Available via license: CC BY 4.0
Content may be subject to copyright.
Journal of Environmental Management 281 (2021) 111846
Available online 2 January 2021
0301-4797/© 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Research article
Environmental risk assessment of using antifouling paints on pleasure crafts
in European Union waters
Erik Ytreberg
a
,
*
, Maria Lagerstr¨
om
a
, Soa N¨
ou
a
, Ann-Kristin E. Wiklund
b
a
Department of Mechanics and Maritime Sciences, Chalmers University of Technology, SE 412 96, Gothenburg, Sweden
b
Department of Envx§ironmental Science, Stockholm University, SE-106 91, Stockholm, Sweden
ARTICLE INFO
Keywords:
Antifouling paints
Environmental risk assessment
Biocides
Coastal management
European union waters
Release rate methods
ABSTRACT
To ensure sustainable use of antifouling paints, the European Union have developed a new environmental risk
assessment tool, which a product must pass prior to its placement on the market. In this new tool, environmental
concentrations are predicted based on estimated release rates of biocides to the aquatic environment and risk
characterization ratios are calculated in regional spreadsheets. There are currently two methods in use to predict
release rates of biocides; a calculation method and a laboratory method. These methods have been believed to
overestimate environmental release of biocides and therefore xed correction factors to reduce the release rate
can be applied. An alternative method, known as the XRF method, has recently been developed and used to
derive eld release rates from antifouling paints. The aim of this study was to review the new environmental risk
assessment tool and assess how the choice of release rate method and application of correction factors impact the
approval of antifouling paint products. Eight coatings were environmentally risk assessed for usage in four
European marine regions; Baltic, Baltic Transition, Atlantic and Mediterranean; by applying release rates of
copper and zinc determined with the different methods. The results showed none of the coatings to pass the
environmental risk assessment in the Baltic, Baltic Transition and the Mediterranean if eld release rates were
used. In contrast, most of the coatings passed if the correction factors were applied on the release rates obtained
with the calculation or laboratory method. The results demonstrate the importance of release rate method choice
on the outcome of antifouling product approval in EU. To reduce the impact of antifouling paints on the marine
environment it is recommended that no correction factors should be allowed in the environmental risk assess-
ment or preferably that site-specic eld release rates are used. If the regulation in the European Union (and
elsewhere) continues to allow correction factors, the pressure of biocides to the environment from leisure boating
will result in degradation of marine ecosystems.
1. Introduction
An unprotected surface area immersed in seawater will within mi-
nutes to days be fouled by different organisms (Bixler and Bhushan,
2012). This so-called biofouling can be of considerable concern, pri-
marily for shipping and leisure boating, as it increases fuel consumption
and maintenance costs as well as shortens dry-docking intervals
(Davidson et al., 2020; Schultz et al., 2011). The most common strategy
to prevent biofouling is to coat the hull with antifouling paint that
contains and leaches biocides (Almeida et al., 2007; Amara et al., 2018).
Historically, many unsustainable biocides have been used in antifouling
paints including arsenic, lead, mercury and organotin compounds such
as tributyltin (TBT) (Antizar-Ladislao, 2008; Yebra et al., 2004). These
biocides were all efcient in preventing biofouling, but they also
impacted non-target organisms and created adverse effects on marine
ecosystems and have therefore been phased out from the antifouling
paint market (Miller et al., 2020; Thomas and Brooks, 2010).
Today, most antifouling paints are based on copper compounds, e.g.
cuprous oxide (Amara et al., 2018). Copper is an essential element for all
living organisms and play important roles in many metabolic processes
(Ochoa-Herrera et al., 2011). However, copper may also be toxic to most
species when concentrations exceed levels that are physiologically
required (Strivens et al., 2020). Due to anthropogenic activities such as
mining and smelting, municipal wastes, agricultural and industrial
emissions the concentration of copper in the environment has increased
(Morroni et al., 2019). In addition, emissions of copper from antifouling
* Corresponding author.
E-mail address: erik.ytreberg@chalmers.se (E. Ytreberg).
Contents lists available at ScienceDirect
Journal of Environmental Management
journal homepage: http://www.elsevier.com/locate/jenvman
https://doi.org/10.1016/j.jenvman.2020.111846
Received 14 October 2020; Received in revised form 10 December 2020; Accepted 13 December 2020
Journal of Environmental Management 281 (2021) 111846
2
paints have shown to impact water quality negatively. For example,
copper concentrations exceeding acute and chronic water quality have
been found in recreational boat marinas located in the US (Schiff et al.,
2004), Sweden (Kylin and Haglund, 2010), and Finland (Lagerstr¨
om
et al., 2020a). As a response, legislation has been passed in Washington
State that will ban the use of certain copper based products in both
freshwater and marine environments (Heine and Nestler, 2019). Cali-
fornia is also considering similar regulations and local restrictions are in
place in e.g. San Diego Bay (Miller et al., 2020).
