Chemical status classification in biota Illustrative examples on the practice of status assessments based on environmental target levels

(1)

Chemical status classification in biota

Illustrative examples on the practice of status assessments based on environmental target levels

Elin Boalt, Caroline Ek, and Anders Bignert

Rapport nr 6:2011

Naturhistoriska Riksmuseet Enheten för miljögiftsforskning Box 50 007

104 05 Stockholm

1

(2)

Introduction ... 3

Aim of study ... 3

Target levels for chemical status assessment in biota ... 3

Scope and definitions ... 3

Internationally agreed target levels... 4

Environmental Quality Standards (EQS)... 4

Environmental Assessment Criteria (EAC)... 5

Guidelines for human consumption ... 5

Availability, internationally agreed target levels ... 5

Target levels with different aims ... 7

Availability and differences between target levels, examples CB‐153 and mercury ... 9

CB‐153... 9

Mercury ... 11

Target levels recommended for the Swedish Monitoring Programme in Marine Biota ... 12

Metals ... 12

PAHs... 13

Pesticides ... 13

PCBs ... 13

Other substances... 14

Illustrative examples using target levels for chemical status evaluation ... 16

How to achieve the same unit between chemical concentrations and target levels... 17

Adjusting for geographical or biological differences ... 20

Salinity ... 20

Fish body length... 21

Different target levels for different matrices... 22

References ... 24

2

(3)

Introduction

The Marine Strategy Framework Directive 2008/56/EC (MSFD), sets the ambitious goal to achieve Good Environmental Status (GES) in regard to pressures exerted by chemical pollution onto marine ecosystems. GES is determined from quality assessments based on target levels that represent a threshold that should not be exceeded. Such target levels are necessary tools in order to enable interpretation of chemical monitoring. This report focus on target levels developed for assessments of chemical status in biota or foodstuffs. With a few exceptions, target levels for sediment and water are not dealt with here.

As an assignment by the Environmental Protection Agency (EPA) in Sweden, this study was conducted by the Department of Contaminant Research at the Swedish Museum of Natural History. The study was also financed by the Swedish EPA

Aim of study

This aim of this study was to provide a background document for the ongoing process of developing and utilizing target levels for status assessments. The ambition is to recom‐

mend target levels for status assessments within the National Swedish Monitoring Pro‐

gramme in Marine Biota 2010. Another task is to provide illustrative examples on the usage of target levels within international working groups such as HELCOM CORESET.

Target levels for chemical status assessment in biota

Scope and definitions

Target levels reported here are for substances included in the National Swedish Moni‐

toring Programme in Marine Biota 2010. Substances/group of substances included are:

mercury (Hg), lead (Pb), cadmium (Cd), nickel (Ni), chromium (Cr), copper (Cu), zinc (Zn), polychlorinated biphenyls (PCB’s), dichlorodiphenylethanes (DDT’s), hexachlorocyclo‐

hexanes (HCH’s), hexachlorobenzene (HCB), polychlorinated dioxins and dibenzofurans (PCDD/PCDF), polybrominated flame retardants (BRFs), polyaromatic hydrocarbons (PAH’s), and perfluorinated compounds (PFC’s).

In order to assess chemical status in water bodies, various types of quality standards or assessment criteria have been developed. These include Environmental Quality Stan‐

3

(4)

dards (EQSs) developed through the Water Frame Work Directive (WFD) and Environ‐

mental Assessment Criteria (EACs) and Background Assessment Concentrations (BACs) developed within the OSPAR convention. EQSs and EACs have a similar definition and are set to protect sensitive organisms from harmful effects due to hazardous sub‐

stances. In this report these definitions are referred to as “target levels” and also include recommendations based on human health, providing recommendations for food intake.

Illustrative examples of environmental status classifications provided in this report are based on concentrations in herring, cod, and mussels from the National Swedish Moni‐

toring Programme in Marine Biota and concentrations in sediments from the Geological Survey of Sweden.

Internationally agreed target levels

Generally, EQSs and EACs are developed through dose‐response relationships between chemical concentrations and documented biological effects in sensitive species. The EQS are set to define GES in respect to chemical status within a 12 nautical miles coastal zone for EU member states. EACs are developed for chemical status assessments, mainly in the North Sea. For more detailed information regarding the various target levels, we refer to background documents where specifics regarding species, tissues, or environ‐

mental conditions may be discussed more in depth.

Environmental Quality Standards (EQS)

The Environmental Quality Standards Directive (2008/105/EC) lays Environmental Qual‐

ity Standards (EQSs) for priority substances and certain other pollutants, as provided for in Article 16 of the WFD, with the aim of achieving good surface water chemical status.

The objective is to protect pelagic and benthic freshwater and marine ecosystems, as well as human beings from adverse impacts of chemical contaminants.

The annual average concentration (AA‐EQS) refers to the annual arithmetic mean con‐

centration providing protection against chronic exposure and cover short‐term chemical effects in biota. The methodological framework used in deriving these EQSs is described in Lepper 2005. Substance EQS Data Sheets (SDS) contains background information re‐

garding the development of EQS (available at CIRCAS webpage). Here, data from ecotoxicological studies are compiled to Specific Quality Standards (QS), derived for wa‐

ter, sediment, biota, and human health. To allow shortcomings in the available data an Assessment Factor (AF) which may be as low as 1 or has high as 10000 is applied to the QS. The most restrictive QS calculated for the different compartments will normally adopted as the EQS. So far, EQS have mainly been available for monitoring in water, only a handful EQS have been available for biological samples. Additional EQSs for sev‐

eral substances are in the pipeline, also for monitoring in biota. These are planned to be adopted within the next 1‐2 years. Here these are referred to as draft EQS.

be

4

(5)

Environmental Assessment Criteria (EAC)

Within the OSPAR convention, Environmental Assessment Criteria (EAC) has been devel‐

oped for interpretation of chemical monitoring data in sediments and biota (OSPAR CEMP 2009). Concentrations below the EACs are considered to present no significant risk to the environment and may be considered as related to the EQSs.

