C-3/96

COMMISSION v NETHERLANDS

OPINION OF ADVOCATE GENERAL FENNELLY delivered on 9 October 1997 *

Table of Contents

I — Introduction I - 3034

II — The relevant provisions of Community law I - 3034

III — Pre-litigation proceedings I - 3036

V — Merits of the application I - 3040

(a) The arguments of the parties I - 3040

(i) Number and total area of SPAs I - 3040

(ii) Fall in the numbers of bird populations I - 3042

(b) Analysis I -3042

(i) Interpretation of Article 4(1) of the Directive I - 3042

OPINION OF MR FENNELLY - CASE C-3/96

cient diversity and area of habitats is essen­ I — Introduction tial to the conservation of all species of birds ... [that] certain species of birds should be the subject of special conservation measures con­ cerning their habitats in order to ensure their survival and reproduction in their area of 1. In the present infringement action under distribution ... [and that] such measures mus- Article 169 of the EC Treaty, the Commis­ t... be coordinated with a view to setting up sion is seeking a declaration that the a coherent whole' (ninth recital). Kingdom of the Netherlands has not sufficiently complied with the obligation imposed on it by Article 4(1) of Council Directive 79/409/EEC of 2 April 1979 on the conservation of wild birds 1 to designate spe­ cial protection areas (hereinafter 'SPAs') for endangered species of wild birds. The Neth­ erlands contests the admissibility of the 4. Article 1 is complemented by Article 2 action and rejects the Commission's claims which reads as follows: on their merits.

II— The relevant provisions of Community law 'Member States shall take the requisite mea­ sures to maintain the population of the spe­ cies referred to in Article 1 at a level which corresponds in particular to ecological, scien­ tific and cultural requirements, while taking 2. The general structure and objectives of account of economic and recreational 2 requirements, or to adapt the population of the Directive are well known to the Court, and I will only reproduce below those provi­ these species to that level.' sions which are directly relevant to the present proceedings.

In this connection, account shall be taken of: 2. Member States shall take similar measures for regularly occurring migratory species not listed in Annex I bearing in mind their need for protection in the geographical sea and (a) species in danger of extinction; land area where this Directive applies, as regards their breeding, moulting and winter­ ing areas and staging posts along their migra­ tion routes. To this end, Member States shall pay particular attention to the protection of wetlands and particularly to wetlands of (b) species vulnerable to specific changes in international importance. their habitat;

(c) species considered rare because of small 3. Member States shall send the Commission populations or restricted local distribu­ all relevant information so that it may take tion; appropriate initiatives with a view to the

I - 3035

OPINION OF MR FENNELLY — CASE C-3/96

coordination necessary to ensure that the tive within two years of its notification . For areas provided for in paragraphs 1 and 2 the Netherlands, this period expired on above form a coherent whole which meets 6 April 1981. the protection requirements of these species in the geographical sea and land area where this Directive applies.

III — Pre-litigation proceedings

7. Article 4(4) of the Directive has been amended by Article 7 of Council Directive 92/43/EEC of 21 May 1992 on the conserva­ tion of natural habitats and of wild fauna and 3 10. A reasoned opinion was sent to the flora, though that amendment is not Netherlands on 14 June 1993, repeating the directly at issue in the present proceedings. claim that it had not designated sufficient SPAs for the endangered species listed in Annex I to the Directive. This sets a deadline for compliance of two months from the date of notification of the opinion. The Nether­ lands claims to have replied to the reasoned opinion (letter of 1 December 1993, Annex 1 8. In accordance with Article 18, the Mem­ to the statement of defence); the Commis­ ber States were obliged to 'bring into force sion states that it never received any reply to the laws, regulations and administrative pro­ the reasoned opinion. The present proceed- visions necessary to comply with this Direc­

4 — Case 236/85 Commission v Netherlands [1987] ECR 3989, 3 — OJ 1992 L 206, p. 7, hereinafter the 'Habitats Directive'. paragraph 2 of the judgment.

I - 3036

COMMISSION v NETHERLANDS

ings were initiated by an application regis­ that the letter also set out legal arguments, of tered at the Court on 5 January 1996. which the Commission has not taken account, in particular to justify the non- designation of certain individual sites, and that the Commission should at least have enquired of the Netherlands Government why, since it had requested two extensions of IV — Admissibility the deadline for replying to the reasoned opinion, it had not done so.

