Biobased Chemistry and Technology

Thesis Biobased Chemistry and Technology

Aquaponic nutrient model

Thomas Peterhans

15.12.2015
 Aquaponic nutrient model

A daily material flow analysis

approach

Name course: Thesis project Biobased Chemistry and

Technology
Number: BCT-80436
Study load: 36 ects
Date: 15.12.2015

Student: Thomas Peterhans

Registration number:
Study programme: MAM (Flexible master's program)
Report number: 041BCT

Supervisor(s): Dr. ir. Karel Keesman

Examiners: Dr. ir. Gerard van Willigenburg
Group: Biobased Chemistry and Technology
Address: Bornse Weilanden 9
6708 WG Wageningen
the Netherlands
Tel: +31 (317) 48 21 24
Fax: +31 (317) 48 49 57
Table of Contents
Abstract..................................................................................................................................1
1 Introduction.........................................................................................................................2
2 Materials and methods........................................................................................................3
2.1 Material flow analysis...................................................................................................3
2.2 Aquaculture..................................................................................................................4
2.3 Hydroponics.................................................................................................................4
2.4 Aquaponics..................................................................................................................4
2.5 Water quality................................................................................................................5
2.6 System imports and exports........................................................................................5
2.7 Assumptions.................................................................................................................5
2.8 Model equation............................................................................................................7
3 Material flow analysis..........................................................................................................7
3.1 Aquaculture..................................................................................................................7
3.2 Hydroponics...............................................................................................................10
3.3 Aquaponics................................................................................................................10
3.3.1 Decoupled aquaponic systems..........................................................................12
3.3.2 Coupled Aquaponics..........................................................................................13
4 Discussion.........................................................................................................................14
5 Conclusion........................................................................................................................16
6 References........................................................................................................................17
A Appendix tables.................................................................................................................19
A.1 Feed analysis............................................................................................................19
A.1.1 Premix analysis..................................................................................................19
A.1.2 Sulphur analysis based on feed protein composition........................................19
A.1.3 Sodium chloride analysis...................................................................................20
A.1.4 Feed substance overview..................................................................................20
A.1.5 Used feed substance composition.....................................................................22
A.2 Faeces composition..................................................................................................23
A.2.1 Faeces composition overview............................................................................23
A.2.2 Used faeces composition...................................................................................24
A.2.3 Faeces dry weight factor....................................................................................24
A.3 Tilapia body composition...........................................................................................25
A.3.1 Reported body compositions for Nile tilapia......................................................25
A.3.2 Used body composition for Nile tilapia...............................................................26
A.4 Overall substance partition........................................................................................27
A.5 Water quality concentrations in fish rearing systems................................................29
A.5.1 RAS water concentration limits..........................................................................29
A.5.2 RAS minimal discharge requirement.................................................................30
A.5.3 Daily supplied substance mass by feed.............................................................30

I
A.6 Hydroponics...............................................................................................................31
A.6.1 Hydroponic water concentration limits...............................................................31
A.6.2 Hydroponic substance uptake............................................................................32
A.6.3 Predicted plant uptake.......................................................................................33
A.7 pH control table.........................................................................................................34
A.8 Decoupled System....................................................................................................35
A.8.1 RAS effluent concentration................................................................................35
A.8.2 Decoupled hydroponics uptake.........................................................................36
A.9 Coupled aquaponic water concentrations.................................................................37
A.9.1 Non-accumulation discharge requirement.........................................................37
A.9.2 Accumulation discharge requirement.................................................................37

II
Abstract
In recirculation aquaculture it is important to keep the amount of water discharged, as
small as possible for environmental reasons, which can be achieved by introducing
nutrient absorption or conversion. By including plants (e.g. tomatoes) for nutrient uptake,
such a system is called aquaponics. There are two different ways to assemble an
aquaponic system: either coupled or decoupled.
This study is showing the differences of the nutrient flows inside such systems and their
behaviour on a daily basis. Currently available knowledge in literature was used to model
material flows in an aquaponic system. Based on the given feed input, the necessary area
for soilless plant cultivation (hydroponics) was calculated and incorporated with an
recirculating aquaculture system.

1
1 Introduction
To increase sustainability, the use of water in aquaculture has to be reduced either through
integration of other trophic levels or through additional water treatments (Martins et al.,
2010). “Aquaponics” is the integration of usually two trophic levels into one system, where
one level consists of fish and the other of hydroponic plants (Goddek et al., 2015; Rakocy
et al., 2006). The direct re-use of the aquaculture effluent and incorporation with growing
plants in an aquaponic system is considered part of the future of European aquaculture
(Aller, 2015). Currently there are two different system designs known (see fig. 1). “Coupled
systems” consist of one connected water layer like the UVI system (Rakocy et al., 2006),
while “decoupled systems” consist of separated aquaculture and hydroponic systems with
a controlled connection in between (Goddek et al., 2015).

Figure 1: a) coupled b) decoupled aquaponic system

In aquaculture the discharge of water is considered to be an environmental problem which

can be tackled by the use of recirculating aquaculture systems (RAS) leading to a
reduction of the amount of water per kilogram of feed (Martins et al., 2010). A further
reduction can be achieved by integrating greenhouse technology to improve the overall
nutrient use efficiency (Kloas et al., 2015; Oberdieck and Verreth, 2009).
The aim of this study is to present and analyse the mass balance inside aquaponic
systems with a material flow analysis approach, to better understand the resulting
behavior. For this a daily interval has been chosen. Based on available literature a daily
material flow analysis (MFA) was created (Brunner and Rechberger, 2004), to investigate
the nutrient development inside such systems.
With a daily material flow analysis the spatial hydroponic requirement connected to a RAS
has been determined, after a literature study to evaluate the nutrient behaviour in an
aquaculture system. These results are used to make a mass balance between the
resulting nutrients in the water with the uptake of the hydroponic plants. According to the
given production plan an outlook for further investigations is made, to improve the system
behaviour and the understanding of it.
Starting with an analysis of the fish feed and its nutrient partition into faeces, fish uptake
and release into the water, the mixture of the available substances is determined. Based
on this and the plant uptake in the hydroponics, the required area for the plants is
calculated based on nitrogen. Due to the expected discharge of water, an differentiation
between a coupled and a decoupled aquaponic system is made, to see, whether the
system design has an influence on the discharged water and its nutrient content.

2
2 Materials and methods
2.1 Material flow analysis
Material flow analysis (MFA) is an assessment tool for the investigation of compounds
based on the principle of conservation of matter. The objective of MFA is to identify
material flows and stocks inside a system and increase the understanding of such, to
provide a basis for decision making (Brunner and Rechberger, 2004). This study covers
the cycle from system import to system export with an MFA approach to differentiate
between coupled and decoupled system designs (Goddek et al., 2015) and to identify the
advantages or disadvantages of either design based on literature data. The fish and plants
are treated as sinks, which take up nutrients and store them. Energy is not considered in
this study (see fig. 2).
Based on the systems of ASTAF-PRO and INAPRO the selected fish species is Nile tilapia
(Oreochromis niloticus) (Kloas et al., 2015; Slinkert et al., 2015). While the values of the
temperature and pH are not relevant for the developed model, they can be expected to be
between 24 and 32 °C (DeLong et al., 2009; Eding et al., 2006) and the pH between 6 and
9 (DeLong et al., 2009). Like the system of ASTAF-PRO, in the greenhouse environment
tomato plants of the species Solanum lycopersicum are used (Kloas et al., 2015).
The plant nutrients consist of 16 chemical elements which can be differentiated in
macronutrients (N, K, Ca, Mg, P, S) and micronutrients, also called trace elements (Cl, Cu,
Mn, Fe, Zn, Co, Mo, Ni) and sodium and silicon (Maathuis and Diatloff, 2013). Due to the
limitation of the used dataset for the tomato substance uptake to N, S, P, K, Ca, Mg and
water (Voogt, 1993), it has been enhanced by sodium and chloride, because of their
importance in tilapia fish feed (Cnaani et al., 2010).
Sodium and chloride can have either beneficial (Rush and Epstein, 1981) or lethal effects
on tomato plants (Rush and Epstein, 1976). In the MFA nutrients are called substances,
which include chemical substances and compounds (Brunner and Rechberger, 2004). This
study lays focus on the chemical substances (e.g. nitrogen) and does not incorporate
different chemical species (e.g. nitrate, ammonia), although the nitrification conversion of
these substances is incorporated because of its system importance to maintain the pH
level. All compounds are considered to be a good, in the sense of having a certain
economic value (Brunner and Rechberger, 2004)

Figure 2: Material flow analysis of aquaponic systems

3
For the aquacultural system and the hydroponics, the water is essential and the used
water (system import) is assumed to be clear of any substances. Therefore, this study
focuses on water quantity and quality, in terms of substances, as the connecting link
containing the substances while being substance (H 2O) and good, and not on the yields of
the respective systems (plants or fish). The nutrients are assumed to be equally distributed
inside the water body of the respective system.

2.2 Aquaculture
The nitrogen in aquacultural systems in form of ammoniacal nitrogen, is toxic to tilapia at
very low levels (Timmons and Ebeling, 2010). That is why RAS contain a nitrifying bio-filter
to convert the total ammoniacal nitrogen (TAN) into less toxic nitrate (NO 3) (DeLong et al.,
2009; Eding et al., 2006; Goddek et al., 2015; Masser et al., 1999; van Rijn et al., 2006) .
The resulting release of H +-ions by the nitrification process of Nitrosomonas and
Nitrobacter species, leads to a drop in pH for which a compensation is necessary (Eding et
al., 2006; Masser et al., 1999; Tyson et al., 2011). This drop in pH and the discharge of
water from the system depend on the amount of food consumed by the fish (Einen et al.,
1995). Due to the nitrogen uptake by the bacteria a factor of 0.98 is used instead of 1.00,
and for hydrogen release 1.98 instead of 2.00 (Eding et al., 2006). In this analysis, the
nutrient solution (a solution of substances in a good) provided to the plants is defined by
the RAS effluent.

2.3 Hydroponics
Nutrient film technique (NFT), aeroponics and continually aerated nutrient solution are
viable growing techniques for systems with a water based solution (Larsen, 1982).
ASTAF-PRO and INAPRO operate their hydroponic part of the system with NFT (Kloas et
al., 2015; Slinkert et al., 2015). According to Sprengel's law of the minimum, the deficiency
of one required mineral prevents further development, even if all other substances are
abundantly available (van der Ploeg et al., 1999). To counteract such a shortage, fertilizer
would be needed. In this model it is assumed, that a shortage of substances does not
affect the uptake of the plants, why a linear uptake is incorporated independent of changes
in the substance concentration, as well as time depending nutrient uptakes.