In addition to EACs, OSPAR has also developed tools to demonstrate low concentrations (Background Concentrations (BC), Low Concentrations (LC) and Background Assessment Concentrations (BAC). BACs are statistical tools defined in relation to the BCs or LCs, providing a 90% probability that the observed mean concentration will be below the BAC when the true mean concentration is at the BC.

Guidelines for human consumption

In addition to EQSs and EACs, chemical status can also be assessed from the point of human consumption. Maximum levels for contaminants in food are set in Commission Regulation (EC) No 1881/2006 and for national recommendations in Sweden by the Swedish National Food Administration (LIVSFS 1993:36).

Availability, internationally agreed target levels

Internationally agreed target levels for chemical status classification in biota are not available for all substances included in the Swedish National Marine Monitoring Pro‐

gramme. Target levels for human consumption and the environment (EC foodstuffs, EQSs and EACs) are listed in table 1. Note that for all substances included in the Swedish National Marine Monitoring Programme, EQSs in biota are all still under development (Quality Standards).

5

(6)

Table 1. Internationally agreed target levels for chemical status assessments in biota. The target levels have been de‐

veloped for food recommendations or to protect sensitive species in the marine environment from harmful effects. The EQS values are still under development and are essentially draft EQS or QS values.

Group of sub-

stances Substance Food recommenda-

tions EQS EAC

Cadmium X X

Copper

Lead X X

Mercury X X

Nickel X

Metals

Zink

Fluoranthene X X

Anthracene X

Napthalene X

Phenantrene X

Pyrene X

Benz(a)antracene X

Chrysene

Benzo(a)pyrene X X

Benzo(ghi)perylene X

PAHs

Indenol(1,2,3-cd)pyrene X

DDE X

γ-HCH X

Pesticides

α-HCH X

Aroclor mix X X

CB-28 X

CB-52 X

CB-101 X

CB-105

CB-118 X

CB-138 X

CB-153 X X

CB-156

CB-180 X

PCBs

ICES (6) X

Dioxins X X

PBDE X

HCB

PFOS

HBCDD X

BDE -incl.octa X

Other

TBT X X

6

(7)

Target levels with different aims

Target levels are designed to assess status in relation to various environmental goals. It is important to choose the appropriate target levels for the scope and aim of the as‐

sessment in mind.

Target levels developed for consumption may be a good option to include in status as‐

sessments. It is however important to keep in mind that these recommendations are set for parts of biota that is consumed and these may not be the parts of the fish that are most suitable for monitoring purposes. In general, recommendations for concentrations in foodstuffs are set for concentrations measured in the fish muscle. Data collected for environmental monitoring purposes often measure concentrations in other parts of the fish. For example, when evaluating levels of dioxins in regard to target levels set for food recommendations, concentrations in herring are measured for the whole fish (as the fish is consumed whole with skin). For environmental status assessments, concentra‐

tions of dioxins are measured in the muscle with the skin and the subcutaneous fat layer removed. As dioxins accumulate in fatty tissue this provides lower concentrations not directly comparable with target levels designed for human consumption. For metals, target levels are designed for human consumption set for concentration in fish muscle.

As metals such as lead and cadmium accumulate in the liver, data collected for envi‐

ronmental monitoring are often measured in the liver. Concentrations of lead in the liver are about eight times higher compared to muscle tissue and over 500 times higher for cadmium (Strandmark et al. 2008)

Regarding chemicals, the goal of Good Environmental Status (GES) within MSFD and WFD differs slightly from the goals within the Baltic Sea Action Plan (BSAP) and national aims in Sweden. GES in MSFD, Descriptor 8, is defined as “concentrations of contami‐

nants are at levels not giving rise to pollution effects”. This is in line with the definition of hazardous substances used in the Water Framework Directive 2000/60/EC (WFD).

EQS and EAC have been developed to identify these targets and are set to protect the most sensitive species in the system from harmful effects. In Sweden, the aim is to achieve an environment free from pollutants where concentrations of alien

contaminants are close to zero and naturally occurring substances close to background levels. This stricter goal is in line with the agreed goal of HELCOM on hazardous

substances and the BSAP. BSAP states the goal of a Baltic Sea undisturbed by hazardous substances described by the ecological objective as “concentrations of hazardous substances close to natural levels”.

7

(8)

Status assessments in this study apply the color codes used within the HELCOM assessment on hazardous substances in the Baltic Sea (HELCOM 2010). Environmental objectives of chemical status is color coded and yields a status classification (“high”,

“good”, “moderate”, “poor”, or “bad”). “High” and “good”

status classes indicates that areas are not disturbed by hazard‐

ous substances, while “moderate”, “poor”, and “bad” indicate different degrees of disturbance of hazardous subs

Bad Poor Moderate Good High Bad Poor Moderate Good High

tances.