I - 3037

OPINION OF MR FENNELLY - CASE C-3/96

of the present action. Indeed, if the Nether­ (c) New pleas in law lands' view were correct, a Member State would in effect be able to prevent the Com­ mission from bringing the matter before the Court by simply refusing to reply to the rea­ soned opinion. I am therefore of the view that the Netherlands' objection to the admis­ sibility on this point should be rejected. 15. Thirdly, the defendant contends that the Commission's complaint regarding the insuf­ ficiency of the total area of SPAs, and their qualitative insufficiency, as well as specific complaints regarding the Friesian IJs- selmeerkust and Hooge Platen on the West­ ern Scheldt, were set out for the first time in the application, and that it was therefore unable to answer these at the pre-litigation stage of the proceedings.

(b) Nature of the obligation under Article 4(1) of the Directive

I - 3038

COMMISSION v NETHERLANDS

not invited to make specific findings in rela­ IBA94 should be discounted in so far as the tion to either area. Commission relies upon it to show a breach of the Directive.

17. However, in so far as the Commission's complaint concerns the financing the Neth­ erlands has received in respect of these two 19. The Commission argues that its allega­ areas under Council Regulation (EEC) tions are based on IBA89, and that its reli­ No 1872/84 of 28 June 1984 on action by the ance on IBA94 was unnecessary. It expresses 8 Community relating to the Environment, I surprise that the Netherlands should object am of the opinion that it is inadmissible, as to its citing all the scientific evidence avail­ neither of these areas is mentioned in either able, and particularly the most recent source the letter of formal notice or the reasoned whose scientific reliability has not been chal­ 9 opinion. lenged in these proceedings.

I - 3039

OPINION OF MR FENNELLY - CASE C-3/96

reports on the situation in the Netherlands essentially on two factors to support its view, in a period subsequent to the period laid each of which is contested by the defendant, down in the reasoned opinion for compli­ supported by the Federal Republic of Ger­ ance therewith. many.

(i) Number and total area of SPAs 21. The Commission has also argued that the SPAs classified by the Netherlands do not conform to the qualitative criteria set by the Directive. In particular, it alleges that the inclusion of fresh water lakes and marshes and moorland in such areas is insufficient. 23. A study completed in 1989 by the Inter­ The only evidence put forward by the Com­ national Council of Bird Preservation (here­ mission in support of this specific aspect of inafter ΊBA89') identified 70 sites in the its claim is taken from IBA94, and is hence, Netherlands, covering an area of 797 920 in my view, inadmissible. hectares, as qualifying for classification on ornithological grounds. IBA94, which is an updated version of IBA89 drawn up by a number of Netherlands organisations pub­ lished in December 1994, identified 87 sites, covering 1 089 357 hectares, as suitable for V — Merits of the application classification as SPAs. A list drawn up by the Netherlands Ministry of Agriculture and Fisheries in 1991, the reliability of which is contested by the Commission, identified 53 suitable sites covering 398 180 hectares.

