Content uploaded by Tom Dolan
Author content
All content in this area was uploaded by Tom Dolan on Nov 28, 2023
Content may be subject to copyright.
1Is the EU Drinking Water Directive Standard for Pesticides in
2Drinking Water Consistent with the Precautionary Principle?
3Tom Dolan, Peter Howsam, David J. Parsons,*and Mick J. Whelan
4School of Applied Sciences, Cranfield University, Cranfield, Bedfordshire, MK43 0AL
5ABSTRACT: Regulations based on the precautionary principle should
6undertake a comprehensive assessment of all available scientific and
7technical data to identify sources of epistemic uncertainty. In the European
8Union (EU), environmental regulation is required to fulfill the principles
9established in Article 174 of the EU Treaty, such that it offers a high level
10 of protection and is consistent with the precautionary principle. Pesticides
11 in drinking water are currently regulated by the Drinking Water Directive
12 using a maximum allowable concentration of 0.1 μg/L. This standard (a
13 surrogate zero) was consistent with the precautionary principle when it
14 was originally set in 1980 and remained consistent when retained in 1998.
15 However, given developments in EU pesticide and water policy,
16 international experience in regulating pesticides, and an increasing
17 knowledge of pesticide toxicity, it can be argued that the level of epistemic
18 uncertainty faced by regulators has substantially decreased. In this paper, we examine the extent to which such developments now
19 challenge the basis of European drinking water standards for pesticides and whether, for substances for which there is good
20 toxicological understanding, a regulatory approach based upon the World Health Organization (WHO) Guideline Value (GV)
21 methodology would be more consistent with the principles underpinning European environmental policy.
1. INTRODUCTION
22 Article 174 (formerly Article 130r) of the 1992 Maastricht
23 Treaty
1
defines a set of principles for the formation of
24 environmental policy in the European Union (EU). Central to
25 Article 174 is the aim to provide a high level of protection for
26 human health and to develop policy based upon available
27 scientific and technical data, the precautionary principle,
28 preventive action at source, and the “polluter pays”principle.
29 The revised Drinking Water Directive (DWD)
2
was
30 established in 1998 in line with the principles of Article 175
31 (formerly Article 130s1). Article 1.2 of the DWD states: “The
32 objective of this Directive shall be to protect human health
33 from the adverse effects of any contamination of water intended
34 for human consumption by ensuring that it is wholesome and
35 clean.”
36 To achieve this objective, the DWD sets maximum allowable
37 concentration (MAC) values for several chemical parameters in
38 drinking water. For pesticides, the MAC is 0.1 μg/L for any
39 individual active substance (a 0.03 μg/L standard applies to
40 four exceptions: aldrin, dieldrin, heptachlor, and heptachlor
41 epoxide) and 0.5 μg/L for the total pesticide concentration.
42 The MAC of 0.1 μg/L was intended as a surrogate zero because
43 it was indicative of a typical limit of quantification for trace
44 organic compounds when it was first established (1980).
3−5
45 Historically, compliance with these standards in mainly
46 agricultural catchments used for water supply has been
47 primarily achieved through the installation of water treatment
48 infrastructure. However, Article 7 of the Water Framework
49 Directive (WFD)
6
is driving a prevention-led approach to
50 compliance with drinking water standards, spreading responsi-
51bility for DWD compliance around all stakeholders in a
52catchment and aligning the DWD more closely with the
53polluter pays principle. WFD Article 7 has cast fresh scrutiny on
54the pesticide standard in the DWD, in part because the DWD
55MAC is applied absolutely with no allowance for low frequency
56periodic exceedence.
57This paper examines whether a 0.1 μg/L MAC for every
58active substance is consistent with the principles of Article 174
59and the precautionary principle. To do this, it addresses four
60key questions:
61
•What does the EU understand by the term precautionary
62
principle, and is this consistent with the reasons to use
63
the precautionary principle given in the academic
64literature?
65
•Why was a 0.1 μg/L MAC for pesticide active substances
66
in drinking water set in 1980 and retained in 1998? Were
67
these decisions compatible with the precautionary
68
principle, available scientific and technical data, and
69with EU Treaty Article 174?
70
•Have advances in scientific understanding and the
71
availability of technical data since 1998 been sufficient
72
to undermine the original justification for the 0.1 μg/L
73
MAC standard in terms of its consistency with Article
74174 and interpretations of the precautionary principle?
Received: December 4, 2012
Revised: April 5, 2013
Accepted: April 16, 2013
Policy Analysis
pubs.acs.org/est
© XXXX American Chemical Society Adx.doi.org/10.1021/es304955g |Environ. Sci. Technol. XXXX, XXX, XXX−XXX
slw00 |ACSJCA |JCA10.0.1465/W Unicode |research.3f (R3.5.i1:3915 |2.0 alpha 39) 2012/12/04 10:21:00 |PROD-JCA1 |rq_1299057 |4/25/2013 11:29:32 |9
75 •Are alternative regulatory options for protecting drinking
76 water quality available to the EU, and are these
77 compatible with Article 174?
78 2. Policy Analysis. 2.1. EU Interpretation of the
79 Precautionary Principle and Its Use. What does the EU
80 understand by the term precautionary principle, and is this
81 consistent with the reasons to use the precautionary principle
82 given in the academic literature?
83 There are many definitions for the precautionary principle
7,8
84 and “a lively debate has arisen concerning its actual meaning
85 and practical application”.
9,10
Therefore, it is important to
86 define, for the purposes of this paper, the precautionary
87 principle, the situations in which it is an appropriate risk
88 management tool, and the EU position regarding criteria for
89 regulatory decisions based upon it.
90 In 1992, the Rio Declaration on Environment and Develop-
91 ment
11
declared 27 principles for sustainable development.
92 Principle 15 defines the precautionary principle as follows:
93 “In order to protect the environment, the precautionary
94 approach shall be widely applied by States according to their
95 capabilities. Where there are threats of serious or irreversible
96 damage, lack of full scientific certainty shall not be used as a
97 reason for postponing cost-effective measures to prevent
98 environmental degradation.”
11
99 Except for the addition of the term “cost effective”, this is the
100 same definition as that agreed as part of the Bergen Declaration
101 in 1990.
12
102 2.1.1. Precautionary Principle in the EU. In 1992, Article
103 174.2 of the Maastricht Treaty, which defined the principles for
104 environmental policy in the EU, made the first reference to the
105 precautionary principle in European policy:
106 “Community policy on the environment shall aim at a high
107 level of protection .... It shall be based on the precautionary
108 principle and on the principles that preventive action should be
109 taken, that environmental damage should as a priority be
110 rectified at source and that the polluter should pay.”
