From Fragmented Plans to a Public Regulatory Knowledge Infrastructure: Harmonizing Flood and Landslide Risk Planning in the Central Apennine River Basin District, Italy

Abstract

Italy’s transition to river basin district governance within the framework of Directive 2000/60/EC unified institutional responsibilities while leaving existing flood and landslide plans in force. District authorities therefore had to integrate instruments with different knowledge bases, classifications and land-use rules. This article examines the integration of nine Hydrogeological Setting Plans (PAIs) into two coordinated plans for the Central Apennine district (42,275 km²), adopted in final form in May 2026. Document analysis compared classifications, intervention conditions and assessment responsibilities, examining 175 public submissions and 118 first-round regional determinations. Geographic information system analysis quantified current classification correspondences and coverage: approximately 1860 km² within the fluvial planning framework, including safeguard proposals, and 107,045 landslide phenomena occupying approximately 4970 km². Initial flood transposition retained inherited boundaries, while landslide classification followed a common type–activity matrix. Regulatory comparisons identified conditional extensions of admissibility alongside more selective urban restructuring rules; the plans also established formal updating procedures. Institutional review connected technical choices with regulatory decisions, producing amendments while leaving some precautionary disagreements unresolved. The framework coordinated inherited knowledge without eliminating differences in detail or completeness. The article offers a procedural and regulatory reference for other authorities facing fragmented planning systems; implementation and risk-reduction outcomes require subsequent evaluation.

1. Introduction

Floods and landslides threaten settlements, infrastructure and economic activities. European assessments identify substantial losses from weather and climate extremes and growing risks to infrastructure, ecosystems and economic security [1,2]. Prevention depends on both knowledge of hazards and decisions governing exposure and vulnerability.

Italy has extensive exposure to diverse flood and landslide processes. The 2025 assessment by the Italian Institute for Environmental Protection and Research (ISPRA) records more than 636,000 landslide phenomena and approximately 69,500 km² in landslide-hazard or attention areas. High and very high landslide-hazard classes cover 9.5% of the national territory. For flooding, the national mosaic based on 2020 data identifies 6.8 million residents in areas covered by the medium-hazard P2 scenario [3]. These figures establish the relevance of land-use regulation alongside structural risk-reduction measures.

The increase in risk, however, is not driven solely by changes in climate-related phenomena. Urbanization, soil sealing, occupation of flood-prone areas, slope modification, aging infrastructure, and maintenance deficiencies continually alter exposure and vulnerability. Hydrogeological risk must therefore be understood as the product of interactions between natural processes and the ways in which land is developed and transformed.

In this context, land-use planning plays a key preventive role. It can prevent new exposure, preserve the space required for flood conveyance and attenuation, protect slope stability, and govern transformations of the existing built environment. To fulfill this role, however, planning must be supported by reliable knowledge, understandable classifications, consistent rules, and clearly organized responsibilities. The availability of maps alone does not ensure that risk is incorporated into decision-making.

European water policy establishes the river basin district as a coordinating scale through Directives 2000/60/EC and 2007/60/EC [4,5]. In Italy, basin planning under Law No. 183/1989 and the subsequent framework of Hydrogeological Setting Plans (Piani di Assetto Idrogeologico, PAIs) connects hazard identification with binding land-use provisions [6,7,8].

The district-authority reform reorganized institutions without immediately replacing existing planning instruments [8,9,10]. District authorities thus brought together plans developed at basin and sub-basin scales, at different dates and with different classifications, knowledge bases and regulatory approaches. Harmonization required integrating these instruments into a common district framework.

Fragmentation affected knowledge bases, maps, regulatory frameworks, procedures, institutional responsibilities, and updating capacity. It consequently influenced both the conditions governing land transformation and the equitable application of rules to comparable physical conditions. Section 4 reconstructs the dimensions and consequences of this fragmentation.

Flood-risk governance research examines institutional arrangements, strategies and coordination [11,12,13]. Within this field, integrating pre-existing plans that remain legally operative raises a specific question: how can heterogeneous hazard knowledge be connected to common classifications and binding land-use provisions?

Harmonization must address differences in both the underlying evidence and the way it is classified. European research has documented variation in flood-mapping practices and their use in spatial planning [14], while landslide research shows how inventory methods, completeness and mapping quality shape the basis for hazard assessment [15]. These differences are also embedded in established institutions: comparative analysis of Floods Directive implementation found that national flood-risk management traditions sometimes prevailed over harmonization objectives [16]. A common classification therefore requires explicit decisions about how different sources and practices are brought together.

ISPRA’s national mosaics reconcile heterogeneous hazard representations to support national assessments and risk indicators [3,17]. District PAI harmonization addresses a further task: establishing common hazard classifications and associated land-use rules through consultation and formal adoption. This requires linking technical classification choices to institutional decisions about their regulatory consequences. By providing a common district framework at the source of the national mosaics, the adopted plans may also facilitate their subsequent compilation.

The 2023 national Policy Guidance identified district PAIs and harmonized safeguards and provisions as priorities [18]. The district draft plans pursued this objective alongside public access to territorial information, supported by the district web-based geographic information system (WebGIS) [19,20,21].

This article analyzes the development of the district Flood and Landslide PAIs prepared by the Central Apennine River Basin District Authority (AUBAC). The district covers 42,275 km², includes parts of seven regions, 22 provinces and 901 municipalities, and has approximately 8.6 million inhabitants [21]. Its inherited planning framework comprised nine PAIs. Draft plans were adopted on 31 July 2025 through Resolutions Nos. 57 and 58 of the Permanent Institutional Conference (CIP) [19,20]. Following consultation and regional institutional review, the two plans were adopted in final form on 26 May 2026 through Resolutions Nos. 67 and 68 [22,23]. The case permits examination of two hazard-specific harmonization methods within the same institutional setting.

Figure 1 situates the district geographically and illustrates the initial institutional fragmentation.

Figure 1. The Central Apennine River Basin District and the nine pre-existing PAIs. Source: prepared by the author using AUBAC WebGIS and planning information [24]. PAI boundaries and thematic layers: AUBAC. Basemap credits: Esri, USGS, TomTom, Garmin, FAO and NOAA.

This study addresses three research questions (RQ1–RQ3): (RQ1) How were heterogeneous inherited classifications harmonized while preserving the underlying knowledge and its provenance? (RQ2) How were the resulting classifications connected to common rules, procedures and responsibilities? (RQ3) What changes, disagreements and limits are demonstrated by the documentary record at adoption?

This study makes four connected contributions. Empirically, it documents how classifications and land-use provisions changed and how institutional review led to amendments, clarifications or retained disagreement. Methodologically, it combines document analysis and geographic information system (GIS) comparisons to trace the relationships between source information, district classifications and regulatory provisions. Theoretically, it explains how a common classification can accommodate differences in the underlying evidence and applicable rules while leaving some disagreements over precaution unresolved. In practical terms, it documents how plan preparation was organized through classification harmonization, progressive regulatory drafting, coordinated technical and institutional review, and differentiated handling of observations. Other planning authorities can draw on these procedural choices and regulatory solutions, adapting them to their own data, legal frameworks and administrative capacities. Six analytical functions organize the reconstruction of this process and make its decision points explicit.

Section 2 defines the analytical framework; Section 3 explains the methods; Section 4 and Section 5 reconstruct the inherited system and harmonization operations; Section 6 presents the findings; Section 7 discusses their implications, transferability and limitations; and Section 8 concludes with this study’s contributions and directions for future research.

2. Theoretical Framework

Harmonizing hazard plans requires decisions about which information to retain, how to classify it and what land-use rules to associate with each class. The theoretical framework examines how these decisions are connected and how they can be reconsidered when objections or new evidence arise.

Three bodies of research inform this analysis. Classification and knowledge-infrastructure research helps explain how information from different sources is organized into shared categories while retaining differences in its origin and meaning. Policy integration research addresses how these categories are connected to land-use provisions, administrative procedures and institutional responsibilities. Adaptive planning research examines how planning arrangements allow decisions to be revisited as knowledge and conditions change.

These perspectives guide the examination of three observable relationships: how the district classifications incorporate inherited knowledge; how those classifications shape the conditions under which land-use interventions are permitted; and how institutional review addresses disagreement and additional evidence. The analytical framework summarized at the end of this section links these relationships to the theoretical concepts and documentary evidence used in the analysis.

2.1. Classification and Regulatory Knowledge

Hazard knowledge is produced and maintained through the work of technical experts, public organizations, information systems and shared standards. Knowledge-infrastructure research examines how these elements operate together [25]. In land-use planning, their relationship matters because the information they produce supports decisions about where and under what conditions development may occur. Research on spatially integrated policy infrastructure provides a basis for examining how this knowledge is connected to planning decisions [26].

Classification is a central part of that connection. It organizes diverse observations into categories that make particular conditions visible and relevant to decisions [27]. When plans use different sources and criteria, establishing a common classification requires choices about which distinctions to retain and how to represent them. Shared categories can then serve as “boundary objects”: common references that different professional and institutional communities use together, even when their interpretations or priorities differ [28]. Agreement to use a category therefore need not imply agreement on all its regulatory consequences.

This study uses the term public regulatory knowledge infrastructure to describe the institutional arrangement connecting documented hazard knowledge to official classifications, land-use provisions and formal procedures for revision. The concept focuses on the decisions linking these elements: how source information supports an official class, how provisions determine the interventions admissible within that class, and how objections or new evidence can lead to reconsideration. It allows integration to be examined through these relationships, including where differences or disagreements persist.

Three expectations guide the analysis. First, a common classification may bring together information from different sources without making the underlying evidence equally detailed or reliable. Second, a category may remain a shared reference even when its precautionary adequacy is contested. Third, the rules applicable within a class may be differentiated without changing the class itself. Classification crosswalks, adopted provisions and documented responses to observations are used to examine these expectations in Section 6 and Section 7.

2.2. Integration, Land Use and Revision

Policy integration concerns how institutions coordinate their understanding of a problem, the actors involved and the instruments used to address it [29]. In basin planning, this coordination must connect territories that are physically linked by water flows and slope processes but governed by different administrations [30,31]. A district framework can establish common classification criteria and regulatory provisions, while regional and local authorities contribute territorial knowledge and exercise their respective responsibilities. Integration therefore depends on how these roles work together and how decisions account for effects beyond administrative boundaries.

Land-use planning gives this coordination practical consequences by determining where development may occur and under what conditions existing settlements may be transformed. Research on spatial planning and flood-risk management examines this connection [26,32,33], while landslide-zoning guidelines relate hazard assessment to land-use planning [34]. Italian research on dispersed settlement patterns highlights the importance of these decisions in territories where buildings and infrastructure extend across large areas [35,36]. New development, transformation of existing buildings and mitigation works raise different planning questions: preventing additional exposure, adapting existing settlements and reducing existing hazards or vulnerabilities. These different purposes help explain why a common hazard classification must be connected to provisions tailored to the intervention concerned.

These decisions also depend on how institutions assess uncertain or incomplete knowledge [37]. An inventory entry, a detailed study and an officially classified area may provide different kinds of evidence and have different regulatory status. The analysis therefore examines how planning procedures handle missing attributes and additional studies, including the conditions under which new information can support a change in classification or applicable provisions.

Adaptive planning extends this question over time by examining how decisions can be revised as knowledge and conditions change [38,39,40]. For PAIs, this involves procedures for assessing additional evidence, updating official maps and amending rules. This study examines both the revision procedures established by the plans and their use where documented. Recursion refers here to a documented reconsideration of an earlier classification or regulatory choice in response to objections or additional evidence. Formal updating provides a procedure for revising maps or provisions as new knowledge becomes available. Table 1 summarizes this analytical framework.

Table 1. Analytical framework of the study.

Concept	Documentary Indicator	Analytical Use
Classification and knowledge infrastructure [25,27,28]	Flood crosswalk; Tiber Basin Hydrogeological Setting Sub-Plan (Piano Stralcio 6, PS6) transposition; landslide type–activity matrix; source definitions and attributes	Examine how common categories bring together information from different sources while retaining its provenance
Policy integration and land-use planning [26,29,34]	Technical Implementation Provisions (NTA) on admissibility, assessment procedures and institutional responsibilities	Trace how hazard classes are connected to conditions for land-use interventions and to decisions by the competent authorities
Multilevel governance [30,31] and adaptive planning [38,39,40]	Issue–response chains; technical and institutional review records; formal updating provisions	Examine how objections and additional evidence lead to reconsideration, amendment, retained disagreement or referral to formal updating

3. Research Design and Methodology

3.1. Research Strategy and Units of Analysis

This study adopts a single-case design, with the Central Apennine district planning process as the case and flood and landslide harmonization as two embedded units of analysis [41]. Comparing the two units within the same institutional setting makes it possible to distinguish hazard-specific choices about source information and classification from shared arrangements for regulatory drafting, institutional review and formal updating.

The case was selected because it brings together nine pre-existing PAIs across seven regions and documents their integration into two coordinated district plans. It provides evidence of how different technical methods were developed within a common planning process and how classification and regulatory choices were examined through consultation and institutional review.

The comparison is used to develop analytical propositions about how harmonization works and to identify procedural choices that other planning authorities can assess and adapt. Their transferability depends on the available knowledge, legal framework and administrative capacity.

3.2. Temporal Scope and Sources

The reconstruction of plan formation covers the period from the Policy Guidance of July 2023 [18] to final adoption on 26 May 2026 [22,23]. Earlier planning documents establish the inherited baseline; the draft General Reports document the initial district proposals [19,20], while the final reports and NTA establish the adopted framework [42,43]. A complementary spatial analysis uses datasets obtained on 3 October 2026 to describe hazard extent and examine correspondences between the preceding and district planning systems.

The primary research technique is document analysis: a systematic examination and interpretation of sources that considers their provenance, purpose and function [44]. The corpus comprises planning reports, maps, appendices, resolutions, procedural records and Technical Implementation Provisions (Norme Tecniche di Attuazione, NTA). Table 2 summarizes the source groups and their role in the analysis.

Table 2. Sources and their functions in the analysis.

