Content uploaded by Philippe Loubet
Author content
All content in this area was uploaded by Philippe Loubet on Sep 09, 2021
Content may be subject to copyright.
Life cycle inventory of plastics losses from seafood supply chains:
Methodology and application to French fish products
Philippe Loubet ⁎,Julien Couturier, Rachel Horta Arduin, Guido Sonnemann
Université de Bordeaux, CNRS, Bordeaux INP, ISM, UMR5255, F-33400 Talence, France
HIGHLIGHTS
•Amethodtoincludemicroand
macroplastic losses in LCI of seafood
products is proposed.
•Fishing gear, marine coating, tire abra-
sion, pellets and mismanaged waste
are considered.
•Fourteen fish products from the French
Agribalyse database are analysed.
•Plastic losses range from 75 mg to
4345 mg per kg of fish at the consumer.
•Main plastic losses come from lost fish-
ing gears and tire abrasion.
GRAPHICAL ABSTRACT
abstractarticle info
Article history:
Received 15 July 2021
Received in revised form 19 August 2021
Accepted 31 August 2021
Available online 4 September 2021
Editor: Damià Barceló
Plastic debris into theenvironment is a growing threat for the ecosystems and human health.The seafood sector
is particularly concerned because it generates plastic losses and can be endangered by plastic contamination. Life
cycle assessment (LCA)does not properly considerplastic losses and related impacts,which is a problem in order
to find relevant mitigation strategies without burden shifting.
This work proposes a methodology for quantifying flows of plastics fromthe life cycle of the seafood products to
the environment. It is based on loss rate and final release rate considering a pre-fate approach as proposed by the
Plastic L eak Project. They a re definedfor 5 types of micro and macro plastic losses: lost fishing gears, marinecoat-
ings, plastic pellets, tire abrasion and plastic mismanaged at the end-of-life. The methodology is validated with a
case study applied to French fish products for which relevant data are available in the Agribalyse 3.0 database.
Results show that average plastic losses are from 75 mg to 4345 mg per kg of fish at the consumer, depending on
the species and the related fishing method. The main plastic losses come from lost fishing gears (macroplastics)
and tire abrasion (microplastics). Results show high variability: when mismanaged, plastic packaging at the end-
of-life (macroplastics) is the main loss to the environment.
As a next stepthe methodologyis to be applied to otherfish or shellfish products, or directlyimplemented in a life
cycle inventory database. Further research should characterize the related impacts to the environment when life
cycle impact assessment methodologies will be available, and identify eco-design solutions to decrease the major
flows to the environment identified.
© 2021 Published by Elsevier B.V.
Keywords:
Life cycle assessment
Marine debris
Plastic pollution
Lost fishing gears
Microplastics
Macroplastics
1. Introduction
Marine debris has been recognized as an environmental concern in
the past decades. Large quantities of plastics leak to the ocean (estima-
tions of 4.8 to 12.7millions of tons in 2010) and generate adverse effects
Science of the Total Environment 804 (2022) 150117
⁎Corresponding author at: ISM-CyVi, Université de Bordeaux, 351 Cours de la
Libération, F-33400 Talence, France.
E-mail address: philippe.loubet@u-bordeaux.fr (P. Loubet).
https://doi.org/10.1016/j.scitotenv.2021.150117
0048-9697/© 2021 Published by Elsevier B.V.
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
on the ecosystems (Jambeck et al., 2015;Ryberg et al., 2019;Werner
et al., 2016). The seafood sector is particularly concerned by this chal-
lenge. On the one hand, a significant proportion of marine litter origi-
nates from fishing activities such as lost gears loss (Lebreton et al.,
2018). On the other hand, marine litter contributes to socioeconomic
and environmental impacts that have implications on fisheries: con-
tamination of fish with ingested plastics, restricted catch due to litter
in nets, vessel damage and staff downtime, reduced earnings and lost
fishing time (Werner et al., 2016).
The sustainability of the seafood sector involves reduction of plastic
use and losses all along the supply chain. In order to support and
strengthen the sector towards a systemic reduction in environmental
impacts, a life cycle thinking is required (Ruiz-Salmón et al., 2020). In
this context, life cycle assessment (LCA) is the most established meth-
odology to assess environmental impacts of products. LCA has been in-
creasingly applied to assess the environmental impacts of seafood
productswith more than 60 casestudies published in the scientific liter-
ature (Ruiz-Salmón et al., 2021). Also, an increase of seafood datasets is
observed in life cycle inventory (LCI) databases (ADEME, 2020). How-
ever, the quantification of plastic emissions to the environment and
their related impacts are still not considered in LCA (in both LCI data-
base and LCIA methods).
Currently, LCA is not adequately addressing the impacts due to ma-
rine debris plastics and microplastics (Sonnemann and Valdivia,
2017). The LCI datasets do not take into account plastic leakages to the
environment. With regard to a methodology to assess the environmen-
tal impacts associated with these plastic emissions Woods et al. (2021)
have published a first framework for the assessment of marine litter im-
pacts in life cycle impact assessment (LCIA), which is planned to be
made operational as part of the MariLCA working group (Verones
et al., 2020). Several set of effect factors have already been developed,
e.g., for macroplastic entanglement impact (Woods et al., 2019), or
physical impact of microplastics on ecosystem quality (Lavoie et al.,
2021). Another challenge is to quantify the transport and the degrada-
tion of plastics within the environment in order to provide a complete
fate factor (Saling et al., 2020).
Concerning LCI, the Plastic Leak Project (PLP) has been an important
step forward as they provide several data and factors that can be used
for building life cycle inventories and modelling the pre-fate (Peano
et al., 2020). Maga et al. (2021) state that LCA datasets containing product
or rather process specificinventoryflows to address the initial release of
plastics into the environment have not yet been generated, and that de-
veloping a stringent methodology to address plastic emissions as part of
an LCA bears multiple complex challenges. Also, GreenDelta aims to cre-
ate a plastic litter extension for the ecoinvent database, as part of its
PLEX project (Ciroth and Kouame, 2019). The PLP and the GreenDelta ap-
proaches have already been applied and compared by the European Com-
mission in the frame of the LCA of alternative feedstocks for plastics
production. In their work, they also expand the PLP approach with an al-
ternative bottom up estimation procedure based on beach litter observa-
tions at the EU level (Nessi et al., 2021, 2020).
Activities related to the seafood sector have not been studied yet
with regard to considering plastics emissions within life cycle inventory
data. This is particularly the case of fishingactivities that can entail the
release of important amounts of plastics to the oceans through lost fish-
ing gears and marine coatings.
Therefore, the aim of this article is to develop a methodology for
quantifying flows ofplastics from the lifecycle of seafood products (cov-
ering fish and shellfish and further on called seafood product life cycle)
to the environment. The methodology is validated with a case study ap-
plied to French seafood products for whichrelevant data are available in
the Agribalyse 3.0 database. The Agribalyse 3.0 is an open source data-
base that provides several datasets for the seafood sector in France
(ADEME, 2020). This approach can be a basis for developing more LCI
datasets, for seafood products and beyond, that include plastic emis-
sions to the environment.