(a) The arguments of the parties

Annex I. The extent of the Netherlands' fail­ 26. It concludes that 40% of the total area o ings in this regard appears even more strik­ the territories, and 40 out of the 87 site ing in the light of IBA94; only 35 of the (46%), listed in IBA94 benefit from naturt 87 sites, and less than one-third of the total conservation measures. Furthermore, in indi­ area, suitable for classification have been so cating only two individual sites which classified. should have been classified, the Commissior has not shown that the Netherlands hai exceeded the margin of appreciation it enjoys under the Directive to choose 'the mosi appropriate territories'; the Court has acknowledged that the Member States are better placed than the Commission to deter­ mine which of the Annex I species live on their territory. Nor has the Commission challenged the validity of the criteria on the basis of which the Netherlands selects SPAs; that each of the three lists mentioned is dif­ ferent shows that the application of such cri­ 25. The Netherlands' main line of defence is teria can give different results and results that Article 4(1) of the Directive does not which vary over time. Supported on this require it to classify a given number or total point by Germany, the defendant argues that area of SPAs. In its view, the classification of the rule on which the Commission relied, to SPAs is only one of the measures a Member wit, that Member States must classify at least State may take in order to comply with half of the suitable sites in their territory, Article 4(1); a breach of this provision can does not appear in the Directive. only arise if a Member State has not adopted any special conservation measures. It is therefore the whole body of measures adopted in respect of a specific site which is decisive. The defendant provides a list of other conservation measures which arc rel­ evant in this regard, such as the 1967 Wet houdende voorziening in het belang van de natuurbescherming (Nature Conservation n Law), the purchase of sites by nature con­ servation organisations, nature management contracts with agricultural organisations, the classification of wetlands under the Conven­ tion on Wetlands of International Impor­ tance especially as Waterfowl Habitat ('the ,2 Ramsar Convention'), and the Nether­ 27. The Commission argues in reply that lands' bird conservation plans. Article 4(1) creates a specific obligation to classify SPAs, which is not satisfied by the adoption of other measures. It also contends 11 — Stb. 572, 1967. that the Netherlands has not shown that the 12 — United Nations Treaty Scries Volume 996, p. 245; sec also measures on which it relies provide a suffi­ Commission Recommendation 75/66/EEC of 20 December 1974 to Member States concerning the protection of birds cient level of protection for the species con­ and their habitats (OJ 1975 L 21, p. 24). cerned.

29. The defendant argues that bird popula­ 30. The correct interpretation of Article 4(1) tions are in their nature subject to fluctua­ of the Directive must first be resolved. tion, and cites eight species whose numbers According to the Netherlands, the classifica­ have greatly increased, and one, the Great tion of SPAs is an important conservation egret (Egretta alba), which has been sighted measure, but is not rendered mandatory by in the Netherlands for the first time. It fur­ this provision; a Member State could only be ther argues, as regards the species listed by found to be in breach of Article 4(1) if it had the Commission, that four of these hibernate not adopted any special conservation mea­ in the African Sahel swamps, and that the fall sures. It follows, under this view, that the in the numbers may be due to the situation mere finding that a Member State had classi­ there; the populations of all the species men­ fied less than half the territories in number and area would not suffice to establish that the Member State was in breach of its obliga­ tions under the Directive. The Netherlands 13 — Case C-355/90 Commission v Spain [1993] ECR. I-4221, describes a number of other conservation 'Santoña Marshes', paragraph 15 of the judgment.

I - 3042

COMMISSION v NETHERLANDS

33. If the Netherlands' interpretation of its obligations under Article 4(1) were taken to its logical conclusion, a Member State could escape the obligation to classify SPAs, where it took the view that other special conserva­ 32. The rather extreme hypothesis pro­ tion measures were sufficient to ensure the pounded by the Netherlands does not appear survival and reproduction of the endangered to me to be justified either by the wording species. Member States would thus be able to or objectives of the Directive, and is not sup­ escape the obligations imposed upon them ported by the Court's case-law in this area. by Article 4(4) to take appropriate steps to The 'preservation, maintenance and avoid the deterioration of habitats or distur­ re-establishment of biotopes and habitats', bances affecting the birds in protected areas- including the creation of protected areas, is . Such an interpretation would also render an obligation which applies in respect of all Article 4(3) nugatory, as there would be no species of wild birds covered by the Direc­ special protection areas to form 'a coherent tive, in accordance with Article 3(2)(a). The whole'.

I - 3043

OPINION OF MR FENNELLY — CASE C-3/96

17 — Case C-44/95, cited in footnote 2 above, paragraph 23 of 14 — Case C-355/90, cited in footnote 13 above, Opinion of the judgment. Advocate General Van Gerven, paragraph 14 (p. I-4249). 18 — Case C-57/89 Commission v Germany (hereinafter 'Ley- bucht Dykes'), Opinion of Advocate General Van Gerven, 15 — Ibid., paragraph 26 of the judgment paragraph 26, footnote 24, [1991] ECR I-883, at p. 1-914. 16 — Paragraph 15.

