Multilevel Governance of Urban Climate Adaptation in the European Union: An Overview

Abstract

Europe is warming faster than the global average, making climate change adaptation a central concern for urban policy and planning. This article develops and applies an analytical framework to assess the maturity of multilevel adaptation governance across European Union Member States as of 2025. Governance is operationalised through eight dimensions: (i) National Adaptation Strategies/Plans; (ii) Regional Adaptation Plans; (iii) Local Adaptation Plans; (iv) Sectoral Adaptation Plans; (v) integration in National Urban Policies; (vi) adaptive content in Long-Term Strategies; (vii) adaptation relevance in climate laws; and (viii) participation in the Covenant of Mayors. The results reveal pronounced heterogeneity: many Member States have up-to-date national strategies but display incomplete territorial diffusion, weak legal anchoring, or limited urban policy standards. By linking auditable rules to urban-facing instruments, this study offers a practical tool for benchmarking governance capacities, prioritising reforms, and tracking progress towards integrated, multilevel adaptation systems that support resilient urban development across the European Union.

1. Introduction

Climate risks in Europe are becoming more frequent, severe, and widespread. These trends generate uneven impacts across sectors and territories and intensify pressure on public institutions to organise effective responses [1,2,3]. The literature consistently shows that effective adaptation depends on the alignment of responsibilities, resources, and information across administrative levels and policy areas. Both fundamental and applied research on barriers, enabling conditions and multilevel coordination, particularly in the European context and corroborated by international evidence, reach similar conclusions [2,4,5,6,7,8,9,10,11]. Beyond the academic literature, practice-oriented city networks similarly stress the importance of vertical and horizontal coordination, including capacity building, information sharing, strategic guidance, and tailored support for local authorities [8,12]. Comparative analyses further confirm recurring role ambiguity, capacity gaps and coordination deficits, alongside corrective approaches and toolkits documented across OECD countries [10,11].

The EU policy framework provides direction and expectations. The EU Strategy on Adaptation to Climate Change and the European Climate Law set out expectations and requirements for systematic monitoring, cross-sectoral coherence, and long-term planning, thereby extending the strategic horizon of the European Green Deal [13,14,15]. Comparative studies and EU-level assessments suggest that these instruments have acted as agenda-setters and coordination devices, prompting initial waves of national strategies, standardising core components (risk assessments, cross-sectoral objectives, coordination bodies), and introducing monitoring and reporting expectations that encouraged periodic updates [13,14,15,16,17]. From a theoretical perspective, and in line with this policy evolution, the major analytical models for climate change adaptation explain how adaptation planning can proceed under conditions of uncertainty by integrating considerations of exposure, sensitivity, and adaptive capacity into decision-making [18,19]. Subsequent scholarship shows that effective implementation is closely linked to processes of transformational change and to governance dynamics operating across scales, involving governance interventions, the allocation of responsibilities, and network-based forms of coordination [20,21,22,23]. Comparative analyses between Member States reveal markedly different adaptation pathways and highly uneven policy uptake across territories [16,24,25,26]. At the same time, reviews of reported actions indicate that successful implementation relies on clear mandates, a purpose-fit institutional setup, and effective coordination mechanisms that function in practice rather than just in theory [17]. This distinction underpins the maturity-based approach adopted in this study. Shared evidence infrastructures—such as EEA reporting and Climate-ADAPT—alongside legal/policy databases support harmonised evidence gathering and comparative analysis [27,28,29,30,31]. Urban and planning scholarship complements this picture by detailing municipal instruments and ecosystem-based approaches, and examining their interactions with national frameworks [32,33]. In parallel, the Sendai Framework provides a complementary lens for risk reduction [34]. Beyond Europe, international assessments indicate that similar multilevel design levers—such as legal anchoring, periodic review, and information systems that connect national priorities to local delivery—recur in North America and Asia [35,36,37,38,39,40,41,42].

Against this backdrop, the paper examines how national adaptation governance arrangements across EU Member States differ in their ability to support coherent multilevel implementation and whether recurring patterns of governance maturity can be identified across heterogeneous institutional contexts. Rather than treating maturity as a proxy for policy performance or effectiveness, the study adopts a governance-oriented maturity lens to capture the degree of institutionalisation, coherence and multilevel integration of adaptation as formally codified in laws, strategies, and planning instruments.

The guiding research question is, therefore, how do national adaptation governance arrangements differ in their capacity to enable coherent multilevel implementation, and which maturity patterns emerge from these differences?

This article makes three main contributions. First, the paper conceptualises maturity as an analytical construct rooted in multilevel governance and adaptation scholarship, clarifying why specific institutional features—such as legal anchoring, territorial mandates, urban integration and monitoring cycles—are decisive for effective implementation under conditions of uncertainty [18,19,20,21,22,23]. Second, it operationalises this concept through a transparent, rule-based and documentary framework applied to all EU Member States, enabling systematic comparison under criteria aligned with EU expectations for accountability and periodic review [13,14,15,17]. Third, by identifying recurrent configurations of governance maturity rather than linear rankings, the framework moves beyond descriptive heterogeneity and provides a basis for differentiated and context-sensitive policy guidance on legislation, budgeting, and oversight across governance levels [13,14,15,17,43,44].

By positioning maturity as a diagnostic lens for multilevel governance capability—rather than as an outcome-based assessment—this article advances a conceptual adaptation of maturity thinking to the field of climate adaptation governance and links comparative diagnosis to reflections on practical pathways for capacity building. This article is structured as follows: Section 2 reviews the literature and develops the theoretical framework; Section 3 details the Materials and Methods; Section 4 presents the Results for the European Union Member States; Section 5 discusses the implications for policy design and capacity building; and Section 6 concludes, with key takeaways and immediate operational insights.

2. Literature Review and Theoretical Framework

Despite notable progress in European adaptation policy, systematic, maturity-based comparisons of how national frameworks support local delivery remain limited, even where national strategies are formally in place [16,17]. Addressing this gap requires operationalising a governance-oriented maturity lens for the European Union, using official documentary evidence to classify countries into archetypes based on the configuration of roles, coordination, monitoring, and municipal support [2,24]. The approach is anchored in contemporary EU expectations for resilience and accountability—the European Green Deal, the EU Adaptation Strategy and the European Climate Law—which emphasise demonstrable progress, multilevel coherence and review cycles [13,14,15]. This gap raises questions about how national adaptation governance arrangements differ in their capacity to enable coherent multilevel implementation, and whether recurring maturity patterns can be identified across heterogeneous institutional contexts. The theoretical starting point treats adaptation as a governance challenge that spans multiple levels and sectors: authority and problem-solving capacity are dispersed, so effectiveness hinges on coordination instruments, role clarity and institutional fit rather than on hierarchy alone [4,10,45,46].

In this study, maturity is employed as an analytical construct to assess the degree of institutionalisation, coherence, and multilevel integration of climate adaptation governance, rather than as a measure of policy performance or the effectiveness of outcomes. Maturity models originate in organisational and governance studies as diagnostic tools used to characterise the formalisation, coordination, and learning capacity of complex systems through observable institutional features, rather than through impact metrics [47,48]. This approach is particularly relevant for climate adaptation, where outcomes are uncertain, impacts are lagged, and the attribution of policy actions to risk reduction remains limited [2,18,19,43,44].

Within multilevel governance, a maturity lens captures how responsibilities, mandates, monitoring, and resources are structured and aligned across administrative levels, recognising that adaptation effectiveness depends on institutional fit and coordination rather than on hierarchical control alone [4,20,21,22,23]. Within this framework, maturity can be conceptualised as a governance capacity condition: the extent to which adaptation is legally anchored, territorially diffused and operationalised through urban and sectoral instruments within a coherent multilevel architecture. Crucially, this does not imply a single linear progression. Instead, drawing on multilevel governance theory, maturity is expected to manifest in recurring configurations shaped by administrative traditions and policy pathways [4,5,6,45,46]. The contribution of this paper, therefore, lies in adapting maturity thinking to multilevel climate adaptation governance, providing a theoretically grounded and documentary diagnostic lens for comparative analysis across Member States.

Foundational analyses of adaptation decision-making under uncertainty motivate iterative learning, proportionality and the transparent articulation of objectives, options and criteria, thereby avoiding technocratic framings detached from feasibility [2,18,19]. Comparative studies consistently document barriers—ambiguous mandates, fragmented responsibilities, weak vertical–horizontal coordination and capability constraints—and show that clear legal bases, reliable information infrastructures, and alignment across tiers consistently act as enabling conditions [7,10]. Evidence from urban governance research further demonstrates that role clarity and coordination tools are decisive for translating national ambitions into local action. At the same time, persistent gaps tend to open at the national–local interface [21,24]. Legal analyses of flood-risk management illustrate how fragmentation undermines coherence unless bridged by integrative instruments and explicit allocation of duties [11].

Within Europe, policy evolution has raised the bar for coherence and traceability. The Adaptation Strategy clarifies expectations for mainstreaming, monitoring and multilevel coordination, while the Climate Law codifies long-term objectives and review obligations [13,14]. However, comparative EU assessments still find convergence in formal strategies but divergence in institutional design, sectoral integration, and local delivery capacity [17]. A shared knowledge backbone—EEA reporting and the Climate-ADAPT platform—offers harmonised access to strategies, laws and indicators, thereby enabling cross-country comparison of governance conditions and reported adaptation actions [1,27]. Complementary databases such as Climate Change Laws of the World and Climate Watch add structured legal/policy metadata and target trajectories to triangulate documentary evidence [28,31]. Network standards, notably the Global Covenant of Mayors’ Playbook and Common Reporting Framework, define reporting fields that align municipal data with national backbones and EU expectations, thereby strengthening the comparability of local planning and monitoring [8,12]. At the city scale, the JRC’s analyses of SECAP practice and the wider urban policy literature document how guidance, finance and templates condition the quality and diffusion of plans [33,49].

A brief extra-European perspective clarifies how multilevel arrangements shape adaptation in federal and unitary settings. In North America, the IPCC AR6 assessment documents a dense but uneven patchwork of measures across federal, state/provincial and municipal levels, with significant progress in city and state planning but persistent coordination and capacity gaps that affect vertical coherence and monitoring under uncertainty [35]. Reviews of United States adaptation law and practice similarly indicate that much implementation occurs at sub-national scales, particularly in municipalities and counties. At the same time, federal frameworks and finance remain crucial for setting standards, gathering information and tracking progress [36,37]. Recent syntheses of local-level policies highlight advances alongside gaps in equity, metrics and enforceability, again pointing to the need for clearer role allocation and learning cycles across levels of government [38]. Across Asia, the IPCC AR6 regional chapter reports rapid growth in national strategies and subnational initiatives, yet emphasises the barriers of finance, capacity, and cross-scale coordination that hinder mainstreaming and monitoring [39]. OECD guidance on climate-risk management reinforces the case for transparent rulesets and periodic review—design features directly relevant to multilevel systems in the region [40]. World Bank diagnostics for South Asia echo these points, underlining the importance of place-based data architectures and institutional arrangements that link national priorities to local delivery and private co-investment [41]. Beyond these regional syntheses, comparative work on multilevel governance and sub-national finance in Asia and the Pacific (OECD/ADB) shows how intergovernmental mandates, fiscal frameworks and reporting systems condition the scalability of local plans—especially where national urban policy and planning codes provide binding hooks [42]. Country-focused analyses illustrate the mechanisms: in Japan, the Adaptation Act’s national-to-local architecture, including Local Climate Change Adaptation Centres, creates structured interfaces between knowledge infrastructures and municipal planning, while revealing heterogeneity in local capacities that must be addressed through standards and support [50,51]; in the Republic of Korea, studies of national adaptation planning and governance identify both the consolidation of cross-ministerial roles and the persistence of implementation barriers that impede iterative monitoring and territorial diffusion [52,53]. Region-wide, UNFCCC and multilateral assessments converge on the same levers—legal anchoring, periodic review, vertical coordination and robust information systems—while also documenting a substantial adaptation-finance gap that undermines local execution unless domestic public finance and MDB support are aligned with clear, auditable standards [54,55]. Taken together, these sources support the analytical stance adopted here: robust national backbones (law/strategy), explicit territorial mandates, and monitored local standards are complementary; without their alignment, voluntary local activity rarely aggregates into integrated multilevel systems [40,41,42].

