Environmental impact assessment of local decoupled multi-loop aquaponics in an urban context

Abstract

Fresh vegetables available on Northern European markets usually originate from a high number of sources. Environmental impacts for these goods typically arise from the resources used in production and the long-distance transport in air-conditioned trucks. As such, environmental impacts are mainly attributed to direct energy consumption, water use and nutrient supply. The aim of this paper was therefore to investigate and evaluate possible solutions to reduce the environmental impacts of vegetables available on urban markets in Northern Europe. We hypothesise that for the production of lettuce and tomatoes in Northern Europe, a 4-step solution, i.e. 1) local production, 2) climate-controlled efficient greenhouses, 3) decoupled aquaponics, and 4) combined building architecture with waste heat and green waste reuse, will enable a low environmental impact. We defined the metropole Berlin as case example, and used simulation results from a proven greenhouse simulator as input to a comparing life cycle assessment of fresh lettuce and tomato. The assessment included a list of 12 midpoint environmental impact categories, e.g. global warming potential with 100 year horizon (GWP100; kg CO2 eq.), depletion of fossil fuel reserves (FRS; kg oil eq.), and water use (WCO; m³ water). Most impact categories decreased systematically when increasing the complexity of the local vegetable production. Compared to the mix of vegetables from different locations available on the market, the complete 4-step solution reduced WCO from water consumption to water saving: i.e. from 14.2 L or 3.3 L to −10.1 L or −0.21 L per package of 500 g tomatoes or 150 g lettuce, respectively. GWP100 and FRS were below the values of the available market mix, e.g. GWP100 decreased with 8.7% in tomatoes and 49.9% in lettuce. In conclusion, with the right set-up, local vegetable productions in urban regions can surpass the imported mix on environmental performance in Northern Europe.

Full text

Journal of Cleaner Production 313 (2021) 127735
Available online 5 June 2021
0959-6526/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Environmental impact assessment of local decoupled multi-loop aquaponics
in an urban context
Oliver K¨
orner
a
,
*
, Mehdi B. Bisbis
a
, G¨
osta F.M. Baganz
b
,
d
, Daniela Baganz
b
, Georg B.O. Staaks
b
,
Hendrik Monsees
b
, Simon Goddek
c
, Karel J. Keesman
c
a
Leibniz-Institute of Vegetable and Ornamental Crops (IGZ), Next-Generation Horticultural Systems, Großbeeren, Germany
b
Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Berlin, Germany
c
Wageningen University, Mathematical and Statistical Methods - Biometris, Wageningen, the Netherlands
d
RWTH Aachen University, Faculty of Architecture, Germany
ARTICLE INFO
Handling editor; Yutao Wang
Keywords:
Tomato
Simulation model
Life cycle impact assessment
Lettuce
Greenhouse production
Hydroponics
ABSTRACT
Fresh vegetables available on Northern European markets usually originate from a high number of sources.
Environmental impacts for these goods typically arise from the resources used in production and the long-
distance transport in air-conditioned trucks. As such, environmental impacts are mainly attributed to direct
energy consumption, water use and nutrient supply. The aim of this paper was therefore to investigate and
evaluate possible solutions to reduce the environmental impacts of vegetables available on urban markets in
Northern Europe. We hypothesise that for the production of lettuce and tomatoes in Northern Europe, a 4-step
solution, i.e. 1) local production, 2) climate-controlled efcient greenhouses, 3) decoupled aquaponics, and 4)
combined building architecture with waste heat and green waste reuse, will enable a low environmental impact.
We dened the metropole Berlin as case example, and used simulation results from a proven greenhouse
simulator as input to a comparing life cycle assessment of fresh lettuce and tomato. The assessment included a list
of 12 midpoint environmental impact categories, e.g. global warming potential with 100 year horizon (GWP
100
;
kg CO
2
eq.), depletion of fossil fuel reserves (FRS; kg oil eq.), and water use (WCO; m
3
water). Most impact
categories decreased systematically when increasing the complexity of the local vegetable production. Compared
to the mix of vegetables from different locations available on the market, the complete 4-step solution reduced
WCO from water consumption to water saving: i.e. from 14.2 L or 3.3 L to −10.1 L or −0.21 L per package of 500
g tomatoes or 150 g lettuce, respectively. GWP
100
and FRS were below the values of the available market mix, e.
g. GWP
100
decreased with 8.7% in tomatoes and 49.9% in lettuce. In conclusion, with the right set-up, local
vegetable productions in urban regions can surpass the imported mix on environmental performance in Northern
Europe.
1. Introduction
Aquaponics is the combined cultivation of sh in recirculating
aquaculture systems (RAS) and plant hydroponics (HP). In these com-
bined systems, two products are produced simultaneously with almost
the same amount of resource input when the system is optimally
balanced (Goddek and K¨
orner, 2019). Additional advantages are syn-
ergistic effects with substantially increased crop production that have
been observed in some crops (e.g., in lettuce) (Delaide et al., 2016;
Goddek and Vermeulen, 2018), decreased energy use (K¨
orner et al.,
2017), and a lower environmental impact compared to production in
independent systems (Ghamkhar et al., 2020). However, common
single-loop aquaponic systems (CAPS) often fail to provide the necessary
quantity and composition of nutrients to the HP system (Goddek et al.,
2019). Signicant amounts of crop produced in CAPS are also often of
lower quality or gain only a reduced yield. The improvements of various
environmental impacts, such as eutrophication, water usage, and
geographic footprint (Cohen et al., 2018), are commonly quantied for
CAPS in a given environment. Compared to CAPS, where RAS water is
directly recirculated via the HP subsystem, decoupled aquaponic sys-
tems treat the water by adjusting the quantity and quality to the actual
crop demands. Since not all nutrients required by the crop are available
from the RAS subsystem (Kloas et al., 2015), decoupled aquaponic
* Corresponding author.
E-mail address: koerner@igzev.de (O. K¨
orner).
Contents lists available at ScienceDirect
Journal of Cleaner Production
journal homepage: www.elsevier.com/locate/jclepro
https://doi.org/10.1016/j.jclepro.2021.127735
Received 24 November 2020; Received in revised form 23 April 2021; Accepted 27 May 2021

Journal of Cleaner Production 313 (2021) 127735
2
systems, showing great advantages, have been developed (Goddek,
2017) and can be economically viable (Baganz et al., 2020). In modern
decoupled aquaponic systems (DAPS), multiple loops are used for water
treatment, and additional nutrients are dosed into the HP system to
maintain high-quality crop production (Goddek and Keesman, 2018).
Closed loop systems tend to accumulate salts; thus, a periodic refreshing
of the HP system and an environmentally harmful discharge of the used
nutrient solution is needed (Savvas et al., 2008). On the other hand, an
improved nutrient balance in HP increases the production amount and
the share of high-quality produce, e.g., lettuce with a higher dry weight
fraction (Goddek and Vermeulen, 2018) or tomato with less
blossom-end rot (Delaide et al., 2019; Schmautz et al., 2016). Since the
environmental impact assessment allocates the total resource use to the
total quantity of products, this in turn reduces the total environmental
impact of a product unit.