2.4 Aquaponics
Aquaponics is the integration of fish culture with hydroponics (Goddek et al., 2015), while
other trophic level combinations also exist (Nobre et al., 2010). The binding link between
the systems is the water body, also called effluent or discharge water on the RAS side
(Eding et al., 2006), and is an ingoing flow in the hydroponic part, called nutrient solution
(Goddek et al., 2015). The difference of coupled and decoupled systems consists in the
control of the water flow from the RAS to the plants (Goddek et al., 2015; Kloas et al.,
2015). A coupled system, such as the UVI system, has the hydroponic part integrated in
the circuit (Rakocy, 2012), while in the decoupled system the hydroponic part is separated
from a RAS with a one-way valve (Kloas et al., 2015). In the coupled system, plants
directly remove the substances from the water. But in a decoupled system, the amount of
water, and subsequently the substances, is controlled by a valve.

4
2.5 Water quality
Water quality parameters are usually given by concentrations (mg / L), except salinity
which is often given in percent or parts per thousand (ppt) (table 1) (Kamal and Mair, 2005;
Timmons and Ebeling, 2010). In a mass balance these information have to be converted
into actual masses. In this study, the aquaponic system is based on a RAS with 40 m 3
volume, which is kept constant through make-up water (Slinkert et al., 2015). For the
hydroponics a maximal volume of 10 m 3 is given (Slinkert et al., 2015). Due to the
fluctuation of the water because of its evaporation or plant transpiration
(evapotranspiration), the volume of the hydroponic basin changes over the day (Goddek et
al., 2015; Seawright et al., 1998; Slinkert et al., 2015).
RAS Hydroponics
Substance Maximal Reference Maximal Reference
conc. conc.
[mg/L] [mg/L]
N 100 Eding et al., 2006 434 Kipp, 1997
K 106 Goddek et al., 2014 414 Kipp, 1997
Ca 180 Goddek et al., 2014 533 Kipp, 1997
Mg 44 Goddek et al., 2014 158 Sonneveld and Voogt, 2009
P 17 Goddek et al., 2014 62 Sonneveld and Voogt, 2009
S 50 Timmons and Ebeling, 2010 289 Kipp, 1997; Sonneveld and
Voogt, 2009
Cl 18200 Kamal and Mair, 2005 531 Kipp, 1997
Na 11820 Kamal and Mair, 2005 275 Kipp, 1997
Table 1: Water quality requirements for RAS and tomatoes in hydroponics
Depending on the aquaponic system system design, the relevant water quality constraint
depends on the overall minimal value (coupled system) or can be differentiated between
fish tanks and hydroponics (decoupled system). The water quality is assumed to be stable
and not to have any internal processes like precipitation.

2.6 System imports and exports

The import and export processes are limited to the fish feed, pH control, faeces removal
and the water, transferred from the aquaculture system to the hydroponics. This study
limits the focus on the macro- and micronutrients (N, K, Ca, Mg, P, S, Cl, Na) inside the
water which are brought in through the fish feed and taken up by the plants. Other
nutrients from gaseous sources (e.g. oxygen, carbon dioxide, elementary nitrogen) are not
part of the MFA, likewise energy use (e.g. heating, cooling, light), is not taken into account.

2.7 Assumptions
The aquaponic system is driven by the feed input for the fish. This amount is assumed to
be consumed with the same FCR over all cohorts or size classes. Additionally the partition
of the substances into uptake, faeces and water (see table 2) are assumed to be constant
overall sizes, without any leeching of the feed. In our analysis the faeces are assumed to
be removed from the system. Thus, all faecal substances are removed by a solids removal
treatment, including the suspended solids. Additionally, the feed spills are assumed to be
zero.
The selected values for temperature and pH of the aquaponic system have no importance

5
for the model itself, while maintaining the pH within certain boundaries is essential for the
fish and plants. The added substances to maintain the pH (ph control) are included in the
MFA for nitrogen. Other substances which might affect the pH are not included in this
study.
Substance Feed content Reference
N 51.80 g/kg feed Moccia et al., 2007
K 2.00 g/kg feed Shiau and Hsieh, 2001
Ca 8.00 g/kg feed Köprücü and Özdemir, 2005
Mg 1.80 g/kg feed Moccia et al., 2007
P 6.83 g/kg feed Guimarães et al., 2008
S 2.83 g/kg feed Köprücü and Özdemir, 2005
Cl 18220.00 g/kg feed Cnaani et al., 2010
Na 11780.00 g/kg feed Cnaani et al., 2010
Substance Body content Reference
N 36.23 g/kg BW Gonzales and Brown, 2006
K 0.06 g/kg BW Gonzales and Brown, 2006
Ca 4.76 g/kg BW Gonzales and Brown, 2006
Mg 0.13 g/kg BW Gonzales and Brown, 2006
P 0.26 g/kg BW Gonzales and Brown, 2006
S 2.45 g/kg BW Köprücu and Özdemir, 2005
Cl - g/kg BW
Na 0.39 g/kg BW Gonzales and Brown, 2006
Substance Faeces content Reference
N 28.30 g/kg DW Naylor et al., 1999
K 1.00 g/kg DW Naylor et al., 1999
Ca 6.53 g/kg DW Köprücü and Özdemir, 2005
Mg 5.30 g/kg DW Naylor et al., 1999
P 6.69 g/kg DW Köprücü and Özdemir, 2005
S 0.38 g/kg DW Köprücü and Özdemir, 2005
Cl - g/kg DW Naylor et al., 1999
Na - g/kg DW Naylor et al., 1999
Table 2: The used feed, body and faeces composition (see tables A.1.5, A.2.2 and A.3.2)

Based on the system design of a coupled aquaponic system, it is not possible to maintain
different levels of pH for a coupled system. Therefore a pH change for the hydroponic part
is not included, despite this is possible in the decoupled system. Due to the scarcity of
detailed time differentiating nutrient uptakes of plants and nutrient supply from the fish in
aquaponic systems, a 24 hours time frame is used for the mass balances. Due to the low
TAN tolerance of the fish, it is assumed to be fully converted into nitrate within 24 hours.
Because of the scarcity of information about the detailed partition of the single nutrients for
the uptake of the plants, a constant nutrient ratio is assumed, independent of the
development stage. This study does not cover any energy balance. Temperature and light
conditions are assumed to be in the optimal range at all times, thus not restricting the
growth of fish and plants. Oxygen supply and degassing are not covered in this study, as
they would need the inclusion of gaseous balances. Likewise, and for a fair comparison
between coupled and decoupled systems, any addition of fertilizer or minerals to change
the conductivity are excluded, as well as pH changes by the plants and water re-use in the
decoupled system.

6
2.8 Model equation
The material flow model is based on the conservation of mass without temporal storage
and is given by m
m water =m feed− body −m faeces∗F faeces DW (1)
FCR

The mass mwater released to the water [g / kg feed] equals the imported m feed [g / kg feed]
minus the partitioned masses of m body / FCR [g / kg Bodyweight / (g Feed / g Bodyweight)],
where the FCR is needed to convert the bodyweight (BW) into the dry weight mass of the
feed, minus the mass of the faeces mfaeces [g / kg faeces wet weight] multiplied by its dry-
weight factor [g faeces wet weight / kg feed].
For the RAS, the masses (see table 2) of the feed (m feed,), the body composition (mbody), the
fish faeces (mfaeces) have to be balanced. Due to the distribution of substances inside the
fish, the FCR is used to distinguish between dry weight of the feed and the body weight of
the fish. Additionally the dry weight factor for the faeces (F faecesDW) has been determined, to
match the substances of the feed, to the substance content of the faeces (Rafiee and
Saad, 2005). With FCR = 1.11 (Kamal and Mair, 2005) and FfaecesDW = 0.214 (see table
A.2.3) (Rafiee and Saad, 2005) being constants, the masses of each of the substances,
have to be conserved. As the feed is pelleted, the moisture content is expected to be close
to zero, thus the feed dry weight is assumed to equal the fish feed.

3 Material flow analysis

A material flow analysis of an aquaponic system can be done for different time durations
(e.g. day, per cohort, year, production cycle). The best choice of the time frame to look at,
depends on the tasks which have to be performed and the available data to incorporate
into the balance. Due to the scarcity of knowledge about detailed processes of the
digestion in fish and the uptake in plants, as yet a daily interpretation is appropriate.

3.1 Aquaculture
The daily system imports of the RAS are water, fish feed and a base for pH control. The
daily exports are water with its soluble and particulate compounds and solids in the form of
faeces (see fig. 2). The feed conversion ratio (FCR) is the reported mean at 0 ppt salinty of
different tilapia species (Kamal and Mair, 2005). The feeding protocol is based on
published data of a feed company (Coppens international bv, The Netherlands,
http://www.coppens.eu) and the unpublished production plan of the INAPRO project.
Currently there is no detailed faeces analysis for tilapia available, why the findings in
rainbow trout are used as a starting point (Moccia et al., 2007; Naylor et al., 1999). Due to
the difference in macronutrient recommendation per species (Figueiredo-Silva et al.,
2013), these data have been adapted with other findings in Nile tilapia (Cnaani et al.,
2010; Kandeepan, 2013; Köprücü and Özdemir, 2005; Moccia et al., 2007; Ng and
Romano, 2013; Robinson et al., 1987; Shiau and Hsieh, 2001). Also there is no specific
dataset available for the conversion of the substances of the diet weight into the respective
dry matter weight for Nile tilapia, therefore it is assumed to be equal over all nutrients.
The fish take up nutrients through the ingested feed (Clement and Lovell, 1994; Dale et al.,

7
2004; Gonzales and Brown, 2006; Köprücü and Özdemir, 2005). For the protein nitrogen
conversion the Kjeldahl method has been used (Eding et al., 2006; Köprücü and Özdemir,
2005). Due to the differences in reported sulphur body content, the smallest one has been
used (Gonzales and Brown, 2006; Köprücü and Özdemir, 2005), while other substance
values have been calculated from the corresponding mass balance. By applying the mass
balance formula (Eq. 1) to all documented nutrients, the following partition results (see fig.
3 and table A.4 for numerical results).
100%

90%

80%

70%

60%
Water
50%
Faeces
40% Fish

30%

20%

10%

0%
N K Ca Mg P S Cl* Na*
Figure 3: Feed substance partition into fish, faeces and water for N, K, Ca, Mg, P, S, Cl
and Na. (*) incomplete dataset
The results of the partition for nitrogen differ from literature for tilapia (Endut et al., 2009;
Rafiee and Saad, 2005). This is a consequence of the different species used in the
studies, as trout (Moccia et al., 2007; Naylor et al., 1999) and red tilapia (Endut et al.,
2009; Rafiee and Saad, 2005) differ from each other and have a different body and faecal
composition, also Nile tilapia is expected to have a different composition.
As it can be seen for chloride, the whole mass is appearing in the water and none in the
faeces and the body (see fig. 3). The original datasets for faeces and body composition do
either not include chloride or chloride is not present in the body (Moccia et al., 2007;
Naylor et al., 1999).
According to the production plan, the average daily feed import is an average of the cohort
feed input. The cohort has a length of 45 days and the average feed import into the system
is 21.9 kg/day (fig. 4). The fluctuations of the system feed input are beyond a daily interval,
therefore these fluctuations, as well as the strong decrease on day 44 (fig. 4), are not
included in this analysis.