On this conceptual basis, the framework used here draws on a set of recurring dimensions that the literature consistently associates with multilevel adaptation performance. These include the structuring of adaptation governance through roles and coordination mechanisms; the presence of barriers and enabling conditions shaping implementation; the diffusion of mandates and instruments across national, regional and local tiers; and the operation of monitoring and learning processes under conditions of uncertainty and lagged outcomes [4,7,24,33]. From these dimensions, the study derives a focused set of governance features that are both conceptually central and consistently documented in pan-European policy analysis. These relate to the existence and legal anchoring of national adaptation strategies and plans; the territorial diffusion of adaptation mandates to regional and local levels; the integration of adaptation into urban and sectoral policy frameworks; and the presence of monitoring, reporting and review mechanisms that support accountability and learning over time [13,14,16,17,43,44,56]. To operationalise these features comparatively—without entering into policy appraisal or outcome measurement—the study follows established standards in comparative policy analysis and qualitative content analysis: explicit constructs and descriptor-anchored levels, ordinal scales consistent with documentary evidence, transparent coding rules, and a documented audit trail [47,48,57,58,59,60].

On these foundations, the five archetypes used in this research—Lagging/Fragmented, Top-Down Centralised, Region-Driven, Local/Urban Bottom-Up and Integrated Multilevel—do not form a single linear ladder but instead recurring configurations anticipated by multilevel governance theory, in which the relative strength of the national core, territorial mandates, and urban–local implementation density varies [4,23,45,46]. The literature suggests, in particular, that: (i) strong legal mandates with review cycles and indicators institutionalise action, but without territorial obligations tend to remain top-down [13,14,43]; (ii) regional strength can bridge part of the gap, yet rarely delivers full integration without binding local coverage and shared risk methodologies [17,22]; (iii) local and urban implementation density increases when adaptation is embedded in planning instruments and supported by networks and common standards, but remains fragile if not linked to predictable finance and a coherent national legal core [8,12,33,49]; and (iv) integrated systems emerge when these pillars co-evolve and are reinforced by credible information infrastructures and monitoring, evaluation and learning processes, in line with European policy agendas and international frameworks such as Sendai and UNFCCC [1,13,34,44].

In sum, this theoretical framework explains why the governance features operationalised in the study are indicative of adaptation-governance maturity, and why the empirical results organise into five non-overlapping archetypes. These archetypes reflect the institutional mechanisms—legal anchoring, territorial diffusion, urban integration, and monitoring and learning—that the European and international literature consistently identifies as decisive for a state’s capacity to orchestrate coherent multilevel implementation [17,38,40,41,42,43,50,51,52,53,54,55,61,62].

3. Materials and Methods

This study employs a documentary, rule-based comparative design to analyse and classify multilevel adaptation governance across the Member States, with data coverage up to July 2025. The framework operationalises governance by linking (i) the national strategic and legal backbone, (ii) territorial mandates at regional and local levels, and (iii) urban-local diffusion and network mobilisation, and categorises Member States into five theoretically grounded archetypes (K1–K5) using deterministic, explicitly rule-based criteria. To improve transparency and support reader navigation, Figure 1 provides a schematic overview of the research framework. The flowchart summarises the sequential logic of the study, from document identification and coding to indicator construction, rule-based classification, and diagnostic interpretation.

The analytical approach is explicitly documentary, indicator-driven, and rule-based. The unit of analysis is not the linguistic or discursive content of policy documents, but their institutional and procedural features as formally codified in strategies, plans, and laws. Accordingly, the analysis does not involve discourse analysis or the interpretation of tone, rhetoric, framing, or policy attitude, nor does it employ machine-learning, probabilistic, or automated text classification techniques of any kind. The classification relies exclusively on observable and verifiable features of policy instruments—such as legal status, binding mandates, territorial coverage thresholds, review clauses, indicators/KPIs, and dedicated finance provisions—which are coded through descriptor-anchored decision rules. Each indicator level corresponds to explicit documentary evidence, defined ex ante in the codebook. This design prioritises transparency, auditability, and replicability: given the same publicly available documents and coding rules, the classification can be reproduced without model training, parameter tuning, or subjective interpretation of language.

The eight governance aspects retained for analysis reflect recurrent enabling conditions for multilevel adaptation identified in both the literature and European policy frameworks. Their selection is based on: (i) pan-European relevance, as established by EEA syntheses and EU policy scaffolding [1,2,3,13,14,15,16,17]; (ii) documentary traceability in national and sub-national strategies (NAS/NAP/RAP) and legal instruments, supported by reporting repositories and policy databases [12,27,28,29,30,31,32]; (iii) cross-country comparability, consistent with standards in comparative policy analysis and content analysis [43,44,56]; and (iv) sufficient data coverage to ensure reproducible coding and interpretable results [1,2,3,12,17,27,28,29,30,31,32,43,44,56].

The selection of indicators and threshold levels is theory-informed and policy-grounded rather than statistically optimised or data-fitted. Specifically, the indicators operationalise governance features that recur across adaptation and multilevel governance scholarship as enabling conditions for implementation, including legal anchoring, territorial diffusion of mandates, and monitored review cycles [2,4,10,16,17,34,43,44]. Their operational focus is also aligned with European Union policy expectations—most notably the EU Adaptation Strategy and the European Climate Law—which emphasise accountability, periodic review, and multilevel coherence [13,14]. Threshold levels (0–4) are descriptor-anchored, with each level corresponding to observable and documentable institutional features (e.g., binding mandates, indicators/KPIs, dedicated finance, mandatory review clauses), consistent with established principles for transparent indicator construction in comparative settings [43,44,56,57]. Evidence includes official national strategies and plans (NAS/NAP), climate framework laws, long-term strategies (LTS), regional adaptation plans (RAPs), local adaptation plans (LAPs—either stand-alone or embedded in SECAPs), sectoral adaptation plans (SAPs), national urban policies (NUPs), and monitoring, evaluation, and finance annexes where available. A comprehensive, country-by-country inventory of all analysed documents (including title, issuing authority, year, language, source/URL, and access date) is provided in Appendix A,

Table A1. National and sub-national adaptation-related documents analysed. Titles of official policy and legal documents are reported in their original language, as used in official national repositories. This choice facilitates traceability and replication of the documentary analysis.

