Press release no. 38/2019 of 15 May 2019

Remittal of the legal dispute concerning the Lünen hard coal-fired power plant

The Higher Administrative Court (Oberverwaltungsgericht) for the Federal State of North Rhine-Westphalia must once more decide on the action brought against the Lünen hard coal-fired power plant. That was decided by the Federal Administrative Court (BVerwG, Bundesverwaltungsgericht) in Leipzig today.


The claimant, a recognised nature conservation association, is contesting an administrative act deciding beforehand on several aspects of an application concerning immission control law (immissionsschutzrechtlicher Vorbescheid) issued to the power supply company summoned as a third party to attend the proceedings as a party whose rights may be affected (hereinafter summoned third party) as well as the 1st and 7th partial approvals for the hard coal-fired power plant in Lünen, which has in the meantime been built and is in regular operation.


The Higher Administrative Court dismissed the action. It held that, in assessing whether the hard coal-fired power plant, in combination with other projects, may adversely affect Natura 2000 sites, the material date was the date of submission of a verifiable application for an approval. Projects applied for later but approved in the meantime were not to be taken into account. The cut-off criterion used to determine the area of influences of the planned plant and thus the area to be examined in the assessment of the implications for a site was to be based on an amount of no more than 0.5% of the critical load (Grenzbelastung) for the respective habitat type under consideration. All projects since the Natura 2000 sites were protected in December 2004 are to be included in the assessment of the additional load (Zusatzbelastung).


Following the claimant's appeal on points of law, the Federal Administrative Court set aside the judgment of the Higher Administrative Court. When including further projects in the assessment of the implications for the site under the Habitats Directive, as a rule, all projects for which an approval has already been granted must be taken into account. The approach adopted by the Higher Administrative Court, namely to disregard in the cumulative assessment those projects which have been approved in the meantime but were applied for later, breaches the requirements of EU law to be taken into account in the interpretation and application of the national provisions. Since the Higher Administrative Court did not make any findings on the effect caused by nitrogen inputs from a copper recycling plant which was approved before the Lünen power plant and is to be included in the cumulative assessment, the matter had to be referred back to the Higher Administrative Court for a further hearing and decision. In its new decision, the Higher Administrative Court will have to take into account the fact that there is no reason to modify the project-related cut-off criterion of 0.3 kg N/(ha·a), which is a generally recognised value in nature conservation, even in the case of cumulative loads. Furthermore, it is not always necessary, when assessing whether a Natura 2000 site is subject to gradual deterioration through minor inputs, to go back to the time when the site was first protected.


BVerwG 7 C 27.17 - judgment of 15 May 2019


Judgment of 15 May 2019 -
BVerwG 7 C 27.17ECLI:DE:BVerwG:2019:150519U7C27.17.0


Please note that the official language of proceedings brought before the Federal Administrative Court of Germany, including its rulings, is German. This translation is based on an edited version of the original ruling. It is provided for the reader’s convenience and information only. Please note that only the German version is authoritative. Page numbers in citations have been retained from the original and may not match the pagination in the English version of the cited text. Numbers of paragraphs that have completely been omitted in the edited version will not be shown.
When citing this ruling it is recommended to indicate the court, the date of the ruling, the case number and the paragraph: BVerwG, judgment of 15 May 2019 - 7 C 27.17 - para. 16.

Integration of new projects into the assessment of the implications for the site under the Habitats Directive (cumulative assessment)

Headnotes

1. Other plans and projects are to be integrated into the assessment of the implications for the site (cumulative assessment) under section 34 (1) first sentence BNatSchG, if their impact, and therefore the extent of the cumulative impact, can be predicted reliably. As a matter of principle, this is not already the case upon submission of verifiable documents or public display of the documents, but only when the necessary approval decisions have been issued (confirmation of the established jurisprudence of the BVerwG, e.g., judgments of 21 May 2008 - 9 A 68.07 - (...) and of 9 February 2017 - 7 A 2.15 - BVerwGE 158, 1 para. 219).

2. The project-related cut-off value (Abschneidewert) for eutrophying nitrogen inputs of 0.3 kg N/(ha·a) does not require correction, also in view of the cumulative effects of several projects.

3. As a rule, it is not necessary to relate the cumulative assessment back to the date on which the Habitats Directive sites were granted protection in December 2004.

  • Sources of law
    Habitats Directivearticle 6 (3)
    Environmental Appeals ActUmwRG, Umwelt-Rechtsbehelfsgesetzsection 3
    Federal Nature Conservation ActBNatSchG, Bundesnaturschutzgesetzsection 34 (1) and (2)
    Federal Immission Control ActBImSchG, Bundes-Immissionsschutzgesetzsection 10 (3) and (5)
    Environmental Impact Assessment ActUVPG, Gesetz über die Umweltverträglichkeitsprüfungsection 12 (2)
    Administrative Procedure ActVwVfG, Verwaltungsverfahrensgesetzsection 10 second sentence
    Ninth Ordinance for the Implementation of the Federal Immission Control Act9. BImSchV, Neunte Verordnung zur Durchführung des Bundes-Immissionsschutzgesetzessections 4 and 4e

Summary of the facts

The claimant, an environmental protection association, contests an administrative act deciding on one or several aspects of an application beforehand (hereinafter preliminary notice) issued to the operator who has been summoned to attend the proceedings as a party whose rights may be affected (hereinafter summoned third party) as well as partial approvals for the construction of a new hard coal-fired power plant in Lünen. The power plant has been built in the meantime and has been in regular operation since January 2014.