In aquaponics, some resources can be allocated to both vegetables
and sh. The major environmental burden of sh production in RAS
consists of sh-feed and wastewater and is largely independent of
location. For greenhouse-produced vegetables (i.e., LCA-term: farm),
the most signicant environmental impact of the products available in
the supermarket (i.e., LCA-term: gate) is often fossil fuel use either in the
form of heating or transport, depending on the location of the farm
(cradle) and the distance to the consumer (gate) (Pluimers, 2001).
In the European Union market, Spain, the Netherlands and Italy are
the three largest exporters of fresh vegetables (De Cicco, 2019). Despite
the long distances of Southern European produce to the market, the high
heating load in Northern European greenhouses greatly overshoots the
energy-related environmental footprints of greenhouse produce. How-
ever, modern and highly insulated greenhouse systems have greatly
reduced the environmental impact of fossil energy consumption (Cuce
et al., 2016). The modernisation of greenhouses with a package of
high-tech equipment, such as combined heat and power units, heat
pumps, underground seasonal and daytime energy storage systems, and
air treatment units, such as those used in the closed or semi-closed
greenhouse concepts (Opdam et al., 2005), has strongly reduced en-
ergy consumption (Cuce et al., 2016; Gruda et al., 2019). These new
technologies enable local production of vegetables all year round in
almost all climatic zones (Ntinas et al., 2020).
The objective of this study was to investigate and evaluate possible
solutions to reduce environmental impacts of vegetables produced in
greenhouses in Northern Europe (Germany, the Netherlands) to or
below the levels in Southern Europe (Italy, Spain). For the given case of
Berlin (Germany), with a mild temperate climate, i.e., Cfb after the
K¨
oppen classication, the environmental impact of local, year-round
produced greenhouse vegetables such as lettuce or tomatoes can be
reduced by technology alone (Vadiee and Martin, 2012). The installa-
tion of modern DAPS can further improve the environmental impacts
per unit of production, while additional roof-top farming can yield in
Abbreviations
ALU agricultural land use
AP aquaponics
CAPS coupled single-loop aquaponics systems
DAPS decoupled multi-loop aquaponics system
DLI daily light integral control
CLCA Consequential life cycle assessment
FRS fossil resource scarcity
FEP freshwater eutrophication
GWP
20
global warming potential with a 20-year horizon
GWP
100
global warming potential with a 100-year horizon
HP hydroponic cultivated plants
HCT human carcinogenic toxicity
HNT human non-carcinogenic toxicity
LCA Life cycle assessment
MRS mineral resource scarcity
MGS moving gutter system
NFT nutrient lm technique
ODP ozone depletion
RAS recirculating aquaculture system
TAP terrestrial acidication
WCO water consumption
WSC Water scarcity
Table 1
Simulation model scenarios used for the comparing LCA study with hydroponics (HP), decoupled multi-loop aquaponics (DAPS); letter code in scenario: T (tomato); L
(lettuce); R (rooftop, combined building); +(active energy transfer).
Scenario Crop Location System Climate Control Building situation Usage Type
T_HP_ES Tomato ES HP Semi Self-contained Benchmark HP
T_HP_IT IT
T_HP_NL NL Full
T_HP_DE DE
T_HP_MIX ES, IT, DE, NL
T_HP_DE
local
DE Berlin DAPS Case
T_AP_DE DAPS
T_HP-R_DE HP Roof-top HP-R
T_HP-R+_DE HP-R+
T_AP-R_DE DAPS DAPS-R
T_AP-R+_DE DAPS-R+
L_HP_ES Lettuce ES HP Semi Self-contained Benchmark HP
L_HP_IT IT
L_HP_NL NL Full
L_HP_DE DE
L_HP_MIX ES, IT, DE, NL
L_AP_DE
local
DE Berlin DAPS Case
L_AP_DE DAPS
L_HP-R_DE HP Roof-top HP-R
L_HP-R+_DE HP-R+
L_AP-R_DE DAPS DAPS-R
L_AP-R+_DE DAPS-R+
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
3
energy savings (Torres Pineda et al., 2020). The combination of DAPS in
an urban environment such as Berlin with the broad possibilities of
reusing waste heat from industrial or private households is thus likely to
outperform imported products in some categories of the environmental
impacts, such as climate change. Therefore, it was hypothesised that a
4-step approach with 1) local production systems, 2) high technology
greenhouses, 3) DAPS, and 4) waste heat re-usage and biogas will allow
an environmentally friendly production of greenhouse vegetables such
as lettuce and tomatoes in Northern Europe. Therefore, for the rst time
this study examines the environmental impact of fresh lettuce and to-
mato from cradle to gate with these preconditions. Three main scenarios
were analysed for both crops: 1) produced locally in HP, 2) produced in
DAPS, and 3) produced in rooftop DAPS using waste heat. All cases were
compared to benchmark scenarios. As a benchmark, mixtures of lettuce
or tomato available in Germany were used, which were produced in and
imported from Spain (El Ejido, Almería), the Netherlands (Westland
region), Italy (Rome area) or in different locations in Germany. The
products were transported to the gate, dened as a supermarket in the
centre of Berlin. An extended greenhouse and aquaponics simulator was
used as the data source for the four scenarios (Goddek and K¨
orner, 2019;
K¨
orner and Hansen, 2012). The simulation results were then used as part
of the input for a follow-up study on life-cycle assessment (LCA).
2. Materials and methods
2.1. Functional unit and LCA scope
A comparative LCA of tomato or lettuce produced in DAPS or produced
in HP was performed with different benchmark scenarios (Table 1). Four
producing countries of both tomato and lettuce (Spain, Italy, the
Netherlands and Germany), including their transports to the nal gate,
were analysed. The main benchmark was calculated from a mix of data
from these four countries present on the German market (see T_HP_MIX,
L_HP-MIX; Table 1). For HP tomatoes or lettuce, separate analyses were
conducted. Eventually, a normalised mixture of tomatoes or lettuce,
available in the supermarket in Berlin (i.e., the gate) and based on the
lettuce/tomato origins from years 2009–2018 (Behr, 2019), was analysed.
German imports of fresh tomatoes came mainly from the Netherlands
(58% of value) and Spain (25% of value) (Workman, 2020a), and then
from Belgium, Morocco and Italy (Behr, 2019). Lettuce was imported from
Spain (39% of value), Italy (25% of value), and the Netherlands (16% of
value) (Workman, 2020b). For tomatoes, the production in Belgium and
Morocco, with 5.0% and 6.6% of the fresh tomatoes consumed in Ger-
many, was attributed to the Netherlands and Spain, respectively. The
remaining 3.7% was distributed evenly among the four countries of origin.
The four countries of Italy, Spain, the Netherlands and Germany accounted
for 90% of the lettuce consumed in Germany. The remaining 10% were
distributed among these four countries at parity.
In this study, the combination of two fully functional food produc-
tion systems with RAS and HP greenhouse production was modelled.