Country	Document Type	Full Citation	Year (as Stated in File)	Language
Austria	NAS	Die Österreichische Strategie zur Anpassung an den Klimawandel. Teil 1—Kontext. Bundesministerium für Klimaschutz, Umwelt, Energie, Mobilität, Innovation und Technologie (BMK), Wien. Stand: 18. März 2024.	2024	German
	NAP	Die Österreichische Strategie zur Anpassung an den Klimawandel. Teil 2—Aktionsplan. Handlungsempfehlungen für die Umsetzung. BMK, Wien. Stand: 18. März 2024.	2024	German
	LTS	Langfriststrategie—Österreich. Periode bis 2050. Bundesministerium für Nachhaltigkeit und Tourismus, Wien, Dezember 2019.	2019	English
	LTS	Long-Term Strategy 2050—Austria. Period through to 2050. Federal Ministry for Sustainability and Tourism, Vienna, December 2019 (As of: 1 December 2020).	2019	German
Belgium	NAS	Belgian National Climate Change Adaptation Strategy. National Climate Commission, December 2010.	2010	English
	LTS	Stratégie à long terme de la Belgique. Document fédéral, Décembre 2019.	2019	French
	LTS	Belgische Langetermijnstrategie. Federaal document, December 2019.	2019	Dutch
	Federal adaptation measures (FAP)	Federale adaptatiemaatregelen 2023–2026—Naar een klimaatbestendige samenleving in 2050. Federaal niveau.	2023	Dutch
	RAP (Flanders)	Vlaams Klimaatadaptatieplan—Vlaanderen wapenen tegen de klimaatverandering. Vlaamse overheid, depotnummer D/2022/3241/266.	2022	Dutch
	RAP (Brussels)	Plan régional Air-Climat-Énergie (PRACE). Bruxelles Environnement, Juin 2016—Axe 7 « Adaptation ».	2016	French
Bulgaria	NAS & NAP (strategy + action plan)	Ministry of Environment and Water (Republic of Bulgaria). National Climate Change Adaptation Strategy and Action Plan. “Final Report …” (text references 2018 context and adoption pathway).	2018	English
	LTS (Long-Term Strategy)	Republic of Bulgaria. Long-term Strategy for Mobilizing Investments for Climate-friendly, Competitive and Secure Economy in the Republic of Bulgaria 2050. (document header indicates 2022).	2022	English
Croatia	Climate law	Republic of Croatia. Zakon o klimatskim promjenama i zaštiti ozonskog sloja (Official Gazette NN 127/2019, 27 December 2019).	2019	Croatian
	LTS (Long-Term Strategy/Low-carbon strategy)	Republika Hrvatska, Ministarstvo gospodarstva i održivog razvoja. Strategija niskougljičnog razvoja Republike Hrvatske do 2030. s pogledom na 2050. godinu (nacrt, Zagreb, lipanj 2021).	2021	Croatian
	NUP (National Urban Policy/Spatial planning strategy)	Republic of Croatia. Spatial Development Strategy of the Republic of Croatia (Zagreb, 2017).	2017	English
Cyprus	NAS (National Adaptation Strategy)	Republic of Cyprus, Department of Environment, Ministry of Agriculture, Rural Development and Environment. National Adaptation Strategy to Climate Change. Nicosia, 2017.	2017	English
	NAP (National Adaptation Plan)	Υπουργείο Γεωργίας, Αγροτικής Ανάπτυξης και Περιβάλλοντος. Σχέδιο Δράσης Προσαρμογής στην Κλιματική Αλλαγή (Παράρτημα ΙΙ του CYPADAPT). Λευκωσία, 2017.	2017	Greek
	LTS (Long-Term Low GHG Emission Development Strategy)	Department of Environment, Ministry of Agriculture, Rural Development and Environment. Cyprus’ Long-Term Low Greenhouse Gas Emission Development Strategy—2022 Update. Nicosia, September 2022.	2022	English
Czechia	NAS (National Adaptation Strategy)	Ministerstvo životního prostředí České republiky. Strategie přizpůsobení se změně klimatu v podmínkách České republiky—1. aktualizace pro období 2021–2030. Prague: Ministerstvo životního prostředí, 2021.	2021	Czech
	NAP (National Adaptation Plan)	Ministerstvo životního prostředí České republiky. Národní akční plán adaptace na změnu klimatu—1. aktualizace pro období 2021–2025. Prague: Ministerstvo životního prostředí, 2021.	2021	Czech
	NUP (National Urban Policy)	Ministerstvo pro místní rozvoj České republiky. Zásady urbánní politiky—Aktualizace 2017. Prague: MMR ČR, 2017.	2017	Czech
	LTS (Long-Term Strategy/Low-Emission Strategy)	Ministerstvo životního prostředí České republiky. Politika ochrany klimatu v České republice. Prague: Ministerstvo životního prostředí, 2017.	2017	Czech
Denmark	NAS (National Adaptation Strategy)	Regeringen. Strategi for tilpasning til klimaændringer i Danmark. Copenhagen: Energistyrelsen, March 2008. ISBN 978-87-7844-719-7.	2008	Danish
	NAP (National Adaptation Plan)	Regeringen, Miljøministeriet. Handlingsplan for klimatilpasning i Danmark. Copenhagen, 2012.	2012	Danish
	LTS (Long-Term Strategy/Low GHG Emission Development Strategy)	Ministry of Climate, Energy and Utilities. Denmark’s Long-Term Strategy for Reducing Greenhouse Gas Emissions. Copenhagen, 2020.	2020	English
	NUP (National Urban Policy/Spatial Planning Act)	Ministry of the Environment. The Planning Act in Denmark: Consolidated Act No. 813 of 21 June 2007. Copenhagen: Ministry of the Environment, 2007. ISBN 978-87-7279-795-3.	2007	English
Estonia	LTS	Government of the Republic of Estonia. General Principles of Climate Policy until 2050. English version (LTS submitted to the EU).	2017	English
	NAS/NAP	Ministry of the Environment (Republic of Estonia). Development Plan for Climate Change Adaptation until 2030 (with Implementation Plan 2017–2020).	2017	English
	NUP (National Urban Policy/Territorial Development Plan)	Republic of Estonia, Ministry of Finance. Territoriaalse tegevuskava 2030 atlas—Euroopa territoriaalse arengu kaardid. Tallinn: Ministry of Finance, 2021. Source: www.bmi.bund.de (European Territorial Development Atlas).	2021	Estonian
Finland	NAS	Ministry of Agriculture and Forestry. Finland’s National Strategy for Adaptation to Climate Change. Helsinki: Ministry of Agriculture and Forestry, 2005.	2005	English
	NAP	Ministry of Agriculture and Forestry. Finland’s National Climate Change Adaptation Plan 2022. Government Resolution 20 November 2014. Publications of the Ministry of Agriculture and Forestry 5b/2014. ISBN 978-952-453-862-6.	2014	English
	LTS	Työ-ja elinkeinoministeriö (Ministry of Economic Affairs and Employment). Suomen pitkän aikavälin strategia kasvihuonekaasujen vähentämiseksi. Helsinki, 2020.	2020	Finnish
	CRA (Climate Risk Assessment)	Finnish Meteorological Institute. National Climate Change Risk Assessment for Finland—2022 Update. Helsinki: FMI, 2022.	2022	English
	RAP	Centre for Economic Development, Transport and the Environment (ELY Centres). Regional Adaptation Plans for Climate Change—Finland’s Regional Climate Work Framework. Helsinki, 2022.	2022	English
	Climate Law	Parliament of Finland. Ilmastolaki (Climate Act) 423/2022. Helsinki: Finlex Data Bank. Available online: https://www.finlex.fi/fi/laki/alkup/2022/20220423 (accessed on 23 July 2025).	2022	Finnish
France	LTS	Ministère de la Transition Écologique. Stratégie Nationale Bas-Carbone (SNBC). Mars 2020.	2020	French
	NAS	ONERC—Observatoire national sur les effets du réchauffement climatique. Stratégie nationale d’adaptation au changement climatique. (La stratégie et ses éclairages sectoriels.)	2007	French
	NAP	Ministère de la Transition Écologique et Solidaire. Plan national d’adaptation au changement climatique 2018–2022 (PNACC-2).	2018	French
Germany	NAS	Federal Government of Germany. Deutsche Anpassungsstrategie an den Klimawandel (German Strategy for Adaptation to Climate Change). Adopted by the Federal Cabinet on 17 December 2008.	2008	German
	NAP	Federal Government of Germany. Aktionsplan Anpassung der Deutschen Anpassungsstrategie an den Klimawandel (Action Plan for the German Strategy for Adaptation to Climate Change). Adopted by the Federal Cabinet on 31 August 2011.	2011	German
	LTS	Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU). Langfristige Klimaschutzstrategie Deutschlands (Long-Term Climate Strategy of Germany). Submitted to the European Commission, Berlin.	2020	German
	Progress Report	Bundesministerium für Umwelt, Naturschutz und nukleare Sicherheit. Fortschrittsbericht zur Deutschen Anpassungsstrategie an den Klimawandel (Progress Report on the German Adaptation Strategy to Climate Change). Berlin: BMU, 2015.	2015	German
	NUP	Bundesministerium des Innern, für Bau und Heimat. Nationales Umsetzungsprogramm zur Territorialen Agenda 2030. Berlin: BMI, 2021.	2021	German
Greece	NAS	Ministry of Environment and Energy (MEEN). National Climate Change Adaptation Strategy (NAS) of Greece. Athens: MEEN, 2016.	2016	Greek
	LTS	Ministry of Environment and Energy. Long-Term Strategy for 2050 (LTS)—National Energy and Climate Plan. Athens, 2019.	2019	Greek
	NC/BR	Hellenic Republic, Ministry of Environment and Energy. 8th National Communication and 5th Biennial Report under the UNFCCC. Athens: MEEN, December 2022.	2022	English
	NUP	Ministry of Environment and Energy. National Implementation Programme for the Territorial Agenda 2030. Athens, 2021.	2021	Greek
Hungary	NAS	Ministry of Agriculture. Nemzeti Alkalmazkodási Stratégia (National Adaptation Strategy) 2018–2030, with Outlook to 2050. Budapest: Ministry of Agriculture, 2018.	2018	Hungarian
	LTS	Ministry for Innovation and Technology (ITM). Nemzeti Tiszta Fejlődési Stratégia 2020–2050 (National Clean Development Strategy 2020–2050). Budapest: ITM, 2021.	2021	Hungarian
Ireland	NAS/NAP	Department of the Environment, Climate and Communications. National Adaptation Framework—Planning for a Climate Resilient Ireland. Dublin: DECC, 2024.	2024	English
	LTS	Government of Ireland. Ireland’s Long-Term Strategy on Greenhouse Gas Emissions Reduction. Dublin: DECC, 2024.	2024	English
	Climate Law	Government of Ireland. Climate Action and Low Carbon Development Act 2015 (No. 46 of 2015); Amendment Act 2021 (No. 32 of 2021). Dublin: Government Publications.	2015/2021	English
	NUP	Government of Ireland, Department of Housing, Local Government and Heritage. National Planning Framework—First Revision. Dublin: DHLGH, April 2025.	2025	English
Italy	NAS/NAP	Ministero dell’Ambiente e della Tutela del Territorio e del Mare (MATTM). Strategia Nazionale di Adattamento ai Cambiamenti Climatici (SNAC). Rome: MATTM, 2015.	2015	Italian
	LTS	Ministero dell’Ambiente e della Tutela del Territorio e del Mare, Ministero dello Sviluppo Economico, Ministero delle Infrastrutture e dei Trasporti, Ministero delle Politiche Agricole, Alimentari e Forestali. Strategia Italiana di Lungo Termine sulla Riduzione delle Emissioni dei Gas a Effetto Serra. Rome: Governo Italiano, 2021.	2021	Italian
	National Report (NC/BR)	Government of Italy. Seventh National Communication and Third Biennial Report under the UNFCCC. Rome: Ministry for the Environment, Land and Sea, 2021.	2021	English
	Technical Report	SNPA—Sistema Nazionale per la Protezione dell’Ambiente. Rapporto sugli Indicatori di Impatto dei Cambiamenti Climatici—Edizione 2021 (Report SNPA 21/2021). Rome: ISPRA, 2021.	2021	Italian
Latvia	NAS/NAP	Ministry of Environmental Protection and Regional Development (VARAM). Latvian National Plan for Adaptation to Climate Change until 2030. Riga: VARAM, 2019.	2019	English
	LTS	Vides aizsardzības un reģionālās attīstības ministrija (VARAM). Latvijas stratēģija klimatneitralitātes sasniegšanai līdz 2050. gadam (Latvia’s Strategy for Achieving Climate Neutrality by 2050). Riga: Ministry of Environmental Protection and Regional Development, 2019.	2019	Latvian
	NUP	Ministry of Environmental Protection and Regional Development (VARAM). National Development Plan of Latvia for 2021–2027. Riga: VARAM, 2020.	2020	English
Lithuania	NAS/NAP	Ministry of Environment of the Republic of Lithuania. National Strategy for Climate Change Management Policy (Nacionalinė klimato kaitos valdymo politikos strategija). Vilnius: Ministry of Environment, 2012 (updated 2019).	2019	Lithuanian
	NECP	Government of the Republic of Lithuania. National Energy and Climate Action Plan for 2021–2030. Vilnius: Ministry of Energy, 2019.	2019	English
	LTS	Lietuvos Respublikos Energetikos Ministerija. Lietuvos ilgos trukmės strategija iki 2050 m. dėl mažai anglies dioksido išskiriančios ekonomikos (Lithuania’s Long-Term Strategy for a Low-Carbon Economy until 2050). Vilnius: Ministry of Energy, 2021.	2021	Lithuanian
	NUP	Government of the Republic of Lithuania. National Progress Plan for 2021–2030 (Nacionalinis pažangos planas). Vilnius: Government of Lithuania, 2020.	2020	Lithuanian
Luxembourg	NAS/NAP	Ministère de l’Environnement, du Climat et du Développement durable. Stratégie et plan d’action pour l’adaptation aux effets du changement climatique au Luxembourg 2018–2023. Luxembourg: Gouvernement du Grand-Duché de Luxembourg, 2018.	2018	French
	LTS	Gouvernement du Grand-Duché de Luxembourg. Stratégie nationale à long terme en matière d’action climat « Vers la neutralité climatique en 2050 ». Octobre 2021.	2021	French
	Climate Law	Journal officiel du Grand-Duché de Luxembourg. Loi du 15 décembre 2020 relative au climat (modifiant la loi modifiée du 31 mai 1999 instituant un fonds pour la protection de l’environnement), A994, Luxembourg: Gouvernement du Grand-Duché de Luxembourg, 2020.	2020	French
Malta	LTS	Government of Malta. Long-Term Strategy for Malta’s Climate Neutrality by 2050, Valletta: Ministry for the Environment, Energy and Enterprise, 2023.	2023	English
	Climate Law	Government of Malta. Climate Action Act (CAP. 543), Laws of Malta, Ministry for the Environment, Sustainable Development and Climate Change, Valletta, 2015.	2015	English
	National Spatial/Planning Strategy	Government of Malta. Strategic Plan for the Environment and Development (SPED), Planning Authority, Valletta, 2015.	2015	English
Netherlands	LTS	Government of the Netherlands. Langetermijnstrategie Klimaat (Long-Term Strategy). The Hague: Government of the Netherlands, 2019.	2019	Dutch
	Other (Adaptation Communication)	Government of the Netherlands, Ministry of Infrastructure and Water Management. Adaptation Communication of the Netherlands. The Hague, 2021. Submission date: October 2021.	2021	English
Poland	NAS/NAP	Ministry of the Environment of the Republic of Poland. Strategic Adaptation Plan for Sectors and Areas Vulnerable to Climate Change in Poland until 2020, with a perspective until 2030 (SPA 2020). Warsaw: Ministry of the Environment, 2013.	2013	Polish/English (summary available)
Portugal	NAS/NAP	Government of Portugal. Resolução do Conselho de Ministros n.º 56/2015—Quadro Estratégico para a Política Climática (QEPiC), Programa Nacional para as Alterações Climáticas 2020/2030 (PNAC 2020/2030) e Estratégia Nacional de Adaptação às Alterações Climáticas (ENAAC 2020). Diário da República, 1.ª série, n.º 147, 30 July 2015.	2015	Portuguese
	LTS	Government of Portugal, Ministry of the Environment and Energy Transition. Roadmap for Carbon Neutrality 2050 (RNC2050): Long-Term Strategy for Carbon Neutrality of the Portuguese Economy by 2050. Lisbon, 2019.	2019	English
	Climate Law	Government of Portugal. Lei de Bases do Clima (Climate Framework Law). Lei n.º 98/2021, Diário da República, 1.ª série, n.º 239, 10 December 2021.	2021	Portuguese
	NUP	Government of Portugal, Direção-Geral do Território. Programa Nacional da Política de Ordenamento do Território (PNPOT). Lisbon, 2019.	2019	Portuguese
Romania	LTS	Long Term Strategy of Romania. Prepared for the Ministry of Energy and the Ministry of Environment, Waters and Forests by PwC; current version elaborated 23 April 2023 and presented to CISC 24 April 2023 (public consultation launched 18 April 2023).	2023	English
Slovakia	NUP	The Urban Development Policy of the Slovak Republic by 2030 (short version). Ministry of Transport and Construction; Government Resolution no. 5/2018 of 10 January 2018.	2018	English
	LTS	Low-Carbon Development Strategy of the Slovak Republic until 2030 with a View to 2050. Ministry of Environment of the Slovak Republic (draft, November 2019).	2019	English
	NAP	Akčný plán pre implementáciu Stratégie adaptácie SR na zmenu klímy. Ministerstvo životného prostredia SR, August 2021.	2021	Slovak
	LTS	Resolucija o dolgoročni podnebni strategiji Slovenije do leta 2050 (ReDPS50). Državni zbor, sprejeta 13. julija 2021.	2021	Slovenian
	NAS	Strateški okvir prilagajanja podnebnim spremembam. Vlada Republike Slovenije, December 2016.	2016	English
	NUP	Spatial Management Policy of the Republic of Slovenia. Government of the Republic of Slovenia, adopted December 20, 2001.	2001	English
Spain	Climate Law	Government of Spain. Law 7/2021, of 20 May, on Climate Change and Energy Transition.	2021	Spanish
	NAS	Ministerio para la Transición Ecológica y el Reto Demográfico (MITECO). Plan Nacional de Adaptación al Cambio Climático 2021–2030 (PNACC 2021–2030).	2021	Spanish
	NAP	Ministerio para la Transición Ecológica y el Reto Demográfico (MITECO). Climate Change Adaptation: Work Programme 2021–2025 (Programa de Trabajo 2021–2025 del PNACC).	2021	Spanish
	SAP	ADIF—Administrador de Infraestructuras Ferroviarias. Climate Change Plan 2018–2030 (Plan de Cambio Climático 2018–2030).	2018	Spanish
	SAP	Ministerio para la Transición Ecológica y el Reto Demográfico (MITECO). Adaptation Strategy for the Spanish Coast (Estrategia de Adaptación de la Costa al Cambio Climático).	NA	Spanish
	SAP	Ministerio para la Transición Ecológica y el Reto Demográfico (MITECO). Strategic Guidelines on Water and Climate Change (Directrices Estratégicas en materia de agua y cambio climático).	NA	Spanish
Sweden	NAS	Nationell strategi för klimatanpassning (Prop. 2017/18:163). Government of Sweden.	2018	Swedish
	LTS	Sveriges långsiktiga strategi för minskning av växthusgasutsläppen. Ministry of the Environment.	2019	Swedish
	NUP	Voluntary National Review—New Urban Agenda (Sweden). Ministry of Finance; Boverket.	2021	Swedish
	SAP	Handlingsplan för klimatanpassning 2022–2025. Socialstyrelsen.	2022	Swedish
	RAP	Regional plan för klimatanpassning i Dalarna (Rapport 2021:09). Länsstyrelsen Dalarnas län.	2021	Swedish
	RAP	Regional handlingsplan för klimatanpassning i Västernorrlands län (Rapport 2018:01). Länsstyrelsen Västernorrland.	2018	Swedish
	RAP	Anpassning till ett förändrat klimat—Blekinges regionala handlingsplan (2014:12). Länsstyrelsen Blekinge län.	2014	Swedish
	RAP	Regional handlingsplan för klimatanpassning i Gotlands län 2018–2020 (uppdaterad 2019-04-04). Länsstyrelsen Gotlands län.	2018 (upd. 2019)	Swedish
	RAP	Regional handlingsplan för klimatanpassning i Gävleborgs län (2014:11). Länsstyrelsen Gävleborg.	2014	Swedish

