OPINION OF ADVOCATE GENERAL
KOKOTT
delivered on 28 February 2019 (1)
Case C‑723/17
Lies Craeynest and Others
v
Brussels Hoofdstedelijk Gewest and Others
(Request for a preliminary ruling
from the Nederlandstalige rechtbank van eerste aanleg Brussel (Dutch-speaking Court of First Instance of Brussels, Belgium))
(Request for a preliminary ruling — Directive 2008/50/EC — Ambient air quality — Limit values — Locations of sampling points — Discretion — Judicial review — Criteria for establishing an exceedance of limit values)
I. Introduction
1. The city of Brussels, together with the city of Paris, recently took a stand for ambient air quality by obtaining a judgment from the General Court which declared the emission limits set by the Commission in respect of oxides of nitrogen for the new real driving emission tests for light passenger and commercial vehicles to be invalid. (2) Nevertheless, in the present case, a number of residents and an environmental organisation are taking action against the Brussels‑Capital Region with regard to the assessment of ambient air quality.
2. The dispute relates to the measurements by which compliance with or breach of the ambitious limit values for ambient air quality under Directive 2008/50 (3) are to be determined. It must be clarified, first, to what extent the siting of sampling points is subject to the control of national courts and, second, whether the results from different sampling points may be averaged in order to assess compliance with the limit values. The first question in particular is of fundamental legal importance because it seeks clarification of the intensity of judicial review which national courts must ensure under EU law.
II. Legal framework
3. Article 1(1) of Directive 2008/50 sets out its essential objective:
‘This Directive lays down measures aimed at the following:
1. defining and establishing objectives for ambient air quality designed to avoid, prevent or reduce harmful effects on human health and the environment as a whole;
2. …’
4. Article 2(25) and (26) of Directive 2008/50 defines certain measurement methods:
‘25. “Fixed measurements” shall mean measurements taken at fixed sites, either continuously or by random sampling, to determine the levels in accordance with the relevant data quality objectives;
26. “indicative measurements” shall mean measurements which meet data quality objectives that are less strict than those required for fixed measurements’.
5. Article 6 of Directive 2008/50 contains assessment criteria for ambient air quality:
‘1. Member States shall assess ambient air quality with respect to the pollutants referred to in Article 5 in all their zones and agglomerations, in accordance with the criteria laid down in paragraphs 2, 3 and 4 of this Article and in accordance with the criteria laid down in Annex III.
2. In all zones and agglomerations where the level of pollutants referred to in paragraph 1 exceeds the upper assessment threshold established for those pollutants, fixed measurements shall be used to assess the ambient air quality. Those fixed measurements may be supplemented by modelling techniques and/or indicative measurements to provide adequate information on the spatial distribution of the ambient air quality.
3. In all zones and agglomerations where the level of pollutants referred to in paragraph 1 is below the upper assessment threshold established for those pollutants, a combination of fixed measurements and modelling techniques and/or indicative measurements may be used to assess the ambient air quality.
4. In all zones and agglomerations where the level of pollutants referred to in paragraph 1 is below the lower assessment threshold established for those pollutants, modelling techniques or objective-estimation techniques or both shall be sufficient for the assessment of the ambient air quality.
5. …’
6. Under Article 7(1) of Directive 2008/50, the location of sampling points for the measurement of sulphur dioxide, nitrogen dioxide and oxides of nitrogen, particulate matter (PM10, PM2.5), lead, benzene and carbon monoxide in ambient air are to be determined using the criteria listed in Annex III.
7. The number of sampling points is to be determined, in accordance with Article 7(2) and Annex V to Directive 2008/50, by reference to the population of the agglomeration or zone in question.
8. Article 13(1) of Directive 2008/50 lays down an obligation to comply with various limit values:
‘Member States shall ensure that, throughout their zones and agglomerations, levels of sulphur dioxide, PM10, lead, and carbon monoxide in ambient air do not exceed the limit values laid down in Annex XI.
In respect of nitrogen dioxide and benzene, the limit values specified in Annex XI may not be exceeded from the dates specified therein.
Compliance with these requirements shall be assessed in accordance with Annex III.
…’
9. Article 23(1) of Directive 2008/50 provides that where limit values are exceeded in given zones or agglomerations, air quality plans must be established in order to achieve those values.