The aquaponic system was designed as a DAPS four-loop system (God-
dek, 2017). The size of the HP greenhouse was set at 5000 m
2
. The RAS
system was predicted for tomatoes and lettuce for a location in Berlin
using the method presented by Goddek and K¨
orner (2019) that resulted
in 180 m
3
or 112 m
3
for tomato and lettuce, respectively.
When allocating the environmental impacts in complex systems be-
tween products and co-product, according to the ISO standards
(ISO14044; ISO14049), the rst option is to avoid allocation by making
use of a subdivision or to expand the systems investigated (Fitwi, 2012).
Therefore, in this paper system expansion was used (Weidema, 2003;
Weidema and Schmidt, 2010). System expansion as part of consequen-
tial LCA (CLCA) is often used in complex systems with co-products, as it
is the case in modern greenhouse horticulture and in aquaponics (Box-
man et al., 2017). In food systems, CLCA is increasingly used as favourite
method (Brand˜
ao et al., 2017; Gava et al., 2018).
Capital goods (e.g., roads and maintenance) were included in the
majority of the background data (i.e., data for processes that are not part
of the immediate product chain, such as electricity production, and
packaging), while infrastructural processes were excluded. The func-
tional units were 0.5 kg of packed tomatoes or 1 bag of sliced lettuce
Fig. 1. Overview of the life cycle system and its borders, with main plant production (1), RAS system with aquaponics combination (2), combined market-roof (3)
and system expansion (4).
Table 2
Core cultivation settings for lettuce and tomatoes in greenhouse production in
Germany (DE), Netherlands (NL), Italy (IT) and Spain (ES).
Lettuce Tomato
DE NL IT ES DE NL IT ES
Set-point heat; ◦C 16–17 18–19
Set-point vent; ◦C 17–18 19–20
Set-point RH; % 80 80
Set-point DLI; mol day
-1
10 – 10 –
Cultivation; months year
−1
11.5 9 11.5 9
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
4
(150 g), available on the supermarket shelf in the centre of Berlin, which
was dened as the gate. In this analysis, the cradle-to-gate principle was
used; i.e., no complete life cycle including waste management was
performed.
2.2. Process boundaries
The boundaries are shown in Fig. 1. The lower boundary of the
analysis was the production of raw material, such as the production of
seeds. The upper boundary was the delivery of products (tomato, let-
tuce) to the gate. System expansion was used to include the conse-
quences of, e.g., the green waste by transformation to biogas. Unless
otherwise stated in the Inventory section, data from the Ecoinvent 3
database (ver 3.6; consequential approach; Ecoinvent, 2019) were used
and, if necessary, adapted to the specic case.
This LCA ended at the environmental impact midpoint categories, i.e.
no weighing or normalisation was performed. To address the major im-
pacts of horticultural crop production with its known high consumption of
energy, nutrients and water, we selected the most suitable LCA midpoint
impact categories from three sources. The ReCiPe 2016 method (Huij-
bregts et al., 2017; ver. 1.1, H for Europe) was used for resource scarcity
(FRS; kg oil eq.), water consumption (WCO; m
3
), freshwater eutrophica-
tion (FEP; kg P eq.), mineral resource scarcity (MRS; kg Cu eq.), strato-
spheric ozone depletion (ODP; kg CFC-11 eq.), land use (ALU; m
2
a crop
eq.), human carcinogenic toxicity (HCT; kg 1,4-DB eq.), human
non-carcinogenic toxicity (HNT; kg 1,4-DB eq.), and terrestrial acidica-
tion (TAP; kg SO
2
eq.). Global warming potential (GWP; kg CO
2
eq.) was
analysed with two different time horizons of 20 and 100 years (GWP
20
or
GWP
100
, respectively) based on the latest IPCC method (clima-
techange2013.org) relevant for Greenhouse Gas Protocol, ISO14067 and
PAS2050. Water scarcity (WSC, m
3
) published by Berger et al. (2014) was
applied. Calculations were performed with the software tool SimaPro (ver.
9.1.1, PR´
e_Consultants, Amersfoort, Netherlands).
2.3. Inventory
The investigations included the three main components of the sys-
tem: RAS and HP, as well as roof-top installation where applicable (see
Table 1). Technical processes of both HP lettuce and tomato production
where included, with seed production, plant cultivation, harvest and
packaging, internal and external transport, plastic for wrapping, the
production of plastic foil for packaging and of cardboard packaging
boxes, etc. The life cycle inventory of the main processes is provided in
Supplementary Table 1. Sources of data and the main processes are
summarised in Supplementary Table 2.
To better compare the main environmental burden of greenhouse HP
production, a natural gas driven heat-power co-generation unit with 200
kWe (allocation exergy) was used for heating according to the global
approach (Ecoinvent, 2019) and was independent of the greenhouse
location. In the DAPS-R, HP-R, DAPS-R+and HP-R+scenarios (see
Table 1), heat demand was partly covered by the waste heat of the
connected industrial supermarket building. The energy consumed in the
supermarket was completely attributed to the supermarket. The biogas
produced from tomato green waste was burned in a combined
Fig. 2. Simulation results for hydroponically produced tomatoes in four locations (Germany, DE; the Netherlands, NL; Italy, IT; Spain, ES) for one year simulated
over 10 years (2009–2018) with heat energy consumption (A), electrical energy consumption for supplementary lighting (B), fresh water demand (C), fresh crop yield
(D), product water use (E), and product energy use (F).
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
5
heat-power co-generation unit (Jenbacher GE, Type 2) with an elec-
tricity share of 39.1% and a thermal share of 46%. Heat production was
used as input for the energy mix, while replacing coal energy that was
identied as a marginal heat resource according to the consequential
LCA theory. These shares were then attributed to HP tomato production
as an avoided product (see Fig. 1). For lettuce production, no green
waste was assumed. In the Netherlands and Germany, the production of
biogas from system waste was modelled using Ecoinvent 3.6 processes,
where the leakage of methane and dinitrogen monoxide was reduced to
0, i.e., a 100% sealed biogas reactor was assumed. No biogas production
was assumed for Spain and Italy. For electricity, country-specic energy
mixes with medium voltage were used (Ecoinvent, 2019). For HP, fer-
tiliser with NPK plus micronutrients was used, which is a typical mix
used for tomatoes (De Kreij et al., 1997), and the Hoagland solution was
used for lettuce. N was modelled as ammonium nitrate, P as triple su-
perphosphate, and K as potassium chloride.
For tomatoes and lettuce in greenhouse cultivation, a production
system either in a modern greenhouse heated with biogas and exhaust
heat for Berlin, or in a semi-climate-controlled greenhouses in Spain and
Italy was used. For both crops, the CLCA data based on the studies of
Stoessel et al. (2011) was used and adjusted to the specic conditions of
each system (yields, energy consumption, etc.) of the current study.
Crop production in Germany and the Netherlands was assumed to pro-
ceed throughout the year with ongoing planting of lettuce and
replanting of young tomato plants in weeks 51 and 52 of the year
(Table 2). In Spain and Italy, a summer break of 92 days for HP green-
house production was assumed. During this period, the German product
mix was covered from Germany and the Netherlands. For all scenarios,
lettuce in hydroponics was produced using nutrient lm technique
(NFT) and moving gutter systems (MGS) with data reported by K¨
orner
et al. (2018).
