Bioresource Technology xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Effects of pH on nitrogen transformations in media-based aquaponics

Yina Zou a, Zhen Hu a,⇑, Jian Zhang a, Huijun Xie b, Christophe Guimbaud c, Yingke Fang a
a
School of Environmental Science and Engineering, Shandong University, Jinan, China
b
Environmental Research Institute, Shandong University, Jinan, China
c
Laboratoire de Physique et de Chimie de l’Environnement et de l’Espace, Université d’Orléans, Orléans, France

h i g h l i g h t s

Aquaponics could tolerate a wide range of pH from 6.0 to 9.0.

Higher nitrogen utilization efficiency and N2O emission were achieved at pH 6.0.

Denitrification accounted for 75.2–78.5% of N2O emission from aquaponics.

Unfavorable conditions for denitrifiers led to higher N2O emission at pH 6.0.

a r t i c l e i n f o a b s t r a c t

Article history: To investigate the effects of pH on performance and nitrogen transformations in aquaponics, media-
Received 17 November 2015 based aquaponics operated at pH 6.0, 7.5 and 9.0 were systematically examined and compared in this
Received in revised form 24 December 2015 study. Results showed that nitrogen utilization efficiency (NUE) reached its maximum of 50.9% at pH
Accepted 28 December 2015
6.0, followed by 47.3% at pH 7.5 and 44.7% at pH 9.0. Concentrations of nitrogen compounds (i.e., TAN,
Available online xxxx
NO 
2 -N and NO3 -N) in three pH systems were all under tolerable levels. pH had significant effect on
N2O emission and N2O conversion ratio decreased from 2.0% to 0.6% when pH increased from 6.0 to
Keywords:
9.0, mainly because acid environment would inhibit denitrifiers and lead to higher N2O emission.
Aquaponics
pH
75.2–78.5% of N2O emission from aquaponics was attributed to denitrification. In general, aquaponics
Nitrogen transformations was suggested to maintain pH at 6.0 for high NUE, and further investigations on N2O mitigation strategy
N2O emission are needed.
Microbial abundance Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction microbes are three main components of aquaponics, and microbes

Aquaculture has become one of the fastest-growing food-
producing sectors since 1980s and accounted for almost half
(49%) of global fish consumption in 2012 (FAO, 2014). In order to
meet the growing human demands for aquatic products, aquacul-
ture scale is bound to continue expand. However, aquaculture is
a high-polluting industry. On average, only 25% of its nitrogen
and phosphorus inputs could be recovered by target organisms
(Crab et al., 2007), and the rest of nutrients are discharged into sur-
rounding water. This is not only a waste of nutrients, but also
causes serious pollution to the surrounding environment.
Aquaponics is considered to have potentials to solve the
abovementioned problems. Aquaponics is the combination of
conventional aquaculture and hydroponics, which could achieve
co-culture of fish and plants at the same time. Fish, plants and
⇑ Corresponding author.
E-mail address: huzhen885@sdu.edu.cn (Z. Hu).
play the bridge role of converting fish waste to plant nutrients
(Somerville et al., 2014). Usually there are three common types
of aquaponics designs, i.e., floating raft, nutrient film technology,
and media-based bed, mainly classified according to hydroponics
(Nelson and Pade, 2007). Of which, media-based bed could act as
a filtration unit and provide surface area for microbial growth at
the same time. This makes it popular in currently running
aquaponics. A survey conducted by Love et al. (2014) discovered
that 86% of their respondents adopted media-based aquaponics.
Nitrogen is a vital element for all living organisms. In aquapon-
ics, fish feed that contains high content of protein is added into sys-
tem and digested by fish. Most of the nitrogen is then excreted in
the form of total ammonia (TAN), which is toxic to fish. Fortu-
nately, nitrifying bacteria in aquaponics could first convert ammo-
nia to nitrite (NO 
2 ) and then into nitrate (NO3 ) through
nitrification. Nitrate would be reduced to N2 through denitrifica-
tion, but more importantly, it is an important fertilizer for
plant growth. The establishment of cooperation among three

http://dx.doi.org/10.1016/j.biortech.2015.12.079
0960-8524/Ó 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Zou, Y., et al. Effects of pH on nitrogen transformations in media-based aquaponics. Bioresour. Technol. (2016), http://dx.
doi.org/10.1016/j.biortech.2015.12.079
2 Y. Zou et al. / Bioresource Technology xxx (2016) xxx–xxx

Table 1 of hydroxylamine, intermediate during the oxidation of TAN to

Performance of aquaponics under different pH treatments. NO 2 , would also lead to N2O production (Kampschreur et al.,