8
 30

kg feed/day
20

10 Feed input
Average feed input
0
0 10 20 30 40 50
day

Figure 4: Averaged cohort feed plan

It is assumed that within one day, all the ammoniacal nitrogen is converted into nitrate by
the nitrification processes in the bio-filter. Because of the release of TAN by the fish in
relation to feeding, all the nitrate has its origin in the TAN (Eding et al., 2006). Based on
the overall nitrification, the necessary pH compensation can be calculated, to keep its level
constant. Per mol of TAN, 2 mol of hydrogen ions (H+) are released (Eding et al., 2006).
Due to the bacterial biomass gain, this number is in reality slightly smaller with 1.98 mol
H/mol NH4-N (Eding et al., 2006). To counteract this alkalinity consumption, sodium
bicarbonate (NaHCO3) (Eding et al., 2006) or limestone (CaCO3) (Goddek et al., 2015) can
be used. For every alkali metal (e.g. Na +, K+, etc.) and alkaline earth metal (Mg 2+, Ca2+,
etc.) a bicarbonate and a carbonate compound exists. This degree of freedom can be
used to counteract a shortage of nutrients supplied by the RAS effluent and therefore
improve the overall suitability of the substance solution. The ratio of the nutrients is
constant, as the fish feed composition does not change (fig. 5 and table A.4).
N
K
Ca
[kg / kg feed]

Mg
P
S
Na
Cl

0.00E+00 5.00E-03 10.00E-03 15.00E-03 20.00E-03

[kg / kg feed]

Figure 5: The ratio [kg/kg feed] between the substances

The effluent provided to the hydroponics depends on the water quality requirement for the
tilapia and nutrients added through the fish feed. The minimal discharge per kilogram of
feed Qdischarge [L / kg feed], can be calculated by dividing the amount of substance per
kilogram feed msubstance [g / kg feed] by cmax,substance [g / L]. That is,
m (2)
Q discharge = substance
C max, substance
This discharge follows from a steady state mass balance and gives the minimal amount of
water needed, to transport a certain substance out of the system (fig. 6).

9
 N
K
Ca
Mg
P
S
Na
Cl

0 20 40 60 80 100 120 140

L/kg feed

Figure 6: Minimal RAS discharge requirement based on feed partition and maximal
allowable water concentration (see table A.5.2)
The requirement for discharge is because of the nitrogen (in form of nitrate) (also Eding et
al., 2006), while the other substances stay below the maximal concentration tolerable for
tilapia (see Eq. (2) and table A.5.2).
Based on the average feed input of 21.9 kg/day and the minimal discharge of 128.4 L/kg
feed, a total daily discharge of 2812.2 L/day is required for the RAS.

3.2 Hydroponics
There have been reports of greenhouse tomato yields of 56.2 kg/m 2 (De Gelder et al.,
2005). A fixed substance partition for N, S, P, K Ca, Mg and water uptake (Voogt, 1993) is
most valuable for this study, as it provides insight in the actual uptake ratio and does not
compare different substances against each other. As the model is based on a daily
material flow, the average substance uptake over the whole growth period has been taken
into account, including the water evaporation. Seasonal or daily changes have not been
taken into account. Based on the data from Voogt (1993) and De Gelder et al. (2005), a
daily uptake per m2 of 0.404 g-N, 0.110 g-S, 0.098 g-P, 0.707 g-K, 0.295 g-Ca and 0.069
g-Mg is predicted (see tables A.6.2 and A.6.3). Due to the lack of data for sodium and
chloride, no uptake is considered.
Tomatoes do not only take up nutrients, but also evaporate water in which the nutrients are
solved. The amount of evaporated water is assumed to be 2.9 mm/m 2/day. Due to the high
sensitivity of tomatoes to sodium and chloride in the provided substance solution, these
nutrients are the drivers for the required discharge. Based on maximal allowable
concentrations in the hydroponic system, the discharge driver is sodium (274.8 mg / L) or
chloride (531.0 mg / L). The nutrients provided and the maximal allowable concentration,
allow the calculation of the minimal required discharge (Eq. 2), resulting in a minimal
discharge for chloride of 751 L / day and for sodium of 911 L / day. Thus, every day
approximately 1 m3 waste water has to be discharged.

3.3 Aquaponics
Given the mass of the substances in the effluent from the RAS and the substance uptake
of the tomatoes, the spatial requirement for the hydroponic area can be calculated from

10
 msubstance , effluent (3)
A substance , req=
m substance ,uptake
where Asubstance,req [m2] is the minimal spatial requirement the substance mass in the effluent
msubstance,effluent [g / day] and msubstance,uptake [g / m2 / day] the uptake of the substance by the
plants.
Substance RAS Predicted Spatial
effluent uptake requirement
[g/day] [g/m2/day] [m2]
N 281.2 0.4 695.4
K 38.0 0.7 53.7
Ca 50.7 0.3 171.8
Mg 12.1 0.1 175.6
P 25.4 0.1 258.5
S 11.9 0.1 108.6
Table 3: Minimal spatial requirement per substance (sodium and chloride have been left
out, as they are not taken up by the model)

From the spatial requirement (see table 3) it can be seen, that the effluent of the RAS is
extremely short in potassium, compared to other substances, especially nitrogen. The
same has been reported earlier in other studies (Graber and Junge, 2009; Kloas et al.,
2015).
The sensitivity of tomatoes to salinity is depending on the cultivar, but also on the ratio of
the available substances (Satti and Al‐Yahyai, 1995). To improve the fitting of the RAS
effluent to the tomato uptake, pH control can be used. Due to the decreased calcium
content of tomato fruits with increased salinity (Satti and Al‐Yahyai, 1995), the use of
sodium bicarbonate (NaHCO3) does not seem useful, as it would increase the sodium
content of the solution. Potassium bicarbonate seems like the best choice, to improve to
overall ratio of the solution.
A daily amount of 281.2 g-N is supplied to the RAS in the form of fish feed, which will result
as TAN in the water. The conversion of ammonia into nitrate requires an pH compensation,
to prevent the pH from dropping, due to the H +-ion release of the nitrification process.
Based on the atomic weight of nitrogen of 14 g / mol and the daily input of 281.2g-N / day,
daily 20.09 mol-N/day have to be converted. Due to the biomass of the nitrifying bacteria,
an alkalinity compensation 1.98 mol-H / mol-N has to be introduced (instead of 2.00)
(Eding et al., 2006). Thus 39.78 mol-H / day have to be bound. Given the atomic weight of
potassium of 39.1 g / mol, potassium bicarbonate (KHCO 3) is used in this study, which has
a molar weight of 100.1 g / mol. The addition of 39.8 mol equals a total weight of 3982.1 g-
KHCO3 / day, which adds 1555.2 g-K / day. As an alternative also magnesium carbonate
might be added, which results in 1676.7 g-MgCO 3 / day or 483.3 g-Mg / day, given the
atomic weight of magnesium of 24.3 g / mol. The shortage of potassium, calcium and
magnesium can thus be counteracted by strategic choosing of (bi-)carbonate compounds
(see table A.7).

11
 600
500
400

[g/day]
300
200 RAS effluent
100 Plant uptake
0
N K Ca Mg P S
[g/day]

Figure 7: Daily system substance balance (see tables 4 and A.5.3)

Based on the findings for nitrogen, a spatial area of the hydroponic system of 695 m 2 was
calculated, under the assumption that the shortages of substances are not limiting for the
plants (see table 4 and fig. 7).
Substance Uptake Area Daily uptake
[g/m^2/day] [m^2] [g/day]
N 0.404 695 281.1
S 0.110 695 76.3
P 0.098 695 68.4
K 0.707 695 491.4
Ca 0.295 695 204.9
Mg 0.069 695 47.7
Water 2920.030 695 2029420.9
Table 4: Daily substance uptake by the plants

3.3.1 Decoupled aquaponic systems

Based on the discharge requirement for nitrogen and a maximal nitrogen concentration of
100 mg-N/L in the RAS (40 m 3), a total of 4000 g-N (100 mg-N / L * 40000 L) can
accumulate. With 12.8 g-N / kg feed a total of 311.5 kg feed (4000 g-N / 12.8 g-N / kg
feed) can be added into the RAS before discharge to the decoupled hydroponic system is
required, which corresponds the feeding of 14.2 days (311.5 kg feed / 21.9 kg feed / day).
Due to the constant water replacement and discharge, the composition of the effluent is
constant, based on the assumption of a maximal concentration of nitrogen and using Eq.
(2), a discharge of (128.4 L / kg feed * 21.9 kg feed / day =) 2812.2 L/day from the RAS is
needed (table A.5.2).
Substance Substance Daily feeding System Days RAS Discharge
added netto volume concentration
[g/kg feed] [kg feed/day] [m3] [day] [mg/L]
N 12.8 21.9 40.0 14.2 100.0
K 1.7 21.9 40.0 14.2 13.5
Ca 2.3 21.9 40.0 14.2 18.0
Mg 0.6 21.9 40.0 14.2 4.3
P 1.2 21.9 40.0 14.2 9.0
S 0.5 21.9 40.0 14.2 4.2
Na 11.4 21.9 40.0 14.2 89.0
Cl 18.2 21.9 40.0 14.2 141.8
Table 5: Substance composition of the RAS effluent in the decoupled aquaponics

12
All the values of the effluent are within the water quality requirements of the hydroponics
(see table A.8.2). With the above mentioned concentrations, it can be expected to get a
daily flow of effluent from the RAS of 2812.22 L / day, as shown earlier. The amount of
water left in the hydroponic tank (2812 L – 2029 L = 783 L) requires additional water to not
exceed the maximally allowed concentrations for sodium in the plant nutrient solution (see
table 1), which requires a minimal discharge of 911 L/day to prevent accumulation (see
table 6). Thus (911 L – 783 L = ) 128 L of water have to be added to the hydroponics to not
exceed the maximal allowable sodium concentration.
Substance Substance Maximal Volume Substance Days before Minimal
amount concentration system mass discharge discharge
volume
[g/day] [mg/L] [L] [g] [days] [L]
Na 250.2 274.8 40000.0 10992.0 43.9 910.6
Cl 398.9 531.0 40000.0 21240.0 53.3 751.2
Table 6: Minimal discharge volume for a coupled aquaponic system based on sodium and
chloride accumulation
A maximum of 783 L / day * 274.8 mg-Na / L = 215.2 g-Na / day can be discharged by the
given water volume. The minimal water discharge for a decoupled aquaponic system with
the given substance composition for the fish feed is 911 L / day / 21.9 kg feed / day = 41.6
L / kg feed due to sodium.

