Life cycle assessment of heterotrophic algae omega-3

Abstract

Fish oil has traditionally been the primary source of long chain omega-3 fatty acids, which are essential nutrients for human diets as well as many aquaculture and animal feeds. The demand for fish oil is growing rapidly, due to an expanding aquaculture sector as well as rising demand in pet and livestock feeds, while the availability of fish oil from wild caught fish has leveled off over the past decade. Fish oil is not easily replaced and alternative sources are required to meet the growing global demand. One of the most promising alternatives is microalgae - the original source of long chain omega-3s. To understand the environmental impacts of omega-3s produced by heterotrophic algae, a comprehensive Life Cycle Assessment (LCA), assessing six impact categories, was conducted for two algae omega-3 DHA (docosahexaenoic acid) products, in powder and liquid suspension formats. These products are manufactured at industrial scale using sugarcane for both feedstock and a renewable energy source. The life cycle impact assessment results for algae omega-3 DHA indicate that sugarcane cultivation has the largest contribution for most of the categories. A sensitivity analysis revealed that the sugarcane yield, and the sugar to omega-3 DHA yield, were the most relevant parameters, and that the choice of allocation methodology did not have a strong influence on the results. A comparison with fish oil, using data publicly available in LCA databases, indicated that the two formats of the commercial algae omega-3 DHA product offer about 30–40% lower impact on climate change than fish oil.

Full text

Algal Research 60 (2021) 102494
Available online 9 October 2021
2211-9264/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Life cycle assessment of heterotrophic algae omega-3
Dillon Davis
a
, Ana Mor˜
ao
b
,
*
, Jill Kauffman Johnson
c
, Li Shen
a
a
Copernicus Institute of Sustainable Development, Utrecht University, Utrecht, the Netherlands
b
Corbion Sustainability department, Gorinchem, the Netherlands
c
Corbion Algae Ingredients, South San Francisco, CA, USA
ARTICLE INFO
Keywords:
Omega-3
Feed
Life cycle assessment (LCA)
Fish oil
Sugarcane
Algae
ABSTRACT
Fish oil has traditionally been the primary source of long chain omega-3 fatty acids, which are essential nutrients
for human diets as well as many aquaculture and animal feeds. The demand for sh oil is growing rapidly, due to
an expanding aquaculture sector as well as rising demand in pet and livestock feeds, while the availability of sh
oil from wild caught sh has leveled off over the past decade. Fish oil is not easily replaced and alternative
sources are required to meet the growing global demand. One of the most promising alternatives is microalgae -
the original source of long chain omega-3s. To understand the environmental impacts of omega-3s produced by
heterotrophic algae, a comprehensive Life Cycle Assessment (LCA), assessing six impact categories, was con-
ducted for two algae omega-3 DHA (docosahexaenoic acid) products, in powder and liquid suspension formats.
These products are manufactured at industrial scale using sugarcane for both feedstock and a renewable energy
source. The life cycle impact assessment results for algae omega-3 DHA indicate that sugarcane cultivation has
the largest contribution for most of the categories. A sensitivity analysis revealed that the sugarcane yield, and
the sugar to omega-3 DHA yield, were the most relevant parameters, and that the choice of allocation meth-
odology did not have a strong inuence on the results. A comparison with sh oil, using data publicly available in
LCA databases, indicated that the two formats of the commercial algae omega-3 DHA product offer about
30–40% lower impact on climate change than sh oil.
1. Introduction
Long chain omega-3 fatty acids—eicosapentaenoic (EPA) and doco-
sahexaenoic (DHA) acids—are essential for human diets as well as many
aquaculture and animal feeds. The developmental and health benets of
long chain omega-3s for brain, eye and heart health for humans is well
documented [1]. Many of these same benets are important for animal
development and growth as well.
Fish oil has traditionally been the primary source of long chain
omega-3s, but is sourced from nite marine sheries, and supply has
leveled off at around 1.1 million tons in recent years [2]. A recent study
tracked ows of long chain omega-3 stocks through global production,
supply and utilization pathways and found a signicant gap in supply
[3]. This supply gap will continue to widen as demand rises, especially
from the aquaculture sector which consumes approximately 75% of the
available crude sh oil [4]. Aquaculture is one of the fastest growing
food systems in the world, with fed species representing nearly 70% of
global aquaculture production [5]. Aquafeed, especially for carnivorous
species, historically relied on forage sh for sh meal and sh oil, to
provide the required nutrients such as protein and omega-3 fatty acids.
With the global supply of forage sh at a plateau, there has been rising
demand to limit the use of wild marine sh in aquaculture feed in order
to reduce pressure on already stressed wild sheries [6]. A recent study
of salmon feed in Norway showed that over 70% of aquafeed is now
sourced from plant origins [7]. For example, sh meal was the original
source of protein in aquafeed and now is partially being replaced by soy
protein [7]. Novel ingredients such as insect meal and protein derived
from single cell bacteria are emerging as alternative protein sources to
sh meal and soy [8].
In an effort to reduce the use of sh oil in aquafeed, some sh oil has
been replaced by rapeseed or camelina oil which meets the energetic
needs of farmed sh, but not the omega-3 needs [7]. Dropping levels of
long chain omega-3s in carnivorous sh, especially salmon, can impact
sh health and welfare [9]. In addition, the change in nutritional quality
of salmon, an increasingly popular species in the global market, can
impact public health due to less intake of long chain omega-3s [10].
Fish oil contains key long chain omega-3 polyunsaturated fatty acids
(PUFAs), eicosapentaenoic acid and docosahexaenoic which are not
* Corresponding author.
E-mail address: ana.morao@corbion.com (A. Mor˜
ao).
Contents lists available at ScienceDirect
Algal Research
journal homepage: www.elsevier.com/locate/algal
https://doi.org/10.1016/j.algal.2021.102494
Received 12 March 2021; Received in revised form 7 September 2021; Accepted 15 September 2021

Algal Research 60 (2021) 102494
2
easily replaced. Therefore, alternative sources are required to meet the
growing demands, not only of aquaculture, but also livestock production
and direct human consumption [11]. Thus, scientists have looked to the
original source of long chain omega-3 PUFAs – microalgae –as a po-
tential solution.
Heterotrophic microalgae from genera such as Schizochytrium,
Ulkenia, and Crypthecodinium produce algae products that are rich in
DHA [12]. Algae from these genera have been selected based on their
ability to produce large quantities of DHA from a carbon feedstock,
through conventional fermentation.
Corbion has developed and commercializes omega-3 DHA products
which utilize an algae of the Schizochytrium genus to produce biomass
that is rich in DHA and is applied as an alternative to sh oil in feed for
aquaculture and livestock, as well as pet food. This heterotrophic
microalgae product is manufactured via fermentation in Orindiúva,
Brazil, using sugar from sugarcane as feedstock. The production site is
co-located with a sugar mill enabling the efcient use of energy from
sugarcane by-products. This novel omega-3 DHA source shows promise
for reducing the amount of wild sh needed in feed for farmed sh and
livestock, as well as pet food, while maintaining or improving animal
health.