8

(9)

Availability and differences between target levels, examples CB‐153 and mercury

CB‐153

Several target levels have been developed for PCBs, including EQSs, EACs, and recom‐

mendations for foodstuffs (Tab. 2, fig. 1 f). Which target levels is the most appropriate to use depends on the environmental goals to be reflected in the assessment. It is also im‐

portant to establish that the target level is designed to evaluate concentrations in the same tissues for which our monitoring data has been sampled (or that it can be trans‐

formed between tissues, see recommended target levels for metals pp. 12).

Table 2. Several target levels with different environmental goals in mind have been developed for PCBs, including EQSs, EACs, and recommendations for foodstuffs.

Matrix Substance Unit Low conc.

BC/LC

Background Reference

Background (BAC)

Background

Reference Threshold Threshold Reference

PCB-28 µg/kg ww 0.0/0.05 BC/LC 0.1 BAC 64 (lw) OSPAR 2009 EAC passive

PCB-52 µg/kg ww 0.0/0.05 BC/LC 0.08 BAC 108 (lw) OSPAR 2009 EAC passive

PCB-101 ug/kg ww 0.0/0.05 BC/LC 0.08 BAC 120 (lw) OSPAR 2009 EAC passive

PCB-105 ug/kg ww 0.0/0.05 BC/LC 0.08 BAC

PCB-153 µg/kg ww 100 (ww) Swedish EPA. LIVFS 1993

PCB-156 ug/kg ww 0.0/0.05 BC/LC 0.08 BAC

PCB-153 (flatfish fillet) ug/kg ww 2.5 OSPAR 2005 EAC

PCB-153 (roundfish) ug/kg ww 0.25 OSPAR 2005 EAC

PCB (Aroclor mix) ug/kg ww 0.0033 AA EQS. EC 2010

sPCB (flatfish fillet) ug/kg ww 10 OSPAR 2005 EAC

sPCB (roundfish) ug/kg ww 1 OSPAR 2005 EAC

PCB-28 µg/kg dw 0.0/0.25 BC/LC 0.75 BAC 3.2 OSPAR 2009 EAC

PCB-101 µg/kg dw 0.0/0.25 BC/LC 0.7 2009BAC 6 OSPAR 2009 EAC

PCB-105 µg/kg dw 0.0/0.25 BC/LC 0.75 BAC

PCB-153 µg/kg dw 0.0/0.25 BC/LC 0.6 BAC 80 OSPAR 2009 EAC

PCB-156 µg/kg dw 0.0/0.25 BC/LC 0.6 BAC

PCB-180 µg/kg dw 0.0/0.25 BC/LC 0.6 BAC 24 OSPAR 2009 EAC

PCB-153³ mg/kg ww 0.1 LIVFS 1993

PCB (Aroclor mix)³ ug/kg ww 0.003 AA EQS. EC 2010

Total PCB ng/g (ppb) dw 22.7 ERL. NOAA 1999

Total PCB ng/g (ppb) dw 180 ERM. NOAA 1999

PCB-28 ug/kg dw 0.0/0.05 BC 0.22 BAC (2.5%TOC) OSPAR 2009 EAC

Tot-PCB¹ ug/kg dw 30 LIVFS 1993

Tot-PCB² ug/kg dw 20 LIVFS 1993

FishMusselsSediment

1) freshwater 9

2) Marine waters 3) Food stuffs

(10)

In order evaluate the difference between target levels, several assessments of environ‐

mental status for CB‐153 measured in herring along the Swedish coast, was evaluated.

Figure 1 (a‐e) presents the results of the various environmental status assessments of CB‐153. Status evaluated in relation to EQS, BAC, and EACroundfish varies between moder‐

ate and bad (Fig. 1 a‐c). Using the more recent EACpassive, or recommendations for food‐

stuffs from the national food administration in Sweden, indicates good status through‐

out the coast of Sweden (Fig. 1 d‐e).

a) b) c)

Figure 1. The figures (a‐e, illustrate status assessment of CB‐153 in herring along the Swedish coast. The figures demonstrate status evaluated against a) EQS, b) BAC, c)EACroundfish, d) EACpassive, and e) EC foodstuffs. At the bottom right (f), the different target levels are compared on a scale from 1‐100 μg/kg ww

d) e)

EQS ug/kg ww. BAC ug/kg ww. EAC round. ug/kg ww.

250 km

< 0.00165 EQS/2 0.00165 - 0.003 EQS 0.003 - 0.03 EQS x 10 0.03 - 0.3 EQS x 100

> 0.3 EQS ug/kg ww.

TISS - 11.03.24 12:44, CB-153_EQS

250 km

< 0.05 BAC/2 0.05 - 0.1 BAC 0.1 - 1 BAC x 10 1 - 10 BAC x 100

> 10 BAC ug/kg ww.

TISS - 11.03.24 12:45, CB-153_BAC

0.5 - 1 EAC round. x 4 0.25 - 0.5 EAC round. x 2 0.1 - 0.25 EAC round.

< 0.1 BAC

250 km

> 1

EC round ug/kg ww.

TISS - 11.03.24 12:44, CB-153_round

EACpassive ug/kg ww.

250 km

< 0.1 BAC 0.1 - 48 EACp.

48 - 96 EC Food x 2 96 - 192 EC Food x 4

> 192

EACpassive ug/kg ww.

TISS - 11.03.24 12:45, CB-153_EACp

EC Food ug/kg ww.

200 - 400 EC Food x 4 100 - 200 EC Food x 2 0.1 - 100 EC Food

< 0.1 BAC

250 km

> 400 EC Food ug/kg ww.