1
111 However, despite having been written into primary EU Law,
112 the precautionary principle is itself not defined in the EU
113 Treaty.
13
A European Commission (EC) communication on
114 the precautionary principle
14
is the clearest available guidance
115 its use in the EU. This communication recommends a
116 structured approach to the analysis of risk, comprising
117 assessment, management, and communication. It identifies
118 the precautionary principle as a risk management tool that is
119 appropriate where the risk assessment has identified scientific
120 uncertainty. When invoking the precautionary principle, the
121 rationale for a decision must be made transparent, the standards
122 set may not be arbitrary, and standards must be in keeping with
t1 123 the five principles of risk management shown in Table 1.
124
Therefore, it can be argued that, in the EU, once the decision
125
has been made to invoke the precautionary principle in support
126
of achieving a high level of protection, the standards set must
127
be in keeping with the above principles. The environmental
128
policy will then be compatible with the principles of both the
129precautionary principle and EU Treaty Article 174.
130
2.1.2. Application of Precautionary Principle. Academic
131
literature on the precautionary principle makes it clear that
132
there should be no conflict between the precautionary principle
133
and the scientific principles of risk assessment, and that its use
134
in environmental policy formation can be consistent with a
135
scientific approach.
7,9,10,15,16
This is in line with the EC
136
communication and the Rio Declaration, both of which
137
recommend recourse to the precautionary principle when
138scientific uncertainty is present. The subject of when recourse
139
to the precautionary principle should be made is widely covered
140 t2in the literature (Table 2).
141
Many authors distinguish two or more types of uncertainty.
142
In the terminology used by Aven,
7
epistemic uncertainty is
143
defined as arising from insufficient knowledge and aleatory
144uncertainty is caused by natural random variation. Because it is
145
generally agreed
7,15,16
that the precautionary principle should
146
be invoked whenever there is epistemic uncertainty, the
147
decision to invoke it should be preceded by a comprehensive
148
assessment of available scientific and technical data in order to
149
identify potential sources of such uncertainty. The level of
150
regulation set under the precautionary principle must be
151
sufficient to manage the epistemic uncertainty identified. These
152
concepts are used later in the paper to assess whether current
153
policy approaches to protecting drinking water quality are
154consistent with the precautionary principle.
155
2.2. Is the EU MAC for Pesticides Justified? Why was a
156
0.1 μg/L MAC for pesticide active substances in drinking water
157
set in 1980 and retained in 1998? Were these decisions
158
compatible with the precautionary principle, available scientific
159understanding and technical data, and with EU Treaty Article
160174?
161
The DWD sets a MAC of 0.1 μg/L for all individual pesticide
162
active substances in drinking water at the point of supply
163
(except for 0.03 μg/L for aldrin, dieldrin, heptachlor, and
164
heptachlor epoxide). The origin of this standard predates both
165the current DWD and the decision to embody the precau-
166
tionary principle in Article 174 (originally Article 130r) of the
167
Maastricht Treaty. The 0.1 μg/L standard first appeared in
168
1980 in Directive 80/778/EEC.
17
Prior to this, a 0.5 μg/L
169
standard was set for total pesticides in Directive 75/440/
170
EEC.
18
Both of these standards are believed to have their origin
171
in the philosophy of the first and second environmental action
172
plans running from 1973 to 1977 and 1977 to 1981, i.e., that
173
pesticides should not be present in drinking water regardless of
174the actual risks posed.
4,19
Table 1. Principles of Risk Management for Precautionary Principle Policy Design
a
Principle Detail
Proportionality Measures should be proportional to the desired level of protection.
Nondiscrimination Measures should not be discriminatory in their application.
Consistency Measures should be consistent with the measures already adopted in similar circumstances or using similar approaches.
Examination of the
benefits and costs of
action
Measures adopted presuppose examination of the benefits and costs of action and lack of action. Benefit and cost evaluation may involve
economic, efficacy, and the socio-economic impact analysis, as well as evaluation of noneconomic considerations.
Examination of scientific
developments Measures, although provisional, shall be maintained as long as the scientific data remain incomplete, imprecise, or inconclusive and as
long as the risk is considered too high to be imposed on society.
a
Adapted from European Commission COM/2000/1
14
.
Environmental Science & Technology Policy Analysis
dx.doi.org/10.1021/es304955g |Environ. Sci. Technol. XXXX, XXX, XXX−XXXB
175
In 1980, at the time the standards for pesticides in drinking
176
water were adopted, relatively little was known about the
177
impacts of pesticides on human health.
4
Pesticides were poorly
178
regulated in the EU, and there was the perception that the scale
179
of health impacts could be large. Therefore, when analyzed
180
against the reasons for recourse to the precautionary principle
181as given by Stirling and Gee,
16
Klinke and Renn,
15
and Aven,
7
182the decision to adopt 0.1 μg/L as a surrogate zero in order to
183
prevent exposure to pesticides through drinking water can be
184justified.
185
In the language of Stirling and Gee,
16
policy makers were in a
186
position of “ignorance”because of insufficient knowledge
187
regarding health outcomes from exposure. In other words, the
188health risk from exposure could not be classified as “normal”
189
because the certainty of assessment for both extent of damage
190
and probability of occurrence was low (Table 2), and there was
191
a perception of high catastrophic potential from exposure and
192
the need for regulators to make a decision at the limits of
193
human knowledge. Using the classification from Aven,
7
the
194
uncertainty present was epistemic, warranting recourse to the
195
precautionary principle. Furthermore, the decision to set a
196
standard as a surrogate zero could also be justified because, for
197
many pesticides (but not all), the only exposure level where
198
“ignorance”/epistemic uncertainty could be replaced with
199scientific certainty was zero.
200
In 1998, Directive 80/778/EEC was replaced by the current
201
Drinking Water Directive (98/83/EC). Following much
202
debate, the decision was taken to retain the 0.1 μg/L standard
203
for pesticides.
4
At the stage of this decision, pesticide active
204
substances were beginning to be much more tightly regulated in
205
Europe following the introduction of Directive 91/414/EEC,
20
206
but the pesticide review initiated by this Directive was not
207
complete until 2009.
21
Therefore, in 1998 many pesticides
208remained available for sale despite significant uncertainty
209
regarding their potential effects on human health. Furthermore,
210
the EC had not formally stated its interpretation of the
211
precautionary principle,
13
EU chemical policy was still
212
grappling with insufficient knowledge,
22
and the WFD
6
did
213
not yet exist. It can be argued, therefore, that the judgment in
214
1998 to extend the 0.1 μg/L standard was also justifiable based
215upon a shortage of available scientific and technical data.
