower biogas production can be attributed to lower
TSS content of the sludge, which was a result of failure to
concentrate solids in the commercial RAS from where the
samples were collected prior to disposal, whereas during the
previous study by Mirzoyan et al. (2008), the concentration
was successfully performed. Sludge thickening should be
considered for optimal anaerobic digestion of BAS.
5. Summary and conclusions
Anaerobic digestion can efficiently reduce saline aquaculture
sludge volume and its potential negative impact on the envi-
ronment, while enhancing overall water reuse. Here we suc-
cessfully demonstrate that UASB reactors can achieve stable
operation and very high digestion efficiencies of BAS under its
natural C:N ratio, 6e8 d HRT, and stable and ambient tem-
peratures. Overall, this practice reduces the environmental
footprint of BAS by improving water recycling and lowering
nutrient and salt discharge into the environment (e.g. soils
and water bodies).
Digestion of BAS in UASB reactors has additional advan-
tages for RASs, such as lower cost of sludge transport from the
facility to another point of disposal or reuse, which is a major
factor in the cost of sludge management (Reed et al., 1995).
The possibility of completely reutilizing the treated effluent in
RASs promises further independence of aquaculture systems
2849
from their traditional environment (i.e. large water bodies)
as it decreases water and salt use, as well as allowing for
additional decreases in energy (heating and pumping)
expenditure.
Lastly, it should be mentioned that under the current
tested conditions, methane-production rates were very low
and cannot be considered a significant energy source. Further
improvement in biogas quantity and quality can be achieved
by thickening the sludge, a practice that is already common in
RASs to reduce water usage. Increasing the C:N ratio is also
expected to significantly enhance methane production.
Optimization, economic evaluation and installation of
large-scale UASB reactors in intensive RASs must follow in
order to validate their feasibility for brackish RASs.
Acknowledgments
This study was funded by the USeIsrael Binational Agricul-
tural Research And Development Fund, the University of
Maryland (MB-8707-04), and the Rosenzweig-Coopersmith
Foundation (RCF). We are especially grateful to Yulia Rafaelov,
Anya Mamontov, Ofer Guy, Michael Mischurow and Doron
Shainer for their technical assistance and laboratory analyses.
Lastly, we sincerely thank Nir Calif from the R.A.M. fish farm
for his assistance with the sludge sources.
references
Ahring, B.K., 2003. Biomethanation I and II. Springer-Verlag,
Berlin, Germany.
Alvarez, R., Liden, G., 2008. The effect of temperature variation on
biomethanation at high altitude. Bioresource Technology 99
(15), 7278e7284.
Angelidaki, I., Ellegaard, L., Ahring, B.K., 2003. Applications of the
anaerobic digestion process. In: Ahring, B.K. (Ed.),
Biomethanation II. Springer-Verlag, Berlin, Germany,
pp. 1e34.
APHA, 1998. Standard Methods for the Examination of Water and
Wastewater, nineteenth ed. American Public Health
Association, Washington DC, USA.
Appels, L., Baeyens, J., Degreve, J., Dewil, R., 2008. Principles and
potential of the anaerobic digestion of waste-activated sludge.
Progress in Energy and Combustion Science 34 (6), 755e781.
Appels, L., Lauwers, J., Degreve, J., Helsen, L., Lievens, B.,
Willems, K., Impe, J.V., Dewil, R., 2011. Anaerobic digestion in
global bio-energy production: potential and research
challenges. Renewable and Sustainable Energy Reviews 15 (9),
4295e4301.
Arbiv, R., van Rijn, J., 1995. Performance of a treatment system for
inorganic nitrogen removal in intensive aquaculture systems.
Aquacultural Engineering 14 (2), 189e203.
Boyd, C.E., Tucker, C.S., 1998. Pond Aquaculture Water Quality
Management. Kluwer Academic Publisher, Boston, MA, USA.
Chen, Y., Cheng, J.J., Creamer, K.S., 2008. Inhibition of anaerobic
digestion process: a review. Bioresource Technology 99 (10),
4044e4064.
FAO, 2008. The State of World Fisheries and Aquaculture. Food
and Agricultural Organization of the United Nations, Rome,
Italy.
Ferree, M.A., Shannon, R.D., 2001. Evaluation of a second
derivative UV/visible spectroscopy technique for nitrate and
 Author's personal copy
2850 w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 2 8 4 3 e2 8 5 0
total nitrogen analysis of wastewater samples. Water
Research 35 (1), 327e332.
Gebauer, R., 2004. Mesophilic anaerobic treatment of sludge from
saline fish farm effluents with biogas production. Bioresource
Technology 93 (2), 155e167.
