Hazardous substances in marine organisms

Indicator definition

This indicator describes the levels and trends in European seas of concentrations of eight hazardous substances in marine biota, based on the individual assessment of monitoring data for the following substances:

• Mercury (Hg) and its compounds

• Cadmium (Cd) and its compounds
• Lead (Pb) and its compounds

• Hexachlorobenzene (HCB) 

• Polychlorinated biphenyl (PCB), using chlorinated biphenyls CB28, CB52, CB101, CB118, CB138, CB153, and CB180 as representatives
• The pesticide DDT (using pp’DDE as a representative of DDT)
• The pesticide Lindane- 1,2,3,4,5,6-hexachlorocyclohexane (HCH)
• The polyaromatic hydrocarbon (PAH) Benzo(a)pyrene (BAP)

The indicator is based on data for substances measured in organisms from the regional seas as follows:

• Baltic Sea – Atlantic herring (Clupea harengus)

• North-East Atlantic Ocean – blue mussel (Mytilus edulis), Atlantic cod (Gadus morhua), flounder (Platichtys flesus)

• Mediterranean Sea – Mediterranean mussel (Mytilus galloprovinicialis)

• Black Sea - Mediterranean mussel (Mytilus galloprovinicialis)

Units

The classification in the maps is based on concentrations in µg/kg, which are then classified into one of three classes: green (Low concentration), yellow (Moderate concentration) or red (High concentration). In addition a pie chart is presented showing the percentage of each class within each of the four regional seas.

Policy context and targets

Context description

A range of EU, regional and national legislation has been implemented in Europe to address the use of chemicals and their emission to the environment, including fresh and marine waters. The regulation of chemical pollutants in water began with the Dangerous Substances Directive (76/464/EEC), which has been integrated into the Water Framework Directive (2000/60/EC). The WFD represents the single most important piece of EU legislation relating to the quality of fresh, transitional and coastal waters, which aims to attain good ecological and chemical status of these waters by 2015. It requires the establishment of a list of priority substances (Decision 2455/2001/EC gave way to the First list of priority substances), to be selected from amongst those presenting a significant risk to or via the aquatic environment at EU level. It also requires the designation of a subset of priority hazardous substances, and proposals for controls to reduce the emissions, discharges and losses of all the substances and to phase out the emissions, discharges and losses of the subset of priority hazardous substances.

The chemical status of Europe's surface waters is currently addressed by the Environmental Quality Standards Directive – EQSD (2008/105/EC), a 'daughter' directive of the WFD, whose Annex II has replaced the first list of priority substances set out in the Decision 2455/2001/EC. The EQSD defines environmental quality standards (EQSs) in fresh and coastal waters for pollutants of EU-wide relevance known as priority substances (PSs). Member States are required to take actions to meet the quality standards in the EQSD by 2015.

Furthermore, emissions of hazardous chemicals from industrial installations and agricultural activities are regulated in the EU through the Integrated Pollution Prevention and Control (IPPC) Directive, whose abatement measures have contributed to a decline in metal emissions to water and air. Of relevance is also the EC Regulation 1907/2006 on the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), which aims to improve the protection of human health and the environment from the risks of chemicals. REACH gives greater responsibility to industry to manage the risks from chemicals and to provide safety information on substances used.

More recently combating this type of pollution in the open sea has been addressed by the Marine Strategy Framework Directive (2008/56/EC), which requires the achievement or maintenance of good environmental status in European seas by the year 2020 at the latest, through the adoption of national marine stategies based on 11 qualitative descriptors. Descriptor 8 (“Concentrations of contaminants are at levels not giving rise to pollution effects”) and Descriptor 9 (“Contaminants in fish and other seafood for human consumption do not exceed levels established by Community legislation or other relevant standards”) refer specifically to contaminants.

The EQSD list of priority substances includes cadmium, mercury, lead, hexachlorobenzene (HCB), lindane and DDT and benzo(a)pyrene, but not PCB. The WFD EQSD was revised in 2013 and the number of hazardous substances listed has increased from 33 to 45. Except for HCB, these substances are also on the list of chemicals for priority action for the OSPAR Marine Convention (OSPAR 1998). 

