ArticlePDF Available

Life cycle inventory of plastics losses from seafood supply chains: Methodology and application to French fish products

Authors:

Abstract and Figures

Plastic debris into the environment 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 consider plastic 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 from the 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 Leak Project. They are defined for 5 types of micro and macro plastic losses: lost fishing gears, marine coatings, 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 step the methodology is to be applied to other fish or shellfish products, or directly implemented 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.
Content may be subject to copyright.
Life cycle inventory of plastics losses from seafood supply chains:
Methodology and application to French sh 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 sh products from the French
Agribalyse database are analysed.
Plastic losses range from 75 mg to
4345 mg per kg of sh at the consumer.
Main plastic losses come from lost sh-
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 nd relevant mitigation strategies without burden shifting.
This work proposes a methodology for quantifying ows of plastics fromthe life cycle of the seafood products to
the environment. It is based on loss rate and nal release rate considering a pre-fate approach as proposed by the
Plastic L eak Project. They a re denedfor 5 types of micro and macro plastic losses: lost shing 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 sh 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 sh at the consumer, depending on
the species and the related shing method. The main plastic losses come from lost shing 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 othersh or shellsh 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
ows to the environment identied.
© 2021 Published by Elsevier B.V.
Keywords:
Life cycle assessment
Marine debris
Plastic pollution
Lost shing 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 signicant proportion of marine litter origi-
nates from shing 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 sheries: con-
tamination of sh with ingested plastics, restricted catch due to litter
in nets, vessel damage and staff downtime, reduced earnings and lost
shing 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 scientic 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 quantication 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 rst 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 specicinventoryows 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 shingactivities that can entail the
release of important amounts of plastics to the oceans through lost sh-
ing gears and marine coatings.
Therefore, the aim of this article is to develop a methodology for
quantifying ows ofplastics from the lifecycle of seafood products (cov-
ering sh and shellsh 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 ows from the seafood product life
cycle
The proposed methodology is based on the LCA terminology in
terms of scope denition and inventory analysis (Hauschild et al.,
2017;ISO, 2006). Firstly, we dene 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 quantied. 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: shing 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 identication of inputs and outputs related to
plastics materials. Two types of input and output are considered in
LCI: intermediate ows which are exchanged within the technosphere,
and elementary ows which are exchanged betweentechnosphere and
ecosphere. Intermediate ows are already well dened in LCI datasets.
However, there is none information on the elementary ows of plastics
that reach the environment. These ows are also named plastic losses.
The methodology we propose aims to identify and quantify these
ows for seafood product life cycles. Based on existing framework
(Peano et al., 2020) we identied ve types of plastic losses from the
seafood product life cycle:
- Abandoned, lost or discarded shing gears during shing activities
(macroplastics),
- Marine coatings applied to boats that can leak during shing 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 denes elementary ows that reach theenvironment in
different compartments: ocean, freshwater, soil, other terrestrial envi-
ronment, air.
Once in the environment, plastic ows 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-
fatemodelling is considered to determine the transport of plastics to
nal environmental compartment. The pre-fate includes two consecu-
tive parameters: initial release compartment and redistribution to the
nal 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 nal 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 dene two types of factors for each type
of loss as proposed by the PLP approach:
- Loss rate (LR) dened as the relative amount of plastics from the
technosphere that leave to the environment. As these loss rates
can have a large variability, we identied 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 ve main types of plastic losses identied for the sea-
food life cycle.
2.2. Identication of initial plastic loss rates and nal release rates
2.2.1. Fishing gear (macroplastics)
Abandoned, lost or otherwise discarded shing gear represents a
signicant 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 shing 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 quantication of sea-based plastics
leakages that include shing gears, but a recent review proposes shing
gear loss rates at a global scale based on a meta-analysis (Richardson
et al., 2019). They present several types of shing gears, which is useful
information in the LCA context in order to compare different product
systems based on diverse shing activities.
The authors dene average loss for nets, pots and trap, and lines. We
selected the shing gears most commonly found in Agribalyse 3.0 LCI
datasets. Fishing boats also use plastic shing 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 shing gears were
considered with the same loss rates as packaging material (see
Section 2.2.5).