In March 2007, the operator applied to the Arnsberg Regional Government (Bezirksregierung) for a preliminary notice to determine whether a hard coal-fired power plant with a rated thermal input of up to 1,705 megawatts (MW) and a net electrical output of 750 MW is eligible for approval and for several partial approvals. In its notice of 6 May 2008, the Regional Government issued the preliminary decision and the first partial approval. On 14 March 2011, the operator applied for the seventh partial approval to operate the coal-fired power plant for the use of hard coal of different qualities.

The preliminary notice and the first partial approval were set aside by the Higher Administrative Court (Oberverwaltungsgericht) in its judgment of 1 December 2011.

On 9 July 2012, the operator once more applied for a preliminary notice to determine the eligibility of the planned hard coal-fired power plant for approval, and, on 15 May 2013, for a new first partial approval, and updated its application for the seventh partial approval on 5 July 2013.

On 20 November 2013, the Regional Government issued the preliminary notice, on 21 November 2013 the first partial approval and on 22 November 2013 the seventh partial approval.

In its judgment of 16 June 2016, the Higher Administrative Court dismissed the action filed on 30 December 2013. The claimant's appeal on points of law was successful.

Reasons (abridged)

16 The claimant's admissible appeal on points of law is well-founded. The Higher Administrative Court's judgment is based on a breach of federal law (section 137 (1) no. 1 of the Code of Administrative Court Procedure (VwGO, Verwaltungsgerichtsordnung)). Furthermore, it is not correct for other reasons (section 144 (4) VwGO). In the absence of sufficient factual findings by the lower court, the case is to be remitted to the Higher Administrative Court (section 144 (3) first sentence no. 2 VwGO).

18 2. The Higher Administrative Court's considerations on the assessment of the implications for the site under the Habitats Directive do not satisfy (...) the requirements of federal law in every respect. This applies to the question under which conditions the assessment of the implications for the site is to be extended to such impairments which may arise "in combination with other plans and projects" (a), to the determination of the cut-off criteria (Abschneidekriterium) to be applied in the context of this assessment (b) and to the specific point in time of the cumulative assessment (c).

19 a) (aa) The assessment of the implications for the site under section 48d (3) of the Landscape Act of the Federal State of North Rhine-Westphalia (LG NRW, Landschaftsgesetz Nordrhein-Westfalen), section 34 (1) of the Federal Nature Conservation Act (BNatSchG, Bundesnaturschutzgesetz) and article 6 (3) of Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora (Habitats Directive - OJ L 206 p. 7) may not be limited to the question of whether a plan or project is likely, by reason of the impact it produces itself, to have a significant effect on the conservation objectives of a site of Community importance. Rather, it must also be extended to cover such effects which may arise "in combination with other plans and projects". The cumulative assessment is intended to prevent gradual effects as a result of successively approved plans and projects which, taken individually, do not affect the site, in so far as their combined impact may lead to a significant effect on conservation objectives (Federal Administrative Court (BVerwG, Bundesverwaltungsgericht), decision of 5 September 2012 - 7 B 24.12 - (...) para. 12; see also Court of Justice of the European Union (CJEU), judgment of 24 November 2011 - C 404/09 [ECLI:EU:C:2011:768], Commission/Spain - para. 103 to 107, 76 to 78 and of 26 April 2017 - C-142/16 [ECLI:EU:C:2017:301], Moorburg - para. 57 to 61; (...)). According to the established jurisprudence of the Federal Administrative Court, the impacts of the other plans and projects on the conservation objectives of the site of Community importance must be reliably foreseeable. The necessary certainty in principle exists only once the approval decisions for the other plans and projects have been issued (BVerwG, judgments of 21 May 2008 - 9 A 68.07 - (...) para. 21, of 14 June 2011 - 9 A 12.10 - (...) para. 81, of 24 November 2011 - 9 A 23.10 - Rulings of the Federal Administrative Court (BVerwGE, Entscheidungen des Bundesverwaltungsgerichts) 141, 171 para. 40, of 15 July 2016 - 9 C 3.16 - (...) para. 56 and of 9 February 2017 - 7 A 2.15 - BVerwGE 158, 1 para. 219).