As a substrate for hydroponic production in tomatoes in Spain and
Italy, expanded perlite was used. For tomato production in the
Netherlands and Germany, stone wool was used and calculated similar
to stone wool insulation material (Schmidt et al., 2003). For DAPS, to-
matoes were grown in NFT in a stone wool cube with an inert eece
underneath (Ramírez et al., 2019). Fleece was modelled as viscose bre
from the global market. Dinitrogen monoxide (N
2
O) was added as a
direct supplement of 0.127 g kg
−1
in both DAPS and HP cultivations
(Aigner, 2003). It has been assumed that transplant production is per-
formed locally within countries using greenhouses. Seed production was
implemented as regular plant production of tomato and lettuce up to the
seed-bearing stage, and transplants were produced in nursery green-
houses for 21 and 14 days for tomato and lettuce, respectively.
For local transport, location-independent standard values for tractors
and trailers (including diesel consumption, construction, maintenance,
shed, etc.) were used. The same technology was used in all countries.
Transport was included using standard values for a EURO5 transport
truck of 7.5–16 t for short-distance transport (<50 km), of 16–32 t for
medium-distance transport (50–200 km), and EURO5 >32 t for long-
distance transport, including international transport (>200 km). For
all trucks, cooling was realised by means of a combined electricity/
diesel driven engine (TS600e, Thermo King). The transport and pack-
aging losses of the products (tomato and lettuce) were considered to be
Fig. 3. Simulation results for hydroponically produced lettuce in four locations (Germany, DE; the Netherlands, NL; Italy, IT; Spain, ES) for one year simulated over
10 years (2009–2018) heat energy consumption (A), electrical energy consumption for supplementary lighting (B), fresh water demand (C), fresh crop yield (D),
product water use (E), and product energy use (F).
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
6
7.5% in long-distance transport from Italy and Spain to Berlin, 3.5%
from the Netherlands to Berlin, and 2% within Germany. Waste man-
agement was included in the modelling for processes demanding waste
treatment.
3. Theory and calculation
3.1. Simulator
A numerical simulation model of HP and DAPS systems in green-
houses (Goddek and K¨
orner, 2019; K¨
orner and Hansen, 2012) was used
to compare virtual production of lettuce or tomato for all scenarios.
Simulations were performed with 5-min time-step and average over 1 h
for 10 independent yearly scenarios for the years 2009–2018. Location
specic hourly climate data were used as model input (MeteoBlue.com).
For direct comparison between locations and cultivation systems (i.e.,
HP and DAPS), basic theoretical greenhouse structures were created.
Multi-span glass greenhouses congured according to commercial
practice with passive heating and ventilation (K¨
orner et al., 2008), and
climate set-points as used in commercial practice were used (Table 2).
Supplementary lighting was installed in Germany and the Netherlands
(no supplementary lighting in Italy and Spain) with LED lamps installed
under the roof above the crop with an installed capacity of 80 W m
−2
power and an output of 192
μ
mol m
−2
s
−1
. Light was controlled
dynamically with set points generated using a daily light integral (DLI,
mol m
−2
d
−1
) (K¨
orner et al., 2006).
3.2. Crop growth
Crop growth and yield were simulated with a photosynthesis-driven
growth model based on a large collection of studies found in the litera-
ture and summarised by K¨
orner (2004). A commonly applied
biochemical-based leaf photosynthesis model was used as a basis (Yu et al.,
2020), while crop-specic parameters for lettuce and tomatoes were used.
Fresh mass and yield were calculated from simulated dry weight with a
fraction dry matter of 6.0% and 4.8% for lettuce and tomato, respectively.
The standard model incorporated the higher fresh yield of DAPS in lettuce
with a conservative approach using an increase fraction of 0.25, while
fractions of 0.1 and 0 were also tested in a sensitivity analysis. The ratio of
discarded tomato fruits due to quality problems was set to 5% and 1% in
HP and DAPS, respectively. Sensitivity analysis on the inuences of crop
production efciency in each of the four producer countries was done for
tomato as example. For each producing country, the total inuence on the
market available tomato mix was analysed with 25%, 50%, 75% and
100% production efciency.
3.3. Urban context scenarios
The greenhouses were either placed on the ground or on the roof of a
supermarket. When set on the ground, energy exchange to the oor was
calculated from the temperature difference between the soil and
greenhouse oor. For roof-top placement, two scenario options were
used: passive and active thermal energy transfer (denoted with ‘R’ or
‘R+’ in Table 1). In the passive scenarios, heat exchange was calculated
according to the supermarket roof, while with active heat transfer, the
exhaust heat from refrigerated cooling rooms in the supermarket was
used as supplementary air heating in the greenhouse when needed. For
Fig. 4. Comparing LCA for 500 g of packed tomatoes as available at the Berlin supermarket with different origins as Germany (DE), the Netherlands (NL), Italy (IT)
and Spain (ES) for 12 LCA impact categories with their indicators.
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
7
the latter case, 10% of the supermarket oor area was assumed with
refrigerated chambers with a constant temperature of 6 ◦C. The rooms
were controlled with an active cooler based on the ANSI/AHRI Stan-
dards 1200 (I-P) using R410A as the refrigerant and an isentropic
compressor efciency of 65%. The energy released in the condenser was
obtained and led to the greenhouse via isolated pipes, i.e., no transport
heat losses were assumed.
Single components of DAPS, i.e., RAS, desalination unit, and sludge
bioreactor, were modelled and optimally sized (Goddek and K¨
orner,
2019). The RAS system was physically located in the same air system as
HP, which enabled heat and CO
2
exchanges between the two main
sub-systems of aquaponics (K¨
orner et al., 2017).
4. Results and discussion
4.1. Differences in production origin
The market available fresh produce originates from various
geographical sources. Production in Northern Europe usually uses more
energy for heating while South European vegetable production is highly
water consuming (Fig. 2, Fig. 3). For instance, for tomatoes produced in
Germany, heating has a total consumption of 1.38 GJ m
−2
per year,
which is more than eight times higher than that in Spain. Electrical
energy in Spain and Italy only uses general electrics, such as pumps,
while a minimum DLI is maintained for electrical lighting in the
Netherlands and Germany. Hence, product energy use in Northern
Europe was more than four times higher than that in Spain and Italy,
which is in agreement with earlier reporting (Ant´
on et al., 2012; K¨
orner
et al., 2008; Torrellas et al., 2012). In contrast, water consumption is
relatively higher in Southern Europe. Consequently, Spain and Italy
have lower product energy use, despite the lower yields, but a higher
product water use of 70–90%.