10. Point 1 of Section B of Annex III to Directive 2008/50 concerns the siting of measurements directed at the protection of human health:
‘(a) Sampling points directed at the protection of human health shall be sited in such a way as to provide data on the following:
– the areas within zones and agglomerations where the highest concentrations occur to which the population is likely to be directly or indirectly exposed for a period which is significant in relation to the averaging period of the limit value(s);
– levels in other areas within the zones and agglomerations which are representative of the exposure of the general population;
(b) Sampling points shall in general be sited in such a way as to avoid measuring very small micro-environments in their immediate vicinity, which means that a sampling point must be sited in such a way that the air sampled is representative of air quality for a street segment no less than 100 m length at traffic-orientated sites and at least 250 m × 250 m at industrial sites, where feasible;
…
(f) Sampling points shall, where possible, also be representative of similar locations not in their immediate vicinity;
…’
11. It is clear from the case file that the Brussels-Capital Region has correctly transposed the relevant provisions of Directive 2008/50.
III. Facts and request for a preliminary ruling
12. Ms Craeynest, Ms Lopez Devaux, Mr Mertens, Ms Goeyens and Ms De Schepper are or were all resident in the Brussels-Capital Region. Ms Goeyens is now deceased, however, and her case has been taken up by Mr Vandermeulen. ClientEarth is a non-profit association established under English law, with the centre of its activities in Belgium. It seeks, inter alia, to advance environmental protection by means of raising awareness and instituting legal proceedings.
13. They are in dispute with the Brussels-Capital Region and the Brussels Institute for Management of the Environment before the Dutch-speaking Court of First Instance of Brussels as to whether an adequate air quality plan has been established for the Brussels zone. In those proceedings that court has addressed the following questions to the Court of Justice:
‘(1) Should Article 4(3) and the second subparagraph of Article 19(1) of the Treaty on European Union, read in conjunction with the third paragraph of Article 288 of the Treaty on the Functioning of the European Union and Articles 6 and 7 of Directive 2008/50 be interpreted as meaning that, when it is alleged that a Member State has not sited the sampling points in a zone in accordance with the criteria set out in point B.1.(a) of Annex III to Directive 2008/50, it is for the national courts, on application by individuals who are directly affected by the exceedance of the limit values referred to in Article 13(1) of that directive, to examine whether the sampling points were sited in accordance with those criteria and, if they were not, to take all necessary measures against the national authority, such as an order, with a view to ensuring that the sampling points are sited in accordance with those criteria?
(2) Is a limit value within the meaning of Article 13(1) and Article 23(1) of Directive 2008/50 exceeded in the case where an exceedance of a limit value with an averaging period of one calendar year, as laid down in Annex XI to that directive, has been established on the basis of the measurement results from one single sampling point within the meaning of Article 7 of that directive, or does such an exceedance occur only when this becomes apparent from the average of the measurement results from all sampling points in a particular zone within the meaning of Directive 2008/50?’
14. Written observations were submitted by Ms Craeynest and Others, the Brussels Region, the Kingdom of the Netherlands, the Czech Republic and the European Commission. With the exception of the Netherlands, these parties also took part in the hearing on 10 January 2019.
IV. Legal assessment
15. The request for a preliminary ruling is intended to clarify, first, the extent to which national courts may review the siting of sampling points and, second, whether the results from different sampling points may be averaged in order to assess compliance with the limit values.
A. Location of sampling points
16. The first question seeks to clarify whether national courts may review the location of sampling points in order to check compliance with the limit values laid down in Directive 2008/50 and which measures they may or must take if the criteria for determining the location laid down in the directive have been infringed.
17. This question can be construed as asking whether the national courts must have certain powers in the enforcement of Union law, in particular the power to impose orders on the authorities. The question also arises, more acutely, in another current request for a preliminary ruling from Germany in which the Court is asked whether national courts may be obliged to impose detention on public officials in order thereby to enforce the obligation to update an air quality plan within the meaning of Article 23 of Directive 2008/50. (4)
18. The answer would have to be that in principle EU law is not intended to create new remedies in the national courts to ensure the observance of EU law other than those already laid down by national law. (5) It would be otherwise only if it were apparent from the overall scheme of the national legal system in question that no legal remedy existed which made it possible to ensure, even indirectly, respect for rights under EU law. (6)
19. In the present case, however, there is no need for a more detailed examination of this aspect of the question because it is undisputed that the national court has the competence to impose orders. Rather, it must be ascertained what standard of review must be applied by the national court in respect of the siting of sampling points.