Gebauer, R., Eikebrokk, B., 2006. Mesophilic anaerobic treatment
of sludge from salmon smolt hatching. Bioresource
Technology 97 (18), 2389e2401.
Halalsheh, M., Koppes, J., den Elzen, J., Zeeman, G., Fayyad, M.,
Lettinga, G., 2005. Effect of SRT and temperature on biological
conversions and the related scum-forming potential. Water
Research 39 (12), 2475e2482.
Hardy, R.W., 2001. Urban legends and fish nutrition, Part 2.
Aquaculture Magazine 27 (2), 57e60.
Klas, S., Mozes, N., Lahav, O., 2006. Development of a single-
sludge denitrification method for nitrate removal from RAS
effluents: lab-scale results vs. model prediction. Aquaculture
259 (1e4), 342e353.
Kugelman, I.J., Van Gorder, S., 1991. Water and energy recycling in
closed aquaculture systems. In: Engineering Aspects of
Intensive Aquaculture. Proceedings of the Aquaculture
Symposium, Cornell University, 4e6 April 1991. Northeast
Regional Agricultural Engineering Service (NRAES)-49, Ithaca,
NY, USA, pp. 80e87.
Lanari, D., Franci, C., 1998. Biogas production from solid wastes
removed from fish farm effluents. Aquatic Living Resources 11
(4), 289e295.
Marchaim, U., 1992. Biogas Processes for Sustainable
Development. FAO, Rome, Italy.
Masse, D.I., Masse, L., Croteau, F., 2003. The effect of temperature
fluctuations on psychrophilic anaerobic sequencing batch
reactors treating swine manure. Bioresource Technology 89
(1), 57e62.
McDermott, B.L., Chalmers, A.D., Goodwin, J.A.S., 2001.
Ultrasonication as a pre-treatment method for the
enhancement of the psychrophilic anaerobic digestion of
aquaculture effluents. Environmental Technology 22 (7),
823e830.
Mirzoyan, N., Parnes, S., Singer, A., Tal, Y., Sowers, K., Gross, A.,
2008. Quality of brackish aquaculture sludge and its suitability
for anaerobic digestion and methane production in an upflow
anaerobic sludge blanket (UASB) reactor. Aquaculture 279
(1e4), 35e41.
Mirzoyan, N., 2009. Waste Treatment for Brackish Water
Recirculated Aquaculture Systems: Reduction of Organic Load
Followed by Methane Production. PhD dissertation, Ben
Gurion University of the Negev, Beer Sheva, Israel.
Mirzoyan, N., McDonald, R.C., Gross, A., 2012. Anaerobic
treatment of brackishwater aquaculture sludge: an alternative
to waste stabilization ponds. Journal of World Aquaculture
Society 43 (2), 238e248.
Omil, F., Mendez, R.J., Lema, J.M., 1996. Anaerobic treatment of
seafood processing waste waters in an industrial anaerobic
pilot plant. Water SA 22 (2), 173e181.
OSullivan, C.A., Burrell, P.C., Clarke, W.P., Blackall, L.L., 2005.
Structure of a cellulose degrading bacterial community during
anaerobic digestion. Biotechnology and Bioengineering 92,
871e878.
Pind, P.F., Angelidaki, I., Ahring, B.K., Stamatelatou, K.,
Lyberatos, G., 2003. Monitoring and control of anaerobic
reactors. In: Ahring, B.K. (Ed.), Biomethanation II. Springer-
Verlag, Berlin, Germany, pp. 135e182.
Reed, S.C., Crites, R.W., Middlebrooks, E.J., 1995. Natural Systems
for Waste Management and Treatment. McGraw-Hill, New
York, NY, USA.
Sharrer, M.J., Tal, Y., Ferrier, D., Hankins, J., Summerfelt, S.T.,
2007. Membrane biological reactor treatment of a saline
backwash flow from a recirculating aquaculture system.
Aquacultural Engineering 36 (2), 159e176.
Soares, M.I.M., Brenner, A., Yevzori, A., Messalem, R., Leroux, Y.,
Abeliovich, A., 2000. Denitrification of groundwater: pilot-
plant testing of cotton-packed bioreactor and post-
microfiltration. Water Science and Technology 42 (1e2),
353e359.
SPSS, 1997. SigmaStat Statistical Software. SPSS, Chicago, IL, USA.
Tal, Y., Schreier, H.J., Sowers, K.R., Stubblefield, J.D., Place, A.R.,
Zohar, Y., 2009. Environmentally sustainable land-based
marine aquaculture. Aquaculture 286 (1e2), 28e35.
Timmons, M.B., Ebeling, J.M., 2007. Recirculating Aquaculture.
Cayuga Aqua Ventures, LLC, Ithaca, NY, USA.