3.3.2 Coupled Aquaponics

Based on the findings for the decoupled system, the substance concentration of the water
is the same after 14.22 days. Also the plant uptake is the same, as the hydroponic area is
also 695 m2. Based on the assumption that discharge of the sodium and chloride takes
place before they accumulate, the discharge of the system would need at least 2711.4 L /
day (see table A.9.1), which is similar to the standalone RAS discharge requirement. In an
aquaponic system the cause of the discharge is sodium, while in the RAS it is nitrate
(Eding et al., 2006). To improve the situation (see tab. 7), the accumulation of sodium and
chloride has to be tolerated.
Substance Maximal daily Theoretical Loss per day
Concentration discharge
[mg/L] [L/day] [g/day]
N 93.0 2812.2 261.4
K 11.6 2812.2 32.6
Ca 16.3 2812.2 45.8
Mg -8.0 2812.2 -22.5
P 3.9 2812.2 11.0
S 3.0 2812.2 8.6
Na 89.0 2812.2 250.2
Cl 141.8 2812.2 398.9
Table 7: Lost substances in a coupled aquaponic system without sodium accumulation

Accumulating sodium and chloride before discharge, increases efficiency due to lower
water usage. This has a negative effect on the tomatoes, as they prefer as little sodium
and chloride as possible (see tables A.6.1a and A.6.1b) (Komosa and Górniak, 2015; Satti
and Al‐Yahyai, 1995). This would result in a daily discharge of 2711.4 L/day.

13
Due to the requirement of minimal water level in a coupled aquaponic system because of
the fish, a minimal water strategy (complete discharge) as in the decoupled system is not
possible. Thus an accumulation of sodium and chloride is necessary, to keep the
discharge as little as possible ('little' means as much substance per litre as possible).
The analysis for a coupled system with sodium an chloride accumulation shows, that after
a maximal period of 44 days the discharge is necessary, to keep the sodium level below
the maximal concentration (see tab. A.9.2). At this concentration, a minimum of 911 L/day
is required to discharge the daily imported sodium, which equals 41.6 L / kg feed.

4 Discussion
The material flow analysis has proven to be useful to investigate the aspects needed in
aquaponic systems. Due to the limited number of inputs (fish feed, water, pH control) in
such systems, the composition of each is extremely important. The data needed to form
such a complete analysis is scarce, but results from Kloas et al. (2015) show similarity to
the results of this study. To further improve the used numbers, experiments are needed to
identify further differences in the known system designs. The dry weight factor of the
faeces (Rafiee and Saad, 2005) and the composition of the faeces (Moccia et al., 2007;
Naylor et al., 1999) are based on single datasets (Eq. 1), thus these may not be accurate
for this analysis. The model (see table A.4) results in some compounds (Fe, Mn, Si, B, Se)
in negative mass balances, due to either errors of the measurements or the differences in
fish species.
In general we know, that the FCR is highly dependent on the used feed, while the faeces
depend on the used binder in the fish feed. Thus, the factors for FCR and faeces dry
weight in the mass balance (Eq. 1) contain uncertainties and need verification for Nile
tilapia through experiments. Additional knowledge from fish nutrition might also affect the
substances in the body composition, due to their high plasticity (see table A.3.1). But the
actual origin of the plasticity is at the moment unknown. The big number of substances
requires a validation for each of the substances.
In this study the water processes have been excluded. It is necessary to include further
details about the behaviour of substances in the water. Due to a continuous water flow,
changes can be expected in different system parts depending on the water flow. Solids
removal is part of the RAS. But we know, that faeces can decompose if they are not
removed in short time and affect the function of the bio-filter. Thus the removal of the
faeces is important. The necessary pH control for counteracting the bio-filter conversion
enables some degree of freedom in the control of the substance solution. But further
knowledge in plant-fish interaction is necessary to investigate the interaction between the
faecal treatment and the plants inside an aquaponic system to include it in a model.
Most of the hydroponic research is based on controlled nutrient solutions without any
incorporation of suspended solids. The availability of the nutrients from the RAS has to be
verified, as this study does not focus on the different chemical species. This difference
might make it necessary to evaluate other approaches than NFT, such as aeroponics, to
provide the plants the needed substances that might be beneficial as a response to the
high sodium and chloride content. The results of this study show the influence of the

14
system design on the sodium discharge requirement and might be further improved to
reduce the necessary water discharge. Because of the trade-off between water usage and
high sodium and chloride concentration (salinity), the overall development of the minimal
water discharge in the coupled and decoupled system is with 911 L / 21.9 kg feed = 41.6
L / kg feed higher, than reported in current low water exchange RAS (30 L / kg feed), which
include denitrification (Martins et al., 2009). Because of the linear relationship between the
substance concentrations, a lower maximal daily concentration of sodium can be achieved
by increasing the spatial area of the plants (with addition of nitrogen fertilizer), a higher
allowable maximal concentration for sodium or a reduction of the sodium content of the
fish feed. Due to the link of the FCR to sodium chloride (Cnaani et al., 2010), a new feed
composition also affects other aspects of the system design. It is also necessary to
consider other fish species with less sodium chloride affinity to prevent these imports to
the system (e.g. rainbow trout (Moccia et al., 2007)).
The system design of the decoupled aquaponic system can be used to provide the
hydroponics a higher concentrated nutrient solution with less salinity by accumulating the
nutrients before the discharge to the hydroponics, although this requires make-up water to
compensate for the evaporation. In the coupled system there is no system differentiation
between RAS and hydroponics, as they share the common water layer. In both system
types the pH control provides a degree of freedom to steer the nutrients provided to the
plants beyond the fish feed. Further degrees of freedom may be in the pH relevant
processes of the plants, which have been excluded from the model.
By providing the best possible nutrient solution, the plants are assumed to grow ideally as
the deficit of nutrients is as little as possible. With ideal growth, also the best possible
nutrient removal and assumed yield should be achieved. By properly selecting the pH
control with calcium, potassium and magnesium, additional fertilizer can be limited to
sulphur and phosphorous, which are the second and third limiting nutrients (see fig. 7).
Although at the moment the ideal composition of the selected compounds for pH control
has to be determined through experiments. If the provided nutrient solution meets the
needs perfectly, the environmental impact of the discharged water is assumed to be
minimal (only sodium and chloride), due to maximal nutrient removal. This removal
increases profits as they are converted into yields and decrease fertilizer use.
The maximal allowable concentrations are based on literature, which may not be suitable
for aquaponic systems. To further decrease water usage (41.6 L / kg feed), more detailed
knowledge is needed for the water quality requirements of the single parts (fish and
plants). Especially in the decoupled system such information is needed for sodium, to
prevent the use of additional water. Without make-up water need because of sodium, the
achievable water discharge is 783 L / day / 21.9 kg feed / day = 35.8 L / kg feed based on
the given maximal allowable nitrate concentration.
The fish feed (fish production plan), the pH control and the spatial requirement of the
hydroponics share a linear relationship (see Fig. 7 and tables 1 and A.7). Thus the size of
a system is defined by the amount of fish feed per day. The harmonization of the RAS
(feed and pH control) to the nutrient requirements of the plants (including fertilizer) is a
requirement on system design level independent of coupled of decoupled system design.

15
With the conditions of this study, a system recommendation on water usage is not
possible. But for the substance loading of the discharged water, the decoupled system is
better (see tables A.8.2 and A.9.1), as all substances except sodium and chloride are
removed by the plants (table 8).
Substance Coupled Dcoupled
[mg/L] [mg/L]
N 97.9 0.1
K 12.2 -42.1
Ca 17.2 -19.5
Mg -8.4 -526.2
P 4.1 -197.0
S 3.2 -39.3
Na 93.7 274.7
Cl 149.4 437.8
Table 8: Discharge concentrations of aquaponic systems (negative values represent
depleted substances)

5 Conclusion
Hydroponics are a viable way to reduce substance loading of RAS effluents. The
discharged concentrations are depending on the system design and further investigation of
the detailed substance behaviour is needed, to fully understand the system. Many details
of internal processes are currently unknown. To further reduce the discharged water, the
development of aquaponic fish feed is necessary to optimise the nutrient composition of
the fish to the tomato plants. More substances have to be incorporated in models and
experiments, to identify critical system substances besides sodium and chloride.
Understanding the internal processes requires more research on fish nutrient behaviour
and experiments on nutrient interaction in the water depending on the nutrient
concentrations. Future research should focus on system design, fish feed composition and
plant uptake to improve the overall performance of aquaponic systems, including pH
control to steer the nutrient solution.