TISS - 11.03.24 12:46, CB-153_Food

10

(11)

Mercury

In order to assess environmental status of mercury along the Swedish coast, concentra‐

tions in herring were evaluated against target levels EQS (22 μg/kg), BAC (35 μg/kg), and EC foodstuffs (500 μg/kg). Figure 2 demonstrates how environmental status varies de‐

pending on target level. Concentrations in herring evaluated against the QS indicate that status is high to moderate (Fig. 2 a). Evaluated against the BAC, status ranges between high and good (Fig. 2 b). Based on the EC food recommendations, status is good (Fig. 2 c).

Figure 2. The figure presents environmental status classification in mercury evaluated against target levels, QS (22 μg/kg ww) (a), OSPAR Background concentrations (b), and EC recommendations for foodstuffs (500 μg/kg ww) (c).

f)

a) c)

BAC ug/kg ww.

250 km

< 22 BAC 22 - 35 Swe.

35 - 220 QS x 2 500 - 1000 QS x 4

> 1000 BAC ug/kg ww.

TISS - 11.03.24 13:38, Hg_BAC

b)

EC Food ug/kg ww.

QS ug/kg ww.

250 km

< 11 QS/2 11 - 22 QS 22 - 44 QS x 2 44 - 88 QS x 4

> 88 QS ug/kg ww.

TISS - 11.03.24 13:35, Hg_QS

1000 - 2000 EC fish x 4 500 - 1000 EC fish x 2 35 - 500 EC Food

< 35 BAC

250 km

> 2000 EC Food ug/kg ww.

TISS - 11.03.24 13:34, Hg_Ec

11

(12)

Target levels recommended for the Swedish Monitoring Programme in Marine Biota

The ambition of this report is to recommend target levels appropriate to use for chemi‐

cal status assessments within the National Swedish Monitoring Programme in Marine Biota 2010. Since the development of target levels is presently intensive, it should be stressed that this report covers the situation to the current date. Assessments within the monitoring programme have the ambition to detect changes over time and relate these to Swedish environmental goals. Target levels recommended in this report are therefore selected based on the following requirements:

1) Primarily, internationally agreed target levels are recommended.

2) Where more than one target level is available, agreed EQSs are prioritized before EACs.

3) If recommendations for foodstuffs are lower than EQSs, these are preferred.

4) If no agreed EQS or EAC are available, QS are used if these are lower than recommendations for foodstuffs.

5) If reliable target levels have been produced with specific regard to Swedish environmental conditions, these are selected prior to internationally agreed target levels (e.g. HCH).

6) Only one type of target level is applied within each substance groups (e.g.

we do not mix EQS and EAC depending on availability for different conge‐

ners).

Table 3 presents an overview of internationally agreed target levels (Food recommenda‐

tions, EQSs, and EACs) along with additional options for target levels, such as EQS for biota on the draft stage, Quality Standards (QS) and other recommendations. Target levels recommended for chemical status assessments within the Swedish National Ma‐

rine Monitoring Programme are indicated with bold letters. More detailed information on the process behind the recommendation of target levels for the different groups of substances are provided below:

Metals

Higher accumulation of metals in liver and a higher between year variations in muscle than in liver, makes liver a more suitable than muscle for time trend studies of metals such as cadmium, lead, and nickel. As target levels for these substances are set for con‐

centrations in fish muscle (EC foodstuffs and QS human health), or for the whole fish (QS secondary poisoning), concentrations in the liver need to be translated to the appropri‐

ate tissue. The concentrations of the analyzed metals are six to nineteen times higher in liver compared to muscle, except for cadmium which is more than 500 times higher in

12

(13)

liver tissue (Strandmark et al. 2008). Ratios enabling recalculating of metal concentra‐

tions in liver and muscle values are available for herring and perch (Strandmark et al.

2008).

For mercury, concentrations are measured in fish muscle and the recommended target level (QS) is set to protect against secondary poisoning and are provided for concentra‐

tions in the whole fish (prey tissue). At present we have no data allowing translations between muscle concentrations and the whole fish, instead we let the muscle concen‐

trations represent the whole fish.

Target levels are lacking entirely for copper and zinc. These are essential trace metals regulated within certain ranges to maintain homeostasis. Secondary poisoning is not considered relevant for these metals (Swedish EPA, report No 5799).

PAHs

For PAHs, draft EQS are available for a few congeners, but for consistency, we use EAC as these are available for a larger group of PAHs.

Pesticides

In this report the current EAC for DDE is recommended, but we fear that the EAC is set to high and may not fully protect Baltic top predators. The EAC for DDE dates back to 1997 and it is set to protect the peregrine falcon in regard to thinning of eggshells.

Predators are usually sensitive to pollutant effects and the harmful effects of DDE and the consequences of eggshell thinning are well established (Helander et al. 2008). The peregrine falcon represents terrestrial species and due to differences in biomagnifica‐

tion, we believe that the effects in a marine top‐predator such as the white‐tailed sea eagle may be a more appropriate to consider when developing a target level for marine environments.

For HCH, we recommend target levels suggested by the Swedish Environmental Re‐

search Institute (IVL). They have developed specific levels for the situation in Sweden regarding for example organic carbon in the sediments and factors for bioconcentration (BCF) and biomagnification (BMF), they have performed translations between EQS for surface water to biota (Lilja et al. 2010). The level is set for the sum of HCH (including lindane), but is still lower than QS levels set for more limited sets of congeners.