In the European Union, a new harmonized product approval process
has been developed under the Biocidal Product Regulation (BPR,
Regulation (EU) 528/2012) requiring antifouling paints to pass an
environmental risk assessment (ERA) prior to their placement on the EU
market. In addition to the biocides, substances added into the paint
formulation labelled as environmental Substances of Concern (SoCs), e.
g. zinc oxide, shall be considered. In the ERA, environmental concen-
trations are modelled based on the estimated release rate of the biocides
and SoCs from the paint surface to the water. The modelling is per-
formed in a newly developed Excel calculation tool which automatically
generates predicted environmental concentrations (PECs) of the bio-
cides and SoCs in pleasure craft marinas within four European marine
regions (Baltic, Baltic Transition, Atlantic and Mediterranean). The PEC
values are subsequently divided with pre-dened predicted no effect
concentrations (PNECs) to produce risk characterization ratios (RCRs) of
the individual biocides/SoCs as well as cumulative RCSs if more than
one biocide/SOC is included in the product. If the RCR is less than unity
(<1), the concentration in the environment is likely to be lower than the
critical threshold value; the risk of adverse effects is considered low. If
the ratio is higher than unity (>1), risk for adverse effects exists and
actions to reduce the risk are recommended. This includes higher tier
renements where correction factors (CFs) can be applied to the biocidal
release rate (ECHA, 2017). For product approval, an efcacy assessment
of the product is also required where the applicant must demonstrate
that the product meet certain efcacy criteria in preventing biofouling.
This antifouling paint product approval scheme under the EU BPR is
shown in Fig. 1.
Since the outcome of the ERA depends on the release rates of the
biocides used in the paint formulation, it is important that the submitted
release rates accurately predict environmental release in the relevant
region of intended use. Currently, two standardized release rates
methods are recommended for risk assessments in EU. The rst is a
laboratory “rotating cylinder” method where release rates are deter-
mined in articial seawater (ASTM, 2005) and the second one is a
calculation method (ISO, 2010). Besides these methods, a new method
allowing for determination of eld release rates using an X-Ray Fluo-
rescence spectrometer (XRF) has been developed by Ytreberg et al.
(2017) and further modied by Lagerstr¨
om et al. (2018) and Lagerstr¨
om
and Ytreberg (2020).
The main aim of the current study was to review the new ERA
calculation tools for antifouling paints developed under the EU BPR and
assess how different release rate methods (the rotating cylinder method,
the calculation method and the XRF method) impact the authorization of
antifouling products in the EU. A second aim was to investigate differ-
ences in environmental sensitivity in terms of acceptable biocidal input
between the different marine regions and how that will affect the
product approval and impact the antifouling paint market. Finally, the
highest acceptable release rates of copper and zinc from antifouling
paints were determined in the different marine regions. Eight commer-
cial coatings were used in the current study. The PECs of copper and zinc
in different pleasure craft marinas located in the marine regions Baltic,
Baltic Transition, Atlantic and Mediterranean were modelled using
release rates obtained from the three release rate methods. Risk char-
acterization ratios were subsequently calculated to determine if usage of
the coatings would pose an unacceptable risk for the marine environ-
ment depending on the choice of release rate method.
2. Materials and methods
2.1. Study area and modelling tool
MAMPEC (Marine Antifouling Model to Predict Environmental
Concentrations) is a widely accepted model used in regulatory purposes
worldwide to predict environmental concentrations of biocides in the
marine environment. The model is used by several EU member states to
Fig. 1. Product approval process for antifouling paints under the European Union Biocidal Products Regulation (BPR, Regulation (EU) 528/2012).
E. Ytreberg et al.
Journal of Environmental Management 281 (2021) 111846
3
determine PEC-values of biocides in different environments such as
harbors, marinas and ship lanes. In MAMPEC, several scenarios exist,
including two OECD scenarios representing a typical European com-
mercial harbor and a pleasure craft marina, respectively. However, as
these OECD scenarios do not reect environments with low tidal dif-
ferences, e.g. the Baltic Sea, several countries have developed their own
scenarios to be used for the ERA of antifouling products, while other
countries have approved all products irrespectively of the biocidal
release rate, as long as the biocides used are approved by the EU.
Therefore, the antifouling paint market differs substantially between EU
countries. In order to harmonize the regulation and to capture the wide
range of conditions across EU waters, the Environment Working Group
of the Biocidal Product Committee developed an agreed set of pleasure
craft scenarios (ECHA, 2017) representative of the marine regions of the
Atlantic (47 marinas), Mediterranean (46 marinas), Baltic Transition (17
marinas) and Baltic Sea (38 marinas) (Fig. 2). The purpose of this
harmonization is to allow for Mutual Recognition under the BPR as the
ERA is performed by region, rather than by country. For the product
approval applications, the applicant lls in the release rates of the
different substances into newly developed Excel calculation tools which
automatically generate PECs and RCRs for all individual marinas within
a region. The Excel workbooks, one for each approved substance, can be
accessed via the PT21 Emission Scenario Document pages on the ECHA
website (ECHA, 2020). The PECs and RCRs are produced for the dis-
solved and particulate fractions of the substance, both inside the marina
and its surrounding environment. For the risk characterization, the
dissolved concentration inside the marina is proposed to be used. The
90th percentile concentration, based on the PECs determined for each
individual pleasure craft marina, is used to calculate an overall RCR per
region. Thereby, the applicant directly receives information about the
risk characterization of the product in the different regions. If more than
one biocide or SoCs is used in the product, a PEC/PNEC summation
approach shall be used in a separately provided Excel calculation tool to
determine the cumulative RCR (ECHA, 2017).