16
6 References
Aller, J.G., 2015. The future of European aquaculture (DRAFT OPINION). Bruxelles.
Brunner, P.H., Rechberger, H., 2004. Practical handbook of material flow analysis, Practical handbook of material flow
analysis. CRC Press LLC, Boca Raton, Florida. doi:10.1016/j.wasman.2004.11.002
Clement, S., Lovell, R.T., 1994. Comparison of processing yield and nutrient composition of cultured Nile tilapia
(Oreochromis niloticus) and channel catfish (Ictalurus punctatus). Aquaculture 119, 299–310. doi:10.1016/0044-
8486(94)90184-8
Cnaani, A., Barki, A., Slosman, T., Scharcanski, A., Milstein, A., Harpaz, S., 2010. Dietary salt supplement increases the
growth rate in freshwater cultured tilapia hybrids. Aquac. Res. 41, 1545–1548. doi:10.1111/j.1365-
2109.2009.02438.x
Dale, N.M., Zumbado, M., Gernat, A.G., Romo, G., 2004. Nutrient value of tilapia meal. J. Appl. Poult. Res. 13, 370–372.
De Gelder, A., Heuvelink, E., Opdam, J.J.G., 2005. Tomato yield in a closed greenhouse and comparison with simulated
yields in closed and conventional greenhouses. Acta Hortic. 691, 549–552.
DeLong, D.P., Losordo, T.M., Rakocy, J.E., 2009. Tank Culture of Tilapia, SRAC publication No. 282.
Eding, E.H., Kamstra, A., Verreth, J.A.J., Huisman, E.A., Klapwijk, A., 2006. Design and operation of nitrifying
trickling filters in recirculating aquaculture: A review. Aquac. Eng. 34, 234–260.
doi:10.1016/j.aquaeng.2005.09.007
Einen, O., Holmefjord, C., Asgard, T., Talbot, C., 1995. Auditing nutrient discharges from fish farms: theoretical and
practical considerations. Aquac. Res. 26, 701–713.
Endut, A., Jusoh, A., Ali, N., Wan Nik, W.N.S., Hassan, A., 2009. Effect of flow rate on water quality parameters and plant
growth of water spinach (Ipomoea aquatica) in an aquaponic recirculating system. Desalin. Water Treat. 5, 19–28.
doi:10.5004/dwt.2009.559
Figueiredo-Silva, A.C., Saravanan, S., Schrama, J.W., Panserat, S., Kaushik, S., Geurden, I., 2013. A comparative study
of the metabolic response in rainbow trout and Nile tilapia to changes in dietary macronutrient composition. Br. J.
Nutr. 109, 816–826. doi:10.1017/S000711451200205X
Goddek, S., Delaide, B., Mankasingh, U., Ragnarsdottir, K., Jijakli, H., Thorarinsdottir, R., 2015. Challenges of
Sustainable and Commercial Aquaponics. Sustainability 7, 4199–4224. doi:10.3390/su7044199
Gonzales, J.M., Brown, P.B., 2006. Nile tilapia Oreochromis niloticus as a food source in advanced life support systems:
Initial considerations. Adv. Sp. Res. 38, 1132–1137. doi:10.1016/j.asr.2005.11.002
Graber, A., Junge, R., 2009. Aquaponic Systems: Nutrient recycling from fish wastewater by vegetable
production. Desalination 246, 147–156. doi:10.1016/j.desal.2008.03.048
Guimarães, I.G., Pezzato, L.E., Barros, M.M., 2008. Amino acid availability and protein digestibility of several protein
sources for Nile tilapia, Oreochromis niloticus. Aquac. Nutr. 14, 396–404. doi:10.1111/j.1365-2095.2007.00540.x
Kamal, A.H.M.M., Mair, G.C., 2005. Salinity tolerance in superior genotypes of tilapia, Oreochromis niloticus,
Oreochromis mossambicus and their hybrids. Aquaculture 247, 189–201. doi:10.1016/j.aquaculture.2005.02.008
Kandeepan, C., 2013. Dietary calcium requirement of Oreochromis mossambicus. Int. J. Curr. Microbiol. Appl. Sci. 2, 89–
97.
Kipp, J.A., 1997. Voedingsoplossingen voor de teelt van tomaat in gesloten teeltsystemen. Naaldwijk.
Kloas, W., Groß, R., Baganz, D., Graupner, J., Monsees, H., Schmidt, U., Staaks, G., Suhl, J., Tschirner, M.,
Wittstock, B., Wuertz, S., Zikova, A., Rennert, B., 2015. A new concept for aquaponic systems to improve
sustainability, increase productivity, and reduce environmental impacts. Aquac. Environ. Interact. 7, 179–
192. doi:10.3354/aei00146
Komosa, A., Górniak, T., 2015. The Effect of Chloride on Yield and Nutrient Interaction in Greenhouse Tomato (
Lycopersicon Esculentum Mill.) Grown in Rockwool. J. Plant Nutr. 38, 355–370.
doi:10.1080/01904167.2014.934466
Köprücü, K., Özdemir, Y., 2005. Apparent digestibility of selected feed ingredients for Nile tilapia (Oreochromis niloticus).
Aquaculture 250, 308–316. doi:10.1016/j.aquaculture.2004.12.003
Larsen, J.E., 1982. Growers problems with hydroponics. J. Plant Nutr. 5, 1077–1081. doi:10.1080/01904168209363039
Maathuis, F.J.M., Diatloff, E., 2013. Plant Mineral Nutrients 953, 1–21. doi:10.1007/978-1-62703-152-3
Martins, C.I.M., Eding, E.H., Verdegem, M.C.J., Heinsbroek, L.T.N., Schneider, O., Blancheton, J.P., d' Orbcastel, E.R.,
Verreth, J. a J., 2010. New developments in recirculating aquaculture systems in Europe: A perspective on

17
 environmental sustainability. Aquac. Eng. 43, 83–93. doi:10.1016/j.aquaeng.2010.09.002
Martins, C.I.M., Ochola, D., Ende, S.S.W., Eding, E.H., Verreth, J. a. J., 2009. Is growth retardation present in Nile tilapia
Oreochromis niloticus cultured in low water exchange recirculating aquaculture systems? Aquaculture 298, 43–50.
doi:10.1016/j.aquaculture.2009.09.030
Masser, M.P., Rakocy, J., Losordo, T.M., 1999. Fact Sheet. J. Am. Diet. Assoc. 99, 119–120. doi:10.1016/S0002-
8223(99)00856-1
Moccia, R., Bevan, D., Reid, G., 2007. Composition of Fecal Waste from Commercial Trout Farms in Ontario : Macro and
Micro Nutrient Analyses and Recommendations for Recycling Final Report Submitted to the : Ontario Sustainable
Aquaculture Working Group Environment Canada. Management.
Naylor, S.J., Moccia, R.D., Durant, G.M., 1999. The Chemical Composition of Settleable Solid Fish Waste (Manure) from
Commercial Rainbow Trout Farms in Ontario, Canada. N. Am. J. Aquac. 61, 21–26. doi:10.1577/1548-
8454(1999)061<0021:TCCOSS>2.0.CO;2
Ng, W.K., Romano, N., 2013. A review of the nutrition and feeding management of farmed tilapia throughout the culture
cycle. Rev. Aquac. 5, 220–254. doi:10.1111/raq.12014
Nobre, A.M., Robertson-Andersson, D., Neori, A., Sankar, K., 2010. Ecological-economic assessment of aquaculture
options: Comparison between abalone monoculture and integrated multi-trophic aquaculture of abalone and
seaweeds. Aquaculture 306, 116–126. doi:10.1016/j.aquaculture.2010.06.002
Oberdieck, A., Verreth, J., 2009. SustainAqua handbook – A handbook for sustainable aquaculture.
Rafiee, G., Saad, C.R., 2005. Nutrient cycle and sludge production during different stages of red tilapia (Oreochromis
sp.) growth in a recirculating aquaculture system. Aquaculture 244, 109–118.
doi:10.1016/j.aquaculture.2004.10.029
Rakocy, J.E., 2012. Aquaponics-Integrating Fish and Plant Culture, in: Aquaculture Production Systems. Wiley-Blackwell,
Oxford, UK, pp. 344–386. doi:10.1002/9781118250105.ch14
Rakocy, J.E., Masser, M.P., Losordo, T.M., 2006. Recirculating aquaculture tank production systems: Aquaponics-
integrating fish and plant culture., SRAC publication - southern regional aquaculture center. doi:454
Robinson, E.H., LaBomascus, D., Brown, P.B., Linton, T.L., 1987. Dietary calcium and phosphorus requirements of
Oreochromis aureus reared in calcium-free water. Aquaculture 64, 267–276. doi:10.1016/0044-8486(87)90189-X
Rush, D.W., Epstein, E., 1981. Comparative studies on the sodium, potassium, and chloride relations of a wild halophytic
and a domestic salt-sensitive tomato species. Plant Physiol. 68, 1308–1313. doi:10.1104/pp.68.6.1308
Rush, D.W., Epstein, E., 1976. Genotypic Responses to Salinity: Differences between Salt-Sensitive and Salt-Tolerant
Genotypes of the Tomato. Plant Physiol. 57, 162–166.
Satti, S.M.E., Al‐Yahyai, R.A., 1995. Salinity tolerance in tomato: Implications of potassium, calcium, and phosphorus.
Commun. Soil Sci. Plant Anal. 26, 2749–2760. doi:10.1080/00103629509369484
Seawright, D.E., Stickney, R.R., Walker, R.B., 1998. Nutrient dynamics in integrated aquaculture-hydroponics systems.
Aquaculture 160, 215–237. doi:10.1016/S0044-8486(97)00168-3
Shiau, S.Y., Hsieh, J.F., 2001. Quantifying the dietary potassium requirement of juvenile hybrid tilapia (Oreochromis
niloticus x O. aureus). Br. J. Nutr. 85, 213–218. doi:10.1079/BJN2000245
Slinkert, T., Lastiri, D.R., Cappon, H., Keesman, K.J., 2015. Optimisation of water, energy and nutrient requirements in an
aquaponic system with interacting production loops, in: New Developments in IT & Water. Rotterdam, pp. 1–8.
Sonneveld, C., Voogt, W., 2009. Plant Nutrition of Greenhouse Crops. doi:10.1007/978-90-481-2532-6
Timmons, M.B., Ebeling, J.M., 2010. Recirculating Aquaculture 2nd Ed, NRAC Publi. ed. Cayuga Aqua Ventures.
Tyson, R. V., Treadwell, D.D., Simonne, E.H., 2011. Opportunities and challenges to sustainability in aquaponic systems.
Horttechnology 21, 6–13.
van der Ploeg, R.R., Böhm, W., Kirkham, M.B., 1999. On the Origin of the Theory of Mineral Nutrition of Plants and the
Law of the Minimum. Soil Sci. Soc. Am. J. 63, 1055. doi:10.2136/sssaj1999.6351055x
van Rijn, J., Tal, Y., Schreier, H.J., 2006. Denitrification in recirculating systems: Theory and applications. Aquac. Eng. 34,
364–376. doi:10.1016/j.aquaeng.2005.04.004
Voogt, W., 1993. Nutrient uptake of year round tomato crops. Acta Hortic. 339, 99–112.

18
A Appendix tables
A.1 Feed analysis
A.1.1 Premix analysis
The substance analysis of the premix gives an insight into the substances of fish feed for
Nile tilapia (Guimarães et al., 2008).
Mineral Premix
Na2SeO3 0.7 mg/kg diet 172.8 g/mol
MnO 50 mg/kg diet 70.9 g/mol
ZnO 150 mg/kg diet 81.3 g/mol
FeSO4 20 mg/kg diet 151.9 g/mol
CoSO4 0.5 mg/kg diet 155 g/mol
I2Ca 1 mg/kg diet 293.9 g/mol
NaCl 1000 mg/kg diet 58.3 g/mol
CaCO3 18500 mg/kg diet 100.1 g/mol
CaHPO4 30000 mg/kg diet 136.1 g/mol
Cr2O3 1000 mg/kg diet 152 g/mol
Sum
Na 22.9 g/mol 392.981 mg/kg diet
Se 79 g/mol 0.320 mg/kg diet
Mn 54.9 g/mol 38.717 mg/kg diet
Zn 65.3 g/mol 120.480 mg/kg diet
Fe 55.8 g/mol 7.347 mg/kg diet
S 32.1 g/mol 4.330 mg/kg diet
I 126.9 g/mol 0.864 mg/kg diet
Ca 40.1 g/mol 16250.314 mg/kg diet
Cl 35.4 g/mol 607.204 mg/kg diet
P 31 g/mol 6833.211 mg/kg diet
Cr 52 g/mol 684.211 mg/kg diet
Table A.1.1: Premix analysis

A.1.2 Sulphur analysis based on feed protein composition

Sulphur analysis based on feed protein composition, as the provided sulphur content of
the premix analysis has to be incomplete based on the mass balance assumption
(Köprücü and Özdemir, 2005).
per kg feed uptake excretion Molar weight Sulphur
% % % g/mol g/mol
Methionine 0.7 86.6 13.4 149.21 32.1
Cysteine 0.5 86.2 13.8 121.15 32.1

uptake excretionSulphur uptake Sulphur excretion

g/kg feed g/kg feed g-S/kg feed g-S/kg feed
Methionine 6.06 0.94 1.30 0.20
Cysteine 4.31 0.69 1.14 0.18
Table A.1.2: Sulphur analysis based on feed protein composition

19
A.1.3 Sodium chloride analysis
The estimated sodium chloride content of the tilapia feed is based on the findings on
improved FCR based on dietary salt supplementation. An addition of 3 % salt improved the
FCR (Cnaani et al., 2010), as the composition of the diet is assumed to be based on
current knowledge.