2. Materials and methods
2.1. Methodology to account for plastic flows from the seafood product life
cycle
The proposed methodology is based on the LCA terminology in
terms of scope definition and inventory analysis (Hauschild et al.,
2017;ISO, 2006). Firstly, we define the main unit processes of the sea-
food product life cycle for which plastics can be lost to the environment.
Unit processes are the smallest elements considered in a LCI model for
which inputs and outputs are quantified. For simplicity, unit processes
will be named processes.
The processes may occur either on the foreground system, whichare
directly related to the seafood product system, or in the background
(support activities that supply the foreground system with required
goods and services).
The following foreground processes were considered for a typical
seafood product life cycle: fishing activities, processing/packaging, con-
sumption,end-of-life, as well astransport activities between these pro-
cesses. The considered background processes are the plastic production
activities for packaging and other plastic materials used in the life cycle.
The next step is the identification of inputs and outputs related to
plastics materials. Two types of input and output are considered in
LCI: intermediate flows which are exchanged within the technosphere,
and elementary flows which are exchanged betweentechnosphere and
ecosphere. Intermediate flows are already well defined in LCI datasets.
However, there is none information on the elementary flows of plastics
that reach the environment. These flows are also named plastic losses.
The methodology we propose aims to identify and quantify these
flows for seafood product life cycles. Based on existing framework
(Peano et al., 2020) we identified five types of plastic losses from the
seafood product life cycle:
- Abandoned, lost or discarded fishing gears during fishing activities
(macroplastics),
- Marine coatings applied to boats that can leak during fishing activi-
ties (microplastics),
- Plastic pellets that can be lost to the environment during plastic pro-
duction in the background (microplastics),
- Plastics from tire abrasion during transport activities
(microplastics),
- Mismanaged plastics during the end-of-life (macroplastics).
Processes and plastic leakages are summarized in Fig. 1.
A typical LCI defines elementary flows that reach theenvironment in
different compartments: ocean, freshwater, soil, other terrestrial envi-
ronment, air.
Once in the environment, plastic flows can be transported, degraded
and fragmented before being exposed and harming human health and
ecosystems. This is typically captured during the LCIA phase through
characterization factors including fate factor, exposure factor and effect
factor. LCIA methodologies for micro and macroplastics are currently
under development (Woods et al., 2021). They are out of scope of this
paper.
In a previous study done as part of the Plastic Leak Project (Peano
et al., 2020) dealing with plastic leakages in the context of LCA, a “pre-
fate”modelling is considered to determine the transport of plastics to
final environmental compartment. The pre-fate includes two consecu-
tive parameters: initial release compartment and redistribution to the
final release compartment.
The methodology proposed in this paper takes into account the pre-
fate in order to be in line with the framework proposed by Peano et al.
(2020), and with the aim to quantify the final amount of plastics
reaching the environmental compartments (ocean, freshwater, soil,
other terrestrial environment).
P. Loubet, J. Couturier, R. Horta Arduin et al. Science of the Total Environment 804 (2022) 150117
2
In the following section, we define two types of factors for each type
of loss as proposed by the PLP approach:
- Loss rate (LR) defined as the relative amount of plastics from the
technosphere that leave to the environment. As these loss rates
can have a large variability, we identified three scenarios with aver-
age, minimum and maximum values found in the literature. When
loss rates present geographic variability, France was chosen as rep-
resentative for the current methodology.
- Final release rate (FRR) to the different environmental compart-
ments (ocean, freshwater, soil, and other terrestrial environment).
These factors are based on the literature presented hereafter and are
detailed for the five main types of plastic losses identified for the sea-
food life cycle.
2.2. Identification of initial plastic loss rates and final release rates
2.2.1. Fishing gear (macroplastics)
Abandoned, lost or otherwise discarded fishing gear represents a
significant sourceof plastic debris in the ocean. It includes macroplastics
that have high potential impacts on marine wildlife through entangle-
ment, ensnare or ingestion. They can eventually degrade into
microplastics.
PLP guidelines acknowledge that fishing gear is an important source
of direct plastic loss but they do not includeloss rates due to lack of data.
There is limited literature on the quantification of sea-based plastics
leakages that include fishing gears, but a recent review proposes fishing
gear loss rates at a global scale based on a meta-analysis (Richardson
et al., 2019). They present several types of fishing gears, which is useful
information in the LCA context in order to compare different product
systems based on diverse fishing activities.
The authors define average loss for nets, pots and trap, and lines. We
selected the fishing gears most commonly found in Agribalyse 3.0 LCI
datasets. Fishing boats also use plastic fishing boxes, that can also be
lost to the sea. However, there is no available data on this type of loss.
Therefore, plastic items used on boats that are not fishing gears were
considered with the same loss rates as packaging material (see
Section 2.2.5).
For active fishing activities, data for loss rates are reported as frag-
ments of nets lost, as opposed to whole net loss (Richardson et al.,
2019). This is because such nets are fixed to the boats and an entire
net loss is rare for these gear types, while the incidence of net tear offs
is more common. Nevertheless, the size of these fragments in relation
to the total size of the net is not defined by the authors. We therefore as-
sumed an average value of 50%, a minimum value of 10%, and a maxi-
mum value of 100%.
In its turn, data for passive fishing activities are directly reported as
net loss rates. Also, fishing aggregating devices (FADs) that are gears
used to attract ocean-going pelagic fish are made of plastic based
buoys and can be lost. Richardson et al. (2019) considers a unique
value of 9.9% as loss rate.
Fig. 1. Identification of plastic losses and final releases from seafood life cycle. Final release flows to environmental compartment are not exhaustive.
P. Loubet, J. Couturier, R. Horta Arduin et al. Science of the Total Environment 804 (2022) 150117
3
We considered that 100% of these losses are initially and finally re-
leased to the oceans since fishing activities occur directly in this envi-
ronment compartment. We acknowledge that part of fishing gears can
be transported from the ocean to the terrestrial environment (beaches).
Due to lack of data, we did not consider this redistribution from ocean to
terrestrial environment.
Table 1 summaries the loss rates and final release rates for active and
passive fishing activities.
2.2.2. Marine coatings (microplastics)
Marine coatings are protective layers applied to surfaces, as fishing
boats, exposed to or immersed in salt water. They are composed of
polymer-based paints and can generate microplastics when it migrates
to the ocean. According to Verschoor et al. (2016), emissions from paint
particles occur during:
- maintenance of the boats because of sandingand abrasive blasting of
the coating;
- use of the boats due to regular wear of the coating and occasional
damage.
Verschoor et al. (2016) considered that 1% of the coating is emitted
during maintenance, and that 1% additional is emitted during use.