20 In contrast, in the opinion of the Higher Administrative Court, the impacts of a project are usually already sufficiently specifically foreseeable when a verifiable application is submitted. It held that, where participation by the public takes place, it could be assumed that the application is verifiable at the latest when the documents are publicly displayed. Projects applied for later but approved in the meantime were not to be taken into account. It held that this applied even if such projects are already operating (...). In addition, even if the relevant approval was subsequently annulled, a priority position, once obtained, was maintained, provided that the project initiator maintained its project, unless the annulling judgment bindingly determined that the project could definitely not be carried out on the planned site (...).

21 (bb) The opinion of the Higher Administrative Court cannot be endorsed. It is not consistent with the requirements of the assessment of the implications for the site under the Habitats Directive, as they result from section 34 (1) BNatSchG and article 6 (3) of the Habitats Directive. As part of the assessment of the implications for the site under the Habitats Directive (FFH-Verträglichkeitsprüfung), the approval authority must be certain at the time of the decision approving the project that from a scientific point of view there is no reasonable doubt that the project will not adversely affect the site as such, even in combination with other plans and projects. Such an assessment may not have lacunae and must contain complete, precise and definitive findings (CJEU, judgments of 24 November 2011 - C-404/09 - para. 103 to 106, of 17 April 2018 - C-441/17 [ECLI:EU:C:2018:255], Commission/Poland - para. 114 and of 8 November 2018 - C-461/17 [ECLI:EU:C:2018:883], Holohan et al. - para. 33) This is the only way of complying with the aim of article 6 (3) of the Habitats Directive, which is to record the impacts as realistically as possible at the time of approval.

22 (1) It is not compatible with this if the Higher Administrative Court considers that a cumulative assessment of the "priority project" does not need to take into consideration the impacts of "subordinate projects" solely because these projects were applied for later than the project awaiting approval. With this privileged treatment of the first project applied for, derived by the Higher Administrative Court from the priority principle, a part of the impacts which have been determined or can be foreseen with sufficient certainty on the basis of the approvals granted in the meantime for "subordinate projects" and which may affect the conservation objectives is deliberately ignored at the material time of the approval decision for the "priority project". This means that a lacunary and incomplete cumulative assessment for the last approved project is being accepted. This applies in particular if - as was done by the Higher Administrative Court - indefinite priority is given to the project first applied for. This cannot be countered by the argument that, according to the approach of the Higher Administrative Court, the foreseeable effects of the project applied for first and thus the priority project must be taken into account in the approval procedures for the subordinate projects. Such a "transfer of conflict" to approval procedures applied for later but completed earlier cannot prevent with the necessary reliability that lacunae in the recording and assessment of cumulative impacts may occur in more than merely exceptional cases.

23 Upon submission of verifiable documents, all documents required for the assessment of the implications for the site under the Habitats Directive and for the environmental impact assessment (EIA) must be handed in and examined for completeness by the approval authority (see sections 4 and 4e of the Ordinance on the Approval Procedure - Ninth Ordinance for the Implementation of the Federal Immission Control Act (9. BImSchV, Neunte Verordnung zur Durchführung des Bundes-Immissionsschutzgesetzes)). However, this does not yet finally determine whether the documents fully comply with the substantive requirements of the Habitats Directive and whether the assessment carried out therein contains complete, precise and definitive findings in line with the best scientific knowledge in the field (...). Consequently, at the time of submission of the verifiable documents, the assessment of the implications for the site of the priority project does not provide a sufficient basis for including the impact of this project into the cumulative assessment of another project.

24 The same applies until the approval procedure for the priority project has been completed. After submission of the documents, the procedure for the participation of authorities and the public (section 10 (3) and (5) of the Federal Immission Control Act (BImSchG, Bundes-Immissionsschutzgesetz)) must be conducted. The comments of the authorities and the objections of the public, including associations, will often give rise to new and more in-depth nature conservation assessments. Furthermore, it can - this applies in particular to major projects such as the present one - not be ruled out that in the course of the approval procedure, which takes a long time, there may be a need to modify the plant itself or parts of the plant, for which the approval is sought, for economic reasons or because of technical developments. Modifications of the plant, in particular changes in plant capacity, in turn usually influence the type and extent of emissions from the plant and immissions. Especially with regard to the implications for the site under the Habitats Directive, this will often entail changes in the protection concept and possible coherence measures. In as far as the Higher Administrative Court intends to counteract this by making worst-case considerations of the impact of the project prioritised under aspects of time, this is not convincing. On the one hand, there is a risk that repeated and far-reaching worst-case scenarios overestimate both the impact of a project and the actual effects on a Habitats Directive site, thus making it more difficult for all subsequent projects to meet the approval requirements (...). On the other hand, there is no justification for unilaterally burdening the subordinate project which is ready for approval at an earlier stage with the problems arising from the fact that the planning of the priority project has not yet been sufficiently consolidated. This applies in particular to prevention, protection and coherence measures that become necessary to compensate for the cumulative impact.