Document Group	Content	Function in the Analysis
Pre-existing PAIs	Planning documents and reconstruction of the pre-existing PAIs in official comparative documentation [24,42,43]; source regulatory provisions used in the selected comparisons [45,46,47]	Identification of the initial heterogeneity
2025 draft plans	Reports, maps, matrices and NTA adopted at the draft-plan stage, documented in the flood and landslide General Reports [19,20]	Analysis of the initial district-wide proposal
2026 PAIs adopted in final form	General reports, technical annexes, maps and NTA [42,43]; comparative landslide appendix [24]; final-adoption resolutions [22,23]	Analysis of the framework adopted in final form
Procedural records	Explanatory report [48] and Operational Conference minutes [49]; public-comment assessments [50,51]; regional response reports [52]; adoption resolutions [22,23] and Official Gazette notices [53,54]	Reconstruction of the decision-making sequence and verification of procedural outputs
Quantitative summaries and GIS datasets	Official inventory summaries [24,42,43]; district hazard extracts and preceding-plan maps and safeguard layers obtained through AUBAC and its WebGIS [21], with extraction details in Appendix A	Describe hazard extent and feature counts, and quantify source-to-district correspondences
National and European planning context	The 2023 Policy Guidance [18]; European water and flood directives [4,5]; Eastern Alps planning [55], Po plans and extensions [56,57], and Northern and Southern Apennine instruments [58,59]	Identify the institutional context and conditions relevant to transferability

Sources were selected for their relevance to the preparation, amendment and adoption of the two PAIs and to comparison with the preceding plans. For each document, the issuing body, date, procedural status, subject matter and analytical function were recorded. Intermediate versions were included where they documented changes between draft preparation, consultation and final adoption. Institutional communication materials provided context; technical findings were based on the underlying planning documents and records.

The regional corpus comprises twelve reports containing regional observations and AUBAC’s reasoned responses, transmitted to the Regions on 9 March 2026 [52]. These were organized into an analytical matrix comprising 118 issue-level determinations. For descriptive aggregation, mixed determinations containing both acceptable and partially acceptable components were classified as partially accepted; the original wording was preserved in the internal verification record. The analytical matrix compiled for this study forms part of the internal verification record and is not published. Public consultation reports count submissions, whereas the analytical matrix records issue-level determinations; these units are therefore reported separately.

The spatial analysis uses AUBAC district landslide, fluvial flood and marine flood datasets exported on 3 October 2026, together with preceding-plan maps and safeguard layers retrieved from the public WebGIS on the same date. For flooding, spatial intersections compare the preceding and district categories; the hydraulic extract includes updates under Decrees Nos. 139 and 140/2026. For landslides, retained source-plan and source-class attributes allow comparison with district classifications on the mapped inventory features.

The landslide extract incorporates updates through the July 2026 Operational Conference. Its acquisition date is distinguished from this update cutoff and from the publication status of the live WebGIS. The spatial comparison describes the datasets available at the common acquisition date, with differences in update coverage and unmatched areas recorded explicitly. Initial harmonization is reconstructed through the documented classification rules and adoption-stage sources, while current spatial correspondences may also reflect subsequent studies and updates.

3.3. Analytical Procedure

The analysis combined documentary comparison, reconstruction of selected review cases and spatial calculations. These established how classifications and provisions changed, how observations were assessed, and the territorial extent of current source-to-district relationships. Table 3 specifies the information extracted and the operation performed for each component.

Table 3. Information extracted and operations performed in the analysis.

Analytical Component	Information Extracted	Operation and Output
Flood classification	Source-zone label; return period or scenario; meaning of the mapped boundary; corresponding district class	Compare source definitions with district correspondence rules to identify retained categories, mergers and gaps in scenario coverage
Landslide classification	Phenomenon type; activity state; source classification criteria; district matrix assignment; missing attributes	Compare assignments for the same documented type–activity combinations and identify information requiring further characterization
Regulatory provisions	Intervention type; admissibility; required studies and opinions; competent authority; territorial qualifications, including the Integral Protection Zone (TI), marginal areas and transitional P3 under Article 26(4)	Compare selected previous and adopted provisions for the same intervention, identifying changes in conditions and review responsibilities
Institutional review	Subject and grounds of the observation; institutional response; relevant adopted provision or updating requirement	Trace issue–response chains and distinguish verified amendments, clarifications, retained disagreement and referral to further assessment
Transition and updating	Safeguard provisions; relationship between preceding and district regimes; requirements, competent bodies and procedures for map updates and regulatory amendments	Distinguish adoption from legal effectiveness and identify how additional evidence can lead to formal changes in maps or provisions
Spatial coverage and correspondence	Geometry and district class; landslide source-plan and source-code attributes; preceding flood categories and safeguard overlays	Calculate geometric unions, feature counts and flood-layer intersections; report coverage and current correspondences separately

Source: author’s analytical protocol applied to the sources in Section 3.2. Classification information: Appendix 1 to the Landslide General Report [24] and district planning documents [42,43]; previous regulatory provisions: [45,46,47]; review evidence: regional response reports [52] and public reports [50,51]; transition and updating: adoption resolutions [22,23] and NTA [42,43].

Documentary information was organized in a comparison matrix by source/version, hazard domain, analytical dimension and issue–response chain. Each entry retained the relevant article, page or response item. Identical labels were not assumed to represent equivalent physical conditions across plans, and hazard and risk classifications were kept distinct.

An ex post comparison applied source criteria and the district matrix to the same documented combinations of quiescent and inactive slides. Source assignments came from the Marche and Tiber schemes reconstructed in Appendix 1 to the Landslide General Report [24]; district assignments came from Annex 1 to the Landslide NTA [43]. Holding type and activity constant isolated differences between assignment rules. This comparison was conducted for the study and was not presented as an option appraisal undertaken during plan preparation.

Regulatory comparisons used the intervention type as the common basis for reading selected previous and adopted provisions. The analysis distinguished changes in class labels from changes in admissibility, assessment requirements and competent authorities, taking account of territorial and transitional qualifications. It examined adopted rules rather than project authorizations or realized transformations.

Process-tracing literature informed the reconstruction of sequences and documentary links [60]. Selected cases linked observations from regional response reports [52] and public reports [50,51] to institutional assessments and relevant adopted texts. Cases examined amendment with a retained class, unresolved precautionary disagreement and conditional incorporation of knowledge. A source-extraction correction documented in Appendix 1 to the Landslide General Report was considered separately as evidence of source verification [24]. A favorable response was not counted as an implemented amendment unless the change was verified in the adopted text; correspondence between an observation and final wording did not establish exclusive causal attribution.

The 118 regional determinations were aggregated using the categories in Section 3.2. Selected cases explain mechanisms rather than estimate their frequency. Public-consultation reports and regional response documents are identified separately in the case descriptions and references. The author checked reported documentary entries against source texts and numerical results against the analytical datasets. No inter-coder reliability statistic was calculated because coding was undertaken by a single researcher.

Spatial calculations used ETRS89/UTM zone 33N (EPSG:25833). Polygons were dissolved within each class to calculate class coverage and across classes to calculate total coverage. Class areas may overlap and are not necessarily additive; each total union counts overlapping territory once. Fluvial flooding, landslides and marine flooding were analyzed separately. Counts refer to inventory features; coverage percentages use the approximately 42,275 km² district as the denominator. District-scale areas are generally rounded to the nearest 10 km², and smaller areas and correspondences to the nearest km².

For landslides, features were grouped by recorded source plan, source code and district class, with counts and union areas calculated for each group. For flooding, geometric unions of preceding-plan categories were intersected with district categories, keeping approved mapping and safeguard overlays distinct. The Misa and Pescara updates were already included in the district adoption layer. These calculations describe current correspondences, which may include subsequent updates; documentary rules establish initial harmonization. Appendix A reports the complete aggregated provenance and spatial correspondence matrices and summarizes the extraction procedures; geometry checks and detailed extraction records are retained in the internal verification record.

Previous comparisons of landslide inventories identify mapping methods, scale and source information as factors affecting comparability [61,62]. These considerations informed interpretation of recorded correspondences rather than calculation of their areas.

The reconstruction distinguished six functions according to the decision addressed and the link between input and output. Inventory organized source information into a provenance record; classification correspondence related source attributes to district categories; regulatory design connected categories to admissibility provisions; validation linked proposals and objections to institutional assessments; transition defined the relationship between preceding and adopted instruments; and updating established formal routes for incorporating additional evidence. Inventory and classification were reconstructed from the planning reports and Appendix 1 to the Landslide General Report [24,42,43]; regulatory design, transition and updating from the NTA [42,43]; and validation from regional response reports and public-consultation reports [50,51,52]. These functions may overlap and are not separate chronological stages.

Findings were organized by research question: classification and spatial correspondences for RQ1; provisions, procedures and responsibilities for RQ2; and review responses for RQ3. Comparison with the theoretical expectations in Section 2 informed the analytical propositions in Section 7.1. Transferability was examined by separating source-comparison methods and organizational arrangements from Italian statutory requirements and hazard-specific criteria. Other Italian planning pathways provided context rather than an independent test of effectiveness. Figure 2 summarizes the analytical workflow.

Figure 2. Research workflow from source and version identification to interpretation. Solid arrows indicate the main analytical sequence; dashed arrows indicate analytical and source-verification iterations. The embedded comparison includes GIS coverage calculations and source-to-district class correspondence checks.

The spatial calculations were implemented in custom Python scripts using Shapely and verified using Python 3.12.14 and Shapely 2.2.0 (GEOS 3.14.1).

During the preparation and revision of this manuscript, the author used ChatGPT (OpenAI; GPT-5 series for initial preparation and GPT-6 Astra for revision) and Claude (Anthropic; Opus 5.5) primarily to support English-language refinement and improve the clarity and organization of the author’s text, with additional assistance in figure preparation and the writing and checking of GIS analysis scripts. The author critically reviewed, revised or discarded AI suggestions, verified all retained content, and takes full responsibility for the published article.

3.4. Outputs, Outcomes, and Impacts

The analysis distinguishes planning outputs, implementation outcomes and risk impacts. Outputs include the adopted classifications, regulatory provisions and procedures, together with the mapped areas to which the compared provisions relate. Outcomes concern how these provisions are applied through assessments, authorizations and actual land-use changes. Impacts concern changes in exposure, vulnerability and risk. This study documents planning outputs and their territorial extent; outcomes and impacts require evidence of subsequent implementation.

The territorial extent of the regulatory comparison is identified by linking documented classification rules and current spatial correspondences to the preceding and adopted provisions, taking account of applicable territorial and transitional regimes. Where an adopted provision allows reconstruction subject to a favorable compatibility assessment, that assessment is a condition for authorizing individual projects within the area governed by the provision. The area measure therefore captures the territorial scope of this conditional pathway, irrespective of whether applications have already been submitted or approved. It does not count eligible buildings or completed transformations. Current spatial correspondences are interpreted separately from the initial harmonization where subsequent studies have revised hazard mapping.

3.5. Source Verification

Reported interpretations were checked against primary documents and successive versions, distinguishing proposed changes, institutional responses and adopted provisions. Contextual knowledge of the planning process (see Conflicts of Interest) was used to support reconstruction but was distinguished from the documentary and spatial evidence supporting the findings. The analysis reports non-accepted proposals and unresolved disagreements alongside documented amendments and differences in regulatory treatment.

4. The Initial System: Nine PAIs and Fragmented Planning

At the start of district plan preparation, AUBAC managed a planning system developed over more than two decades by former national, interregional and regional basin authorities. It comprised nine PAIs: Abruzzo–Sangro, Fiora, Lazio Regional Basins, Liri–Garigliano, Marecchia–Conca, Marche Regional Basins, Ombrone, Tiber and Tronto [42,43]. Each brought its own knowledge base, maps, regulatory provisions and procedures, reflecting different planning histories and territorial contexts.

Integrating these instruments required coordination across seven regions and multiple levels of administration. The institutional setting combines AUBAC’s technical bodies, the Operational Conference, the Permanent Institutional Conference and regional programmatic conferences under Article 68 of Legislative Decree No. 152/2006 [8]. The latter involve provinces, municipalities and basin authorities in examining the draft plans and their relationship to territorial planning. These forums supported the examination of technical choices and their regulatory implications.

The starting point was therefore an extensive but heterogeneous planning system. Harmonization required identifying which knowledge could be retained, how existing classifications could be related to common district categories, and which provisions and updating procedures needed coordination. The following subsections examine these differences and their consequences for planning.

4.1. Knowledge-Base and Cartographic Heterogeneity

The pre-existing landslide plans drew on regional inventories, field surveys, geological and geotechnical studies, and monitoring activities conducted at different times and using different methods. Their maps represented different aspects of instability, including hazard, risk and susceptibility, sometimes within combined schemes. The phenomena and features recorded also varied: alongside landslides, they could include badlands, scarps, cavities, sinkholes, avalanches, point features and attention areas. Terminology for activity states differed between plans [43].

These differences extended to the meaning of mapped boundaries and classes. Polygons could delineate an existing instability, its potential evolution, a surrounding buffer or an area combining several phenomena. Mapping scales and levels of detail also varied [43]. Hazard scales reached five levels in some landslide plans. In certain plans, polygon symbology represented risk, while hazard was recorded separately in accompanying lists or labels [24,43]. Comparing these maps therefore required examining both what each polygon represented and how its attributes had been classified.

For flooding, heterogeneity concerned the scenarios and return periods used to delineate areas, the terminology of flood zones and the relationship between hazard and risk. The plans used categories such as A, B and C; A1–A2 and B1–B2; E1–E4; and P.I.2–P.I.4, while some represented a single floodable zone. These schemes were not directly comparable, and not all followed the three probability scenarios used in the Floods Directive [42]. Establishing district correspondences therefore required relating each source category to its underlying scenario and definition.

4.2. Regulatory, Procedural, and Institutional Heterogeneity

Differences between the pre-existing PAIs extended from mapping to the rules governing land use. Their NTA specified different protection thresholds, permitted and prohibited interventions, study requirements, assessment procedures and competent authorities. Similar interventions could follow different administrative pathways. Exceptions for existing settlements, residual plots, agricultural areas, infrastructure and works without feasible alternative locations also varied.

These differences reflected the institutional arrangements of the former national, interregional and regional basin authorities. Each plan allocated responsibilities and established procedures for incorporating new studies, monitoring results and cartographic changes. At district scale, these inherited arrangements provided no common procedure for assessing additional knowledge and incorporating it into updated maps and provisions.

Table 4 summarizes these dimensions and their implications for comparing plans and coordinating decisions.

Table 4. Dimensions of fragmentation among the pre-existing PAIs.