A comprehensive understanding of the environmental impacts of
algae feed ingredients is lacking [8]. Although algae-based ingredients
reduce the demand of wild sh, other environmental impacts associated
with the production of these algae ingredients must be evaluated. The
potential environmental trade-offs need to be well understood to ensure
the sustainability of algae-based omega-3 production. Life cycle assess-
ment (LCA) is a well-established methodology that uses a holistic
approach to identify trade-offs between environmental impacts to avoid
shifting burdens over the value chain. Due to characteristics such as fast
growth, high cell density, and high oil productivity, heterotrophic algae
systems are of increasing interest [13], however, to our knowledge,
there are no LCAs available in the public domain about commercialized
heterotrophic algae systems [13–15]. The literature studies on the LCA
of heterotopic algae systems focus on biodiesel production. Other LCA's
on feed and food production of protein and oil ingredients are based on
lab or pilot data [14–16].
This paper aims at providing cradle-to-gate LCA information on
commercially produced omega-3 DHA from heterotrophically grown
microalgae to scientists, feed producers, and aquaculture and animal
farmers, who seek to understand the environmental sustainability as-
pects of heterogeneous algae products. The novelty of this LCA study is
the use of industrial data for algae omega-3 production. Additionally,
the LCA results are compared with the environmental impact of pro-
ducing omega-3s from sh oil. The commercial production of omega-3
DHA via heterotrophic microalgae has optimization potential that can
be realized through continuous product and process development. In
this study, the impact of some of these improvements are quantied to
provide an outlook of the potential of algae-based products.
2. Materials and methods
This LCA was performed according to the standard methodology
described in ISO 14040 series by the International Organization of
Standardization [17,18]. The LCA model was created in SimaPro
developer software version 9.1 (PR´
e Sustainability, The Netherlands).
2.1. Goal and scope denition
The goal of this LCA is to quantify the environmental footprint of
omega-3 DHA produced from heterotrophically grown microalgae,
commercialized under the name of AlgaPrime™ DHA. The results are
intended to be used as input in the LCA of feed formulations and to
understand the environmental sustainability of the sh oil alternative
for omega-3s. The study relied on an attributional LCA approach,
therefore the potential impact of feed and food system transformation
due to heterotrophic microalgae production was not considered.
The functional unit was dened as 1 kg omega-3 fatty acids. Although
other components of the algae biomass, such as proteins, carbohydrates,
and other fatty acids, have nutritional value for feed, omega-3s are the
reason for inclusion of the algae product in feed. For this reason, the
choice of the functional unit reects the primary function of the algae
omega-3 DHA product. Furthermore, this functional unit allows com-
parison between the algae omega-3 products and sh oil. In the case of
the algae product studied, DHA is the only long chain omega-3 fatty
acid, whereas in the case of sh oil, both EPA and DHA are present.
Traditionally, aquafeed contains varying ratios of EPA and DHA
depending on the type and amount of sh oil included. In a study con-
ducted on salmon, it was shown that omega-3s are essential to the
growth and health of the sh and when omega-3s were excluded from
the diet entirely, the sh growth rate was slower and overall health of
the sh decreased [9]. DHA appears to be the more essential of the two
PUFAs and has been suggested that the long chain PUFA requirement of
healthy salmon can be met by DHA alone [19]. Consistent with previous
research, a recent study found that DHA has higher omega-3 retention in
the let of salmon versus EPA, which is typically metabolized by the sh
[20]. In additional studies by Noma, one of the leading aquaculture
research institutes in Norway, it was determined that DHA rich Schizo-
chytrium microalgae effectively replaces sh oil in diets of Atlantic
salmon in both freshwater and sea-water life stages [21].
The omega-3 DHA algae is produced in two feed ingredient formats -
powder and liquid suspension - to offer exibility for feed producers. A
diversity of feed formulation formats is needed for aquaculture species,
livestock, and pets due to consumption patterns, digestive and absorp-
tion dynamics, and operational application needs. Both formats are
included in the scope of this study:
1) Powder or biomass - Oil high in omega-3 DHA is encapsulated within
the algae cell to form a free-owing dry powder.
2) Liquid suspension – Blend of omega-3 DHA-rich algae and vegetable
oil.
Hereafter, we will refer to these products as “algae omega-3 DHA
powder” and “algae omega-3 DHA liquid suspension”. Both products
have a similar DHA content of 28–30 wt%.
The scope of this LCA is cradle-to-gate and includes (1) sugarcane
cultivation and harvest, (2) processing sugarcane to sugar (with energy
and ethanol as co-products) and (3) conversion of sugar into algae
omega-3 DHA by Corbion. The production of the algae omega-3 DHA
products is located at a Corbion facility in Brazil where current pro-
duction is ongoing. The input data for production used in this study is
based on calculations for full scale production, using techno-economic
models and manufacturing production data from 2019. Manufacturing
of production equipment, buildings, and other capital goods on the
manufacturing site of Corbion are not included in the scope. Due to the
long lifetime of the plant, the contributions are expected to be small.
2.2. Environmental impact categories
The characterization method used was the EF (Environmental foot-
print) 3.0 impact assessment method [20] adapted to SimaPro 9.1. This
is the impact method published for use during transition phase of the
Environmental Footprint initiative [22]. The study covers the six most
relevant impact categories dened by the product environmental foot-
print guidelines category rules (PEFCR) for animal feed [23]: (1) Climate
change (GWP100a based on the Intergovernmental Panel on Climate
Change - IPCC, 2013 [24]), (2) Particulate matter (impact on human
health [25]), (3) Acidication of terrestrial and freshwater (Accumu-
lated Exceedance [26]), (4) Land Use - LANCA model [27], (5) Terres-
trial Eutrophication (Accumulated Exceedance [26]) and (6) Water use -
AWARE model [25].
D. Davis et al.

Algal Research 60 (2021) 102494
3
2.2.1. Biogenic carbon
Sugar used for algae omega-3 DHA production contains biogenic
carbon which is stored in the plant tissue during plant growth and
converted to omega-3 DHA. In this LCA, biogenic carbon does not earn a
credit for carbon uptake because most of the carbon will eventually be
released again as CO
2
or CH
4
after animal or human consumption in less
than a few years. This means that the carbon uptake of the crop during
cultivation is not considered and neither are the CO
2
emissions after
consumption. This approach is consistent with the PEF (Product envi-
ronmental footprint) guidelines [28].
2.2.2. Additional information
As additional information, the water scarcity risk was assessed using
a water risk assessment tool - Aqueduct Water Risk atlas [29].
3. Process description and life cycle inventory
The production of algae omega-3 DHA has ve main steps: (1)
cultivation of sugarcane and transportation to sugar mill, (2) processing
of sugarcane into multiple products: sugar, ethanol, and electricity, (3)
transformation of sugar by algae into DHA via fermentation and (4)
downstream processing of algae into the nal products. The production
system can be seen in Fig. 1. Below is a detailed description of each step.
Primary data for algae omega-3 DHA production and downstream
processing (see the system boundary line of “Corbion” in Fig. 1) was
obtained from the engineering model (Corbion engineering team in
Orindiúva, Brazil). The model is based on kinetic models for the
fermentation validated at commercial scale, thermodynamics and mass
and energy balances for the existing process and equipment. These data
sources are chosen to ensure the consistency and comparability between
both formats of the products. For the main inputs, sugar, electricity and
steam, primary data was collected from the actual suppliers. Background
data were obtained from LCI (Life cycle inventory) databases and
modied if necessary, as described in the next sections. For the agri-
cultural products, the Agri-footprint database (Blonk Consultants, The
Netherlands) was used and partially modied (e.g., sugarcane cultiva-
tion, see Section 3.1) to reect the farm and sugar mill-specic data from
the suppliers. For all other background processes, Ecoinvent v3.6 APOS
was used (Ecoinvent, Switzerland).