For active shing 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 xed 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 dened 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 shing activities are directly reported as
net loss rates. Also, shing aggregating devices (FADs) that are gears
used to attract ocean-going pelagic sh 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. Identication of plastic losses and nal releases from seafood life cycle. Final release ows 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 nally re-
leased to the oceans since shing activities occur directly in this envi-
ronment compartment. We acknowledge that part of shing 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 nal release rates for active and
passive shing activities.
2.2.2. Marine coatings (microplastics)
Marine coatings are protective layers applied to surfaces, as shing
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 nal 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 shing 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 t32 t), considered to be long haul: 2.74E5 kg/
tkm (min: 1.51E5, max: 5.79E05)
- Freight lorry (7.5 t16 t), considered to be short haul: 2.59E5kg/
Table 1
Summaryoflossrateandnal 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 1632 t (kg/tkm) 2.74E05 1.51E05 5.79E05 1.68% 15.15% 65.66% 3.61%
Truck 7, 516 t (kg/tkm) 2.59E05 2.10E05 3.40E05 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: 2.10E5, max: 3.40E05)
- 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 nal 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) y tipping which is the deliberate disposal of larger quantities of
litter in the environment outside of ofcial 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 landlls 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 ushable, in-home ushable and on-the-go). Plastic packaging are
considered as in-h ome (non-ushable) 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 ushable plastics (e.g., cotton swabs).
For y-tipping, dumping and non-sanitary landlls, 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 dened 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 sh 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
1kgofsh. 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 specic LCI datasets of raw prod-
ucts, including agriculturalbut also shery products. LCI datasets of sh-
ery products were collected through the specic project ICV Pêche
(Cloâtre, 2018). It provides specic data for the shing activities
(including quantity of shing nets and shing boats). Data are available
for 12 different species (Table 2), through the denition of triplets re-
lated to: target species, shing area and shing gear. It should be
noted that two species originally referred as seineshing (skipjack
and yellown tuna) were considered as purse seineshing because
they also rely on shing 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/standardproducts 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 ows 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 ve 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 LCItab). It mostly includes foreground processes
but also background processes related to plastic production, as dened
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 identied 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 nal 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 sh products.
We selected 14 different sh 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 specicsh product was also selected to show all plastic ows in
detail (Fresh saithe).
Table 2
List of selected sh products life cycles for plastic ows quantication.
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
Yellown tuna (Thunnus albacares) Eastern Central Atlantic Purse seine Filleting PS
Anchovy (Engraulis encrasicolus) Eastern Central Atlantic Seine Filleting PS
Swordsh (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 specicsh product
This section focuses on the results for one specicsh product life
cycle, i.e., fresh saithe packaged in polystyrene (PS), the functional
unit being 1 kg of sh at the consumer.
Fig. 3 shows theows of plastics within the technosphere, which are
directly available from Agribalyse datasets. The FU requires 63.7 g of
plastics/kg of sh 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
sh) due to products losses in the distribution chain.
The second highest plastic requirement is the shing gear as it rep-
resents 9 g of plastics/kg of sh at the consumer. Fresh saithe is shed
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 sh up to 2185 t of saithe over its life time,
resulting in 4 g of plastics/kg of sh at shery. In this case, 2.23 kg of
sh at the sheryisrequiredfor1kgofsh at the consumer (consider-
ing the losses and non-edible parts of sh).
Remaining ows of plastics in the technosphere are below 1.5 g/kg
of sh at the consumer and include plastics for marine coatings, other
plastics materials in the shing 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 ows 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 sh) since Agribalyse and ecoinvent do
not consider emissions of plastics to the environment. However, the re-
ported value is 60.05 g/kgof sh. 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 specicsh 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 sh at the consumer, which represents 0.27% of the
total plastic requirements (63 g).
Major plastic losses are macroplastics from lost shing 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 emissionfor transporta-
tion. The dataset transport, freight, lorry 1632 metric ton, EURO6
{RER}considers 2.2E04 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 2.74E05 kg/tkm for 1632 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 rened (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 shing 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
specicsh product
Fig. 4 shows the nal 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 shing gears (i.e., macroplastics).