The production and use of DDT, HCB, lindane and PCBs are banned or voluntarily withdrawn within Europe. However, for PCBs, use is still permitted in closed system equipment, manufactured before the ban, until the end of service. Sales to consumers of nickel-cadmium batteries have been banned in Europe.  

At a regional level, international regional sea conventions (OSPAR, HELCOM, Barcelona Convention and Black Sea Convention) are also addressing pollution due to hazardous substances in European marine waters.

Article 5 of the revised Helsinki Convention of 1992 requires that the Contracting Parties undertake to prevent and eliminate pollution of the marine environment of the Baltic Sea area caused by harmful substances from all sources (HELCOM 2008). Under this convention, DDT and PCBs are banned and there is agreement that every contracting party shall endeavour to minimise and, whenever possible, to ban the use of cadmium, lead and mecury compounds. HELCOM has also Recommendation 31E/1, adopted May 2010, that Contracting Parties apply the Strategy to implement the HELCOM Objective for hazardous substances, and make use of the principles and methodologies contained therein to move towards the target of the cessation of discharges, emissions and losses of hazardous substances to achieve the Baltic Sea in good environmental status by 2021. 

Parties to the Convention for the protection of the Mediterranean Sea against Pollution have identified contaminants or groups of contaminants whose dumping or land-based discharges are prohibited or limited (Barcelona Convention and protocols). Parties to the Convention on the Protection of the Black Sea against Pollution have identified similar groups of contaminants and have protocols to reduce pollution by these harmful substances. However, it should be noted that even though relatively high contamination levels of some pesticides, heavy metals and PCBs are present, these substances are not monitored routinely (BSC 2009). 

Complementing these efforts are the Basel, Rotterdam and Stockholm Conventions - multilateral environmental agreements, which share the common objective of protecting human health and the environment from hazardous chemicals and wastes. The Basel Convention has dealt with control of transboundary movements of hazardous wastes and their disposal since 1989. The Rotterdam Convention text, adopted in 2004, promotes shared responsibility and cooperative efforts among Parties in the international trade of certain hazardous chemicals (which include mercury, HCB, lindane,  PCB and DDT) in order to protect human health and the environment and to contribute to the environmentally sound use of those hazardous chemical by, inter alia, providing for a national decision making process on their import and export. The Stockholm Convention, which entered into force in 2004, requires Parties to take measures to eliminate or reduce the release of POPs into the environment for which HCB, lindane, PCB and DDT are addressed. 

Targets

The aim of the Water Framework Directive is to achieve zero, near zero or background concentrations (more specifically defined in a daughter directive on ecological quality standards, i.e. the EQS-directive 2008/105/EC), depending on the contaminant. This is to be achieved through abatement actions on inputs, with the objective of reaching good ecological and chemical status by 2015 of fresh, transitional and coastal waters. However, the WFD only applies to the transitional and coastal environment. Goals similar to the Water Framework Directive have also been outlined by OSPAR and HELCOM. For the Mediterranean Sea, similar targets have been adopted. The reduction and phasing-out targets are formulated in accordance with related regional and international Conventions and programmes, such as the EU Directives, policies and strategies, and the Stockholm and Basel Conventions. However, similar targets have yet to be formulated for the Black Sea, although discussions are under way.

Within the scope of the Marine Strategy Framework Directive, hazardous substances are the relevant criteria and indicators in marine waters under Descriptor 8 (“Concentrations of contaminants are at levels not giving rise to pollution effects”) and 9 (“Contaminants in fish and other seafood for human consumption do not exceed levels established by Community legislation or other relevant standards”). In this regard, Member States are required to take into account relevant existing environmental targets. This would imply, inter alia, the environmental quality standards set out in the EQSD, since it applies to waters common to both the Water Framework Directive and the Marine Strategy Framework Directive, i.e seaward side of the baseline to the extent of territorial waters. A process is underway to define standards for the listed hazardous substances in biota so that these can apply to the marine and coastal environment.