Based on the same reference, we considered that 60% of the paint is
composed of polymer as the solid phase.
Consequently, the initial plastic leakage ratefrom coatings is equal to
1.2% of the total mass of coatings. We considered variability from 0.5% to
3%.
We also acknowledge that the report of Boucher and Friot (2017)
considers 6% losses from marine coatings but their estimation is based
on a former reference (OECD, 2009). PLP report did not include marine
coatings due to lack of data.
As presented in Table 1, we assumed that this leakage is fully re-
leased to the ocean.
2.2.3. Plastic pellets (microplastics)
Plastic goods, such as packaging, are manufactured from small pel-
lets that are melted toproduce the final product. Pellets can be consid-
ered as primary microplastics with an average size of 5 mm. PLP
methodological guidelines estimates leakages associated to pellets en-
tering drains near plastic facilities: at compounders, master batch
makers, distributors, resellers, storage locations, processors and recy-
clers (Peano et al., 2020). Although other types of losses may occur,
such as at the periphery of plastic facilities, due to lack of data they
were not considered in PLP.
Based on this assumption, Peano et al. (2020) assume leakage rates
of pellets from 0.001% to 0.1% along their supply chain. From this
range, we considered in the present methodology an average of 0.01%
initial leakage rate from pellets for all plastics produced and involved
in the product seafood life cycle (mainly for fishing gears and plastic
packaging). Finalrelease rates are also taken fromPLP and are reported
in Table 1.
2.2.4. Tire abrasion during transport (microplastics)
Abrasion of tire on road surfaces is one of the main sources of
microplastic losses to the environment (Jan Kole et al., 2017). Resulting
particles are a mix of polymers including styrene butadiene rubber, nat-
ural rubber and other additives. They are found in the environment as
embedded with pavement particles, forming tire and road wear parti-
cles (TRWP). However, the fraction from road only includes mineral
material and do not participate in plastic leakages.
Peano et al. (2020) also estimate leakagesfrom tire abrasion, consid-
ering a broad range of parameters including typeof vehicle, type of road,
among others. They also propose two modelling strategies for tire-
related and non-tire related studies. We used the second type of model-
ling strategy and considered the case “Goods transport by truck (light,
medium and heavy trucks)”for computing LR, which is also recom-
mended by the European Commission (Nessi et al., 2021):
LRtruck tires ¼
Dtruck prod ∗Mprod
Loadaverage
∗Losstruck tires ∗ShPolymertruck tires
where D
truck prod
∗M
prod
is the distance over which the products are
transported multiplied by the mass of product transported (in t·km).
This is usually provided in LCI.
Load
average
is the average load from trucks. It is considered to be
12,000 kg per medium or heavy truck.
Loss
truck tires
is the loss of tired tread per kilometre travelled by the
vehicle. It is considered to be 517 mg/km for a medium/heavy truck
long haul and 658 mg/km for a medium/heavy truck short haul.
ShPolymer
truck tires
is the share of polymer (synthetic
rubber + natural rubber) in tire tread. It is considered to be 60% for
medium/heavy truck long haul and 50% for medium/heavy truck short
haul.
We considered two types of trucks for transport in the seafood
product life cycle, with the resulting LR (computed with data from PLP):
- Freight lorry (16 t–32 t), considered to be long haul: 2.74E−5 kg/
tkm (min: 1.51E−5, max: 5.79E−05)
- Freight lorry (7.5 t–16 t), considered to be short haul: 2.59E−5kg/
Table 1
Summaryoflossrateandfinal release rate to the environmental compartments.
Type of losses (and source of data) Variations Loss rate LR (%) Final release rate FRR (%)
Average Min Max Ocean Fresh water Soil Terrestrial
Fishing gears (macroplastics)
(Richardson et al., 2019)
Dredge 50%∙1.8% =.9% 10%∙1.60% =0.16% 1.90% 100% 0% 0% 0%
Gillnet/Trammel net 5.8% 5% 6.50%
Longline 20% 19% 22%
Purse seine 50%∙6.60% =3.3% 10%∙5.90% =0.59% 7.30%
Seine 50%∙2.30% =1.15% 10%∙1.90% =0.19% 2.80%
Trap/pot 19% 18.00% 20%
Trawl bottom 50%∙1.80% =0.9% 10%∙1.60% =0.16% 1.90%
Trawl pelagic 50%∙0.70% =0.35% 10%∙0.58% 0.058% 0.82%
FADs 9.90% 9.90% 9.90%
Marine coatings (microplastics)
(Verschoor et al., 2016)
1.20% 0.50% 3% 100% 0% 0% 0%
Plastic pellets (microplastics)
(Peano et al., 2020)
0.01% 0.001% 0.10% 11.74% 5.09% 65.66% 3.61%
Tire abrasion (microplastics)
(Peano et al., 2020)
Truck 16–32 t (kg/tkm) 2.74E−05 1.51E−05 5.79E−05 1.68% 15.15% 65.66% 3.61%
Truck 7, 5–16 t (kg/tkm) 2.59E−05 2.10E−05 3.40E−05 1.68% 15.15% 65.66% 3.61%
Mismanaged plastics at the end-of-life
(macroplastics)
(Peano et al., 2020)
France 0.02% 0.02% 4% 25% 0% 0% 75%
P. Loubet, J. Couturier, R. Horta Arduin et al. Science of the Total Environment 804 (2022) 150117
4
tkm (min: 2.10E−5, max: 3.40E−05)
- Data for othertypes of trucks not included in this study(such as light
truck) are available in the PLP report (Peano et al., 2020).
The final release rates are presented in Table 1.
2.2.5. Mismanagement at the end-of-life (macroplastics)
Plastic waste that leaks to the environment at the end-of-life in-
cludes uncollected and poorly managed waste. PLP methodological
guidelines proposes loss rates for both mechanisms.
Uncollected waste includes (i) littering which is the disposal of
small, one-off items in the environment (such as throwing a cigarette),
and (ii) fly tipping which is the deliberate disposal of larger quantities of
litter in the environment outside of official waste collection and treat-
ment locations. Poorly managed waste includes (i) dumping, which
mainly occurs in low-income countries where waste can end up in
open dump and (ii) non-sanitary landfills where waste can end up
being mismanaged.
For littering, they differentiate several size of plastic products (small
or detachable, medium and large), and different types of use (in-home
non flushable, in-home flushable and on-the-go). Plastic packaging are
considered as in-h ome (non-flushable) medium size for which the litter
rate is considered to be 0%. PLP guidelines only considers littering for
on-the-go (e.g., cups) or flushable plastics (e.g., cotton swabs).
For fly-tipping, dumping and non-sanitary landfills, PLP guidelines
propose loss rates that are geographically differentiated (on a country
level). These rates are based on a previous report from the World
Bank Group (Kaza et al., 2018). For example, France presents a loss
rate of 0.02%, which is the value that was considered for minimum
and average scenario. PLP alsoconsiders a default value forhigh income
countries which is defined to be 4.3%, and is considered for the maxi-
mum scenario.