Asignicant 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 nal 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 nal 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 sh products
Table 4 shows the plastic requirements, and the plastic losses for the
12 different sh 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: lleting,
polystyrene (PS) packaging, no preparation at the user. Total plastic
losses vary from 74.1 mg/kg of sh (mackerel/pelagic trawling) to
4344.9 mg/kg of sh (sole/gillnetting).
The life cycles of two sh species product systems (sole/gillnet and
swordsh/longline) generate plastic losses with one order of magnitude
higher than the other ones. This is because they rely on high quantity of
shing gear perkg of sh (72.1 g and 14.9 g, respectively) and their sh-
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 shing gear is re-
quired for 1 kg of sh with these techniques, except for cod (39.9 g/kg of
sh). In general, there is a high variability in the quantity of plastics re-
quired for shing gears.
Marine coatings requirements and associated losses also present high
variability because they depend on the shing boats (size and quantity of
sh per boat). The microplastics losses from marine coatings range from
4.5 mg (mackerel/pelagic trawl) to 187.6 mg (swordsh/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
1kgofsh 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 owsfrom 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 ows 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 shing gear (sword-
sh, sole, cod, and saithe) generate higher macroplastics from
mismanaged nets at the end-of-life.
The inuence 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 sh 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 sh species product systems (FU: 1 kg of sh at the consumer, scenario: average).
Species Mackerel Saithe Albacore Herring Yellown
tuna
Anchovy Swordsh 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 ows
(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: shing 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 sh (because of sh losses),
thus resulting in an increase of shing activities and transport require-
ments. For sh products which are letedand packaged in PS, the plastic
losses associated with plastic production and end-of-life are higher but
do not counterbalance the other plastic ows.
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 nal release rates for the studied seafood
product life cycles are available in SI (Resultstab).
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 dened 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 specic 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 specic material or resource,
e.g., water (Pster et al., 2015).
In order to expand data on plastic ows, 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 quantication 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 specic (such as seafood) top-
down data, which is not yet the case.
Complementary to the quantication 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 specic environmental compartment. Therefore, only the
initial loss rate presented in this paper should be considered as LCI,
since the nal 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 specically 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 inuence 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 specically stated here but can
be easily specied, 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 shing 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 nal releases from the life cycle of seafood
Fig. 5. Plastic losses from the life cycle of fresh saithe as reference and 2 sh species product systems with different packaging/processing (FU: 1 kg of sh 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
specic to the seafood sector (shing gear and marine coatings).
Results have shown a high variability in plastic losses depending on
the sh species and the associated shing method. Plastic losses com-
puted with average rates range from 74 mg to 4344 mg per kg of sh at
the consumer. Most of the sh species result in plastic losses around
100 mg per kg of sh at the consumer (Mackerel, Saithe, Albacore,
Herring, Yellown tuna, Anchovy, Sardine and Skipjack tuna). The highest
plastic losses are related to species that require high mass of passive sh-
ing gear (e.g. Swordsh 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 shing 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 nancial
interests or personal relationships that could have appeared to inu-
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 nancial 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,Modelsand 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, 497510. 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 sheries.
Gilman, E.,2015. Status ofinternational monitoring and management of abandoned, lost
and discardedshing gear and ghost shing. Mar. Policy 60, 225239. 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-.),
768771. 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 quantication of their physical impact
on biota. J. Ind. Ecol. 113. 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 Pacic Garbage
Patch is rapidly accumulating plastic. Sci. Rep. 8, 115. 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. 1201.
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.
Pster, 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 shing gear loss rates at a
global scale: a literature review and meta-analysis. Fish Fish. 20, 12181231.
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, 101106. 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 sh 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, 20082026. 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, 16371639.
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, 12181230. 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, 6166. 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
... Also, current life cycle inventory (LCI) database do not consider plastic losses to the environment. Since a significant proportion of marine plastic debris originates from fishing activities, it is worthwhile to quantify these losses and to identify the main sources of macroplastics and microplastics (Loubet et al. 2022). ...