216
However, it is pertinent to point out that, at this stage,
217
perception in the general public and among politicians was that
218
the human health risks associated with pesticides were high. As
219
a consequence of this prevailing negative public opinion about
220
pesticides, it needs to be acknowledged that the decision in
2211998 is likely to have had a political dimension.
222A key question now is whether there has been a significant
223
change in scientific understanding and available technical data
224
since 1998 that might be sufficient to prompt a review of the
225
DWD standard in terms of its consistency with the principles of
226
European environmental policy (Article 174 and the precau-
227tionary principle).
2282.3. Advances in Policy and Scientific Understanding
229Since 1998. Have advances in scientific understanding and the
230
availability of technical data since 1998 been sufficient to
231
undermine the original justification for the 0.1 μg/L MAC
232
standard in terms of its consistency with Article 174 and
233interpretations of the precautionary principle?
234
Any case for the DWD pesticide standard to be reviewed
235must demonstrate that increased scientific understanding and
236technical knowledge is now available to address the epistemic
237uncertainty present in 1998. Scientific developments in the
Table 2. Synthesis of Influential Discussion of Precautionary Principle in Literature
Paper: Stirling and Gee
16
Synopsis: Based upon risk as a function of likelihood and magnitude, Stirling and Gee distinguish between four states of incertitude: “risk”,“ambiguity”,“uncertainty”, and “ignorance”. These states are based upon the ability to define outcomes
and assign probabilities to these outcomes. In these definitions, “risk”is where outcomes are known and there is some basis to assign probabilities allowing “risk”to be managed without recourse to the precautionary principle. Where
uncertainty, ambiguity, or ignorance exist they must be acknowledged, and policy based upon the precautionary principle is needed to avoid unexpected outcomes.
Paper: Klinke and Renn
15
Synopsis: Klinke and Renn
15
define three risk types: “normal”,“intermediate”, and “intolerable”, which are based upon knowledge of the “extent of damage”“E”and the “probability of occurrence”“P”.“Normal”risk types are those that can be
managed using the conventional tools of risk management. Implementation of the precautionary principle is most strongly recommended to manage either intermediate or intolerable risk where the certainty of assessment for either E or P is
low or where there is believed to be high catastrophic potential or incomplete systematic knowledge, or where the evaluation of risk has identified that regulatory decisions need to be made at the limits of human knowledge.
Paper: Aven 2011
7
Synopsis: Aven 2011
7
critiques the position taken by Stirling and Gee with the argument that risk assessment can use both objective and subjective probabilities, and therefore, the term “risk”as defined by Stirling and Gee is too narrow. Aven
argues that the precautionary principle should only be used when faced with scientific uncertainty defined as epistemic uncertainty and arising from insufficient knowledge. He proposes two circumstances where the precautionary principle
should be used in policy design to manage risks arising from epistemic uncertainty (incomplete knowledge). These situations are where it is difficult to specify a set of possible consequences or difficult to establish an accurate prediction model.
Environmental Science & Technology Policy Analysis
dx.doi.org/10.1021/es304955g |Environ. Sci. Technol. XXXX, XXX, XXX−XXXC
238 toxicology of pesticides occur periodically.
23−25
However,
239 significant uncertainties remain, for example, about mixture
240 toxicity,
26,27
which can sometimes be accounted for using
241 conservative safety factors, and about the relevance of
242 metabolites.
28
In addition, three factors have increased our
243 knowledge of potential pesticide health impacts, chemical
244 properties, and use patterns, and improved the effectiveness of
245 pesticide regulation since 1998. These are (1) developments in
246 EU Policy, (2) increased application of the WHO GV method
247 for regulating threshold chemicals in drinking water, and (3)
248 international experience in regulating pesticides.
249 2.3.1. Developments in EU Policy. Analysis of EU policy
t3 250 (Table 3) demonstrates that several policy barriers are now in
251 place to control exposure to pesticides in drinking water. In all
252 cases, the strength of each barrier has increased since 1998.
253 Thus, taken together, these barriers decrease dependence on a
254 stringent drinking water standard for the protection of human
255 health and increase the knowledge base from which drinking
256 water quality can be regulated. The significance of each barrier
257 is detailed below.
258 Barrier 1: EU Pesticide Approval Legislation. The first
259 barrier to control the presence of pesticides in drinking water is
260 the EU pesticide approval legislation. This legislation governs
261 which chemicals can and cannot be marketed for use as
262 pesticides. Initially, in the form of Dir. 79/117/EEC,
29
this
263 barrier allowed any chemical to be marketed as a pesticide
264 provided it did not contain mercury or one of eight persistent
265 organo−chlorine compounds specified in Annex I of the
266 Directive. In 1991, Dir. 91/414/EEC
20
strengthened this
267 barrier by introducing the need for all active substances to
268 gain prior approval before they could be marketed as pesticides
269 in Europe. Implementation took the form of a 16 year review,
270 during which each active substance was subjected to a
271 comprehensive evaluation of hazards posed to various end
272 points and the likelihood of human and environmental
273 exposure. This evaluation also included assessment of the
274 propensity for land to water movement (Article 5).
20
The data
275 requirements for the evaluation were specified in Annex II of
276 the Directive. The review was completed in 2009, at which
277 point 26% of active substances had passed the review, 7% were
278 not approved, and 67% had been removed from the market
279 prior to submission for review.
30
The introduction of Dir. 91/
280 414/EEC gave greater certainty about the chemistry and
281 toxicology of those substances being used as pesticides,
282 removed the most dangerous chemicals from the market,
283 shifted the burden of proof to pesticide producers, and gave a
284 knowledge-base from which pesticides could be regulated into
285 the future. In addition, the review solved for pesticides the
286 problem of different regulations for new and existing
287 substances, which persisted in some areas of EU Chemical
288 policy
22
until REACH (Regulation 1907/2006),
31
the Euro-
289
pean Community Regulation on general chemicals and their
290safe use, entered into force on first June 2007.
291
As part of the EU thematic strategy for pesticides, new
292
legislation was introduced in 2009.