Dimension	Main Differences	Implication for District Planning
Knowledge base	Different sources, survey dates, levels of detail and assessment methods	Establish the coverage and comparability of the available evidence
Cartographic	Different mapped variables, classification systems, scales and delineation criteria	Interpret source categories and differences in mapping across former PAI boundaries
Regulatory	Different protection thresholds, admissibility conditions and exceptions	Compare the treatment of similar interventions under different plans
Procedural	Different study requirements, assessment steps and administrative decisions	Coordinate the evidence and procedures required to assess interventions
Institutional	Different allocations of responsibilities among competent authorities	Clarify responsibilities and coordination across administrative levels
Updating	Different procedures for incorporating studies, monitoring results and map revisions	Establish common requirements for assessing and incorporating new knowledge

Source: prepared by the author based on the documentation of the pre-existing PAIs and the general reports of the two district PAIs [42,43].

The inherited system therefore posed two linked problems: establishing correspondences between classifications based on different sources and criteria, and coordinating the provisions and procedures associated with them. Comparable physical conditions could receive different regulatory treatment under different plans. Classification harmonization provided a basis for comparison, while regulatory coordination required separate decisions about admissibility, assessment and responsibilities. The following section reconstructs how these tasks were addressed.

5. Construction of the District Framework

Preparation of the district PAIs connected three tasks: reviewing the inherited knowledge and classifications, developing common regulatory provisions, and examining both through technical and institutional review. Flood and landslide specialists used different classification methods within a shared coordination process. Section 6 presents the resulting correspondences, regulatory comparisons and documented responses to observations.

5.1. Organizing Inherited Knowledge into District Classes

The initial review identified the source plans, the phenomena and areas represented, their classification criteria and the supporting information. This established the basis for relating existing categories to district classes while preserving the provenance and level of detail of the source knowledge [24,42,43].

Preparation progressed from harmonizing hazard representations and classes to developing provisions for permitted, conditional and prohibited interventions, followed by coordination of the NTA as a whole. Separate flood and geomorphological groups worked within this common process. The AUBAC–Regions Technical Working Group provided a forum for examining technical choices and their regulatory implications [48].

For flooding, the initial transposition retained inherited flood-zone boundaries and assigned them to the common P1–P3 system. Correspondences were established by examining the definition and mapped meaning of each source category. The district return-period ranges are up to 50 years for P3, over 50 to 200 years for P2, and over 200 to 500 years for P1 [42,63]. Because the preceding plans used different thresholds, categories assigned to the same district class could retain differences in their underlying flood scenarios. New delineations based on additional studies were addressed through separate assessment and updating operations.

For landslides, classification used information on phenomenon type and activity state through the common district matrix, with incomplete characterization handled separately. Preparing the district inventory also required data extraction, replacement of some source inventories and corrections documented in Appendix 1 to the Landslide General Report [24,43]. The work therefore combined classification harmonization with the documented integration and revision of the underlying inventories.

These two methods established the classification framework used to develop the district provisions. The flood crosswalk and landslide type–activity matrix are presented in Section 6.1.

5.2. Regulatory Design and Institutional Review

Building on the classification framework, regulatory drafting addressed the conditions for permitted interventions, the studies and assessments required, and the responsibilities of the competent authorities. The provisions also defined the relationship with existing planning instruments and the procedures for subsequent updates [42,43].

Following draft adoption on 31 July 2025, public consultation and regional institutional review examined the proposed classifications and provisions before final adoption on 26 May 2026. The public response reports address 133 flood submissions and 42 landslide submissions, each potentially containing several issues [50,51]. Regional observations were examined through a separate review process, documented in twelve reasoned response reports transmitted on 9 March 2026 [52].

The review addressed proposals to amend regulatory provisions, objections to classification choices and requests supported by additional territorial information. Depending on their subject and supporting evidence, issues were addressed through amendments, clarifications, retention of the proposed approach or referral to further technical assessment and formal updating. Section 6.3 examines these responses and their relationship to the adopted texts, including cases in which disagreement remained unresolved. Appendix B documents the procedural sequence, the common NTA architecture, the graded review pathways and the distribution of functions among AUBAC, the Regions and municipalities.

5.3. Adoption and Revision Routes

Final adoption was accompanied by safeguard measures limited to the provisions and maps enumerated in CIP Resolutions 67 and 68 [22,23]. Publication of the notices on 4 June 2026 activated those measures; full plan effectiveness remains distinct from this selected safeguard regime [53,54].

The adopted framework provides dedicated technical and participatory procedures for map updates under Flood Article 22 and Landslide Article 20 [42,43]. During the transition, Article 4 of both adoption resolutions retains the preceding-plan procedures for updates initiated before the district plans enter into force, subject to safeguards; their results also update the district maps [22,23]. New evidence therefore requires the applicable formal procedure to change official delineations or classifications. Regulatory amendments follow a distinct procedure.

Separating map updates from amendments to the general provisions combines regulatory continuity with an evolving knowledge base. Maps can be revised through technical assessment, public notice, participation and agreement with the competent Region, without reopening the general regulatory framework each time. Changes in the territorial application of the rules therefore follow formal map revisions, rather than arising automatically from the availability of new data [42,43].

These updating procedures are supported by common technical references for subsequent studies: the hydrological–hydraulic guidance in Annex 2 to the Flood NTA and the landslide classification and survey requirements in Annexes 1 and 5 to the Landslide NTA [42,43].

6. Results

The results examine how inherited information was organized into common hazard classes, how those classes were connected to land-use provisions, and how institutional review addressed proposed changes and disagreements. Section 6.1, Section 6.2 and Section 6.3 answer the three research questions; Section 6.4 quantifies the territorial extent of the mapped categories and their current spatial correspondences.

6.1. RQ1: Classification Harmonization and Retained Knowledge Differences

The two district plans established common hazard classifications through different operations. Flood harmonization related existing zones to the district P1–P3 system while retaining their delineations in the initial transposition. Landslide harmonization assigned P1–P4 classes using a common matrix of phenomenon type and activity state. In both domains, the resulting categories brought together information whose sources, detail and classification criteria had previously differed [24,42,43].

For flooding, Table 5 shows how each source category was related to a district class according to its definition and underlying scenario. The correspondence was not always one-to-one. In Abruzzo–Sangro, for example, the former P4 and P3 categories, associated with 50–100-year scenarios in the source framework, were both assigned to district P3 [42]. This merged two planning categories while retaining the mapped flood extent. Because source thresholds differed between plans, the common district label did not imply that all contributing areas had been delineated using the same hydraulic scenario.

Table 5. Correspondence between the hazard classes of the pre-existing instruments and the classes of the District Flood PAI for river and coastal flooding.

Pre-Existing Planning Instruments	P3—High Hazard (Tᵣ ≤ 50 Years)	P2—Medium Hazard (50 < Tᵣ ≤ 200 Years)	P1—Low Hazard (200 < Tᵣ ≤ 500 Years)
River-flood hazard
Tiber	PAI Zones A and AA (Tᵣ = 50); R4 areas in the PAI Atlas of Flood-Risk Situations; PS1 Zone A	PAI Zone B (Tᵣ = 200, including indirect and marginal Tᵣ = 50 areas); PS1 Zone B	PAI Zone C (Tᵣ = 500 and marginal Tᵣ = 200 areas)
Lazio Regional Basins	Areas A1 and A2 (Tᵣ = 30); attention areas for flood hazard outside land-reclamation areas	Areas B1 and B2 (Tᵣ = 200); attention areas within land-reclamation areas	Areas C (Tᵣ = 500); attention areas for interventions
Marche Regional Basins	—	Floodable zone (Tᵣ = 200)	—
Tronto	Areas E4 and E3 (Tᵣ = 50 and Tᵣ = 100)	Areas E2 (Tᵣ = 200)	Areas E1 (Tᵣ = 500)
Fiora	Areas P.I.4 (Tᵣ = 30)	Areas P.I.3 (Tᵣ = 200)	Areas P.I.2 (Tᵣ = 500)
Abruzzo–Sangro	Areas P4 and P3 (Tᵣ = 50–100)	Areas P2 (Tᵣ = 100, marginal)	Areas P1 (Tᵣ = 200)
Ombrone	P.I.M.E. areas	P.I.E. areas	—
Coastal-flood hazard
Abruzzo—Integrated Coastal Zone Management Plan	—	—	—
Lazio—Integrated Coastal Zone Management Plan	—	—	—
Lazio—Sectoral Coastal Erosion Plan for the Domitian Coast	Zone P3	Zone P2	Zone P1
Marche—Integrated Coastal Zone Management Plan	Zone Tᵣ = 20	Zone Tᵣ = 100	Zone Tᵣ > 100

Source: prepared by the author based on the General Report of the District Flood PAI [42]. PS1: Tiber Basin Sub-Plan for the floodable areas from Orte to Castel Giubileo (Piano Stralcio 1). The term “marginal” in inherited legends denotes source-plan categories and does not by itself establish equivalence with the district marginal areas defined in Article 7(9) and discussed in Section 6.2. The source correspondence table appears in Section 4.2 of the final Flood General Report (pp. 11–12). A dash denotes no entry in that source table and does not establish an absence of hazard. The Marche entry records correspondence with P2, with no entry under P3; the separate transitional regulatory treatment is examined in Section 6.2. Marecchia–Conca and Liri–Garigliano are absent from the river-flood section because no previously delineated flood-hazard areas requiring transposition were identified in their peripheral AUBAC portions. The coastal-flood section records correspondences for the Domitian Coast and Marche; blank entries for Abruzzo and Lazio indicate no correspondence recorded in the source table [42].

The crosswalk also reveals differences in scenario coverage. For much of the Marche river network, the available mapping represented only the 200-year floodable zone, assigned to district P2. Additional studies had introduced other scenarios in specific areas: the Misa mapping included 50- and 500-year scenarios, while low-probability scenarios were also available in parts of Abruzzo–Sangro [42]. Harmonization organized this uneven coverage within a common framework, leaving gaps where the corresponding scenarios had not been mapped.

Figure 3 shows the cartographic application of this operation in the PS6 area of the Tiber basin. The paired maps retain the illustrated delineations while expressing the inherited categories through district classes. Table 5 and Figure 3 therefore document complementary aspects of the same result: the correspondence rule and its application to existing planning maps.

Figure 3. PS6 in the Tiber basin: (a) inherited planning zones; (b) their transposition into district flood classes P3, P2 and P1 on the retained delineations. Panel (a) retains the source-layer legend; zone A is absent from the illustrated extent, which contains zones AA, B and C. The figure illustrates classificatory transposition based on inherited planning knowledge, not new hazard modeling, a GIS before/after analysis or a measured spatial effect of harmonization. Source: prepared by the author based on AUBAC [19,42]. Basemap: Regione Lazio, Carta Tecnica Regionale (CTR), licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/), displayed through the AUBAC WebGIS and overlaid with AUBAC planning layers.

For landslides, the common matrix combines twelve phenomenon types with three harmonized activity states—active, quiescent and inactive—to assign hazard classes (Table 6). Classification depends on these descriptive attributes rather than on the previous class label alone. The codes and terminology in Table 6 reproduce Annex 1 to the adopted Landslide NTA [24,43].

Table 6. Harmonized landslide hazard matrix adopted in the district PAI.

PAI Code	Landslide Type	Active (A)	Quiescent (Q)	Inactive (I)
CR	Rockfall or toppling–landslide crown zone	P4	P3	P2
SC	Rotational or translational slide	P3	P2	P1
CO	Flow	P3	P2	P1
FC	Complex landslide	P3	P2	P1
FS	Area affected by diffuse shallow landslides	P3	P2	P1
DG	Area affected by deep-seated gravitational slope deformation (DGPV in Italian)	P3	P2	P2
SO	Area affected by slow surficial deformation or solifluction	P2	P2	P1
FA	Talus or debris cone	P2	P2	P1
DF	Debris flow	P4	P3	P2
CL	Badland or eroding area	P3	P3	P2
FR	Setback zone from the landslide crown or scarp edge	P3	P2	P2
VL	Avalanche	P4	P4	P4

Note: The hazard class derives from the combination of landslide type and activity status. A = active; Q = quiescent; I = inactive. The PAI codes reproduce those established in Annex 1 to the Technical Implementation Provisions. Source: prepared by the author based on AUBAC [43].

The applied comparison in Table 7 shows what this choice means for two documented combinations. Under the source schemes described in Appendix 1 to the Landslide General Report, quiescent slides were assigned to P2 in Marche and P3 in the Tiber plan; inactive slides were assigned to P1 and P2, respectively [24] (pp. 31, 34). Applying the district matrix assigns P2 to quiescent slides and P1 to inactive slides in both cases. Retaining the source criteria would preserve the different assignments; applying the common matrix removes that difference.

Table 7. Applied comparison of source retention and district classification for documented slide types.

Type and Activity	Source Scheme	Source Assignment Retained	District Matrix Applied
Quiescent slide	Marche	P2	P2
Quiescent slide	Tiber	P3	P2
Inactive slide	Marche	P1	P1
Inactive slide	Tiber	P2	P1

Source: Appendix 1 to the Landslide General Report [24], pp. 31 and 34, for source schemes; final Landslide NTA, Annex 1 [43], for district rotational/translational slides (SC). The comparison concerns documented assignment rules, not the frequency of every assignment in the GIS inventory or identical physical conditions at individual sites.

In these examples, convergence follows the Marche assignments rather than retaining the higher Tiber categories. Appendix 1 to the Landslide General Report [24] describes calibration of the harmonization threshold with reference to less restrictive legacy schemes, while the final NTA establish harmonized minimum protection levels and allow more restrictive territorial provisions under the applicable conditions [24,43]. The comparison thus identifies both the gain in classificatory consistency and the choice underlying it: standardization did not preserve every higher source assignment. Its precautionary adequacy remained open to institutional scrutiny, as the disagreement examined in Section 6.3 demonstrates.

The integration is also visible across adjoining planning jurisdictions. Figure 4 compares the former Tiber, Marche and Tronto representations with district hazard mapping in the same area. The Tiber representation combined hazard classes, landslide inventory information and avalanche risk. Official Marche and Tronto sheets used risk classes for polygon symbology, recording hazard separately in the accompanying event lists and in H0–H4 polygon labels, respectively. The district map represents these territories through a common hazard-classification framework, with the former PAI boundaries retained in both panels to make the comparison legible. This is integration of the underlying information into a shared representation, rather than a direct conversion of risk labels into hazard classes.