I - 3044

COMMISSION v NETHERLANDS

38. In my opinion, Article 4(1) creates both obligations of a general character, and spe­ 39. Much argument has been devoted to the cific obligations in relation to individual scope of Member States' margin of discretion sites. In particular, the requirement that the in selecting SPAs. The Netherlands argues most suitable territories 'in number' be clas­ that the application of Article 4(1) is based sified as SPAs can only be judged taking on a concrete appreciation of whether a par­ account of the general degree of a Member ticular site is amongst the most suitable ter­ State's compliance with the fourth subpara­ ritories, and points out that the previous graph of Article 4(1); for an individual site, cases dealt with by the Court have all con­ numerical suitability is irrelevant. The cerned the question of whether a Member English rendition of the criterion 'most suit­ State should have classified an individual site able ... in size' is slightly ambiguous, and appears to be inconsistent with the ninth recital in the preamble and with some of the other language versions. The French version, 19 — The Danish ('til ... udstrækning er bedst egnede'), German ('die ... flächenmäßig geeignetsten'), Italian ('i ... più idonei for example, reads 'les plus appropriés ... en ... in superficie'), Greek ('τα πιο κατάλληλα, σε ... επιφάνεια'), Spanish ('los ... más adecuados ... en superfi­ superfície', which corresponds to 'most suit­ cie'), Portuguese ('os ... mais apropriados ... em extensão'), able ... in area', while the Dutch-language Finnish ('kooltaan sopivimmat') and Swedish ('storlek är mest lämpade') ... all either refer to 'area' or use a term version, which reads 'naar ... oppervlakte ... which can mean 'area' or 'size'.

tive factor is that the areas must be, as regards their number and area, suitable for the conservation of the species concerned and for the establishment, along with those classified by the other Member States, of a coherent network of protection areas. In its view, this provision does not require that a particular number of SPAs be classified, but 42. In the present case, there was no dispute rather obliges Member States to ensure that as to the identification of the wild bird spe­ the SPAs which are classified be appropriate cies which require protection on the terri­ for the conservation of endangered bird spe­ tory of the Netherlands. As the German cies. Government argued in Leybucht Dykes, the selection of an SPA entails an extremely complex assessment of the most varied facts 22 and requires considerable scientific work. In the present case, the Commission has rec­ ognised that Member States are not obliged to classify a separate SPA for each Annex I species. Some species require more protec­ 41. Though the margin of discretion is not tion than others, and the classification of a mentioned anywhere in the text of particular site as an SPA could provide pro­ Article 4(1), the Court noted in Ley bucht tection for different endangered species at Dykes that 'the Member States do have a cer­ the same time. It appears to me that the dis­ tain discretion with regard to the choice of cretion of the Member States operates in the territories which are most suitable for respect of the evaluation, according to objec­ 20 classification as special protection areas'. tive ornithological criteria, of the suitability In Commission v Italy, the Court had explained that 'the management of the com­

43. Whatever the scope of a Member State's discretion concerning the classification of an individual site, I do not sec how it can assist the defendant in the present proceedings. The Commission is seeking to prove that the Netherlands has not classified a sufficient number and area of SPAs to comply with its general obligations under Article 4(1) of the 45. The Netherlands adds that the Member Directive; the Netherlands is not arguing States are obliged, when adopting special that it has a general discretion not to comply conservation measures, to take account of with these obligations. the economic and recreational requirements to which Article 2 refers. It modified this affirmation somewhat in its rejoinder, fol­ lowing the judgment of the Court in 2 RSPB, 4 to argue that the scope of the obli­ gations arising from Article 4(1) should be interpreted in the light of Articles 1 and 2 of 44. Furthermore, the reliance put by the the Directive. Such a contention is in my Netherlands on its margin of appreciation in view in evident contradiction with the first the present case is inconsistent. On the one paragraph of the operative part of the judg­ hand, it argues that the Member States are ment in question, which states that better placed than the Commission to iden­ 'Article 4(1) or (2) of [the Directive] is to be tify sites which are deserving of protection; interpreted as meaning that a Member State is not authorised to take account of the economic requirements mentioned in Article 2 thereof when designating a Special 23 — Case C-44/95, cited in footnote 2 above, paragraph 26 of Protection Area and defining its bound­ the judgment. See also Case C-72/95 Kritaijcveid and Oth- 25 ers, where the Court adopted a similar approach to Member aries'. States' discretion under Article 4(2) of Council Directive 8 337 / EC of 27 Jun 1985 on th 5/ E e e assessment of the effects of certain public and private projects on the environ­ ment (OJ 1985 L 175, p. 40); an analogy between these two types of discretion had been suggested by the Netherlands Raad van State, the referring court ([1996] ECR I-5403) 24 — Case C-44/95, cited in footnote 2 above. paragraphs 44, 49 and 50 of the judgment. 25 — Ibid., [1996] ECR I-3805, at pp. 1-3856 and 1-3857.

