1 Deselection criteria for existing Priority Substances based on STE assessment method Contents 1. Data quality required for the deselection of existing priority substances.................................................................. 2 2. Criteria for deselection ............................................................................................................................................... 2 3. Identification of priority substances fulfilling criteria 1 .............................................................................................. 4 4. List of priority substances that could be proposed for deselection ........................................................................... 6 5. Detailed analysis for the substances that fulfil the deselection criteria..................................................................... 7 5.1. Simazine .................................................................................................................................................................. 8 5.2. Alachlor ................................................................................................................................................................. 13 5.3. Chlorfenvinphos .................................................................................................................................................... 17 6. Questions to the WG Chemicals ............................................................................................................................... 22 Annex .................................................................................................................................................................................... 23 7. Comments on the criteria for deselection ................................................................................................................ 23 8. Examples of PS (or "other pollutants") with very low STE score which do not fulfil criteria 2 or 3 ......................... 26 8.1. Atrazine ................................................................................................................................................................. 26 8.2. Benzene................................................................................................................................................................. 29 8.3. 1,2-Dichloroethane ............................................................................................................................................... 32 8.4. Diuron ................................................................................................................................................................... 35 8.5. Isoproturon ........................................................................................................................................................... 38 8.6. Trichlorobenzenes................................................................................................................................................. 41 8.7. Carbon tetrachloride (tetrachloromethane)......................................................................................................... 43 8.8. Tetrachloroethylene ............................................................................................................................................. 46 8.9. Trichloroethylene .................................................................................................................................................. 49
51
Embed
Contents€¦ · 1 Deselection criteria for existing Priority Substances based on STE assessment method Contents 1. Data quality required for the deselection of existing priority
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Deselection criteria for existing Priority Substances based on STE assessment method
Contents 1. Data quality required for the deselection of existing priority substances.................................................................. 2
2. Criteria for deselection ............................................................................................................................................... 2
3. Identification of priority substances fulfilling criteria 1 .............................................................................................. 4
4. List of priority substances that could be proposed for deselection ........................................................................... 6
5. Detailed analysis for the substances that fulfil the deselection criteria ..................................................................... 7
6. Questions to the WG Chemicals ............................................................................................................................... 22
7. Comments on the criteria for deselection ................................................................................................................ 23
8. Examples of PS (or "other pollutants") with very low STE score which do not fulfil criteria 2 or 3 ......................... 26
1. Data quality required for the deselection of existing priority substances
It is important to base the deselection of an existing priority substance (PS) on a comprehensive dataset, to avoid
overlooking ongoing issues in certain Member States. In this regard, the data gathered by the JRC for the review of the
list of priority substances1 may need to be complemented for certain substances, for which monitoring data are missing
for several Member States.
2. Criteria for deselection
The substances are considered for deselection if they belong to the list of priority substances (and 8 "other pollutants")
identified by Directive 2008/105/EC, to be monitored preferentially in water according to Directive 2013/39/EU.
The following substances will not be considered for deselection, as the full picture of contamination at EU level, in the
relevant matrix, will only be available at the end of the second or third river-basin management plan:
- Priority substances recently added to the list by Directive 2013/39/EU,
- Priority substances listed under Directive 2008/105/EC, for which the preferred monitoring matrix in Directive
2013/39/EU is biota,
- Priority hazardous substances (PHS). For PHS, the objectives of the WFD include not only reaching good status, but also
phasing out emissions, discharges and losses. More information would be needed to check whether this last
requirement has been fulfilled, before PHS can be considered for deselection.
Priority substances will be considered for deselection if they do not pose a significant risk at EU level, and if the risk they
pose is not likely to increase significantly as a result of their deselection. This is reflected in the criteria below.
Priority Substances could be considered for deselection if they fulfil all of the following criteria:
1. STE score is below 0.6 (see Table 1) according to the STE run for Scenario2–PNEC-QC for whole water for
organic substances or dissolved phase for metals. A STE score of < 0.6 is taken to indicate a very low risk
to/via the aquatic environment2. Before calculating the STE score, a check should be made of whether the
EQS needs to be revised, and the STE score should be implemented based on the updated EQS.3
2. The substance is measured in more than (≥) 20 countries (Scenario2-PNEC-QC) and the number of countries
with observed EQS exceedances is equal or less than (≤) 4.