Final release rates (considering initial rate and redistribution) for all
mismanaged waste are gathered for medium size plastics which have
low value (since there are packaging materials). The rates are presented
in Table 1.
2.3. Application to French fish products of the Agribalyse database
Aiming to validate the approach presented in Section 2.2, we have
applied the factors to existing LCI datasets in Agribalyse 3.0 database.
For this case study, the functional unit (FU) is “the consumption of
1kgoffish”. System boundaries are the ones presented in Fig. 1. The
quantities of plastics in the technosphere were extracted from the
Agribalyse database.
Agribalyse contains more than 200 specific LCI datasets of raw prod-
ucts, including agriculturalbut also fishery products. LCI datasets of fish-
ery products were collected through the specific project ICV Pêche
(Cloâtre, 2018). It provides specific data for the fishing activities
(including quantity of fishing nets and fishing boats). Data are available
for 12 different species (Table 2), through the definition of triplets re-
lated to: target species, fishing area and fishing gear. It should be
noted that two species originally referred as “seine”fishing (skipjack
and yellowfin tuna) were considered as “purse seine”fishing because
they also rely on fishing aggregating devices that are usually used
with purse seine.
Agribalyse also provides reference data on 2500 food products con-
sumed in France. These data rely on many assumptions and approxima-
tions and correspond to “medium/standard”products that are detailed
in the Agribalyse methodological report (ADEME, 2020). It includes
data related to: transport, processing, packaging(only primary), house-
hold consumption and end-of-life of food products. Product losses are
also considered with generic values (from the Agribalyse methodologi-
cal report) at the processing stage, retail stage and consumer stage.
Thus, the LCI provides elements related to flows of plastics in the
technosphere during these stages of the seafood supply chain. It is to
be noted that Agribalyse is composed of unit processes interconnected
for the agri-food value chain. Supporting products and services such as
energy, packaging materials and infrastructure are modelled based on
ecoinvent 3 system processes (Wernet et al., 2016).
After identifying the different processes of the seafood supply chain,
the goal is to identify and quantify the relevant unit and/or system pro-
cess that may have plastic losses. For doing so, theseprocesses should be
associated with one of the five type of plastic losses presented in
Section 2.2. The list of processes and related type of plastics losses is pro-
vided in SI (“Plastic LCI”tab). It mostly includes foreground processes
but also background processes related to plastic production, as defined
in the methodology.
The collection of these processes as well as the computation of plas-
tic losses has been automated in anExcel tool (available in SI), with the
different steps explained hereafter (Fig. 2):
- All process requirements for 1 kg of Fish are obtained through
openLCA software and imported in the Excel tool.
- The relevant processes are automatically identified in the Excel tool.
The quantities of each process are multiplied with the correct LRand
FRR from Table 1.
- Resulting quantities of plastic loss and final release are assessed
through Sankey diagrams, contribution analysis (to identify the
stages that generate most plastics into the environment) and com-
parison analysis between different types of fish products.
We selected 14 different fish product life cycles to be analysed
(Table 2): 12 of them for the different species with a common packaging
(polystyrene-PS); 2 additional datasets were selected to study the effect
of packaging and processing. Inorder to simplify the analysis of results,
one specificfish product was also selected to show all plastic flows in
detail (Fresh saithe).
Table 2
List of selected fish products life cycles for plastic flows quantification.
Fish species Fishing area Fishing gear Processing Packaging
Mackerel (Scomber scombrus) Northeast Atlantic Trawl pelagic Filleting PS
Saithe (Pollachius virens) North Sea Trawl bottom Filleting PS
Albacore (Thunnus alalunga) Northeast Atlantic Trawl pelagic Filleting PS
Herring (Clupea harengus) Northeast Atlantic Trawl pelagic Filleting PS
Yellowfin tuna (Thunnus albacares) Eastern Central Atlantic Purse seine Filleting PS
Anchovy (Engraulis encrasicolus) Eastern Central Atlantic Seine Filleting PS
Swordfish (Xiphias gladius) Mediterranean Sea Longline Filleting PS
Scallop with coral (Pecten maximus) Saint-Brieuc Bay Dredge No preparation PS
Gadidae-cod (Gadus morhua) Celtic Sea Trawl bottom Filleting PS
Eur. Pilchard (Sardina pilchardus) Eastern Central Atlantic Seine Filleting PS
Skipjack tuna (Katsuwonus pelamis) Eastern Central Atlantic Purse seine Filleting PS
Sole (Solea solea) Bay of Biscay Gillnet Filleting PS
Mackerel (Scomber scombrus) Northeast Atlantic Trawl pelagic Filleting + caning Aluminium
Albacore (Thunnus alalunga) Northeast Atlantic Trawl pelagic Filleting + caning Aluminium
P. Loubet, J. Couturier, R. Horta Arduin et al. Science of the Total Environment 804 (2022) 150117
5
3. Results and discussion
3.1. Typical plastic requirements within the life cycle of a specificfish product
This section focuses on the results for one specificfish product life
cycle, i.e., fresh saithe packaged in polystyrene (PS), the functional
unit being 1 kg of fish at the consumer.
Fig. 3 shows theflows of plastics within the technosphere, which are
directly available from Agribalyse datasets. The FU requires 63.7 g of
plastics/kg of fish at the consumer. The highest plastic requirement is
the polystyrene packaging (52 g). This value is higher than the packag-
ing mass involved during the packaging phase (50 g of packaging/kg of
fish) due to products losses in the distribution chain.
The second highest plastic requirement is the fishing gear as it rep-
resents 9 g of plastics/kg of fish at the consumer. Fresh saithe is fished
with bottompair trawl (63 m of headline) which is composed of
350 kg ethylene vinyl acetate copolymer, 5000 kg polyethylene high
density, 3500 kg synthetic rubber (as well as 5934 kg of steel). Such
bottompair trawl can fish up to 2185 t of saithe over its life time,
resulting in 4 g of plastics/kg of fish at fishery. In this case, 2.23 kg of
fish at the fisheryisrequiredfor1kgoffish at the consumer (consider-
ing the losses and non-edible parts of fish).
Remaining flows of plastics in the technosphere are below 1.5 g/kg
of fish at the consumer and include plastics for marine coatings, other
plastics materials in the fishing boats, and truck tires. As the Agribalyse
datasets is connected to ecoinvent 3 system process, it is not possible to
identify all plastic requirements in the background (for example within
the infrastructure). We assume these plastics flows are minor.
Quantity of plastics at the end-of-life should be the sum of all plastic
requirements (i.e., 63.7 g/kg of fish) since Agribalyse and ecoinvent do
not consider emissions of plastics to the environment. However, the re-
ported value is 60.05 g/kgof fish. This is becausethe datasets do not nec-
essarily consider plastics waste management at the end-of-life or they
do not provide correct mass balance for plastics.