The aggregated LCI for algae omega-3 DHA powder and liquid sus-
pension is provided as supplementary information both as *.csv le for
SimaPro 9.1 - Supplementary material.
3.1. Sugarcane production
Production of algae omega-3 DHA uses sugarcane sugar from Brazil
as feedstock. Brazil is the world's leading producer of sugarcane with
41% of the global production in 2017 [5,30]. Sugar is supplied by a
sugar mill located adjacent to the Corbion facility. This mill, located in
the state of S˜
ao Paulo, uses sugarcane from the surrounding elds and
has been in operation for over 20 years.
Sugarcane production consists of land preparation, planting of cut-
tings, application of synthetic and organic fertilizers and plant protec-
tion agents, maintenance of the crop and harvesting. The organic
fertilizers consist of vinasse and lter cake which are byproducts of
sugar and ethanol production. The application of fertilizers to the elds
have become a focus of many studies in attempting to reduce the envi-
ronmental footprint of sugarcane [31,32]. In this region, the sugarcane
elds are primarily rain-fed, with application of the mill by-products
and returned treated water from the mill. This irrigation technique
was shown to increase the crop yields [33].
Harvesting of the sugarcane is mechanized and burning cane resi-
dues is prohibited by the local legislation [33]. The prohibition to burn
cane residue in open air has led to a major decrease in CO
2
emissions,
decrease in particulate matter and improvements in human health.
Currently, the crop residues (tops and leaves) are left in the elds after
harvesting, but there is on-going research to understand the best prac-
tices for the management of residues, aiming at reducing emissions,
mostly N
2
O, from crop residues [32].
The inventory for sugarcane production was modelled by modifying
the Agri-footprint v4 unit process of “Sugar cane, at farm/BR Energy”
with available primary data provided by the supplier, to make the LCI
input site-specic. The dataset modications are summarized in Table 1.
Sugarcane production is not allocated. After harvesting, the entire cane
is transported to the sugar mill. The sugarcane residues left on the eld
are not considered as by-products and the emissions related to the res-
idues left on the eld are included in the LCI.
Direct land use change (DLUC) refers to the change or conversion of
the original land use (forest, grassland, pastureland, etc.) to another land
use, i.e. sugarcane plantations. Depending on the type of conversion that
occurs, DLUC can unlock carbon that is stored in soil and vegetation,
which is released as CO
2
. The total GHG (greenhouse gas) emissions and
removals arising from DLUC over a period of 20 years must be included
in the quantication of GHG emissions. Knowledge of the prior land use
may be demonstrated, for example, by using satellite imaging to identify
and measure historical land use change. With this purpose, GRAS
(Global Risk Assessment Services GmbH
1
) conducted a study for Corbion
based on high-resolution satellite images and enhanced vegetation index
time series analysis over the period between 1999 and 2019, covering
the areas of the sourcing plantations of the sugar mill. The results
showed that, for the sourcing area of the mill, the only signicant land
transformation was from degraded pastureland to sugarcane (48%)
[34]. The GHG emissions from DLUC were calculated based on the
measured transformation areas and types of land transformation [35].
The conversion of degraded pastureland with the lowest biomass carbon
stock to sugarcane cultivation can potentially be a carbon sink,
depending on the soil types and agricultural management practices,
resulting in negative DLUC [33]. In order to make a conservative esti-
mate, a DLUC value of zero kgCO
2
eq/t sugarcane was used.
The second component of land use change is indirect land use change
(iLUC), which considers secondary effects induced by large-scale
expansion. The displacement of existing crops potentially leads to the
expansion of cropland elsewhere. There are different models used to
assess iLUC with no consensus yet on iLUC methodology [36]. In the
case of Brazilian sugarcane, most of the land displacement occurring is
related to conversion of pasture lands for cattle into sugarcane elds
enabled by livestock intensication. Additionally, improvements in
agricultural yield and expansion of sugarcane into regions with higher
potential for agricultural productivity are pointed out as explanations
for the low iLUC impact related to sugarcane in the region of S˜
ao Paulo
[33,36,37]. iLUC is excluded from this study to reect the attributional
nature of this study and also in alignment with the PEF and the PEFCR
guidance [22,23].
Fertilizers and direct emissions from fertilizer application are key
aspects for the LCI of agricultural products. Specic information pro-
vided by the supplier was used to modify the original Agri-footprint
dataset:
•The amount of fertilizers was adjusted based on the total values of
NPK (kg N, P and K/ha) provided by the supplier.
•The emissions from vinasse and lter cake application were calcu-
lated based on supplier information regarding the amounts and
composition, based on Macedo et al. 2018 [38]. The IPCC method-
ology was applied, consistently with the Agri-footprint methodology
[39].
3.2. Sugar mill
After harvest, the cane is transported by truck to the mill and
1
https://www.gras-system.org/.
D. Davis et al.

Algal Research 60 (2021) 102494
4
processed immediately.
In the sugar mill, sugarcane is pressed to extract the juice and a
brous residue remains, known as bagasse. The sugar juice is used to
make sugar and ethanol while bagasse is burned in the CHP (combined
heat and power) plant. The juice is puried to remove the suspended
matter and the resulting lter cake. The claried juice is evaporated and
crystallized until the required sugar purity. Molasses is obtained as a by-
product from the crystallization steps. Both molasses and sugar are in-
puts to ethanol production. Ethanol is produced through yeast fermen-
tation with the main by-product being vinasse [40].
Sugarcane is supplied by farms around the sugar mill, from a distance
<50 km. The products of this sugar mill are sugar, hydrous and anhy-
drous ethanol, electricity, and steam. Other by-products are lter cake
and vinasse, which are applied to the sugarcane elds as fertilizers. The
steam and electricity generated from bagasse in the CHP plant is suf-
cient to run the mill and the excess electricity is sold to Corbion or to the
Brazilian electricity grid. Corbion also purchases steam from the sugar
mill, produced exclusively from bagasse. More information on the sugar
mill by-products can be found in the Multi-functionality section.
All inputs and outputs for the sugar mill, except the emissions to air,
were based on supplier information (condential). Emissions to air from
bagasse combustion were included, based on the Agri-footprint V4
dataset “Sugar, from sugar cane, from sugar production, at plant/BR”.
3.2.1. Multi-functionality
The sugar mill is a multifunctional process, with the output products
Fig. 1. Production system of algae omega-3 DHA (docosahexaenoic acid) considered for the life cycle assessment study.
Table 1
Overview of the modications to the Agri-footprint sugarcane dataset, to reect
the supplier specic information. DLUC – direct land use change.