... We propose to map the plastic flows within the technosphere and to the environment for the production of octopus from the vessels studied. We used the methodology from Loubet et al. (2022) in order to compute plastic losses to the environment. Considering the boundaries of this study, we only consider three activities (fishing, plastic production, and plastic waste management) and four types of plastic losses: ...
... Loss rates for abandoned, lost, or discarded fishing gears during fishing activities were directly computed from primary data gathered in this study (Table 1). Loss rates for other plastic losses were taken from literature sources identified in Loubet et al. (2022). Two scenarios were set up, with average and max values, as there can be a large variability in these loss rates (Table 2). ...
Article
Full-text available
Purpose Common octopus is the fishing species with highest economic revenue in Portugal, and its consumption per capita is very high. The majority of catches come from the small-scale fleet with pots and traps. The aims were to assess main environmental impacts of common octopus’ fishery with traps and pots in the Algarve region, where the most important fleet size and landings volume occurs, and to find if there are significant differences between both fishing gears. Methods The assessment includes standard LCA impact categories, fishery-specific impact categories, and quantification of macroplastics and microplastics emitted to the environment. The functional unit selected was 1 kg of octopus and the study was a ‘cradle to gate’ system. The scope included fishing operations until the product is landed at the harbour. Primary data was obtained by face-to-face questionnaires from 22 vessels, with an average of 1005 pots and 1211 traps per vessel, and 372 pots and 234 traps lost annually to the environment. Plastic pots have a concrete block and traps are a metal framed covered by plastic netting. Each trap or pot is connected to the main line at regular intervals. Unlike traps, pots do not need bait. Results and discussion Fuel contribution to global warming is very high and where the highest potential exists to lower down the carbon footprint. The fuel use intensity resulted in 0.9 L/kg of octopus. The bait used in traps is significant and raises further environmental costs related with fuel consumption. The use of traps represents more than two times the impacts found for pots in all the categories studied except ecotoxicity categories. Zinc use was the main contributor to ecotoxicity categories, but it has not been included in other fishery LCA studies. It was estimated that 12.2 g of plastics is lost to the environment per kg of octopus. The loss of macroplastics from fishing gears was the highest contributor. Conclusions The carbon footprint obtained was 3.1 kg CO 2 eq per kg of octopus, being lower compared to other seafood products, and less than half compared to octopus caught with trawling. Pots and traps are highly selective fishing gears, causing negligible disturbance to the seafloor. The stock is not assessed, but management measures exist and can be improved. A drawback exists related with gears lost to the environment.
... In addition, around 52% of plastic pollution is carried by rivers, and plastic leakage into the ocean from land-based sources has emerged as an important cause of different marine ecosystems damages [70]. These impacts are still not considered in LCA studies, and therefore, the subsequent fate of polymers and their products in the marine environment might be underestimated [71]. Viable alternatives to plastic from bio-based materials have emerged [72], but they still face challenges in their disposal effectiveness Sustainability 2022, 14, 3054 9 of 25 and impose changes in plastic separation systems. ...
... Concerning life cycle inventory, Loubet et al. (2022) [71] proposed the first methodological framework to measure the plastic flows of the life cycle of seafood products based on the suggestions of the Plastic Leak Project [115], consisting of quantifying the loss rate and the final release rate. In this methodology, applied to French seafood products, loss rates are defined for 5 types of micro-and macro-plastic losses occurring at different life cycle stages of seafood products: (1) abandoned, lost, and discarded fishing gear (ALDFG) and (2) marine coatings (during fishing activities), (3) polymer pellets (during the production of plastic), (4) tire abrasion (occurred in transportation), and (5) plastic bungle (during end-of-life). ...
... Concerning life cycle inventory, Loubet et al. (2022) [71] proposed the first methodological framework to measure the plastic flows of the life cycle of seafood products based on the suggestions of the Plastic Leak Project [115], consisting of quantifying the loss rate and the final release rate. In this methodology, applied to French seafood products, loss rates are defined for 5 types of micro-and macro-plastic losses occurring at different life cycle stages of seafood products: (1) abandoned, lost, and discarded fishing gear (ALDFG) and (2) marine coatings (during fishing activities), (3) polymer pellets (during the production of plastic), (4) tire abrasion (occurred in transportation), and (5) plastic bungle (during end-of-life). ...