32
Regulation 1107/2009
32
293
further increased the stringency of the prior-approval barrier
294
put in place under Directive 91/414/EEC by introducing
295
several hazard-based criteria (Annex II),
32
which have the
296
intention of removing active substances with properties such as
297
carcinogenicity, mutagenicity, and reproductive toxicity (in-
298
cluding endocrine disruptors) from the market, regardless of
299
their potential for human or environmental exposure. All
300
pesticide active substances scheduled for approval from 2011
301
onward will need to satisfy the requirements of Regulation
302
1107/2009, as will all new active substances. Relevant
303
metabolites, i.e., those judged to have comparable intrinsic
304
properties to the active substance,
33
are also evaluated as part of
305
the pesticide approval process defined in Directive 91/414/
306
EEC and Regulation 1107/2009 and may prevent approval of
307an active substance.
308
In addition, Barrier 1b requires that any plant protection
309
product containing an approved active substance must also gain
310
approval at member state or zonal level before the product can
311be used (Article 28 of Regulation 1107/2009).
32
312
Barrier 2: Water Framework Directive (WFD). Various
313
elements of the WFD
6
can potentially have an impact on the
314
concentration of pesticide active substances in”raw”(untreated)
315
water. The WFD can, thus, be considered as Barrier 2, which
316
can be broken down into those elements acting at receptor
317
(Barriers 2a−c) and those elements acting at source (Barrier
318
2d). The primary purpose of Barriers 2a−c is not the protection
319
of drinking water from pesticides, although they do, never-
320
theless, have some impact on the presence of pesticides in
321
“raw”water. Barrier 2a is the requirement for “no deterioration”
322
as specified in Articles 1 and 4 of the WFD; it is the starting
323
point for the WFD objective to “protect, enhance, and restore”
324
(Article 4.1a) all surface water bodies and to “prevent or limit
325
the input of pollutants into groundwater”(Article 4.2b). Barrier
326
2b is the requirement, in Article 16, for the identification of
327
priority substances and priority hazardous substances to enable
328
assessment of the chemical status of a water body in order to
329
support the “progressive reduction of discharges, emissions, and
330
losses of priority substances and the cessation or phasing-out of
331
discharges, emissions, and losses of the priority hazardous
332
substances”.
6
Environmental quality standards (EQS) have
333
been set for 11 pesticide active substances.
34
However, only
334
one of these active substances remains approved for use.
35
The
335
ability to designate priority substances ensures that any active
336
substance considered a potential problem for water quality is
337
identified and targeted under the WFD. That 10 of the 11
338
pesticide active substances identified as priority substances or
339hazardous priority substances have now been removed from the
Table 3. EU Policy as a Multi-Barrier Approach to Pesticide Regulation and Drinking Water Policy
Number Barrier Point of Action Range of Influence
1a EU Pesticide Approval Legislation: Dir. 79/117/EEC, Dir. 91/414/EEC, and Reg. 1107/2009 source pathway receptor supply
1b Member state plant protection product approval policy source pathway receptor supply
2a WFD “No Deterioration Objective”(Art. 1 and 4) receptor source pathway
2b WFD “EQS for Chemical Status”(Art. 16) receptor source pathway
2c WFD “EQS for Ecological Status”receptor source pathway
2d WFD Art. 7 on waters used for the abstraction of drinking water supply pathway receptor source
3a EU Dir. 2009/128/EC on the sustainable use of pesticides source pathway receptor
4 Drinking Water Legislation: Dir. 75/440/EEC, Dir. 80/778/EEC, Dir. 98/83/EC supply point N/A
Environmental Science & Technology Policy Analysis
dx.doi.org/10.1021/es304955g |Environ. Sci. Technol. XXXX, XXX, XXX−XXXD
340 market under the independent action of pesticide approval
341 regulation is testament to the strength of Barrier 1 and its
342 ability to prevent the release of highly hazardous substances
343 into the environment. Barrier 2c is linked to the assessment of
344 ecological status and involves the identification at member state
345 level of “other substances being discharged in signif icant quantities
346 into the body of water”(Annex V).
6
These “specific pollutants”
347 are distinct from the priority substances (Article 16), and
348 environmental quality standards (EQS) are set for these by the
349 member state in order to support compliance with WFD
350 ecological status targets.
351 Barrier 2d is WFD Article 7, which refers to waters used for
352 the abstraction of drinking water. It requires the creation of
353 protected areas where water is abstracted, emphasizes the need
354 for no deterioration in “raw”water quality, and aspires toward
355 DWD compliance without installation of additional treatment.
356 Therefore, in WFD Article 7, protected areas the management
357 of “raw”water quality becomes a priority for river basin
358 managers, water suppliers, and all pesticide users. Although
359 compliance with the DWD is measured at the point of drinking
360 water supply to customers, the intention of WFD Article 7 is to
361 create the incentive for action at the point of application (i.e.,
362 the primary pollution source), action to minimize movement to
363 the water body (the pollution pathway), and action in
364 abstracted water bodies (the pollution receptor)
35
to reduce
365 pesticide concentrations in “raw”water to levels that can be
366 removed by current treatment, thereby ensuring DWD
367 compliant drinking water. Barrier 2d is an aspiration, and it
368 remains to be seen whether WFD Article 7 compliance will be
369 achieved across Europe in 2015 for all pesticide active
370 substances without the need for increased treatment.
371 Barrier 3: EU Directive 2009/128/EC on Sustainable Use of
372 pesticides. The Sustainable Use Directive (09/128/EC)
36
is
373 the second element of the EU thematic strategy for pesticides.
374 It is intended to act at source to promote best practice for
375 responsible pesticide use and along pathways to reduce the
376 movement of pesticides to water. Many of the requirements of
377 this Directive are very similar to the good practice and
378 responsible pesticide use recommended in the U.K. by
379 voluntary industry-run schemes such as the Voluntary Initiative
380 and the Metaldehyde Stewardship Group.
35
The Directive
381 applies to the aquatic environment and drinking water through
382 Article 11.1 and aims to deliver actions that strengthen Barriers
383 1 and 2. However, at the time of writing, the degree to which
384 this Directive is likely to be effective remains uncertain because
385 the December 2011 deadline for transposition into member
386 state law has only recently passed.
387 Barrier 4: Drinking Water Directive. The DWD
2
could now
388 be seen as a final barrier acting at the point of supply to prevent
389 the presence of pesticides in drinking water. However, the
390 DWD standard for pesticides has its origins at a time before
391 Barriers 1, 2, and 3 were in place and when insufficient scientific
392 knowledge or technical data were available regarding the
393 human health impacts of exposure to pesticides at any level. In
394 addition, little was known about which pesticides were likely to
395 be found in “raw”water. For this reason, it regulates based
396 upon the assumption that it is the only barrier and that all
397 exposure to pesticides must be avoided.