Parameter Modication Reference
DLUC Zero CO
2
emissions from
DLUC are considered based
on the results of satellite
imagery between 1999 and
2019
(GRAS, Global Risk
Assessment Services,
2019) [34]
Amount of fertilizers
and direct emissions
from fertilizer
application
Described in the text Supplier information
(condential)
Irrigation No water input from natural
sources. Only efuent from
the sugar mill, lter cake
and vinasse are applied
Supplier information
Harvesting All mechanized harvest Supplier information
Sugarcane yield Average yield in Brazil in
2014–2018: 73.8 ton/ha/
year (range: 70.6–75.2 ton/
ha/year)
(FAO, 2020 [30])
Emissions from crop
residues
Direct N
2
O emissions from
crop residues: 0.8 kg N
2
O/
ha
Supplier information &
N
2
O emissions calculated
according to Agri-footprint
v5
D. Davis et al.

Algal Research 60 (2021) 102494
5
shown in Fig. 2: sugar, ethanol, electricity, steam, lter cake and
vinasse. From all these outputs, only sugar, ethanol, and electricity are
sold directly to the market, including Corbion. Filter cake and vinasse
are returned to the sugarcane cultivation elds around the mill and the
emissions from their application are included in the sugarcane dataset,
as described in the previous paragraph. Additionally, it is assumed that
the benet of this practice in terms, for example, of improved yield is
already reected in the sugarcane dataset. Corbion also uses steam from
the sugar mill because both sites are co-located. Steam is not sold to any
other external parties and is not considered as a by-product in the data
sources provided by the mill. If steam was not used by Corbion, it would
either be wasted or used to generate additional electricity, with very
limited benet. For these reasons, no burdens are allocated to steam.
Therefore, relevant co-products of the sugar mill in terms of alloca-
tion are:
−Sugar
−Ethanol
−Electricity
As shown in Fig. 2, the production of these products is interlinked
and cannot be sub-divided. In attributional modelling, when sub-
division is not possible, allocation between co-products should be
applied [41]. Economic allocation is applied as default based on the
PEFCR for animal feed [23]. The economic allocation factors, based on
average prices in 2016–2019, are 45%, 45% and 9% for sugar, ethanol
and electricity, respectively. In the sensitivity analysis (see Section 4.3)
energy allocation is applied, which results in slightly higher allocation
factor for sugar and lower allocation factor for electricity, i.e. 48%, 45%
and 7%.
3.3. Algae production and downstream processing
At the Corbion facility, algae of the genus Schizochytrium sp. are
cultivated in large-scale stainless-steel agitated aerobic fermenters.
Schizochytrium sp. belongs to the family, Thraustochytriaceae, and is a
member of the Chromista kingdom. Historically, it has been grouped
with algae (CEN denition EN 17399 2020) and for this reason the term
algae is commonly used.
Sugar syrup is prepared and mixed with other nutrients, including
nitrogen sources, to makeup the fermentation medium for the algae
growth and oil production. Oil production occurs intracellularly. The
main co-product of the fermentation is CO
2
which is released to the
atmosphere.
The algae broth is harvested and dried using steam from bagasse. The
dry algae biomass is a free-owing dry powder with oil high in DHA
encapsulated within the algae biomass. The water removed in the drying
steps is condensed and reused in the process. Algae omega-3 DHA liquid
suspension is produced by mixing algae biomass with vegetable oil. The
liquid suspension format allows for several benets including more
efcient bulk transportation and, in some cases, easier and higher levels
of incorporation into feed.
As described earlier, the sugar, electricity and steam consumed at the
Corbion facility are generated from sugarcane and sourced from the
nearby sugar mill. The wastewater efuent is also returned to the sugar
mill where it is treated and applied to the cane plantations. Additional
background data, for example for the fermentation nutrients and other
auxiliary chemicals, is included using suitable Ecoinvent 3.6 APOS
datasets. The production of vegetable oil is based on the Agri-footprint
database, including DLUC based on crop and sourcing country [42].
For the default case, consistently with the sugar mill model, economic
allocation is used.
The LCI for algae omega-3 DHA production is based on the con-
dential process and for this reason detailed input/output data cannot be
disclosed.
4. Results
4.1. Life Cycle Impact Assessment results
The Life Cycle Impact Assessment (LCIA) results for the sixmost
relevant impact categories recommended by the feed PEFCR are shown
in Table 2. The impacts of algae omega-3 DHA powder and liquid sus-
pension are quite similar. Algae omega-3 DHA liquid suspension has a
slightly larger impact than algae omega-3 DHA powder for the climate
change, acidication and eutrophication impact categories, and a
slightly lower impact is observed for particulate matter, land use and
water use categories. Fig. 3 shows the process contribution to each of
these impact categories which is described in more detail in the next
paragraphs.
Fig. 2. Process scheme for the mill with inputs in yellow, products internally recycled in orange and outputs in purple. (For interpretation of the references to colour
in this gure legend, the reader is referred to the web version of this article.)
D. Davis et al.

Algal Research 60 (2021) 102494
6
4.2. Process contributions
4.2.1. Climate change
The GHG emissions related to sugarcane cultivation contribute 65%
to the total climate change impact of algae omega-3 DHA powder.
Sugarcane is the input used for both the feedstock and energy produc-
tion for algae omega-3 DHA production. Most of these emissions (57%)
were due to direct N
2
O and CO
2
emissions resulting from crop residues,
synthetic fertilizer application and the application of the by-products,
vinasse and lter cake. The remaining emissions are attributed to
diesel use for agricultural machinery (20%), fertilizer production (13%),
and production of crop protection products(5%). Based on the study
conducted by GRAS, the CO
2
emissions related to direct land use change
were determined to be 0.
The second largest contributor to the GHG emissions of algae omega-
3 DHA powder is the sugar mill which represents 22% of the product
emissions. These emissions are driven by the transport of the cane by
truck to the mill (49%) and the N
2
O and CH
4
emissions from bagasse
combustion (50%). The small amount of chemicals used in sugarcane
processing represent the remainder of the GHG emissions in the sugar
mill.
For algae omega-3 DHA powder production, some additional nutri-
ents and chemicals are used which represent 11% of the total GHG
emissions. Direct process emissions are mostly biogenic CO
2
formed in
Table 2
Results for the relevant impact assessment categories for 1 kg of omega-3 in
algae omega-3 DHA (Docosahexaenoic acid) powder and liquid suspension,
using Environmental footprint (EF) 3.0 method.
Impact category Unit Algae omega-3
DHA powder
Algae omega-3 DHA
liquid suspension
Climate change kg CO
2
eq 4.12 4.70
Particulate matter Disease inc. 1.35E-06 1.23E-06
Acidication mol H+eq 5.65E-02 6.91E-02
Eutrophication,
terrestrial
mol N eq 1.82E-01 2.50E-01
Land use Pt (soil index
quality)
2090 1802
Water use m3 depriv. 0.926 0.757
Fig. 3. Contribution of different process stages for (a) algae omega-3 DHA (Docosahexaenoic acid) powder and (b) algae omega-3 DHA liquid suspension.
D. Davis et al.

Algal Research 60 (2021) 102494
7
the fermentation, which is not considered for GWP (global warming
potential).
Regarding algae omega-3 DHA liquid suspension, the production and
transport of vegetable oil contributes 31% to the impact on GHG emis-
sions. Cultivation of vegetable oil is the most relevant step, again driven
by direct N
2
O and CO
2
emissions from fertilizer application, fertilizer
production, energy used for agricultural machinery and to a smaller
extent to DLUC. According to FAO statistics, over the last 20 years, there
was about 0.1% land transformation from perennial crop to the oil crop
resulting in some GHG emissions from DLUC [42].
4.2.2. Particulate matter
The impact category of particulate matter refers to the emission of
ne particles and its precursors (e.g., NO
x
and SO
2
) and the adverse
potential impact on human health.