Fig. 2. Distribution of pleasure craft marinas, per marine region, included in the Excel calculation tool. The three exposure sites (Nyn¨
ashamn, Baltic Sea; Malm¨
o, the
Sound and Kristineberg, Kattegat) used by Lagerstr¨
om et al. (2020b) to measure eld release rates are also shown.
E. Ytreberg et al.
Journal of Environmental Management 281 (2021) 111846
4
2.2. Methods for release rate estimation
Two methods are currently approved to be used for release rate
determination in the EU, a mass-balance calculation method developed
by CEPE (EU, 2006), from now on referred to as the “calculation
method” and a rotating cylinder laboratory method, from now on
referred to as the “rotating cylinder method”. In addition, two in situ
release rate methods exists: the “Dome method” developed by the US
Navy (Finnie, 2006; Valkirs et al., 2003), and the “XRF method” rst
developed by Ytreberg et al. (2017) and later modied by Lagerstr¨
om
et al. (2018) and Lagerstr¨
om and Ytreberg (2020). A detailed description
of the different release rate methods is shown in Supporting Material.
If the coating fails to pass the initial Tier 1 assessment, correction
factors (CFs) of 2.9 (calculation method) and 5.4 (rotating cylinder
method) can be used through simple division (EU, 2006). These cor-
rected release rates can be used for a further Tier 2 assessment. The CFs
were originally proposed by Finnie (2006) as the calculation and
rotating cylinder method have shown to overestimate environmental
release rates of copper determined in eld. However, the proposed CFs
are based on eld data from one coating only and the applicability of a
universal CF to predict environmental release rates of biocides for any
type of coating has been questioned (Lagerstr¨
om et al., 2018). In addi-
tion, neither the rotating cylinder nor the calculation method and the
CFs consider changes in environmental parameters known to govern
leaching of biocides such as salinity, pH and temperature (Lagerstr¨
om
et al., 2020a, 2020b; Ytreberg et al., 2017).
2.3. Antifouling coatings investigated and release rates of copper and zinc
The release rates of copper and zinc from eight antifouling coatings
were used in the study. The coatings are authorized to be used on
pleasure crafts in Sweden and were selected based on their Cu
2
O con-
centrations to achieve a wide range (6.1–31.9% Cu
2
O) (Table 1). In
Sweden, regional restrictions exist for antifouling paints where stricter
regulation is applied for coatings used in the Baltic Sea as compared to
the Swedish west coast. Consequently, different antifouling paints exist
for the Baltic Sea market as compared to the Swedish west coast. The
release rates of copper and zinc, determined using the calculation
method or the rotating cylinder method, were obtained from the product
authorization reports submitted to the Swedish Chemicals Agency
(Table 1). Five of the coatings were authorized to be used on the Swedish
West coast only, i.e. in the Baltic transition region, while the other three
were authorized to be used both in the Baltic Sea and on the Swedish
West coast. Therefore, release rates are available for all eight coatings in
the Baltic transition. As ve of the coatings were environmentally risk
assessed for exposure in the Baltic transition only, release rates for
exposure in the Baltic Sea were not submitted during the product
authorization.
Site-specic eld release rates from the eight coatings determined
with the XRF method were obtained from Lagerstr¨
om et al. (2020b) and
used for assessment against the release rates submitted in the product
authorization reports (Table 1). These average daily in situ release rates
between day 14 and 56 of exposure were determined during the summer
season of 2018 at three sites along the Swedish coast: Nyn¨
ashamn,
(Baltic Sea, salinity 6.4 PSU), Malm¨
o (the Sound, Baltic transition,
salinity 7.5 PSU) and Kristineberg Marine Research and Innovation
Centre (Kattegat, Baltic transition, salinity 27 PSU) (Table 1). The
exposure sites are shown in Fig. 2.
2.4. Exposure assessment and risk characterization
The release rates used in the application for product approval (Tier 1
and 2) as well as the site-specic eld release rates determined with the
XRF method were used in the new Excel calculation tool to environ-
mentally risk assess the different coatings for use on pleasure crafts in
the four marine regions. Noteworthy, zinc was not included in the
Table 1
Tier 1 and Tier 2 release rates for Cu and Zn, as submitted to the Swedish Chemicals Agency for the risk assessment of the products, and site specically eld release rates using the XRF method obtained from Lagerstr¨
om
et al. (2020b).