Figure 4. Landslide-related representations across the adjoining territories of the former Tiber, Marche and Tronto PAIs: (a) legacy PAI layers, including hazard, risk and inventory information; (b) district Landslide PAI hazard classification. Both panels show the same geographical extent and retain the former Tiber and Tronto PAI boundaries as reference lines. Colors and patterns in panel (a) represent different variables and are interpreted through their respective legends; panel (b) uses the common district hazard classes P1–P4. In panel (a), Marche and Tronto landslide areas are shown by risk class, as in the official PAI map sheets; their hazard levels (P1–P4 and H0–H4) were recorded in event lists and polygon labels, respectively, rather than through polygon fill colors. Source: AUBAC WebGIS, accessed 4 October 2026; composition by the author. Official map sheets consulted: Marche PAI, Risk Map Sheet RI 70d (update approved by DPCM of 14 March 2022); Tronto PAI, Sheet 10/20, Landslide and Floodable Area Map (update by Secretarial Decree No. 263/2023).

The district framework also distinguishes classified hazards from information requiring further characterization. Insufficiently characterized phenomena and potentially hazardous geological or geomorphological elements were retained separately from definitive P1–P4 assignments, with the relevant verification or updating requirements [43]. Risk classes R1–R4 were derived separately by combining hazard with potential damage, preserving the distinction between the characteristics of the phenomenon and its consequences for exposed elements [24] (p. 46).

The current technical inventory contains 107,045 mapped landslide phenomena and covers approximately 4970 km², or 11.8% of the district (Table 8). P2 accounts for 70,176 phenomena, or 65.6% of the total, and approximately 2990 km². P3 contains 27,276 phenomena across approximately 1300 km², while P4 contains 4271 across approximately 180 km². The predominance of P2 is consistent with the assignment of several quiescent landslide types to that class under the district matrix. Table 8 describes the inventory composition, while Table 6 and Table 7 explain the relevant assignment criteria.

Table 8. Landslide inventory by hazard class and approximate share of district territory, based on the extract obtained on 3 October 2026.

Class	Mapped Phenomena	Share of Phenomena (%)	Area (Approx. km2)	District Area (%)
P4	4271	4.0	180	0.4
P3	27,276	25.5	1300	3.1
P2	70,176	65.6	2990	7.1
P1	5322	5.0	510	1.2
Total	107,045	100.0	4970	11.8

Source: AUBAC landslide inventory extract obtained on 3 October 2026, incorporating updates through the July 2026 Operational Conference. Areas are approximate polygon-union areas; percentages use unrounded values and a district area of approximately 42,275 km². Shares of mapped phenomena may not sum to 100% because of rounding. The total mapped area counts overlapping territory only once. Individual class areas may overlap and therefore are not strictly additive. Counts describe mapped inventory phenomena, not separate triggering events.

The principal result is a common classification framework in which source differences remain traceable. Flood correspondence preserves the relationship to inherited zones and scenarios; landslide classification relates recorded types and activity states to shared assignment rules. The resulting maps can be read through common district categories while retaining information about differences in source coverage, characterization and detail.

6.2. RQ2: Regulatory Consequences of Harmonization

The adopted district framework connects common hazard classes to provisions governing interventions, assessment procedures and institutional responsibilities. The comparison identifies changes in both directions: some provisions broaden conditional admissibility, while others establish more selective conditions. Their consequences depend on the preceding rules, the intervention considered and the applicable territorial regime.

Article 6 of both sets of NTA regulates the relationship with urban and spatial planning according to a principle of hierarchical coordination. The binding force and precedence of landscape plans for landscape protection remain unaffected under Legislative Decree No. 42/2004 [64]. In the event of conflict, the PAI takes precedence within its remit over socioeconomic development plans and programs and over instruments governing spatial organization and land use. More restrictive regional and local provisions remain applicable, provided that they are consistent with and do not conflict with the district framework [42,43].

The common provisions connect admissibility to prevention of additional hazard and risk, preservation of flood conveyance and attenuation, protection of slope stability and control of new exposure [42,43]. Review requirements vary with both hazard class and intervention type. The Flood NTA distinguish interventions requiring an AUBAC opinion and assessment by the competent hydraulic authority, assessment by that authority alone, or no specific PAI opinion. The Landslide NTA distinguish requirements for AUBAC and regional opinions, separately or both, and cases requiring no PAI opinion. These exemptions do not remove other applicable technical and authorization requirements. Under Landslide Article 8, a Technical Verification (“Verifica Tecnica”) supports assessment of compatibility; only specified interventions meeting the conditions of paragraph 2 may instead use a certified statement of no aggravation (“Attestazione asseverata di non aggravamento”) [42,43].

Table 9 examines these changes through selected provisions from the Marche regional-basin PAI [45], Tiber PAI [46] and Abruzzo–Sangro Flood Defence Sub-Plan (PSDA) [47], following the classification correspondences in Table 5 and the adopted district provisions [42]. The cases illustrate conditional expansion of admissibility and greater selectivity. They compare adopted provisions; the distinction between adoption and legal effectiveness is set out in Section 5.3.

Table 9. Selected comparisons of inherited and adopted flood-planning provisions.

Source Regime and Intervention	Previous Provision	Adopted District Provision and Consequence
Marche: building renovation	Art. 9(1)(c): renovation without volume increase, except specified adjustments; compatibility verification under Art. 9(2) [45].	P3 Art. 9(7)(c) and P2 Art. 10(7)(d): conditional vertical extensions, subject to paragraph 4 hydraulic safety conditions. P3 permits no increase in urban-development load; P2 permits an increase of up to 15% under the specified reconstruction conditions. Apply Art. 26(4) where relevant [42].
Tiber B → P2: urban restructuring	Art. 29(2)(b), (3): admissible subject to hydraulic conditions [46].	Art. 10(1)(d), (7)(e): specified pre-existing planning provisions for residual plots in built-up A, B and D zones and general-interest facilities in F zones, with Annex 2 studies. Separate conditional route under Art. 6(8). More selective conditions for admissibility [42].
Abruzzo–Sangro P4/P3 → P3: reconstruction	P4 Art. 18(1)(a), (h): demolition without reconstruction and restricted repair. P3 Art. 20(1)(b): reconstruction with unchanged volume and shape, access and first floor more than 1 m above ground level [47].	Art. 9(7)(c): conditional reconstruction. Expansion relative to former P4. For former P3, changed conditions: vertical extensions permitted without additional enclosed ground-floor coverage or increased urban-development load. Subject to TI and other applicable restrictions [42].

Note: The table compares previous and adopted provisions. Their territorial scope can be quantified by linking the relevant source and district categories to mapped areas. Conditional admissibility remains subject to the prescribed studies and assessments for individual projects. The arrow (→) indicates the source-to-district classification correspondence; it does not denote a temporal change in physical hazard. The comparison of adopted rules is distinct from their legal effectiveness under the safeguard and approval arrangements described in Section 5.3; territorial and transitional qualifications remain relevant. For the Abruzzo–Sangro comparison, the calculated intersection is 63.98 km²: the approved preceding-plan P4 group in the PSDA layer retrieved on 3 October 2026 intersected with consolidated district P3. Separately identified safeguard overlays are excluded.

The Marche case is informative because source terminology, cartographic correspondence and regulatory treatment diverge. Article 8 of the previous NTA associated a single high–very high hazard level with the delineated floodable areas, while distinguishing four risk levels. Article 9(1)(c) admitted statutory building renovation without volume increases, apart from specified adjustments; Article 9(2) required technical compatibility verification [45]. The district crosswalk places the inherited 200-year zone under P2, but Article 26(4) prescribes the P3 regime for specified morphologically delineated single-zone areas identified by map hatching. Subsequent assessment need not assign those areas to P2: the applicable class must follow the approved assessment and updating procedure [42].

The adopted P3 and P2 provisions both expressly address reconstruction and permit conditional vertical extensions without additional enclosed ground-floor coverage. Their conditions differ: Article 9(7)(c) requires no increase in urban-development load, whereas Article 10(7)(d) allows a conditional increase of up to 15% within the specified reconstruction provision, with unchanged use and the required compatibility findings. This is a load threshold, not an automatic volume bonus.

Both reconstruction provisions require compliance with the hydraulic safety conditions in paragraph 4 of the respective article, including a first-floor elevation above the 200-year flood level plus the prescribed safety margin. P3 reconstruction excludes basement and semi-basement rooms. In P2, their retention or reinstatement is allowed only for existing lawful rooms in reconstruction for seismic improvement or upgrading, with unchanged dimensions, non-residential and non-commercial use, no continuous occupancy, flood protection and no increase in exposure or vulnerability. The Marche comparison thus shows how the opportunities for adapting existing buildings depend on the applicable regime, including the transitional P3 treatment [42].

The Tiber band B case shows a different direction of change. Former Article 29(2)(b) admitted urban restructuring subject to the hydraulic conditions in Article 29(3). District Article 10(1)(d) restricts this category to the cases in Article 10(7)(e). These concern specified pre-existing planning provisions for residual plots in wholly or partly built-up A, B and D zones and facilities of general interest in F zones, supported by studies conforming to Annex 2. Article 6(8) assigns the Regions a separate conditional route for qualifying pre-existing planning provisions, without introducing additional building entitlement [42,46].

The Abruzzo–Sangro merger illustrates why one common class can have different consequences according to provenance. Previous P4 provisions listed demolition without reconstruction and limited repair of damaged buildings; reconstruction was expressly contemplated under conditions in previous P3. Both source categories enter district P3, whose Article 9(7)(c) explicitly provides for conditional reconstruction. The new ordinary provision therefore extends the listed building interventions relative to the previous P4 provision. For former P3, the change concerns conditions rather than first admissibility. Article 20(1)(b) required unchanged volume and shape, with access and the first floor more than 1 m above ground level. District Article 9(7)(c) permits vertical extensions without additional enclosed ground-floor coverage or increased urban-development load. This comparison concerns the reconstruction provision; former Article 20 also contained separate provisions for specified extensions and built-up centers. Site applicability still depends on TI and other relevant restrictions [42,47].

The spatial comparison gives the Abruzzo–Sangro case a measurable territorial extent: approximately 64 km² of approved preceding-plan P4 mapping correspond to consolidated district P3. This is the area addressed by the comparison between the previous P4 reconstruction provision and the adopted district pathway, independently of whether individual projects have been proposed or authorized.

Regulatory treatment also depends on territorial qualifications within the mapped hazard classes. The Integral Protection Zone (TI) lies within mapped P1–P3 areas; its Article 8 regime does not apply in urban zones A and B, where the hazard-band provisions and applicable hydraulic checks remain applicable [42].

Marginal areas link differentiated regulatory treatment to flood intensity within the common probability-based classes. Article 7(9) identifies portions of P3 and P2 where water depth and flow velocity fall within Domain 1 (“Ambito 1”) of the Annex 2 depth–velocity graph, evaluated for the 50-year and 200-year scenarios, respectively. Delineation also considers the expected hydraulic dynamics and the surrounding morphological and territorial context. Within formally delineated P3 marginal areas, the P2 provisions apply to the regulation of interventions; within P2 marginal areas, specified new construction and expansion may be admitted under the conditions established by the NTA [42]. These provisions retain the mapped hazard class while differentiating regulatory treatment according to more detailed hydraulic information.

For inherited mapping, Article 7(9) provides for progressive delineation where detailed hydraulic information consistent with Annex 2 is available. Until the first formal cartographic identification, the specific provisions dependent on marginal-area status are not operative. Delineation follows Article 22(5); the graph cannot independently establish marginal-area status through an individual project assessment [42].

The preventive framework also extends beyond mapped hazard areas. Flood Article 16 applies hydraulic invariance to territorial transformations wherever located, requiring that peak stormwater discharge to receiving water bodies does not increase. Article 17 includes integrated ecosystem and biodiversity restoration among mitigation measures. The Landslide NTA identify limiting land take and promoting the renaturation of urban and periurban soils as objectives for prevention and vulnerability reduction [42,43].

The principal result is that common hazard classes support a coordinated regulatory framework with differentiated conditions for intervention. Compared with the preceding plans, the adopted provisions introduce both conditional extensions of admissibility and more selective requirements. The applicable treatment depends on the intervention, territorial qualifications and transitional provisions, so the direction of regulatory change cannot be inferred from the class label alone.

6.3. RQ3: Amendment, Disagreement and Knowledge Incorporation

The documentary record shows that review produced different responses according to the issue examined: regulatory provisions were revised, classification choices were retained despite objections, and additional evidence was directed to technical assessment and formal updating. These differences reveal how institutional coordination addressed proposals without requiring agreement on every classification or regulatory consequence.

This study’s analytical matrix classifies 59 of the 118 first-round regional determinations as accepted, 46 as partially accepted and 13 as not accepted, based on AUBAC’s reasoned responses [52]. The distribution demonstrates differentiated assessment, including explicit non-acceptance. These classifications record the response to each issue; amendments are identified separately through comparison with the adopted documents.

A public example concerns the joint professional submission under Protocol No. 14460/2025 of 20 November 2025. The objection to merging A1 and A2 into P3 led to a response that retained the common class while reporting changes to definitions, differentiated admissibility and the role of detailed studies (item 87(1), p. 106). The response also identified methodological clarifications concerning the proponent’s study and compatibility verification, the hydraulic authority’s evaluation and the definition of marginal areas (item 87(4), p. 108). However, the proposal to assess interventions within Domain 2 (“Ambito 2”) of the depth–velocity graph as admissible subject to mitigation and compensation was not accepted (item 87(5), p. 109) [50].

The adopted Annex 2 corroborates the distinction of technical roles at p. 70 and the criteria for marginal-area delineation at p. 76 [42]. Read together, the response report and adopted provisions document reconsideration of the relationship between a retained class and its regulatory treatment. They do not establish that this submission alone determined the final wording.

A different regulatory response concerned interventions already provided for by existing planning instruments. The territorial issue was reformulated as a general district-wide provision in Article 6(8), rather than retained as an individual territorial exception [42,43,48]. In this case, review changed the scope of the response by expressing it through a provision applicable across the district.

Classification review also exposed disagreement that remained unresolved. A regional administration questioned the assignment of certain quiescent landslide types to P2 as insufficiently precautionary. The common matrix was retained, while its role as a minimum district threshold and the possibility of more restrictive territorial provisions were clarified [52]. Retaining the shared classification therefore did not establish technical consensus on its precautionary adequacy.

Additional knowledge followed another route. Regional historical flood-extent records were recognized as relevant inputs, but their incorporation into official hazard maps remained subject to district technical standards and the formal update procedure [52]. Recognition of their evidential value thus provided a basis for further assessment without automatically modifying the maps.