Fig. 3 shows that most of the emissions related to algae omega-3 DHA
powder come from the sugar mill (76%). Particulates and sulfur dioxide
emissions from bagasse burning contribute the most for these emissions.
Sugarcane cultivation is the second largest contributor (21%) with
ammonia emissions from fertilizer having the highest inuence. Like-
wise, for algae omega-3 DHA liquid suspension, ammonia emissions
from fertilizer application in oil crop cultivation contribute to particu-
late matter.
4.2.3. Acidication
Acidication is caused by the emission of the compounds NH
3,
NO
2
and SO
x
which are the precursors to acid rain. When these compounds
are deposited to the soil or freshwater, the acidity can change. As most
ecosystems operate at an optimal acidity level, changes in the acidity
can have drastic repercussions on ecosystem functionality and individ-
ual species.
According to Fig. 3, the largest impact to acidication potential is the
cultivation of the sugarcane. Emissions of ammonia and nitrous oxide
due to the application of fertilizer, both synthetic and organic, are the
largest contributors, representing 65% of the acidication impact of
algae omega-3 DHA powder. The main reason for acidication at the
sugar mill is the sulfur dioxide emitted from the combustion of the
bagasse. The additional emissions that impact the acidication potential
can be linked to transportation and production of fertilizers and chem-
icals used throughout the product system.
For algae omega-3 DHA liquid suspension additional impacts are
related to vegetable oil crop production and emissions from fertilizer
application.
4.2.4. Eutrophication
Terrestrial eutrophication refers to the deposition of aerial nitrogen
compounds such as NO
x
and NH
3
to terrestrial environments resulting in
increased nutrient availability. As seen in Fig. 3, sugarcane cultivation
accounts for 89% of the impacts in terrestrial eutrophication of algae
omega-3 DHA powder production. These emissions are related to ni-
trogen and phosphate fertilizer application. Again, for algae omega-3
DHA liquid suspension, additional impacts are related to vegetable oil
crop production and emissions from fertilizer application.
The remaining impacts contribution to eutrophication are associated
with the emission of NO
x
throughout the product system. Most of these
remaining emissions are credited to the combustion of fuel during
transportation.
4.2.5. Land use and water use
Land use is an essential topic when discussing food systems and
development of alternative feed ingredients. The land use impact of
algae omega-3 DHA liquid suspension product is dominated by sugar-
cane cultivation (87%) and vegetable oil production (8%), see Fig. 3.
The water scarcity for each of the process steps can be seen in Fig. 3.
The water use impact from sugarcane cultivation is not signicant
~6–7% of total water use impact because sugarcane is not an irrigated
crop and in the region of Orindiúva, the water stress level is low (see
Fig. 4). More than 80% of the water use impact is contributed to the
water used in the algae production. The water consumed at the algae
production facility is withdrawn from the local river and in the same
region as the sugarcane plantations, therefore from a low water stress
level. The water consumption is mainly due to the evaporation in the
cooling towers. Water used in the fermentation is largely recycled within
the process and a smaller amount is returned to the sugar mill where it is
treated and applied in the sugarcane elds.
4.3. Sensitivity analysis
Sensitivity analyses are performed to estimate the effect of the
choices made regarding methods and data assumptions on the outcome
of a study. The variables considered in the sensitivity analysis were
determined based on key uncertain assumptions and on the relevant
contributions to the LCIA results. The variables include key aspects
related to algae production, the sugar mill, and sugarcane cultivation:
1) Sugar conversion yield to omega-3 DHA
2) Allocation of by-products using energy allocation factor
3) Amount of electricity exported from the sugar mill
4) Sugarcane yield
The algae production data has low uncertainty because of the high
maturity level of the process and the accuracy of the engineering
models. Sugar conversion yield to omega-3 DHA is included in the
sensitivity analysis since it can potentially be further improved as an
outcome of the R&D efforts on Schizochytrium strain development.
Considering an increase of 5–20% in the conversion yield of sugar to
omega-3 DHA, a corresponding 5–20% decrease in the environmental
footprint of algae omega-3 DHA is expected. The yield increase reduces
the environmental impact for all categories because the sensitivity
analysis assumes that all other process parameters remain unchanged.
The choice of the allocation method is an important assumption in
any LCA and the ISO 14044 requires that a sensitivity analysis is con-
ducted to illustrate the consequences of the selected approach [18]. The
default results in this study are based on economic allocation of the by-
products and, as part of the sensitivity analysis, the results were also
calculated using energy allocation. This alternative calculation provides
insights on the impact of the modelling choice for allocation and showed
that energy allocation results in lower environmental impacts. The dif-
ference between both allocation approaches is less than 3% for algae
omega-3 DHA powder and less than 9% for algae omega-3 DHA liquid
suspension, for all impact categories investigated. Algae omega-3 DHA
liquid suspension is slightly more sensitive to the allocation approach
because of the vegetable oil.
The amount of electricity exported by the sugar mill is a potential
measure to reduce the environmental impact of the sugar mill products.
This can be achieved by increasing the energy efciency of the sugar and
ethanol production, by implementing energy intensication measures or
by improving the efciency of the CHP. In the sensitivity analysis, higher
values of electricity export were considered, up to 120 kWh/ton sug-
arcane (about 200% larger than the default case). The higher value of
electricity export considered is achieved for modern mills with high
pressure co-generation [43]. For all impact categories, the emissions
decrease with the amount of electricity exported by the mill. The effect
of increasing the electricity export is the largest for land use and par-
ticulate matter and almost negligible for water (See Fig. 5). Even if a
maximum electricity export assumed, the GHG emissions associated
with algae omega-3 DHA liquid suspension decreases by less than 7%.
Crop yield can have a signicant impact on the overall emissions
from crops and is highly dependent on the farming practices and
weather conditions, such as rainfall. As mentioned in Section 3.1, the
sugarcane yield was based on the most recent 4-year national average.
Even though this average is aligned with the information provided by
D. Davis et al.

Algal Research 60 (2021) 102494
8
the supplier for 2019, it was also indicated that the yields varied within
the plantation areas. Fig. 6 shows the change in the impacts of algae
omega-3 DHA liquid suspension resulting from the variation of the
sugarcane yield. The range of the sugarcane yield considered (60–90
ton/ha/year, baseline yield 73.8 ton/ha/year) covers the variation in
the last 5 years (Table 1) and the expected yield improvement in the next
5–10 years. A zero change in input corresponds to the default case, a
positive change in input corresponds to an increase in yield. The yield
increase has a positive impact in reducing all environmental categories.
Impact categories that are most sensitive to sugarcane yield are: land
use, eutrophication (marine and terrestrial) and climate change.
From the sensitivity analysis, it is concluded that the sugarcane yield
and the conversion yield of sugar to omega-3 DHA are the most sensitive
parameters. The allocation choice and amount of electricity export have
a lower impact on the results.
5. Discussion
5.1. Comparison with sh oil omega-3
The production of sh oil relies on the capture of small pelagic ma-
rine sh and sh trimmings from sh processing facilities. The wild sh
are caught and transported to shmeal plants and then processed into
sh meal and sh oil (FMFO).