3.2. Initial plastic losses from the life cycle of a specificfish product
Fig. 3 also shows plastic losses to the environment calculated based
on factors from Table 1. The average value of plastic losses is
168.8 mg/kg of fish at the consumer, which represents 0.27% of the
total plastic requirements (63 g).
Major plastic losses are macroplastics from lost fishing gears
(81.3 mg) and microplastics from tire abrasion during transport
(62.6 mg). The value for tire abrasion can also be compared with
ecoinvent dataset that considers “tyre wear emission”for transporta-
tion. The dataset “transport, freight, lorry 16–32 metric ton, EURO6
{RER}”considers 2.2E−04 kg/tkm of tyre wear emissions (Spielmann
et al., 2007). Considering 60% of this emission as plastic, this is 5 times
higher than the average loss rate considered by PLP guidelines (which
is 2.74E−05 kg/tkm for 16–32 t truck as shown in Table 1).
The three remaining sources of losses are in the same order of mag-
nitude: macroplastics from mismanaged waste (12.5 mg), microplastics
from marine coatings (7.1 mg) and microplastics from plastic pellets
(5.4 mg).
Plastic losses for average, min and max scenarios are shown in
Table 3. The losses of macroplastics from mismanaged plasticwaste pre-
sents a large variability de pending on the different s cenarios. In the max
scenario, this value is 2689.7 mg, which is 4.3% (default value for high
income countries as presented in Section 2.2.5) of the mass of plastic
at the end-of-life, mainly composed of packaging. In the average and
min scenarios, 0.02% (value for France) of plastic at the end-of-life is
considered to be lost in the environment. This is 215 times less than
the default value for high-income countries. Once these data originally
come from a single source, it should be refined (Kaza et al., 2018). Fur-
thermore, in two case studies conducted by PLP (dairy and textile sec-
tors) as well as in the case studies conducted by Nessi et al. (2020) on
plastic products, most of the plastic losses were macroplastics occurring
during the end-of-life of the products (Peano et al., 2020).
Microplastics from pellets production also presents a high variability
(2 orders of magnitude between min and max values) but remains a
low contributor in all scenarios.
Macroplastics from lost fishing gears presents one order of
magnitude variability because of the uncertainty associated with the
size of net fragments that is reported by Richardson et al. (2019).
3.3. Final environmental compartments for plastics from the life cycle of a
specificfish product
Fig. 4 shows the final release to the environmental compartments
after application of the pre-fate factors. Such factors consider initial re-
lease rate and redistribution in different environmental compartments.
Most plastics end up into the ocean (84.38 mg), mainly because of the
direct loss of fishing gears (i.e., macroplastics).
Asignificant amount of plastics is released into the soil (51.82 mg)
and other terrestrial environment (12.23 mg). Both environmental
compartments have been merged in Fig. 4 for clarity reasons. Tire abra-
sion is the main source of release of microplastics to these environmen-
tal compartments.
Fig. 2. Procedure to compute plastic losses and final release rates from existing LCI datasets.
P. Loubet, J. Couturier, R. Horta Arduin et al. Science of the Total Environment 804 (2022) 150117
6
Final releases to freshwater represents 11.33 mg, as 11 mg (out of the
initial loss of 22.3 mg) are removed from wastewater treatment plant.
The model does not consider any final release to the air compart-
ment because all plastics initially emitted to air have low residential
time, and end up in one of the other compartments.
3.4. Comparison between different life cycles of fish products
Table 4 shows the plastic requirements, and the plastic losses for the
12 different fish species studied from the Agribalyse datasets, computed
with average loss rates. The downstream life cycle stages (processing,
packaging and use) are considered equivalent in all life cycles: filleting,
polystyrene (PS) packaging, no preparation at the user. Total plastic
losses vary from 74.1 mg/kg of fish (mackerel/pelagic trawling) to
4344.9 mg/kg of fish (sole/gillnetting).
The life cycles of two fish species product systems (sole/gillnet and
swordfish/longline) generate plastic losses with one order of magnitude
higher than the other ones. This is because they rely on high quantity of
fishing gear perkg of fish (72.1 g and 14.9 g, respectively) and their fish-
ing gears have high loss rates (5.9% for gillnet and 20% for longlining).
The other species rely on bottompair trawl, pelagic trawl, seine or
purse seine that have lower loss rates. In addition, less fishing gear is re-
quired for 1 kg of fish with these techniques, except for cod (39.9 g/kg of
fish). In general, there is a high variability in the quantity of plastics re-
quired for fishing gears.
Marine coatings requirements and associated losses also present high
variability because they depend on the fishing boats (size and quantity of
fish per boat). The microplastics losses from marine coatings range from
4.5 mg (mackerel/pelagic trawl) to 187.6 mg (swordfish/longline).
Plastic pellets microplastics losses are similar for the different spe-
cies (around 5 mg) because they rely on the same PS packaging mass
which is the main plastic produced. However, scallop is associated
with a packaging mass 7 times higher than the other species because
it is distributed and sold with shells and requires more packaging for
1kgoffish at the consumer. It results in 33.1 mg of pellets losses.
Microplastics from tire abrasion are also similar for all species once
Agribalyse considers similar transport distances and mass of products
for all species, with the exception of scallops. Similarly, to what we ob-
served with the packaging, this is because shells are also transported
and generate 7 times more microplastics from tire abrasion than the
other seafood product life cycles.
Table 3
Average,min and max plastic leakage flowsfrom the life cycle of fresh Saithepackaged in
PS (FU: 1 kg at the consumer).
mg of plastics/kg of fresh saithe at the consumer Average Min Max
Fishing gears (macroplastics) 81.3 14.4 171.5
Marine coatings (microplastics) 7.1 3.0 17.7
Plastic pellets from production (microplastics) 5.4 0.5 53.5
Tire abrasion from transport (microplastics) 62.6 50.9 82.6
Mismanaged plastic waste (macroplastics) 12.5 12.5 2689.7
Total macroplastics 93.8 27.0 2861.2
Total microplastics 75.0 54.3 153.8
All plastics 168.8 81.3 3015.0
% of plastic leakage to the total plastic requirements 0.27% 0.13% 4.76%
Fig. 3. Major plastic flows in the technosphere and leaching to the environment from the life cycle of Saithe packaged in PS (FU: 1 kg at the consumer, scenario: average loss rates as
indicated in Table 1).
P. Loubet, J. Couturier, R. Horta Arduin et al. Science of the Total Environment 804 (2022) 150117
7
Mismanaged waste losses are mostly dependent on the packaging
mass. Also, species that rely on higher quantity of fishing gear (sword-
fish, sole, cod, and saithe) generate higher macroplastics from
mismanaged nets at the end-of-life.