Article
Full-text available
Fisheries and aquaculture are becoming a focus of societal concern driven by globalization and increasing environmental degradation, mainly caused by climate change and marine litter. In response to this problem, the European Atlantic Area NEPTUNUS project aims to support and inform about the sustainability of the seafood sector, boosting the transition towards a circular economy through defining eco-innovation approaches and a steady methodology for eco-labelling products. This timely trans-regional European project proposes key corrective actions for positively influencing resource efficiency by addressing a life cycle thinking and involving all stakeholders in decision-making processes, arnessing the water-energy-seafood nexus. This paper presents inter-related objectives, methodologies and cues to action that will potentially meet these challenges that are aligned with many of the United Nations Sustainable Development Goals and European policy frameworks (e.g., Farm to Fork, European Green Dea
... As a starting point, the Plastic Leak Project (PLP) developed life cycle inventory guidelines to quantify plastic inventory flows based on industrial data and the expertise of public, private, and scientific organizations (Peano et al., 2020). Recently, these guidelines have been further refined for the specific case of plastic losses from seafood supply chains (Loubet et al., 2021). However, a methodology to assess the impacts of plastic emissions in life cycle impact assessment (LCIA) is still missing (Boucher et al., 2019;Loubet et al., 2021;Nessi et al., 2021;Peano et al., 2020). ...
... Recently, these guidelines have been further refined for the specific case of plastic losses from seafood supply chains (Loubet et al., 2021). However, a methodology to assess the impacts of plastic emissions in life cycle impact assessment (LCIA) is still missing (Boucher et al., 2019;Loubet et al., 2021;Nessi et al., 2021;Peano et al., 2020). This limits the application of LCA as a tool for highlighting potential trade-offs between different impact categories (e.g., ecotoxicity, human toxicity, global warming, and water scarcity) and the relative significance of their contribution to a specific area of protection (e.g., ecosystems quality, human health). ...
Article
To date, life cycle assessment (LCA) does not include a methodology for assessing the impacts of plastic litter leaked to the environment. This limits the applicability of LCA as a tool to compare the potential impacts of single‐use plastics and their alternatives on ecosystem quality and human health. As a contribution to tackle this issue, this work proposes simplified fate and characterization factors (CFs) for modeling the impacts of two types of microplastics—expanded polystyrene and tire and road wear particles—in the marine environment. In terms of fate mechanisms, this work explores different sedimentation, degradation, and fragmentation rate scenarios, based on literature values and expert estimates. Whereas the fate of expanded polystyrene is sensitive to the different fragmentation, degradation, and sedimentation scenarios, for tire and road wear particles the fate is primarily sensitive to sedimentation. The fate factors are integrated into CFs using an existing exposure and effect factor for microplastics in aquatic environments. Since the CFs of the two studied microplastics show important differences, these results reveal the need for developing polymer‐specific CFs. Finally, the CFs are tested in a case study of on‐the‐go food containers (one single‐use plastic, two compostable alternatives, and one reusable plate). Depending on the fate scenario, plastic litter impacts range from barely noticeable to more than doubling the total potential damage to ecosystem quality, compared to no plastic litter impact assessment. The high uncertainty of the results encourages further research on modeling microplastic fate and impacts in detail.