Paint Area of use Product Manufacturer Cu
2
O (wt %,
ww)
ZnO (wt %,
ww)
Baltic Sea Baltic Transition
Tier 1 Tier 2 Field,
Baltic Sea
Tier 1 Tier 2 Field,
Oresund
Field,
Kattegat
Cu Zn Cu Zn Cu Zn Cu Zn Cu Zn Cu Zn Cu Zn
A Baltic Sea and Baltic
Transition
Lefant Nautica
Copper
(C)
Lefant 7.0 20 – 100 3.0
a
- 1.03
a
- 1.9 8.2 3.0
a
- 1.03
a
- 1.9 7.7 5.4 16.5
B Baltic Sea and Baltic
Transition
Mille Light Copper
(C)
International 6.1 10 – 25 3.0
a
4.0 0.97
a
1.32
a
2.2 5.8 3.0
a
4.0
a
0.97
a
1.32
a
3.9 5.4 7.1 6.3
C Baltic Sea and Baltic
Transition
Cruiser One
(C)
Jotun 8.5 2.5 – 25 2.5
a
5.8
a
0.9
a
2.0
a
4.4 6.4 2.5
a
5.8
a
0.9
a
2.0
a
3.6 5.9 4.4 10.1
D Baltic Transition VC17m
(C)
Hempel 17.96
c
- N/
A
N/
A
N/A N/A 8.0 0 15.11
a
- 5.21
a
- 7.9 0 13.3 0
E Baltic Transition Racing VK
(C)
Hempel 22.02 10 – 25 N/
A
N/
A
N/A N/A 7.0 9.4 13.0
a
- 4.48
a
- 7.2 6.7 18.9 10.3
F Baltic Transition Hard Racing Xtra
(R)
International 33.1 10 – 25 N/
A
N/
A
N/A N/A 8.1 6.2 32.0
b
- 6.2
b
- 6.9 4.4 16.7 4.2
G Baltic Transition Biltema Antifouling
(R)
Biltema 13 20 – 25 N/
A
N/
A
N/A N/A 5.0 7.7 12.8
b
- 2.37
b
- 6.3 5.5 12.1 9.7
H Baltic Transition Micron Superior
(C)
International 31.93 2.5 – 25 N/
A
N/
A
N/A N/A 7.8 2.7 12.44
a
3.66
a
4.29
a
1.26
a
12.0 2.9 27.5 4.1
N/A denotes release rates not available in the product approval since the products were not risk assessed for usage in the Baltic Sea
- denotes Zn not to be included in the product approval and therefore release rates were not available
a
denotes release rates calculated with the CEPE mass-balance method.
b
denotes release rates obtained using the Rotating Cylinder method.
c
denotes Cu powder
E. Ytreberg et al.
Journal of Environmental Management 281 (2021) 111846
5
application for product approval for paints A, D, E, F and G. However,
zinc was shown to be released from four of these coatings (A, E, F and G)
when measured with XRF (Table 1). This will impact the outcome of the
risk assessment, as the exposure assessment using the eld derived
release rates considers both Cu and Zn for all coatings (except paint D
which did not contain zinc) while that using Tier 1 and Tier 2 release
rates from product approval for paints A, D, E, F and G is based on
copper only.
The release rates obtained with the rotating cylinder (Product F and
G) are applicable to all marine waters and would hence be used in the
ERA for all four marine regions. However, the release rates obtained
with the calculation method (Product A–E and H) are derived based on
the recommended paint lm thickness and lifetime of the coating (see
equation 1 and 2) and reects emissions during usage recommendations
for the regions Baltic Sea and Baltic Transition (Table 1). It is unknown if
other dry paint lm thicknesses would have been recommended for the
Atlantic and Mediterranean market which would have increased or
decreased the release rates and impacted the result of the ERA. There-
fore, the assessment in the Atlantic and Mediterranean was performed to
explore if the coatings theoretically would pass the ERA or not.
For the Baltic Sea region, eld determined release rates obtained in
Nyn¨
ashamn were used. For the Baltic transition region, release rates
from both the Sound and Kattegat were used. For the Atlantic and
Mediterranean, no site-specic led release rates were available. How-
ever, as the salinity at the Kattegat exposure site (27 PSU) is close to
what is expected in most Atlantic and Mediterranean marinas (35 PSU),
the eld release rates obtained at Kattegat were also used for the ERA of
the Atlantic and Mediterranean regions.
The application factors (the fraction of pleasure crafts assumed to use
the product), PNEC-values and background concentrations are already
xed in the Excel calculation tool (Table 2) and the release rates of
biocides are the only parameters required for the tool to derive PECs and
RCRs. The 90th percentile concentration, based on the PECs determined
for each individual pleasure craft marina, is used to calculate PEC/
PNEC-ratios per region. When the coatings contain more than 1
biocide and/or SoC a PEC/PNEC summation approach was used to
calculate the cumulative risk of the product (ΣRCR).
3. Results and discussion
3.1. ERA in the marine regions
The three coatings approved for use on pleasure crafts in the Swedish
Baltic Sea (Paint A, B and C) all failed the risk assessment when Tier 1
release rates of the standardized methods were used in the Baltic Sea
region (Fig. 3). Tier 2 release rates showed ΣRCR <1 for Paint A and
Paint B, while Paint C had a ΣRCR of 1.05. Although not intended for use
in the Baltic Sea, the remaining ve coatings also fail the ERA using Tier
1 release rates. Only one coating (Paint G, ΣRCR =0.99) would pass in a
Tier 2 assessment, but as zinc was not included in the product applica-
tion of this paint (as well as those of paints A, D, E and F) its exposure
assessment is lacking given that the XRF method showed zinc to indeed
be released from all coatings except paint D (Table 1). Only one coating
(paint B) out of the three coatings passing the Tier 2 assessment (paints
A, B and G) included zinc in the product approval. When Baltic Sea eld
release rates were used, all eight coatings showed ΣRCRs above 2, thus
failing the ERA. Notably, the PEC/PNEC ratios of zinc alone was >1 for
all paints except paint H (0.56) and paint D which did not contain any
zinc (Supporting Material Table S1), highlighting the importance of
including SoCs in the ERA of antifouling products.