25 (2) These considerations also apply in the event that projects applied for earlier - which in the opinion of the Higher Administrative Court are to be included in the cumulative assessment - are only approved after the project which had been applied for later. In this constellation, too, the cumulative assessment for the project that was applied for later but was ready for approval at an earlier stage is necessarily based on provisional findings relating to the projects applied for earlier. This circumstance may also force worst-case assumptions to be made, which may entail derogation assessments and coherence measures, even though the eligibility for approval of the projects applied for earlier is uncertain.

26 (3) The approach taken by the Higher Administrative Court, which is not primarily aimed at the requirements of site protection but rather at the protection of legitimate expectations for the project developer, therefore does not meet the requirements of article 6 (3) of the Habitats Directive. Therefore, the practical considerations made by the Higher Administrative Court cannot change the relevance of the finalisation of the approval procedure for the inclusion of a project into the cumulative assessment. Notwithstanding this, the "priority approach" cannot be based on the assumption of improved practicability either. The Higher Administrative Court correctly assumes that the assessment of the implications for the site under the Habitats Directive is based on the factual and legal situation at the time of the last decision by the authorities so that after submission of a verifiable application, any changes, modifications, etc. of the projects applied for earlier must be taken into account as well as the current condition and development of the Habitats Directive site. Any cumulative assessment - irrespective of the approach taken - faces the practical difficulties associated with these legal requirements. In particular, the problem of the comparability and usability of older assessments of the implications for the site is independent of whether an approval has already been granted or whether the approval procedure is still ongoing. In accordance with the jurisprudence of the Federal Administrative Court, it is in principle not the task of the approval authority to review in the context of the cumulative assessment the lawfulness of the assessment of the implications for the site under the Habitats Directive of other projects at the time approval was granted (BVerwG, judgment of 9 February 2017 - 7 A 2.15 - BVerwGE 158, 1 para. 223). In principle, the effects of the project identified therein may therefore be used as a basis for the cumulative assessment. However, if new scientific or nature conservation findings and/or methods have emerged in the meantime, particularly with regard to the individual impact factors and cause-effect relationships, these must be taken into account in the cumulative assessment, which must assess the impacts of all projects to be considered applying uniform standards on the basis of an up-to-date inventory. The same applies to the progress made in setting standards, for example through research projects that have resulted in technical conventions and other working aids. The cumulative assessment for the last approved project must be based on the best scientific knowledge. Frictions cannot be avoided in view of the fact that the cumulative backdrop cannot be kept up to date "on a daily basis" until the project is approved. To a certain extent, however, such frictions can be countered by forward-looking planning. Experience shows that projects that compete under temporal aspects and that are or may become relevant for the cumulative assessment are usually known at an early stage or can be identified upon request by the competent authorities.

27 (4) In view of this finding, the considerations of the Higher Administrative Court on the protection of legitimate expectations are not decisive. Both can only exist within the scope of what is legally permissible. Moreover, the consideration of the Higher Administrative Court that using the submission of verifiable documents in the cumulative assessment as the relevant factor would prevent larger projects from being "slowed down" by other ("smaller") emission-intensive projects in the course of the complex approval procedure is not acceptable.

28 It is true that the project developer initially did what was necessary to obtain the desired approval. However, the developer has no right to have his project approved before others. The approval authority must carry out every administrative procedure in a timely manner (section 10 second sentence of the Administrative Procedure Act (VwVfG, Verwaltungsverfahrensgesetz)) and may not treat similar procedures differently without objective reasons (article 3 (1) of the Basic Law (GG, Grundgesetz)), so that, as a rule, applications received earlier must processed first (...). However, the timing of the approval of a project depends on the readiness of the application for approval, so that an application filed at a later date may be ready for decision at an earlier stage because it raises fewer questions or requires fewer participation processes (...). In such a case, the approval decision regarding the earlier application must take into account the "overtaking" project.

29 Section 12 (2) of the Environmental Impact Assessment Act (UVPG, Gesetz über die Umweltverträglichkeitsprüfung) does not lead to a different conclusion either (...). According to that provision, the developer of an earlier project who has already submitted complete documentation without approval having been granted is not required to carry out an environmental impact assessment retrospectively if his or her project only becomes subject to EIA approval in combination with a later project. At this advanced stage of the procedure, the developer of the earlier project is to be protected from being retroactively subjected to an EIA obligation as a result of cumulative projects which are added at a later date (see draft act of the Federal Government on the modernisation of the law on environmental impact assessment, Bundestag printed paper (BT-Drs., Bundestagsdrucksache) 18/11499 p. 86). However, the protection of the earlier project in terms of procedural law cannot be transferred to the substantive law question as to whether a necessary assessment of the implications for the site under the Habitats Directive must take into account the cumulative impacts of other projects.

30 b) The judgment also breaches federal law to the extent that the Higher Administrative Court has set a separate value for eutrophying and acidifying nitrogen inputs that is well below the cut-off value assumed by the approval authority.