Parameters pH = 6.0 pH = 7.5 pH = 9.0

2009). While in denitrification, N2O which failed to be reduced in
time might be emitted to the atmosphere. Previous study has
Phase I Plant biomass increase (kg/m2) 2.70a 2.48a 1.94b
Fish biomass increase (kg/m3) 0.81a 0.87b 0.83a
shown that 1.5–1.9% of nitrogen input was lost in the form of
SGR* (%) 0.22 0.24 0.23 N2O in floating raft aquaponics, while 1.3% was found in conven-
FCR** 3.10 3.02 3.14 tional aquaculture (Hu et al., 2013, 2015). However, to date, no
Phase II Plant biomass increase (kg/m2) 3.01a 2.59b 2.72b N2O emission investigation has been conducted in media-based
Fish biomass increase (kg/m3) 1.31a 1.43b 1.39b aquaponics.
SGR (%) 0.39 0.37 0.37 In this study, media-based aquaponics was established to inves-
FCR 3.35 3.13 3.22
tigate the effects of pH on its nitrogen transformations, and special
*
SGR (Specific Growth Rate) = (ln Wf  ln Wi)  100/days, Wf is final weight of attention was paid to N2O emission. 15N labeling experiment was
fish and Wi is initial weight. used to determine the main source of N2O emission, and quantita-
**
FCR (Food Conversion Ratio) = total feed given (g) of fish/total wet weight gain
tive polymerase chain reaction (Q-PCR) technology was applied to
(g) of fish.
a,b
Different letters show significant differences at p < 0.05 (Duncan). quantify the abundance of nitrifies and denitrifies to reveal the
influence of pH on microbial community.

components increases nitrogen utilization efficiency (NUE) and

avoids nitrogen-rich wastewater discharge. To achieve higher
productivity and better water quality in aquaponics, many kinds
of regulation attempts have been conducted. Liang and Chien
(2013) found that better fish growth, plant growth, and nutrients
removal efficiency from water were obtained in aquaponics under
24-hour light than 12-hour light, and Endut et al. (2010) reported
similar results at loading rate of 1.28 m/d. However, thorough
study on nitrogen transformations in aquaponics is still lacking.
pH is one of the most important regulation factors for aquapo-
nic systems, and it is needed to be balanced for fish, plants and
microbes at the same time. Usually, recommended pH for plant
cultivation was slightly acid (5.5–5.8) (Bugbee, 2003), while the
optimal pH for nitrification was 7.5–8.0 (Kim et al., 2007). Fish
can tolerate a wide pH range, and the optimal pH was different
for different species (Arimoro, 2006; Lemarie et al., 2004). In
aquaponics, providing the pH optima for every part is impossible,
but knowing optimal pH range for the best overall performance
is necessary. Tyson et al. (2008) had claimed that reconciling pH
for aquaponics should be 7.5–8.0, but no difference in plant yields
was detected in their research, which was unreasonable. Essential
study is required to investigate the aquaponic performance under
different pH conditions. In addition, to achieve the best sustainabil-
ity, neither yield nor environment impacts could be ignored.
Since biological nitrogen transformations play the key role of
bridge in aquaponics, it may cause environmental harms. Nitrous
oxide (N2O), the third biggest greenhouse gas with a global-
warming potential 296 times higher than CO2, is often generated
from biological nitrification and denitrification processes. In nitri-
fication, heterotrophic ammonia oxidation bacteria (AOB) could
conduct nitrifier denitrification to produce N2O, and the oxidation
Fig. 1. Nitrogen distribution among different pH treatments.
2. Methods
2.1. Aquaponic microcosms
Experimental aquaponic systems were operated side by side
under natural conditions in Jinan, China. A transparent rainproof
shed was installed above the aquaponics. All systems shared same
setup design. Each system was mainly consisted of two parts, fish
tank and hydroponic bed. These two parts were both made of plas-
tic box, i.e., 65 cm  45 cm  50 cm for fish tank and
80 cm  55 cm  45 cm for hydroponic bed. Fish tank was placed
on ground. The effective water volume in fish tank was 100 L. There
was no water exchange in fish tank during study period except for
water loss through evaporation, transpiration and sampling, which
was replenished with freshwater every day. Peristaltic pumps
(BT100-2J; Baoding Longer, China) were applied to lift water into
hydroponic bed, which was placed above fish tank. Water flow rate
was 200 L/d. About 30 cm-deep of perlite with particle size ranging
from 1 mm to 3 mm was filled in hydroponic bed. Water from fish
tank flowed through hydroponic bed for purification and then puri-
fied water flowed back to fish tank under gravity. Air compressors
were used to supply air for fish growth and gas flow meters were
installed to guarantee that dissolved oxygen (DO) concentration
in fish tank was above 5 mg/L.
Common carp (Cyprinus carpio) and pakchoi (Brassica chinensis),
which are very popular aquatic product and vegetable in northern
China, were selected to be cultured in this study. Fish with initial
weight of 50–70 g was distributed randomly into each fish tank
at a stocked density around 10 kg/m3. Commercial fish feed was
used in present study. At the beginning of study, artificial feeding
was employed. Fish feed was added into fish tank twice a day,
and the unconsumed fish feed was taken out ten minutes later to
prevent water from being polluted. From day 31, automatic fish
feeders (AF-2005D; Resun, China), which could feed fish four times
per day, were introduced into systems. The amount of fish feed was
recorded every day. Plant seeds were germinated in seedling-
raising plates two weeks before the present experiment began,
and then healthy seedlings with similar size were transplanted
carefully into hydroponic bed at 20 cm  20 cm spacing. In order
to achieve better plant growth, iron-chelator and hoagland
microelement solution were added to supply minerals required
for plant growth (Soetan et al., 2010).
A pre-experiment was conducted with present aquaponic sys-
tems before study began, in order to accumulate microbes. Present
study was carried on for 70 days, and aquaponic systems were
operated continuously. Three pH gradients were set as 6.0, 7.5
and 9.0, and each treatment contained three replicates. H2SO4
and KOH of 1.0 M were applied to adjust pH every day from the