Molar weight Total amount

g/mol % g
Na 22.9 1.18 11.78
Cl 35.4 1.82 18.22
Sum 58.3 3 30
Table A.1.3: Sodium chloride analysis

A.1.4 Feed substance overview

The reports about substance composition for tilapia feed only cover part of the nutrients.
Thus a the knowlegde from different sources has been combined, based on the literature
(see table A.1.1, A.1.2 and A.1.3) (Cnaani et al., 2010; Guimarães et al., 2008; Köprücü
and Özdemir, 2005; Rafiee and Saad, 2005; Robinson et al., 1987; Shiau and Hsieh,
2001).

20
 Substance Amount Reference
N 5.18 % Moccia et al., 2007
3.40 % Rafiee and Saad, 2005
K 0.88 % Moccia et al., 2007
0.53 % Rafiee and Saad, 2005
2000.00 mg/kg diet Shiau and Hsieh, 2001
Ca 1.53 % Moccia et al., 2007
0.80 % Köprücü and Özdemir, 2005
1.74 % Rafiee and Saad, 2005
16250.31 mg/kg diet Guimarães et al., 2008
Mg 0.18 % Moccia et al., 2007
0.43 % Rafiee and Saad, 2005
P 1.12 % Moccia et al., 2007
1.48 % Rafiee and Saad, 2005
6833.21 mg/kg diet Guimarães et al., 2008
5000.00 mg/kg diet Robinson et al., 1987
S 4.33 mg/kg diet Guimarães et al., 2008
2830.00 mg/kg diet Köprücü and Özdemir, 2005
Cl - mg/kg DW Moccia et al., 2007
607.20 mg/kg diet Guimarães et al., 2008
18220.00 mg/kg diet Cnaani et al., 2010
Cu 0.0024 % Rafiee and Saad, 2005
20.67 mg/kg DW Moccia et al., 2007
Mn 78.00 mg/kg DW Moccia et al., 2007
0.003 % Rafiee and Saad, 2005
38.72 mg/kd diet Guimarães et al., 2008
Fe 186.00 mg/kg DW Moccia et al., 2007
0.1094 % Rafiee and Saad, 2005
7.35 mg/kg diet Guimarães et al., 2008
Zn 156.67 mg/kg DW Moccia et al., 2007
0.0056 % Rafiee and Saad, 2005
120.48 mg/kg diet Guimarães et al., 2008
Co 1.50 mg/kg DW Moccia et al., 2007
Mo 2.50 mg/kg DW Moccia et al., 2007
Ni 4.00 mg/kg DW Moccia et al., 2007
Na 392.98 mg/kg diet Guimarães et al., 2008
11780.00 mg/kg diet Cnaani et al., 2010
Si -
B -
C 49210.00 mg/kg DW Moccia et al., 2007
H -
O -
As 1.00 mg/kg DW Moccia et al., 2007
Cd 1.00 mg/kg DW Moccia et al., 2007
Cr 1.33 mg/kg DW Moccia et al., 2007
684.21 mg/kg diet Guimarães et al., 2008
Hg 0.05 mg/kg DW Moccia et al., 2007
Pb 5.00 mg/kg DW Moccia et al., 2007
Se 1.00 mg/kg DW Moccia et al., 2007
0.32 mg/kg diet Guimarães et al., 2008
Al -
Ba -
Table A.1.4: Feed substance overview

21
A.1.5 Used feed substance composition
In literature there a different reports on substance combinations used for tilapia feed. This
plasticity makes it difficult to select the right value. Based on table A.1.4 the substances
have been selected to fulfill the mass balance (Eq. 1 and table A.4).
Substance Amount Reference
N 51.80 g/kg feed Moccia et al., 2007
K 2.00 g/kg feed Shiau and Hsieh, 2001
Ca 8.00 g/kg feed Köprücü and Özdemir, 2005
Mg 1.80 g/kg feed Moccia et al., 2007
P 6.83 g/kg feed Guimarães et al., 2008
S 2.83 g/kg feed Köprücü and Özdemir, 2005
Cl 18.22 g/kg feed Cnaani et al., 2010
Cu 0.02 g/kg feed Moccia et al., 2007
Mn 0.08 g/kg feed Moccia et al., 2007
Fe 0.19 g/kg feed Moccia et al., 2007
Zn 0.16 g/kg feed Moccia et al., 2007
Co 0.00 g/kg feed Moccia et al., 2007
Mo 0.00 g/kg feed Moccia et al., 2007
Ni 0.00 g/kg feed Moccia et al., 2007
Na 11.78 g/kg feed Cnaani et al., 2010
Si - g/kg feed
B - g/kg feed
As 0.00 g/kg feed Moccia et al., 2007
Cd 0.00 g/kg feed Moccia et al., 2007
Cr 0.68 g/kg feed Guimarães et al., 2008
Hg 0.00 g/kg feed Moccia et al., 2007
Pb 0.01 g/kg feed Moccia et al., 2007
Se 0.00 g/kg feed Guimarães et al., 2008
Table A.1.5: Used feed substance composition

22
A.2 Faeces composition
A.2.1 Faeces composition overview
Based on the reports in literature a review of reports of faeces composition has been made
(Köprücü and Özdemir, 2005; Naylor et al., 1999).
Substance Feces content Reference
N 2.83 % Naylor et al., 1999
K 0.1 % Naylor et al., 1999
Ca 6.99 % Naylor et al., 1999
6.528 g/kg DW Köprücü and Özdemir, 2005
Mg 0.53 % Naylor et al., 1999
P 2.54 % Naylor et al., 1999
6.687 g/kg DW Köprücü and Özdemir, 2005
S -% Naylor et al., 1999
0.38 g/kg DW Köprücü and Özdemir, 2005
Cl - mg/kg DW Naylor et al., 1999
Cu 33.4 mg/kg DW Naylor et al., 1999
Mn 487.8 mg/kg DW Naylor et al., 1999
Fe 1942 mg/kg DW Naylor et al., 1999
Zn 604.9 mg/kg DW Naylor et al., 1999
Co 1.82 mg/kg DW Naylor et al., 1999
Mo - mg/kg DW Naylor et al., 1999
Ni 4.94 mg/kg DW Naylor et al., 1999
Na - mg/kg DW Naylor et al., 1999
Si - mg/kg DW Naylor et al., 1999
B - Naylor et al., 1999
C - Naylor et al., 1999
H - Naylor et al., 1999
O - Naylor et al., 1999
As 2.2 mg/kg DW Naylor et al., 1999
Cd 1.13 mg/kg DW Naylor et al., 1999
Cr 3.86 mg/kg DW Naylor et al., 1999
Hg 0.05 mg/kg DW Naylor et al., 1999
Pb 5.54 mg/kg DW Naylor et al., 1999
Se 0.5 mg/kg DW Naylor et al., 1999
Al - Naylor et al., 1999
Ba - Naylor et al., 1999
Table A.2.1: Faeces composition overview

23
A.2.2 Used faeces composition
Based on the review of available data (see table A.2.1) and the mass balance approach
(table A.4), a representative faeces composition has been chosen from Rainbow trout
(Naylor et al., 1999) and Nile tilapia (Köprücü and Özdemir, 2005).
Substance Faeces content Reference
N 28.30 g/kg DW Naylor et al., 1999
K 1.00 g/kg DW Naylor et al., 1999
Ca 6.53 g/kg DW Köprücü and Özdemir, 2005
Mg 5.30 g/kg DW Naylor et al., 1999
P 6.69 g/kg DW Köprücü and Özdemir, 2005
S 0.38 g/kg DW Köprücü and Özdemir, 2005
Cl - g/kg DW Naylor et al., 1999
Cu 0.03 g/kg DW Naylor et al., 1999
Mn 0.49 g/kg DW Naylor et al., 1999
Fe 1.94 g/kg DW Naylor et al., 1999
Zn 0.60 g/kg DW Naylor et al., 1999
Co 0.00 g/kg DW Naylor et al., 1999
Mo - g/kg DW Naylor et al., 1999
Ni 0.00 g/kg DW Naylor et al., 1999
Na - g/kg DW Naylor et al., 1999
As 0.00 g/kg DW Naylor et al., 1999
Cd 0.00 g/kg DW Naylor et al., 1999
Cr 0.00 g/kg DW Naylor et al., 1999
Hg 0.00 g/kg DW Naylor et al., 1999
Pb 0.01 g/kg DW Naylor et al., 1999
Se 0.00 g/kg DW Naylor et al., 1999
Table A.2.2: Used faeces composition

A.2.3 Faeces dry weight factor

The faeces dry weight factor is based on one report for Red tilapia (Rafiee and Saad,
2005), as no other dataset has been found.
Fish groups feed input dry sludge dry solid faeces
[g] [g] [g] [g] [%]
20 2025 182.5 349.1 26.25%
40 2167 113.5 444.8 25.76%
80 2702 159.41 334.8 18.29%
120 3579 169.1 436.2 16.91%
180 2868 224.1 440 23.16%
Total 13341 848.61 2004.9 21.39%
Table A.2.3: Faeces dry weight factor

24
A.3 Tilapia body composition
A.3.1 Reported body compositions for Nile tilapia
The reported substance contents for Nile tilapia (Clement and Lovell, 1994; Dale et al.,
2004; Gonzales and Brown, 2006; Köprücü and Özdemir, 2005).
Dale et al., 2004 Gonzales and Clement and Lovell, Köprücu and
Brown, 2006 1994 Özdemir, 2005

whole body whole body fillet Uptake

Substance mg/100g DW mg/100g DW mg/100g raw fillet ** mg/100g
N 3425 * 3623.13 3248.00 *
K 380 5.69 324.00
Ca 8400 476.15 17.50 147.20
Mg 150 12.75 26.26
P 4100 25.87 169.00 231.30
S 600.00 244.61

Cl
Cu 0.09 0.05 0.09
Mn 0.139 0.02 0.01
Fe 1.87 0.03 1.76
Zn 0.675 1.35 0.70
Co 0.06 0.04
Mo 0.57 0.01
Ni
Na 380 39.47 34.70
Si 0.16
B 0.04 0.06
C
H
O
As
Cd
Cr 7.10 0.04
Hg
Pb 0.01
Se 0.71
Al 0.36
Ba 0.05