1 In the course of gathering data for the selection of new priority substances, the JRC asked MS for data on existing priority substances; these data were used e.g. to test the methodology for the prioritisation. 2 See document summarising the monitoring and modelling methodologies, or full report : Raquel Negrão Carvalho, Dimitar Marinov, Robert Loos, Dorota Napierska, Nicola Chirico, Teresa Lettieri. “Monitoring-based Exercise: Second Review of the Priority Substances List under the Water Framework Directive”. Report: https://circabc.europa.eu/w/browse/52c8d8d3-906c-48b5-a75e-53013702b20a 3 After a comparison of the RQ and STE score for non-PS, it appeared that in some rare cases, the RQ can be above one while the STE score is below 0.6. Obviously we wouldn't want to deselect PS with a RQ above one so the JRC checked and concluded that for all substances considered here, when the STE score was below 0.6, the RQ was also below 1.
3. The substance is banned in the EU, and the ban is not likely to be lifted soon. The substance shows a
temporal decrease in measured mean concentrations or at least no upward trend neither in mean
concentrations nor in the number of exceedances.
Table 1. Risk classification based on STE score (Carvalho, et al. 2016, draft report on Monitoring based exercise).
STE score Risk classification Risk rank
≥ 2.4 and ≤3 Very high 1
≥ 1.8 and < 2.4 High 2
≥ 1.2 and < 1.8 Intermediate 3
≥ 0.6 and < 1.2 Low 4
< 0.6 Very low 5
In some cases, there might still be illegal use of banned substances in some European countries, which could be shown
by almost stable concentrations (for instance annual means) in the monitoring data. Even banned substances that are no
longer used may still be circulating in the environment, and measures to avoid their continued circulation/emission from
one compartment to another (e.g. sediments, soils, groundwater) might be implementable. It may still be necessary to
list/retain such substances to ensure that such measures are taken, including decontamination if necessary. This more
in-depth assessment can only be done once the criteria above have been implemented.
Limitations in trend analysis :
The observed trend in annual mean concentrations could be influenced by the variability in the number of samples,
stations, and countries per year inherent to the dataset. However, it was not possible to identify a clear link between the
number of samples and the mean concentration (see examples provided in section 5 and the annex), nor to detect an
impact of the other factors.
In addition to this, the rainfall can impact the concentrations found in the environment, and thus the observed trend,
by :
- increasing river discharge leading to more dilution (and vice versa)
- increasing or decreasing the amount released to the environment (e.g. for pesticides).
During the 5th SG-R meeting, the experts suggested not to consider the rainfall due to the difficulties to collect the data overall Europe. In this document, for two of the substances proposed for deselection, the data of the annual mean precipitation are shown, however no the correlation with the mean concentration was observed.
When deciding on the deselection of a substance, it is also necessary to assess whether the substance has been
monitored with analytical methods with appropriate LOQs. If not, the monitoring data will not show a full picture of the
risk linked to the substance.
4
3. Identification of priority substances fulfilling criteria 1
In accordance with section 2, the PS and groups of PS considered below are substances to be monitored in the water
compartment, that have not been added by Directive 2013/39/EU, and that are not PHS. This section identifies the PS
and groups of PS fulfilling criteria 1 (STE score <0.6) in the inland whole water compartment. Additional information is
also given on whether the substance is approved or not.
Inland whole water
Table 1: Identification of priority substances from compartment inland whole water with a very low STE score < 0.6
based on the 95th percentile (P95) concentrations (i.e. fulfilling criteria 1). The table shows also: EQS (µg/L), number of
countries with exceedances (by P95 and max concentration), total number of countries with measurements (Sc2-
CAS countries sites samples #samples<lod #samples<loq Substance # in the PS list
#470-90-6 13 224 448 0 0 Chlorfenvinphos 8
21
No need to be performed another additional test considering the precautionary drinking water limit value 0.1 µg/L
(request by the drinking water suppliers to give an indication on whether it could be relevant for MS to consider the
substance under article 7) since EQS is equal to 0.1 µg/L.
Aggregated data from EEA for missing Member States
The aggregated monitoring data for chlorfenvinphos was extracted and found from one additional country with an EQS
exceedance for a lake monitoring station, with a maximum concentration of 0.13 µg/L in 2010 as reported below. This
country is then added to the 5 countries from the STE run.
Year 2010
# of samples 3
Min concentration (µg/L) 0.0033
Mean value (µg/L) 0.0455
Max concentration (µg/L) 0.13
Groundwater concentrations > 0.1 µg/L (from SoE)
The disaggregated data set of EEA contains 41609 monitoring records for chlorfenvinphos in groundwater in 12
countries; only one groundwater monitoring measurement is above the groundwater limit value of 0.1 µg/L (0.138 µg/L)
(years 2006 – 2012), which is 0.002 % of all records.