Note: Document types follow the taxonomy used in the Methods (NAS/NAP, RAP, LAP/SECAP, SAP, LTS, NUP, Climate Law, Other), and sources are limited to official government publications or the institutional platforms listed above [1,3,25,26,27,28,29,30,36,38,43,44,45,49].

. Document identification and cross-checks draw primarily on official repositories, notably EEA reporting and Climate-ADAPT [1,3,17,27], supplemented by Climate Change Laws of the World, the Covenant of Mayors (CoM) platform, the Urban Policy Platform, and Climate Watch [28,29,30,31,32].

Where inconsistencies arise between repositories and national sources, the original national legal or policy text takes precedence, following established standards on source hierarchy and documentary traceability (laws > decrees > plans > portals) [57]. Instruments that have been formally adopted but are not yet in force are recorded as adopted, while draft instruments under consultation and not yet adopted are excluded. For the Covenant of Mayors indicator, population coverage is calculated using a fixed cut-off date of 30 June 2025, and SECAPs are considered only where adaptation measures are included in the adopted plan [12,30,49,59].

Non-English policy and legal documents were translated using DeepL (2025 version) solely to facilitate screening and coding. No formal intercoder reliability testing was conducted on translated passages, as coding decisions were rule-based and descriptor-anchored. To minimise translation-related bias, three safeguards were applied: (i) systematic precedence of the original source-language legal or policy text whenever discrepancies were identified; (ii) targeted manual verification of decision-relevant clauses (e.g., legal mandates, review cycles, finance provisions, indicator requirements), with explicit attention to legal modality (shall/must versus may/should); and (iii) conservative, descriptor-anchored coding rules based on explicit, documentable features (e.g., the existence of a statutory review clause or a dedicated budget line) rather than nuanced linguistic interpretation. Where ambiguity persisted after re-examining the source-language text, the indicator was coded at the lower descriptor level, consistent with the framework’s non-interpretive, feature-based nature. A detailed codebook specifies inclusion and exclusion criteria, extraction units (strategy clauses, legal provisions, programme measures), and descriptor anchors, with illustrative coded excerpts; the full document inventory is reported in Appendix A, Table A1.

Eight governance dimensions (B–I) are coded on an ordinal scale from 0 to 4 using indicator-specific documentary anchors. These anchors are summarised in

Table 1. Base governance indicators used in the analysis.

Code	Indicator	Conceptual Focus	Scale
B	National Adaptation Strategy/Plan (NAS/NAP)	Robustness and update cycle of national adaptation strategies and plans	0–4
C	Regional Adaptation Plans (RAPs)	Institutionalisation and coverage of regional adaptation planning	0–4
D	Local Adaptation Plans (LAPs)	Diffusion and mandate of local adaptation planning	0–4
E	Sectoral Adaptation Plans (SAPs)	Sectoral embedding of adaptation across key policy domains	0–4
F	Adaptive content in Long-Term Strategy (LTS)	Integration of adaptation into long-term national strategies	0–4
G	Integration in National Urban Policy (NUP)	Integration of adaptation into national urban policy frameworks	0–4
H	Adaptation Relevance in Climate Law	Legal anchoring of adaptation in framework climate legislation	0–4
I	Covenant of Mayors adhesion (CoM) %	Population coverage of municipalities with adopted SECAPs including adaptation	0–4

and fully specified in Appendix B.1 and align with international guidance on adaptation monitoring and learning [34,43,44] as well as European Union policy expectations regarding accountability and periodic review [13,14]. The indicators capture: (B) robustness of NAS/NAPs; (C) regional adaptation plans (RAPs); (D) local adaptation plans (LAPs); (E) sectoral adaptation plans (SAPs); (F) integration of adaptation in long-term strategies (LTS); (G) integration in national urban policy (NUP); (H) legal anchoring of adaptation in climate framework laws; and (I) Covenant of Mayors population coverage. Ordinal coding supports precise, evidence-traceable classification and explicitly excludes composite scoring or compensatory aggregation. Where repository summaries diverge from national texts, the latter prevail [1,3,27,28,29,57]. To address potential sensitivity to threshold choices, the stability of country classification under alternative plausible assumptions is assessed through a set of robustness checks reported in Section 4.3 and Appendix C, including moderate threshold shifts and stress tests applied to the Multilevel Integration Index. These checks show invariance of cluster membership across the tested scenarios [43,56].

The eight base governance indicators (see Table 1) operationalise institutional features that recur across adaptation research as preconditions for implementation, learning, and vertical coherence, rather than aiming for exhaustive coverage of policy. Specifically, the indicators reflect dimensions repeatedly highlighted in the literature, including: national strategic and legal anchoring of adaptation responsibilities [13,14,16,17]; the diffusion and institutionalisation of adaptation mandates across regional and local tiers [4,22,23,32]; sectoral embedding of adaptation in key policy domains [7,10,20,22]; and the integration of adaptation into urban policy frameworks and transnational municipal networks as vehicles for local uptake and horizontal diffusion [8,9,12,24,49,61]. Together, these indicators translate well-established theoretical constructs into observable, documentable features suitable for transparent and replicable comparative analysis.

Coverage of indicator I is classified on an ordinal scale from 0 to 4, consistent with JRC coverage tiers and the GCoM reporting framework [12,49,59]: 0 < 1.00% (Very Low); 1 = 1.00–4.99% (Low); 2 = 5.00–9.99% (Medium); 3 = 10.00–14.99% (High); 4 ≥ 15.00% (Very High) (see

Table 2. CoM coverage in municipalities with an adopted SECAP addressing both mitigation and adaptation (cut-off 30 June 2025): scoring scheme for indicator I.

Coverage (%)	Score (I)	Label
≥15.00	4	Very High
10.00–14.99	3	High
5.00–9.99	2	Medium
1.00–4.99	1	Low
<1.00	0	Very Low

).

Derived indices (J–P) are calculated from the base indicators (B–I) to synthesise vertical and horizontal coordination patterns across governance levels.

Table 3. Derived indices summarising multilevel governance patterns.

Code	Index	Conceptual Meaning	Role in Analysis
J	National Core Strength	Strength of the national strategic–legal backbone	Used in clustering rules
K	Territorial Mandate	Degree of institutionalisation of regional and local mandates	Used in clustering rules
L	Local Synergy	Density of the local–urban implementation ecosystem	Used in clustering rules
M	Advanced Breadth	Share of governance pillars at advanced levels	Diagnostic (non-compensatory)
N	Multilevel Integration	Balance across national, territorial and local tiers	Key clustering threshold
O	Top-down differential	Relative strength of national vs. local tier	Diagnostic only
P	Bottom-up differential	Relative strength of local vs. national tier	Diagnostic only

provides a concise conceptual overview of these indices and their analytical role, while detailed definitions and formulae are reported in Appendix B.1. The indices operationalise the deterministic classification rules used to assign Member States to the five governance archetypes (K1–K5) (see Section 3.1). Two indices (O and P) function as directional diagnostics, capturing centre-heavy versus local-heavy configurations, and are included solely for interpretive transparency; they do not influence class assignment. As an additional sensitivity probe, the Multilevel Integration Index (N) was recalculated under a zero-indicator condition, subtracting 0.25 whenever any base indicator equals zero. Supplementary visual summaries, archetype listings, and level distributions are provided in Appendix B.

The construction of the derived indices (Table 3) is theory-informed and grounded in the comparative literature on multilevel climate adaptation governance and in established principles for governance-oriented maturity models. Rather than introducing additional variables, the indices synthesise core institutional domains that recur in the literature as enabling conditions for coherent multilevel implementation. Specifically, prior research consistently identifies: (i) a national strategic–legal backbone that anchors responsibilities, accountability mechanisms and periodic review cycles [13,14,16,17,43]; (ii) the territorial diffusion of mandates through statutory regional and local planning duties that translate national commitments into binding or widely institutionalised sub-national action [4,22,23,32]; and (iii) local–urban implementation ecosystems supported by municipal planning instruments, integration in national urban policy frameworks and network mobilisation, which enable operational uptake and horizontal diffusion [8,9,12,24,32,49,61].

The derived indices (J–N) operationalise these domains by aggregating descriptor-anchored base indicators into diagnostic measures of institutionalisation, coherence and multilevel integration. Consistent with maturity-model design principles, the indices prioritise structural consolidation and balance across governance levels over performance measurement or outcome ranking and are implemented through a non-compensatory logic to avoid offsetting deficits in one domain with strengths in another [43,44,47,48,56].

3.1. Rationale for the Five Clusters and Rule Order

The five archetypes reflect recurrent multilevel governance configurations anticipated by the theory [9,45,46]. Their empirical recurrence is documented in comparative studies and EU-level comparative assessments [4,16,17,24]. Each class is designed to be diagnostic (clear signal), non-compensatory (no single pillar offsets deficits elsewhere), and replicable (descriptor-anchored thresholds). The rule order reflects increasing degrees of vertical balance and legal–strategic consolidation, consistent with European Union expectations on periodic review and accountability [13,14,16,43,56].

• K1—Lagging/Fragmented. Several foundational pillars are low; neither regional nor local tiers are consolidated, and the national core is not steering multilevel diffusion. Rule: count {B–I ≤ 1} ≥ 4, M ≤ 0.25, J < 2.5, C ≤ 2, D ≤ 2, L ≤ 2, and not higher classes.
• K2—Top-Down Centralised. Robust national strategic–legal backbone with incomplete territorial diffusion and local collaboration; not yet integrated. Residual rule (applied after excluding K5/K3/K4/K1): J ≥ 2.5 and N < 3.
• K3—Region-Driven. Territorial (regional) leadership precedes full national consolidation; the local tier is present but not yet scaled. Rule: C ≥ 3, D ≥ 1, K ≥ 2, J < 3, and not K5.
• K4—Local/Urban Bottom-Up. City-led mobilisation and urban policy instruments lead, while the regional tier remains weak and the national core is moderate. Rule: L > 2 driven by at least one advanced local lever (D ≥ 3 or G ≥ 2 or I ≥ 2), C ≤ 2, J < 3, and not K5/K3.
• K5—Integrated (Multilevel). Balanced, mature multilevel architecture in which a strong national core is coupled with mandatory or widely institutionalised sub-national layers and active local/urban integration. Rule: (i) vertical balance N ≥ 2.5; (ii) advanced local tier D ≥ 3 and urban integration G ≥ 2; (iii) regional presence C ≥ 2; (iv) advanced climate law H ≥ 3; (v) functional NAS/NAP B ≥ 2.