31 aa) In breach of federal law, the Higher Administrative Court considered the cut-off value for project-related additional loads (Zusatzbelastungen) from eutrophying nitrogen inputs of 0.3 kg N/(ha·a) to be too high in cumulative assessments, and based its calculations on a project-related cut-off value of 0.05 kg N/(ha·a).

32 It assumes that the concept of critical loads (Grenzbelastungen) in the assessment of the implications for the site under the Habitats Directive for eutrophying nitrogen inputs represents a suitable standard and is legally unobjectionable. It also considers the application of a project-related cut-off value to be technically and legally justified. This assessment of the nitrogen load using critical loads and a cut-off value of 0.3 kg N/(ha·a) nitrogen has met with approval in the established jurisprudence of the Federal Administrative Court (BVerwG, judgment of 23 April 2014 - 9 A 25.12 - BVerwGE 149, 289 para. 34 et seqq.), which, however, in the past had no reason to distinguish between cut-off criterion and de minimis threshold. The cut-off values as published in both the 2013 Federal Highway Research Institute Report (BASt Report 2013, Balla et al., "Untersuchung und Bewertung von straßenverkehrsbedingten Nährstoffeinträgen in empfindliche Biotope", Final Report on Research and Development Project (FE-Vorhaben) 84.0102/2009 of the Federal Highway Research Institute (BASt, Bundesanstalt für Straßenwesen), series Forschung Straßenbau und Straßenverkehrstechnik, vol. 1099, November 2013 - Nitrogen Research and Development Report (FE-Bericht Stickstoff) -) and in the current 2019 Nitrogen Guidelines for Roads (Notes on the Examination of Nitrogen Inputs in the Assessment of the Implications for the Site under the Habitats Directive for Roads 2019 [H PSE] p. IX, 74), reflect the best scientific knowledge on the determination of the load of nitrogen inputs into protected habitats (...). The view of the Higher Administrative Court cannot be endorsed insofar as it considers the cut-off value of 0.3 kg N/(ha·a) to be too high in view of the necessary cumulative assessment and assumes a project-related cut-off value of 0.05 kg N/(ha·a) and, in addition, for exceptional cases, a cut-off value relating to the relevant protected habitat of 0.5% of the critical load of the respective (specific) habitat type under consideration (...).

33 (1) The cut-off criterion serves to determine the area of influences of a planned plant and thus the area and scope of the assessment of the implications for the site under the Habitats Directive. At the same time, it determines the projects to be included in the cumulative assessment (...). Systematically, this takes place before the examination of de minimis thresholds and is to be determined independently of these. If the cut-off value for very low critical loads is above the 3% de minimis threshold, the BASt Report 2013 of the Federal Ministry of Transport, Building and Urban Development (Bundesministerium für Verkehr, Bau und Stadtentwicklung), which is designed as an expert convention, states that priority should be given to the cut-off value because additional loads and detection limits are only theoretical.

34 (2) Contrary to the opinion of the Higher Administrative Court, the cut-off value of 0.3 kg N/(ha·a) must also be applied in cases of cumulative nitrogen loads from several projects (left open in BVerwG, judgment of 27 November 2018 - 9 A 8.17 - (...)).

35 (2.1) According to the findings of the Higher Administrative Court, the cut-off value of 0.3 kg N/(ha·a) is oriented at measurement uncertainties (...). Below this threshold, the additional load from a project can no longer be determined with reasonable accuracy or cannot be clearly distinguished from the background level (Hintergrundbelastung). Nitrogen inputs below the cut-off value can no longer be measured and thus the modelled values cannot be validated. These are of such small orders of magnitude that specific effects have not been observed in vegetation populations. The BASt Report 2013 speaks of theoretical additional loads that cannot be attributed to a project even under conservative assumptions. On the basis of the lowest detection limit, the cut-off value for nitrogen inputs converts into the order of 0.5 kg N/(ha·a) (...) To be on the safe side, the cut-off value has been set in the order of half the (rounded) detection limit at 0.3 kg N/(ha·a) (...).