The comparison with sh oil was based on data available in the
databases Agri-footprint v5 economic (arithmetic average of the data-
sets available for the seven different countries “Fish oil, at processing/
countries Economic”, countries =Denmark, Chile, Peru, Norway, Ger-
many, Great Britain, The Netherlands) and Ecoinvent 3.6 (Fish oil, from
anchovy {GLO}| market for sh oil | APOS). The Agri-footprint datasets
were modied to reect the average market prices of sh oil and sh
meal in the period of 2015–2019 [44]. Based on these prices, the allo-
cation factors are 22% and 78% for sh oil and sh meal, respectively.
For the Ecoinvent economic allocation, the original factors were used:
31% and 69% for sh oil and sh meal, respectively. For the sh oil
Fig. 4. Water stress map of the area of Orindiúva (World Resource Institutue, WRI – [29]).
Fig. 5. Effect of electricity export on the environmental impact of algae omega-3 DHA (Docosahexaenoic acid) liquid suspension. x-axis: percent change in the input
parameter electricity export (kwh/ton sugarcane) and y-axis: relative change in the environmental impact. The 0% change for electricity export corresponds to the
default case.
D. Davis et al.

Algal Research 60 (2021) 102494
9
system, these datasets consider the entire production process, including
capture of wild sh, use of offal and processing of sh into sh meal and
sh oil.
The omega-3 content of sh oil varies based on species of sh, time of
year, diet, and many other factors. Fish oil sourced from the North
Atlantic generally has a lower omega-3 content (maximum 16%) while
sh oil from anchovies in the South Pacic can be much higher (up to
24%). The omega-3 content of sh oil used for comparison was 19%, this
represents the higher end of the spectrum of omega-3 content of sh oil
that is being used in aquafeed and is representative of the global
average.
The reference ow for the comparison of sh oil with algae omega-3
DHA is 1 kg of omega-3 fatty acids, assuming 30% of long chain omega-3
fatty acids for the algae omega-3 DHA products and 19% of long chain
omega-3 fatty acid for sh oil. Fig. 7 shows that algae omega-3 DHA has
lower impact than sh oil in the climate change impact category. The
impacts of sh oil production are mostly related to the diesel energy
consumption for the shing vessels (diesel engine and generating of
power for cooling), and the electricity and heat to operate the FMFO
plant. Both the type and efciency of the shing vessel, and the ef-
ciency and source of energy used at the FMFO plant, have a signicant
inuence on the results for sh oil. The most in depth LCA study of
shmeal plants to date highlighted the processing phase as the most
impactful stage of sh oil production [45].
Fig. 6. Effect of sugarcane yield on the environmental impact of algae omega-3 DHA (Docosahexaenoic acid) liquid suspension. x-axis: percent change in the input
parameter sugarcane yield and y-axis: relative change in the environmental impact. The average slope of the curves is the relative sensitivity parameter.
Fig. 7. Results for the environmental footprint of algae omega-3 DHA (Docosahexaenoic acid) and omega-3 from sh oil. The sh oil datasets are based on Ecoinvent
3.6 “Fish oil, from anchovy {GLO}| market for sh oil | APOS” and the average of the Agri-footprint v5 datasets (Chile, Perú, Norway, Denmark, UK, Netherlands and
Germany) The different products are compared per 1 kg of omega-3s and the maximum value is set to 100%.
D. Davis et al.

Algal Research 60 (2021) 102494
10
The algae omega-3 DHA products have relatively higher impact on
particulate matter than sh oils, which is related with the energy pro-
duction. Land use for sh oil omega-3 is low – instead of agricultural
land, the sheries use marine resources which are not dened by the
LANCA model. Regarding water use, the results from sh oil differ
substantially between both databases and conclusions cannot be drawn.
Water consumption in the Ecoinvent dataset is related to water use in the
FMFO plant. For all other impact categories, results are driven by the
signicant fossil fuel consumption to produce sh oil and diesel for the
sh capture.
5.2. Comparison with other algae systems
The LCAs available in the literature from heterotrophic algae are
sparse and focus on biodiesel [13–15] and food or feed ingredient pro-
duction [16]. These studies are based on lab, pilot or literature data, and
include additional processing steps (e.g., oil extraction, biodiesel pro-
duction, lyophilization, etc.) which are not required for algae omega-3
DHA production and have different functional units. For these reasons,
the results cannot directly be compared.
Studies on biodiesel production show that the GWP of the hetero-
trophic and autotrophic production systems are within the same range,
depending on the scenarios evaluated [14,15]. In terms of land use, the
autotrophic systems have a better performance, and for water, the re-
sults depend on the feedstock. [14]. The study of Smetana et al. (2017)
concluded that the heterotrophic production system results in the lowest
environmental impact. The use of renewable energy for algae omega-3
DHA production is one of the main factors for achieving a low envi-
ronmental footprint [16]. For the heterotrophic systems, the choice of
feedstock and sourcing geography is very impactful [13] and substan-
tiated in this study by the high contribution of the sugarcane cultivation
to the LCA.
5.3. Limitations of the study and recommendations
Heterotrophic omega-3 DHA microalgae have been produced at in-
dustrial scale for several years at Corbion. The algae production facility
also produces other algae products currently. Measured plant data is not
directly used for the LCI because it is not representative of the algae
facility producing solely omega-3 DHA at full capacity. For this reason,
the LCI was based on calculated values based on heat and mass balances
combining both historical data and process design knowledge. It is
recommended that the LCA is updated with measured plant data for the
products when this is available and representative. However, no sub-
stantial change is expected because the impact of algae omega-3 DHA
manufacturing was shown to be low in most impact categories. The LCA
results are representative only for heterotrophic algae production using
this commercial process and integrated with the sugar mill, both for
feedstock and energy supply.
Although this study followed the guidelines of the PEFCR measure-
ment for the most relevant impact categories, the study does not quan-
tify all environmental impacts. For example, the depletion of marine
resources is not included in any LCA impact categories and the impact
caused by unsustainable sheries cannot be immediately assessed.
Additionally, one of the most important impact categories to be
considered in the future will be biodiversity loss. Agriculture and shing
are major contributors to biodiversity loss and should be considered
when discussing impacts of these industries [46]. Measuring biodiver-
sity and properly dening its worth has been a challenge and even more
challenging when trying to compare between terrestrial and marine
aquatic environments. There are currently methodologies in develop-
ment that may make this possible in the near future and should be
considered when making decisions about the environmental impact of
both of these systems [6,47,48].
6. Conclusions
With the global supply of forage sh at a plateau, there is growing
pressure to reduce dependence on the use of sh oil in aquaculture and
livestock feeds, as well as pet food. The use of algae omega-3s as a
scalable and alternative to sh oil is growing. This LCA provides a
detailed description of the environmental impacts of two commercial
omega-3 DHA products produced from heterotrophically grown
microalgae.
For both products, algae omega-3 DHA powder and liquid suspen-
sion, sugarcane cultivation has the largest contribution for most of the
categories, except for particulate matter and water use. The last two
impacts are, respectively, dominated by the sugar mill operation and
algae production. The sensitivity analysis showed that the environ-
mental impacts are highly sensitive to sugarcane yield. In contrast, the
allocation methods and amount of electricity export as a credit, had a
limited inuence on the results.