Descriptor-anchored cut-offs (e.g., D ≥ 3 = “mandatory/reporting with ≥50% population and ≤6-year update”) tie each rule to observable features in laws and plans. Balance thresholds on N (≥2.5) prevent compensation across tiers, while exclusion predicates (e.g., “not K5/K3”) avoid ambiguous assignments. Sensitivity checks for threshold choices and alternative partitions are documented in Appendix B; visual diagnostics (heatmaps and colour-coded membership tables) are provided for transparency and remain descriptive and subordinate to the rules [12,43,49,56]. For transparent replication, the rule-based cluster assignment can be implemented as a single nested Excel formula (without macros) in en-US and it-IT locales. The expressions directly operationalise the deterministic rules and do not imply any statistical modelling. For readability, the complete formulas are provided in Appendix B.2.

3.2. Coding and Documentary Verification

The complete document inventory analysed in this study—organised by Member State and including titles, issuing authorities, years, languages, sources/URLs and access dates—is reported in Appendix A, Table A1. Where discrepancies emerged between repository summaries and national sources, the official national legal or policy text took precedence (hierarchy: laws > decrees > plans > portals), in line with established standards in documentary and content analysis [57].

Non-English policy and legal texts were translated using DeepL (2025) solely to support screening and coding. To minimise translation-related bias, three safeguards were applied: (i) priority was given to the original source-language text whenever discrepancies were identified; (ii) decision-relevant clauses (e.g., mandates, review cycles, finance provisions and indicators) were manually cross-checked with particular attention to legal modality (e.g., shall/must versus may/should); and (iii) conservative, descriptor-anchored coding rules were applied, defaulting to the lower level in cases of unresolved ambiguity [43,44,56,57].

The analytical scope is explicitly documentary and rule-based. The analysis does not involve discourse analysis or the interpretation of tone, rhetoric, framing, or policy attitude, and does not employ machine-learning, probabilistic, or automated text-classification techniques. Classification relies exclusively on observable and verifiable institutional features—such as legal status, binding mandates, territorial coverage thresholds, review clauses, indicators/KPIs, and finance provisions—coded through descriptor-anchored decision rules. This design prioritises transparency, auditability, and replicability: given the same publicly available documents and coding rules, the classification can be reproduced without model training, parameter tuning, or subjective interpretation of language.

Coverage rules were applied consistently across countries: indicator C was coded at the highest binding regional tier in mixed arrangements; indicator D was derived from national inventories and/or statutory mandates with explicit population thresholds; an indicator I (Covenant of Mayors) was calculated as the proportion of national residents living in municipalities with an adopted SECAP including adaptation, as of 30 June 2025 (numerator: Covenant of Mayors platform; denominator: Eurostat 2024 provisional population [52]), classified according to Table 2.

Given the ordinal (0–4) nature of the base indicators (B–I), results are reported as level distributions and presence/status tables rather than as cardinal measures. Derived indices (J–P) support the deterministic classification rules only and are not interpreted as continuous scores. The full Excel-based implementation of the coding rules is provided in Appendix B.2, while sensitivity checks and robustness tests are documented in Appendix C.

3.3. Methodological Limitations

This section outlines the main methodological limitations associated with the framework’s documentary, indicator-driven, and rule-based design. They concern issues of measurement, coding, comparability and sensitivity, rather than substantive interpretations of national adaptation performance. Interpretative and policy-related limits, as well as the implications of these methodological constraints for the reading of the results, are addressed explicitly in Section 5.

First, compressing heterogeneous national evidence into an ordinal 0–4 scale prioritises cross-Member-State comparability over within-level institutional nuance. The resulting summaries should therefore be read as diagnostic signposts of governance configuration rather than as precise measurements. To mitigate spurious compensation across pillars, the framework applies non-compensatory rules and a sensitivity adjustment to the multilevel integration index N (subtracting 0.25 whenever any base indicator equals 0); these design choices reduce, but do not fully eliminate, threshold effects [43,56].

Second, the reliance on machine-assisted translation for non-English documents (DeepL, 2025) introduces residual uncertainty, particularly regarding legal nuance and institutional terminology. This risk was mitigated through source-text verification, targeted manual checks of decision-relevant clauses, and conservative descriptor-anchored coding. Importantly, sensitivity and robustness checks indicate that plausible translation noise does not alter cluster membership or comparative patterns [43,56].

Third, the framework prioritises documentable institutional features to maximise transparency and auditability. As a result, it does not directly assess implementation quality or the effectiveness of outcomes. For example, the Covenant of Mayors indicator captures the extent of SECAP adoption, including adaptation, rather than policy impacts [12,49,59]. Finally, timing and availability constraints may lead to conservative coding, in which very recent sub-national instruments are not yet reflected in European repositories or national portals [27,28,29,30,31,32].

These limitations are intrinsic to the documentary, rule-based nature of the framework and should be interpreted as methodological constraints. Their broader interpretative implications—particularly regarding how institutional configurations shape implementation capacity—are addressed explicitly in the Discussion (Section 5). To support transparency and replicability, the document inventory, coding logic, illustrative examples, and replication rules are documented in the Appendices, consistent with international guidance on adaptation monitoring, learning, and NAP good practices [44,62].

4. Results

4.1. Aggregate Patterns and Descriptive Statistics

The comparison of the eight governance dimensions across the European Member States (reference date: July 2025) indicates a marked asymmetry between a relatively consolidated national strategic–legal backbone and much weaker regional and local layers. This pattern is visible in the complete numerical distributions and heatmaps (Appendix B.3, Appendix B.4 and Appendix B.5), which make explicit the heterogeneity otherwise concealed by the ordinal scales.

At the national level, adaptation strategies and plans (B, NAS/NAP) cluster towards the upper end of the spectrum, with a median of 3.0 (IQR = 1.0): almost half of the Member States reach levels 3–4 (44.4%), while none score 0 and only 14.8% remain at level 1. By contrast, regional adaptation plans (C, RAPs) and local adaptation plans (D, LAPs) are concentrated in the lower categories. RAPs display a median of 1.0 (IQR = 2.0), with 59.2% of countries at levels 0–1 and no case at level 4, indicating that regional mandates are either absent or only partially developed. LAPs also have a median of 1.0 (IQR = 1.0), and two-thirds of the Member States (66.7%) are at level 1, with only 22.2% reaching levels 3–4. Sectoral adaptation plans (E, SAPs) are even more weakly developed: the distribution is heavily skewed towards level 1 (70.4%), the median is 1.0 (IQR = 1.0), and very few countries have either no sectoral planning (22.2% at level 0) or fully institutionalised sectoral frameworks (3.7% at level 4).

The adaptive components of long-term strategies (F, LTS) and the integration of adaptation into national urban policies (G, NUP) occupy a more intermediate position. Both record a median of 2.0 (IQR = 1.0), with a majority of Member States distributed between levels 1 and 2 (89.0% for F and 77.7% for G), and no country reaching level 4, suggesting that long-term and urban policy integration is present but rarely fully mature. Climate laws with explicit adaptation provisions (H) reveal a more polarised picture: the median is 1.0 (IQR = 2.0), with 22.2% of Member States lacking any specific climate law (level 0) and 37.0% reaching levels 3–4, pointing to significant divergence in the legal consolidation of adaptation. Local engagement through the Covenant of Mayors (I) remains modest overall: the median is 1.0 (IQR = 1.0), with 74.0% of Member States at levels 0–1 and only 7.4% at levels 3–4, indicating limited national coverage of CoM signatories with adaptation commitments in many countries.

Simple arithmetic summaries, used as internal diagnostics, reinforce this structurally unbalanced configuration. The national core index (J, NAT_CORE) has a median of 1.70 (IQR = 1.00), slightly higher than the territorial mandate index (K, TERR_MAN; median = 1.50, IQR = 1.00) and the local synergy index (L; median = 1.30, IQR = 1.00). Advanced governance scores (M, ADV_SCORE—share of pillars at level ≥ 3/8) are generally low, with a median of 0.10 (IQR = 0.25), indicating that cases combining high scores across multiple pillars are infrequent. The multilevel integration index (N, MULTI_INT) clusters around intermediate values (median = 1.70, IQR = 0.73), consistent with partial but incomplete alignment between national, regional, and local tiers. Directional differentials are modest but systematic: the median for O is +0.30, while P is −0.30, reflecting a slight but widespread tendency for national instruments to be more developed than their subnational counterparts.

The co-occurrence patterns visible in the heatmaps and distributional diagnostics (Appendix B.5) help qualify these aggregate tendencies. First, the heatmap reveals that, while some Member States combine relatively advanced national strategies (B at level 3) with strong climate-law provisions (H ≥ 3), a majority of this group still display low values for climate laws (H = 0–1), pointing to a recurrent strategic–legal asymmetry within the national layer. Second, combinations in which both regional and local mandates are weak (C = 1 and D = 1) are common, indicating a recurrent pattern in which subnational tiers remain underdeveloped relative to the national layer; within these configurations, the rule-based classification distinguishes “regional-first” (K3) and “local-first” (K4) trajectories. Third, higher local synergy (L), primarily driven by stronger LAPs (D), more systematic NUP integration (G), and broader Covenant of Mayors participation (I), tends to co-occur with the more synergistic configurations (K4–K5). These combine stronger, more balanced territorial mandates with more even centre–local dynamics (O closer to zero, with K4 slightly bottom-heavy and K5 mildly centre-heavy) and comparatively high multilevel integration scores (N). Taken together, these results describe a governance landscape in which national frameworks are comparatively consolidated, yet the territorial diffusion of adaptation responsibilities remains uneven, and upward–downward integration varies substantially across the Member States.

4.2. Country Archetypes

Applying the deterministic rules defined in Section 3, five mutually exclusive archetypes (rule-based clusters K1–K5) were identified (

Table 4. Cluster synopsis (membership, medians and IQR for composite signposts). Medians and IQRs are computed from country-level values and serve as ordinal summaries; cluster assignment follows the deterministic rules in Section 3.

Cluster	No. of Countries	Members	Median J (IQR)	Median K (IQR)	Median L (IQR)	Median M (IQR)	Median N (IQR)
K1—Lagging/Fragmented	14	AT, BG, CY, CZ, EE, HU, IT, LV, LT, MT, PL, RO, SK, SI	1.5–1.8 (≈0.7–1.0)	1.0 (≈0.5–1.0)	1.0 (≈0.7–1.3)	0.1 (≈0.0–0.1)	1.2 (≈1.1–1.3)
K2—Top-Down Centralised	3	FI, LU, ES	3.0 (≈0.3)	1.0 (≈0.5–1.0)	1.7 (≈1.3–2.3)	0.3 (≈0.3–0.4)	1.9–2.4 (≈0.3–0.5)
K3—Region-Driven	5	BE, HR, DE, EL, SE	1.3 (≈0.4)	2.0–2.5 (≈0.5–1.0)	2.0 (≈0.7)	0.3–0.4 (≈0.1)	1.7–1.9 (≈0.2–0.3)
K4—Local/Urban Bottom-Up	1	DK	1.7 (–)	2.0 (–)	2.3 (–)	0.3 (–)	2.0 (–)
K5—Integrated (Multilevel)	4	FR, IE, NL, PT	2.9–3.0 (≈0.3)	2.5–3.0 (≈0.5–1.0)	2.7–3.0 (≈0.7–1.0)	0.6 (≈0.5–0.6)	2.7–3.1 (≈0.3–0.6)

; Figure 2). Memberships are invariant under a conservative sensitivity stress test on N (subtracting 0.25 when any base indicator equals 0) and remain stable under reasonable non-compensatory adjustments [12,43,47,48,49,56]. This stability supports the analytical use of the archetypes as recurrent governance configurations, rather than as artefacts of borderline threshold choices [43,44,47,48,56,57].