Article
Full-text available
Purpose The assessment of potential environmental impacts associated to mineral resource use in LCA is a highly debated topic. Most current impact assessment methods consider the extraction of resources as the issue of concern, while their dissipation is an emerging concept. This article proposes an approach to account for mineral resource dissipation in life cycle inventories (LCIs), with application to a case study. Methods The definition of mineral resources is first discussed considering both current main LCA practice and the context of resource dissipation. Secondly, the approach is described: considering a short-term perspective (25 years), any flow of resources to (i) environment, (ii) final waste disposal facilities, and (iii) products-in-use in the technosphere, with the resources not providing any significant function anymore (including due to non-functional recycling), is suggested to be reported as dissipative at the level of unit processes. This approach first requires to map the flows of mineral resources into and out of the unit processes under study (“resource flow analysis”), before identifying the dissipative flows and reporting them in LCI datasets. Results and discussion The approach is applied to analyze the direct dissipation of mineral resources along the primary production of copper, using Ecoinvent (v3.5) datasets. The production of 1 kg of copper cathode generates 0.88 kg of direct dissipative flows of resources (primarily calcium carbonate, copper, and to a lower extent iron), with important contributions of “tailings disposal,” “pyrometallurgy,” and “mining and concentration.” Moreover, this article discusses (i) how the developed approach would change the interpretation of results regarding mineral resources in LCA, (ii) how far some key methodological aspects of this approach (e.g., the temporal perspective) can affect the inventory results (e.g., in the case of the primary production of copper, considering a long-term perspective implies a significant shift in main contributions regarding both unit processes and resource flows), and finally (iii) the issue of new data requirements, in terms of availability and adequacy. Conclusions As demonstrated in the case study, existing LCI datasets and supporting documentation contain at least part of the data and information required to consistently compile the dissipative flows of resources at the unit process level, yet with the need for some complementary data and assessments. This approach may be particularly relevant to better support the development of more resource-efficient techniques or product designs. It is still open how to adapt characterization approaches to account for the impact induced by these resource dissipative flows.
Article
Full-text available
PurposePlastic pollution in marine environments is a severe problem in the world due to misuse and mismanagement of the materials. Microplastics are a specific form of pollutants in this context and its handling is very difficult due to its very small size of particulates. Currently, the impacts of marine plastic debris are not considered. However, this type of particulates can be assessed like other emissions with the systematic and quantifiable approach in life cycle assessment (LCA). It was our goal to find and test first methodological approaches for including impacts of marine litter of microplastics to LCA.Methods The Medellin Declaration on Marine Litter in Life Cycle Assessment and Management raised this issue in 2017 and called for LCA to address the challenges of marine litter. The present research paper focuses on how to integrate plastic debris impacts with focus on microplastics into LCA and gives a suggestion for an assessment approach. Based on a literature review, we considered various impacts to the marine environment of microplastics linked with their kinetics of the fragmentation and degradation. Subsequently, we developed a characterization LCA model for microplastics in the marine environment. We addressed therein the fate of microplastics and their specific eco-toxic effects to different organisms. We compared the impacts of different types of polymers as well and showed how these can be integrated in an assessment using the new characterization model.Results and discussionThe assessment of marine litter impacts in LCA was strongly dependent on the number of microplastic particles produced from the original litter over time. These impacts were derived from measurements of the number of microparticles, their densities in the marine environment and their impacts to different organisms. The new characterization model includes the relationship between fragmentation and degradation and can be used for impact assessments within LCA.Conclusion The question where we did not find a finally satisfying solution is the issue of the length of the time horizon of the assessment or the discounting. Those are regarded as subjective and are encountered with sensitivity or scenario analysis. Results from different time horizons can be aggregated to one figure or can be compared separately. Further investigations should be taken for a better understanding of this issue and for concrete solutions because their influence on the results of life cycle assessments is often fundamental.
Article
Full-text available
Abandoned, lost or otherwise discarded fishing gear (ALDFG) represents a significant, yet ultimately unknown amount of global marine debris, with serious environmental and socioeconomic impacts. This study reviews 68 publications from 1975 to 2017 that contain quantitative information about fishing gear losses. Gear loss estimates reported by the studies ranged widely, with all net studies reviewed reporting annual gear loss rates from 0% to 79.8%, all trap studies reporting gear loss rates from 0% to 88%, and all line studies reporting gear loss rates from 0.1% to 79.2%. Information obtained from this review was used to perform a meta‐analysis that provides the first synthetic, statistically robust estimates of global fishing gear losses. The meta‐analysis estimates global fishing gear losses for different major gear types. We estimate that 5.7% of all fishing nets, 8.6% of all traps, and 29% of all lines are lost around the world each year. Furthermore, we identified key gear characteristics, operational aspects and environmental contexts that influence gear loss. These estimates can be used to support sustainable fisheries development through informing risk assessments for fisheries and monitoring and assessment efforts to reduce gear losses.