398 2.3.2. World Health Organization (WHO) Guideline Value
399 (GV). The WHO has published international standards for
400 drinking water since 1958. In 1984, the WHO published the
401 first edition of “Guidelines for Drinking Water Quality”, which
402 was subsequently followed by further editions in 1993,
37
403
2004,
38
and 2011.
39
The aim of these Guidelines is to “provide
404
a scientific point of departure for national authorities to develop
405
drinking water regulations and standards appropriate for the
406
national situation.”
38
The WHO Guidelines are recognized as
407
representing the UN position on issues of drinking water
408quality and health by “UN-Water”.
38
409
The most recent editions of the guidelines
37−39
include
410
guidance on chemical aspects. This differentiates between
411
threshold and non-threshold chemicals when determining a
412
regulatory approach to protect human health. Non-threshold
413
chemicals are substances that pose a theoretical risk to human
414
health at any exposure level (e.g., genotoxic carcinogens) and
415
should be regulated at source. Threshold chemicals are those
416
where daily exposure below a certain level will have no adverse
417
health effects and can, thus, be regulated with reference to
418
guideline values (GVs), which are designed to prevent chronic
419
health effects over a 70 year lifetime. A method for the
420
calculation of guideline values was first proposed and applied in
421
the second edition (1993); this has been applied to a broader
422
range of chemicals in subsequent editions (2004 and 2011) of
423the guidelines.
424
For threshold chemicals, the GV methodology uses no
425
observed adverse effect level (NOAEL) or lowest observed
426
adverse effect level (LOAEL) data to calculate a tolerable daily
427
intake (TDI). The TDI accounts for four sources of residual
428
uncertainty in the toxicological data via uncertainty factors
429
(UFs). These arise from interspecies variations (the use data
430
from animal studies), intraspecies variation (difference between
431
individual humans), data quality, and uncertainty regarding the
432
nature or severity of exposure above the NOAEL/LOAEL.
40
433
When calculating a GV from a TDI, it is recognized that
434
drinking water is not the only source of daily exposure to a
435
threshold chemical. The GV method is applicable to many
436
chemicals, including pesticide active substances. Unlike the
437
DWD MAC, it recognizes that different active substances have
438
different toxicities; i.e., at the same level of exposure, not all
439pesticides pose an equal risk to human health.
440
2.3.3. International Experience Regulating Pesticides. The
441
WHO reports GVs for 32 pesticide active substances. Forty
442
other active substances were evaluated, 27 of which were
443
judged as “unlikely to occur in drinkin water”and 13 of which
444
were judged to “occur in drinking water at concentrations well
445
below those of health concern”; therefore, GVs for these are
446
not given. In Australia, the WHO method has been applied as
447
part of the National Water Quality Management Strategy
41
to
448
calculate GVs for 154 pesticide active substances, including
449
several pesticides judged “unlikely to be found in drinking water
450
at levels that may cause health concerns”. The current
451
Australian standards
41
include information on the derivation
452
of the GV and include a full and transparent justification of all
453
the assumptions made, including the selection of UFs and the
454fraction of exposure assumed to occur via drinking water.
455
In the United States, as part of the Safe Drinking Water Act,
456
the Environmental Protection Agency (EPA) is responsible for
457
the identification of contaminants for inclusion in national
458
primary drinking water regulation. To be included in primary
459
drinking water regulation, a contaminant must first be included
460
on a contaminant candidate list (CCL) for further evaluation of
461
health effects, occurrence, and analytical methods.
42
If
462
prioritized for inclusion in primary regulation, a maximum
463
contaminant level goal (MCLG) is calculated for the
464contaminant. To date, only 20 pesticide active substances
Environmental Science & Technology Policy Analysis
dx.doi.org/10.1021/es304955g |Environ. Sci. Technol. XXXX, XXX, XXX−XXXE
465 have been subject to MCLG values in national primary drinking
466 water regulation.
43
467 The MCLG is equivalent to the WHO GV and is calculated
468 using a similar methodology.
40
NOAEL or LOAEL and UF
40
469 values are applied to calculate a reference dose (RfD)
470 (equivalent to a TDI.) The RfD is then converted to a
471 MCLG using standard assumptions of adult body mass, daily
472 water consumption, and the level of exposure through potable
473 water. Unlike the WHO GV process, an additional step is taken
474 to derive a legally enforceable standard or maximum
475 contaminant level (MCL). Whereas the MCLG is based purely
476 on public health considerations, the MCL considers both best
477 available treatment technology and cost in order to set a
478 standard that “maximizes health risk reduction benefits at a cost
479 that is justified by the benefits”.
44
It is possible, therefore, for an
480 MCL to be set at a higher concentration than the MCLG.
481 However, like the WHO, for known carcinogens the EPA
482 reverts to the precautionary approach and sets the MCL at zero
483 because any exposure could present a cancer risk.
484 In the EU, all pesticide active substances currently on the
485 market are regulated by the DWD, whereas GVs have been
486 derived for just 32 substances by the WHO and for 154
487 substances in Australia. In the United States, MCLs have been
488 derived for just 20 substances. Furthermore, the MAC is a more
489 stringent standard than the equivalent GVs and MCLs for all
t4 490 substances (Table 4).
491 With only six exceptions, the WHO GVs allow concen-
492 trations at least 20 times the DWD MAC. None of those
493 exceptions (aldrin and dieldrin, chlordane, 1,2-dibromoethane,
494 cynazine, endrin) is currently approved for use in the EU.
495 In Australia, with only seven exceptions, standards allow
496 concentrations at least 10 times the DWD MAC. The
497 exceptions are carbophenothion, fenamiphos, fipronil, para-
498 thion-methyl, pirimiphos-ethyl, profenofos, and terbufos, each
499 of which is allowed at concentrations at least 3 times the DWD
500 MAC. Only two of these exceptions are approved in Europe,
501 and these are expected to be candidates for substitution when
502 reassessed under Regulation 1107/2009.
44
In the United States,
503 most MCLs are at least 20 times the DWD MAC.
504 In summary, the introduction of prior approval regulation for
505 pesticides and other developments in EU pesticide and water
506 policy since 1998 have collectively increased the stringency of
507 legislation influencing which active substances can be used and,
508 in principle, how these chemicals are managed. In addition,
509 developments in toxicological understanding have reduced
510 some of the scientific uncertainties that existed previously with
511 respect to the risks associated with exposure to pesticides in
512 drinking water. It follows that the assumption of total epistemic
513 uncertainty regarding the human health impacts arising from
514 any level of exposure to any pesticide active substance can no
515
longer be justified. If this is the case, then it could be argued
516
that a surrogate zero for pesticide active substance concen-
517
trations in drinking water is no longer consistent with the
518
precautionary principle and EU Treaty Article 174. This is not
519
in itself necessarily a justification for the revision of the DWD
520
MAC for pesticides; it simply challenges the claim that this
521value is still based upon the precautionary principle.