Conclusion
Chlorfenvinphos could be a potential candidate for deselection. Chlorfenvinphos has been banned in the EU since 2012. For chlorfenvinphos, the EQS exceedances criteria is not totally fulfilled because exceedances were found in 5 countries,
plus one additional country found in the aggregated data from EEA. Exceedence is reported in 2010. If an exception is
made to this criteria it would be possible to consider also chlorfenvinphos for deselection, because 3 of the 5 countries
have only a single exceedance. However from aggregated data we could not conclude if it is only one exceedance,
reported in 2010.
CAS min mean SD median P25 P75 P90 P95 P99 max Substance # in the PS list
- Do you have comments on the criteria outlined in section 2 and on the substances identified as potential candidates for
deselection ?
23
Annex
7. Comments on the criteria for deselection
Eurometaux association comments on the PS deselection criteria:
The STE score limit below 0.6 seems very low and arbitrary. We suggest to derive a limit fixing the STE score
below a certain percentile of STE scores of all investigated chemicals in the exercise.
When determining the STE for metals where the EQS is bioavailability-based, the exceedance needs to be
defined within the bioavailability context. That is, an exceedance is defined as a situation where dissolved
ambient concentrations exceed the bioavailability normalized PNEC. Importantly, ambient dissolved
concentrations that are greater than the reference EQS should not automatically be interpreted as an
exceedance.
When observing the exceedances, the 95th or even 99th percentile should be considered; using the maximal
value is too sensitive to outliers (e.g. unit mistakes).
It seems more sensible to consider a temporal decrease in the number of exceedances, rather than the
average/median concentrations. Average/median concentrations are typically insensitive to reductions of
exceedances because exceedances are usually only in a small fraction of all water samples.
The “banned” criterion is not relevant for metals which occur naturally. A full ban on metals is not appropriate,
because:
under annex X of the WFD, the metals figure as a “metal group” e.g. “Nickel and Nickel compounds”, not as
a single substance (e.g. “NiO”). As such, all different substances (the metal and all its chemical salts) and all
different uses and applications are comprised in the one entry under Annex X. It is highly unlikely that there
would be a general ban on all forms and compounds of a given metal.
metals are natural elements and as such always present in all environmental compartments, including the
water. So they will always be measured in monitoring programmes. At some locations influenced by local
metalliferous geology, the natural background concentrations will be higher, and may exceed the EQS
because of natural occurrence.
metals are present as trace elements in many natural basic products such as fossil fuels and coal. The
industrial process using such basic materials will result in emissions that have nothing to do with the use of
the metals as such, but the emissions will influence water concentrations
Answers to Eurometaux' comments on the PS deselection criteria
The concept for deselection is looking the susbtances that pose very low concern. According to the agreed method
for assessment in the monitoring exercise, an STE score ≤0.6 is taken to represent to very low concern. Currently,
substances showing STE score below 0.6 are 21% (excluding those in biota matrix or being recently added) of the all
substance which have been considered in this exercise while represents the 50% of the not approved PS .
JRC admits that the metals have to be evaluated considering the bioavailability and background concentrations.
However, the data reported by MS don’t provide records neither for the parameters (at same location and time)
needed to calculated the bioavailable factor nor for the background.
JRC considers the procedure for treatment of outliers as suffiently robust to cleaned “obvious” outliers from the
dataset, so, there is not need for further filtering of data
JRC took into account the temporal aspect of the number of exceedances when evaluating the time-trends. We
found unstable (oscilating) number of exceedances per year that cannot be used for a time trend analyses.
JRC agrees that the condition for banning is not applicable for naturally occurring substances (for instance metals).
24
Comments from RIVM (NL) on the criteria for deselection:
1. Data quality for deselection. MS that have found PS below the EQS for a number of years (3?) can reduce the monitoring frequency. It is possible that monitoring data are not available for that reason.4 A reduced monitoring frequency in a number of MS may in itself also be a valid reason for deselection and this should be taken into account when evaluating the monitoring data for PS.
2. We are not in favor of asking MS to provide addition information for PS. This information was already reported by MS.
3. The criterium 1. Is not relevant for deselection and should be left out. 4. One of the criteria for deselection is the STE score. The STE scoring method has changed (or should be changed)
because of the comments we gave in the last SG-R meeting (jun 9-10) This may influence the outcome of the deselection procedure
5. Number of countries with observed EQS exceedance below 4: this is the criterion use for inclusion of compounds in a list of new PS. This was chosen as a very precautionary value not to miss possibly relevant compounds. I suggest to consider the following: number of countries with observed EQS exceedance below 8 (= 30% of the MS) AND a decreasing trend.
6. Suggest NOT all of the listed criteria should be fulfilled: a substance that is not banned but is below EQS in all MS should also be deselected!