20. As is indicated in the request for a preliminary ruling and will be explained in greater detail below, the application of the rules on the siting of sampling points requires discretion to be exercised in connection with a complex assessment of scientific matters and a weighing up exercise.
21. There could also be reason to doubt in this regard whether these rules are sufficiently precise to be applied directly. (7) However, even where direct application of provisions of a directive is ruled out, there is an entitlement to a judicial review of whether the national legislation and its application have remained within the limits of the discretion set by that directive. (8)
22. As the Commission rightly explains, it is ultimately not a matter of the direct application of EU law in this case because the provisions of Directive 2008/50 were transposed into domestic law. It is nevertheless also important for the application of national law in the main proceedings to determine the limits of the discretion of the competent authorities. To that end, the minimum level of judicial review required by EU law with regard to the application of the relevant rules must be clarified.
23. I will therefore begin by examining below the rules on the identification of locations and then consider the intensity of judicial review required by EU law.
24. First of all, however, it should be made clear that the provisions of Directive 2008/50 on the Commission’s scrutiny of the application of the directive cannot diminish the responsibility of the national courts. These rules merely specify the general task conferred on the Commission in the second and third sentences of Article 17(1) TEU of overseeing the application of Union law in general.
1. The rules on the identification of locations
25. The national court questions in particular whether the rules on the location of sampling points imply any unconditional obligations, compliance with which can easily be monitored by a court at the behest of individuals. According to the national court, there do not appear to be more detailed rules regarding the way in which the ‘areas … where the highest concentrations occur’ should be identified or defined.
26. It thus refers to Article 7(1) and the first indent of point 1(a) of Section B of Annex III to Directive 2008/50. Under those provisions, sampling points are to be sited in such a way as to provide data on the areas within zones and agglomerations where the highest concentrations occur to which the population is likely to be directly or indirectly exposed for a period which is significant in relation to the averaging period of the limit value(s).
27. Furthermore, under Article 7(1) and point 1 of Section B of Annex III(b) to Directive 2008/50, sampling points are to be sited in general in such a way as to avoid measuring very small micro-environments in their immediate vicinity. It is further specified that the air sampled is to be representative of air quality for a street segment no less than 100 m length at traffic-orientated sites and at least 250 m × 250 m at industrial sites, where feasible.
28. The dimensions of possible areas where the highest concentrations occur are therefore defined. They cannot be identified solely by having recourse to jurisprudence, but only by applying the relevant scientific methods, although this does not preclude a judicial review.
29. Directive 2008/50 does not expressly specify the scientific methods for identifying the areas where the highest concentrations occur but, on the basis of the legislative context of the directive, the competent authorities must have recourse in that regard to measurements, models and other information.
30. This is demonstrated in particular by the rules on the basis of which the methods for monitoring compliance with the limit values for ambient air quality are defined. Under Article 5 and Section B of Annex II to Directive 2008/50, pollution of agglomerations and zones is to be determined either using measured values over five years or at least on the basis of a combination of information from emission inventories and modelling and measurement campaigns of short duration during the period of the year and at locations likely to be typical of the highest pollution levels.
31. Using the values thus obtained, it is possible to assess whether a lower or an upper assessment threshold is exceeded. Below the lower threshold, that is, where exceedance of the limit values is very unlikely, it is sufficient under Article 6(4) of Directive 2008/50 to monitor compliance with the limit values by modelling techniques or objective-estimation techniques or both. Between the two thresholds, where the limit values can be exceeded more readily, under Article 6(3) a combination of fixed measurements (Article 2(25)) and modelling techniques and/or indicative measurements (Article 2(26)) can be used. In the case of exceedance of the upper threshold, where exceedance of the limit values is most likely, fixed measurements are to be used to assess the ambient air quality under Article 6(2). Those measurements may be supplemented by modelling techniques and/or indicative measurements to provide adequate information on the spatial distribution of the ambient air quality.
32. These possible methods for determining ambient air quality, in particular measurements and modelling techniques, must be used for the siting of fixed sampling points.
33. If, as in the main proceedings, there is dispute over the locations of sampling points, the competent authorities must therefore explain the adequate information on the spatial distribution of the ambient air quality on the basis of which they determined the locations in question and how they obtained that information.