PCBs

Several target levels have been developed for PCBs, including EQSs, EACs, and recom‐

mendations for foodstuffs. At the present state, recommending a target level for envi‐

ronmental status assessments for PCBs is problematic. The EACs struggle with credibility as the target levels for PCBs have been updated several times with very fluctuating lev‐

els (Tab. 2, fig. 1 f). The most recent, EACpassive (Tab. 2), is the result of ICES’s recommen‐

dations. For the EACpassive, PCBs in sediment were used to calculate concentrations of

13

(14)

14 PCBs in fish liver (on a lipid weight basis) and mussels in equilibrium with sediment con‐

taining PCB concentrations equal to the EACs in sediment (Webster et al. 2009). The EACpassive can be regarded as tentative.

Recently, an EQS for PCBs in an Aroclor mixture has been presented as a draft within the EU. However, this draft EQS is set to protect against the effects of non‐dioxin like PCBs as well as planar dioxin‐like PCBs which results in a concentration that often falls below the limit of quantification (LOQ). In reality, practical use of this target level within status assessment is limited.

In addition to EACs and EQS, recommendations for concentrations in foodstuffs are also available but the levels are so high that they cannot be recommended for assessing en‐

vironmental status. Awaiting a legally binding EQS, we recommend that the BACs for the individual CB congeners are used for status assessments. The BAC is in line with Swedish national goals of chemical concentrations close to natural levels.

Other substances

For PBDEs, there is a common draft EQS for the sum of BDE congeners linked to c‐ pen‐

taBDE and c‐octaBDE (e.g. tetra‐ to nona‐BDE congeners). In the preparation of this re‐

port, we have not been able to clarify which congeners that are agreed to be included in the sBDE. Instead, the QS set for food uptake in humans from fishery products is rec‐

ommended. For PFOS and HBCDD we recommend the draft EQS, and for HCB we rec‐

ommend using the QS set for food uptake by man from fishery products.

(15)

15

Draft EQS (µg/kg w et w eight)

Other (µg/kg w et w eight)

Fish Bivalves Fish

AA-EQS BAC EAC BAC EAC

Cadmium General 50 1000 26 160 ²

Lead 300 1500 26

Mercury General 500 (tot Hg) 35 20 (Methyl Hg)²

Nickel 670 ¹

Fluoranthene 30 ¹ 12.2 110

Anthracene 290 2435¹

Naphtalene 340 2435¹

Phenantrene 11 1700

Pyrene 9 100

Benz(a)anthrazene 2.5 80

Chrysene 8.1

Benzo(a)pyrene 2 ¹ 1.4 600

Benzo(ghi)perylene 2.5 110

Indenol(1,2,3-cd)pyrene 2 ¹ 2.4

Aroclor mix 0.003 ²

CB-28 0.1 64 lw 0.75 3.2

CB-52 0.08 108 lw 0.75 5.4

CB-101 0.08 120 lw 0.7 6

CB-105 0.08 0.75

CB-118 0.1 24 lw 0.6 1.2

CB-138 0.09 316 lw 0.6 15.8

CB-153 100 LIVS 0.1 1600 lw 0.6 80

CB-156 0.08 0.6

CB-180 0.11 480 lw 0.6 24

CB (ICES 6) 75 ng/g

ΣPCDDs+PCDFs General

ng WHO98-TEQ / kg 4 ng/kg 0.23 ng/kg ²

ΣPCDDs+PCDFs+dl-PCBs General, Fish liver:

25.0 ng WHO98-TEQ / kg

8.0 ng/kg

Tributylin compounds 15.2 12

BDE (congeners 28, 47,

99, 100, 153, and 154) 274 ng/g ¹

HCB 9.7 ¹

PFOS 9.1 ¹

HBCDD 167 ²

DDE (p,p') 0.1 5

HCH (incl. lindane) 11 0.97 1.45 33 ² 2.6³

Other MetalsPCBsPAHs

Pesticides

QS, CIRCA SDS (µg/kg w w ) Group of substance

EAC (µg/kg w et w eight)

Fish Mussels (dw )

EC, food stuffs (µg/kg w et w eight)

General = excluding species listed in paragraphs in Commission Regulation (EC) No 1881/2006 1) Food uptake by man via fishery products

2) Protection against secondary poisoning, amount / prey tissue

3) Swedish Environmental Institute, EQS set for water translated to concentrations in biota

(16)

Illustrative examples using target levels for chemical status evaluation

Many target levels are specific to basis (lipid or wet weight), organs, and species etc.

This requires adjustments either of the chemical concentrations in the monitoring ma‐

trix, or vice versa. In this section, examples of methods for adjusting data or target lev‐

els, and the consequent outcome in status assessments are provided.

Other situations that may affect the outcome of status classifications evaluated against target levels are when chemical concentrations measured in biota are effected by con‐

founding factors. Examples of this are may be when measured concentrations, beside anthropogenic impact, also reflect factors such as variations in salinity or size of the monitoring species (Johansson et al. 2010). Without consideration to confounding fac‐

tors, status classifications may come to reflect differences in biological or environmental factors rather than contamination load. This section provides illustrative examples of scenarios where status may be misinterpreted and where extra caution may be neces‐

sary.