Please cite this article in press as: Zou, Y., et al. Effects of pH on nitrogen transformations in media-based aquaponics. Bioresour. Technol. (2016), http://dx.
doi.org/10.1016/j.biortech.2015.12.079
 Y. Zou et al. / Bioresource Technology xxx (2016) xxx–xxx 3

Fig. 2. Nitrogen compounds variations among different pH treatments during the study period. (a) TAN; (b) N2O-N; (c) N3O-N.

eighth day. Two batches of plants were harvested during the entire
experiment, and experiment period was divided into Phase I and
Phase II according to the time of first plant harvest.
2.2. Water quality monitoring
simultaneously. About 50 ml of water sample was collected into
polyethylene bottles and taken back to laboratory. Nitrogen com-
pounds measurements including TAN (NH3 and NH+4), NO 2 and
NO3 were accomplished in 12 hours according to the methods
described in APHA (2005).

Monitoring of water quality was carried out every other day at
9:00–10:00 AM., except for pH, which was measured daily. pH and
DO concentration were determined in suit using pH meter (PHS-
3C; Leici, China) and DO meter (HQ30d 53LEDTM; HACH, USA),
respectively; water temperature was also measured by DO meter
2.3. N2O sampling and measurement
Since there was no stable pH condition in Phase I due to pH
adjustment, N2O emission was measured weekly during Phase II
(five times). Open chamber and closed chamber methods revised