In the Baltic Transition, only one coating (Paint A) had ΣRCR ≤1
when Tier 1 release rates were used whereas the Tier 2 assessment
showed half the coatings (paints A, B, C and G) to obtain ΣRCRs ≤1
(Fig. 3 and Supporting Material Table S2). The remaining four coatings
displayed ΣRCRs between 1.2 and 1.5. All coatings posed an unaccept-
able risk to the marine environment when eld release rates obtained in
the Sound were used (ΣRCR =1.8–2.9). The ΣRCRs were even higher
when the Kattegat-derived release rates were used (ΣRCR =2.5–5.6).
The higher ΣRCRs for Kattegat are a result of the higher release rates
obtained in the more saline Kattegat (27 PSU) as compared to the Sound
(7.5 PSU).
All coatings except coating F showed ΣRCRs <1 when Tier 1 release
rates were used in the ERA for the Atlantic. When Tier 2 release rates
were used, all coatings showed ΣRCRs well below 1 (around 0.5) (Fig. 3
and Supporting Material Table S3). All coatings except paints E and H
would pass the ERA if eld derived release rates were used. In the
Mediterranean, only the paints designed for the Swedish Baltic Sea
(Paint A, B and C) pass the assessment using Tier 1 release rates, while
all eight coatings showed ΣRCR <1 when Tier 2 release rates were used
(Fig. 3 and Supporting Material Table S4). The results showed no coat-
ings to pass the ERA in the Mediterranean when eld derived release
rates were used.
3.2. Acceptable environmental risk
The results from the ERA show all paints to pose an unacceptable risk
to the environment in the Baltic, Baltic Transition and the Mediterra-
nean when region-specic eld release rates were used. As such, the
antifouling paints assessed here would need to reduce the release rates
of copper and/or zinc in order to be approved for these regions. The
maximum allowable release rate combinations of copper and zinc to
pass the ERA, dened as ΣRCR =1, in the four marine regions were
derived from the Excel calculation tool and are shown in Fig. 4. The
differences in the extent of acceptable releases between marine regions,
i.e. the positioning of the lines representing ΣRCR =1, reect differ-
ences in environmental sensitivity in terms of biocidal input. For
example, acceptable risks for the Baltic marinas can only be obtained if
the release rate of copper is ≤2.4
μ
g/cm
2
/d (assuming no zinc to be
present in a coating). The corresponding maximum allowable release
rates of copper in the Baltic Transition, Mediterranean and the Atlantic
are; 3.4
μ
g/cm
2
/d, 9.0
μ
g/cm
2
/d and 21.4
μ
g/cm
2
/d, respectively. The
differences can be explained by differences in water exchange capacity
where the average tidal difference for the pleasure craft set used for the
Baltic Sea is 0.2 m as compared to the higher average tidal differences in
Mediterranean (0.53 m) and the Atlantic (2.97 m) (Shan-I et al., 2013).
3.3. Implications for the antifouling paint market and its regulation
Antifouling paints are considered effective if static eld tests show a
surface coverage of macrofouling <25% (European Chemical Agency,
2018). However, as macrofouling on ship and pleasure boat hulls
translates to severe economic costs due to fuel penalties (Schultz et al.,
2011), the critical release rate of copper (RR
crit
) to prevent the attach-
ment of macrofouling (dened as 0% surface coverage of macrofouling)
is a more accurate efcacy parameter for the boating and the shipping
sector. The antifouling efcacy and RR
crit
of copper has been investi-
gated by Lindgren et al. (2018) where a generic biocide and zinc free
coating was spiked with cuprous oxide. The release rates of copper were
determined with the XRF method during exposure time d14-d56. The
RR
crit
of copper was determined based on a number of experimental
Table 2
Predicted no-effect concentrations (PNEC), background concentrations and
application factor used for the different regions in the Excel calculation tool.
Region PNEC (
μ
g/L) Background
concentrations
Application
factor
Cu Zn Cu Zn Cu Zn
Atlantic 2.6 3.4 1.1 0 0.95 0.9
Mediterranean 2.6 3.4 1.1 0 0.95 0.9
Baltic Transition 2.6 3.4 1.1 0 0.95 0.9
Baltic 2.6 3.4 1.1 0 0.95 0.9
E. Ytreberg et al.
Journal of Environmental Management 281 (2021) 111846
6
coatings containing increasing cuprous oxide concentrations that were
exposed in a pleasure craft marina in Gothenburg, i.e. in the region
Baltic Transition (Lindgren et al., 2018). All investigated copper coat-
ings were efcient in preventing macrofouling and the RR
crit
of copper
was determined to be ≤4.7
μ
g/cm
2
/d, indicating that coatings with
even lower release rates of copper may be sufcient in preventing
macrofouling. Hence, low-leaching efcient coatings could pass the ERA
in the Baltic Transition where the maximum allowable release rate is 3.4
μ
g/cm
2
/d (Fig. 4).