2. Intensity of judicial review
34. There has been no mention thus far, however, of how the national courts review compliance with the rules on the location of sampling points.
35. The competent authorities clearly infringe those rules if, against their better knowledge, they do not site sampling points where the highest concentrations occur or if there is no scientific basis for the locations determined. National courts must be able to establish such infringements.
36. Furthermore, it follows from the above considerations that the areas where the highest concentrations occur must be identified at least as a rule by a combination of measurements, modelling techniques and other information. However, in this context there is considerable scope for differences of opinion, for example regarding the siting, timing and frequency of measurements and above all regarding the modelling techniques used.
37. Consideration therefore needs to be given to the intensity of judicial review required by EU law, that is to say, the margin of discretion enjoyed by the competent authorities in applying the criteria for the siting decision. The national court would like to know to what extent Directive 2008/50 permits the (national) principle of the division of powers to restrict its power to review administrative action in connection with the siting of sampling points.
(a) The procedural autonomy of the Member States
38. This question must be answered against the background of the procedural autonomy of the Member States. In the absence of detailed European Union rules governing judicial review of administrative decisions implementing Directive 2008/50, it is for the domestic legal system of each Member State to determine those requirements in accordance with the principle of procedural autonomy provided, however, that those requirements are not less favourable than those governing similar domestic situations (principle of equivalence) and that they do not render impossible in practice or excessively difficult the exercise of rights conferred by the EU legal order (principle of effectiveness). (9)
39. It is conceivable that, in exercising procedural autonomy, Member States will grant their courts very extensive powers of review which might even permit them to substitute for the administrative authorities and to modify or replace their decisions. There can be no objection if such judicial decisions satisfy the requirements laid down by EU law for the administrative decision and, in particular, have an adequate scientific basis, but also comply with the procedural rules. (10)
40. The request for a preliminary ruling is not, however, based on such a legal situation, but is seeking to identify the necessary minimum level of powers of judicial review. As the request does not make any suggestion that the principle of equivalence might be affected, regard must be had solely to the principle of effectiveness, that is to say, what standard of judicial review is necessary so that reliance on the applicable provisions of EU law is not rendered excessively difficult.
(b) The review of scientifically complex assessments
41. As far as the effective review of the siting of a sampling point is concerned, it must be stated that the rules outlined require a complex scientific assessment. It must first be decided by what methods adequate information is to be obtained as the basis for selecting the location and that information must then be evaluated in order to determine a location.
42. The minimum standards which the review of such a decision by the national courts must satisfy can be inferred from the review criteria applied by the EU Courts in assessing similar measures of the institutions. EU law does not require the Member States to establish a procedure for judicial review of national decisions applying rules of EU law which involve a more extensive review than that carried out by the Court in similar cases. (11)
43. A characteristic of the resultant criteria is that in complex scientific or technical assessments and weighing up there is, as a rule, broad discretion which can be reviewed only to some degree. That discretion is nevertheless limited in certain cases and must therefore be reviewed more intensively, in particular where they are particularly serious interferences with fundamental rights.
44. The broad discretion enjoyed by the Union institutions where they are called upon to make complex scientific and technical assessments relates in particular to the assessment of the facts in order to determine the nature and scope of the measures but also, to a certain extent, to the finding of facts underlying their action. (12)
45. Substantive review by the European Union judicature is limited to verifying whether the authorities have manifestly exceeded the limits of their discretion. In such a context, the European Union judicature cannot substitute its assessment of scientific and technical facts for that of the institutions on which alone the legislature has placed that task. (13)
46. However, in such circumstances, it is the duty of the competent institution to examine carefully and impartially all the relevant aspects of the individual case, (14) to comply with the procedural rules (15) and, in particular, to give adequate reasoning for its decision in order to enable the European Union judicature to verify whether the factual and legal elements upon which the exercise of the power of assessment depends were present. (16)
47. Furthermore, the Court has ruled that, where interferences with fundamental rights are at issue, the extent of the EU legislature’s discretion may prove to be limited, in the light of the principle of proportionality, depending on a number of factors, including, in particular, the area concerned, the nature of the right at issue guaranteed by the Charter, the nature and seriousness of the interference and the object pursued by the interference. (17) Thus, in view of the important role played by the protection of personal data in the light of the fundamental right to respect for private life and the extent and seriousness of the interference with that right caused by an unjustified comprehensive retention of call data, Directive 2006/24 (18) was subject to strict review (19) and the Court ultimately declared it invalid.