The main differences between Northern and Southern European
greenhouse production can thus be identied by higher energy use for
heating and supplementary light in the North (using the current carbon-
bound energy mix) and higher water consumption in the South, where
desalinated sea water or unsustainably exploited groundwater often is
used (Custodio et al., 2019). Substantially higher use of energy for to-
mato production in Northern Europe leads to a lower global warming
potential and fossil resource scarcity potential for Spain and Italy (Fig. 4,
Fig. 5. Comparing LCA for 150 g of packed lettuce as available at the Berlin supermarket with different origins as Germany (DE), the Netherlands (NL), Italy (IT) and
Spain (ES) for 12 LCA impact categories with their indicators.
Table 3
Comparison between 150 g lettuce at production site in DAPS with yield increase
of 0%, 10% and 25% (i.e. standard) compared to local produced lettuce in HP
with the main inuenced impact indicators.
Cultivation Method DAPS to HP Reduction
(%)
DAPS HP –
DAPS
related
yield
increase →
0% 10% 25% 0% 0% 10% 25%
GWP
20
0.083 0.075 0.066 0.089 6.8 15.3 25.4
GWP
100
0.060 0.055 0.048 0.066 8.4 16.7 26.7
FRS 0.025 0.023 0.020 0.027 5.9 14.4 24.7
WCO −0.400 −0.364 −0.320 1.309 188.1 177.0 164.8
MRS −0.005 −0.005 −0.004 0.010 309.3 273.7 238.5
ODE 0.138 0.126 0.111 0.189 26.8 33.4 41.4
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
8
Fig. 5). As such, e.g. GWP
100
for Spanish or Italian packed tomatoes in
German supermarkets is approximately 44% and 48% of local produce,
respectively.
4.2. Combined systems
One advantage of DAPS is the recycling of nutrients, and
environmental impact reductions can partly be attributed to this
(Monsees et al., 2019). A signicantly higher lettuce yield of 25%
compared to HP was incorporated in DAPS, which was below reported
increase possibilities of 35% and more (Delaide et al., 2016; Goddek and
Vermeulen, 2018). This had a clear impact, while the general picture
does not change (Table 3). Thus, in terms of environmental impact,
locally producing greenhouse vegetables with DAPS is the best method,
Fig. 6. Percentage from absolute maximum case (100%) of 12 LCA impact categories with their indicators for availability of (A) 500 g of packed tomatoes or (B) 150
g of lettuce in Germany (DE) with different origins as Germany (DE, black), the Netherlands (NL, blue), Italy (IT, green) and Spain (ES, red).
Fig. 7. Simulation results for one-year tomato fruit production in four production systems of hydroponics (HP), decoupled aquaponics (DAPS), decoupled aquaponics
on a roof with passive energy exchange (DAPS-R), decoupled aquaponics on a roof with active energy exchange (DAPS-R+) in Berlin (Germany, DE) with heat energy
consumption (A), fresh crop yield (B), product water use (C), and product energy use (D).
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
9
preferably in combination with waste heat usage (Fig. 6).
DAPS mainly reduce the water- and energy-related midpoint cate-
gories compared to HP, and all the DAPS scenarios showed the same
trend. Mainly four key issues lead to these reductions. First, the sh
tanks in the current study, which were kept at 30 ◦C (assuming tilapia
culture), were located within the plant production area and released low
energy heat to the greenhouse environment. Consequently, the DAPS
used less greenhouse heating energy (K¨
orner et al., 2017). Second, most
Fig. 8. Simulation results for one-year lettuce production in four production systems of hydroponics (HP), decoupled aquaponics (DAPS), decoupled aquaponics on a
roof with passive energy exchange (DAPS-R), decoupled aquaponics on a roof with active energy exchange (DAPS-R+) in Berlin (Germany, DE) with heat energy
consumption (A), fresh crop yield (B), product water use (C), and product energy use (D).
Fig. 9. Comparing LCA for 500 g of packed tomatoes from import mix (MIX), local hydroponics (HP), decoupled aquaponics (DAPS), and decoupled aquaponics on a
roof with active energy exchange (DAPS-R+) for 12 LCA impact categories with their indicators.
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
10
plant nutrients originate from the sh system. Hence, the energy needed
to produce a considerable amount of fertiliser can be signicantly
reduced (Baganz et al., 2020; Goddek, 2017). Third, all the DAPS were
located close to the market, and thus transport was minimised. Fourth,
in DAPS, no water is spoiled, and besides the power needed for irrigation
pumps, no additional water supply is needed for the plants when the
system is optimally balanced (Keesman et al., 2019).
Thermal energy consumption in the different systems under
consideration shows a similar behaviour for tomato and lettuce pro-
duction (Fig. 7, Fig. 8). The highest energy use was calculated for regular
HP, while energy reductions of 8% and 10% (lettuce and tomato) were
possible replacing HP with DAPS. Additional reduction of energy con-
sumption were achieved when combining DAPS with active waste heat
transfer from a connected building, hence under scenarios HP-R+and
DAPS-R+(for both tomato and lettuce). Combining greenhouses with
buildings has a high potential on energy savings, which depends on
individual set up and location (Nadal et al., 2017). In our case, the total
energy consumption could be reduced by 18% from regular HP to
HP-R+in both crops and by 19% and 30% with DAPS-R+in tomato and
lettuce, respectively. Energy savings of 13% were earlier reported for
roof-top greenhouses in humid continental climates and roof-adjusted
structures (Torres Pineda et al., 2020), while the actual settings and
climate conditions strongly inuence the possible impact reductions. In
our case, the passive reuse of waste heat reduced energy consumption
only marginally; i.e., ca. 1% in HP and 3% in DAPS.
4.3. Impact of DAPS on environmental impact categories
In aquaponics, heat, electricity, equipment, and sh feed are re-
ported as the four main environmental impact hotspots (Ghamkhar
et al., 2020). However, in the current LCA, a sharp line was drawn be-
tween the two main sub-systems of DAPS, i.e., HP and RAS. As such, RAS
was used as a provider of water, nutrients, heat and, to a lesser extent,
CO
2
for hydroponic crop production. For both lettuce and tomato,
moving from regular HP to DAPS reduced the global warming potential
(and fossil resource scarcity) and hence the emissions of greenhouse
gases to the atmosphere. Water consumption (WCO, m
3
) was highest for
available market mixes (Fig. 9, Fig. 10), with a high contribution of
imported products. Product water use in Figs. 2 and 3 illustrates that,
Fig. 10. Comparing LCA for 150 g of packed and sliced lettuce from import mix (MIX), local hydroponics (HP), decoupled aquaponics (DAPS), and decoupled
aquaponics on a roof with active energy exchange (DAPS-R+) for 12 LCA impact categories with their indicators.
Table 4
Percentage saving on impact indicators of different production methods
compared to the market available bag of cut lettuce of (150 g) or a box of 500 g
tomatoes at a local market in Berlin.