36 (2.2) On the basis of this derivation, it is not justified to set the cut-off value of 0.3 kg N/(ha·a) for eutrophying nitrogen inputs at a lower level in deviation from the BASt Report 2013 in view of the cumulative effects of several projects, and, additionally, to carry out special assessments in individual cases related to the protected habitats. The assessment of the implications for the site according to section 48d (3) LG NRW, section 34 (1) BNatSchG and article 6 (3) of the Habitats Directive is linked to a project or plan likely to affect the Natura 2000 site. In this respect it is not necessary to prove a causal link between the project or plan and the significant effect on the conservation objectives. However, a certain probability of such an disturbance must exist ("be possible to establish") (CJEU, judgments of 14 January 2016 - C-141/14 [ECLI:EU:C:2016:8], Commission/Bulgaria - para. 58 and of 24 November 2011 - C-404/09 - para. 142; BVerwG, judgment of 23 April 2014 - 9 A 25.12 - BVerwGE 149, 289 para. 45). Purely theoretical concerns are therefore not to be taken into account. The precautionary principle under EU law, which is reflected in article 6 (3) of the Habitats Directive (article 191 (2) second sentence of the Treaty on the Functioning of the European Union (TFEU), see CJEU, judgment of 7 September 2004 - C-127/02 [ECLI:EU:C:2004:482], Raad van State/Netherlands - para. 58), also does not require the assessment of the implications for the site to be based on zero risk, because scientific evidence could never be provided for this. Rather, a project is permissible if, after completion of the assessment of the implications for the site, there is no reasonable doubt from a scientific point of view that significant effects will be avoided. In order to arrive at a reliable evaluation, the assessment of the implications for the site must take into account the "best relevant scientific knowledge" and thus presupposes the "exhaustion of all scientific means and sources" (BVerwG, judgment of 28 March 2013 - 9 A 22.11 - BVerwGE 146, 145 para. 41; CJEU, judgment of 26 October 2006 - C-239/04 [ECLI:EU:C:2006:665], Commission/Portugal - para. 20). Measured against this standard, the determination of the site to be examined in an assessment of the implications for the site under the Habitats Directive and thus the projects to be considered in the cumulative assessment cannot depend on nitrogen input that cannot be measured.

37 (2.3) An input value determined on the basis of calculation models which is neither empirically verifiable nor attributable with regard to its impact, represents a purely theoretical concern. Contrary to the opinion of the Higher Administrative Court, this approach underlying the BASt Report 2013 is not only plausible insofar as it relates to one project. A large number of concerns that cannot be measured and validated do not lead to an actual amount to be charged to a project with regard to impact either. Measurement uncertainties of cumulative projects can therefore not be added up, contrary to the assumption of the Higher Administrative Court. An addition of modelled and thus hypothetical inputs lying below the detection limit would not change the lack of proof of impact with regard to a project. As long as it is not possible to clarify whether, and if so to what extent, inputs exist at all and which sources they originate from, no sufficient likelihood of significant effect can be determined. Rather, such "inputs" have a diffuse impact and can at best be represented as part of the background level (...). However, it is not part of the assessment of the implications for the site to consider and establish measures to improve the background nitrogen level on a project basis. This rather is the task of site management (...).

38 (2.4) The fact that sites below the detection limit of 0.3 kg N/(ha·a) are not taken into account in the assessment of the implications for the site under the Habitats Directive, even if several projects are to be implemented in a close spatial context, is not objectionable against the background of the precautionary principle under EU law. The cut-off criterion takes this into account by using a value that is clearly below the detection limit and thus itself already shifts the consideration of the impacts of the project into a hypothetical range and therefore, as a precaution, considerably enlarges the assessment area to be included. The fact that the cut-off value used of 0.3 kg N/(ha·a) is far below a safe cause-effect relationship is also illustrated by the circumstance that most experimental scientific studies on the effects of additional nitrogen inputs on vegetation work with nitrogen releases in steps of at least 5 to 10 kg N/(ha·a) (...).

39 (2.5) The view of the Higher Administrative Court that a reduction of the cut-off value is necessary in view of the special cases in which the cut-off value of 0.3 kg N/(ha·a) is equal to or even above the 3% de minimis threshold cannot be endorsed either. Even for habitat types with very low critical loads, purely theoretical concerns are not sufficient to justify the denial of approval for a project. The lack of ability to validate modelled inputs and the lack of distinctness from random variations in the background level does not change in these cases either.

40 (3) The CJEU judgment of 7 November 2018 - C-293/17 [ECLI:EU:C:2018:882], Raad von State - on the Dutch "offsetting model" for nitrogen assessment in Natura 2000 sites does not state that the cut-off criterion of 0.3 kg N/(ha·a) is incompatible with the requirements of EU law under article 6 (3) of the Habitats Directive. The proceedings concerned a different concept (see BVerwG, judgment of 27 November 2018 - 9 A 8.17 - (...)). The CJEU judgment was based on a Dutch national programmatic approach to tackling nitrogen deposition in Natura 2000 sites, which should at the same time enable the maintenance or development of economic activities by considering half of the projected reduction in nitrogen deposition as a "room for deposition" for new economic activities. In this context, an additional nitrogen load of 1 mol/ha·a is standardised as the threshold for an exception from the individual approval requirement.

41 This value, which is significantly lower than the cut-off value of the BASt Report 2013, but also compared to the cut-off value assumed by the Higher Administrative Court, is an arithmetical figure that can be explained by the design of the programme. It is based on the premise that, as a result of certain measures, nitrogen loads will decrease in the future to a certain calculated extent. The Dutch model represents a management programme which, as part of a programmatic global assessment, "coordinates" nitrogen inputs by determining, on the basis of a prior assessment, a certain amount of nitrogen that can be deposited in the protected site concerned. Within the framework of this model, the contribution to nitrogen deposition of the emission sources that the user inputs into the system is calculated, thus determining whether the project will have a significant effect on an Habitats Directive site (CJEU, judgment of 7 November 2018 - C-293/17 - para. 30 to 39 and 104, 112). With such a purely arithmetical model approach, which furthermore assumes a reduction of nitrogen inputs through site management measures, it is system-compatible to include even small contributions that can only be determined arithmetically. These values are not comparable with the cut-off values, which are recognised as the best scientific knowledge in the established jurisprudence of the Federal Administrative Court.