For each cluster, Table 4 reports the number of Member States, the member list, and the cluster-level medians and IQRs for the composite signposts J (NAT_CORE), K (TERR_MAN), L (Local synergy), M (Advanced breadth) and N (Multilevel integration). These ordinal summaries enable a concise verification of the rule logic (e.g., K5 is characterised by higher vertical balance—median N around 3—and higher territorial/local pillars—median K and L ≳ 2.5—whereas K1 shows lower J, K, L simultaneously).

• K1—Lagging/Fragmented (14 MS): AT, BG, CY, CZ, EE, HU, IT, LV, LT, MT, PL, RO, SK, SI. Profiles combine multiple low pillars (≥4 bases ≤ 1), weak national cores (J < 2.5) and thin territorial layers (C ≤ 2, D ≤ 2, L ≤ 2). N typically ≤ 1.7; positive TOP_DOWN indicates centre-led yet shallow systems. Sub-national obligations are often non-binding or absent; sectoral institutionalisation is at an early stage; LTS content tends to be narrative rather than KPI-based; urban integration is at the guidance level; and CoM leverage is generally low. These profiles do not satisfy the gates for other archetypes (not K2/K3/K4/K5 for the stated reasons). This profile does not meet the gate conditions for K2–K5 because multiple base indicators remain at low levels (≤ 1), resulting in low values for J, K, L and N.
• K2—Top-Down Centralised (3 MS): FI, LU, ES. Robust national cores (J ≥ 2.5) coexist with N < 2.5 and at least one territorial gate below K5 (often D < 3 and/or C < 2–3), typically with O = J − L > 0. Finland and Luxembourg pair strong legal backbones with voluntary or thin territorial layers. Spain is closest to K5 but remains below D ≥ 3 and N ≥ 2.5. K2 profiles do not meet K5 because multilevel balance remains below the integration threshold (N < 2.5) and at least one territorial gate is not satisfied (typically D < 3 and/or C below the K5 requirement).
• K3—Region-Driven (5 MS): BE, HR, DE, EL, SE. Regional institutions are the primary engine (C ≥ 3, K ≥ 2), while the national core is non-dominant (J < 3) and N ≈ 1.7–1.9; local synergy is present but uneven (L ≈ 2.0). Under the rule set, K3 profiles do not meet K5 because local coverage remains below the threshold (D < 3), the national legal core does not meet the gate (H < 3, with J remaining below the dominant range), and multilevel balance remains below the integration threshold (N < 2.5).
• K4—Local/Urban Bottom-Up (1 MS): DK. High local synergy (L > 2) driven by mandatory LAPs with full population coverage (D = 4), weak/absent regional tier (C ≤ 2) and a non-dominant national core (J < 3), yielding N = 2.0. Under the rule set, this configuration does not meet K5 because the legal core and regional tier gates are not satisfied (H < 3 and C < 2), and multilevel balance remains below the integration threshold (N < 2.5).
• K5—Integrated (Multilevel) (4 MS): FR, IE, NL, PT. Membership requires all high bars—B ≥ 2, C ≥ 2, D ≥ 3, G ≥ 2, H ≥ 3, N ≥ 2.5—so neither strong laws nor dense local activity alone suffice. These systems exhibit convergent medians across J, K, and L, and at high N, indicating a configuration characterised by balanced pillars rather than single-lever dominance. Differences at the margins (sectoral articulation E, network mobilisation I) are secondary to the rule-gates.

The mapped distribution of archetypes across the Member States shows recognisable geographical groupings. K1 is prevalent in parts of Central and Eastern Europe; K2 combines strong national backbones with limited downward diffusion; K3 displays region-led configurations with uneven local mobilisation; K4 is a local-first configuration; K5 shows vertically aligned systems with broader coverage. These descriptions summarise observed memberships and do not imply causal interpretation or performance ranking. The interpretative implications of these configurations, including links to institutional and administrative conditions discussed in the literature, are examined in Section 5.

4.3. Robustness and Sensitivity

We assessed the stability of the rule-based cluster memberships through three stress-test scenarios: (i) a conservative non-compensatory sensitivity adjustment applied to the Multilevel Integration Index N (subtracting 0.25 when any base indicator equals 0), (ii) moderate shifts in the high-bar threshold for N (±0.25 around the reference value of 2.50), and (iii) an inverted ordering of the tiebreaker variables. Across all three scenarios, no Member State changed cluster membership, indicating full invariance of the classification to alternative threshold and tie-breaking assumptions. The detailed scenario definitions and outputs are reported in Appendix C (

Table A6. Sensitivity of rule-based cluster assignments to alternative thresholds and tiebreakers (EU-27, reference July 2025).

Scenario ID	Change vs. Baseline	Rationale Tested	Countries Reclassified (n)	Reclassification Details	Adjusted Rand Index vs. Baseline	% Stable
S1	Zero-penalty on N (−0.25 if any base = 0)	Non-compensatory stress (penalises zero pillars)	0	—	1.00	100
S2	N threshold stricter: 2.75 (vs. 2.50)	Tougher integration bar for K5	0	—	1.00	100
S3	N threshold looser: 2.25 (vs. 2.50)	Easier integration bar for K5	0	—	1.00	100
S4	Tiebreakers inverted (J before N)	Sensitivity to tie resolution	0	—	1.00	100

).

4.4. Typologies of Adaptation Governance (Five Country Clusters)

4.4.1. Cluster 1: Lagging/Fragmented

This cluster groups Member States where adaptation governance is simultaneously limited at the national core and thin across subnational tiers. Assignment follows non-compensatory, descriptor-anchored rules. Countries fall here when the national core is modest (J < 2.5), territorial mandates are low (K ≲ 1.5; typically, RAP ≤ 1 and LAP ≤ 1–2), and the local/urban ecosystem remains underdeveloped (L ≤ approx. 1.7). These conditions are associated with low vertical balance (N ≤ approx. 1.3) and yield positive O values (J − L > 0) indicating centre-heavy yet shallow configurations.

Across AT, BG, CY, CZ, EE, HU, IT, LV, LT, MT, PL, RO, SK, SI, standard features include non-binding or absent subnational obligations, early-stage sectoral institutionalisation, and largely narrative (rather than KPI-based) LTS content, weak integration into national urban policy, and limited network leverage via the Covenant of Mayors.

Illustrative country examples:

• Austria: Mature NAS/NAP, but no adaptation clauses in the federal Climate Protection Act (Klimaschutzgesetz) and voluntary Länder instruments (RAP = 1; SAP = 1; G = 1; I very low).
• Czechia: Comprehensive NAS/NAP, no climate act; RAP recommendations are voluntary (RAP = 1) and LAP/CoM uptake is extremely low (LAP = 0; I ≈ 0.37%).
• Estonia: Solid NAS and an LTS chapter (F = 2), but no mandated RAP/LAP/SAP and I = 0.
• Hungary: County-level strategies in the 2014–2020 EU-funding window (temporarily RAP = 2), but no binding framework at present (LAP = 1; H = 0).
• Romania, Latvia, Lithuania: Adaptation embedded narratively in LTS/NUP (F ≈ 2; G ≈ 1–2), without enforceable subnational duties and with low CoM coverage.
• Italy: updated PNACC, no framework climate law; regional/local diffusion remains voluntary, and CoM coverage is moderate (approx. 8%).
• Malta: H = 4 (strong law), yet no RAP/LAP/SAP and I = 0; under the non-compensatory rules, the profile remains in K1.

Stand-alone, binding SAPs are uncommon (most E = 0–1; SK/SI = 2 only). LTS content is largely qualitative (F = 0–2). NUP integration is typically guidance-level (G = 0–2). CoM engagement is generally very low to low, with a few medium cases and isolated higher values; detailed percentages are reported with the country codes (see Appendix B). Profiles do not meet the alternative gates: not K2 (no consistently high J together with sufficient N), not K3 (regional mandates generally below C = 3), not K4 (local coverage neither mandatory nor ≥50%, i.e., D < 3), and not K5 (integration gates—N ≥ 2.5, D ≥ 3, C ≥ 2, H ≥ 3, B ≥ 2, G ≥ 2—not jointly satisfied).

4.4.2. Cluster 2: Top-Down Centralised

This cluster includes Finland, Luxembourg and Spain, where a comparatively strong, legally anchored national core coexists with incomplete vertical diffusion to regional and local tiers. Assignment follows the descriptor-anchored, non-compensatory rules: high J (≥2.5)—typically reflecting up-to-date NAS/NAP, adaptation-relevant climate law (H ≥ 3), and adaptive content in the LTS—combined with sub-threshold multilevel balance (N < 2.5) and at least one unmet territorial gate for K5 (notably D < 3 and/or C < 2–3), usually with positive O values (J − L > 0), indicating centre-heavy configurations.

Illustrative country profiles:

• Finland: The 2022 Climate Act includes a whole adaptation chapter (H = 4) and NAS/NAP are current (B = 4); RAPs are voluntary with <50% coverage (C = 1), LAPs are voluntary with limited diffusion (D = 1), and NUP integration is reference-level (G = 1). The resulting configuration (J = 3.3, N = 1.9, L = 1.3) does not meet the K5 gates.
• Luxembourg: A robust Climate Law (H = 4), an operational national plan, and a strong strategy (B = 3) coexist with an absent regional tier (C = 0) and non-mandated LAPs (D = 1); despite a coherent central policy (J = 3.0), N = 1.7 and L = 1.7 remain below K5 thresholds.
• Spain: A consolidated NAS/NAP cycle codified in Law 7/2021 (H = 3; B = 4) and structured monitoring is accompanied by widespread yet voluntary regional/local engagement (C = 2; D = 2), G = 2, and N = 2.4—closest to K5, yet still below the D ≥ 3 and N ≥ 2.5 gates.

These profiles do not meet the criteria for the following:

• K3 (Region-Driven): Regions are not the primary engine (FI C = 1; LU C = 0; ES C = 2 with voluntary diffusion);
• K4 (Local/Urban Bottom-Up): LAPs are not mandatory with ≥50% coverage (D < 3), and L is moderate;
• K5 (Integrated): The joint non-compensatory gates—N ≥ 2.5 together with D ≥ 3 (and C ≥ 2, H ≥ 3, B ≥ 2, G ≥ 2)—are not simultaneously satisfied (all three cases fall short on N and D, and FI/LU also on C).

Overall, K2 systems exhibit clear national legal backbones and strategy cycles (elevating J/H/F), alongside incomplete territorial mandates and local coverage (constraining K, L, N). This rationale clarifies why high J alone does not yield K5 membership under non-compensatory, descriptor-anchored rules.

4.4.3. Cluster 3: Region-Driven

Cluster 3 comprises systems where regional institutions constitute the primary locus of adaptation governance, while the national core remains non-dominant. Assignment follows the descriptor-anchored, non-compensatory rules: comparatively strong territorial mandates, operationalised by RAP = 3–4 and K ≥ 2.0, coexist with a non-dominant national core (J < 3) and intermediate multilevel balance (N ≈ 1.7–1.9), with local synergy present but uneven (L ≈ 2.0). These conditions distinguish K3 from adjacent archetypes: K2 (high J ≥ 2.5 but voluntary C/D and moderate L), K4 (requires D ≥ 3, i.e., legally mandated LAP coverage ≥ 50%), and K5 (requires the full set of non-compensatory gates: N ≥ 2.5, D ≥ 3, C ≥ 2, H ≥ 3, B ≥ 2, G ≥ 2). Illustrative country profiles:

• Belgium illustrates a constitutionally decentralised model: each region has an autonomous, up-to-date RAP (C = 3; K = 2.5), CoM participation is very high (I = 4; L = 2.3), yet the federal core is weak (J = 1.3) and there is no national climate law on adaptation (H = 0).
• Croatia and Greece display asymmetric profiles: statutory or fully rolled-out regional planning (EL C = 4; HR C = 3; both K ≥ 2) coexists with underpowered national cores (HR J = 1.3, H = 2; EL J = 1.0, H = 1) and non-mandatory LAPs (D = 1), with mid-range N (≈1.7–1.8).
• Germany’s Länder drive adaptation (C = 3; K = 2.0), but the federal layer remains primarily strategic (J = 1.7), with no binding national adaptation law (H = 1), no stand-alone SAPs (E = 0), limited LAP diffusion (D = 1) and very low CoM coverage (I = 0)—all of which limit L (=1.3) and N (=1.7).
• Sweden combines mandatory regional assignments (C = 3; K = 2.5) and strong sectoral agency plans (E = 4) with integration in urban planning (G = 3; D = 2); the national core remains non-dominant (J = 1.3; H = 1) and N (=1.9) remains below the integrated threshold.