Article
Full-text available
Plastic waste from anthropogenic activities is accumulating in the marine environment and poses a threat to marine biodiversity. Nevertheless, tools to assess the potential ecosystem damage from plastic waste are currently lacking from sustainability assessment approaches, such as life cycle assessment (LCA) methodologies. However, despite incomplete knowledge of the environmental mechanisms involved, the LCA community (researchers and practitioners) is calling for methodological developments to close this gap. We present a preliminary effect factor (EF) for working towards including the impacts of entanglement in plastic waste on marine biodiversity in life cycle assessment (LCA). Our preliminary EF modelling approach couples spatially-differentiated and taxon-specific estimates of the current fraction of species affected by entanglement with spatially-differentiated floating macroplastic density estimates. Our results indicate that the effect of macroplastic density on the fraction of species potential affected by entanglement is highest in areas with low estimated plastic density, most prominently the Southern Ocean and equatorial Pacific. However, in parameterising our approach, we discovered trade-offs between data source options, e.g. species coverage versus range extent accuracy. In addition, we identify knowledge gaps, e.g. defining species sensitivity effect thresholds to enable statistically relating pressure (density of floating marine macroplastic) with effect (the potentially affected fraction of species), and set out options for future methodological development for achieving quantification of an effect factor ready for incorporation in to a life cycle impact assessment modelling approach.
Article
Purpose Marine litter, mostly plastics, is a growing environmental problem. Environmental decision makers are beginning to take actions and implement regulations that aim to reduce plastic use and waste mismanagement. Nevertheless, life cycle assessment (LCA), a tool commonly used to assist environmental decision making, does not yet allow for considering the consequences of plastic waste leaked into the environment. This limits the application of LCA as a tool for highlighting potential tradeoffs between impact categories and the relative significance of their contribution on a specific Areas of Protection (AoP). A coordinated research effort to cover various parts of the marine litter impact pathway is required to ultimately produce characterisation factors that can cover this research gap. Here, we design a consistent and comprehensive framework for modelling plastic litter impact pathways in LCIA models. This framework is to support such coordinated research progress towards the development of harmonized pathways to account for impacts of plastic litter, specifically to the marine environment. The framework includes an overview of life cycle inventory requirements (leakage to the environment; a focus of other research efforts), and a detailed description of possible marine litter impact pathways, modelling approaches and data(-type) requirements. We focus on marine plastic litter and consider the potential contribution of different impact pathways to overall damage in the main operational AoPs, as well as recently proposed ones. Results and conclusions The proposed framework links inventory data in terms of kg plastic leaked to a specified environmental compartment (air, terrestrial, freshwater, marine) to six AoPs: ecosystem quality, human health, socio-economic assets, ecosystem services, natural heritage and cultural heritage. The fate modelling step, which includes transportation, fragmentation and degradation processes, is common to all included impact pathways. Exposure and effect modelling steps differentiate between at least six exposure pathways, e.g. inhalation, ingestion, entanglement, invasive species rafting, accumulation, and smothering, that potentially compromise sensitive receptors, such as ecosystems, humans, and manmade structures. The framework includes both existing, e.g. human toxicity and ecotoxicity, and proposed new impact categories, e.g. physical effect on biota, and can be used as a basis for coordinating harmonized research efforts.