42 bb) The cut-off value of 0.05% of the respective critical load, or the equivalent of 4 eq assumed by the Higher Administrative Court for the additional loads from acidifying inputs in the context of the cumulative assessment, faces the same - above-mentioned - fundamental concerns under federal law. In this respect, too, the cut-off value is not to be modified in view of other plans or projects. It can therefore be left open whether the Higher Administrative Court - (...) made a procedural error (...) in concluding that the "isolated" cut-off value in the BASt Report 2013 was only to be set at 24 eq and that it also included a percentage for acidifying sulphur inputs (...).

43 c) The view of the Higher Administrative Court that all projects since a site was granted protection must be taken into account as a further additional load cannot be endorsed either.

44 According to the Federal Administrative Court's jurisprudence, the impacts of projects that have already been implemented or of previous uses that have been included in the determination of the current status are not to be included in the cumulative assessment, but are to be attributed to the initial load (Vorbelastung; see BVerwG, judgments of 15 July 2016 - 9 C 3.16 - (...) para. 55 and of 9 February 2017 - 7 A 2.15 - BVerwGE 158, 1 para. 220).

45 As a rule, the inclusion of completed projects into the initial load does not result in an impermissible reduction in the level of protection. Initial loads can worsen the conservation status to such an extent that only minor additional loads can be tolerated (BVerwG, judgment of 9 February 2017 - 7 A 2.15 - BVerwGE 158, 1 para. 220). However, merely taking into consideration the initial load does not always allow the identification and prevention of a gradual deterioration of a Habitats Directive site and the habitat types and protected species found there, especially in the event of repeated use of de minimis thresholds ("death by a thousand cuts", see Opinion of Advocate General Sharpston of 22 November 2012 - C-258/11 (ECLI:EU:C:2012:743) - para. 67 and Opinion of Advocate General Kokott of 25 July 2018 - C-293/17 and C-294/17 (ECLI:CU:C:2018:622) - para. 107).

46 Such a supplementary assessment may be necessary in particular where, as in the case of nitrogen, the impact of the project is only reflected with a certain time lag in the conservation status of the habitat types and protected species (...) and where there is no direct link between the assessment of the status of a site in the context of site management and the assessment of the effect in the context of project approval (...). However, the view of the Higher Administrative Court is not to be endorsed in so far as it considers it necessary to relate the assessment to the date on which the Habitats Directive sites were granted protection in December 2004 (see Commission Decision of 7 December 2004 - 2004/813/EC - OJ L 387 p. 1). Although this approach can be based on the precautionary principle of EU law, it does not take sufficient account of the principle of proportionality, which is also founded in EU law.

47 aa) The approach of the Higher Administrative Court is based, as far as eutrophying and acidifying nitrogen inputs are concerned, on the idea of a "quota regime" according to which the de minimis threshold can only be exhausted once after the time the Habitats Directive site is designated as a protected area, irrespective of the respective conservation status and the development of nitrogen inputs. In this respect, it can be argued that multiple use of the de minimis threshold may lead to effects that are in sum no longer negligible and thus significant. Insofar, the approach takes account of the precautionary principle in the context of habitat law. However, even multiple use of the de minimis threshold does not necessarily result in significant effects. The defined conservation objectives of the protected site are decisive for the assessment of whether a project, either individually or in combination with other projects, can significantly affect a Habitats Directive site. Pursuant to section 34 (2) BNatSchG and article 6 (3) second sentence of the Habitats Directive, a plan or project can only be approved if the competent national authorities, taking into account the best relevant scientific knowledge, have made certain that the plan or project, even in combination with other plans or projects, will not adversely affect the integrity of the site. As shown above, this is so when there is no reasonable doubt from a scientific point of view as to the absence of such adverse effects (CJEU, judgments of 7 September 2004 - C-127/02 - para. 54 et seqq. and of 26 April 2017 - C-142/16 - para. 33; BVerwG, judgment of 17 January 2007 - 9 A 20.05 - BVerwGE 128, 1 para. 62).