Please cite this article in press as: Zou, Y., et al. Effects of pH on nitrogen transformations in media-based aquaponics. Bioresour. Technol. (2016), http://dx.
doi.org/10.1016/j.biortech.2015.12.079
4 Y. Zou et al. / Bioresource Technology xxx (2016) xxx–xxx
Fig. 3. Total N2O emission fluxes among different pH treatments during Phase II
period.
according to Sun et al. (2013) were used to take samples simulta-
neously from fish tank and hydroponic bed, respectively. Briefly, in
aerated fish tank, the polymethyl methacrylate chamber covered
up almost entire water surface and a sampling port was installed
on the chamber roof. Sampling port was open during sampling
time in case that air pressure increased too much due to aeration.
A small fan fixed on side wall was used to blend air in the chamber.
For non-aerated hydroponic bed, similar chamber but with closed
sample port was applied. In each sampling process, a pump was
used to take about 200 ml of gas into air bag every 20 min for a
total 120 min. Subsequently, gas chromatograph (7890A; Agilent,
USA) equipped with electron capture detector was used to detect
N2O concentration according to Xia et al. (2013).
N2O emission flux was obtained based on N2O concentrations of
seven gas samples during each sample period. Formulas are listed
as follows:
In fish tank:
F f ¼ ðQ  C  M  PÞ=ðR  T  AÞ ð1Þ
where, Ff is the N2O emission flux in fish tank; Q is the aeration rate;
C is the average N2O concentrations at 0, 20, 40, 60, 80, 100, 120 min
of process; M is the molecular weight of N2O; P is the atmospheric
pressure; R is the gas constant; T is the temperature; A is the area
covered by open-chamber.
In hydroponic bed:
F h ¼ V=A  dC=dt ð2Þ
where, Fh is the N2O emission flux in hydroponic bed; V is the vol-
ume of gas chamber; A is its cover area; dC/dt is the linear slope of
N2O concentrations at 0, 20, 40, 60, 80, 100, 120 min of process.
2.4. Nitrogen content determination
After fish or plants were harvested, their biomass (wet weight)
increases were recorded. Three fish and three plants were ran-
domly selected to determine the nitrogen content. Fish and plants
were firstly dried to the constant weight in oven at 80 °C to calcu-
late their moisture content, and subsequently grinded to pass 100
mesh sieves. Elementary analyzer (Vario Macro Cube; Elementar,
Germany) was used to determine the nitrogen concentration in
dried samples. Nitrogen in fish feed was also obtained with the
above methods. As for nitrogen remained in water body, it was
estimated on basis of the monitored nitrogen compounds concen-
trations and the volume of water (i.e., 100 L). Average N2O emis-
sion flux of five measurements was regarded as the average
Please cite this article in press as: Zou, Y., et al. Effects of pH on nitrogen transformations in media-based aquaponics. Bioresour. Technol. (2016), http://dx.
doi.org/10.1016/j.biortech.2015.12.079
emission flux during entire experiment and was used to estimate
the share of N2O (i.e., conversion ratio).
2.5. Microbial analysis
Total DNA from fish tank sample and hydroponic bed sample
was extracted separately. Two hundred milliliter of water sample
from fish tank was first filtered through 0.22 mm nitrocellulose fil-
ters as described by Cébron et al. (2004), and then PowerSoil DNA
Isolation Kit (12888-50; MOBIO, USA) was used to extract DNA
from filters following the manufacturer’s protocol. Perlite samples
from hydroponic bed were grabbed by five-point sampling method
at 15 cm below the surface and were mixed completely. Then, 0.5 g
perlite was used to extract total DNA. DNA concentration and qual-
ity were estimated by a NanoVue Plus Spectrophotometer (K5500;
Kaiao, China).
To assess the effects of pH on aquaponics performance and its
nitrogen transformations, functional genes involved in biological
nitrification and denitrification were measured using Q-PCR tech-
nology. Q-PCR has been widely adopted to detect microbes in nat-
ural samples without laboratory culture. amoA, nirS, nirK and nosZ
genes were chosen in present study. amoA is the gene encoding
ammonia monooxygenase in nitrification; nirS and nirK are two
main genes encoding nitrite reductases in denitrification, and nosZ
is the gene encoding nitrous oxide reductase in denitrification. Pri-
mer fragments and PCR programs for these genes were cited from
Zhi and Ji (2014). Plasmids containing target gene fragment were
used to obtain standard working curves as Wang et al. (2014)
described. The standard curve efficiency of four genes was kept
in 90–105%. Q-PCR was performed in a real-time PCR system
(LC-480; Roche, Switzerland) with a 20 lL of reaction mixture.
The reaction mixture consisted of 10 lL of SYBR Premix Taq,
0.5 lL of each of forward and reverse primer (10 lM), 1 lL of tem-
plate DNA, and 8.2 lL of RNase-free water.
15
2.6. N labeling experiment
To identify the source of N2O in aquaponics, modified 15N label-
ing experiment described as Zhu et al. (2013) was conducted in
aquaponics at pH 7.5. In brief, K15NO3 (99 atom%) was injected to
fish tank at an initial amount of 10% of original NO
3 -N concentra-
tion to avoid the shock to the system. To ensure enough time for
uniform distribution of 15N in system, gas and water sampling
were conducted at 24 h, 48 h, and 72 h after 15N was added. Water
was collected before gas sampling began. 15N abundance measure-
ments in N2O and NO 3 were performed using TraceGas-Isotope
Ratio Mass Spectrometer system in Environmental Stable Isotope
Lad, Chinese Academy of Agricultural Sciences. N2O emission from
denitrification was obtained through comparing 15N abundance in
N2O with that in NO 3 during sampling period (Zhu et al., 2013).
2.7. Statistics analysis
All data shown in the manuscript was the average values of
three replicates. SPSS 17.0 (IBM, USA) was used to perform statis-
tical analysis of fish and plant production, using analysis of vari-
ance (ANOVA) followed by Duncan’s multiple range test at p < 0.05.
3. Results and discussion
3.1. Aquaponic performance
Survival rates for fish and plants were 100% in all systems, indi-
cating that aquaponics can adapt to a wide range of pH. In the
entire study period, water temperature and DO concentrations
 Y. Zou et al. / Bioresource Technology xxx (2016) xxx–xxx 5

Fig. 4. Functional gene copy numbers among different pH treatments. (a) amoA; (b) nirS; (c) nirK; (d) nosZ.

were both maintained at similar values in three pH treatments,
which were 23.0 ± 1.9 °C and 6.2 ± 0.9 mg/L, respectively. Another
important index for aquaponics evaluation was daily water replen-
ishment ratio, which also didn’t show significant difference. The
average water replenishment ratio of all treatments was about
1.8%. pH adjustment started on day 8, however, given that
aquaponics had certain buffering capacity, stable difference in pH
didn’t appear until day 20. This result proved that aquaponics
had self-regulation ability to cope with pH changes in environ-
ment. pH fluctuated near the set value in the following days.
Table 1 shows the fish and plant production parameters in
Phase I and Phase II. It can be found that fish biomass increase in
Phase II was higher than that in Phase I, which was mainly attrib-
uted to the modification of feeding methods. Increased feeding fre-
quency and feeding rate significantly promoted the fish growth
rate, e.g. specific growth rate (SGR) in Phase II was enhanced by
54.2–77.3%. But feed conversion ratio (FCR) was about 8% lower
Table 2
N2O source partition in aquaponics.
in Phase II than that in Phase I. Similar result has also been
reported by Silva et al. (2007), and possible reason was that fish
would extract more nutrients under low feeding rate through opti-
mizing their digestion. As to plants, higher yield, i.e., plant biomass
increase, was also obtained in Phase II, which was attributed to
higher nutrient concentrations in water in Phase II (Fig. 2).pH
had significant effect on aquaponics production (p < 0.05). No mat-
ter in Phase I or Phase II, fish biomass increase was highest when
pH was 7.5 followed by 9.0 and 6.0, indicating that neutral or
slightly alkaline environment was more suitable for common carp
growth, which was consistent with the research of Heydarnejad
(2012), who claimed that the optimal pH range for the growth of
common carp was 7.5–8.0. Under optimal environment, physiolog-
ical activity of fish would increase and thus enhance its digestion.
Different trend was obtained in plant yield. The highest plant yield
appeared at pH 6.0. This indicated that plant growth in aquaponics
was consistent with that in hydroponics, where pH of nutrient
solution was always adjusted to slightly acid, e.g. 6.5 (Liao et al.,
2006).