Related policy documents

• 2013/39/EC.
  Directive 2013/39/EU of the European Parliament and of the Council of 12 August 2013 amending Directives 2000/60/EC and 2008/105/EC as regards priority substances in the field of water policy.
• Co-ordinated Environmental Monitoring Programme - Assessment manual for contaminants in sediment and biota, OSPAR 2008
  OSPAR Commission, 2008, publication no. 379/2008. ISBN 978-1-906840-20-4. 39 pp..
• Commission Decision 2010/477/EU of 1 September 2010
  Commission Decision of 1 September 2010 on criteria and methodological standards on good environmental status of marine waters.
• Commission Decision 2455/2001/EC
  2455/2001/EC. Decision No 2455/2001/EC of the European Parliament and of the Council of 20 November 2001 establishing the list of priority substances in the field of water policy and amending Directive 2000/60/EC [cf. Annex 10 of Water Framework Directive].
• Commission Directive 2009/90/EC of 31 July 2009
  Commission Directive 2009/90/EC of 31 July 2009 laying down, and pursuant to 2000/60/EC of the European Parliament and of the Council, technical specifications for chemical analysis and monitoring of water status.
• Commission Regulation 221/2002/EC of 6 February 2002
  221/2002/EC. Commission Regulation (EC) NO. 221/2002 of 6 February 2002 amending Regulation (EC) No. 466/2001 setting maximum levels for certain contaminants in foodstuffs. 3 pp.
• Commission Regulation 1881/2006/EC of 19 December 2006
  Commission Regulation (EC) No. 1881/2006 of 19 December 2006 setting maximum levels for certain contaminants in foodstuff.
• Convention on the Protection of the Marine Environment of the Baltic sea area (HELCOM).
  1992 (entered into force on 17 January 2000)
• Council Directive (76/464/EEC) of 4 May 1976
  Council Directive of 4 May 1976 on pollution caused by certain dangerous substances discharged into the aquatic environment of the Community. (76/464/EEC). Dangerous Substances Directive.
• EQS Directive 2008/105/EC of 16 December 2008
  Directive 2008/105/EC of the European Parliament and of the Council of 16 December 2008 on environmental quality standards in the field of water policy, amending and subsequently repealing Council Directives 82/176/EEC, 83/513/EEC, 84/156/EEC, 84/491/EEC, 86/491/EEC, 86/280/EEC and amending Directive 2000/60/EC of the European Parliament and of the Council.
• Marine Strategy Framework Directive 2008/56/EC
  Directive 2008/56/EC of the European Parliament and of the Council of 17 June 2008 establishing a framework for community action in the field of marine environmental policy (Marine Strategy Framework Directive)
• Strategic Action Plan for the Environmental Protection and Rehabilitation of the Black Sea
  This document represents an agreement between the six Black Sea Coastal states (Bulgaria, Georgia, Romania, the Russian Federation, Turkey and Ukraine) to act in concert to assist in the continued recovery of the Black Sea.  Black Sea Commission 2009
• Water Framework Directive (WFD) 2000/60/EC
  Water Framework Directive (WFD) 2000/60/EC: Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy.

Methodology

Methodology for indicator calculation

Data sources and coverage

The data used in this indicator is part of the WISE - State of the Environment (SoE) data, available in Waterbase - TCM (Transitional, Coastal and Marine) waters. Waterbase is the generic name given to the EEA´s database on status, quality and quantity of Europe´s water resources. Waterbase – TCM waters contains data collected both from EEA member countries (i.e. belonging to the EIONET) and from the Regional Seas Conventions through the WISE-SoE TCM data collection process (WISE-SoE was formerly known as Eionet-Water and Eurowaternet). The resulting WISE SoE TCM dataset is therefore made of sub-samples of national data assembled for the purpose of providing comparable indicators of state and impact of transitional, coastal and marine waters on a Europe-wide scale.

Geographical coverage

There is generally good data coverage for concentrations in the North-East Atlantic, except for Portugal, and in the Baltic. The Mediterranean Sea was only represented by data from a few countries. For the Black Sea, the only available data was Romanian mussel data, but because of the requirement for conversion to a preferred basis, too little data was available for the 2013 assessment. However, in the 2014 assessment some stations that were quite close to each other were grouped in order to get the minimum three years of data (see further below), so an assessment could be made. 