522
The development by the WHO of a method for calculating
523
guideline values to regulate exposure to threshold chemicals via
524
drinking water and lessons from international experience of
525
regulating pesticides in drinking water could form the basis of a
526
DWD review. However, any such review should be careful to
527
establish whether alternative regulatory approaches offer a
528
sufficiently high level of protection to human health and are
529
themselves consistent with the precautionary principle and EU
530
Treaty Article 174. The compatibility of the WHO GV and
531
EPA MCL approaches with the precautionary principle and EU
532Treaty Article 174 are analyzed in the next section of this paper.
533
2.4. Regulatory Alternatives to Surrogate Zero. Are
534
alternative regulatory options for protecting drinking water
535
quality available to the EU and are these compatible with
536Article 174?
537
The regulatory approach taken by the WHO and EPA does
538
not tolerate any level of exposure to non-threshold chemicals.
539
For threshold chemicals, these approaches assume the existence
540
of two levels of exposure: one level at which risk can be
541
managed and one at which epistemic uncertainty regarding
542
human health outcomes must be acknowledged and regulated.
543
In this respect, both approaches are in keeping with the
544precautionary principle.
545
The DWD, on the other hand, assumes epistemic uncertainty
546
for all pesticide active substances at any level of exposure
547
greater than zero. It then applies the precautionary principle to
548
invoke a surrogate zero (0.1 μg/L) to prevent significant
549
exposure via drinking water. If the assumption of total
550
epistemic uncertainty is not valid, then the standard is no
551
longer justified by the precautionary principle, although it may
552still be justified on other grounds.
553
Thus, although both the WHO and DWD approaches
554
recognize the existence of epistemic uncertainty, the WHO
555
approach is more in keeping with EU Treaty Article 174 and
556
the precautionary principle. This is because it acknowledges the
557
existence of available scientific and technical data, recognizes
558
the heterogeneity of active substances, sets nonarbitrary
559
standards in a transparent way, and is consistent with the five
560
principles of risk management in the EC communication on the
561precautionary principle
14
(as presented in Table 1).
562
In the United States, the approach taken by the EPA to
563
calculate a level of exposure below which no adverse effects will
564occur (the MCLG) uses available scientific and technical data
Table 4. Comparison of Drinking Water Values for Some Common Pesticide Active Substances
Active Substance DWD value (μg/L) WHO GV
a
(μg/L) USA MCL
a
(μg/L) Australian value
a
(μg/L)
Metaldehyde 0.1 −− 20
Propyzamide 0.1 −− 70
Clopyralid 0.1 −− 2000
Pendimethalin 0.1 20 −400
Chlortoluron 0.1 30 −−
Glyphosate 0.1 −700 1000
Atrazine 0.1 2 3 20
Simazine 0.1 2 4 20
a
Symbol “−”means active substance not included in standards.
Environmental Science & Technology Policy Analysis
dx.doi.org/10.1021/es304955g |Environ. Sci. Technol. XXXX, XXX, XXX−XXXF
565 in the same way as the WHO GV calculation. However, the
566 next step, converting the MCLG to a legally enforceable MCL,
567 has the potential to set an MCL above the MCLG, thereby
568 allowing exposure at a level where epistemic uncertainty is
569 known to exist. It follows that the EPA approach is not fully
570 compatible with the precautionary principle.
3. DISCUSSION
571 In the absence of available scientific understanding and
572 technical data, a surrogate zero for pesticides in drinking
573 water can be justified under the precautionary principle and is
574 in keeping with both EU Treaty Article 174 and DWD Article
575 1.2. The decision to use a surrogate zero standard in the 1998
576 DWD can be justified based upon the state of scientific and
577 technical knowledge at the time this Directive was agreed.
578 However, since 1998, there have been advances in available
579 scientific understanding and technical knowledge (Section 2.3)
580 as well as significant developments in EU pesticide and water
581 policy, which have essentially strengthened protective barriers
582 and reduced overall risks to human health arising from
583 exposure to pesticides. Therefore, whether the change in
584 available scientific understanding and technical data since 1998
585 is sufficient to prompt a review of the DWD standard in terms
586 of its consistency with the principles of European environ-
587 mental policy (Article 174 and the precautionary principle)
588 becomes a key question. It can be argued that the WHO GV
589 method for pesticide regulation, like the DWD standard, also
590 offers a high level of protection to prevent adverse effects of
591 pesticides on human health but, unlike the DWD, makes
592 explicit use of available scientific and technical data to set
593 regulatory standards in a way that is consistent with the
594 principles of EU Environment policy as defined in Article 174.
595 Given continuing scientific and legislative developments, it
596 can be argued that a review of drinking water quality standards
597 in the EU is required. A future regulatory approach could better
598 utilize scientifically robust toxicological understanding, where
599 this exists, and still be consistent with the precautionary
600 principle. Where relevant (i.e., where a NOAEL or LOAEL can
601 safely be established through toxicological studies), revised
602 regulation could be based on the WHO GV method. In some
603 cases (for highly toxic compounds), this may, in principle,
604 actually reduce the MAC. The 0.1 μg/L MAC would be
605 retained for those active substances for which reliable NOAEL
606 or LOAEL values were not currently available. In addition, all
607 non-threshold active substances would be banned under the
608 independent action of EU pesticide approval Regulation 1107/
609 2009. We suggest that this would be a more accurate reflection
610 of Article 174 principles and the precautionary principle than
611 using the same surrogate zero for all active substances.
612 The primary purpose of the observations made in this paper
613 is to open an objective debate about whether the current
614 standard for pesticides in drinking water remains consistent
615 with the principles of European Environmental Policy (Treaty
616 Article 174 and the precautionary principle) that originally gave
617 rise to the standard. To be objective, this debate should focus
618 solely upon how available scientific understanding and technical
619 data can be most effectively used to develop environmental
620 policy that is consistent with both Article 174 and the
621 precautionary principle. Where current policy is inconsistent
622 with these principles and where robust scientific evidence exists
623 to support regulatory change, alternatives may need to be
624 formulated. Of course, a key factor in reopening the debate will
625 be whether or not any apparent relaxation of environmental
626and public health protection (real or not) is acceptable to the
627public and to interest groups. This is beyond the scope of this
628paper but could be critical for any ultimate changes in the
629regulation regardless of the scientific arguments.