48. In broader terms, the rights established by EU law may not be deprived of all useful effect and, in particular, they may not be undermined, that is, deprived of their very substance. (20)
49. These criteria constitute the required minimum standard for judicial review of compliance with EU law in the Member States. The Court has nevertheless also held that any national procedure for judicial review of decisions of national authorities must enable the court or tribunal seised effectively to apply the relevant principles and rules of EU law when reviewing its legality. (21)
50. I understand this latter condition as being a reminder that the national courts must carefully discern whether a certain matter falls within the scope of the broad discretion and requires only limited judicial review or whether other matters requiring stricter judicial review are at issue, in particular the limits of discretion or procedural complaints.
(c) Application to the point at issue
51. The basis for the application of these criteria to the decision on the siting of sampling points for assessing ambient air quality is the complex assessment which must be made by the competent authorities in order to decide on the siting of sampling points. They must select scientific methods to obtain the necessary information and, in doing so, weigh up the investigation effort required and evaluate the results.
52. In that assessment, EU law would, in principle, allow them broad discretion and require only limited judicial review.
53. However, note should be taken of the considerable importance of the rules on ambient air quality highlighted by the Commission. Directive 2008/50 is based on the assumption that exceedance of the limit values leads to a large number of premature deaths. (22) The rules on ambient air quality therefore put in concrete terms the Union’s obligations to provide protection following from the fundamental right to life under Article 2(1) of the Charter and the high level of environmental protection required under Article 3(3) TEU, Article 37 of the Charter and Article 191(2) TFEU. Measures which may impair the effective application of Directive 2008/50 are thus comparable, in their significance, with the serious interference with fundamental rights on the basis of which the Court made the rules on the retention of call data subject to strict review.
54. If the sampling points were not sited in the areas where the highest concentrations actually occur, the effectiveness of Directive 2008/50 could be seriously impaired, as the most ambitious limit values remain ineffective if compliance with them is monitored in the wrong place. In that case it would be possible that an exceedance of the limit values would be undetected and the necessary measures to ensure ambient air quality would not therefore be taken.
55. This risk and the objective, laid down in Article 1 of Directive 2008/50 in particular, of ensuring appropriate ambient air quality with a view to the protection of the life and health of residents limit the discretion available to the competent authorities in the complex assessment underlying the siting of sampling points. (23) In case of doubt, they must choose a strategy which reduces the risk of exceedances of limit values not being detected.
56. Accordingly, the judicial review must also be directed at the exercise of discretion. The courts may confine themselves to establishing manifest errors if this could result in an excessively strict application of Directive 2008/50, but doubts calling into question the achievement of the protective aims of the directive must be reviewed more intensively.
57. What does this mean for the standard of review to be applied?
58. In the light of the precautionary principle, the Court has specified a strict standard of review in connection with the appropriate assessment under Article 6(3) of the Habitats Directive. (24) It must contain complete, precise and definitive findings and conclusions capable of removing all reasonable scientific doubt as to the effects of the proposed works on the protected site concerned. (25) Otherwise it is not possible to authorise the plan or project under Article 6(3) and it is conceivable at most to grant an authorisation for imperative reasons of overriding public interest pursuant to Article 6(4).
59. This standard has the same effect as a presumption that plans or projects adversely affect areas of conservation and should not therefore be implemented in principle. It can be refuted only by removing all reasonable scientific doubt.
60. However, such a standard cannot be applied to the review of decision on siting sampling points because it is highly likely that all existing methods for selecting locations are subject to reasonable scientific doubt. The competent authorities must apply some method, otherwise — contrary to the requirements of Directive 2008/50 — no sampling points at all could be sited. In other words, in choosing the location for sampling points there is no presumption that a certain location would be particularly suitable.
61. Nevertheless, the competent authorities may, in principle, apply the ‘best’ available method. This should be the method which is subject to the least reasonable scientific doubt. Identifying this method is not unproblematic from a scientific point of view, as the different doubts must be weighed in order to ascertain which doubts have less significance.
62. In addition, it must be assumed that the methods for identifying the areas where the highest concentrations occur can be improved through greater examination effort. In particular, it is to be expected that a larger number of measurements, that is, an extension of the measurement period but also a larger number of indicative measurements at different locations, produces more precise results. However, greater effort generally entails higher costs and can cause delays. The decision on the effort undertaken must thus be weighed up, which is also apparent from the fact that under the second sentence of Article 6(2) of Directive 2008/50 adequate information is to be obtained.