Two cross-cutting patterns are apparent in the rule-based classification. (i) Strong regional mandates alone do not produce K5 membership in a non-compensatory design: without D ≥ 3, H ≥ 3 (with J rising), and N ≥ 2.5, high C/K cannot offset lower J/L/H. (ii) Descriptor gates set the boundary between K3 and adjacent archetypes: K3 differs from K2 by territorial strength (C/D), from K4 by the absence of legally mandated local coverage at or above 50% (D < 3), and from K5 by not meeting all simultaneous integration gates.

4.4.4. Cluster 4: Local Bottom-Up

This cluster is illustrated by Denmark, where legally mandated local instruments constitute the primary drivers of adaptation governance. Rule-based assignment reflects the non-compensatory gates: high local synergy (L = 2.3) underpinned by mandatory LAPs with full population coverage (D = 4), a weak/absent regional tier (C = 0), a non-dominant national core (J = 1.7 < 3; H = 0), and intermediate multilevel balance (N = 2.0 < 2.5). This configuration differentiates K4 from adjacent archetypes: (i) K3, where regional mandates are strong (C ≥ 3), and (ii) K5, which requires all high bars to be met simultaneously (N ≥ 2.5, D ≥ 3, C ≥ 2, H ≥ 3, B ≥ 2, G ≥ 2). Denmark’s country values (C = 0, D = 4, E = 1, F = 3, G = 2, H = 0, I = 1; J = 1.7, K = 2.0, L = 2.3, M = 0.3, N = 2.0) are reported in Appendix B.3 and meet the K4 criteria by design.

A city-led model was operationalised through the 2012 national initiative, which required all 98 municipalities to prepare climate-adaptation plans and embed them in statutory spatial planning, later reinforced by cloudburst planning duties. These obligations resulted in universal LAP coverage (D = 4), systematic risk mapping and actionable measures, generating a dense local implementation ecosystem captured by L > 2. The regional layer is not used as a planning instrument (C = 0) following county reform, so the territorial-mandate pillar remains modest (K = 2.0) despite strong municipal implementation.

Denmark’s NAS (2008) and NAP (2012) provide strategic direction but have not been recently refreshed with binding, review-backed duties; the Climate Act (2020) is mitigation-centred and does not codify adaptation mandates (H = 0). Adaptive content is present in the long-term strategy (F = 3), which references governance arrangements and mainstreaming, raising J above the lowest band without establishing a legal core. Urban policy integration is prescriptive but not fully binding (G = 2): adaptation is woven into guidance and flood-risk provisions but not enforced through national KPI-based standards. Sectoral articulation is limited (E = 1), reflecting reliance on integration through water, spatial and infrastructure regimes rather than stand-alone SAPs. Network mobilisation via the Covenant of Mayors is modest (I = 1; ≈4% population coverage), partly offset by a national technical backbone (e.g., guidance portals, coastal/flood tools and templates) that standardises local practice outside CoM channels.

Denmark fits K4 under the rule-based classification: local legal mandates and planning integration deliver high L and D in the absence of regional RAPs and a binding national adaptation law, producing a bottom-up but not yet fully integrated multilevel system (N = 2.0).

Inizio modulo

Fine modulo

4.4.5. Cluster 5: Integrated (Multilevel)

Assignment follows the non-compensatory gates defined in Section 3 and documented in Appendix B.3: membership requires that all high bars be met (B ≥ 2, C ≥ 2, D ≥ 3, G ≥ 2, H ≥ 3, N ≥ 2.5). Therefore, neither robust legislative frameworks nor intensive local activities are sufficient on their own. All four members satisfy these conditions comfortably and display convergent medians (J ≈ 2.9–3.0; K ≈ 2.5–3.0; L ≈ 2.7–3.0; N ≈ 2.7–3.1; see also Table 4), indicating balanced pillars rather than dominance by a single governance lever.

France implements the rule set through mandatory regional planning and local obligations (C = 3; D = 3) embedded in SRADDET and PCAET cycles, legal codification (H = 3), urban integration (G = 3), and an updated strategic–legal core (B = 2; F = 4), resulting in N = 2.7, J = 3.0, K = 3.0, and L = 2.0. Its sectoral plans remain embedded, not standalone (E = 0), and network leverage via CoM is very low (I = 0); however, these do not affect assignment, as the gates are driven by legal mandates and multilevel balance rather than voluntary networks.

Ireland illustrates statutory local activation through legal LAP/LCAP obligations (D = 4), sectoral plans issued by ministries (E = 3), urban policy coherence (G = 4), and climate law anchoring (H = 4). Although it lacks classic RAP PDFs, the CARO hubs and the National Planning Framework deliver functional regional coordination (C = 2), sufficient for the rule, and the system achieves N = 3.0, with J = 2.7, K = 3.0, and L = 3.3—the highest L in the cluster. CoM uptake is moderate (I = 2), again non-determinant given the legal gates.

The Netherlands meets and often exceeds the thresholds through the Delta Programme/Act and the Environment & Planning Act: universal stress tests → risk dialogues → spatial embedding result in de facto mandatory subnational coverage (C = 4; D = 4; G = 3; H = 4; B = 4), delivering N = 3.1 and J = 3.0/K = 4.0/L = 2.3. Like France, standalone SAPs are not used (E = 0), and CoM is marginal (I = 0); however, the legally enforced hydrological–spatial regime achieves the required multilevel integration without relying on CoM.

Portugal combines an updated NAS/NAP (B = 3) with statutory local obligations under the Climate Framework Law (D = 3; H = 4) and expanding regional coverage (C = 2), supported by urban policy integration (G = 3). This configuration yields N = 2.7, J = 2.7/K = 2.5/L = 3.0, with comparatively high CoM engagement relative to other K5 members (I = 3). Sectoral content is still embedded (E = 1), but the legal gates are nonetheless satisfied.

Across K5, the defining configuration is the simultaneous presence of (i) a legal core, (ii) territorial mandates, and (iii) urban integration, jointly raising N above the 2.5 balance threshold. Differences at the margins—sectoral articulation (E ranges from 0 in FR/NL to 3 in IE) and network mobilisation (I spans 0–3)—are secondary to the rule-based gates. Across K5, variation in sectoral articulation (E) and network mobilisation (I) does not affect class assignment, because membership is determined by the simultaneous satisfaction of the legal, territorial, and urban-integration gates alongside N ≥ 2.5.

4.5. Dominant Limiting Mechanisms Across Adaptation-Governance Archetypes

Although the clusters are derived through deterministic, non-compensatory rules and do not imply causal inference, each archetype can be associated with a dominant limiting mechanism that constrains progression towards integrated multilevel governance:

• K1—Lagging/Fragmented: Absence of binding legal anchoring and accountability. Weak or missing statutory duties, limited review cycles, and underdeveloped monitoring frameworks hinder territorial diffusion, even when strategies exist.
• K2—Top-Down Centralised: incomplete downward diffusion of mandates. Strong national legal–strategic cores are not matched by mandatory regional or local coverage and monitored urban standards.
• K3—Region-Driven: Lack of binding local coverage and national coherence. Robust regional mandates coexist with moderate national cores, but integration stalls without legally mandated local plans and shared monitoring frameworks.
• K4—Local/Urban Bottom-Up: Missing intermediate coordination and national legal anchoring. Dense municipal implementation ecosystems emerge through statutory planning hooks, yet the absence of a functional regional tier and national adaptation duties constrains the balance across levels.
• K5—Integrated (Multilevel): No single dominant limiting mechanism. Legal anchoring, territorial mandates, and monitored urban standards co-evolve, allowing balanced multilevel integration.

This diagnostic summary indicates that progression across archetypes is associated with addressing the binding constraint specific to each configuration, rather than reinforcing already strong governance dimensions.

5. Discussion

This Discussion deliberately separates interpretation from description, building on the empirical results reported in Section 4. This section situates the identified archetypes within the multilevel climate adaptation governance literature, clarifying their theoretical meaning, underlying governance mechanisms, and policy relevance in analytical terms. The purpose is not to prescribe specific policy solutions, but to explain why certain institutional configurations recurrently enable—or constrain—coherent multilevel implementation. In this sense, the classification functions as a diagnostic tool rather than a descriptive inventory or a performance ranking, making visible the institutional configurations that systematically shape multilevel adaptation capacity.

Revisiting the research question—how national adaptation governance arrangements enable coherent multilevel implementation and what maturity patterns emerge—this section interprets the results beyond their descriptive presentation. By linking the empirically observed archetypes to institutional and administrative mechanisms discussed in comparative governance research, the Discussion strengthens the explanatory contribution of the framework. It clarifies the added value of the maturity lens relative to descriptive policy inventories [4,5,6,16,17,43,44,45,46]. Specifically, the maturity lens distinguishes between the formal presence of policies and their degree of institutionalisation, explaining why similar policy portfolios can lead to markedly different multilevel implementation outcomes.

Overall, the results indicate that, across European Union Member States, governance maturity is more frequently concentrated at the national tier, while territorial diffusion and vertical integration remain uneven. The five archetypes—Lagging/Fragmented, Top-Down Centralised, Region-Driven, Local/Urban Bottom-Up, and Integrated (Multilevel)—capture recurrent governance equilibria rather than transitional stages. These equilibria reflect differences in institutional design, administrative traditions, and path dependence rather than the mere presence or absence of policy documents [4,5,6,16,24,25,26,45,46]. In line with the literature, the findings confirm that formal strategies are a necessary but insufficient condition for effective multilevel implementation: without binding mandates, monitoring, and financial support, national ambition often fails to translate into durable territorial coverage [7,10,11,13,14,20,43]. The analytical value of the classification thus lies not in cataloguing differences across Member States but in explaining why specific governance equilibria persist over time despite increasing levels of adaptation activity.

5.1. Binding Constraints and Limiting Mechanisms Across Adaptation-Governance Archetypes

From a diagnostic perspective, the identified archetypes can be interpreted as structured combinations of binding constraints that shape how multilevel climate adaptation governance operates in practice. Building on the descriptive clustering presented in Section 4, this section interprets the archetypes as manifestations of distinct limiting mechanisms within multilevel governance systems. Rather than treating governance maturity as a linear or incremental scale, the framework reveals how different institutional configurations stabilise at different equilibria depending on which constraint—legal anchoring, territorial diffusion, or monitored urban standards—dominates. This diagnostic reading aligns the empirical typology with comparative governance research and explains why similar levels of adaptation activity may result in markedly different multilevel implementation outcomes.

To explain why Member States fall into specific archetypes, the typology can be read as a diagnostic of binding constraints operating across three governance mechanisms that the literature consistently associates with multilevel coherence in climate adaptation:

(i)

legal anchoring and accountability, including binding statutory duties, review cycles, indicators, and finance-linked enforcement mechanisms;

(ii)

territorial diffusion of mandates across regional and local tiers; and

(iii)

urban–local implementation ecosystems, supported through planning instruments, guidance, finance, data infrastructures, and network mobilisation [4,7,10,11,13,14,16,17,21,24,43,44,45,46].

These mechanisms interact with administrative traditions (e.g., centralised versus decentralised systems) and historical trajectories, shaping whether national, regional, or municipal levels emerge as the primary engines of adaptation governance. Importantly, outcomes are not reducible to a single explanatory factor; rather, they reflect the combined effects of institutional design choices and path-dependent governance arrangements [4,5,6,16,23,24,25,26,45,46].

Within this reading, K2 (Top-Down Centralised) systems are characterised by strong legal–strategic national backbones but insufficient downward diffusion. National coherence does not translate into multilevel integration where local coverage obligations and monitored standards remain voluntary or only partially enforced [10,13,14,17,24,43,44]. K3 (Region-Driven) systems capture configurations in which robust regional mandates and planning competences coexist with a weaker national core in decentralised settings; however, integration remains constrained in the absence of binding local coverage and nationally coherent monitoring frameworks [4,17,22,23,45,46]. K4 (Local/Urban Bottom-Up) illustrates how statutory municipal planning hooks can generate dense local implementation ecosystems, while the absence of an intermediate coordinating tier complicates cross-boundary risk management and limits vertical balance for hazards that exceed traditional administrative borders [11,21,32,33,34].