Chemical status classification in biota.DOC

16

(17)

How to achieve the same unit between chemical concentrations and target levels

Almost exclusively, target levels such as EQSs, EACs, and recommendations for foodstuffs, are expressed on a wet weight basis. In the marine environment, most persistent, organic pollutants (POPs) accumulate in fatty tissue and therefore most chemical concentrations in biota are measured on a lipid weight basis (lw). Trans‐

formations to achieve chemical concentrations in the study organisms expressed on the same basis as target levels can be performed in several ways. The procedure will ultimately affect the results of the environmental status assessment.

With an average fat content of approximately 0.6 %, perch is a leaner fish than herring with a fat content of approximately 3 %. Due to this, chemical concentrations on a wet weight basis generally are lower in perch compared to herring. To evaluate the importance of status classifications dealing with different basis between study organism and target level, chemical status of CB‐153, measured in perch and herring are evaluated against OSPAR BAC (0.1 μg/kg ww). Chemical concentrations of CB‐153 are originally expressed on lipid weight and the original BACs value on wet weight. For the status assessment in figure 3, the different bases have been transformed accordingly:

1) The target level is expressed as original values on wet weight. Concentrations of CB‐153 are transformed from lipid weight to wet weight (Fig. 3 a‐b).

2) Concentrations of CB‐153 are expressed as original values on lipid weight. The target level is transformed to lipid weight. In relation to the specific fat content of each of the two species (perch and herring), the BAC is transformed into two, species specific target levels (Fig. 3 c‐d).

3) Concentrations of CB‐153 are expressed on a lipid weight basis. Based on the mean fat content of herring and perch, the BAC is transformed into a common target level (Fig. 3 e‐f).

Due to few sampling sites, concentrations in perch were estimated for a number of stations using a ratio be‐

tween herring and perch (from the last ten years). In order to correct for a gradient in salinity, a trend line equation was applied to calculate new ratios based on salinity at different sampling sites for herring. The new ratios were then combined with herring concentrations to extrapolate perch concentrations along the coast.

(18)

18

Figure 3 e‐f. Chemical status of CB‐153 expressed on a lipid weight basis in herring (e) and perch (f). Status is eval uated against OSPAR BAC express lipid weight. The BACs were trans‐

formed using the mean lipid concentra tion of herring and perch.

‐ ed on

‐ Figure 3 c+d. Chemical status of CB 153 expressed on a lipid weight basis in herring (c) and perch (d).

Status is evaluated against OSPAR BAC expressed on lipid weight. The BACs were transformed using spe‐

cies specific lipid concentrations for herring and perch.

‐ Figure 3 a+b. Chemical status of CB‐

153 expressed on a wet weight basis in herring (a) and perch (b). Status is evaluated against OSPAR BAC ex‐

pressed on wet weight.

a) b)

250 km

< 0.05 BAC/2 0.05 - 0.1 BAC 0.1 - 0.2 BAC x 2 0.2 - 0.4 BAC x 4

> 0.4 BAC x 8 BAC ug/kg ww.

TISS - 11.03.30 12:00, CBww_clup

0.2 - 0.4 BAC x 4 0.1 - 0.2 BAC x 2 0.05 - 0.1 BAC

< 0.05 BAC/2

250 km

> 0.4 BAC x 8 BAC ug/kg ww.

TISS - 11.03.30 12:00, CBww_perc

Herring (lw H) Perch (lw P)

c) d)

250 km

< 0.0015 BAC/2 0.0015 - 0.003 BAC 0.003 - 0.006 BAC x 2 0.006 - 0.012 BAC x 4

> 0.012 BAC x 8 BAC ug/kg lw.

TISS - 11.03.30 11:59, CBlw_clup

0.032 - 0.064 BAC x 4 0.016 - 0.032 BAC x 2 0.008 - 0.016 BAC

< 0.008 BAC/2

250 km

TISS - 11.03.30 11:59, CBlw_perc

Herring (mean lipid weight, H&P) Perch (mean lipid weight, H&P)

e) f)

250 km

< 0.003 BAC/2 0.003 - 0.006 BAC 0.006 - 0.012 BAC x 2 0.012 - 0.024 BAC x 4

TISS - 11.03.30 11:57, CBlwx2_clup

0.012 - 0.024 BAC x 4 0.006 - 0.012 BAC x 2 0.003 - 0.006 BAC

< 0.003 BAC/2

250 km

TISS - 11.03.30 11:58, CBlwx2_perc

(19)

19

Figure 3 (a‐f), demonstrates the significance of a deliberate and consequent approach to tackle differences in the base on which chemical concentrations are measured. Expressed on wet weight basis, chemical status for CB‐153 reflects differences in lipid content between herring and perch rather than differences in contaminant load. Status for CB‐ 153 in herring is overall bad, whereas status for CB‐ 153 in perch is overall high (Fig. 3 a‐b).

When concentrations are evaluated against a BAC created for each of the species based on species specific lipid content, chemical status ranges from poor to bad for herring and between high to bad for perch (Fig. 3 c‐

d). If a common BAC is created for the two species based on the mean of the lipid content of the two species, status of CB‐153 ranges between poor and bad for both species (Fig. 3 e‐f). The differences are most pre‐

dominant on the west coast where status ranges between moderate and poor for herring (Fig. 3 e) and re‐

mains high for perch (Fig. 3 f).

(20)

20

Adjusting for geographical or biological differences

Salinity

Salinity is an abiotic factor known to affect bioaccumulation of metals in biota (Barnes and Barnes 1986, Bro‐

man et al. 1991). This suggests that adjusting chemical concentrations in biota to varying salinity before as‐

sessing environmental status may be appropriate. Here, the outcome of adjusting for salinity or not is tested by evaluating environmental status of cadmium in mussels against OSPAR BAC using concentrations adjusted or unadjusted to salinity.