RR
crit
of copper in the Baltic Sea and Baltic Transition can also be
estimated from the studies by Lagerstr¨
om et al. (2020b) and Lagerstr¨
om
et al. (2018) where commercial coatings were used. The RR
crit
of copper
from these studies, determined using the XRF method, holds a higher
uncertainty since all the coatings contained and released zinc which also
may impact the coatings antifouling performance. Fig. 5 gives a
geographical overview of the RR
crit
estimates and the acceptable release
rates in the two regions. The RR
crit
of copper in the Baltic is lower than
the acceptable release rate to pass the ERA indicating that antifouling
coatings with a substantial lower release rate of copper, than what is
currently available on the market, could be developed without
compromising the requirements of antifouling efcacy. For the Baltic
Transition, the situation is more complex where the RR
crit
is lower than
the acceptable release rate in the Sound, while the northern parts of the
region show RR
crit
to be higher than the acceptable release rate. Hence,
in order to pass the ERA in the Baltic Transition, coatings must have a
lower release rate of copper than expected to be required to deter
macrofouling in the northern part of the region.
The eld determined release rates from the eight coatings studied
here are not only typically above the maximum allowable but also
sometimes greatly in excess of RRcrit. As shown in Lagerstr¨
om et al.
(2020b), copper emission could be reduced by as much as 80% for some
paints without any loss in efciency. The large discrepancy between the
copper release rate of products currently on the market and release rates
needed to deter fouling. This is likely because release rates derived using
inappropriate methods and correction factors have, thus far, been ad-
missible in the ERA. As demonstrated here, application of correction
factors in the Tier 2 assessment nearly always results in a product
passing the ERA. No longer permitting the use of corrections factors
would incentivize the development of lower-leaching, more sustainable
coatings as well as promote the development of biocide free alternatives
such as silicone coatings which currently only hold a small percentage of
the market.
To ensure sustainable management of our coastal environments it is
crucial that we understand the links between human activities, the
pressure they pose on the environment and how that pressure may
impact the environment and human welfare (Elliott et al., 2017; Elliott
and O’Higgins, 2020). Thus, it is important that environmental release
rates of biocides from antifouling paints are used to predict impacts on
Fig. 3. Sum of Risk Characterization Ratios (ΣRCRs) of copper (Cu) and zinc (Zn) for pain A – H depending on release rate method (Tier 1 and Tier 2 release rate
submitted during product approval or region-specic eld release rates determined with the XRF method) and in what marine region (Baltic, Baltic Transition,
Atlantic and Mediterranean) the paints shall be used. For Paint D, that did not contain any Zn, the RCR is based on Cu release rate only for all release rate methods.
The red line shows ΣRCR =1. * denotes RCRs are based on Cu release rates only as release rates of Zn was not included in ERA submitted for product approval. . (For
interpretation of the references to colour in this gure legend, the reader is referred to the Web version of this article.)
E. Ytreberg et al.
Journal of Environmental Management 281 (2021) 111846
7
the environment. In a study by Lagerstr¨
om et al. (2020a) a comparison
between modelled (using MAMPEC) and measured environmental con-
centrations of copper inside a marine in Sweden was performed. The
study used release rates submitted for product approval (determined
with the calculation method and Tier 2) and site-specic eld release
rates determined with the XRF method. The study showed good agree-
ment between measured concentrations and the modelled concentration
when the eld derived release rates were used. However, when the
release rates from the products’ authorization applications were used,
the modelled concentration were 2-fold lower than the measured con-
centrations. This suggest the XRF method to accurately predict envi-
ronmental concentrations while the calculation method (using Tier 2)
underestimated the load of copper. Hence, if the regulation in EU (and
elsewhere) continue to allow Tier 2 release rates, the pressure of copper
to the environment from leisure boating will result in degradation of
marine ecosystems. Reduction of copper emissions is also required to
improve the environmental status of our oceans, seas and coasts. For
example, the latest environmental status assessment of Swedish coastal
water bodies showed 27% of the assessed water bodies not to full the
requirements for good ecological status with respect to copper, i.e. they
displayed copper concentrations in surface seawater or in sediment
exceeding the environmental quality standard (WISS, 2020). Similar
patterns have been reported for other coastal areas, i.e. Burant et al.
(2019) showed 51% of water samples from Californian saltwater ma-
rinas to exceed the chronic environmental quality standard. This further
motivates the need to accurately determine the load of copper from
different anthropogenic activities, including emissions from antifouling
paints.
4. Conclusions
This study shows that the release rate method chosen in the ERA of
antifouling products will have a large impact on the estimated pressure
of biocides as well as the outcome of the ERA. If site-specic release rates
determined in the eld with the XRF method are used, none of the eight
products assessed in the current study would pass the ERA in the Baltic,
the Baltic Transition or in the Mediterranean. However, most of the
coatings would pass the ERA if Tier 2 release rates obtained using the
calculation method or the laboratory method are used. Ideally, it is
recommended that site-specic eld release rates are used to estimate
pressures of biocides in future environmental risk assessments of anti-
fouling products, but as a rst step, Tier 2 correction factors should no
longer be permitted when using existing standardized methods.