The influence of packaging and processing is shown in Fig. 5.Asit
can be noticed, seafood products packaged in aluminium do not gener-
ate less plastic losses over the lifecycle (for 1 kg of fish at the consumer).
This is because the studied aluminium products require more
Fig. 4. Final plastic releases to environmental compartments after redistribution, in mg (FU: 1 kg at the consumer, scenario: average)
Table 4
Summary of plastic requirements and plastic losses from the life cycle of 12 fish species product systems (FU: 1 kg of fish at the consumer, scenario: average).
Species Mackerel Saithe Albacore Herring Yellowfin
tuna
Anchovy Swordfish Scallop with
coral
Cod Sardine Skipjack
tuna
Sole
Fishing method Trawl
pelagic
Trawl
bottom
Trawl
pelagic
Trawl
pelagic
Purse
seine
Seine Longline Dredge Trawl
bottom
Seine Purse
seine
Trammel
net
Packaging PS PS PS PS PS PS PS PS PS PS PS PS
Technosphere
plastic flows
(g/FU)
Fishing gear (g) 0.0 9.0 1.2 0.0 5.2 2.2 14.9 0.0 39.9 0.9 5.1 72.1
Marine coatings (g) 0.1 0.6 2.2 0.1 0.5 0.6 17.2 1.9 3.0 1.1 0.5 6.3
Packaging (g) 52.1 52.1 52.1 52.1 52.1 52.1 52.1 372.0 52.1 52.1 52.1 52.1
Plastic production
others (g)
0.0 1.0 0.1 0.0 0.0 1.2 20.0 13.0 0.1 0.5 0.0 9.2
Tire (g) 1.1 1.2 1.0 1.1 1.0 1.1 1.0 4.9 1.4 1.1 1.0 1.3
Plastics at the
end-of-life (g)
50.0 60.0 51.2 50.0 55.2 53.4 83.8 375.9 89.9 51.3 55.1 131.0
Total plastic
requirements (g)
53.2 63.3 54.3 53.2 58.2 56.7 87.9 389.9 93.4 54.6 58.2 134.7
Environment
plastic losses
(mg/FU)
Macro: fishing gear
(mg)
0.1 81.3 4.1 0.1 179.1 25.7 2976.8 0.0 358.7 10.0 176.4 4184.5
Micro: marine coating
(mg)
1.4 7.1 21.0 1.6 5.6 5.8 187.6 15.6 19.7 6.0 5.5 53.3
Micro: plastic
production (mg)
4.5 5.4 4.6 4.5 4.9 4.8 7.5 33.1 7.9 4.6 4.9 11.5
Micro: tire abrasion
(mg)
57.7 62.6 52.1 59.5 52.1 59.5 52.1 265.1 76.2 59.5 52.1 68.4
Macro: mismanaged
waste (mg)
10.4 12.5 10.7 10.4 11.5 11.1 17.5 78.3 18.7 10.7 11.5 27.3
Total macroplastics
(mg)
10.5 93.8 14.8 10.5 190.6 36.9 2994.2 78.3 377.4 20.7 187.9 4211.7
Total microplastics
(mg)
63.6 75.0 77.7 65.5 62.7 70.0 247.1 313.9 103.8 70.1 62.5 133.2
All plastics (mg) 74.1 168.8 92.5 76.0 253.2 106.9 3241.3 392.2 481.2 90.8 250.4 4344.9
Ratio between plastic
losses and plastic
requirements
0.14% 0.27% 0.17% 0.14% 0.34% 0.09% 3.69% 0.10% 0.52% 0.17% 0.34% 3.23%
P. Loubet, J. Couturier, R. Horta Arduin et al. Science of the Total Environment 804 (2022) 150117
8
processing stages and therefore more raw fish (because of fish losses),
thus resulting in an increase of fishing activities and transport require-
ments. For fish products which are filetedand packaged in PS, the plastic
losses associated with plastic production and end-of-life are higher but
do not counterbalance the other plastic flows.
However, in the case of the max scenario where 4% of packaging is
mismanaged at the end-of-life (instead of 0.02% in the average sce-
nario), the increase in macroplastics loss at the end-of-life is predomi-
nant. It shows again the high variability of plastic losses quantity,
which depends on geographical parameter and consumer behaviour.
Full results in terms of final release rates for the studied seafood
product life cycles are available in SI (“Results”tab).
3.5. Perspectives
LCI data generated in this paper could be implemented into the
Agribalyse datasets concerning the seafood sector. The proposed meth-
odology and procedure presented in Fig. 2 could also be applied for all
types of agri-food products studied in Agribalyse. In this case,the plastic
loss rates defined in this study for plastic production, processing/pack-
aging and end-of-life processes couldbe used for the agri-food products.
New loss rates should be developed for specific activities that use plastic
such as aquaculture (e.g., loss of plastic buoys at sea for oyster produc-
tion) and agriculture (e.g., loss of plastic mulching in the soil). However,
we acknowledge that the implementation of plastic losses would be
partial because it only includes losses in the foreground system, and
on some processes in the background (plastic production).
Wider database such as ecoinvent should also be updated to con-
sider plastic emissions in all sectors and in all background processes.
This is an important step towards the consideration of plastic related
impacts in LCA. A more general challenge is to develop mass balanced
unit processes regarding intermediate and elementary inputs and out-
puts. This is also the case for other types of materials such as metals
for which their dissipation is not well accounted yet (Beylot et al.,
2021). Past initiatives have shown that existing database can be fully
updated to consider mass balance of a specific material or resource,
e.g., water (Pfister et al., 2015).
In order to expand data on plastic flows, several data sources can be
used. Maga et al. (2021) classify 4 types of data sources: survey, statisti-
cal data, empirical measurement and probabilistic models. This paper
mostly includes statistical data on consumption of products or littering
and mismanaged waste. Further data sources such as empirical mea-
surements of plastic parts found in the environment can be used to im-
prove the quality of the data even if it is complex to link retrieved plastic
items to their initial product.
Also, top-down quantification of plastics leaching into the environ-
ment have been conducted in the last years (Jambeck et al., 2015;
Ryberg et al., 2019). Further research could be conducted to reconciliate
top-down data with bottom-up LCIs such as the ones provided here.
This would require producing sector specific (such as seafood) top-
down data, which is not yet the case.
Complementary to the quantification of the plastic emissions from
value chains, it is important to assess its impact in the environment.