Time after N 15
NO
3 -N
15
N2O N2O from 3.2. Nitrogen budget
input (h) abundance (%) abundance (%) denitrification* (%)
24 6.56 4.14 63.09 Nitrogen budget was calculated at the end of study period and
48 6.23 4.68 75.15 is shown in Fig. 1. The most important part was the proportion of
72 5.98 4.69 78.51
fish and plants, in another word, NUE. NUE showed a significant
*
N2O from denitrification = 15N2O abundance/15NO
3 -N abundance. difference among three pH treatments, i.e., 50.9% at pH 6.0, 47.3%

Please cite this article in press as: Zou, Y., et al. Effects of pH on nitrogen transformations in media-based aquaponics. Bioresour. Technol. (2016), http://dx.
doi.org/10.1016/j.biortech.2015.12.079
6 Y. Zou et al. / Bioresource Technology xxx (2016) xxx–xxx
at pH 7.5 and 44.7% at pH 9.0. The highest NUE was achieved at pH
6.0 and it decreased as the pH increased. Plants played the decisive
role for this result, since plant growth was significantly better at
pH 6.0. Nitrogen retained in plants was 34.8% at pH 6.0, 30.3% at
pH 7.5 and 28.5% at pH 9.0, which was a great improvement than
our previous research. Value of 15.0% has been found in a floating-
raft pakchoi-tilapia aquaponics (Hu et al., 2015). This promotion in
plants’ ability of retaining nitrogen indicated that media-based
hydroponic bed had better production performance than floating
raft hydroponics when combined with aquaculture system.
It is noteworthy that nutrient level in water was almost similar
in different treatments, i.e., 19.3% at pH 6.0, 18.3% at pH 7.5 and
19.8% at pH 9.0. That meant water quality in terms of nutrient
removal, didn’t show significant difference among three pH sys-
tems. Interestingly, significant difference in nitrogen loss was
observed in present study, i.e., 27.8% at pH 6.0, 32.8% at pH 7.5
and 34.6% at pH 9.0. Less nitrogen loss at pH 6.0 meant more nitro-
gen was retained within system and could be further used by
plants. More nitrogen loss at pH 7.5 and 9.0 was the combined con-
tribution of enhanced NH3 evaporation and N2 emission. NH3 frac-
tion would increase as pH increased (Rajagopal et al., 2013), and
coupled with the existence of aeration, NH3 loss would be
strengthened at high pH. N2 is the end-product of denitrification,
and since the optimal pH for denitrification rate was 7.5–8.0, more
N2 would be produced under alkaline conditions. N2O emission
was not taken into consideration because of its little amount, and
detailed information about N2O can be found in Section 3.4.
3.3. Variation of nitrogen compounds
Variations of nitrogen compounds concentrations are shown in
Fig. 2. Both NH3 and NO 2 have toxicity effect on fish survive, and
much attention should be paid to control their concentrations.
Since the aquaponic systems used in present study were mature
systems rich in microbes, neither TAN nor NO 2 -N accumulation
was observed at the beginning of experiment. Interaction of sud-
denly increased feeding rate (8–12 g/d) and pH adjustment on
day 8 led to a rise in TAN concentrations. However, considering
that aquaponic systems had buffer capacity to pH changes, the
increases of TAN concentrations were mainly attributed to increase
of feeding rate. Much more excreta was produced by fish when
feeding rate increased, and TAN concentrations increased accord-
ingly. Similar phenomenon appeared again on day 31 when feeding
rate was increased from 12 g/d to 16 g/d. But at this time, changes
of TAN concentrations were different among three pH treatments.
Under pH 7.5, TAN concentration was almost unchanged and main-
tained at 0.7 ± 0.2 mg/L, while under pH 9.0, TAN concentration
first increased to 1.9 mg/L and subsequently dropped down to
0.6 ± 0.1 mg/L. Under pH 6.0, TAN concentration increased and it
was always high In Phase II. Fortunately, NH3 fraction was very
small when pH was 6.0 (Rajagopal et al., 2013), and fish growth
would not be influenced by NH3 under this condition. This implied
that system balance of aquaponics could be easily broken at
extreme pH especially in acidic state. NO 2 -N concentrations were
always low (<0.2 mg/L) and almost had no accumulation. That’s
because effects of pH on ammonium-oxidizing bacteria (AOB)
and nitrite-oxidizing bacteria (NOB) were similar. Since ammonia
oxidation was the first and rate-limiting step in nitrification,
NO 2 -N produced by ammonia oxidation could be converted to
NO 3 -N immediately.
NO 3 -N is the end product of nitrification and also the main
nutrient for plant growth in aquaponics. NO 3 -N concentration in
aquaponics was the equilibrium result between nitrification pro-
duction and plant absorption. In Phase I, when plants were just
transplanted, they had low absorption ability. Nitrification produc-
tion exceeded plant absorption, and resulted in the increases of
Please cite this article in press as: Zou, Y., et al. Effects of pH on nitrogen transformations in media-based aquaponics. Bioresour. Technol. (2016), http://dx.
doi.org/10.1016/j.biortech.2015.12.079
NO 
3 -N concentrations. As plants grew up and more NO3 -N was