Article
Plastic litter of all sizes has been acknowledged as a serious threat to biodiversity, especially in the marine environment. The fact that life cycle assessment (LCA) does not properly consider these issues is a serious problem for the aspirations of LCA in the public sphere. This paper focuses on micro‐ and nano‐sized plastics (MNPs), which have the potential to cause a substantial impact on ecosystem quality because of their increased presence in the marine compartment and capacity to affect a greater range of species. The data regarding MNPs’ effect on different aquatic species were extracted from the academic literature. These data were then explored and analyzed to bring to light the possibilities in terms of effect factor (EF) developments and the existing relations between effect on aquatic ecosystems and different parameters such as particle size, polymer type, and shape. No significant difference could be observed between the effect of the different subgroups of MNPs tested when considering a single species. However, when including many species in the analysis, differences could be noted between polystyrene (PS) and other polymer types. The high uncertainty on the developed EFs combined with this lack of statistical difference among subgroups at the single species level suggest that the use of a single generic EF could be appropriate for now. This EF is provided along with Species Sensitivity Distributions developed to allow for a quick visualization of the gathered data used to generate the EFs. This EF can now be used to quantify the physical impact of all MNPs in life cycle impact assessment.
Article
Life cycle assessment (LCA) has been widely applied in many different sectors, but the marine products and seafood segment have received relatively little attention in the past. In recent decades, global fish production experienced sustained growth and peaked at about 179 million tonnes in 2018. Consequently, increased interest in the environmental implications of fishery products along the supply chain, namely from capture to end of life, was recently experienced by society, industry and policy-makers. This timely review aims to describe the current framework of LCA and it’s application to the seafood sector that mainly focused on fish extraction and processing, but it also encompassed the remaining stages. An excess of 60 studies conducted over the last decade, along with some additional publications, were comprehensively reviewed; these focused on the main LCA methodological choices, including but not limited to, functional unit, system boundaries allocation methods and environmental indicators. The review identifies key recommendations on the progression of LCA for this increasingly important sustaining seafood sector. Specifically, these recommendations include (i) the need for specific indicators for fish-related activities, (ii) the target species and their geographical origin, (iii) knowledge and technology transfer and, (iv) the application and implementation of key recommendations from LCA research that will improve the accuracy of LCA models in this sector. Furthermore, the review comprises a section addressing previous and current challenges of the seafood sector. Wastewater treatment, ghost fishing or climate change, are also the objects of discussion together with advocating support for the water-energy-food nexus as a valuable tool to minimize environmental negativities and to frame successful synergies.
Article
The European seafood and aquaculture sectors are facing important challenges in terms of environmental threats (climate change, marine debris, resources depletion), social development (worker rights, consumer's awareness) or economic growth (market and nonmarket goods and services, global competitiveness). These issues are forcing all stakeholders, from policy-makers to citizens and industries, to move to more sustainable policies, practices and processes. Consequently, an improvement in collaborations among different parties and beyond borders is required to create more efficient networks along the supply chain of seafood and aquaculture sectors. To achieve this, a "nexus thinking" approach (i.e. the analysis of actions in connected systems) combined with a life cycle thinking appears as an excellent opportunity to facilitate the transition to a circular economy.
Article
With the increasing focus on marine plastic pollution, quantification of the environmental losses of plastics in the world, with differentiation into geographic regions, polymers and loss occurrences along the plastics value chains, is required. In this study, we make a global estimation of the losses of plastics to the environment across the entire plastic value chain, using existing literature and databases coupled with improved and additional methodological modelling of the losses. The resulting loss estimates are unprecedented in their detailed differentiations between polymers (23), plastic applications (13), geographical regions (11), and plastic value chain stages. Comprehensive sensitivity and uncertainty analyses were also conducted to identify key drivers in terms of plastic losses. We overall found that approximately 6.2 Mt (95% confidence interval, CI: 2.0–20.4 Mt) of macroplastics and 3.0 Mt (CI: 1.5–5.2 Mt) of microplastics were lost to the environment in 2015. The major macroplastic loss source was identified as the mismanaged municipal solid waste (MSW) management in low-income and lower-middle income countries (4.1 Mt). For microplastics, the major sources were abrasion of tyre rubbers, abrasion of road markings and plastics contributing to city dust generation. To curb marine plastic pollution, such quantified mapping as ours are needed to evaluate the magnitude of the plastics losses to environment from different sources and locations, and enable a further assessment of their environmental damage. Through our uncertainty and sensitivity analyses, we highlight plastics sources that should be prioritized in further research works to obtain a more comprehensive and accurate representation of global plastics losses.