63. For both aspects, the weighting of reasonable scientific doubts and the weighing up of what effort to remove them is justified may not restrict the national courts to identifying manifest errors, on account of the importance of the rules on ambient air quality for human life and health.
64. Rather, it is for the competent authorities to convince the courts in particular by presenting substantiated arguments. These must be mainly scientific in nature, but may be extended to economic aspects in the weighing up exercise. The other party is free to counter such claims with its own scientifically substantiated arguments. Of course, it is also conceivable for the court to have recourse to independent experts in order to find support in appraising such a scientific dispute.
65. If the authorities are unable to convince the Court, they must at least undertake additional investigations, for example by taking further measurements or applying other models of the development of ambient air quality.
66. If the national courts have powers to impose orders, they may order such further investigations. If, however, the courts are permitted only to annul administrative decisions, there must be an obligation on the authorities to draw the necessary conclusions from such annulment and the grounds of the decision.
3. The answer to the first question
67. The answer to the first question is therefore that national courts must, on application by affected individuals, examine whether sampling points were sited in accordance with the criteria set out in point B.1.(a) of Annex III to Directive 2008/50 and, if they were not, must take all necessary measures within the scope of their judicial powers against the national authority with a view to ensuring that the sampling points are sited in accordance with those criteria. Such a judicial decision may give rise to the obligation to site sampling points at certain locations if it is clear from the available information that sampling points must be sited there. Otherwise the competent authorities may be obliged to undertake investigations in order to identify the correct locations.
B. Assessment of compliance with the limit values
68. The second question seeks to clarify whether a limit value within the meaning of Article 13(1) and Article 23(1) of Directive 2008/50 is exceeded in the case where an exceedance of a limit value with an averaging period of one calendar year, as laid down in Annex XI to that directive, has been established on the basis of the measurement results from one single sampling point within the meaning of Article 7 of that directive or whether such an exceedance occurs only when this becomes apparent from the average of the measurement results from all sampling points in a particular zone within the meaning of Directive 2008/50.
1. Implementing Decision 2011/850/EU
69. The Commission relies on its Implementing Decision 2011/850/EU (26) for Directive 2008/50, which provides in Article 10 for the Commission to be notified of the measurements from the individual sampling points. If compliance with the limit values were assessed on the basis of an overall analysis, this would not be necessary.
70. However, the Commission cannot lay down in an implementing decision how compliance with the limit values is to be assessed. It follows from the first paragraph of Article 290 TFEU in conjunction with the second paragraph of Article 291 TFEU that in exercising an implementing power, the Commission may neither amend nor supplement the legislative act. (27) In addition, even if, contrary to its designation, the Implementing Decision were to be regarded as a delegated act in accordance with Article 290 TFEU, it would have to come within the regulatory framework as defined by Directive 2008/50. (28)
71. The interpretation of Directive 2008/50 is thus crucial.
2. The wording of Article 13(1) of Directive 2008/50
72. The wording of Article 13(1) of Directive 2008/50 does not give rise to a compelling answer to the question.
73. According to the German version of the first sentence of Article 13(1) of Directive 2008/50, the Member States must ensure that ‘überall in ihren Gebieten und Ballungsräumen’ the levels of sulphur dioxide, PM10, lead, and carbon monoxide in ambient air do not exceed the limit values laid down in Annex XI. This wording could be understood to mean that these limit values must be complied with at each location, that is to say, they may not be exceeded at any location. Accordingly, exceedance at one sampling point would constitute an infringement of that provision. The English version uses the wording ‘throughout their zones and agglomerations’ and thus has similar substance.
74. On the other hand, the French (‘dans l’ensemble de leurs zones et agglomérations’), Dutch (‘in de gehele zones en agglomeraties’) and Spanish (‘en todas sus zonas y aglomeraciones’) versions could be interpreted as referring to the zones and agglomerations collectively. This does not necessarily mean that the limit values have to be complied with at each location, but does not preclude it.
75. Furthermore, the second sentence of Article 13(1) of Directive 2008/50 does not make reference to a particular location. Under that provision, in respect of nitrogen dioxide and benzene, the limit values specified in Annex XI may not be exceeded from the dates specified therein. This provision is thus open to both interpretations in all language versions.