Local Production Method
HP DAPS DAPS-R DAPS-R+
Lettuce GWP
20
2.2 27.3 29.8 44.0
GWP
100
11.8 35.6 37.8 49.9
FRS 3.5 27.6 30.1 44.2
WCO 59.9 108.8 108.4 106.1
WSC 93.4 106.6 106.3 104.7
FEP 72.3 41.8 35.9 1.4
MRS 40.9 121.9 120.3 111.3
ODE 4.2 43.7 44.6 49.8
ALO 311.1 281.0 271.9 218.6
HCT 69.3 38.4 32.8 −0.1
HNT 491.9 501.4 477.0 333.7
TAP 275.7 214.9 200.9 117.5
Tomato GWP
20
−30.0 −11.6 −8.3 2.1
GWP
100
−19.6 −5.3 −2.0 8.7
FRS −34.3 −5.7 −3.2 4.7
WCO 66.1 175.8 174.7 171.1
WSC 137.6 157.3 156.7 154.6
FEP 68.4 79.9 76.4 65.1
MRS 330.0 339.5 337.9 332.9
ODE 442.1 568.6 590.0 657.9
ALO 99.1 83.8 83.5 82.7
HCT 86.6 122.0 114.6 90.2
HNT 210.2 289.5 276.6 234.1
TAP 131.2 116.5 109.3 85.3
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
11
and it is in agreement with earlier results of tomatoes produced in
Mediterranean greenhouse (Payen et al., 2015). The general picture is
evident: Local Northern European produced lettuce and tomatoes have
strong inuence on the energy related impact categories (GWP
20
,
GWP
100
, FRS); while water based impacts (WCO, WSC) are mainly a
problem in Southern Europe. Some environmental impacts could
improve by solely shifting from import to 100% local production
(Table 4). However, the combination of local production and DAPS (in
particular DAPS-R+), can strongly reduce all environmental impact
categories, with strongest effect on those related to energy, water and
mineral resources (Table 4). Using the complete programme of roof-top
decoupled aquaponics in DAPS-R+compared to the available market
mix, WCO dropped from water consumption to water saving: i.e. from
14.2 L or 3.3 L to −10.1 L or −0.21 L per package of 500 g tomatoes or
150 g lettuce, respectively. GWP
100
and FRS were below the values of
the available market mixes, e.g. GWP
100
decreased with 8.7% in to-
matoes and 49.9% in lettuce (Table 4, Fig. 11).
4.4. Methods and data
In LCA, data quality and choice of method are key issues. Given the
complexity of the system with various co-products, CLCA with system
expansion was dened as best method for our scenarios (Weidema and
Schmidt, 2010). While CLCA was earlier used in aquaponics (Boxman
et al., 2017), only few LCA studies have been done on this topic
(Ghamkhar et al., 2020). The data situation is limited (Boxman et al.,
2017), while it even worsens with aquaponics in urban environments
(Wu et al., 2019). We have addressed that problem incorporating the
output from a model-simulation study to LCA. For the rst time, a
simulation study with a widely validated aquaponics simulator (Goddek
and K¨
orner, 2019) was combined with LCA. Through that, production
data from various geographical locations and individual production
systems could be used for real market scenarios. While the composition
of the market available mixes of lettuce and tomatoes where a result of
the present fresh market situation in Berlin (Behr, 2019; Workman,
2020a, b), production efciency inuences the contributions on envi-
ronmental impacts from each producing country. In addition, the larger
the share of a certain origin country on any impact category, the
stronger is its inuence on the total impact (Table 5). For instance, a
change in production efciency in 25% of the Netherlands would have a
total impact of 22.4% on GWP
100
; the same change in Spain results in a
higher GWP
100
of only 3.4%. Focusing on water turns that picture
around. With the same scenario, the WCO for 500 g packed tomatoes on
the Berlin market would increase by 5.6% or 17.7% due to the
Netherlands or Spain, respectively. Thus, although the reported results
are valid in the current situation, any adjustments in the complex global
producer market will have an impact. However, with the here presented
methodology, also future market scenarios could be analysed.
5. Conclusions and future prospective
Choosing the right set-up, local vegetable productions in urban re-
gions can surpass the import mix on environmental performance in
Northern European centres. Production in Northern European countries
uses more energy (mostly carbon-bound, currently), while product
water use is signicantly higher in Southern Europe (inter alia desali-
nated seawater). When replacing HP with aquaponics, some resources
could be attributed to RAS, which partly reduces the global warming
potential for plant production and water-related environmental impacts
can be strongly reduced. Placing the DAPS in a local urban context using
the rooftop of an existing industrial building further reduces GWP and
other environmental impact through reduced transport, area re-usage
and a combined possibility for waste heat usage. In this scenario,
active reuse of waste heat was the most effective method of environ-
mental impact reductions. Our results clearly illustrate the strong posi-
tive effect of both local food production in combined production
systems. Increase in complexity and technology as shown in the top
solution analysed here, decrease environmental impact. In the next step,
additional co-products need to be added to the system, further
increasing its complexity. Here, we expect increasing synergy and
further improvements of environmental impacts of the produce. Then, to
decrease the error in LCA studies, system expansion becomes increas-
ingly important (Weidema and Schmidt, 2010).
CRediT authorship contribution statement
Oliver K¨
orner: Conceptualization, Methodology, Software, Formal
analysis, Investigation, Data curation, Writing – original draft,
Fig. 11. Percentage from absolute maximum case
(100%) of 12 LCA impact categories with their in-
dicators for availability of (A) 500 g of packed to-
matoes or (B) 150 g of lettuce in Germany with ve
scenarios (from left to right in the column clusters):
available market mix (grey), local hydroponics
(green), decoupled multi-loop aquaponics (DAPS,
light blue)), decoupled multi-loop aquaponics on a
roof with passive energy exchange (DAPS-R, blue),
decoupled multi-loop aquaponics on a roof with
active energy exchange (DAPS-R+, dark blue).
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
12
Visualization. Mehdi B. Bisbis: Writing – original draft, Writing – re-
view & editing. G¨
osta F.M. Baganz: Investigation, Data curation,
Writing – review & editing. Daniela Baganz: Writing – review & edit-
ing, Supervision, Funding acquisition. Georg B.O. Staaks: Writing –
review & editing. Hendrik Monsees: Writing – review & editing. Simon
Goddek: Software, Writing – review & editing. Karel J. Keesman:
Validation, Writing – review & editing, Project administration.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgements
We gratefully acknowledge funding support from the Belmont Forum
and the European Commission via CITYFOOD (grant agreement No
726744).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jclepro.2021.127735.
References
Aigner, H., 2003. Denitrication in Cultures of Potted Ornamental Plants. Universit¨
at
Hannover, Hannover, Germany, p. 133.
Ant´
on, A., Torrellas, M., Montero, J., Ruijs, M., Vermeulen, P., Stanghellini, C., 2012.
Environmental impact assessment of Dutch tomato crop production in a Venlo
glasshouse. Acta Hortic. 927, 781–791.
Baganz, G., Baganz, D., Staaks, G., Monsees, H., Kloas, W., 2020. Protability of multi-
loop aquaponics: year-long production data, economic scenarios and a
comprehensive model case. Aquacult. Res. 2020 (51), 2711–2724.
Behr, H.C., 2019. The Market for Fresh Tomatoes in Germany Global Tomato Congress.
Rotterdam, The Netherlands.