Suggested criteria:
- Substance with observed EQS exceedance in less than 8 countries and decreasing trend OR
- Substance banned, observed EQS exceedance in less than 8 countries and decreasing trend 7. Not in favor of including the criterion substance is banned also in import products. I think this criterion is beyond
the scope of the WFD
Answers to RIVM’s comments:
JRC agrees that the certain MSs may have not monitored a priority substance because it was deemed not relevant in
the MS (e.g. not discharged). If a substance is proposed for deselection, the JRC will make sure with the
corresponding MSs that this is the reason for the lack of monitoring data. If the substance was actually monitored in
several MS that did not send the corresponding data, the JRC may need to request these MSs to provide monitoring
data, to confirm the deselection of the substance (and avoid deselecting a substance that would cause a risk in
several MSs)
The criterion related to the ban of substances will be discussed with the experts from SG-R. Because of the
requirement to phase out emissions, discharges and losses of PHS (in addition to the requirement to reach good
status for these substances), PHS will not be proposed for deselection from the list.
Former criteria 1 was redundant with the introductory sentences to the section and it has been removed.
The criteria with observed EQS exceedances in equal or less than four countries has been proposed for precaution.
We can take into account the proposed criteria by NL (EQS exceedances in less than 8 countries (= 30% of the MS))
and will evaluate and compare the difference (see section 3 of this document) .
The suggestion (comment #6) could be considered at later stage and it should be verified one by one substance
since as we mentioned already the available monitoring data are not from all MS, so in case it is not banned and no
exceedance is only reported from some MS, this would not be enough as deselection criteria. It is also necessary to
make sure that deselecting the substance would not lead to the suppression of the measures put in place to control
4 According to Annex V 1.3.1, the frequency of monitoring can be reduced when "the previous surveillance monitoring exercise showed that the body concerned reached good status and there is no evidence from the review of impact of human activity in Annex II that the impacts on the body have changed."
25
the emissions and that it would not lead to an increase in the risk linked to the substance. At this stage of the
process, it is thus proposed to focus on banned substances for which we are sure that no increase in emissions
would occur following deselection.
Cooment # 7 has been taken into account in the updated version.
Comments from Umweltbundesamt GmbH (AT) on the criteria for deselection
It is questioned, whether the ban of a substance should be a requirement for such a delisting. If concentrations in
European rivers are shown to decrease and no EQS exceeding’s are reported by MS in the RBMPs, substances might be
deselected from the list of priority substances although no ban is established for those substances. Monitoring data for
an evaluation should be available. A requirement for the listing of priority substances is their relevance at European level
with regard to potential risks. The single occurrence/presence of a substance in a water body is not a selection criterion.
Following this reasoning, substances for which risk management measures were implemented and due to this measures
the concentrations decreased under the EQS values, no justification for listing these compounds further as priority
substances can be deduced. It is different for priority hazardous substances, as for these compounds also a phase out is
required. For these substances a reduced monitoring obligation could be discussed.
Answers to Umweltbundesamt GmbH comments:
If we were to deselect a PS still in use, it would be necessary to make sure that deselecting the substance would not
lead to the suppression of the measures put in place to control emissions and that it would not lead to an increase in
the risk linked to the substance. At this stage of the process, it is thus proposed to focus on banned substances for
which we are sure that no increase in emissions would occur following deselection.
26
8. Examples of PS (or "other pollutants") with very low STE score which do not fulfil criteria 2 or 3
Note: Attention should be paid to the fact that for some substances the Fspatial (spatial factor in STE method) could
become null since the country term of exceedances based on 95-percentile of concentrations is zero. . In addition,
following the decisions of the 5th meeting of the SG-R (June 2016), the Temporal and Extent STE factors are synchronized
with Fspatial (when Fspat=0 then automatically Ftemp=0 and Fext=0). Some exceedances may exist while the STE score
is equal to zero.The boxplots in this document show all data for Inland Whole water Sc2-PNECQC (including max
concentrations; without any data filtering).
8.1. Atrazine
STE assessment results: spatial, temporal and extent factors, STE score, risk rank, and AA-EQS (µg/L).
Although the number of countries with exceedances is 5 (there are really a very small number of samples that exceed EQS). 1,2-Dichloroethane could not be a candidate for deselection according to the proposed criteria since it is approved for use.
35
8.4. Diuron
STE assessment results: spatial, temporal and extent factors, STE score, risk rank, and AA-EQS (µg/L). The temporal and
extent factors are harmonised with Fspatial (when Fspat=0 then automatically Ftemp=0 and Fext=0).
CAS Fspat Ftemp Fext STE score Risk Substance EQS
#330-54-1 0 0 0 0.000 5 Diuron 0.2
STE additional information: Fspatial (site and country frequency of exceedances found by P95 and max concentration),
and Fextent (extent of exceedance by Risk Quotient_P95.