(iii) The probative value of the inventories behest by the Muséum National d'Histoire Naturelle, Paris, in 1987. In IBA89, 'for the of important bird areas first time individual sites in each European country were evaluated in a standard way, and a continent-wide network of sites was identified that, if protected, would safeguard a significant proportion of the European 28 populations of many species'. The notes on the inventory for the Netherlands, set out in Annex 7 to the Commission's application, identify the following three categories of cri

26 As Community. Five other sites were included the Court noted in RSPB, 'notwithstanding in the inventory for other reasons, for example Het Zwin because it is contiguous the divergences between the various language with an important bird area in Belgium, or versions of the last subparagraph of Krammer and Volkerak because this site Article 4(1), the criteria in question are orni 27 'could, if managed appropriately, be devel thological criteria'. It follows in my view oped into an important freshwater ecosys that ornithological criteria must also be used

47. In seeking to demonstrate that the Neth erlands has failed sufficiently to implement its obligations in this regard, the Commis sion relies primarily on IBA89, though refer ring also to the modified and updated list IBA94. IBA89 is itself an updated version of 28 — Tucker et al., Birds in Europe: Their conservation status (cited by the Commission in Annex 7 of its application), an inventory drawn up at the Commission's BirdLife International, Cambridge, 1994, p. 20. 29 — The Commission expert explained at the hearing that this phrase referred to distinct fly-away populations of bird spe cies from their breeding grounds to their staging and win tering areas, which may include areas outside the territory 26 — See paragraph 6 of the present Opinion, above. to which the Directive applies. Measures which favour the protection of one such population presumably have no 27 — Case C-44/95, cited in footnote 2 above, paragraph 26 of effect on the other such populations. the judgment.

48. Germany has argued forcefully that IBA89 and IBA94 only contain lists of sites which in accordance with scientific criteria could potentially contribute to the conserva­ tion of endangered species; these lists are nei­ ther part of the Directive, nor legally bind­ ing. Moreover, neither the criteria upon which the lists are based, nor the resulting lists have been agreed upon at the Commu­ nity level. It adds that the fixing of a lower limit of 50% of sites classified is arbitrary 50. It follows, in my view, that IBA89 not and cannot be scientifically verified. only constitutes scientific evidence, the necessity for which Germany appears to accept in principle, but was expressly designed for use in the application of the Directive. It is not itself conclusive or consti­ tutive of a legal obligation, but can be relied upon in demonstrating the extent of a Mem­ ber State's compliance therewith, both as regards the general obligation and specific 30 sites. As regards an individual site, it is 49. This argument seems to confuse the legal open to a Member State to produce better obligation and the evidence required to scientific evidence to show that it is not prove non-compliance. It is, of course, true amongst the 'most suitable' for the conserva­ that IBA89 is not per se binding on the tion of Annex I species. Similarly, it is open Member States; if it were, the present case to a Member State to produce contrary evi­ could have been disposed of much more dence to prove that the total figures for briefly. Though drawn up by the Eurogroup SPAs, in number and in area, which arise for the Conservation of Birds and Habitats from IBA89, or from any other such list in conjunction with the International Coun­ upon which the Commission relies, are erro­ cil of Bird Preservation (now 'BirdLife') neous. rather than through any exclusively Com­ munity procedure, IBA89 was prepared for the relevant Directorate-General of the Commission and in cooperation with Com­ 30 — The Commission declared at the hearing that it had relied mission and national experts; the inventory on IBA 89 in SantoñaMarshes, though this does not appear from the case report.

51. In the present case, the Netherlands does the biogeographical population of species of not directly contest the scientific viability of water birds, seem to me to be much less IBA89, except as regards the matter of the comprehensive and detailed than the list of definition of a minimum viable area for an categories of numerical criteria on which SPA. In its pleadings, it notes that IBA94 IBA89 was based (Table 1 to Annex 7 to the gives some indication of the definition of a Commission's application).