Berger, M., Van der Ent, R., Eisner, S., Bach, V., Finkbeiner, M., 2014. Water accounting
and vulnerability evaluation (WAVE): considering atmospheric evaporation
recycling and the risk of freshwater depletion in water footprinting. Environ. Sci.
Technol. 48 (8), 4521–4528.
Boxman, S.E., Qiong, Z., Doland, B., Trotz, M., 2017. Life cycle assessment of a
commercial freshwater aquaponic system. Environ. Eng. Sci. 35 (5), 299–311.
Brand˜
ao, M., Martin, M., Cowie, A., Hamelin, L., Zamagni, A., 2017. Consequential life
cycle assessment: what, how, and why? In: Abraham, M.A. (Ed.), Encyclopedia of
Sustainable Technologies. Elsevier, pp. 277–284.
Cohen, A., Malone, S., Morris, Z., Weissburg, M., Bras, B., 2018. Combined sh and
lettuce cultivation: an aquaponics life cycle assessment. Procedia CIRP 69, 551–556.
Cuce, E., Harjunowibowo, D., Cuce, P.M., 2016. Renewable and sustainable energy
saving strategies for greenhouse systems: a comprehensive review. Renew. Sustain.
Energy Rev. 64, 34–59.
Custodio, E., Sahuquillo, A., Albiac, J., 2019. Sustainability of intensive groundwater
development: experience in Spain. Sustain. Water Resourc. Manag. 5 (1), 11–26.
De Cicco, A., 2019. The Fruit and Vegetable Sector in the EU - a Statistical Overview
Eurostat.
De Kreij, C., Voogt, W., Van den Bos, A.L., Baas, R., 1997. Nutrient Solutions for the
Growth of Tomato in Closed Systems (In Dutch). Brochure VG 3. Proefstation voor
Bloemistrij en Glasgroenten, Naaldwijk, The Netherlands.
Delaide, B., Teerlinck, S., Decombel, A., Bleyaert, P., 2019. Effect of wastewater from a
pikeperch (Sander lucioperca L.) recirculated aquaculture system on hydroponic
tomato production and quality. Agric. Water Manag. 226, 105814, 1-9.
Delaide, B.P.L., Goddek, S., Gott, J., Soyert, H., Jijakli, M.H., 2016. Lettuce (Lactuca
sativa L. Sucrine) growth performance in complemented aquaponic solution
outperforms hydroponics. Water 8, 467.
Ecoinvent, 2019. Version 3.6. Simapro Database Swiss Centre for Life Cycle Inventories.
Fitwi, B.S., 2012. Environmenta evaluation of aquaculture using life cycle assessment
(LCA), Agrar- und Ern¨
ahrungswissenschaftlichen Fakult¨
at. Christian-Albrechts-
Universit¨
at zu Kiel, p. 145.
Gava, O., Bartolini, F., Venturi, F., Brunori, G., Zinnai, A., Pardossi, A., 2018. A reection
of the use of the life cycle assessment tool for agri-food sustainability. Sustainability
11 (71), 1–16.
Ghamkhar, R., Hartleb, C., Wu, F., Hicks, A., 2020. Life cycle assessment of a cold
weather aquaponic food production system. J. Clean. Prod. 244, 118767.
Goddek, S., 2017. Opportunities and Challenges of Multi-Loop Aquaponic Systems.
Wageningen University, p. 170.
Goddek, S., Bl¨
aser, I., Reuter, M., Wuertz, S., K¨
orner, O., Joyce, A., Keesman, K.J., 2019.
Decoupled aquaponic systems. In: Goddek, S., Joyce, A., Kotzen, B., Burnell, G.
Table 5
Sensitivity analysis on the effects of the yields mixes with individual reductions per country of 25%, 50%, and 75% yields in Germany (Mix DE), The Netherlands (Mix NL), Italy (Mix IT), and Spain (Mix ES) compared to
local production and the market available mix (MIX).
Local MIX Mix DE Mix NL Mix IT Mix ES
HP DAPS DAPS-R+
Prod. Efciency (%) → 100 100 100 100 25 50 75 25 50 75 25 50 75 25 50 75
GWP
20
kg CO
2
eq 1.08 0.93 0.81 0.83 1.09 0.92 0.86 2.62 1.43 1.03 0.88 0.85 0.84 1.08 0.91 0.86
GWP
100
kg CO
2
eq 0.69 0.61 0.53 0.58 0.74 0.63 0.59 1.74 0.96 0.71 0.62 0.59 0.58 0.81 0.65 0.60
FRS kg oil eq 0.45 0.36 0.32 0.34 0.44 0.37 0.35 1.09 0.59 0.42 0.35 0.34 0.34 0.41 0.36 0.35
WCO m
3
(10
−3
) 4.8 −10.7 −10.1 14.2 15.3 14.6 14.3 21.6 16.7 15.0 19.0 15.8 14.7 36.8 21.7 16.7
WSC m
3
(10
−3
) −4.1 −6.2 −5.9 10.9 9.9 10.6 10.8 20.1 14.0 11.9 13.9 11.9 11.2 27.3 16.3 12.7
FEP kg P eq (10
-3
) −0.54 −0.58 −0.53 −0.32 −0.45 −0.36 −0.33 −1.17 −0.60 −0.41 −0.31 −0.32 −0.32 −0.29 −0.31 −0.32
MRS kg Cu eq (10
-3
) −0.47 −0.49 −0.47 0.20 0.10 0.17 0.19 −0.56 −0.05 0.12 0.44 0.28 0.23 1.21 0.54 0.32
ODE kg CFC11 eq (10
-6
) −0.54 −0.67 −0.75 −0.10 −0.22 −0.14 −0.11 −0.99 −0.40 −0.20 0.04 −0.05 −0.08 0.38 0.06 −0.05
ALO m
2
a crop eq −0.63 −0.58 −0.58 −0.32 −0.46 −0.37 −0.33 −1.35 −0.66 −0.43 −0.25 −0.30 −0.31 −0.17 −0.27 −0.30
HCT kg 1,4-DB (10
−3
) −18.3 −21.8 −18.7 −9.8 −14.2 −11.3 −10.3 −36.9 −18.9 −12.8 −9.3 −9.7 −9.8 −7.8 −9.1 −9.6
HNT kg 1,4-DB −0.37 −0.47 −0.40 −0.12 −0.21 −0.15 −0.13 −0.62 −0.29 −0.18 −0.07 −0.10 −0.11 0.04 −0.07 −0.10
TAP kg SO
2
eq (10
−3
) −2.52 −2.36 −2.02 −1.09 −1.68 −1.29 −1.16 −4.76 −2.31 −1.50 −0.90 −1.03 −1.07 −0.34 −0.84 −1.01
O. K¨
orner et al.

Journal of Cleaner Production 313 (2021) 127735
13
(Eds.), Aquaponics Food Production Systems. Springer Nature, SpringerOpen,
pp. 201–229.
Goddek, S., Keesman, K.J., 2018. The necessity of desalination technology for designing
and sizing multi-loop aquaponics systems. Desalination 428, 76–85.