absorbed by plants, the increase of NO3 -N concentrations slowed
down and a platform of NO 3 -N concentrations was observed at
the later period (Fig. 2c). Under pH 6.0, NO3 -N concentration even
dropped a little because of enhanced plant growth. In Phase II,
NO3 -N concentrations didn’t show a platform because of increased
feeding rate. It can be inferred that more plants can be supported
after the increase of nitrogen input. Proper plants/fish ratio was
needed for long-running aquaponics to control the nutrient level.
pH also had notable effect on NO 3 -N concentrations, and lower
NO3 -N concentration was obtained at pH 6.0 in Phase I. This might
because nitrification was inhibited at low pH (Tang and Chen,
2015). In addition, it can be found that NO 3 -N concentration gaps
among three pH systems were narrowed in Phase II. That’s because
more nitrogen loss occurred at pH 7.5 and 9.0. Under pH 7.5, more
NO3 -N was consumed by denitrification, and under pH 9.0, pro-
duction of NO 3 -N decreased because of more NH3 evaporation.
3.4. N2O emission
Nitrogen loss via N2O emission accounted for 2.0%, 1.6% and
0.6% of total nitrogen input at pH 6.0, 7.5 and 9.0, respectively
(Fig. 1). The specific N2O emission fluxes in different pH treatments
during Phase II are shown in Fig. 3. Firstly, it was found that most
of the N2O emission occurred in hydroponic bed, i.e., 89.2%, 93.5%
and 86.2% were obtained at pH 6.0, 7.5 and 9.0, respectively. Thus,
hydroponic bed was responsible for 0.5–1.8% of nitrogen input.
Results in present study were comparable to that in traditional
hydroponics. 1.0–4.6% of applied nitrogen has been reported to
be emitted as N2O in rockwool-based hydroponics (Yoshihara
et al., 2014; Daum and Schenk, 1996). Secondly, total N2O emission
fluxes increased gradually during Phase II at pH 6.0 and 7.5, which
could be related to continuously increasing NO 3 -N concentrations.
The reasons are as follows: N2O was mainly produced through
nitrification and denitrification, which took TAN and NO 3 -N as
substrate, respectively. During Phase II, since all operational
parameters (such as, pH, DO concentration and feeding rate) were
maintained stable, only slight fluctuations in TAN concentrations
were observed. Thus, it was believed that N2O from nitrification
would not change a lot. Whereas NO 3 -N concentrations increased
steadily during Phase II, therefore, denitrification was considered
to be the most possibly responsible way for N2O emission.
To figure out N2O source partition in aquaponics, 15N labeling
experiment was conducted in aquaponics at pH 7.5. Table 2 gives
the percentages of N2O from denitrification. Considering 15NO 3 -N
may not distribute evenly in the system during the first 24 h after
being input, 75.2–78.5% was deemed as the denitrification contri-
bution to N2O emission. Thus denitrification was the dominant
pathway of N2O emission in media-based aquaponics. Given that
DO concentration was over 5 mg/L in fish tank, nitrification would
be the only process accounting for N2O emission from fish tank,
and hydroponic bed accounted for all of N2O emission from deni-
trification. The coexistence of anaerobic microenvironments and
denitrifying bacteria in hydroponic bed provided optimal circum-
stance for denitrification.
Significant effect of pH on N2O emission from aquaponics was
observed. N2O emission fluxes decreased as pH increased. pH is
an important factor influencing denitrification. It is wildly accepted
that N2O reductase would be inhibited at low pH and more N2O
would be produced. Cuhel et al. (2010) found that N2O/(N2O
+ N2) decreased as pH increased from 5.5 to 7.8. Also, study of
N2O reductase activity by Fujita and Dooley (2007) showed that
its specific activities were especially high when pH exceeded 9.0.
This explained why N2O emission flux was lower at pH 9.0, and
why its emission didn’t show an obvious increase during Phase II
(Fig. 3). Under pH 9.0, the enhanced enzyme activity could quickly
 Y. Zou et al. / Bioresource Technology xxx (2016) xxx–xxx
reduce N2O to N2, and the little accumulation of N2O led to low N2O
emission.
3.5. Effect of pH on microbes
Gene number could reflect the microbe number, which was clo-
sely corresponded with nitrogen transformations. The quantities of
functional genes involved in biological nitrogen transformations,
i.e., amoA, nirS, nirK and nosZ are shown in Fig. 4. Firstly, a majority
of microbes existed in hydroponic bed. Large numbers of microbes