It suggests that the difference between the list of sites which qualify for classification according to the IBA inventories, and those which have in fact been classified, can be explained by the nature of ornithological data. It also suggests that the difference in the number of SPAs results from a difference in the delimitation and regrouping of sites, while the difference 53. While the Netherlands has not been able in area is due to the absence of adequate cri to demonstrate the objective superiority of teria for defining the boundaries of sites for its own national list, the Commission has classification. questioned the scientific basis of the Nether lands' list. In an annex to its reply, the Com mission has presented a table comparing the (theoretical) results of the application of the three criteria on which the Netherlands' list is based with the actual classification of SPAs for 26 Annex I species found in the Nether

In par ticular, the terms of the first of the three 31 _ The Commission does not specify the source of the popula criteria as set out in the Netherlands' tion figures; the table is not, however, relied upon specifi defence, which refers solely to the regular cally to show that the Netherlands is in breach of Article 4(1), but to challenge the reliance by the Nether presence on a site of at least one per cent of lands on its own 1991 list.

55. The Netherlands has also sought to rely on the differences between IBA 89 and IBA 94 to show that the application of orni­ 57. The latter part of the Commission's thological criteria gives uncertain results. The answer to the parliamentary question raises Commission has vigorously challenged this the issue of the feasibility of determining the assertion. Seven of the 12 sites in the later list proportion of particular species which missing from IBA 89 were included to take should be within SPAs in the territory of a account of the addition of new species to given Member State. In the annex to its reply Annex I to the Directive, while the other five in the written proceedings, cited above, the are the result of various objective factors Commission table shows the percentage of such as a different division or regrouping of all the Annex I species in the Netherlands sites, an increase in knowledge, or an evolu­ which arc found in the five most suitable tion in the bird populations in the Nether­ sites, and the percentage of these species lands. The Commission's explanations on which are in SPAs. The figures arc telling this point seem to be convincing; moreover, indeed; for example, six species whose per­ the Netherlands has itself pointed out that centage population in the five best sites the situation of bird species is in constant ranges from 19% to 100% have none (0%) evolution over time. of their population in SPAs. However, while

33 — Written question No 131/93 by Mr Floras Wijsenbeek. 32 — Summary of Annex 9 to the application. OJ 1993 C 258, p. 7.

58. At the time of the Commission's appli­ 60. As I have already stated, the obligation cation in the present case, the Netherlands which arises for the Member States by virtue had classified 23 SPAs, covering a total area of the fourth subparagraph of Article 4(1) is of 327 602 hectares. As the equivalent figures to classify all the most suitable territories, which arise from IBA89 are 70 sites covering identified by the application of reliable 797 920 hectares, the Commission is of the objective scientific criteria. The Commis­ view that the Netherlands' failure to comply sion's duty in proceedings such as the present is to indicate the total number and with its obligations to classify SPAs under total area of SPAs which a Member State Article 4(1) is manifest. should classify on the basis of these criteria, compared to the number and area of SPAs which it has in fact classified. The existence of any discrepancy between the two sets of figures can of course be challenged by the Member State in question; if proven before the Court, the discrepancy is sufficient to establish a breach of the defendant's obliga­ 59. In seeking to show that this state of tions under the Directive. The Commission's affairs constitutes a manifest breach of the conclusion on the basis of the evidence in the Netherlands' obligations under the Direc­ present case that the Netherlands has failed tive, the Commission has repeatedly referred to classify sufficient SPAs in number and to the figure of a half of the number and area results from the normal process of legal total area of sites. The Netherlands and Ger­ reasoning, and does not rely on any pre­ many have both pointed out that such a fig­ sumption. ure is not in the Directive, and is arbitrary

VI — Conclusion

In the light of the foregoing, I recommend to the Court that it:

(1) Declare that , by failing to classify a sufficient number and area of special pro­ tection areas in accordance with Article 4(1) of Council Directive 79 / 409 / EEC of 2 April 1979 on the conservation of wild birds , the Kingdom of the Neth­ erlands has failed to comply with its obligations under the EC Treaty;

(2) Order the Kingdom of the Netherlands to pay the costs.

I - 3053