630
■AUTHOR INFORMATION
631Corresponding Author
632*Phone: 01234 750111; fax: +44(0)1234 752971; e-mail d.
633parsons@cranfield.ac.uk.
634Author Contributions
635The manuscript was written through contributions of all
636authors. All authors have given approval to the final version of
637the manuscript.
638Notes
639The authors declare no competing financial interest.
640
■ACKNOWLEDGMENTS
641This work was conducted as part of a UK EPSRC EngD
642studentship, which was supported by Anglian Water Services.
643The views expressed herein solely reflect those of the authors
644and not those of the sponsoring organizations.
645
■ABBREVIATIONS
646DWD Drinking Water Directive
647EC European Commission
648EPA U.S. Environmental Protection Agency
649EQS Environmental Quality Standards
650ESTO European Science and Technology Observatory
651EU European Union
652GV guideline value
653LOAEL lowest observed adverse effect level
654MAC maximum allowable concentration
655MCL maximum contaminant level
656MCLG maximum contaminant level goal
657NOAEL no observed adverse effect level
658RfD reference dose
659TDI tolerable daily intake
660UF uncertainty factor
661UN United Nations
662WFD Water Framework Directive
663WHO World Health Organization 664
665
■REFERENCES
(1) 666Consolidated versions of the treaty on European Union and of
667the treaty establishing the European Community. Off.J.Eur.
668Communities 2002,C 325,1−184
(2) 669Council Directive 98/83/EC of 3 November 1998 on the quality
670of water intended for human consumption. Off. J. Eur. Communities
6711998,L 330,32−54.
(3) 672Keeping Raw Drinking Water Resources Safe from Pesticides;
673EU1-01-56; Eureau: Brussels, 2001.
(4) 674Jordan, A. European community water policy standards: Locked
675in or watered down? J. Common Mark. Stud. 1999,37,13−37,
676DOI: 10.1111/1468-5965.00148.
(5) 677Tye, R. Are water consumers bearing an unfair burden? Pesticides
678News 1997,35, 18 http://www.pan-uk.org/pestnews/Issue/Pn35/
679pn35p18.htm.
(6) 680Council directive 2000/60/EC of 23 October 2000 establishing a
681framework for community action in the field of water policy. Off. J. Eur.
682Communities 2000,L 327,1−72.
(7) 683Aven, T. On different types of uncertainties in the context of the
684precautionary principle. Risk Anal. 2011,31, 1515−1525,
685DOI: 10.1111/j.1539-6924.2011.01612.x.
Environmental Science & Technology Policy Analysis
dx.doi.org/10.1021/es304955g |Environ. Sci. Technol. XXXX, XXX, XXX−XXXG
(8)
686 Sandin, P.; Peterson, M.; Ove Hansson, S.; Rude
n, C.; Juthe, A.
687 Five charges against the precautionary principle. J. Risk Res. 2002,5,
688 287−299, DOI: 10.1080/13669870110073729.
(9)
689 Vlek, C. Judicious management of uncertain risks: I. Develop-
690 ments and criticisms of risk analysis and precautionary reasoning. J.
691 Risk Res 2010,13, 517−543, DOI: 10.1080/13669871003629887.
(10)
692 Vlek, C. Judicious management of uncertain risks: II. Simple
693 rules and more intricate models for precautionary decision-making. J.
694 Risk Res. 2010,13, 545−569, DOI: 10.1080/13669871003629903.
(11)
695 The Rio Declaration on Environment and Development, 1992.
696 United Nations Environment Programme. www.unep.org/Documents.
697 Multilingual/Default.asp?documentid=78&articleid=1163 (accessed
698 April 25, 2013).
(12)
699 Sand, P. H. The precautionary principle: A European
700 perspective. Hum. Ecol. Risk. Assess. 2000,6, 445−458,
701 DOI: 10.1080/10807030091124563.
(13)
702 De Sadeleer, N. The precautionary principle as a device for
703 greater environmental protection: Lessons from EC courts. Rev. Eur.
704 Commun. Int. Env. Law 2009,18,3−10.
(14)
705 Communication from the Commission on the Precautionary
706 Principle; COM/2000/1; European Commission: Brussels, 2000.
(15)
707 Klinke, A.; Renn, O. Precautionary principle and discursive
708 strategies: Classifying and managing risks. J. Risk Res. 2001,4, 159−
709 173, DOI: 10.1080/136698701750128105.
(16)
710 Stirling, A.; Gee, D. Science, precaution, and practice. Public
711 Health Rep. 2002,117, 521−533.
(17)
712 Council Directive 80/778/EEC of 15 July 1980 relating to the
713 quality of water intended for human consumption. Off. J. Eur.
714 Communities 1980,L 229,11−29.
(18)
715 Council Directive 75/440/EEC of 16 June 1975 concerning the
716 quality required of surface water intended for the abstraction of
717 drinking water. Off. J. Eur. Communities 1975,L 194,26−31.
(19)
718 Hey, C. EU Environmental Policies: A short history of the
719 policy strategies. In European Union Environmental Policy Handbook: A
720 Critical Analysis of EU Environmental Legislation; Scheuer, S., Ed.;
721 European Environmental Bureau: Utrecht, 2006; p 18.
(20)
722 Council Directive 91/414/EEC of 15 July 1991 concerning the
723 placing of plant protection products on the market. Off. J. Eur.
724 Communities 1991,L 230, 1-32.
(21)
725 Evaluation of the Active Substances of Plant Protection
726 Products (Submitted in Accordance with Article 8(2) of Council
727 Directive 91/414/Eec on the Placing of Plant Protection Products on
728 the Market); Report from the Commission to the European
729 Parliament and the Council, COM/2001/444; European Commission:
730 Brussels, 2001.
(22)
731 White Paper: Strategy for a Future Chemicals Policy; COM/
732 2001/88; European Commission: Brussels, 2001.
(23)
733 Munro, I. C.; Carlo, G. L.; Orr, J. C.; Sund, K. G.; Wilson, R.
734 M.; Kennepohl, E.; Lynch, B. S.; Jablinske, M.; Lee, N. L. A
735 comprehensive, integrated review and evaluation of the scientific
736 evidence relating to the safety of the herbicide 2,4-D. J. Am. Coll.