Increasing the accuracy of the predicted pressure of biocides, would act
to ensure sustainable leisure boating in European coastal waters in the
future. Since copper pollution is a worldwide problem, particularly in
semi-enclosed pleasure craft marinas, the recommendation to use eld
specic release rates in environmental risk assessments also applies to
other regions and countries outside of EU.
Credit author statement
Erik Ytreberg: Writing – original draft, Conceptualization, Formal
analysis, Methodology, Maria Lagerstr¨
om: Writing – review & editing,
Conceptualization, Visualization, Formal analysis Soa N¨
ou: Investiga-
tion, Writing – review & editing Ann-Kristin E. Wiklund: Writing – re-
view & editing, Funding acquisition, Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Fig. 4. Maximum allowable release rates of copper (Cu) and zinc (Zn) to obtain
a Sum of Risk Characterization Ratios (ΣRCRs) equal 1 in the different marine
regions. Areas above line denotes unacceptable risks, i.e. ΣRCRs >1 while areas
below the line denotes acceptable risks, i.e. ΣRCRs <1.
Fig. 5. Maximum acceptable release rates of copper in the Baltic Sea (green)
and the Baltic Transition (yellow) for a product to pass the environmental risk
assessment (ΣRCR ≤1) and critical Cu release rates (RR
crit
) at ve locations
along the Swedish coast obtained from Lagerstr¨
om et al. (2020b) (a), Lindgren
et al. (2018) (b) and (Lagerstr¨
om et al., 2018) (c). (For interpretation of the
references to colour in this gure legend, the reader is referred to the Web
version of this article.)
E. Ytreberg et al.
Journal of Environmental Management 281 (2021) 111846
8
Acknowledgements
This study was funded by the Swedish Agency for Marine and Water
Management (SwAM) and the foundation BalticSea 2020.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jenvman.2020.111846.
References
Almeida, E., Diamantino, T.C., de Sousa, O., 2007. Marine paints: the particular case of
antifouling paints. Prog. Org. Coating 59, 2–20.
Amara, I., Miled, W., Slama, R.B., Ladhari, N., 2018. Antifouling processes and toxicity
effects of antifouling paints on marine environment. A review. Environmental
Toxicology and Pharmacology 57, 115–130.
Antizar-Ladislao, B., 2008. Environmental levels, toxicity and human exposure to
tributyltin (TBT)-contaminated marine environment. A review. Environment
International 34, 292–308.
ASTM, 2005. Standard Test Method for Determination of Copper Release Rate from
Antifouling Coatings in Substitute Ocean Water. ASTM Method D 6442-05, p. 9.
Bixler, G.D., Bhushan, B., 2012. Biofouling: lessons from nature. Phil. Trans. Math. Phys.
Eng. Sci. 370, 2381–2417.
Burant, A., Zhang, X., Singhasemanon, N., 2019. Antifouling Paint Biocide Monitoring
and Modeling to Support Mitigation, Pesticides in Surface Water: Monitoring,
Modeling, Risk Assessment, and Management. American Chemical Society,
pp. 491–517.
Davidson, I.C., Smith, G., Ashton, G.V., Ruiz, G.M., Scianni, C., 2020. An experimental
test of stationary lay-up periods and simulated transit on biofouling accumulation
and transfer on ships. Biofouling 36, 455–466.
ECHA, 2017. PT 21 Product Authorisation Manual (Environmental Risk Assessment).
Available at:. https://echa.europa.eu/sv/guidance-documents/guidance-on-biocid
es-legislation/emission-scenario-documents.
ECHA, 2020. Accessed 2020-09-14. https://echa.europa.eu/sv/guidance-documents/g
uidance-on-biocides-legislation/emission-scenario-documents.
Elliott, M., Burdon, D., Atkins, J.P., Borja, A., Cormier, R., de Jonge, V.N., Turner, R.K.,
2017. “And DPSIR begat DAPSI(W)R(M)!” - a unifying framework for marine
environmental management. Mar. Pollut. Bull. 118, 27–40.
Elliott, M., O’Higgins, T.G., 2020. From DPSIR the DAPSI(W)R(M) Emerges… a Buttery
– ‘protecting the natural stuff and delivering the human stuff’. In: O’Higgins, T.G.,
Lago, M., DeWitt, T.H. (Eds.), Ecosystem-Based Management, Ecosystem Services
and Aquatic Biodiversity : Theory, Tools and Applications. Springer International
Publishing, Cham, pp. 61–86.
EU, 2006. HARMONISATION of LEACHING RATE DETERMINATION for ANTIFOULING
PRODUCTS under the BIOCIDAL PRODUCTS DIRECTIVE. Workshop Report
Endorsed at the 26th Meeting of Representatives of Members States Competent
Authorities for the Implementation of Directive 98/8/EC Concerning the Placing of
Biocidal Products on the Market (11-14 September 2007). Workshop Report Ispra,
Italy, 12 December 2006.