Temporal coverage

Concentrations in biota were measured during the period 1978 to 2012. Only data from the period starting in 1998 was considered for the trend assessment up to the 2013 assessment. For the latest 2014 assessment, only data from the period starting in 2003 was considered. Furthermore the assessment only includes time series that have data from 2006 or later and that cover at least three years, not necessarily contiguous. Many of these series have gaps, with intervals of two or more years between observations. 

Tissues

For fish, only concentrations from the following tissues were used:

• For mercury: muscle
• For the other metals: liver
• For organic compounds: liver in all fish except herring (Clupea harengus), for which muscle was used.

Conversion to a preferred basis for data in assessment

The classification by which the data are assessed requires conversion to the preferred basis (OSPAR, 2008). In order to create comparability between data within and between stations, and to allow comparison with assessment criteria, it is necessary to choose the bases on which all concentrations must be expressed. The preferred bases applied by OSPAR (2008) are:

• wet weights for metals and for organic compounds in mussels and fish except herring;
• lipid weights for organic compounds in herring.

The choice of bases aimed at meeting several considerations: scientific validity, uniformity for groups of contaminants for particular tissues and a minimum loss of data. As to the latter, the choice of bases will affect the amount of data that can be included in the assessment, depending on available information on dry weights, wet weights, lipid weights and the ratio between dry and wet weight. For example, for the Mediterranean Sea, mussel data from Greece could not be converted to the preferred dry-weight basis on a sample-specific basis, and was therefore excluded from the assessment. For the 2014 assessment, data available on dry weight was converted to wet weight basis, but only when a sample-specific dry weight-wet weight ratio was available. An exception was made for data from Mediterranean mussels (Mytilus galloprovincialis) in France and Italy where it assumed the dry weight-wet weight ratio to be 0.19, as used by OSPAR, in order to increase the geographic coverage. 

Aggregation of data by station

In a primary step, each time series (combination of location, species, tissue and determinand) is aggregated to a median concentration for each year. For years where some values are reported as ´less-thans´ (below reporting limit) and only 50% or less of the observations are real values above the largest ´less-than´ limit, the median can only be specified as a low-high range. The high limit is the estimate found when using the upper limit for ´less-than´ observations, and the low limit is the median estimate found by assuming 0 for these observations. The low limit will often be 0, but may also be a positive value. If the low and high limit is the same value, the median is well-defined, even if there are some ´less-than´ values in the sample.

Some data series contain concentrations given as exact zero. If this occurs for less than 50% of observations within a year, the high limit for the median of that year will still be a positive value. If more than 50% of observations are reported as 0, the smallest value > 0 is used as the best approximation to the median value. It should be noted that reporting concentrations as exact zero is not correct; it should either be a definite positive value or reported as below a detection limit > 0. The procedure described above implicitly assumes that zero values represent concentrations below the reported positive values or detection limits; if this is not the case the procedure may give a misleading result.

In a secondary step, each resulting time series of median values (or ranges) was assessed separately. Only time series with at least three years of data, of which one year must be 2006 or later, were included in the assessments. Only data from the years 1998-2010 was included in the 2013 analysis whereas the period was 2033-2012 for the 2014 one. Years where all values are reported as equal to zero are excluded from the analysis.

Trend assessment for each time series

The trend assessment method depends on whether all median values are well-defined, or whether some median values can only be specified as ranges.

Time series where all median values are well-defined
For such series, the trend is analysed both by regression on log-transformed medians versus year and by non-parametric Mann-Kendall test.

The regression method depends on the length of the time series:

• For series with at least seven years of data, a smooth curve is fitted by local weighted regression (LOWESS) vs. time, assuming a log-normal distribution of residuals. The smoother uses a moving regression window with a time span of seven years, expanded where necessary to cover at least three data points if there are large gaps in the time series. The dominant trend within the last 10 years is assessed by testing if there is a significant difference between the fitted value for the most recent year and the fitted value for the earliest year with data within the last 10 years. Thus, the maximum contrast period is 2001-2010. 
• If the time series has five to six years with data, the trend is assessed by linear regression of log-transformed concentrations instead of the smoother.
• If the time series has less than five years, no regression trend test is done.