Several initiatives have been launched to develop LCIA methods in the
past years as described before. Most of them are listed under the
MariLCA working group (Verones et al., 2020) and can be integrated
into the initial framework recently published (Woods et al., 2021). Ac-
cording to their framework, LCI should be the quantity of plastic initially
leaked to a specific environmental compartment. Therefore, only the
initial loss rate presented in this paper should be considered as LCI,
since the final release rate includes the transport factor that is part of
the fate modelling. It is to be noted that the fate also includes degrada-
tion and fragmentation in the environment. The MariLCA framework
and Maga et al. (2021) also recommend to attribute further details to
the inventory apart from the mass of plastics: (i) receiving environmen-
tal compartment (air, ocean, freshwater, terrestrial), (ii) fragment type
(micro, macro, etc.), (iii) material type (PS, PEHD, etc.), and (iv) loca-
tion. Attributes (i), (ii) are available from our study. The material
types (iii) are not specifically discussed in our results, but are available
in Supplementary Information. However, further properties on the ma-
terial such as the shape can have a strong influence on the fate or the ef-
fect of the plastic and should also be taken into account (Maga et al.,
2021). The location (attribute iv) is not specifically stated here but can
be easily specified, at least for the foreground processes.
The application of these LCIA methods to theLCI data presented here
would complement the LCA of seafood and show the magnitude of plas-
tic related impacts and/or damages compared to other environmental
issues (for example in terms of (eco)-toxicity).
LCI and LCIA results related to plastic can also be used to identify the
types of losses that generate most impacts, and therefore to prioritize
eco-design initiatives. Considering that lost fishing gears is a major plas-
tic loss in seafood product life cycles, several preventive and mitigation
measureshave been developed (Gilman, 2015). This includes, for exam-
ple, gear marking to identify ownership and increase visibility, technol-
ogy to avoid unwanted gear contact with seabed, etc. Plastic packaging
and transport by truck are also two potential sources of plastic losses.
Some actions could also be implemented toreduce them, asuse of bio-
degradable packaging, improvement of end-of-life management, and
reduction of transport distance through local circuits.
4. Conclusion
This paper proposes a methodology and an automatic procedure to
account for plastic losses and final releases from the life cycle of seafood
Fig. 5. Plastic losses from the life cycle of fresh saithe as reference and 2 fish species product systems with different packaging/processing (FU: 1 kg of fish at the consumer, scenario:
average).
P. Loubet, J. Couturier, R. Horta Arduin et al. Science of the Total Environment 804 (2022) 150117
9
products into the environment. It is based on the guidelines and data
from the PLP (Peano et al., 2020) and complements it with loss rates
specific to the seafood sector (fishing gear and marine coatings).
Results have shown a high variability in plastic losses depending on
the fish species and the associated fishing method. Plastic losses com-
puted with average rates range from 74 mg to 4344 mg per kg of fish at
the consumer. Most of the fish species result in plastic losses around
100 mg per kg of fish at the consumer (Mackerel, Saithe, Albacore,
Herring, Yellowfin tuna, Anchovy, Sardine and Skipjack tuna). The highest
plastic losses are related to species that require high mass of passive fish-
ing gear (e.g. Swordfish and Sole, since such gears present high loss rates).
Fishing gear (macroplastics) and tire abrasion (microplastics) are the
main plastic losses when considering average rates for all types of losses.
When considering maximum rates, mismanaged plastic packaging at the
end-of-life is the main plastic loss. The variability of results depending on
the parameters (min, max, average) show that there are research needs
to better quantify several types of losses, mainly lost fishing gears and
mismanaged waste.
This work paves the way to better account for plastic pollution in
LCA, more particularly for the seafood sector. Several perspectives are
foreseen: broadening the methodology to other products, assessing
the associated impacts with preliminary LCIA characterization factors,
identifying and prioritizing relevant eco-design solutions to mitigate
plastic pollution arising from a seafood product life cycle of seafood.
CRediT authorship contribution statement
Philippe Loubet: Conceptualization, Methodology, Validation, Writ-
ing –original draft, Supervision. Julien Couturier: Methodology, Inves-
tigation, Data curation, Resources, Writing –review & editing. Rachel
Horta Arduin: Writing –review & editing, Visualization. Guido
Sonnemann: Writing –review & editing, Funding acquisition.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Acknowledgements
This work was supported by the EAPA_576/2018NEPTUNUS project.
The authors would like to acknowledge the financial support of Interreg
Atlantic Area.
Appendix A. Supplementary data
Supporting information to this article includes the Excel tool with 4
different tabs: “Import tool”,“Plastic LCI”,“Models”and “Results”.Sup-
plementary data to this article can be found online at https://doi.org/
10.1016/j.scitotenv.2021.150117.
References
ADEME, 2020. Agribalyse data v3.0 [WWW Document]. . https://doc.agribalyse.fr/docu-
mentation-en/.
Beylot, A., Ardente, F., Sala, S., Zampori, L., 2021. Mineral resource dissipation inlife cycle
inventories. Int. J. Life Cycle Assess. 26, 497–510. https://doi.org/10.1007/s11367-
021-01875-4.
Boucher, J., Friot, D., 2017. Primary microplastics in the oceans. IUCN.
Ciroth, A., Kouame, N., 2019. Life Cycle Mamangement Conference LCM. Elementary litter
in life cycle inventories, approach and application Poznan.
Cloâtre, T., 2018. Methodological report for the French LCI project on fisheries.
Gilman, E.,2015. Status ofinternational monitoring and management of abandoned, lost
and discardedfishing gear and ghost fishing. Mar. Policy 60, 225–239. https://doi.org/
10.1016/j.marpol.2015.06.016.
Hauschild, M.Z., Rosenbaum, R.K., Olsen, S.I., 2017. Life Cycle Assessment: Theory and
Practice, Life Cycle Assessment: Theory and Practice. Springer International Publish-
ing https://doi.org/10.1007/978-3-319-56475-3.
ISO, 2006. ISO - 14040 —Environmental Management —Life Cycle Assessment —Princi-
ples and framework Geneva.
Jambeck, J.R., Geyer, R., Wilcox, C., Siegler, T.R., Perryman, M., Andrady, A., Narayan, R.,
Law, K.L., 2015. Plastic waste inputs from land into the ocean. Science 347 (80-.),
768–771. https://doi.org/10.1126/science.1260352.
Jan Kole, P.,Löhr, A.J., Van Belleghem, F.G.A.J., Ragas, A.M.J., 2017. Wear and tear oftyres: a
stealthy source of microplasticsin the environment. Int. J. Environ.Res. Public Health
14. https://doi.org/10.3390/ijerph14101265.
Kaza, S., Yao, L.C., Bhada-Tata, P., Van Woerden, F., 2018. What a waste 2.0: a global snap-
shot of solid waste management to 2050. World Bank, Washington, DC https://doi.
org/10.1596/978-1-4648-1329-0.
Lavoie, J., Boulay, A., Bulle, C., 2021. Aquatic micro- and nano-plastics in life cycle assess-
ment: development of an effect factor for the quantification of their physical impact
on biota. J. Ind. Ecol. 1–13. https://doi.org/10.1111/jiec.13140.