Separately, Appendix 1 to the Landslide General Report records the correction of an erroneous extraction from the former Lazio regional PAI database before final adoption [24] (§9, pp. 37–38). This demonstrates verification and correction of source-data handling. The record does not identify who detected the error or associate its detection with a specific participation channel.

The principal finding is that validation connected different issues to different decisions. A shared class could be retained while its regulatory treatment was revised or its precautionary adequacy remained contested; a territorial concern could be addressed through a general provision; and additional knowledge could be recognized while requiring further assessment before incorporation. Coordination therefore combined amendment, retained disagreement and formal routes for further investigation.

6.4. Territorial Coverage, Updating and Evidence Boundaries

The spatial analysis establishes the territorial scale of the framework examined in Section 6.1, Section 6.2 and Section 6.3 and quantifies current correspondences between inherited and district categories. The technical landslide inventory covers approximately 12% of the district (Table 8). Consolidated fluvial mapping covers approximately 1830 km², or 4.3%; safeguard proposals add approximately 27 km² outside that footprint, bringing the combined coverage to approximately 1860 km², or 4.4%. Marine flood mapping covers approximately 35 km², or 0.08% (Table 10). The three hazard domains are reported separately.

Table 10. Approximate current flood-hazard extent and share of district territory, 3 October 2026.

Domain/Category	Area (Approx. km2)	District Area (%)
Fluvial P3	820	1.9
Fluvial P2 (all mapped P2 categories)	560	1.3
Fluvial P1	250	0.6
Reclamation attention areas	210	0.5
Residual hazard and risk	6	0.01
Consolidated fluvial footprint	1830	4.3
Additional footprint of safeguard proposals	27	0.06
Combined fluvial footprint	1860	4.4
Marine P3	15	0.04
Marine P2	7	0.02
Marine P1	12	0.03
Marine total	35	0.08

Source: AUBAC hydraulic GIS export and WebGIS safeguard layers, 3 October 2026. Areas and percentages are approximate; percentages use unrounded areas. The P2 total includes approximately 200 km² subject to transitional P3 provisions. Additional safeguard footprint counts only territory outside the consolidated map; the combined total removes overlaps. Category areas may overlap; the total footprint is calculated from their geometric union. Rounded components need not sum exactly. Hazard domains are not added together.

Within the consolidated fluvial map, approximately 820 km² are classified as P3, 560 km² as P2 and 250 km² as P1. These mapped classes include territorial qualifications relevant to their regulatory treatment. Approximately 200 km² of mapped P2 correspond to inherited floodable areas in the Marche regional basins subject to transitional P3 provisions under Article 26(4), pending determination of their effective hazard levels. A further 210 km² comprise reclamation attention areas, where Article 14(2) applies P2 provisions pending hazard definition, without prejudice to more precautionary sectoral provisions [42]. The smaller extent of mapped P1 partly reflects the uneven availability of inherited low-probability scenarios, including 500-year scenarios.

The correspondence analysis shows how categories from the preceding plans contribute to the district framework (Table 11). In Abruzzo–Sangro, approximately 64 km² of former P4 and 32 km² of former P3 correspond to district P3. In the Lazio regional basins, approximately 66 km² of A1 and 9 km² of A2 correspond to P3, while approximately 74 km² of B1 and B2 correspond to P2. These examples quantify the extent to which different source categories are represented within a common district class. Other entries document the retention of source distinctions through corresponding district categories.

Table 11. Principal source-to-district classification correspondences and their approximate territorial extent in the current mapping.

Domain and Source Plan	Source → District Correspondence (Approx. km2)
Flood—Abruzzo–Sangro	P4 → P3: 64; P3 → P3: 32; P2 → P2: 42; P1 → P1: 17
Flood—Lazio regional basins	A1 → P3: 66; A2 → P3: 9; B1 → P2: 49; B2 → P2: 25; C → P1: 21
Flood—Marche regional basins	Inherited flood-zone mapping → mapped P2: about 200; transitional P3 treatment described in Section 6.2 and Table 9
Flood—Tiber principal network	A → P3: 217; AA → P3: 16; B → P2: 50; C → P1: 32
Flood—Tiber secondary networks	A → P3: 208; B → P2: 125; C → P1: 137
Flood—Tiber PS1	Zone A → P3: 108; Zone B → P2: 4
Flood—Fiora	PI4 → P3: 31; PI3 → P2: 14; PI2 → P1: 22
Flood—Tronto	E4 → P3: 17; E3 → P3: 3; E2 → P2: 19; E1 → P1: 10
Flood—Ombrone	Source P4 → district P3: 2; source P3 → district P2: 14
Landslides—Marche	Recorded source code 1 → district P1: 73; → district P2: 246
Landslides—Tiber	Recorded source code 3 → district P2: 568; → district P3: 277

Source: AUBAC current preceding-plan layers, district hydraulic export and landslide inventory, 3 October 2026. Flood entries quantify the principal spatial correspondences interpreted using Table 5; landslide entries use recorded source codes on current geometries. The arrow (→) denotes the relationship between a recorded source category and a district category in the analyzed datasets; it does not by itself establish a change in physical hazard. Areas are approximate and generally rounded to the nearest km²; the Marche flood-zone correspondence is summarized as approximately 200 km². The table reports the principal current spatial correspondences, not an additive district land budget or a complete transition matrix. Appendix A reports the complete aggregated matrices, distinguishing preceding-plan mapping and safeguard overlays. Risk codes in Marche mapping are not treated as hazard classes. Current intersections can also reflect subsequent studies and updates; they are not all attributed to initial harmonization.

Current mapping also contains intersections that depart from the initial crosswalk. Examples include approximately 19 km² of Tiber principal-network band C intersecting district P2 or P3, 9 km² of Abruzzo–Sangro P2 intersecting district P3, and 8 km² of Tiber band B intersecting district P3. These are reported in the complete matrix in Appendix A, Table A2. Because the compared layers incorporate subsequent studies and updates, the intersections establish current spatial relationships; their causes cannot be inferred from the crosswalk alone.

For landslides, the Tiber and Marche examples show district assignments both higher and lower than the hazard classes represented by the recorded source codes. In the Tiber inventory, 36,475 features with source code 3 are assigned to district P2, covering approximately 570 km²; about 96% are recorded as quiescent. This pattern is consistent with the former Tiber scheme assigning P3 to several quiescent landslide types [24]. Features with the same source code also occupy approximately 277 km² in district P3. In the Marche inventory, 3127 features carrying source code 1 are assigned to district P2, covering approximately 246 km², while approximately 73 km² remain in P1. The complete provenance matrix in Appendix A accounts for all 107,045 inventory features.

These landslide correspondences give territorial scale to the differences between source schemes and district assignments. They complement the comparison of classification rules in Section 6.1, while describing the current inventory, including subsequent updates. Their regulatory implications must be read through the associated provisions.

Continued updating explains why the spatial analysis refers to a dated planning framework. New studies pass through proposal, safeguard and approval stages, during which an approved map and a safeguard overlay may coexist. The overlay can introduce additional mapped areas or revised classes before definitive incorporation into the plan. This process progressively revises the knowledge base beyond the initial harmonization of inherited classifications.

The district WebGIS makes the maps available for consultation alongside other spatial information [21]. Its presentation of preceding-plan safeguards and district counterparts also makes the procedural status of updates visible. The platform provides access to these representations, while the formal planning procedures establish changes to official classifications and provisions.

Taken together, the spatial and documentary analyses quantify the territory represented by common classes and identify the areas addressed by the regulatory comparisons. The territorial scope of a conditional reconstruction provision, for example, can be measured independently of whether projects have already been proposed or authorized. The findings therefore establish the extent of the planning framework and selected regulatory changes. Their uptake and consequences—approval rates, completed regeneration, changes in exposure and damage, and administrative performance—require project-level or longitudinal evidence.

7. Discussion

The results show how common classifications, regulatory provisions and institutional review worked together in the preparation of the district plans. This section examines what these relationships imply for the theoretical framework and what other planning authorities can learn from the process. Three propositions emerge from the case, developing the expectations introduced in Section 2.

7.1. Knowledge as a Public Regulatory Infrastructure

The first proposition is that common classification can make inherited information usable within one planning framework while preserving differences in its evidential basis. The Abruzzo–Sangro merger brought categories based on different flood scenarios into district P3, while the Marche crosswalk incorporated mapping with incomplete scenario coverage. In both cases, a common classification was established without recalculating every area through a uniform hydraulic model. These findings connect knowledge-infrastructure and classification research [25,27] to a practical requirement of planning: the sources and criteria underlying a mapped class remain relevant after harmonization. Retaining provenance allows users to distinguish what has been standardized from what still requires further investigation. A common classification established at the planning source may also support subsequent national mosaics, providing a shared institutional reference for statistical aggregation.

The second proposition is that a common category can remain in use while disagreement persists over its adequacy or consequences. The quiescent-landslide case shows this directly: the district P2 assignment was retained despite an objection to its precautionary adequacy. The public A1/A2 case illustrates a related but different response: the common class was retained while its regulatory treatment was reconsidered. These cases explain how classifications can function as boundary objects [28] in regulatory planning. They provide a shared reference through which institutions can coordinate decisions, while allowing debate to continue over the protection afforded by a class or the interventions permitted within it. Adoption therefore establishes a common framework without necessarily resolving the disagreements expressed during its preparation.

The third proposition is that the regulatory use of hazard knowledge depends on institutional decisions connecting evidence, official classification and applicable provisions. Historical flood records could be recognized as relevant evidence without immediately changing official maps. Detailed hydraulic information can support marginal-area delineation, but the associated provisions require a formal map update. In the Marche case, an inherited zone corresponds to mapped P2 while specified areas remain subject to transitional P3 provisions. These examples show why evidence, mapped class and applicable regime must be examined separately. They give concrete expression to the spatial-policy-infrastructure perspective [26]: integration involves establishing how technical information enters official planning and how that planning governs land-use decisions.

Together, these propositions explain the role of a public regulatory knowledge infrastructure. Its contribution lies in connecting information, classifications and provisions through traceable decisions, while retaining the means to examine source differences, address objections and incorporate new evidence. Figure 5 summarizes these relationships and the validation and updating routes documented in Section 5.3 and Section 6.3.

Figure 5. Analytical propositions derived from the case. Common classification can retain the provenance and differences of inherited knowledge (P1); a common category can remain in use while its precautionary adequacy or regulatory consequences are contested (P2); institutional decisions connect recognition of evidence, official classification and applicable regulatory treatment (P3). Validation can revisit the class-to-rule relationship, while formal updating provides a route for revising maps or provisions. Implementation outcomes lie outside the evidence of the study. Source: prepared by the author.

7.2. Common Thresholds and Precautionary Trade-Offs

Establishing a common classification requires deciding how to handle differences between inherited schemes. Table 7 illustrates the consequences of retaining source assignments or applying the district matrix to the same documented type–activity combinations. Retention preserves the distinction between Marche and Tiber: quiescent slides receive P2 and P3, respectively, while inactive slides receive P1 and P2. The district matrix assigns P2 to quiescent slides and P1 to inactive slides in both territories. Standardization therefore establishes a common explicit rule but does so without retaining the higher Tiber assignments in these examples.

The current inventory gives this issue a substantial territorial dimension. Approximately 570 km² of Tiber features carrying source code 3 are assigned to district P2, and about 96% of these features are recorded as quiescent (Section 6.4). This correspondence describes the current inventory, including subsequent updates, rather than isolating the effect of initial harmonization. Nevertheless, it shows why the treatment of quiescent phenomena is consequential for the district framework. The objection documented in Section 6.3 concerns the adequacy of the common threshold for precisely this type of classification choice.

The comparison of alternatives is an ex post analytical exercise; it does not establish that both options were formally evaluated during plan preparation. Retaining source assignments would preserve their differences, while applying the district matrix makes assignments consistent for the selected inputs. Reassessing the underlying hazard would address a further question, requiring additional evidence about the phenomena themselves. The comparison thus identifies what the common matrix achieves—consistency of assignment—and the question that remains open: whether its precautionary threshold is adequate across the conditions to which it applies.

The regulatory significance of a lower assignment depends on the provisions associated with the source and district classes. The district matrix establishes common class assignments, while the NTA define minimum protection levels and allow more restrictive territorial provisions under the applicable conditions [24,43]. That possibility does not establish that additional protection applies throughout the areas concerned. Assessing changes in protection therefore requires examining the governing rules alongside the classification, including territorial qualifications and requirements for further investigation.

The flood comparisons illustrate the complementary point that regulatory treatment can change while a common class is retained. In the A1/A2 case, review reconsidered the provisions associated with P3 without abandoning the category. More broadly, Section 6.2 identifies conditional opportunities for adapting existing buildings alongside more selective conditions for urban restructuring. Harmonization consequently involves choices about which transformations to permit and under what conditions, connecting hazard classification to the management of the existing built environment [26,29]. Its precautionary implications must be assessed through those specific choices, rather than inferred from a uniform direction of change across the district. Together, the preventive provisions examined in Section 6.2 connect hazard regulation with runoff management, soil protection and ecosystem restoration in ordinary territorial transformation.

7.3. From Institutional Roles to Coordinated Responsibility

Harmonization required connecting decisions taken by institutions with different responsibilities. A common district classification provides a shared reference, but its application depends on how studies are prepared, compatibility is assessed and interventions are authorized. This connects the case to basin-governance research: coordination requires both an overall framework and clear relationships among the authorities responsible for specific decisions [11,30,31].

The district plans organize these relationships through common technical requirements and an allocation of assessment responsibilities. The public response concerning the proponent’s study, compatibility verification and the hydraulic authority’s evaluation shows that clarifying who performs each task was itself part of harmonization [42,50]. District coordination therefore operates through shared criteria and connected procedures, while the relevant institutions retain their respective functions.

This arrangement makes administrative capacity important to implementation. Preparing and assessing detailed studies, processing map updates and applying territorial provisions require technical expertise and resources. The transferable lesson is that common rules need to be accompanied by a workable distribution of tasks and access to the knowledge required to perform them. Whether these conditions are met consistently across administrations remains a question for implementation assessment.

7.4. Updating the Common Framework: Adaptive Governance

The case shows that harmonization must accommodate continuing changes in hazard knowledge. The current mapping examined in Section 6.4 already includes updates beyond the initial transposition, while the adopted plans establish common technical references and formal procedures for subsequent revisions. The resulting framework combines a shared classification with a means of revising its territorial application as new evidence becomes available.