Goddek, S., K¨
orner, O., 2019. A fully integrated simulation model of multi-loop
aquaponics: a case study for system sizing in different environments. Agric. Syst.
171, 143–154.
Goddek, S., Vermeulen, T., 2018. Comparison of Lactuca sativa growth performance in
conventional and RAS-based hydroponic systems. Aquacult. Int. 26, 1377–1386.
Gruda, N., Bisbis, M., Tanny, J., 2019. Impacts of protected vegetable cultivation on
climate change and adaptation strategies for cleaner production – a review. J. Clean.
Prod. 225, 324–339.
Huijbregts, M.A.J., Steinmann, Z.J.N., Elshout, P.M.F., Stam, G., Verones, F., Vieira, M.,
Zijp, M., Hollander, A., Van Zelm, R., 2017. ReCiPe2016: a harmonised life cycle
impact assessment method at midpoint and endpoint level. Int. J. Life Cycle Assess.
22, 138–147.
Keesman, K.J., K¨
orner, O., Wagner, K., Urban, J., Karimanzira, D., Rauschenbach, T.,
Goddek, S., 2019. Aquaponics system modelling. In: Goddek, S., Joyce, A.,
Kotzen, B., Burnell, G. (Eds.), Aquaponics Food Production Systems. Springer
Nature, SpringerOpen, pp. 267–299.
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. Aquacult. Environ. Interact. 7, 179–192.
K¨
orner, O., 2004. Evaluation of crop photosynthesis models for dynamic climate control.
Acta Hortic. 654, 295–302.
K¨
orner, O., Andreassen, A.U., Aaslyng, J.M., 2006. Simulating dynamic control of
supplementary lighting. Acta Hortic. 711, 151–156.
K¨
orner, O., Gutzmann, E., Kledal, P.R., 2017. A dynamic model simulating the symbiotic
effects in aquaponic systems. Acta Hortic. 1170, 309–316.
K¨
orner, O., Hansen, J.B., 2012. An on-line tool for optimising greenhouse crop
production. Acta Hortic. 957, 147–154.
K¨
orner, O., Pedersen, J.S., Jaegerholm, J., 2018. Simulating lettuce production in multi
layer moving gutter systems. Acta Hortic. 1227, 283–290. https://doi.org/
10.17660/ActaHortic.2018.1227.34.
K¨
orner, O., Warner, D., Tzilivakis, J., Eveleens-Clark, B., Heuvelink, E., 2008. Decision
support for optimising energy consumption in European greenhouses. Acta Hortic.
801, 803–810.
Monsees, H., Suhl, J., Paul, M., Kloas, W., Dannehl, D., Würtz, S., 2019. Lettuce (Lactuca
sativa, variety Salanova) production in decoupled aquaponic systems: same yield
and similar quality as in conventional hydroponic systems but drastically reduced
greenhouse gas emissions by saving inorganic fertilizer. PloS One 16 (6), e0218368.
Nadal, A., Llorach-Massana, P., Cuerva, E., L´
opez-Capel, E., Montero, J.I., Josa, A.,
Rieradevall, J., Royapoor, M., 2017. Building-integrated rooftop greenhouses: an
energy and environmental assessment in the mediterranean context. Appl. Energy
187, 338–351.
Ntinas, G.K., Dannehl, D., Schuch, I., Rocksch, T., Schmidt, U., 2020. Sustainable
greenhouse production with minimised carbon footprint by energy export. Biosyst.
Eng. 189, 164–178.
Opdam, J.J.G., Schoonderbeek, G.G., Heller, E.M.B., De Gelder, A., 2005. Closed
greenhouse: a starting point for sustainable entrepreneurship in horticulture. Acta
Hortic. 691, 517–524.
Payen, S., Basset-Mens, C., Perret, S., 2015. LCA of local and imported tomato: an energy
and water trade-off. J. Clean. Prod. 87, 139–148.
Pluimers, J.C., 2001. An Environmental System Analysis of Greenhouse Horticulture in
the Netherlands. Wageningen University, Wageningen, The Netherlands, p. 181.
Ramírez, T., Kl¨
aring, H.P., K¨
orner, O., 2019. Does interruption of supplementary lighting
affect the long-term carbon dioxide exchange of greenhouse tomato crops? Biosyst.
Eng. 187, 69–80.
Savvas, D., Chatzieustratiou, E., Pervolaraki, G., Gizas, G., Sigrimis, N., 2008. Modelling
Na and Cl concentrations in the recycling nutrient solution of a closed-cycle pepper
cultivation. Biosyst. Eng. 99 (2).
Schmautz, Z., Loeu, F., Liebisch, F., Graber, A., Mathis, A., Bulc, T.G., Junge, R., 2016.
Tomato productivity and quality in aquaponics: comparison of three hydroponic
methods. Water 8 (533), 1–21.
Schmidt, A., Ulf, A., Jensen, A.A., Kamstrup, O., 2003. In: Documents, L. (Ed.),
Comparative life lycle assessment of three insulation materials -stone wool, ax and
paper wool. Ecomed publishers.
Stoessel, F., Juraske, R., Pster, S., Hellweg, S., 2011. Life cycle inventory and carbon
and water FoodPrint of fruits and vegetables: application to a Swiss retailer. Environ.
Sci. Technol. 6 (46), 3253–3262.
Torrellas, M., Ant´
on, A., L´
opez, J.C., Baeza, E.J., Parra, J.P., Mu˜
noz, P., Montero, J.I.,
2012. LCA of a tomato crop in a multi-tunnel greenhouse in Almeria. Int. J. Life
Cycle Assess. 17 (7), 863–875.
Torres Pineda, I., Cho, J.H., Lee, D., Lee, S.M., Yu, S., Lee, Y.D., 2020. Environmental
impact of fresh tomato production in an urban rooftop greenhouse in a humid
continental climate in South Korea. Sustainability 12 (21), 9029.
Vadiee, A., Martin, V., 2012. Energy management in horticultural applications through
the closed greenhouse concept, state of the art. Renew. Sustain. Energy Rev. 16,
5087–5100.
Weidema, B., 2003. Market Information in Life Cycle Assessment, Environmental Project
No. 863 2003. Danish Ministry of the Environment, Copenhagen, Denmark, p. 147.
Weidema, B.P., Schmidt, J., 2010. Avoiding allocation in life cycle assessment revisited.
J. Ind. Ecol. 14 (2), 192–195.
Workman, D., 2020a. Tomatoes Imports by Country. worldstopexports.com.
Workman, D., 2020b. Top Lettuce Exports by Country. worldstopexports.com.
Wu, F., Ghamkhar, R., Ashton, W., Hicks, A.L., 2019. Sustainable seafood and vegetable
production: aquaponics as a potential opportunity in urban areas. Integrated
Environ. Assess. Manag. 15 (6), 832–843.
Yu, W., K¨
orner, O., Schmidt, U., 2020. Crop photosynthetic performance monitoring
based on a combined system of measured and modelled chloroplast electron
transport rate in greenhouse tomato. Front. Plant Sci. 11 (1038), 1–15.
O. K¨
orner et al.