in hydroponic bed made it become the main place nitrogen trans-
formations took place, thus more N2O was emitted through these
biological processes. Secondly, amoA gene numbers at pH 6.0 were
only a quarter of that at pH 7.5 and 9.0, indicating that low pH has
significant inhibition effect on ammonia oxidizing bacteria (AOB).
This explained the TAN accumulation at pH 6.0. Though the num-
bers of AOB were lower at pH 6.0, NO 3 -N produced by nitrification
was comparable to that at pH 7.5 and 9.0, indicating that the num-
bers of microbes was not a limiting factor for nitrification in
media-based aquaponics.
What’s more, gene numbers of nirS and nirK demonstrated that
denitrifying bacteria preferred living in alkaline environment. But
interestingly, although denitrification was responsible for N2O
emission, negative correlation between denitrifying bacteria num-
bers and N2O emission was detected in present study. The most
likely reason was that inappropriate environment inhibited deni-
trification process, and more N2O was emitted as intermediate
product. The similar result has also been found by Wunderlin
et al. (2012), who conducted nitrate reduction experiment using
heterotrophic denitrifies and more N2O production was obtained
under non-optimal conditions. Difference in nosZ gene numbers
among three pH systems also confirmed this result. Less nosZ gene
copy numbers were found at pH 6.0, indicating a low ability to
reduce N2O, which leading to more N2O emission.
4. Conclusions
Although aquaponics could tolerate a wide pH range, its perfor-
mance and nitrogen transformations were affected by pH signifi-
cantly. Higher plant yield was obtained at pH 6.0, led to higher
NUE. Unfortunately, this was achieved at the expense of higher
N2O emission, which accounted for 2.0% of nitrogen input. In
media-based aquaponics, N2O emission mainly occurred in hydro-
ponic bed, where most of microbes grew. More importantly, 75.2–
78.5% of N2O emission from aquaponics was attributed to denitri-
fication. And higher N2O emission at lower pH was attributed to
the inhibition of denitrifying bacteria.
Acknowledgements
This work was supported by National Natural Science Founda-
tion of China (No. 21307076 & No. 51578321), and Fundamental
Research Funds of Shandong University (No. 2014TB003 & No.
2015JC056). This work is also supported by the AMIS (Fate and
Impact of Atmospheric Pollutants) project funded by the European
Union, under the Marie Curie Actions IRSES (International Research
Staff Exchange Scheme), within the Seventh Framework Pro-
gramme FP7-PEOPLE-2011-IRSES.
References
Arimoro, F.O., 2006. Culture of the freshwater rotifer, Brachionus calyciflorus, and
its application in fish larviculture technology. Afr. J. Biotechnol. 5, 536–541.
APHA, 2005. Standard Methods for the Examination of Water and Wastewater,
21th ed. APHA, Washington, DC, USA.
Bugbee, B., 2003. Nutrient management in recirculating hydroponic culture. Acta
Hortic. 648, 99–112.
Please cite this article in press as: Zou, Y., et al. Effects of pH on nitrogen transformations in media-based aquaponics. Bioresour. Technol. (2016), http://dx.
doi.org/10.1016/j.biortech.2015.12.079
7
Cébron, A., Coci, M., Garnier, J., Laanbroek, H.J., 2004. Denaturing gradient gel
electrophoretic analysis of ammonia-oxidizing bacterial community structure
in the lower Seine River: impact of Paris wastewater effluents. Appl. Environ.
Microb. 70, 6726–6737.
Crab, R., Avnimelech, Y., Defoirdt, T., Bossier, P., Verstraete, W., 2007. Nitrogen
removal techniques in aquaculture for a sustainable production. Aquaculture
270, 1–14.
Cuhel, J., Simek, M., Laughlin, R.J., Bru, D., Cheneby, D., Watson, C.J., Philippot, L.,
2010. Insights into the effect of soil pH on N2O and N2 emissions and denitrifier
community size and activity. Appl. Environ. Microb. 76, 1870–1878.
Daum, D., Schenk, M.K., 1996. Gaseous nitrogen losses from a soilless culture
system in the greenhouse. Plant Soil 183, 69–78.
Endut, A., Jusoh, A., Ali, N., Nik, W.B.W., Hassan, A., 2010. A study on the optimal
hydraulic loading rate and plant ratios in recirculation aquaponic system.
Bioresour. Technol. 101, 1511–1517.
FAO (Food and Agriculture Organization of the United Nations), 2014. The State of
World Fisheries and Aquaculture: Opportunities and Challenges. FAO, Rome,
Italy.
Fujita, K., Dooley, D.M., 2007. Insights into the mechanism of N2O reduction by
reductively activated N2O reductase from kinetics and spectroscopic studies of