Finnie, A.A., 2006. Improved estimates of environmental copper release rates from
antifouling products. Biofouling 22, 279–291.
Heine, L., Nestler, A., 2019. Promising practices for alternatives assessment: lessons from
a case study of copper-free antifouling coatings. Integrated Environ. Assess. Manag.
15, 867–879.
ISO, 2010. Paints and Varnishes — Modelling of Biocide Release Rate from Antifouling
Paints by Mass-Balance Calculation. International Standards ISO 10890, p. 2010.
Kylin, H., Haglund, K., 2010. Screening of antifouling biocides around a pleasure boat
marina in the Baltic Sea after legal restrictions. Bull. Environ. Contam. Toxicol. 85,
402–406.
Lagerstr¨
om, M., Ferreira, J., Ytreberg, E., Eriksson-Wiklund, A.-K., 2020a. Flawed risk
assessment of antifouling paints leads to exceedance of guideline values in Baltic Sea
marinas. Environ. Sci. Pollut. Control Ser. 27, 27674–27687.
Lagerstr¨
om, M., Lindgren, J.F., Holmqvist, A., Dahlstrom, M., Ytreberg, E., 2018. In situ
release rates of Cu and Zn from commercial antifouling paints at different salinities.
Mar. Pollut. Bull. 127, 289–296.
Lagerstr¨
om, M., Ytreberg, E., 2020. Quantication of Cu and Zn in Antifouling Paint
Films by XRF. Talanta, p. 121820.
Lagerstr¨
om, M., Ytreberg, E., Wiklund, A.-K.E., Granhag, L., 2020b. Antifouling paints
leach copper in excess – study of metal release rates and efcacy along a salinity
gradient. Water Res. 186, 116383.
Lindgren, J.F., Ytreberg, E., Holmqvist, A., Dahlstr¨
om, M., Dahl, P., Berglin, M.,
Wrange, A.-L., Dahlstr¨
om, M., 2018. Copper release rate needed to inhibit fouling on
the west coast of Sweden and control of copper release using zinc oxide. Biofouling
34, 453–463.
Miller, R.J., Adeleye, A.S., Page, H.M., Kui, L., Lenihan, H.S., Keller, A.A., 2020. Nano
and traditional copper and zinc antifouling coatings: metal release and impact on
marine sessile invertebrate communities. J. Nanoparticle Res. 22, 129.
Morroni, L., Sartori, D., Costantini, M., Genovesi, L., Magliocco, T., Ruocco, N.,
Buttino, I., 2019. First molecular evidence of the toxicogenetic effects of copper on
sea urchin Paracentrotus lividus embryo development. Water Res. 160, 415–423.
Ochoa-Herrera, V., Le´
on, G., Banihani, Q., Field, J.A., Sierra-Alvarez, R., 2011. Toxicity
of copper(II) ions to microorganisms in biological wastewater treatment systems. Sci.
Total Environ. 412–413, 380–385.
Schiff, K., Diehl, D., Valkirs, A., 2004. Copper emissions from antifouling paint on
recreational vessels. Mar. Pollut. Bull. 48, 371–377.
Schultz, M.P., Bendick, J.A., Holm, E.R., Hertel, W.M., 2011. Economic impact of
biofouling on a naval surface ship. Biofouling 27, 87–98.
Shan-I, C., Thomason, J., Prowse, G., 2013. Dening Typical Regional Pleasure Craft
Marinas in the EU for Use in Environmental Risk Assessment of Antifouling Products,
Available at:. https://echa.europa.eu/documents/10162/16908203/pt21_regiona
l_marina_scenario_study_en.pdf/be35dd48-86ca-4e30-b6ac-425ed5ffd5a4.
Strivens, J., Hayman, N., Rosen, G., Myers-Pigg, A., 2020. Toward validation of
toxicological interpretation of diffusive gradients in thin lms in marine waters
impacted by copper. Environ. Toxicol. Chem. 39, 873–881.
Thomas, K.V., Brooks, S., 2010. The environmental fate and effects of antifouling paint
biocides. Biofouling 26, 73–88.
Valkirs, A.O., Seligman, P.F., Haslbeck, E., Caso, J.S., 2003. Measurement of copper
release rates from antifouling paint under laboratory and in situ conditions:
implications for loading estimation to marine water bodies. Mar. Pollut. Bull. 46,
763–779.
WISS, 2020. Water Information System Sweden, a Database Developed by the Competent
Authorities of the Swedish Water Districts, the County Administrative Boards and the
Swedish Agency for Marine and Water Management. Accessed: 2020-06-23. https
://viss.lansstyrelsen.se/.
Yebra, D.M., Kiil, S., Dam-Johansen, K., 2004. Antifouling technology—past, present and
future steps towards efcient and environmentally friendly antifouling coatings.
Prog. Org. Coating 50, 75–104.
Ytreberg, E., Lagerstr¨
om, M., Holmqvist, A., Eklund, B., Elwing, H., Dahlstrom, M.,
Dahl, P., Dahlstrom, M., 2017. A novel XRF method to measure environmental
release of copper and zinc from antifouling paints. Environ. Pollut. 225, 490–496.
E. Ytreberg et al.