This arrangement gives practical expression to adaptive planning [40]. New studies must be assessed, translated into official classifications and incorporated through the applicable procedure. The historical flood-record case illustrates the distinction between recognizing additional evidence and completing its incorporation into a plan. The WebGIS makes maps and the procedural status of updates accessible, while formal decisions establish their planning consequences. A common district framework can also support institutional learning: opinions, technical verifications and map updates can be compared against shared classes and provisions, helping identify recurring implementation problems and informing subsequent revisions.

Climate non-stationarity strengthens the importance of maintaining this connection between evidence and revision [65]. Return-period bands organize flood-hazard classes but do not themselves represent projections of future climate conditions. Common technical standards and updating procedures provide a route for incorporating revised assessments. Evaluating the adaptive performance of the district framework will therefore require examining how the new procedures operate over time: which evidence leads to revision, how long assessment and incorporation take, and whether maps and applicable provisions remain aligned.

7.5. Transferability and Conditions for Application

The case offers practical lessons for authorities reorganizing inherited planning systems. Source review and explicit classification correspondences established a shared technical reference for developing regulatory provisions [24,42,43]. Flood and landslide methods remained distinct, while specialist groups and the AUBAC–Regions Technical Working Group connected technical choices with their regulatory implications [48]. Preparation progressed from classification harmonization to the development of admissibility provisions and their final coordination. This sequence provides a documented way of organizing interdependent decisions without requiring every technical and regulatory question to be settled simultaneously (Section 5.1 and Section 5.2).

Institutional review provides a second transferable component. Regional programmatic conferences provided a forum for examining territorial concerns [8,48], while separate response records documented the assessment of regional observations and public submissions [50,51,52]. Responses distinguished issues that could be addressed through revised provisions, classification choices requiring clarification or further discussion, and requests for map changes requiring technical assessment and a formal procedure. This organization allowed plan preparation to proceed while identifying matters requiring further investigation or subsequent updating (Section 6.3). The documented correction of a source-extraction error also distinguishes correction of data handling from reassessment of hazard boundaries, which requires supporting technical evidence and the applicable updating procedure (Section 6.3).

The regulatory comparisons also offer substantive lessons. Common classes accommodate differentiated conditions for existing buildings, urban restructuring and territorial circumstances, while transitional provisions connect inherited mapping to the applicable regime [42,43] (Section 6.2; Table 9). Marginal-area provisions show how water-depth and velocity information can support differentiated land-use rules within a common hazard class, subject to formal cartographic delineation [42]. Other authorities can examine these solutions when designing provisions suited to their own legal responsibilities, territorial conditions and available evidence.

Both district plans reached final adoption on 26 May 2026 [22,23], less than ten months after draft adoption on 31 July 2025 [19,20]. This interval covers consultation and institutional review within a longer preparation process. Completing two coordinated plans demonstrates that the combined technical and institutional approach was feasible in this setting. The documented organization of the work provides a practical reference for other authorities, although the case does not isolate the contribution of individual organizational choices to the time taken. Application elsewhere requires adequate source information, technical capacity and arrangements for interinstitutional coordination.

European studies illustrate why these institutional conditions matter. Holub and Fuchs [66] describe Austrian hazard maps as expert assessments whose binding effect on spatial planning depends on specific legislative provisions. Thiery et al. [67] examine landslide hazard assessment within France’s regulatory risk-prevention plans. These different relationships between mapping and regulation provide useful settings for examining how evidence acquires official status and land-use consequences. Within Italy, the Eastern Alps framework coordinates flood-risk management with the PAIs in force [55]; the Po district combines its PAI with a proposed territorial extension [56,57]; and the Northern and Southern Apennine districts use different combinations of sectoral and inherited instruments [58,59]. These pathways demonstrate the wider relevance of coordinating fragmented planning systems.

The transferable contribution is therefore an approach to organizing decisions: establish correspondences between sources, develop common provisions, coordinate institutional review and maintain formal routes for revision. The river basin district scale and cyclical planning framework of European water policy make these tasks relevant beyond the Italian case [4,5]. Their application requires adaptation to the receiving legal and administrative context, while comparative studies can assess which arrangements work under different conditions.

7.6. Limitations and Research Agenda

This study reconstructs plan formation through documentary and spatial evidence. Selected issue–response cases reveal how particular decisions were handled but do not represent the full range of participants’ experiences. Stakeholder interviews and independent evidence from implementing authorities and users would broaden the institutional perspective and help assess how the adopted framework is understood and applied.

The spatial analysis describes dated analytical extracts from an evolving planning system. Current correspondences can include subsequent studies and updates alongside the initial harmonization. They establish the territorial extent of recorded relationships, while attribution to the initial transposition depends on the documented classification rules and source history. Future comparisons should use consistent spatial boundaries, record-selection criteria and aggregation conventions.

Further research should connect the territorial scope of the compared provisions to their implementation. Building and population overlays would characterize the assets within the mapped areas; project records would document applications, compatibility assessments, authorizations and completed transformations. These data would allow examination of how the regulatory opportunities and restrictions identified here operate in practice.

Administrative evaluation should examine review times, requests for supplementary information, consistency of decisions and use of updating procedures. Territorial evaluation should assess changes in exposure and vulnerability alongside completed mitigation and regeneration interventions. Longitudinal evidence would thus extend the present assessment of plan formation and adopted provisions to their administrative and territorial consequences.

8. Conclusions

The Central Apennine case shows that integrating inherited hazard plans requires coordinated decisions about how knowledge is classified, how classifications govern land use and how both can be revised. Bringing nine PAIs into two district plans established a common framework while retaining identifiable differences in source information and some unresolved disagreements over precaution. The scientific contribution lies in explaining how shared categories can support binding planning decisions despite these differences. This develops the knowledge-infrastructure and boundary-object perspectives by identifying the institutional decisions that connect evidence, official classifications and regulatory consequences. The concept of a public regulatory knowledge infrastructure describes this connection, including the procedures through which it can be reconsidered.

The findings demonstrate how this integration operated across the two hazard domains. Initial flood transposition retained inherited boundaries, whereas landslide harmonization applied a common type–activity matrix. The datasets obtained on 3 October 2026 document the territorial scale of the framework: approximately 1860 km² within the current fluvial planning framework, including safeguard proposals, and 107,045 mapped landslide phenomena occupying approximately 4970 km². Common classifications provided a basis for regulatory coordination, but the consequences depended on the inherited provisions and the interventions concerned. Conditional opportunities for transforming existing buildings coexisted with more selective urban restructuring rules, while transitional regimes and marginal-area provisions differentiated treatment within mapped classes. Linking provisions to spatial correspondences also made their territorial scope measurable: the approximately 64 km² Abruzzo–Sangro correspondence defines the extent of one reconstruction comparison, independently of subsequent project applications or authorizations. Institutional review could amend provisions while retaining classifications and could retain a common category without resolving disagreement over its precautionary adequacy.

Methodologically, this study combines classification rules, comparisons of provisions for the same intervention, issue–response records and current spatial correspondences to distinguish classification changes, regulatory changes and subsequent hazard reassessment. For other planning authorities, the case provides a documented way of organizing interdependent decisions. Source review and classification correspondences established the basis for progressive regulatory drafting; specialist and institutional review connected technical choices with their consequences for land use. The six analytical functions—inventory, classification correspondence, regulatory design, validation, transition and updating—make the tasks, inputs and outputs explicit. Formal map-updating routes allow additional knowledge to be assessed and incorporated without reopening the general provisions, while regulatory amendments follow a distinct procedure. Together, these arrangements offer procedural and regulatory references that authorities can adapt to their available knowledge, legal responsibilities and administrative capacities. Final adoption of both plans in May 2026 marks the outcome examined in this study; subsequent approval, legal effectiveness and implementation remain distinct from that outcome.

Future research should compare how these arrangements operate under different institutional conditions and follow their use over time. Project and administrative records could establish how admissibility provisions, assessment responsibilities and updating procedures are applied, including review times and consistency of decisions. Linking these records to changes in exposure, vulnerability and completed interventions would help assess whether the integrated framework contributes to more effective risk prevention and management.

Appendix A.1. Landslide Provenance Matrix

Table A1. Complete landslide provenance matrix. Each district-class cell reports feature count/within-cell union area (km²).

Source PAI	Source Code	District P1	District P2	District P3	District P4
Abruzzo	1	495/75.95	3981/385.05	3/0.42	0/0.00
Abruzzo	2	89/7.13	8943/840.25	330/6.71	1/0.01
Abruzzo	3	0/0.00	1333/96.78	2939/196.65	162/14.64
Abruzzo	P1	0/0.00	1/0.19	0/0.00	0/0.00
Abruzzo	P2	0/0.00	6/0.29	1/0.08	0/0.00
Abruzzo	P3	0/0.00	3/0.23	11/1.48	1/0.01
Fiora	—	1/0.37	559/14.63	1188/50.82	211/3.10
Lazio	—	0/0.00	20/0.10	2/0.00	0/0.00
Lazio	A	1/0.00	80/3.57	620/13.43	2048/86.31
Lazio	B	0/0.00	10/0.56	54/1.73	3/1.16
Lazio	C	1/0.01	647/30.75	15/0.36	5/0.03
Liri-Garigliano	—	0/0.00	165/3.96	286/9.12	563/11.13
Marche	1	1253/73.40	3127/245.96	0/0.00	0/0.00
Marche	2	1/0.01	8278/562.71	42/14.31	0/0.00
Marche	3	0/0.00	26/1.42	9041/591.55	0/0.00
Marche	4	0/0.00	0/0.00	0/0.00	274/19.42
Marche	P2	0/0.00	1/0.07	0/0.00	0/0.00
Marche	P3	0/0.00	0/0.00	1/0.03	0/0.00
Ombrone	—	0/0.00	171/8.46	0/0.00	0/0.00
Tevere	1	2528/193.05	2871/13.55	0/0.00	2/0.02
Tevere	2	914/151.85	3001/186.97	1379/81.48	1/0.00
Tevere	3	0/0.00	36,475/568.28	8912/277.09	1/0.04
Tevere	4	0/0.00	1/0.00	1229/7.03	776/31.97
Tevere	R4	0/0.00	0/0.00	2/0.08	46/3.77
Tronto	1	0/0.00	0/0.00	5/0.01	0/0.00
Tronto	2	0/0.00	3/0.27	4/0.05	0/0.00
Tronto	3	0/0.00	0/0.00	2/0.31	0/0.00
Tronto	4	0/0.00	0/0.00	1/0.28	0/0.00
Tronto	H0	0/0.00	3/0.05	12/0.04	3/0.01
Tronto	H1	11/0.15	55/0.55	185/1.02	15/0.17
Tronto	H2	28/3.25	262/11.82	644/11.47	28/0.97
Tronto	H3	0/0.00	150/13.38	342/32.02	79/1.11
Tronto	H4	0/0.00	4/0.49	26/4.21	52/6.06
Total features (107,045)		5322	70,176	27,276	4271

Source: AUBAC technical landslide inventory received on 3 October 2026. All 107,045 mapped features are included. Codes are reproduced as recorded and must be interpreted within their source scheme; R4 is a risk code, not a hazard class. There is no group labeled Marecchia–Conca in the extract. A dash denotes a source code not recorded in the extract; these records are retained. The final row reports counts only; its grand total is 107,045. Counts are additive; areas may overlap between cells and are not additive. The inventory incorporates updates through the July 2026 Operational Conference. The matrix is a provenance comparison on current geometries, not a time series of physical hazard changes. Areas below 0.005 km² are displayed as 0.00; a positive feature count with a displayed area of 0.00 therefore does not imply zero mapped area.

Table A1. Complete landslide provenance matrix. Each district-class cell reports feature count/within-cell union area (km²).

Source PAI	Source Code	District P1	District P2	District P3	District P4
Abruzzo	1	495/75.95	3981/385.05	3/0.42	0/0.00
Abruzzo	2	89/7.13	8943/840.25	330/6.71	1/0.01
Abruzzo	3	0/0.00	1333/96.78	2939/196.65	162/14.64
Abruzzo	P1	0/0.00	1/0.19	0/0.00	0/0.00
Abruzzo	P2	0/0.00	6/0.29	1/0.08	0/0.00
Abruzzo	P3	0/0.00	3/0.23	11/1.48	1/0.01
Fiora	—	1/0.37	559/14.63	1188/50.82	211/3.10
Lazio	—	0/0.00	20/0.10	2/0.00	0/0.00
Lazio	A	1/0.00	80/3.57	620/13.43	2048/86.31
Lazio	B	0/0.00	10/0.56	54/1.73	3/1.16
Lazio	C	1/0.01	647/30.75	15/0.36	5/0.03
Liri-Garigliano	—	0/0.00	165/3.96	286/9.12	563/11.13
Marche	1	1253/73.40	3127/245.96	0/0.00	0/0.00
Marche	2	1/0.01	8278/562.71	42/14.31	0/0.00
Marche	3	0/0.00	26/1.42	9041/591.55	0/0.00
Marche	4	0/0.00	0/0.00	0/0.00	274/19.42
Marche	P2	0/0.00	1/0.07	0/0.00	0/0.00
Marche	P3	0/0.00	0/0.00	1/0.03	0/0.00
Ombrone	—	0/0.00	171/8.46	0/0.00	0/0.00
Tevere	1	2528/193.05	2871/13.55	0/0.00	2/0.02
Tevere	2	914/151.85	3001/186.97	1379/81.48	1/0.00
Tevere	3	0/0.00	36,475/568.28	8912/277.09	1/0.04
Tevere	4	0/0.00	1/0.00	1229/7.03	776/31.97
Tevere	R4	0/0.00	0/0.00	2/0.08	46/3.77
Tronto	1	0/0.00	0/0.00	5/0.01	0/0.00
Tronto	2	0/0.00	3/0.27	4/0.05	0/0.00
Tronto	3	0/0.00	0/0.00	2/0.31	0/0.00
Tronto	4	0/0.00	0/0.00	1/0.28	0/0.00
Tronto	H0	0/0.00	3/0.05	12/0.04	3/0.01
Tronto	H1	11/0.15	55/0.55	185/1.02	15/0.17
Tronto	H2	28/3.25	262/11.82	644/11.47	28/0.97
Tronto	H3	0/0.00	150/13.38	342/32.02	79/1.11
Tronto	H4	0/0.00	4/0.49	26/4.21	52/6.06
Total features (107,045)		5322	70,176	27,276	4271

Source: AUBAC technical landslide inventory received on 3 October 2026. All 107,045 mapped features are included. Codes are reproduced as recorded and must be interpreted within their source scheme; R4 is a risk code, not a hazard class. There is no group labeled Marecchia–Conca in the extract. A dash denotes a source code not recorded in the extract; these records are retained. The final row reports counts only; its grand total is 107,045. Counts are additive; areas may overlap between cells and are not additive. The inventory incorporates updates through the July 2026 Operational Conference. The matrix is a provenance comparison on current geometries, not a time series of physical hazard changes. Areas below 0.005 km² are displayed as 0.00; a positive feature count with a displayed area of 0.00 therefore does not imply zero mapped area.