The comparison with sh oil was carried out using data available in
LCI databases. The large range of results obtained for the different
datasets of sh oil, and the variability of the omega-3 content of sh oil,
made the comparison challenging. Despite the ranges observed in the
impacts of sh oil from different LCI databases, both algae omega-3 DHA
products offer lower impacts for climate change and acidication
compared to sh oil. On the other hand, the particulate matter impact is
higher for algae omega-3 DHA products due to the air emissions from
bagasse combustion. In addition, the land use impact for sh oil appears
to be lower – because marine sheries are not dened by the current LCA
land use impact model.
In conclusion, the use of algae omega-3 DHA in feed can contribute
positively to maintaining or improving omega-3 levels in feed, reduce
pressure on marine resources, and play a role in improving the carbon
footprint of feed formulations.
Looking forward, the production of heterotrophic algae omega-3
DHA at commercial scale still has optimization potential, for instance
through the development of higher sugar to omega-3 DHA yield algae
strains and intensication of the production process to increase energy
and water efciencies. Additionally, reductions can be achieved through
engagement with the sugar mill on potential improvements both for
sugarcane cultivation and the sugar mill. Likewise, sourcing low impact
vegetable oil can improve the footprint of the algae omega-3 DHA liquid
suspension format.
CRediT authorship contribution statement
All authors contributed equally to the conception of the manuscript
and to its drafting, and approval. Dillon North-Davis and Ana Mor˜
ao
were the main contributors to the acquisition, analysis and interpreta-
tion of the data.
Statement of informed consent, human/animal rights
No conicts, informed consent, or human or animal rights are
applicable to this study.
Declaration of competing interest
Ana Morao reports a relationship with Corbion BV that includes:
employment. Jill Kauffman Johnson reports a relationship with Corbion
CV that includes: consulting or advisory.
Acknowledgements
The authors want to thank the Corbion operations team in Brazil for
collecting the manufacturing data, to Carlos Veloso for the collaboration
with Corbion suppliers and to all Corbion colleagues who contributed
with their input by reviewing the manuscript.
D. Davis et al.

Algal Research 60 (2021) 102494
11
This research did not receive any specic grant from funding
agencies in the public, commercial, or not-for-prot sectors.
Appendix A. Supplementary data
The LCI datasets can only be accessed with SimaPro 9.1. Importing
the data into other LCA softwares could lead to errors due to incom-
patible elementary ows. Before any reuses of the dataset, please care-
fully check the goal and scope denition, the impact categories covered,
and the choices made in the inventory modelling. The information can
be found in Chapter 2 of the article. These LCI datasets do not offer
information beyond the dened scope of the study. Supplementary data
to this article can be found online at https://doi.org/10.1016/j.algal.20
21.102494.
References
[1] B. Narayan, K. Miyashita, M. Hosakawa, Physiological effects of eicosapentaenoic
acid (EPA) and docosahexaenoic acid (DHA) - a review, Food Rev. Int. 22 (2006)
291–307, https://doi.org/10.1080/87559120600694622.
[2] Annual Data - Infogram, (n.d.). https://infogram.com/annual-data-1ggk26e86
njkpn0 (accessed March 7, 2021).
[3] H.A. Hamilton, R. Newton, N.A. Auchterlonie, D.B. Müller, Systems approach to
quantify the global omega-3 fatty acid cycle, Nat. Food. 1 (2020) 59–62, https://
doi.org/10.1038/s43016-019-0006-0.
[4] N. Auchterlonie, The continuing importance of shmeal and sh oil, in: Aquafeeds
Presented at the Aquafarm Conference, Aquafarm Conference, Pordenone, Italy,
2018.
[5] FAO, The State of World Fisheries and Aquaculture 2018-Meeting the Sustainable
Development Goals, 2018.
[6] J. Alder, B. Campbell, V. Karpouzi, K. Kaschner, D. Pauly, Forage sh: from
ecosystems to markets, Annu, Rev. Environ. Resour. 33 (2008) 153–166, https://
doi.org/10.1146/annurev.environ.33.020807.143204.
[7] T.S. Aas, T. Ytrestøyl, T. Åsgård, Utilization of feed resources in the production of
Atlantic salmon (Salmo salar), in: Norway: An Update for 2016, Aquac. Reports,
2019, https://doi.org/10.1016/j.aqrep.2019.100216.
[8] H. Cottrell, R.S. Blanchard, J.L. Halpern, B.S. Metian, M. Froehlich, Global
adoption of novel aquaculture feeds could substantially reduce forage sh demand
by 2030, Nat. Food. (2020), https://doi.org/10.1038/s43016-020-0078-x in press.
[9] M. Bou, G.M. Berge, G. Baeverfjord, T. Sigholt, T.K. Ostbye, O.H. Romarheim,
B. Hatlen, R. Leeuwis, C. Venegas, B. Ruyter, Requirements of n-3 very long-chain
PUFA in Atlantic salmon (Salmo salar L): effects of different dietary levels of EPA
and DHA on sh performance and tissue composition and integrity, Br, J. Nutr. 117
(2017) 30–47, https://doi.org/10.1017/S0007114516004396.
[10] B. De Roos, A.A. Sneddon, M. Sprague, G.W. Horgan, I.A. Brouwer, The potential
impact of compositional changes in farmed sh on its health-giving properties: is it
time to reconsider current dietary recommendations? Public Health Nutr. 20
(2017) 2042–2049, https://doi.org/10.1017/S1368980017000696.
[11] D.R. Tocher, M.B. Betancor, M. Sprague, R.E. Olsen, J.A. Napier, Omega-3 long-
chain polyunsaturated fatty acids, EPA and DHA: bridging the gap between supply
and demand, Nutrients. 11 (2019) 1–20, https://doi.org/10.3390/nu11010089.
[12] R.J. Winwood, Recent developments in the commercial production of DHA and
EPA rich oils from micro-algae, OCL - oilseeds fats, crop, Lipids. 20 (2013), https://
doi.org/10.1051/ocl/2013030.
[13] S. Parsons, C.J. Chuck, M.C. Mcmanus, Microbial Lipids: Progress in Life Cycle
Assessment (LCA) and Future Outlook of Heterotrophic Algae and Yeast-derived
Oils, 2018, https://doi.org/10.1016/j.jclepro.2017.10.014.
[14] N.D. Oreld, R.B. Levine, G.A. Keoleian, S.A. Miller, P.E. Savage, Growing Algae
for Biodiesel on Direct Sunlight or Sugars: A Comparative Life Cycle Assessment,
2015, https://doi.org/10.1021/sc5004117.
[15] A. Azari, A.R. Noorpoor, O. Bozorg-Haddad, Carbon footprint analyses of
microalgae cultivation systems under autotrophic and heterotrophic conditions,
Int. J. Environ. Sci. Technol. 16 (2019) 6671–6684, https://doi.org/10.1007/
s13762-018-2072-5.
[16] S. Smetana, M. Sandmann, S. Rohn, D. Pleissner, V. Heinz, Autotrophic and
heterotrophic microalgae and cyanobacteria cultivation for food and feed: life
cycle assessment, Bioresour, Technol. 245 (2017) 162–170, https://doi.org/
10.1016/j.biortech.2017.08.113.
[17] ISO 14040, Environmental Management — Life Cycle Assessment — Principles and
Framework 2016, ISO 14040, 2016.
[18] ISO 14044, Environmental Management — Life Cycle Assessment — Requirements
and Guidelines 2016, 2016.