48 This assessment programme does not necessarily require a rigid quota regime of the 3% de minimis threshold, nor does it require the cumulative assessment to be related back to the time protection is granted. Section 34 (1) BNatSchG and article 6 (3) first sentence of the Habitats Directive require - as shown above - a link between the nitrogen input of a project and a significant effect on the conservation objectives of a Habitats Directive site. On this basis, it is not justified from the point of view of proportionality to indefinitely consider a de minimis threshold to be "exhausted" once it has been used. Such a view would lead to disproportionate restrictions in project approval and contradict the principle that realised projects are included in the initial load (background level) and are therefore adequately reflected in the assessment of the implications for the site. High initial nitrogen loads at a site, which in some cases are several times higher than the critical loads, cannot be reduced by means of habitat protection, but only by an effective clean air policy (see BVerwG, judgment of 14 April 2010 - 9 A 5.08 - BVerwGE 136, 291 para. 93). Measures with local effects, such as the removal of highly nitrogen-binding plants as part of site management, can also lead to an improvement in the nitrogen load (see BVerwG, judgment of 6 November 2012 - 9 A 17.11 - (...) para. 93; further examples of measures in the Opinion of Advocate General Kokott in Joined Cases C-293/17 and C-294/17, para. 65). Even after one or more projects approved and implemented after December 2004 have made use of the de minimis threshold of 3% of the critical load, the load situation does not necessarily consolidate or even worsen, but may improve due to certain measures with global and regional effects. If this is the case and the nitrogen load shows a (clear) decreasing trend, it would not be in line with the principle of proportionality if an additional input of up to 3% of the critical load which, according to the best available scientific knowledge, is insignificant, were to be considered impermissible from the outset. This applies even more, the longer the time that has elapsed since the initial use of the de minimis threshold and the more significantly the initial load has decreased.

49 There are no concerns under EU law regarding a repeated application of the de minimis threshold. The judgment of the European Court of Justice of 7 November 2018 - C-293/17 - (para. 104, 112) on the Dutch "offsetting model" for the assessment of nitrogen deposition at Habitats Directive sites shows that article 6 (3) of the Habitats Directive does not preclude consideration of the reduction of excessive nitrogen deposition in Habitats Directive sites in order to enable the maintenance or development of economic activities, provided that the competent authorities made sure that there is no reasonable scientific doubt as to the absence of adverse effects of each approved project on the integrity of the site concerned (see BVerwG, judgment of 27 November 2018 - 9 A 8.17 - para. 81; see also Opinion of Advocate General Kokott of 25 July 2018 - C-293/17 and C-294/17 - para. 79 et seq.).

50 bb) Irrespective of the fact that a rigid quota regime of minor inputs based on the year 2004 is not justified, the approach of the Higher Administrative Court gives cause for concerns because, if an examination of the development of the population of the species and the number of habitat types affected, the respective impact projections and, in particular, the use of the de minimis threshold is carried out going back to 2004, even the identification of the projects to be included will be associated with growing difficulties and uncertainties with increasing lapse of time. Moreover, the stocktakings as well as emission and immission projections contained in older approvals are becoming a less and less reliable and meaningful basis for a retrospective consideration of the development of the site and the protected species and habitat types. It is obvious that the studies underlying the older approvals are based on a site status that is comparable to the current status only to a limited extent (...). Even the determination of the initial status of the Habitats Directive sites will encounter considerable difficulties due to the often missing or incomplete database at the time protection is granted and due to the fact that the information in the registration documents is not primarily directed at the conservation status of the site, but rather at its eligibility for registration (...). In addition, on-site stocktakings represent only snapshots of fauna and flora (BVerwG, judgment of 9 July 2008 - 9 A 14.07 - BVerwGE 131, 274 para. 62) and the comparability of older approval documents with current findings is made more difficult by progress of knowledge in science and nature conservation as well as the progress regarding the setting of standards made since the granting of protection. For example, a rigid reference to the year 2004 meets with practical concerns regarding an almost arbitrary mix of methods, which is held to render the results incomparable (...).

51 cc) The intention to exclude effects on Habitats Directive sites through repeated use of the 3% de minimis threshold could, if necessary, be taken into account by taking recourse to the Federal Environment Agency's (Umweltbundesamt) data sets (UBA data sets). According to the current 2019 Nitrogen Guidelines for Roads (p. 35), the data sets have been recording the background level of nitrogen deposition comprehensively throughout Germany for several years. The input data used for this purpose relate to one year in each case and are combined to form multi-year averages in order to smooth out meteorological fluctuations over a longer period. In the data sets with a spatial resolution of 1 x 1 km, all emission sources and deposition influences relevant for the determination of the background level are taken into consideration. According to the Guidelines for Germany, they represent the best relevant scientific knowledge for the determination of the initial load (background level). It cannot successfully be argued against the use of the UBA data sets that such use inevitably leads to a consolidation and deterioration of the nitrogen load. As shown above, a repeated application of the 3% value can only be taken into consideration if a (clear) positive development of the initial load can be derived - for example from the UBA data sets. It will also have to be examined in this context whether gradual deterioration may occur due to project impacts that have not yet been reflected in these data sets (so-called corrected initial load). Whether the UBA data sets provide a sufficient basis for decision-making in this respect, and whether this applies equally to eutrophying and acidifying nitrogen inputs, is first and foremost a nature conservation issue and cannot be finally decided here.