Status classification of cadmium (Cd) was performed in relation to BAC at four stations along the Swedish coast. To adjust values, concentrations were transformed by means of a regression line before using the slope (‐1.5335) calculated from data found in Broman et al. (1991). In figure 4, we can see that the effects of not adjusting for differences in salinity can lead to an environmental status classification reflecting differences in salinity between monitoring stations rather than differences in contaminant burden. Without adjusting for differences in salinity, cadmium concentrations are high on the west coast and elevated on the east coast.

Adjusting for salinity improves the status on the east coast, resulting in high status at all four stations.

Figure 4. Status classification of cadmium in mussels along the Swedish coast. The status classification reflects differences in salin‐

ity. a) Cd status without adjusting concentrations of cadmium to differences in salinity. b) Status classification when concentrations of cadmium have been adjusted to compensate for differences in salinity.

BAC ug/kg ww.

250 km

< 13 BAC/2 13 - 26 BAC 26 - 50 EC Food 50 - 160 QS

> 160 BAC ug/kg ww.

TISS - 11.03.22 11:45, Cd_BAC

BAC ug/kg ww.

250 km

< 13 BAC/2 13 - 26 BAC 26 - 50 EC Food 50 - 160 QS

> 160 BAC ug/kg ww.

TISS - 11.03.22 11:47, CdB_BAC

Adjusted for varying salinity Unadjusted for salinity

(21)

21

Fish body length

With increasing size, herring changes trophic feeding level affecting the accumulation of mercury (Falandysz et al. 2000). Consequently, comparing mercury loads in fish belonging to different size categories may reflect differences in body length rather than geographical differences in contaminant burden. In order to test the importance of limiting your data to specific size classes (or trophic levels) before evaluating environmental status of mercury between different areas, herring data from 2000‐2010 was grouped into two size classes.

Size class a contain herring with a body length below 15 cm and size class b contain herring with a body length of 15 cm and above.

Comparing status evaluations of mercury between two size classes of herring, reveals that status classification of mercury reflects differences in the fish body length (Fig. 5). At two locations in the Bothnian Bay (indicated with arrows), environmental status differs between the two size classes. Fish in the category with a body length below 15 cm, indicate an overall high status, whereas status for fish in the category with a body length of 15 cm and above, ranges between high and moderate.

Figure 5. Status assessment for mercury (Hg) along the Swedish coast. The status classification reflects differences in size of the fish. The figures demonstrate Hg status in a) small fish (0‐15 cm), b) large fish (>15 cm).

Hg ug/kg ww.

250 km

< 11 EQS/2 11 - 22 EQS 22 -35 BAC 35 - 70 BAC x2

> 70 Hg ug/kg ww.

TISS - 11.03.22 14:13, Hg_EQSa

Hg ug/kg ww.

250 km

< 11 EQS/2 11 - 22 EQS 22 -35 BAC 35 - 70 BAC x2

> 70 Hg ug/kg ww.

TISS - 11.03.22 14:13, Hg_EQSb

Body length ≥ 15 cm Body length 0‐15 cm

(22)

22 Different target levels for different matrices.

Depending on availability, assessments evaluating chemical status may often consist of concentrations meas‐

ured in different matrices, such as biota, sediment or water. The fact that different matrices accumulate chemicals differently should be covered within the scope of the target levels. However, status assessment based on evaluations using target levels developed within separate organizations/conventions may lead to status reflecting different origin of the target levels used for separate matrices rather than differences in con‐

taminant burden. The availability of target levels for naphthalene illustrates a practical example. EQS are only available for concentrations measured in fish. EAC is available for concentrations in mussels, and for concen‐

trations in sediments, an EAC, based on the North American corresponding target level, Effect Range Level (ERL) is available.

In this example, we have tested how chemical status evaluations of naphthalene are affected by using three different assessment tools, EQS for fish, EAC for mussels and ERL (Target level developed in North America, later adapted as an EAC) for sediments. For perch and blue mussel there were not an adequate number of sampling sites to make a good comparison between matrices leading to extrapolation of concentrations. Ra‐

tios were calculated between sediment concentrations and perch and blue mussel concentrations, respec‐

tively at available sites. Since ratios were not consistent between different areas they were plotted versus salinity. A regression equation was created and applied to calculate new ratios based on salinity at different sampling sites for sediment. The new ratios were then combined with sediment concentrations at several sites to extrapolate perch and blue mussel concentrations along the coast. It should be stressed that the aim with extrapolation of concentrations was to create fairly realistic values not to calculate true concentrations.

In Sweden, current concentrations of naphthalene fall well below corresponding target level for fish, mussel and sediment. To illustrate this example, target levels for all matrices (EQS, EAC, ERL) are divided by a factor of ten in order to illustrate differences in concentrations of naphthalene.

Results show that chemical status evaluations for naphthalene concentrations reflect what target level are used. Concentrations of naphthalene measured in herring and evaluated against an EQS receive high status (Fig. 6 a) whereas concentrations in mussels and evaluated against an EAC receive moderate status (Fig. 6 b).

Finally, concentrations of naphthalene measured in sediment evaluated against an ERL results in a moderate or even poor status (Fig. 6 c). Generally, the degree of accumulation of chemical substances differs between biota, sediment and water, and cannot be expected to demonstrate the same status even in samples from the same location. However, in this example, chemical concentrations are deliberately created to reflect the same chemical status as concentrations in fish and mussels are extrapolated from concentrations measured in the sediment.