Lebreton, L., Slat, B., Ferrari, F., Sainte-Rose, B., Aitken, J., Marthouse, R., Hajbane, S.,
Cunsolo, S., Schwarz, A., Levivier, A., Noble, K., Debeljak, P., Maral, H., Schoeneich-
Argent, R., Brambini, R., Reisser, J., 2018. Evidence that the Great Pacific Garbage
Patch is rapidly accumulating plastic. Sci. Rep. 8, 1–15. https://doi.org/10.1038/
s41598-018-22939-w.
Maga, D., Thonemann, N., Strothmann, P., Sonnemann, G., 2021. How to account for plas-
tic emissions in life cycle inventory analysis? Resour. Conserv. Recycl. 168, 105331.
https://doi.org/10.1016/j.resconrec.2020.105331.
Nessi, S., Bulgheroni, C., Konti, A., Sinkko, T., Tonini, D., Pant, R., 2020. JRC Technical Re-
ports. Comparative Life-Cycle Assessment of Alternative Feedstock for Plastics Pro-
duction. European Commission https://doi.org/10.2760/XXXXX.
Nessi, S., Sinkko, T., Bulgheroni, C., Garcia-Gutierrez, P., Giuntoli, J., Konti, A., Sanye-
Mengual, E., Tonini, D., Pant, R., Marelli, L., Ardente, F., 2021. JRC Technical Report.
Life Cycle Assessment (LCA) of alternative feedstocks for plastics production Part 1:
the Plastics LCA method https://doi.org/10.2760/271095.
OECD, 2009. Emission Scenario Document on Coating Industry (Paints, Lacquers and Var-
nishes). Ser. Emiss. Scenar. Doc. 1–201.
Peano, L., Kounina, A., Magaud, V., Chalumeau, S., Zgola, M., Boucher, J., 2020. Plastic Leak
Project. Methodological Guidelines. Quantis and EA. quantis-intl.com/plastic-leak-
project-guidelines.
Pfister, S., Vionnet, S., Levova, T., Humbert, S., 2015. Ecoinvent 3: assessing water use in
LCA and facilitating water footprinting. Int. J. Life Cycle Assess https://doi.org/10.
1007/s11367-015-0937-0.
Richardson, K., Hardesty, B.D., Wilcox, C., 2019. Estimates of fishing gear loss rates at a
global scale: a literature review and meta-analysis. Fish Fish. 20, 1218–1231.
https://doi.org/10.1111/faf.12407.
Ruiz-Salmón, I., Margallo, M., Laso, J., Villanueva-Rey, P., Mariño, D., Quinteiro, P., Dias,
A.C., Nunes, M.L., Marques, A., Feijoo, G., Moreira, M.T., Loubet, P., Sonnemann, G.,
Morse, A., Cooney, R., Clifford, E., Rowan, N., Méndez-Paz, D., Iglesias-Parga, X.,
Anglada, C., Martin, J.-C., Irabien, Á., Aldaco, R., 2020. Addressing challenges and op-
portunities of the European seafood sector under a circular economy framework.
Curr. Opin. Environ. Sci. Health 13, 101–106. https://doi.org/10.1016/j.coesh.2020.
01.004.
Ruiz-Salmón, I., Laso, J., Margallo, M., Villanueva-Rey, P., Rodríguez, E., Quinteiro, P., Dias,
A.C., Almeida, C., Nunes, M.L., Marques, A., Cortés, A., Moreira, M.T., Feijoo,G., Loubet,
P., Sonnemann, G., Morse, A.P., Cooney, R., Clifford, E., Regueiro, L., Méndez, D.,
Anglada, C., Noirot, C., Rowan, N.,Vázquez-Rowe, I., Aldaco, R., 2021. Life cycle assess-
ment of fish and seafood processed products –a review of methodologies and new
challenges. Sci. Total Environ. 761, 144094. https://doi.org/10.1016/j.scitotenv.2020.
144094.
Ryberg, M.W., Hauschild, M.Z., Wang, F., Averous-Monnery, S., Laurent, A., 2019. Global
environmental losses of plastics across their value chains. Resour. Conserv. Recycl.
151, 104459. https://doi.org/10.1016/j.resconrec.2019.104459.
Saling, P., Gyuzeleva, L., Wittstock, K., Wessolowski, V., Griesshammer, R., 2020. Life cycle
impact assessment of microplastics as one component of marine plastic debris. Int.
J. Life Cycle Assess. 25, 2008–2026. https://doi.org/10.1007/s11367-020-01802-z.
Sonnemann, G., Valdivia, S., 2017. Medellin declaration on marine litter in life cycle as-
sessment and management: facilitated by the forum for sustainability through life
cycle innovation (FSLCI) in close cooperation with La Red Iberoamericana de Ciclo
de Vida (RICV) on Wednesday 14 of Jun. Int. J. Life Cycle Assess. 22, 1637–1639.
https://doi.org/10.1007/s11367-017-1382-z.
Spielmann, M., Bauer, C., Dones, R., Tuchschmid, M., 2007. Transport Services. ecoinvent
report No. 14 Dübendorf.
Verones,F., Woods, J., Jolliet,O., Boulay, A.-M.,Vazquez-Rowe, I., 2020. SETAC Europe 30th
Annual Meeting. Drawing a framework to assess marine plastic litter impacts in life
cycle impact assessment: the MarILCA project.
Verschoor, A., de Poorter, L., Dröge, R., Kuenen, J., de Valk, E., 2016. Abrasive cleaning
agents, paints and tyrewear. Emission of microplastics and potential mitigation mea-
sures Bilthoven, the Netherlands.
Werner, S., Budziak, A., Van Franeker, J., Galgani, F., Hanke, G., Maes, T., Matiddi, M.,
Nilsson, P., Oosterbaan, L., Priestland, E., Thompson, R., Veiga, J., Vlachogianni, T.,
2016. MSFD GES TG Marine Litter-Thematic Report, JRC Technical report. Harm
caused by Marine Litter https://doi.org/10.2788/19937.
Wernet, G., Bauer, C., Steubing, B., Reinhard, J., Moreno-Ruiz, E., Weidema, B., 2016. The
ecoinventdatabase version3 (part I): overview andmethodology. Int.J. Life Cycle As-
sess. 21, 1218–1230. https://doi.org/10.1007/s11367-016-1087-8.
Woods, J.S., Rødder, G., Verones,F., 2019. An effect factor approach for quantifying theen-
tanglement impact on marine species of macroplastic debris within life cycle impact
assessment. Ecol. Indic. 99, 61–66. https://doi.org/10.1016/j.ecolind.2018.12.018.
Woods, J.S., Verones, F., Jolliet, O., Vázquez-Rowe, I., Boulay, A.-M., 2021. A framework for
the assessment of marine litter impacts in life cycle impact assessment. Ecol. Indic.
129, 107918. https://doi.org/10.1016/j.ecolind.2021.107918.
P. Loubet, J. Couturier, R. Horta Arduin et al. Science of the Total Environment 804 (2022) 150117
10