* Kjeldahl protein calculation

** the fillet represents 36% of the body
Table A.3.1: Reported body compositions for Nile tilapia

25
A.3.2 Used body composition for Nile tilapia
The used body composition based on the found body substance composition (see table
A.3.1) and the mass balance approach (see table A.4) (Clement and Lovell, 1994;
Gonzales and Brown, 2006; Köprücü and Özdemir, 2005).
Substance Body content Reference
N 3623.13 mg/100g Gonzales and Brown, 2006
K 5.69 mg/100g Gonzales and Brown, 2006
Ca 476.15 mg/100g Gonzales and Brown, 2006
Mg 12.75 mg/100g Gonzales and Brown, 2006
P 25.87 mg/100g Gonzales and Brown, 2006
S 244.61 mg/100g Köprücu and Özdemir, 2005

Cl - mg/100g
Cu 0.09 mg/100g Clement and Lovell, 1994
Mn 0.02 mg/100g Gonzales and Brown, 2006
Fe 0.03 mg/100g Gonzales and Brown, 2006
Zn 1.35 mg/100g Gonzales and Brown, 2006
Co 0.04 mg/100g Clement and Lovell, 1994
Mo 0.01 mg/100g Clement and Lovell, 1994
Ni - mg/100g
Na 39.47 mg/100g Gonzales and Brown, 2006
Si 0.16 mg/100g Clement and Lovell, 1994
B 0.06 mg/100g Clement and Lovell, 1994
C - mg/100g
H - mg/100g
O - mg/100g
As - mg/100g
Cd - mg/100g
Cr 7.10 mg/100g Gonzales and Brown, 2006
Hg - mg/100g
Pb - mg/100g
Se 705.00 mg/100g Gonzales and Brown, 2006
Al 0.36 mg/100g Clement and Lovell, 1994
Ba 0.05 mg/100g Clement and Lovell, 1994
Table A.3.2: Used body composition for Nile tilapia

26
A.4 Overall substance partition
By combining the information from tables A.1.5, A.3.2, A.2.2, A.2.3 the following substance
partition can be found based on formulas 1), 2) and 3). The negative values for Fe, Mn, Si,
B and Se are a consequence of the standard deviation in the original dataset or the
different species used in the study (Moccia et al., 2007; Naylor et al., 1999).
mbody∗10
1) 1000
m bodyF =
FCR
m faecesDW =m faeces∗F FaecesDW
2)
3) m water =(m feed −mbodyF −mfaecesDW )∗F nitrification

100%
90%
80%
70%
60%
50% Water

40% Faeces
Fish
30%
20%
10%
0%
K Mg S Cu Fe** Co Ni* As Cr
N Ca P Cl* Mn** Zn Mo* Na* Cd Hg
Figure A.4: Feed substance partition into fish, faeces and water without Si, B, C, H, O due
to data scarcity (*) incomplete dataset; **) overall balance is negative)

27
 Sub- Feed Body Faeces FCR Fish uptake F,FaecesDW Faeces Nitrification Water
Stance composition composition composition content correction Substances
(table A.1.5) (table A.3.2) (table A.2.2) 1) (table A.2.3) 2) 3)
[g/kg feed] [mg/100g] [g/kg DW] [-] [g/kg feed] [-] [g/kg feed] [-] [g/kg feed]
m.feed m.body m.faeces m.bodyF F.FaecesDW m.faecesDW F.nitrification m.water
N 51.80 3623.13 28.30 1.11 32.64 0.214 6.06 0.98 12.841
K 2.00 5.69 1.00 1.11 0.05 0.214 0.21 1.00 1.735
Ca 8.00 476.15 6.53 1.11 4.29 0.214 1.40 1.00 2.313
Mg 1.80 12.75 5.30 1.11 0.11 0.214 1.13 1.00 0.551
P 6.83 25.87 25.40 1.11 0.23 0.214 5.44 1.00 1.161
S 2.83 244.61 0.38 1.11 2.20 0.214 0.08 1.00 0.545

Cl 18.22 - - 1.11 0.00 0.214 0.00 1.00 18.216

Cu 0.02 0.09 0.03 1.11 0.00 0.214 0.01 1.00 0.013
Mn 0.08 0.02 0.49 1.11 0.00 0.214 0.10 1.00 -0.027
Fe 0.19 0.03 1.94 1.11 0.00 0.214 0.42 1.00 -0.230
Zn 0.16 1.35 0.60 1.11 0.01 0.214 0.13 1.00 0.015
Co 0.00 0.04 0.00 1.11 0.00 0.214 0.00 1.00 0.001
Mo 0.00 0.01 - 1.11 0.00 0.214 0.00 1.00 0.002
Ni 0.00 - 0.00 1.11 0.00 0.214 0.00 1.00 0.003
Na 11.78 39.47 - 1.11 0.36 0.214 0.00 1.00 11.428
Si 0.00 0.16 - 1.11 0.00 0.214 0.00 1.00 -0.001
B 0.00 0.06 - 1.11 0.00 0.214 0.00 1.00 -0.001
C 492.10 - - 1.11 0.00 0.214 0.00 1.00 492.100
H 0.00 - - 1.11 0.00 0.214 0.00 1.00 0.000
O 0.00 - - 1.11 0.00 0.214 0.00 1.00 0.000
As 0.00 - 0.00 1.11 0.00 0.214 0.00 1.00 0.001
Cd 0.00 - 0.00 1.11 0.00 0.214 0.00 1.00 0.001
Cr 0.68 7.10 0.00 1.11 0.06 0.214 0.00 1.00 0.619
Hg 0.00 - 0.00 1.11 0.00 0.214 0.00 1.00 0.000
Pb 0.01 - 0.01 1.11 0.00 0.214 0.00 1.00 0.004
Se 1 705 0.5 1.11 6.35 0.214 0.11 1.00 -5.458
Al - 0.36 - 1.11 0.00 0.214 0.00 1.00 -
Ba - 0.05 - 1.11 0.00 0.214 0.00 1.00 -
Table A.4: Overall substance partition (mass balance)

28
A.5 Water quality concentrations in fish rearing systems
A.5.1 RAS water concentration limits
In literature there are some information about the concentrations used in RAS, which have
been collected (Goddek et al., 2015; Kamal and Mair, 2005; Timmons and Ebeling, 2010)
Substance Concentration Reference
NO3-N 0-400 mg/L Timmons and Ebeling, 2010
20-137 mg/L Goddek et al., 2014
100 mg/L Eding et al., 2006
K <5 mg/L Timmons and Ebeling, 2010
27-106 mg/L Goddek et al., 2014
Ca 4-160 mg/L Timmons and Ebeling, 2010
24-180 mg/L Goddek et al., 2014
Mg <15 mg/L Timmons and Ebeling, 2010
6-44 mg/L Goddek et al., 2014
P 0.01-3.0 mg/L Timmons and Ebeling, 2010
PO4-P 8-17 mg/L Goddek et al., 2014
SO4-S <50 mg/L Timmons and Ebeling, 2010
6 mg/L Goddek et al., 2014
Cl <0.003 mg/L Timmons and Ebeling, 2010
18200 mg/L Kamal et Mair, 2005
Cu 0.18*10E-3 mg/L Timmons and Ebeling, 2010
0.03-0.05 mg/L Goddek et al., 2014
Mn <0.01 mg/L Timmons and Ebeling, 2010
0.06-0.8 mg/L Goddek et al., 2014
Fe <0.15 mg/L Timmons and Ebeling, 2010
0.2-2.5 mg/L Goddek et al., 2014
Zn 2.4*10E-3 mg/L Timmons and Ebeling, 2010
0.34-0.44 mg/L Goddek et al., 2014
Co - mg/L Timmons and Ebeling, 2010
Mo - mg/L Timmons and Ebeling, 2010
0.01 mg/L Goddek et al., 2014
Ni <0.1 mg/L Timmons and Ebeling, 2010
Na <75 mg/L Timmons and Ebeling, 2010
11820 mg/L Kamal et Mair, 2005
14-17 Goddek et al., 2014
Si - mg/L Timmons and Ebeling, 2010
B 0 mg/L Timmons and Ebeling, 2010
0.09-0.19 mg/L Goddek et al., 2014
As <0.05 mg/L Timmons and Ebeling, 2010
Cd 0.01 mg/L Timmons and Ebeling, 2010
Cr - mg/L Timmons and Ebeling, 2010
Hg <0.02 mg/L Timmons and Ebeling, 2010
Pb <0.02 mg/L Timmons and Ebeling, 2010
Se <0.01 mg/L Timmons and Ebeling, 2010
Al <0.01 mg/L Timmons and Ebeling, 2010
Ba <5 mg/L Timmons and Ebeling, 2010
Table A.5.1: RAS water concentration limits

29
A.5.2 RAS minimal discharge requirement
Based on the water concentration limits (table 1) and Eq. (2), the minimal required
discharge per kilogram of feed can be calculated.
Substance Feed Fish water Minimal discharge
substances quality requirement
[g/kg feed] [mg/L] [L/kg feed]
N 12.841 100 128.4
K 1.735 27 64.2
Ca 2.313 160 14.5
Mg 0.551 15 36.7
P 1.161 15 77.4
S 0.545 50 10.9
Cl 18.216 18200 1.0
Na 11.431 11820 1.0
Table A.5.2: RAS minimal discharge requirement

A.5.3 Daily supplied substance mass by feed

The defined feed supply combined with the substances released into the water (table A.4)
gives the daily supplied substance mass to the RAS.
Substance Feed Average Daily substance
substances feed input mass released
into water
[g/kg feed][kg feed/day] [g/day]
N 12.841 21.902 281.242
K 1.735 21.902 37.994
Ca 2.313 21.902 50.667
Mg 0.551 21.902 12.067
P 1.161 21.902 25.436
S 0.545 21.902 11.931
Cl 18.216 21.902 398.970
Na 11.431 21.902 250.372
Table A.5.3: Daily supplied substance mass by feed