(23)

23

EQS, fish EAC, mussels ERL, sediments

Figure 6. The maps illustrate chemical status assessment of naphthalene in fish, mussels, and sediments along the Swedish coast. Environmental status is evaluated against different target levels depending on matrix. Status in herring evaluated against an EQS (a), status in mussels evaluated against an EAC (b), and status in sediments evaluated against an ERL (c).

EQS (sec. poisn) ug/kg ww.

250 km

< 2.345 EQS/10E4 2.435 - 24.35 EQS/10E3 24.35 - 243.5 EQS/10E2 243.5 - 2435 EQS/10E1

> 2435 EQS ug/kg ww.

TISS - 11.03.22 15:05, Naft_percE

ERL ug/kg dw.

250 km

< 0.16 ERL/10E4 0.16 - 1.6 ERL/10E3 1.6 - 16 ERL/10E2 16 - 160 ERL/10E1

> 160 ERL ug/kg dw.

TISS - 11.03.22 14:50, Naft_sedE

EAC ug/kg dw.

34 - 340 EAC/10E1 3.4 - 34 EAC/10E2 0.34 - 3.4 EAC/10E3

< 0.34 EAC/10E4

250 km

> 340 EAC ug/kg dw.

TISS - 11.03.22 14:50, Naft_mytiE

(24)

24

References

Andres, H., Malm, O., Kinjo, Y., Harada, M., Branches, F., Pfeiffer, W., Kato, H. 2000. Interspecific comparison of cadmium and zinc contamination in the organs of four fish species along a polymetallic pollution gradient (Lot River, France). The Science of the Total Environment 248, 11‐25.

Barnes, H and Barnes, M. 1986.Oceanography and marine Biology. Vol 24. Aberdeen University Press.

Broman D., Lindqvist L. and Lundbergh I. 1991. Cadmium and Zinc in Mytilus edulis L. from the Bothnian Sea and the Northern Baltic Proper. Environmental Pollution 74: 227‐244

EC. Commission regulation (EC) No 1881/2006 of 19 December 2006 setting maximum levels for certain con‐

taminants in foodstuffs.

EC. Directive 2008/56/EC of the European parliament and of the council. Establishing a framework for comu‐

nity action in the field of marine environmental policy (Marine Strategy Framework Directive).

Falandysz, J., Chwir, A., Wyrzykowska, B. 2000. Total mercury contaminantion of some fish species in the firth of Vistula and the lower Vistula river, Poland. Pol. J. Env. S. 9 (4): 335‐339.

Helander, B., Olsson, A., Bignert, A., Asplund, L., Litzén, K. 2002. The role of DDE, PCB, Coplanar PCB and egg‐

shell parameters for reproduction in the white‐tailed sea eagle (Haliaeetus albicilla) in Sweden. Ambio. 31:

387‐403.

HELCOM 2010. Hazardous substances in the Baltic Sea. An integrated thematic assessment of hazardous sub‐

stances in the Baltic Sea. Baltic sea environment proceedings No 120 B.

IPCS, International Programme on Chemical Safety (1992b). Cadmium. Environmental Health Criteria; 135.

World Health Organization, Geneva.

Johansson, A.‐K., Boalt, E., and Bignert, A. 2010. Utvärdering av effekten av störningsfaktorer på miljögiftsdata inom den limniska miljögiftsövervakningen. Swedish EPA, report nr 12:2010.

Lilja, K., Andersson, H., Woldegiorgis, A., Jönsson, A., Palm‐Cousins, A., Hansson, K., and Brorström‐Lundén, E.

2010. Bedömning av miljögiftspåverkan i vattenmiljö, samordnad metodutveckling. Rapport B1891.

Lepper, P. 2005. Manual on the methodological framework to derive Environmental Quality Standards for priority substances in accordance with Article 16 of the Water Framework Directive (2000/60/EC). Fraunhofer‐

Institue Molecular Biology and Applied Ecology, Schmallenberg, Germany.

LIVSFS 1993:36. Livsmedelsverkets föreskrifter om visa främmande ämnen i livsmedel.

Strandmark, A., Danielsson, S., Holm, K., and Bignert, A. 2008. Metaller I strömming och abborre‐ en jämförel‐

se mellan retrospektiva analyser I muskel och existerande tidsserier för leverkoncentrationer. Överrenskom‐

melse 212 0639, dnr 721‐7397‐06Mm.

OSPAR Comission 2009. CEMP assessment report: 2008/2009. Assessment of trends and concentrations of selected hazardous substances in sediments and biota.

(25)

25

Swedish EPA, 2009. Förslag till genomförande av direktiv 2008/105/EG om miljökvalitetsnormer inom vatten‐

politikens område. Rapport 5799.

Szefer, P., Falandysz, J. 1985. Trace metals in Muscle Tissue of Fish Taken from the Southern Baltic. Z Lebensm Unters Forsch 181, 217‐220.

Szefer, P., Domagała‐Wieloszewska, M., Warzocha, J., Garbacik‐Welosołowska, A., Ciesielski, T. 2003. Distribu‐

tion and relationships of mercury, lead, cadmium, copper and zinc in perch (Perca fluviatilis) from the Pom‐

eranian Bay and Szczecin Lagoon, southern Baltic. Food Chemistry 81, 73‐83.