76. Therefore, as the first sentence of Article 13(1) of Directive 2008/50 does not necessarily have the same meaning in the different language versions and the second sentence permits different interpretations, the legislative context and the aim of the provision in particular should be examined more closely.
3. The legislative context of Article 13(1) of Directive 2008/50
77. Annex III is of particular importance to the interpretation of the first and second sentences of Article 13(1) of Directive 2008/50, as under the third sentence compliance with the requirements laid down in the first two sentences is to be assessed in accordance with that annex. The detailed rules on measurements laid down in Annex III militate against an overall analysis of zones and agglomerations, however.
78. In particular, under point 1 of Section A of Annex III to Directive 2008/50 ambient air quality is to be assessed at all locations (29) except for certain locations where it is not to be assessed in accordance with point 2. The excluded locations are small-scale, such as locations where members of the public do not have access and there is no fixed habitation, factory premises or industrial installations to which all relevant provisions concerning health and safety at work apply, and the carriageway of roads and the central reservations of roads except where there is normally pedestrian access to the central reservation. Accordingly, no overall analysis is to be undertaken, but a local assessment.
79. It must be recognised, however, that the language versions are not consistent here either, as in the Dutch version ambient air quality is to be assessed everywhere (‘overal’), which would be slightly more open to an overall analysis than the German wording, for example. However, in that version too, certain locations are excluded from assessment, which would not make sense in the case of an overall analysis.
80. Averaging measurements from different locations must be rejected above all, however, because Article 7(1) and point 1(a) of Section B of Annex III to Directive 2008/50 provide for two kinds of fixed sampling points, in the first indent of point 1(a) of Section B of Annex III sampling points at which data are obtained on the ‘areas where the highest concentrations occur’ and in the second indent sampling points where data are obtained on levels in other areas which are representative of the exposure of the general population.
81. Averaging measurement data from different locations may be appropriate for determining the exposure of the general population. But why should an average be obtained from data on the highest concentrations and data on general pollution? By their nature, average values are intended to present the general situation, whereas this is already done by the latter data.
82. Furthermore, the rule on data quality in the fourth indent of Section B of Annex I to Directive 2008/50 in particular shows that ambient air quality is to be assessed locally and not in an overall analysis. Under that provision, the extent of any area or, if relevant, the length of road within the zone or agglomeration over which concentrations exceed any limit value is to be described. No differences in the language versions can be seen in this regard, moreover.
83. Therefore, the legislative context of Article 13(1) of Directive 2008/50 clearly indicates that compliance with the limit values should be assessed by reference to the measurement results from the fixed sampling points, without obtaining an average from all the sampling points.
4. The aims of Article 13(1) of Directive 2008/50
84. The conclusions reached from the legislative context are confirmed by the purpose of the limit values at issue. According to Article 1(1) and the headings of Article 13 and Annex XI to Directive 2008/50, they are aimed at the protection of human health.
85. There is a risk of adverse effects on health wherever the limit values are exceeded. Appropriate measures must be taken there in order to prevent adverse effects. It is only of limited importance to that risk whether an exceedance applies to the entire zone or agglomeration on average. This is clearly illustrated by the joke about the statistician who drowns in a lake even though it averages only a few centimetres in depth.
5. Answer to the second question
86. A limit value under Annex XI to Directive 2008/50 is thus exceeded within the meaning of Article 13(1) and Article 23(1) of the directive where the measurement result exceeds one single sampling point within the meaning of Article 7 of that directive.
V. Conclusion
87. I therefore propose that the Court rule as follows:
(1) The national courts must, on application by affected individuals, examine whether sampling points were sited in accordance with the criteria set out in point B.1.(a) of Annex III to Directive 2008/50/EC of the European Parliament and of the Council of 21 May 2008 on ambient air quality and cleaner air for Europe and, if they were not, must take all necessary measures within the scope of their judicial powers against the national authority with a view to ensuring that the sampling points are sited in accordance with those criteria. Such a judicial decision may give rise to the obligation to site sampling points at certain locations if it is clear from the available information that sampling points must be sited there. Otherwise the competent authorities may be obliged to undertake investigations in order to identify the correct locations.
(2) A limit value under Annex XI to Directive 2008/50 is exceeded within the meaning of Article 13(1) and Article 23(1) of the directive where the measurement result exceeds one single sampling point within the meaning of Article 7 of that directive.