737 Toxicol. 1992,11 (5), 559−664.
(24)
738 Garcia, S. J.; Abu-Qare, A. W.; Meeker-O’Connell, W. A.;
739 Borton, A. J.; Abou-Donia, M. B. Methyl parathion: A review of health
740 effects. J. Toxicol. Environ. Health, Part B 2003,6(2), 185−210.
(25)
741 Silva, M. H.; Carr, W. C. Human health risk assessment of
742 endosulfan: II. Dietary exposure assessment. Regul. Toxicol. Pharmacol.
743 2010,56 (1), 18−27.
(26)
744 Carpy, S. A.; Kobel, W.; Doe, J. Health risk of low-dose
745 pesticides mixtures: A review of the 1985−1998 literature on
746 combination toxicology and health risk assessment. J. Toxicol. Environ.
747 Health, Part B 2000,3(1), 1−25.
(27)
748 Lydy, M.; Belden, J.; Wheelock, C.; Hammock, B.; Denton, D.
749 Challenges in regulating pesticide mixtures. Ecol. Soc. 2004,9(6), 1.
(28)
750 Dieter, H. H. The relevance of “non-relevant metabolites”from
751 plant protection products (PPPs) for drinking water: The German
752 view. Regul. Toxicol. Pharmacol. 2010,56 (2), 121−125.
(29)
753 Council Directive 79/117/EEC of 21 December 1978
754 prohibiting the placing on the market and use of plant protection
755
products containing certain active substances. Off. J. Eur. Communities
7561979,L33,36−40.
(30) 757
EU Action on Pesticides “Our Food Has Become Greener”,
758
2009. Directorate-General for Health and Consumers, European
759
Commission. http://ec.europa.eu/dgs/health_consumer/
760
information_sources/docs/plant/factsheet_pesticides_en.pdf (ac-
761cessed April 25, 2013).
(31) 762
Regulation 1907/2006 of the European Parliament and of the
763
Council of 18 December 2006 concerning the Registration, Evaluation,
764
Authorisation and Restriction of Chemicals (REACH), establishing a
765
European Chemicals Agency, amending Directive 1999/45/EC and
766
repealing Council Regulation (EEC) No 793/93 and Commission
767
Regulation (EC) No 1488/94 as well as Council Directive 76/769/
768
EEC and Commission Directives 91/155/EEC, 93/67/EEC, 93/105/
769EC and 2000/21/EC. Off. J. Eur. Union 2006,L 396,1−849.
(32) 770
Regulation 1107/2009 of the European Parliament and of the
771
Council of 21 October 2009 concerning the placing of plant protection
772
products on the market and repealing Council Directives 79/117/EEC
773and 91/414/EEC. Off. J. Eur. Union 2009,L 309,1−50.
(33) 774
European Commission Guidance Document on the Assessment
775
of the Relevance of Metabolites in Groundwater of Substances
776
Regulated under Council Directive 91/414/Eec; Sanco/221/2000-
777
rev.10-final; Health & Consumer Protection Directorate-General:
778Brussels, 2003.
(34) 779
Council Directive 2008/105/EC of 16 December 2008 on
780
environmental quality standards in the field of water policy, amending
781
and subsequently repealing Council Directives 82/176/EEC, 83/513/
782
EEC, 84/156/EEC, 84/491/EEC, 86/280/EEC and amending
783
Directive 2000/60/EC of the European Parliament and of the
784Council. Off. J. Eur. Union 2008,L 348,84−97.
(35) 785
Dolan, T.; Howsam, P.; Parsons, D. J. Diffuse pesticide
786
pollution of drinking water sources: Impact of legislation and UK
787
responses. Water Policy 2012,14, 680−693, DOI: 10.2166/
788wp.2012.147.
(36) 789
Directive 2009/128/EC of the European Parliament and of the
790
Council of 21 October 2009 establishing a framework for Community
791
action to achieve the sustainable use of pesticides. Off. J. Eur. Union
7922009,L 309,71−86.
(37) 793
Guidelines for Drinking-Water Quality, 2nd ed.; World Health
794
Organization, Geneva, 1993. http://www.who.int/water_sanitation_
795health/dwq/gdwq2v1/en/ (accessed April 25, 2013).
(38) 796
Guidelines for Drinking-Water Quality: Incorporating the First
797
and Second Addenda, 3rd ed.; World Health Organization, Geneva,
798
2008. http://www.who.int/water_sanitation_health/dwq/gdwq3rev/
799en/ (accessed April 25, 2013).
(39) 800
Guidelines for Drinking-Water Quality, 4th ed.; World Health
801
Organization, Geneva, 2011. http://www.who.int/water_sanitation_
802
health/publications/2011/dwq_guidelines/en/ (accessed April 25,
8032013).
(40) 804
Ritter, L.; Totman, C.; Krishnan, K.; Carrier, R.; Ve
zina, A.;
805
Morisset, V. Deriving uncertainty factors for threshold chemical
806
contaminants in drinking water. J. Toxicol. Environ. Health, Part B
8072007,10, 527−557, DOI: 10.1080/15287390600975178.
(41) 808
National Water Quality Management Strategy; Australian
809
Drinking Water Guidelines Paper 6; National Health and Medical
810
Research Council, National Resource Management Ministerial
811Council: Commonwealth of Australia, Canberra, 2011.
(42) 812
Basic Information on CCL and Regulatory Determinations,
813
2012. U.S. Environmental Protection Agency. http://water.epa.gov/
814
scitech/drinkingwater/dws/ccl/basicinformation.cfm (accessed April
81525, 2013).
(43) 816
Regulating Public Water Systems and Contaminants Under the
817
Safe Drinking Water Act, 2012. U.S. Environmental Protection
818
Agency. http://water.epa.gov/lawsregs/rulesregs/
819
regulatingcontaminants/basicinformation.cfm accessed April 25,
8202013).
(44) 821
Proposal for a Regulation of the European Parliament and of the
822
Council Concerning the Placing of Plant Protection Products on the
823Market: Summary Impact Assessment; Pesticide Safety Directorate:
Environmental Science & Technology Policy Analysis
dx.doi.org/10.1021/es304955g |Environ. Sci. Technol. XXXX, XXX, XXX−XXXH
824 York, 2009. http://www.pesticides.gov.uk/Resources/CRD/Migrated-
825 Resources/Documents/O/Outcomes_paper_-_summary_impact_
826 assessment_(Jan_09).pdf (accessed April 25, 2013).
Environmental Science & Technology Policy Analysis
dx.doi.org/10.1021/es304955g |Environ. Sci. Technol. XXXX, XXX, XXX−XXXI