[19] J.A. Emery, F. Norambuena, J. Trushenski, G.M. Turchini, Uncoupling EPA and
DHA in sh nutrition: dietary demand is limited in Atlantic salmon and effectively
met by DHA alone, Lipids. 51 (2016) 399–412, https://doi.org/10.1007/s11745-
016-4136-y.
[20] K. Katerina, G.M. Berge, M. Turid, K. Aleksei, B. Grete, Y. Trine, C. Mats, S. John,
R. Bente, Microalgal Schizochytrium limacinum biomass improves growth and let
quality when used long-term as a replacement for sh oil, in modern salmon diets,
Front, Mar. Sci. 7 (2020) 57, https://doi.org/10.3389/fmars.2020.00057.
[21] B. Ruyter, Latest research on omega-3 and sh performance, in: AquaForum 11-
12th Novemb. Puerto Montt, 2020.
[22] European Comission, The Environmental Footprint Transition Phase, 2020.
[23] E. Commission, PEFCR Feed for Food Producing Animals, 2018.
[24] IPCC, No TitClimate Change 2013: The Physical Science Basis. Contribution of
Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Changele, Cambridge, United Kingdom and New York, NY, USA, 2013,
https://doi.org/10.1017/CBO9781107415324.
[25] UNEP, Global Guidance for Life Cycle Impact Assessment Indicators. Volume,
2016.
[26] J. Sepp¨
al¨
a, M. Posch, M. Johansson, J.P. Hettelingh, Country-dependent
characterisation factors for acidication and terrestrial eutrophication based on
accumulated exceedance as an impact category indicator, in, Int. J. Life Cycle
Assess., Springer (2006) 403–416, https://doi.org/10.1065/lca2005.06.215.
[27] K. Beck, T. Bos, U. Wittstock, B. Baitz, M. Fischer, M. Sedlbauer, LANCA Land Use
Indicator Value Calculation in Life Cycle Assessment – Method Report, 2010.
[28] R. Zampori, L. Pant, Suggestions for updating the product environmental footprint
(PEF) method, JRC Tech. Reports (2019).
[29] WRI, Aqueduct Water Risk Atlas, 2019.
[30] FAO, Brazil - Sugarcane Prole, 2020.
[31] V.P. Vargas, H. Cantarella, A.A. Martins, J.R. Soares, J.B. do Carmo, C.A.
de Andrade, Sugarcane crop residue increases N2O and CO2 emissions under high
soil moisture conditions, Sugar Tech. 16 (2014) 174–179, https://doi.org/
10.1007/s12355-013-0271-4.
[32] J. Fonseca da Silva, A. Moreira de Carvalho, T.A. Rein, T. Rodrigues Coser, W.
Quadros Ribeiro, D. Lino Vieira, D.A. Coomes, Nitrous oxide emissions from
sugarcane elds in the Brazilian Cerrado, Agric, Ecosyst. Environ. 246 (2017)
55–65, https://doi.org/10.1016/j.agee.2017.05.019.
[33] R. de O. Bordonal, J.L.N. Carvalho, R. Lal, E.B. de Figueiredo, B.G. de Oliveira, N.
La Scala, Sustainability of sugarcane production in Brazil. A review, Agron,
Sustain. Dev. 38 (2018), https://doi.org/10.1007/s13593-018-0490-x.
[34] GRAS, Global Risk Assessment Service, Sustainability Risk Assessment of
Sugarcane Sourcing Area in Brazil - Condential, 2019.
[35] Ofcial Journal of the European Union, Decision No 529/2013/EU of the European
Parliament and of the Council of 21 May 2013 on accounting rules on greenhouse
gas emissions and removals resulting from activities relating to land use, land-use
change and forestry and on information concerning ac, 2013.
[36] D. Zilberman, Indirect land use change: much ado about (almost) nothing, GCB
Bioenergy 9 (2017) 485–488, https://doi.org/10.1111/gcbb.12368.
[37] J.A. Verstegen, F. van der Hilst, G. Woltjer, D. Karssenberg, S.M. de Jong, A.P.
C. Faaij, What can and can’t we say about indirect land-use change in Brazil using
an integrated economic - land-use change model? GCB Bioenergy 8 (2016)
561–578, https://doi.org/10.1111/gcbb.12270.
[38] I.C. Macedo, J.E.A. Seabra, J.E.A.R. Silva, Green house gases emissions in the
production and use of ethanol from sugarcane in Brazil: the 2005/2006 averages
and a prediction for 2020, Biomass Bioenergy 32 (2008) 582–595, https://doi.org/
10.1016/j.biombioe.2007.12.006.
[39] P. van Paassen, M. Braconi, N. Kuling, L. Durlinger, B. Gual, Agri-footprint 5 (0)
(2019).
[40] M.O. de Souza Dias, R. Maciel Filho, P.E. Mantelatto, O. Cavalett, C.E.V. Rossell,
A. Bonomi, M.R.L.V. Leal, Sugarcane processing for ethanol and sugar in Brazil,
Environ, Dev. 15 (2015) 35–51, https://doi.org/10.1016/j.envdev.2015.03.004.
[41] International Reference Life Cycle Data System (ILCD) Handbook - General Guide
for Life Cycle Assessment - Detailed Guidance, Luxembourg, 2010. https://eplca.jr
c.ec.europa.eu/uploads/ILCD-Handbook-General-guide-for-LCA-DETAILED-GUID
ANCE-12March2010-ISBN-n-v1.0-EN.pdf (accessed March 7, 2021).
[42] Blonk Consultants, Direct Land Use Change Assessment Tool. https://tools.
blonkconsultants.nl/tool/luc/, 2018.
[43] E. Birru, C. Erlich, A. Martin, Energy performance comparisons and enhancements
in the sugar cane industry, Biomass Convers, Biorenery. 9 (2019) 267–282,
https://doi.org/10.1007/s13399-018-0349-z.
[44] Fish | OECD-FAO agricultural outlook 2020–2029 | OECD iLibrary, (n.d.). https://
www.oecd-ilibrary.org/sites/4dd9b3d0-en/index.html?itemId=/content/compo
nent/4dd9b3d0-en (accessed March 8, 2021).
[45] P. Fr´
eon, H. Durand, A. Avadí, S. Huaranca, R. Orozco Moreyra, Life cycle
assessment of three Peruvian shmeal plants: toward a cleaner production,
J. Clean. Prod. 145 (2017) 50–63, https://doi.org/10.1016/j.jclepro.2017.01.036.
[46] N. Dudley, S. Alexander, Agriculture and biodiversity: a review, Biodiversity. 18
(2017) 45–49, https://doi.org/10.1080/14888386.2017.1351892.
[47] V. Bach, A. H´
elias, A. Wojciechowski, H. Bosch, M. Binder, M. Finkbeiner,
Development of an approach to assess overshing in Europe based on the
ecological scarcity method, 12th Int, in: Conf. Life Cycle Assess. Food 2020 (LCA
Food 2020) 2020, 2020, pp. 181–185.
[48] A.D.M. Smith, C.J. Brown, C.M. Bulman, E.A. Fulton, P. Johnson, I.C. Kaplan,
H. Lozano-Montes, S. Mackinson, M. Marzloff, L.J. Shannon, Y.J. Shin, J. Tam,
Impacts of shing low-trophic level species on marine ecosystems, Science (80-. )
333 (2011) 1147–1150, https://doi.org/10.1126/science.1209395.
D. Davis et al.