Appendix A.2. Fluvial Correspondence Matrix

Table A2. Fluvial spatial correspondence matrix for all retrieved source groups. Areas in km²; preceding-plan groups and safeguard overlays remain separate.

Source/Role	Code	P1	P2	P3	P2t	BA	RR
Abruzzo–Sangro/Base	P1	16.84	1.73	4.20	0.00	0.00	0.00
Abruzzo–Sangro/Base	P2	0.03	41.75	9.21	0.00	0.00	0.00
Abruzzo–Sangro/Base	P3	0.00	0.03	32.11	0.00	0.00	0.00
Abruzzo–Sangro/Base	P4	0.00	0.00	63.98	0.00	0.00	0.00
Abruzzo–Sangro/SG	P1	7.42	1.04	0.68	0.00	0.00	0.00
Abruzzo–Sangro/SG	P2	0.46	19.58	6.02	0.00	0.00	0.00
Abruzzo–Sangro/SG	P3	0.26	0.78	14.98	0.00	0.00	0.00
Abruzzo–Sangro/SG	P4	0.46	0.62	46.89	0.00	0.00	0.00
Fiora/Base	P.I.2	21.81	0.00	0.00	0.00	0.00	0.00
Fiora/Base	P.I.3	0.00	13.90	0.00	0.00	0.00	0.00
Fiora/Base	P.I.4	0.00	0.00	30.93	0.00	0.00	0.00
Lazio/Base	A1	0.00	0.00	65.53	0.00	0.00	0.00
Lazio/Base	A2	0.00	0.00	9.38	0.00	0.00	0.00
Lazio/Base	Attenzione	0.00	0.00	0.06	0.00	207.65	0.00
Lazio/Base	B1	0.00	48.95	0.56	0.00	0.00	0.00
Lazio/Base	B2	0.00	25.05	0.00	0.00	0.00	0.00
Lazio/Base	C	20.50	0.00	0.00	0.00	0.00	0.00
Marche/Base	R1	0.00	0.00	0.31	43.25	0.00	0.00
Marche/Base	R2	0.00	0.00	1.54	62.33	0.00	0.00
Marche/Base	R3	0.00	0.00	0.15	46.96	0.00	0.00
Marche/Base	R4	0.00	0.00	6.34	49.63	0.00	0.00
Marche/SG (risk class)	R1	0.00	0.14	1.13	0.01	0.00	0.00
Marche/SG (risk class)	R2	0.00	0.29	3.20	0.15	0.00	0.00
Marche/SG (risk class)	R3	0.00	0.00	0.00	0.01	0.00	0.00
Marche/SG (risk class)	R4	0.00	1.17	8.09	3.66	0.00	0.00
Marche/SG (risk class)	Residual risk	0.82	0.00	0.00	1.29	0.00	3.37
Marche/SG (risk)	R1	0.00	0.00	0.00	8.40	0.00	0.00
Marche/SG (risk)	R2	0.00	0.00	0.00	7.39	0.00	0.00
Marche/SG (risk)	R3	0.00	0.00	0.00	5.56	0.00	0.00
Marche/SG (risk)	R4	0.00	0.00	0.00	2.45	0.00	0.00
Ombrone/Alternative service	P3	0.00	14.18	0.00	0.00	0.00	0.00
Ombrone/Alternative service	P4	0.00	0.00	1.66	0.00	0.00	0.00
Tiber PS1/Base	Zona A	0.00	0.01	107.81	0.00	0.00	0.00
Tiber PS1/Base	Zona B	0.00	4.36	0.00	0.00	0.00	0.00
Tiber principal network/Base	A	0.00	0.00	216.73	0.00	0.00	0.00
Tiber principal network/Base	AA	0.00	0.00	15.65	0.00	0.00	0.00
Tiber principal network/Base	B	0.00	50.21	7.55	0.00	0.00	0.00
Tiber principal network/Base	C	32.34	12.65	5.92	0.00	0.00	0.00
Tiber principal network/SG	A	0.06	0.93	19.11	0.00	0.00	0.00
Tiber principal network/SG	B	0.16	1.67	1.24	0.00	0.00	0.00
Tiber principal network/SG	C	0.43	1.20	0.55	0.00	0.00	0.00
Tiber secondary networks/Base	A	0.00	0.00	208.25	0.00	0.00	0.00
Tiber secondary networks/Base	B	0.00	124.81	5.35	0.00	0.00	0.00
Tiber secondary networks/Base	C	136.63	3.52	2.16	0.00	0.00	0.00
Tiber secondary networks/SG	A	0.29	2.49	14.73	0.00	0.00	0.00
Tiber secondary networks/SG	B	1.25	2.88	8.34	0.00	0.00	0.00
Tiber secondary networks/SG	C	1.39	2.80	3.98	0.00	0.00	0.00
Tronto/Base	E1	9.56	0.00	0.00	0.00	0.00	0.00
Tronto/Base	E2	0.00	18.68	0.00	0.00	0.00	0.00
Tronto/Base	E3	0.00	1.17	3.07	0.00	0.00	0.00
Tronto/Base	E4	0.00	0.00	16.65	0.00	0.00	0.00

P2t = mapped P2 subject to transitional P3 provisions, reported separately from the other mapped P2 areas in this matrix; Table 10 includes both categories in its overall mapped P2 coverage. BA = reclamation attention areas; RR = residual hazard and risk; Base = preceding-plan source group; SG = separately identified safeguard mapping. Rows describe intersections with the consolidated district extract and do not imply that every intersection is an initial harmonization transition. Source groups and safeguard overlays may overlap and must not be summed. Source attribute fields vary by service layer (see the technical note); the two Marche safeguard groups correspond to separate layers with different risk fields. Source classes and codes retain their original meaning; Fiora hazard labels are abbreviated to their P.I. codes. The Ombrone group is labeled Alternative service because its procedural status could not be independently resolved from its attributes; it is retained as a spatial source without classifying it as approved mapping. Values below 0.005 km² are displayed as 0.00. Small unmatched areas are not assigned to a class by assumption. Marine flooding is reported separately in Table 10 and is not included in this fluvial matrix. Source: public AUBAC preceding-plan WebGIS layers and district flood extract retrieved on 3 October 2026.

Table A2. Fluvial spatial correspondence matrix for all retrieved source groups. Areas in km²; preceding-plan groups and safeguard overlays remain separate.

Source/Role	Code	P1	P2	P3	P2t	BA	RR
Abruzzo–Sangro/Base	P1	16.84	1.73	4.20	0.00	0.00	0.00
Abruzzo–Sangro/Base	P2	0.03	41.75	9.21	0.00	0.00	0.00
Abruzzo–Sangro/Base	P3	0.00	0.03	32.11	0.00	0.00	0.00
Abruzzo–Sangro/Base	P4	0.00	0.00	63.98	0.00	0.00	0.00
Abruzzo–Sangro/SG	P1	7.42	1.04	0.68	0.00	0.00	0.00
Abruzzo–Sangro/SG	P2	0.46	19.58	6.02	0.00	0.00	0.00
Abruzzo–Sangro/SG	P3	0.26	0.78	14.98	0.00	0.00	0.00
Abruzzo–Sangro/SG	P4	0.46	0.62	46.89	0.00	0.00	0.00
Fiora/Base	P.I.2	21.81	0.00	0.00	0.00	0.00	0.00
Fiora/Base	P.I.3	0.00	13.90	0.00	0.00	0.00	0.00
Fiora/Base	P.I.4	0.00	0.00	30.93	0.00	0.00	0.00
Lazio/Base	A1	0.00	0.00	65.53	0.00	0.00	0.00
Lazio/Base	A2	0.00	0.00	9.38	0.00	0.00	0.00
Lazio/Base	Attenzione	0.00	0.00	0.06	0.00	207.65	0.00
Lazio/Base	B1	0.00	48.95	0.56	0.00	0.00	0.00
Lazio/Base	B2	0.00	25.05	0.00	0.00	0.00	0.00
Lazio/Base	C	20.50	0.00	0.00	0.00	0.00	0.00
Marche/Base	R1	0.00	0.00	0.31	43.25	0.00	0.00
Marche/Base	R2	0.00	0.00	1.54	62.33	0.00	0.00
Marche/Base	R3	0.00	0.00	0.15	46.96	0.00	0.00
Marche/Base	R4	0.00	0.00	6.34	49.63	0.00	0.00
Marche/SG (risk class)	R1	0.00	0.14	1.13	0.01	0.00	0.00
Marche/SG (risk class)	R2	0.00	0.29	3.20	0.15	0.00	0.00
Marche/SG (risk class)	R3	0.00	0.00	0.00	0.01	0.00	0.00
Marche/SG (risk class)	R4	0.00	1.17	8.09	3.66	0.00	0.00
Marche/SG (risk class)	Residual risk	0.82	0.00	0.00	1.29	0.00	3.37
Marche/SG (risk)	R1	0.00	0.00	0.00	8.40	0.00	0.00
Marche/SG (risk)	R2	0.00	0.00	0.00	7.39	0.00	0.00
Marche/SG (risk)	R3	0.00	0.00	0.00	5.56	0.00	0.00
Marche/SG (risk)	R4	0.00	0.00	0.00	2.45	0.00	0.00
Ombrone/Alternative service	P3	0.00	14.18	0.00	0.00	0.00	0.00
Ombrone/Alternative service	P4	0.00	0.00	1.66	0.00	0.00	0.00
Tiber PS1/Base	Zona A	0.00	0.01	107.81	0.00	0.00	0.00
Tiber PS1/Base	Zona B	0.00	4.36	0.00	0.00	0.00	0.00
Tiber principal network/Base	A	0.00	0.00	216.73	0.00	0.00	0.00
Tiber principal network/Base	AA	0.00	0.00	15.65	0.00	0.00	0.00
Tiber principal network/Base	B	0.00	50.21	7.55	0.00	0.00	0.00
Tiber principal network/Base	C	32.34	12.65	5.92	0.00	0.00	0.00
Tiber principal network/SG	A	0.06	0.93	19.11	0.00	0.00	0.00
Tiber principal network/SG	B	0.16	1.67	1.24	0.00	0.00	0.00
Tiber principal network/SG	C	0.43	1.20	0.55	0.00	0.00	0.00
Tiber secondary networks/Base	A	0.00	0.00	208.25	0.00	0.00	0.00
Tiber secondary networks/Base	B	0.00	124.81	5.35	0.00	0.00	0.00
Tiber secondary networks/Base	C	136.63	3.52	2.16	0.00	0.00	0.00
Tiber secondary networks/SG	A	0.29	2.49	14.73	0.00	0.00	0.00
Tiber secondary networks/SG	B	1.25	2.88	8.34	0.00	0.00	0.00
Tiber secondary networks/SG	C	1.39	2.80	3.98	0.00	0.00	0.00
Tronto/Base	E1	9.56	0.00	0.00	0.00	0.00	0.00
Tronto/Base	E2	0.00	18.68	0.00	0.00	0.00	0.00
Tronto/Base	E3	0.00	1.17	3.07	0.00	0.00	0.00
Tronto/Base	E4	0.00	0.00	16.65	0.00	0.00	0.00

P2t = mapped P2 subject to transitional P3 provisions, reported separately from the other mapped P2 areas in this matrix; Table 10 includes both categories in its overall mapped P2 coverage. BA = reclamation attention areas; RR = residual hazard and risk; Base = preceding-plan source group; SG = separately identified safeguard mapping. Rows describe intersections with the consolidated district extract and do not imply that every intersection is an initial harmonization transition. Source groups and safeguard overlays may overlap and must not be summed. Source attribute fields vary by service layer (see the technical note); the two Marche safeguard groups correspond to separate layers with different risk fields. Source classes and codes retain their original meaning; Fiora hazard labels are abbreviated to their P.I. codes. The Ombrone group is labeled Alternative service because its procedural status could not be independently resolved from its attributes; it is retained as a spatial source without classifying it as approved mapping. Values below 0.005 km² are displayed as 0.00. Small unmatched areas are not assigned to a class by assumption. Marine flooding is reported separately in Table 10 and is not included in this fluvial matrix. Source: public AUBAC preceding-plan WebGIS layers and district flood extract retrieved on 3 October 2026.

Appendix A.3. Public Data Services

Table A3. Public fluvial services and retrieved polygon layers, 3 October 2026.

Public Service	Retrieved Polygon Layer IDs
PAI_Abruzzo_Sangro_05_11_2025	149, 1815, 1817
PAI_Fiora_05_11_2025	7
PAI_Ombrone_webgis	1770
PAI_Tronto_05_11_2025	6
T05_pai_lazio_20251105	15
T05_pai_marche_25833	6, 7, 8, 9
T05_pai_tevere_25833_20260902	22, 23, 24, 25, 30, 31, 32, 35, 40, 41, 42, 45
T12_Proposta_Paglia_Chiani_20260909_p	4
T12_Proposta_Saline_Fino_Tavo_p	4
T12_ite_proposta_metauro_con_ms_ds116_20260622_p	4

Base URL: https://webgis.abdac.it/server/rest/services/. Layer endpoints are formed by appending the service identifier followed by /MapServer/{layer ID}. Service names and layer identifiers refer to the retrieval of 3 October 2026.

Table A3. Public fluvial services and retrieved polygon layers, 3 October 2026.

Public Service	Retrieved Polygon Layer IDs
PAI_Abruzzo_Sangro_05_11_2025	149, 1815, 1817
PAI_Fiora_05_11_2025	7
PAI_Ombrone_webgis	1770
PAI_Tronto_05_11_2025	6
T05_pai_lazio_20251105	15
T05_pai_marche_25833	6, 7, 8, 9
T05_pai_tevere_25833_20260902	22, 23, 24, 25, 30, 31, 32, 35, 40, 41, 42, 45
T12_Proposta_Paglia_Chiani_20260909_p	4
T12_Proposta_Saline_Fino_Tavo_p	4
T12_ite_proposta_metauro_con_ms_ds116_20260622_p	4

Base URL: https://webgis.abdac.it/server/rest/services/. Layer endpoints are formed by appending the service identifier followed by /MapServer/{layer ID}. Service names and layer identifiers refer to the retrieval of 3 October 2026.