By contrast, K5 (Integrated Multilevel) corresponds to configurations in which legal anchoring, territorial mandates, and monitored urban standards co-evolve within a coherent multilevel architecture and are reinforced through iterative monitoring and learning cycles. This configuration aligns closely with European Union and international guidance that identifies accountability, coherence, and adaptive governance as enabling conditions for durable implementation [1,2,3,13,14,17,34,43,44]. Finally, K1 (Lagging/Fragmented) represents contexts in which multiple mechanisms remain simultaneously weak—limited binding mandates, thin sectoral institutionalisation, and low territorial diffusion—consistent with well-documented barriers such as fragmented responsibilities, capacity gaps, and underdeveloped monitoring and learning systems, even where strategies formally exist [7,10,16,17,43,44].

Two cross-cutting dynamics further clarify these patterns. First, networks tend to mobilise, whereas laws and budgets institutionalise. Participation in city networks—most notably the Covenant of Mayors—often co-occurs with a proliferation of local plans and increased reporting activity. However, its integrative effect depends on domestic enabling conditions such as statutory planning hooks, routine review cycles, and predictable subnational finance [8,9,12,24,25,29,49]. Where these enablers are weak, extensive municipal activity remains voluntary and fails to raise multilevel integration above the thresholds defined by the non-compensatory rules. Spain provides an illustrative example of this dynamic: widespread municipal practice does not elevate local coverage beyond voluntariness unless majority population coverage is both mandated and monitored (D ≥ 3), thereby keeping local synergy (L) and vertical balance (N) below the integrated gate despite high levels of activity [12,13,14,29,49,59,63].

5.2. Diagnostic Implications for Policy Design

The policy relevance of the classification lies in its capacity to identify dominant limiting mechanisms, thereby supporting prioritisation, accountability and targeted intervention rather than prescribing uniform solutions or best practices. The typology clarifies which levers dominate—and which constraints bind—in each equilibrium. Top-Down Centralised systems pair strong national cores with limited downward diffusion, underscoring that territorial mandates and financing instruments must match statutory ambition if integration is to occur [13,14,43]. Region-Driven systems demonstrate that robust regional agency and planning can partially compensate for a moderate national core but rarely deliver full integration without binding local coverage and nationally coherent risk methodologies [17,22,23,32,33]. Local/Urban Bottom-Up profiles highlight the effectiveness of embedding adaptation into municipal land-use instruments, while also revealing the coordination challenges that arise in the absence of an intermediate tier for managing hazards that transcend local boundaries [32,33]. Integrated (Multilevel) systems perform best where legal frameworks, territorial mandates, and urban integration co-evolve and are stabilised through shared information infrastructures and monitoring frameworks [1,2,3,13,17,43].

From an operational perspective, each archetype is associated with a dominant limiting mechanism—for example, missing legal hooks, weak territorial mandates, or insufficiently monitored urban standards. Consequently, policy recommendations target the binding constraints rather than amplifying already-strong pillars [7,10,11,13,14,17,43,44]. Lagging/Fragmented systems benefit from statutory RAP and LAP requirements with five-year review cycles, national technical hubs for standardised risk data, conditional grants, and initial binding sectoral plans in priority domains such as water and health. Top-Down Centralised systems should convert voluntary regional and local plans into mandates, embed binding urban standards, and expand monitored municipal coverage to cross the integration threshold [13,14,43]. Region-Driven settings require national minimum coherence standards for risk assessment and indicators, combined with legal local obligations aligned to regional strategies and pooled regional funding mechanisms to reduce project dependency [17,22,23,32,33]. Local Bottom-Up profiles should complement strong municipal leadership with functional regional coordination (e.g., basin-scale consortia) and codified national adaptation duties. Integrated systems can deepen sectoral mainstreaming, broaden participation, and link local obligations to outcome-oriented metrics supported by stable subnational finance instruments, such as revolving funds or climate bonds [1,2,3,13,14,32,33,43,61].

5.3. Implications for Interpretation and Methodological Scope

Because the framework is explicitly non-compensatory, progress towards higher governance maturity requires relaxing or removing the binding constraints rather than reinforcing already strong dimensions. In practical terms, near-term commitments can be expressed as small, verifiable shifts in indicator levels—for example, converting voluntary LAPs into mandatory instruments with majority population coverage and a five-year review cycle (D → 3), or upgrading planning guidance into monitored standards (G → 3). These changes directly affect local synergy (L) and, in turn, multilevel balance (N). In top-down profiles, integration is achieved not by further fortifying the national core (J), but by raising D and G so that local capacity catches up. In region-driven settings, high regional mandates alone are insufficient unless complemented by national legal hooks (H ≥ 3) and binding local coverage (D ≥ 3). In local bottom-up systems, dense municipal practice must be balanced by functional regional coordination and national recognition if improvements are to register as vertical balance within the multilevel system rather than as local exceptionalism.

Methodologically, the use of descriptor-anchored cut-offs, non-compensatory rules, and targeted sensitivity probes addresses common pitfalls associated with short ordinal scales and composite indicators [43,44,47,48,56,57]. By tying indicators to recognised policy instruments (NAS/NAP, RAP, LAP, SAP, LTS, NUP, climate law, Covenant of Mayors), the framework supports a portfolio perspective in which governance maturity emerges from the co-evolution of legal clarity, territorial mandates, financial incentives, and interoperable data and reporting systems, rather than from the maximisation of any single score [18,19,20,21,44]. These analytical implications complement the methodological limitations outlined in Section 3, clarifying how the framework should be interpreted rather than operationalised. Accordingly, the framework is intended to support comparative diagnosis and strategic reflection, not to function as an operational checklist or implementation manual.

6. Conclusions

Europe is experiencing rapid warming, with temperatures increasing faster than the global average; climate change adaptation (CCA) has therefore become integral to European policy to anticipate and manage unavoidable impacts [1,2,3]. This study assesses the maturity of CCA arrangements across the 27 Member States in 2025, updating earlier stocktakes. Using eight dimensions—(i) National Adaptation Strategies/Plans; (ii) Regional Adaptation Plans; (iii) Local Adaptation Plans; (iv) Sectoral Adaptation Plans; (v) integration in National Urban Policies; (vi) adaptive content in Long-Term Strategies; (vii) adaptation relevance in climate framework laws; and (viii) adhesion to the Covenant of Mayors—we identify five governance clusters: Integrated Multilevel, Top-Down Centralised, Region-Driven, Local Bottom-Up, and Lagging/Fragmented. These clusters derive from descriptor-anchored, non-compensatory rules and reveal substantial variability across the European Union, underscoring the need for integrated multilevel systems [13,14,43,44,56]. Rather than ranking Member States or assessing policy effectiveness, the classification is designed to diagnose how institutional configurations condition the capacity for coherent multilevel adaptation governance.

The maturity framework operationalises policy and legal instruments into transparent, ordinal indicators, and, in doing so, advances maturity thinking as a diagnostic tool for multilevel climate adaptation governance rather than as a performance-ranking exercise [13,14,43,44,56]. Its analytical contribution lies in explaining how and under which institutional conditions adaptation systems tend to stall, fragment or integrate, based on institutional design rather than on stated ambitions or narrative content. It addresses the research aims by (i) developing and applying a governance-maturity framework and (ii) interpreting implementation implications through rule-based archetypes that reveal where—and why—systems stall or advance.

Recent evidence of escalating climate risks highlights the need for institutional designs that align legal mandates, territorial diffusion, and urban integration, rather than relying on a single territorial level effort [1,2,3]. Entry to an integrated multilevel regime requires that all high thresholds are met simultaneously—at minimum B ≥ 2, C ≥ 2, D ≥ 3, G ≥ 2, H ≥ 3, and N ≥ 2.5—so strong performance in one tier cannot compensate for weak statutory or territorial foundations elsewhere [13,14,43,44,56]. In practice, networks mobilise, while laws and budgets institutionalise municipal activity scales and persist when embedded in planning codes, monitored review cycles, and predictable subnational finance; without these, practice remains fragmented despite local dynamism [8,9,49].

The diagnostic reading of the archetypes supports the identification of priority levers for policy design, illustrating how different institutional configurations can be rebalanced in practice:

• Legislate RAP/LAP gates with five-year review and reporting, anchored in national planning law (raises C, D, G, N) [13,14,61].
• Adopt binding, outcome-oriented urban standards (e.g., flood/heat provisions) with monitored compliance (raises G, L) [32,33,34,61].
• Link conditional grants and performance-based transfers to monitored milestones and indicator reporting (supports J, L, N) [43,44,56].
• Establish functional regional consortia at basin/coastal scales to coordinate cross-boundary risks and pool capacity (raises C, K, N) [11,32,33,34].
• Integrate Covenant of Mayors reporting with national/EU monitoring to standardise local data and verification (raises L, N) [8,12,49].
• Strengthen legal recognition of adaptation in framework laws with financing hooks and statutory review cycles (raises H, J, N) [13,14].

The usefulness of this assessment lies in converting comparative diagnostics into indicator-based commitments that align with current planning and budget cycles. Since the framework is non-compensatory, systems progress only when the gate conditions defining integration are collectively met—at a minimum, B ≥ 2, C ≥ 2, D ≥ 3, G ≥ 2, H ≥ 3, and N ≥ 2.5. Short-term priorities can therefore be articulated as clear progress on binding pillars: require majority-coverage local plans with periodic reviews (D → 3), elevate urban provisions from guidance to monitored standards (G → 3), and incorporate adaptation duties into climate-framework law (H → 3). These measures align legal design, territorial spread, and urban practice and are immediately observable through changes in L and N. When capacities are uneven, equity-focused sequencing—using targeted technical support and grants—can help ensure that compliance with D and G thresholds reduces, rather than increases, territorial inequalities [61,62].

Although the empirical analysis is deliberately focused on the European Union, the framework speaks to broader debates on multilevel climate adaptation governance beyond the EU context. The European Union represents a particularly instructive case because it combines strong supranational policy coordination with highly diverse national, regional, and local governance traditions. This makes it possible to observe, within a single analytical setting, how similar strategic commitments translate into markedly different multilevel implementation configurations.

Comparative evidence from North America and Asia indicates that analogous institutional challenges recur across governance systems, including persistent gaps between national adaptation strategies and subnational implementation, uneven diffusion of binding mandates, and a continued reliance on voluntary local action in the absence of enforceable coordination mechanisms [35,39,42]. While legal competences, fiscal arrangements, and planning systems differ substantially across regions, the diagnostic logic advanced here—focusing on legal anchoring, territorial mandates, and local–urban implementation ecosystems—remains transferable as an analytical lens.

Rather than promoting a specific European model, the framework offers a structured way to identify dominant limiting mechanisms in multilevel adaptation governance and to distinguish between activity-rich but weakly institutionalised systems and genuinely integrated arrangements. In this sense, the EU-based analysis provides insights relevant to other multilevel settings, while recognising that threshold levels and institutional pathways must be adapted to the context.

Limitations remain. The indicators measure institutional readiness rather than actual risk reduction; brief ordinal scales and document-based coding may misclassify edge cases despite descriptor anchors and sensitivity checks. Accordingly, the framework should be read as a comparative diagnostic of governance capacity, not as an evaluation of adaptation outcomes or policy performance. Translation of source documents introduces uncertainty; future work will triangulate with expert review and sample-based cross-validation, expand coverage of finance and private actors, and link institutional maturity to quantitative outcomes (e.g., losses avoided, excess-heat mortality) [43,44,56]. Applying and benchmarking the same ruleset outside Europe—especially in the United States and Asia, where state–local agreements vary—will assess external validity and improve transferability [2,61]. In addition, the study does not include a formal expert review or stakeholder validation process, which represents a recognised limitation of the current design. While multiple safeguards were adopted to ensure reliable document-based coding, structured expert consultation would provide a valuable complementary layer of triangulation and is therefore identified as a priority direction for future research. Accordingly, the framework is intended to inform comparative diagnosis and strategic reflection, rather than to prescribe uniform policy pathways or institutional designs.

Advancing climate-resilient governance in Europe ultimately depends on the coevolution of statutory mandates, multilevel coordination platforms, interoperable data/reporting, and stable subnational finance. Aligning these elements with the European Green Deal, the EU Adaptation Strategy, and the Sendai Framework provides a credible pathway to accelerate implementation while ensuring equity and effectiveness [13,15,34,43].