Both types of regression test are done at a 5% two-tail significance level. Extremely low values that deviate from the trend by more than a factor 10 will be excluded if that results in a lower classification level, see below. In addition to the regression, a non-parametric Mann-Kendall trend test is done also with 5% significance level (ICES, 2000). 

The nonparametric Mann-Kendall test is done for all time series with four years or more.

In cases where both regression and Mann-Kendall assessments have been made, the results are combined as follows: If only one of the two methods shows a significant trend, that result is used. If the two methods should both indicate significant trend, but in opposite directions, the series would be flagged as having inconsistent trend indication, however, this does not occur in the present dataset. It is found that the two methods supplement each other: The regression test, in particular the smoother test for series of seven or more years, is better at detecting clear differences in level between the beginning and end of a time series if there are large fluctuations in between (non-monotonic trends). The Mann-Kendall test is better at detecting trends if for instance values are consistently low in the last part of the series, but has large variations in the first part of the time series.

Time series where some median values can only be specified as ranges

Regional trend assessment

The regional trend assessment is based on a tally of significant upward and downward trends and the contaminant-region trend (general trend) or contaminant-region-class trend in question (OSPAR 2009b). The significance of the tendency is determined by using the cumulative binomial distribution probability. This implies an assumption that each trend represents an independent measure of overall trend, which cannot be expected to be met, so the statistical assessment of the regional trend is merely indicative.

Classification of concentrations (i.e. levels) – low, moderate, high

For time series with five or more years of data, the upper 95% confidence limit for the fitted value for the last year in the data series (equal or > 2006) is used to classify environmental status. For time series covering only three or four years of data, the upper 95% confidence limit for the average of the yearly medians is used. The classification is done by comparing the confidence limit to low and high classification levels in Table 1 (see further below). For series with data from only one or two years, no classification is made.

The level is classified as Green if the test value is below the low limit, Yellow if in the low-high classification interval, and Red if it is above the high classification level.

In some series, the normal procedure on the full time series of yearly medians will give unreasonably high assessment levels. In particular, this is the case when there are extreme low values occurring in the series of yearly medians. Such low values will, in many cases, increase the uncertainty estimate so much that the upper 95% confidence level becomes very high, much higher than any observed value. To avoid such unreasonable classification, series where any of the yearly medians are lower than 10% of the fitted value for the same year are reanalysed with such values excluded. If this results in a lower assessment level, the revised analysis is used both for assessment level and time trend. Deviating high yearly medians are never removed, so an extremely high and possibly spurious median for a single year will still increase both fitted value and uncertainty, and may cause very high assessment levels.

It should be noted that classification reflects both observed levels and the uncertainty in trends or levels indicated by data. Thus a Red classification does not necessarily mean that estimated levels are above the limit; it may merely indicate that the uncertainty is so large (because of little data and/or high variability) that it cannot be assessed with reasonable confidence that the true level is below the highest limit.

The regional assessment of current levels is based on a tally of current levels for each of the three categories and for each contaminant and species-tissue in question (OSPAR 2009b). Datasets that are insufficient for trend analyses are weighted by a factor of two. Proportions for different combinations of contaminants and regions are calculated for the categories.

Combined PCB assessment

For PCBs, a combined assessment for each combination of location, species, tissue is made as follows: A significant trend direction (Up/Down) is reported only if at least 50% of all the PCB components where the time trend is assessed (Up/Down/Not significant) has the same significant direction. Missing results (NA for trend ‘test not applied’) do not count; if all components have a trend result NA, that is also the combined result. The combined classification is the next worst classification over the PCB components where a class is specified. NA is also set for time series where it is not possible to assign a direction of change between any two years in the series, no matter how many years the series have. This occurs if all yearly medians given as a specific number are identical, and where all low-high median ranges overlap the ranges and values for all other years. (An example: three years with value 0.1, two years with ranges [0.05-0.2], [0-0.15]). In these cases, it is not possible to define a sensible scale against which to assess the lack of trend indications, and NA seems to be the most relevant classification. NS is reserved for cases where it is possible to assess whether the long-term trend in the time series is significant compared to the short-term (year-to-year) variation.