30
A.6 Hydroponics
A.6.1 Hydroponic water concentration limits
Plants do not only require a certain amount of substances, but also concentration of these
substances within certain boundaries (Kipp, 1997; Sonneveld and Voogt, 2009).
Substance concentration Reference
EC 2.50 - 5.50 mS/cm Kipp, 1997
4.00 dS/m Sonneveld and Voogt, 2009
NH4-N 1.40 - 7.00 mg/L Kipp, 1997
0.00 - 7.00 mg/L Sonneveld and Voogt, 2009
K 207.23 - 414.46 mg/L Kipp, 1997
254.15 - 391.00 mg/L Sonneveld and Voogt, 2009
Na 2.29 - 274.80 mg/L Kipp, 1997
Ca 264.66 - 533.33 mg/L Kipp, 1997
320.80 - 481.20 mg/L Sonneveld and Voogt, 2009
Mg 72.90 - 145.80 mg/L Kipp, 1997
65.61 - 157.95 mg/L Sonneveld and Voogt, 2009
NO3-N 210.00 - 434.00 mg/L Kipp, 1997
238.00 - 392.00 mg/L Sonneveld and Voogt, 2009
Cl 3.54 - 531.00 mg/L Kipp, 1997
S 144.45 - 288.90 mg/L Kipp, 1997
128.40 - 288.90 mg/L Sonneveld and Voogt, 2009
P 21.70 - 40.30 mg/L Kipp, 1997
21.70 - 62.00 mg/L Sonneveld and Voogt, 2009
Fe 0.73 - 2.12 mg/L Kipp, 1997
0.50 - 1.40 mg/L Sonneveld and Voogt, 2009
Mn 0.11 - 0.41 mg/L Kipp, 1997
0.16 - 0.55 mg/L Sonneveld and Voogt, 2009
Zn 0.23 - 0.69 mg/L Kipp, 1997
0.33 - 0.65 mg/L Sonneveld and Voogt, 2009
B 0.27 - 0.81 mg/L Kipp, 1997
0.38 - 0.70 mg/L Sonneveld and Voogt, 2009
Cu 0.03 - 0.07 mg/L Kipp, 1997
0.03 - 0.10 mg/L Sonneveld and Voogt, 2009
Mo 0.03 - 0.08 mg/L Kipp, 1997
0.03 - 0.08 mg/L Sonneveld and Voogt, 2009
Table A.6.1a: Hydroponics water quality concentration limits

31
Kipp (1997) does not only give the boundaries for the concentrations, but also includes an
optimal value for each of the substances.
Substance Concentration Reference
ideal min max
EC 3.70 2.50 5.50 mS/cm Kipp, 1997
NH4-N 1.40 1.40 7.00 mg/L Kipp, 1997
K 312.80 207.23 414.46 mg/L Kipp, 1997
Na 138.55 2.29 274.80 mg/L Kipp, 1997
Ca 401.00 264.66 533.33 mg/L Kipp, 1997
Mg 109.35 72.90 145.80 mg/L Kipp, 1997
NO3-N 322.00 210.00 434.00 mg/L Kipp, 1997
Cl 267.27 3.54 531.00 mg/L Kipp, 1997
S 218.28 144.45 288.90 mg/L Kipp, 1997
P 31.00 21.70 40.30 mg/L Kipp, 1997
Fe 1.40 0.73 2.12 mg/L Kipp, 1997
Mn 0.27 0.11 0.41 mg/L Kipp, 1997
Zn 0.46 0.23 0.69 mg/L Kipp, 1997
B 0.54 0.27 0.81 mg/L Kipp, 1997
Cu 0.05 0.03 0.07 mg/L Kipp, 1997
Mo 0.05 0.03 0.08 mg/L Kipp, 1997
Table A.6.1b: Hydroponic water quality requirement including optimal range

A.6.2 Hydroponic substance uptake

The tomato uptake of substances and water evaporation has been reported (Voogt, 1993).
Substance Unit Average Exp 1 Exp2 Exp3 Exp4 Exp5
NO3 mmol/m² 6143.8 5145 5400 5642 6324 8208
SO4 mmol/m² 728.4 578 810 620 650 984
H2PO4 mmol/m² 675.8 452 648 744 715 820
K mmol/m² 3847 2973 3618 3782 4680 4182
Ca mmol/m² 1564.6 1307 1242 1302 1430 2542
Mg mmol/m² 601 355 486 496 520 1148
Water mm 621 475 540 620 650 820
Plants/m² 2.14 2.2 2.2 2.1 2.1 2.1
Yields kg/m² 40 32 26 48 46 48
Start growth 01/15/99 01/15/99 12/20/98 12/20/98 12/20/98
End growth 10/30/99 10/01/99 10/25/99 11/07/99 11/01/99
Number of days 298.8 288 259 309 322 316
Table A.6.2: Hydroponic substance uptake

32
A.6.3 Predicted plant uptake
The predicted plant uptake is based on findings on overall yield (De Gelder et al.,
2005) and the uptake of each substance and water (Voogt, 1993). The yield factor is
based on the yield of Voog (1993) divided by the yield of De Gelder et al. (2005) which is
40 kg / m2 / 56.2 kg / m2 = 1.405.
Substance Average uptake/day Predicted yield (De Yield factor Predicted uptake
(Voogt, 1993) Gelder et al., 2005)
N 0.288 g/m2/day 56.2 kg/m2 1.405 0.404 g/m2/day
S 0.078 g/m2/day 56.2 kg/m2 1.405 0.110 g/m2/day
P 0.070 g/m2/day 56.2 kg/m2 1.405 0.098 g/m2/day
K 0.503 g/m2/day 56.2 kg/m2 1.405 0.707 g/m2/day
Ca 0.210 g/m2/day 56.2 kg/m2 1.405 0.295 g/m2/day
Mg 0.049 g/m2/day 56.2 kg/m2 1.405 0.069 g/m2/day
Water 2.078 mm/m2/day 56.2 kg/m2 1.405 2.920 mm/m2/day
Table A.6.3: Predicted plant uptake

33
A.7 pH control table
Eding et al. (2006) report the needed calculations for pH compensation for the nitrification
of the bio-filter. Based on these calculations and the information from Goddek et al. (2015),
the necessary amount have been calculated for potassium, sodium, magnesium and
calcium. The calculation is based on the compensation of 1.98 mol HCO 3- / mol NH4-N of
Eding et al. (2006).
Substance Daily atomic Daily Daily Aklinity
substance weight amount alkalinity factor
mass
[g/day] [g/mol] [mol/day] [-]
N 281.24 14.00 20.09 39.78 1
KHCO3 3982.14 100.12 39.78 39.78 1
K 1555.15 39.10 39.78 39.78 1
NaHCO3 3341.15 84.00 39.78 39.78 1
Na 914.04 22.98 39.78 39.78 1
MgCO3 1676.74 84.31 19.89 39.78 2
Mg 483.27 24.30 9.94 19.89 2
CaCO3 1990.37 100.08 19.89 39.78 2
Ca 797.10 40.08 9.94 19.89 2
Table A.7: pH control table

34
A.8 Decoupled System
A.8.1 RAS effluent concentration
Through the given water volume and the added feed per day, the length before reaching
the maximal concentration can be calculated. The lowest number of days is showing the
critical nutrient, thus a discharge after 14.2 days is needed because of the nitrogen
substance.
Substance Substance Daily feeding Substance Water quality System # days before
added netto per day RAS volume discharge
[g/kg feed] [kg feed/day] [g/day] [mg/L] [m3] [day]
N 12.841 21.902 281.242 100 40 14.2
K 1.735 21.902 37.994 106 40 111.6
Ca 2.313 21.902 50.667 180 40 142.1
Mg 0.551 21.902 12.067 44 40 145.9
P 1.161 21.902 25.436 17 40 26.7
S 0.545 21.902 11.931 50 40 167.6
Na 11.428 21.902 250.302 11820 40 1888.9
Cl 18.216 21.902 398.970 18200 40 1824.7
Table A.8.1: RAS effluent concentration

35
A.8.2 Decoupled hydroponics uptake
Because of the installed valve, the hydroponic and fish part are separated from each other. Thus each part of the system can be adressed
by its own specific maximal concentration of substances. This separation allows a strategy to have a minimal amount of water in the
system, to prevent accumulation of nutrients. The concentration of accumulating substances (e.g. sodium and chloride) changes due to the
evaporation of water by the plants. To remove these not take up nutrients, the discharge has to be triggered. Based on the maximal
concentration and the amount of substance, the minimal water discharge is calculated. In the situation, where the water volume in the
hydroponics has to be kept as low as possible, the minimal discharge also represents the minimal water level.
Substance RAS Volume Substan Plant uptake Area Plant Left Maximal Minimal
Discharge ce mass uptake substances concentration water level
concentration
[mg/L] [L] [g] [g/m2/day] [m2] [g/day] [g/day] [mg/L] [L/day]
N 99.982 2812.22 281.170 0.404 695 281.090 0.080 434 -
K 13.507 2812.22 37.985 0.110 695 76.340 -38.355 414.46 -
Ca 18.012 2812.22 50.654 0.098 695 68.397 -17.743 533.33 -
Mg 4.290 2812.22 12.063 0.707 695 491.438 -479.374 157.95 -
P 9.042 2812.22 25.429 0.295 695 204.933 -179.504 40.3 -
S 4.241 2812.22 11.928 0.069 695 47.746 -35.819 288.9 -
Na 88.983 2812.22 250.238 0.000 695 0 250.238 274.8 910.620
Cl 141.834 2812.22 398.867 0.000 695 0 398.867 531 751.163
Nutrient Plant Plant
solution evaporation Area evaporation Left water
[L] [L/m2/day] [m2] [L/day] [L/day]
Water 2812.22 2.92 695 2029.4 783 -
Table A.8.2: Decoupled hydroponics uptake

36
A.9 Coupled aquaponic water concentrations
A.9.1 Non-accumulation discharge requirement
Based on the assumption to discharge the same amount of sodium and chloride as added through the feed, the minimal required discharge
can be calculated.
Substance Substance Daily Substance System End water Minimal
added netto feeding per day volume concentration discharge
[g/kg feed] [kg feed/day] [g/day] [L] [mg/L] [L/day]
N 12.841 21.902 281.242 38556 96.436 2916.355
K 1.735 21.902 37.994 38556 12.033 3157.526
Ca 2.313 21.902 50.667 38556 16.913 2995.772
Mg 0.551 21.902 12.067 38556 -8.296 -1454.542
P 1.161 21.902 25.436 38556 4.066 6255.938
S 0.545 21.902 11.931 38556 3.162 3773.304
Na 11.428 21.902 250.302 38556 92.316 2711.378
Cl 18.216 21.902 398.970 38556 147.146 2711.378
Table A.9.1: Non-accumulation discharge requirement for coupled aquaponics

A.9.2 Accumulation discharge requirement

Based on the maximal allowable substance concentration (see table 1), the following minimal
Substance Substance Daily Substance Maximal Maximal
added netto feeding per day System concentration accumulation
volume time
[g/kg feed] [kg feed/day] [g/day] [L] [mg/L] [day]
N 12.8 21.9 281.2 40000 100.0 14.2
K 1.7 21.9 38.0 40000 106.0 111.6
Ca 2.3 21.9 50.7 40000 180.0 142.1
Mg 0.6 21.9 12.1 40000 44.0 145.9
P 1.2 21.9 25.4 40000 17.0 26.7
S 0.5 21.9 11.9 40000 50.0 167.6
Na 11.4 21.9 250.3 40000 274.8 43.9
Cl 18.2 21.9 399.0 40000 531.0 53.2
Table A.9.2: Accumulation discharge requirement for coupled aquaponics

37