pH effects. Inorg. Chem. 46, 613–615.
Heydarnejad, M.S., 2012. Survival and growth of common carp (Cyprinus carpio)
exposed to different water pH levels. Turk. J. Vet. Anim. Sci. 36, 245–249.
Hu, Z., Lee, J.W., Chandran, K., Kim, S., Sharma, K., Brotto, A.C., Khanal, S.K., 2013.
Nitrogen transformations in intensive aquaculture system and its implication to
climate change through nitrous oxide emission. Bioresour. Technol. 130, 314–
320.
Hu, Z., Lee, J.W., Chandran, K., Kim, S., Brotto, A.C., Khanal, S.K., 2015. Effect of plant
species on nitrogen recovery in aquaponics. Bioresour. Technol. 188, 92–98.
Kampschreur, M.J., Temmink, H., Kleerebezem, R., Jetten, M.S., van Loosdrecht, M.C.,
2009. Nitrous oxide emission during wastewater treatment. Water Res. 43,
4093–4103.
Kim, Y.M., Park, D., Lee, D.S., Park, J.M., 2007. Instability of biological nitrogen
removal in a cokes wastewater treatment facility during summer. J. Hazard.
Mater. 141, 27–32.
Lemarie, G., Dosdat, A., Covès, D., Dutto, G., Gasset, E., Person-Le Ruyet, J., 2004.
Effect of chronic ammonia exposure on growth of European seabass
(Dicentrarchus labrax) juveniles. Aquaculture 229, 479–491.
Liang, J., Chien, Y., 2013. Effects of feeding frequency and photoperiod on water
quality and crop production in a tilapia–water spinach raft aquaponics system.
Int. Biodeter. Biodegr. 85, 693–700.
Liao, C., Yen, J., Wang, Y., 2006. Effects of endocrine disruptor di-n-butyl phthalate
on the growth of Bok choy (Brassica rapa subsp. chinensis). Chemosphere 65,
1715–1722.
Love, D.C., Fry, J.P., Genello, L., Hill, E.S., Frederick, J.A., Li, X., Semmens, K., 2014. An
international survey of aquaponics practitioners. PLoS One 9, e102662.
Nelson, R.L., Pade, J.S., 2007. Aquaponic equipment the clarifier. Aquaponics J., 30–
31, 4th Quarter
Rajagopal, R., Massé, D.I., Singh, G., 2013. A critical review on inhibition of anaerobic
digestion process by excess ammonia. Bioresour. Technol. 143, 632–641.
Silva, C.R., Gomes, L.C., Brandão, F.R., 2007. Effect of feeding rate and frequency on
tambaqui (Colossoma macropomum) growth, production and feeding costs
during the first growth phase in cages. Aquaculture 264, 135–139.
Soetan, K.O., Olaiya, C.O., Oyewole, O.E., 2010. The importance of mineral elements
for humans, domestic animals and plants: a review. Afr. J. Sci. 4, 200–222.
Somerville, C., Cohen, M., Pantanella, E., Stankus, A., Lovatelli, A., 2014. Small-scale
aquaponic food production: integrated fish and plant farming. FAO Fish.
Aquacult. Tech. Paper., 1–262
Sun, S., Cheng, X., Sun, D., 2013. Emission of N2O from a full-scale sequencing batch
reactor wastewater treatment plant: characteristics and influencing factors. Int.
Biodeter. Biodegr. 85, 545–549.
Tang, H.L., Chen, H., 2015. Nitrification at full-scale municipal wastewater
treatment plants: evaluation of inhibition and bioaugmentation of nitrifiers.
Bioresour. Technol. 190, 76–81.
Tyson, R.V., Simonne, E.H., Treadwell, D.D., Davis, M., White, J.M., 2008. Effect of
water ph on yield and nutritional status of greenhouse cucumber grown in
recirculating hydroponics. J. Plant. Nutr. 31, 2018–2030.
Wang, Q., Xie, H., Zhang, J., Liang, S., Ngo, H.H., Guo, W., Liu, C., Zhao, C., Li, H., 2014.
Effect of plant harvesting on the performance of constructed wetlands during
winter: radial oxygen loss and microbial characteristics. Environ. Sci. Pollut.
Res. 22, 7476–7484.
Wunderlin, P., Mohn, J., Joss, A., Emmenegger, L., Siegrist, H., 2012. Mechanisms of
N2O production in biological wastewater treatment under nitrifying and
denitrifying conditions. Water Res. 46, 1027–1037.
Xia, Y., Li, Y., Li, X., Guo, M., She, D., Yan, X., 2013. Diurnal pattern in nitrous oxide
emissions from a sewage-enriched river. Chemosphere 92, 421–428.
Yoshihara, T., Tokura, A., Hashida, S., Kitazaki, K., Asobe, M., Enbutsu, K., Takenouchi,
H., Goto, F., Shoji, K., 2014. A precise/short-interval measurement of nitrous
oxide emission from a rockwool tomato culture. Environ. Control Biol. 52, 137–
147.
Zhi, W., Ji, G., 2014. Quantitative response relationships between nitrogen
transformation rates and nitrogen functional genes in a tidal flow constructed
wetland under C/N ratio constraints. Water Res. 64, 32–41.
Zhu, J., Mulder, J., Bakken, L., Dörsch, P., 2013. The importance of denitrification for
N2O emissions from an N-saturated forest in SW China: results from in situ 15N
labeling experiments. Biogeochemistry 116, 103–117.