DDT assessment

For DDT, only the DDE, p,p' component is used for the assessment. The other two frequently measured components are DDD,p,p’ and DDT,p,p’ but they have in generally smaller concentrations than DDE,p,p’. DDE,p,p’ has also been measured in the largest number of samples. In particular DDT,p,p' is missing in a substantial number of samples.

Classification tables 

Tables 1 and 2 provide an overview of the concentration limits used in the 2013 and 2014 assessments respectively. Based on those limits, three classes are defined: Low, Moderate and High concentrations. The current EQSD (2013/39/EU) has provisions for eleven substances in biota, including mercury, HCB and benzo(a)pyrene BAP. However, further guidance is needed in order to apply the limits provided by the EQSD, because it is not yet clear how the EQS can be directly applicable to the target tissues used in MAR001 (i.e. liver, fillet or soft body). An EU guidance document in this respect is expected in 2014. Thus limits have been defined based on OSPAR methodology, EU legislation concerning concentrations in relevant foodstuffs (EC no.1881/2006 - only applicable to a few substances relevant for this indicator: cadmium, lead and mercury), or in absence of these, expert judgment is used. 

OSPAR has developed Background Assessment Criteria (BAC) and Environmental Assessment Criteria (EAC), which are being applied in this indicator. EACs were preferred over BACs because EACs entail an assessment of harm to the environment, whereas BACs only provide expected concentrations in mussels and fish that have had presumed low exposure to contaminants. Both EAC and EQS are risk based thresholds designed to provide equally good protection of the environment as to the EQSD. Furthermore, there is an ongoing process at OSPAR to harmonise the EACs with the EQS using the same principles for environmental protection. An EQS in the water column calculated from OSPARs EAC does not always agree with the EQS from the directive. One major issue is the choice of assessment factors and biomagnification factors applied. It should also be noted that EAC does not take into account specific long term biological effects such as carcinogenicity, genotoxicity and reproductive disruption and do not include combination toxicity, the assessment criteria in general, and especially for PAHs and PCBs should not be considered final goals or ultimate targets (see OSPAR (2004) for further cautionary notes). 

Two different types of EACs were derived. The first type is based on the derived EACs for water or sediment, and transferred to biota using an appropriate bioconcentration bactor (BCF). The second type takes into account that fish or mussels are food for predators. Concentrations in mussels or fish can be derived that protect against this so-called secondary poisoning using appropriate biomagnification factors (BMF). BCF is the ratio of the factor by which the result of the uptake, distribution and elimination of a substance in an organism due to waterborne exposure (EU/ Technical guidance document). Biomagnification is the accumulation and transfer of chemicals via the food chain, resulting in an increase of the  internal concentration in organisms at higher levels in the trophic chain (EU/ Technical guidance document) (E.C. 2003).

¹) Blue mussel (Mytilus edulis) for the north-east Atlantic, Mediterranean mussel (M. galloprovincialis) for the Mediterranean and Black Sea.

²) Used in the OSPAR statistical assessment (R.Fryer (Marine Lab., UK) pers. comm.)

Table 2: Limit concentration used for classification in figures and maps in 2014 assessment: 

Low/High concentration limits for spatial assessment, which delimits the classes Low, Moderate (Low < Assessment level < High) and High. EU foodstuff limits are derived from Regulation (EC) No. 1861/2006. Except for EU legislation, the limits have no legal application. All values are expressed in units of µg/kg and on a wet weight (W) except for organics in herring which is expressed on a fat weight (L) basis. Unless otherwise noted, all values are derived from OSPAR Background Assessment Concentration (BAC) or Environmental Assessment Criteria (EAC) or use of the foodstuff limits are as applied in the OSPAR assessment 2013; the exceptions being the High values for DDE and HCB and the Low values for Lindane (g-HCH).

¹) Concerns blue mussel (Mytilus edulis) for the North-East Atlantic, Mediterranean mussel (M. galloprovincialis) for the Mediterranean Sea and Black Sea.

²) Used in the OSPAR statistical assessment (R.Fryer (Marine Lab., UK) pers. comm.)

Methodology for gap filling

The assessment method does not require that time series be complete, so no measures are being taken to fill such gaps. Instead, the method adapts to the existing gaps in the time series.

The regional assessments are based on tallies of results for single time series within each region, without further consideration of geographical distribution within the region. Gaps in the geographical coverage will mean that the regional assessment may not be representative for the region as a whole. There is no attempt to extrapolate results from existing data to estimate conditions in geographical areas without data.

Methodology references

• 1881/2006/EC Commission regulation Commission regulation (EC) No 1881/2006 of 19 December 2006 setting maximum levels for certain contaminants in foodstuffs.
• E.C. 2003, technical document Technical guidance document in support of Commission Directive 93/67/EEC on risk assessment for new notified substances and Commission Regulation (EC) No 1488/94 on risk assessment for existing substances and Commission Directive (EC) 98/8 on biocides. European Commission
• ICES, 2000. Report Of The Working Group on Statistical Aspects of Environmental Monitoring Nantes, France, 27- 31 March 2000 91 pp. ICES, 2000.
• OSPAR 2008 Co-ordinated Environmental Monitoring Programme - Assessment manual for contaminants in sediment and biota. OSPAR Commission, 2008, publication no. 379/2008. ISBN 978-1-906840-20-4. 39 pp..
• OSPAR 2009a Trends in waterborne inputs Assessment of riverine inputs and direct discharges of nutrents and selected hazardous substances to OSAPR maritime area in 1990-2006, OSPAR Commission, publication number 448/2009, 113 pp.. ISBN 978-1-906840-88-4
• OSPAR 2009b CEMP: 2008/2009 Assessment of trends and concentrations of selected hazardous substances in sediments and biota. OSPAR Publication Number: 390/2009. ISBN 978-1-906840-30-3.

Uncertainties

Methodology uncertainty

Aggregated data does not necessarily convey the uncertainty these problems cause.  Also, the time coverage is very variable between series and some of the trends shown may be based on mainly older data. The statistical significance of trends is based on a two-sided test with a nominal 5% significance level, separately for each time series, without regard to serial correlation. Assessments of 'No trend' (i.e. no statistically significant trend) may be due both to actual lack of trend and to insufficient data (too few years in the data series; values in general below reporting limit, giving many ties).

Data sets uncertainty

This assessment is based on data reported to the EEA by EEA member countries, which have significant gaps in terms of availability (geographical and temporal) and consistency, especially for the Mediterranean and Black Seas. These data uncertainties, therefore, hinder more adequate assessment of concentrations and trends of hazardous substances in European marine waters. 

Rationale uncertainty

The classification of levels reflects both observed levels and the uncertainty in trends or levels indicated by data. Thus a Red classification does not necessarily mean that estimated levels are above the limit; it may merely indicate that the uncertainty is so large (because of little data and/or high variability) that it cannot be assessed with reasonable confidence that the true level is below the highest limit. It should also be noted that the three-class system applied for concentrations of hazardous substances in fish does not highlight those cases where there is a risk to human consumption.

More generally, it should also be noted that considerable efforts have been made (i.e. policy, management and research levels) to establish and maintain monitoring programmes to assess the level, trends and effects of hazardous substances in biota, and to select the preferred indicator tissues in particular species. However, there is a lack of reliable and consistent data for many hazardous substances and for several regions. Although basic legislation is in place to combat excessive exposure, specific assessment criteria with respect to levels, trends and effects need to be further developed for the indicator matrices. Furthermore, measurement of concentrations in biota are not coordinated with measurements of inputs, which enhances the uncertainty in the correlation between the two.

It should also be noted that this indicator should not be used as an assessment of compliance monitoring in relation to the Water Framework Directive, mainly because the monitoring strategy and assessment criteria are for the status of hazardous substances in biota, whereas the Water Framework Directive EQS concerns concentrations in water for the most part. The exceptions include mercury and HCB in “prey tissue”, i.e. the whole individual, which would only apply to mussels in this indicator. There is a provision under EQSD that allows Member States to establish other EQS in biota (and sediment) for other substances as long as they would provide the same level of protection. The ongoing development of technical guidance for deriving environmental quality standards for the Water Framework Directive should help Member States if they choose this solution. In addition to this, these indicators will also have to take account of new legislation, for example technical specifications for chemical analysis (2009/90/EC).