Appendix B.1. Procedural Sequence

Table A4 lists the main steps from the 2023 Policy Guidance to publication of the final adoption notices. Forums with different functions operated in parallel: public consultation collected submissions from all interested parties; regional programmatic conferences under Article 68 of Legislative Decree No. 152/2006 coordinated the plans with territorial planning; the AUBAC–Regions Technical Working Group examined proposed amendments to the NTA; and the Operational Conference monitored the progress of the review [8,48,49].

Table A4. Main procedural steps in the preparation of the two district PAIs.

Date	Step	Forum or Act
25 July 2023	Policy Guidance identifies district PAIs and harmonized safeguards and provisions as priorities	Permanent Institutional Conference, Resolution No. 38
31 July 2025	Adoption of the draft Flood and Landslide PAIs and related safeguard measures	Permanent Institutional Conference, Resolutions Nos. 57 and 58
22 August 2025	Publication in Official Gazette No. 194; start of the 90-day public consultation	Official Gazette of the Italian Republic
19 September 2025	Launch of the regional programmatic conferences and proposal of a joint technical working group	AUBAC note to the Regions
2 October 2025	Review of progress, working method and timetable	Operational Conference
20 October 2025	First meeting	AUBAC–Regions Technical Working Group
22 November 2025	Close of the public consultation (133 flood and 42 landslide submissions)	Public consultation
8 January 2026	Second meeting	AUBAC–Regions Technical Working Group
26 January–3 March 2026	Transmission of regional proposals on the NTA	Regions
28 January 2026	Review of progress	Operational Conference
18 February 2026	Third meeting	AUBAC–Regions Technical Working Group
5 March 2026	Review of progress	Operational Conference
9 March 2026	Transmission of amended texts and twelve region-by-region response reports	AUBAC to the Regions and the Ministry
17–25 March 2026	Bilateral meetings with Emilia-Romagna, Umbria, Abruzzo, Marche, Tuscany and Lazio	AUBAC–Region meetings
31 March 2026	Presentation of the revised texts and final meeting	AUBAC–Regions Technical Working Group
2 April 2026	Definition of the final proposal of the two PAIs	Technical–Operational Secretariat
13 April 2026	Follow-up meeting	Ministry of the Environment and Energy Security
15 April 2026	Closing session of the Abruzzo programmatic conference	Regional programmatic conference
16 April 2026	Final refinements and proposal of safeguard measures	Technical–Operational Secretariat
20 April 2026	Opinion on the final texts and safeguard measures	Operational Conference
26 May 2026	Final adoption of the two PAIs and related safeguard measures	Permanent Institutional Conference, Resolutions Nos. 67 and 68
4 June 2026	Publication of the adoption notices; safeguard measures take effect	Official Gazette of the Italian Republic

Source: prepared by the author based on the Policy Guidance [18], draft-adoption documents [19,20], final-adoption resolutions [22,23] and publication notices [53,54], the public response reports [50,51], the Explanatory Report updated on 16 April 2026 [48] and the published minutes of the Operational Conference of 20 April 2026 [49] (pp. 4–6).

Table A4. Main procedural steps in the preparation of the two district PAIs.

Date	Step	Forum or Act
25 July 2023	Policy Guidance identifies district PAIs and harmonized safeguards and provisions as priorities	Permanent Institutional Conference, Resolution No. 38
31 July 2025	Adoption of the draft Flood and Landslide PAIs and related safeguard measures	Permanent Institutional Conference, Resolutions Nos. 57 and 58
22 August 2025	Publication in Official Gazette No. 194; start of the 90-day public consultation	Official Gazette of the Italian Republic
19 September 2025	Launch of the regional programmatic conferences and proposal of a joint technical working group	AUBAC note to the Regions
2 October 2025	Review of progress, working method and timetable	Operational Conference
20 October 2025	First meeting	AUBAC–Regions Technical Working Group
22 November 2025	Close of the public consultation (133 flood and 42 landslide submissions)	Public consultation
8 January 2026	Second meeting	AUBAC–Regions Technical Working Group
26 January–3 March 2026	Transmission of regional proposals on the NTA	Regions
28 January 2026	Review of progress	Operational Conference
18 February 2026	Third meeting	AUBAC–Regions Technical Working Group
5 March 2026	Review of progress	Operational Conference
9 March 2026	Transmission of amended texts and twelve region-by-region response reports	AUBAC to the Regions and the Ministry
17–25 March 2026	Bilateral meetings with Emilia-Romagna, Umbria, Abruzzo, Marche, Tuscany and Lazio	AUBAC–Region meetings
31 March 2026	Presentation of the revised texts and final meeting	AUBAC–Regions Technical Working Group
2 April 2026	Definition of the final proposal of the two PAIs	Technical–Operational Secretariat
13 April 2026	Follow-up meeting	Ministry of the Environment and Energy Security
15 April 2026	Closing session of the Abruzzo programmatic conference	Regional programmatic conference
16 April 2026	Final refinements and proposal of safeguard measures	Technical–Operational Secretariat
20 April 2026	Opinion on the final texts and safeguard measures	Operational Conference
26 May 2026	Final adoption of the two PAIs and related safeguard measures	Permanent Institutional Conference, Resolutions Nos. 67 and 68
4 June 2026	Publication of the adoption notices; safeguard measures take effect	Official Gazette of the Italian Republic

Source: prepared by the author based on the Policy Guidance [18], draft-adoption documents [19,20], final-adoption resolutions [22,23] and publication notices [53,54], the public response reports [50,51], the Explanatory Report updated on 16 April 2026 [48] and the published minutes of the Operational Conference of 20 April 2026 [49] (pp. 4–6).

Appendix B.2. Common NTA Architecture

Although the harmonization produced two separate regulatory texts, both adopt the same five-part structure, adapted to the characteristics of flood and geomorphological phenomena (Table A5). The initial parts establish plan documents, definitions and common terminology, together with planning principles and coordination with spatial and urban planning. Part III connects hazard classes to regulatory provisions. Part IV regulates specific areas and elements, conditions of incomplete knowledge and additional planning and mitigation measures: in the Flood PAI, it covers land-reclamation areas, the river network not yet classified, hydraulic invariance, mitigation measures, aggregate extraction and state-owned areas; in the Landslide PAI, it covers potentially hazardous geological and geomorphological elements and the Landslide Risk Atlas of the pre-existing Tiber PAI. Part V governs opinions, coordination, map updates, plan amendments and transitional and final provisions [42,43].

Table A5. Structural and functional correspondence between the NTA of the two district PAIs.

Function	Flood PAI	Landslide PAI
Plan documents, definitions and common terminology	Part I—General provisions (Arts. 1–4)	Part I—General provisions (Arts. 1–4)
Planning principles and relationship with spatial and urban planning	Part II—General principles and framework provisions (Arts. 5–6)	Part II—General principles and framework provisions (Arts. 5–6)
Rules by hazard class and provisions for risk areas	Part III—Hazard zones and risk areas (Arts. 7–13)	Part III—Landslide-hazard and risk areas (Arts. 7–11)
Specific areas and elements, incomplete knowledge, planning and mitigation measures	Part IV—Attention areas and flood-management interventions (Arts. 14–19)	Part IV—Potentially hazardous geomorphological elements (Arts. 12–17)
Opinions, coordination, updating, amendments, transitional and final provisions	Part V—Common provisions (Arts. 20–26); Annexes 1–3	Part V—Common provisions (Arts. 18–24); Annexes 1–5

Source: prepared by the author based on the final NTA of the District Flood and Landslide PAIs [42,43].

Table A5. Structural and functional correspondence between the NTA of the two district PAIs.

Function	Flood PAI	Landslide PAI
Plan documents, definitions and common terminology	Part I—General provisions (Arts. 1–4)	Part I—General provisions (Arts. 1–4)
Planning principles and relationship with spatial and urban planning	Part II—General principles and framework provisions (Arts. 5–6)	Part II—General principles and framework provisions (Arts. 5–6)
Rules by hazard class and provisions for risk areas	Part III—Hazard zones and risk areas (Arts. 7–13)	Part III—Landslide-hazard and risk areas (Arts. 7–11)
Specific areas and elements, incomplete knowledge, planning and mitigation measures	Part IV—Attention areas and flood-management interventions (Arts. 14–19)	Part IV—Potentially hazardous geomorphological elements (Arts. 12–17)
Opinions, coordination, updating, amendments, transitional and final provisions	Part V—Common provisions (Arts. 20–26); Annexes 1–3	Part V—Common provisions (Arts. 18–24); Annexes 1–5

Source: prepared by the author based on the final NTA of the District Flood and Landslide PAIs [42,43].

Appendix B.3. Graded Review Pathways

Hazard class and intervention type jointly determine admissibility and review requirements. The Flood PAI distinguishes among prohibited interventions; interventions subject to an AUBAC opinion and an assessment by the competent Hydraulic Authority; interventions subject only to that assessment; and interventions that do not require a specific opinion under the Plan. The Landslide PAI distinguishes among prohibited interventions; interventions subject to an AUBAC opinion; interventions requiring the opinions of both AUBAC and the Region or delegated authority; interventions subject only to a regional opinion; and interventions for which the NTA do not require a specific opinion. The progression is not identical for every type of intervention. Table A6 and Table A7 present selected examples for the Flood and Landslide PAIs, respectively, without reproducing the full range of cases governed by the NTA [42,43].

Table A6. Examples of graded admissibility and required opinions by hazard class in the District Flood PAI.

Type of Intervention	TI	P3	P2	P1
New construction or extension beyond the existing envelope of public buildings or buildings of public interest; in P3/P2, only where no alternative location is feasible	Prohibited (Art. 8(1)(b))	AUBAC opinion + Hydraulic Authority assessment (Art. 9(3)(e))	AUBAC opinion + Hydraulic Authority assessment (Art. 10(3)(e))	Hydraulic Authority assessment (Art. 11(6)(c))
Building renovation and restoration of collapsed or demolished buildings, without basements or semi-basements	Hydraulic Authority assessment (Art. 8(7)(b))	Hydraulic Authority assessment (Art. 9(7)(c))	Hydraulic Authority assessment (Art. 10(7)(d))	No specific PAI opinion (Art. 11(7)(d))
Basements or semi-basements in demolition and reconstruction, subject to the class-specific conditions	Prohibited (Art. 8(1)(a))	Prohibited (Art. 9(1)(b))	Hydraulic Authority assessment; lawful existing rooms only, for seismic improvement or upgrading (Art. 10(7)(d))	Hydraulic Authority assessment; non-residential/non-commercial use and flood protection (Art. 11(6)(bb))
Public, publicly accessible or private open-air surface parking	Prohibited, except cases in Art. 8(7)(q) (Art. 8(1)(g))	Prohibited, except cases in Art. 9(7)(v) (Art. 9(1)(g))	Hydraulic Authority assessment (Art. 10(7)(m))	No specific PAI opinion (Art. 11(7)(m))

Note: The TI is a functional regulatory zone within flood-hazard zones. “Hydraulic Authority” means the territorially competent authority under the applicable regional or delegated arrangements. “No specific PAI opinion” means that the NTA do not require a specific determination under the Plan; technical assessments and approvals required under urban planning, building, landscape, seismic and sectoral legislation remain applicable. All examples remain subject to the conditions and qualifications of the cited provisions. Hydraulic assessments are acquired where required by the NTA or applicable law. Source: prepared by the author based on the final NTA of the District Flood PAI [42] (Arts. 8–11).

Table A6. Examples of graded admissibility and required opinions by hazard class in the District Flood PAI.

Type of Intervention	TI	P3	P2	P1
New construction or extension beyond the existing envelope of public buildings or buildings of public interest; in P3/P2, only where no alternative location is feasible	Prohibited (Art. 8(1)(b))	AUBAC opinion + Hydraulic Authority assessment (Art. 9(3)(e))	AUBAC opinion + Hydraulic Authority assessment (Art. 10(3)(e))	Hydraulic Authority assessment (Art. 11(6)(c))
Building renovation and restoration of collapsed or demolished buildings, without basements or semi-basements	Hydraulic Authority assessment (Art. 8(7)(b))	Hydraulic Authority assessment (Art. 9(7)(c))	Hydraulic Authority assessment (Art. 10(7)(d))	No specific PAI opinion (Art. 11(7)(d))
Basements or semi-basements in demolition and reconstruction, subject to the class-specific conditions	Prohibited (Art. 8(1)(a))	Prohibited (Art. 9(1)(b))	Hydraulic Authority assessment; lawful existing rooms only, for seismic improvement or upgrading (Art. 10(7)(d))	Hydraulic Authority assessment; non-residential/non-commercial use and flood protection (Art. 11(6)(bb))
Public, publicly accessible or private open-air surface parking	Prohibited, except cases in Art. 8(7)(q) (Art. 8(1)(g))	Prohibited, except cases in Art. 9(7)(v) (Art. 9(1)(g))	Hydraulic Authority assessment (Art. 10(7)(m))	No specific PAI opinion (Art. 11(7)(m))

Note: The TI is a functional regulatory zone within flood-hazard zones. “Hydraulic Authority” means the territorially competent authority under the applicable regional or delegated arrangements. “No specific PAI opinion” means that the NTA do not require a specific determination under the Plan; technical assessments and approvals required under urban planning, building, landscape, seismic and sectoral legislation remain applicable. All examples remain subject to the conditions and qualifications of the cited provisions. Hydraulic assessments are acquired where required by the NTA or applicable law. Source: prepared by the author based on the final NTA of the District Flood PAI [42] (Arts. 8–11).

Table A7. Examples of graded admissibility and required opinions by hazard class in the District Landslide PAI.

Type of Intervention	P4	P3	P2/P1
New construction of public buildings or buildings of public interest, such as hospitals, schools, barracks and elevated parking structures	Prohibited (Art. 9(1)(c))	AUBAC + regional opinions; no alternative solution and assessment or implementation of hazard mitigation (Art. 10(3)(b))	Regional pathway where exposure or occupancy load increases significantly (Art. 11(3)(c); see note)
Urban redevelopment, subject in P3 to prior reduction of hazard or risk	Prohibited (Art. 9(1)(b))	AUBAC opinion + regional opinion (Art. 10(3)(a))	Regional pathway where Art. 11(3) applies; otherwise no specific PAI opinion if all Art. 11(4) conditions are met
New road, railway and port infrastructure	AUBAC opinion; prior alternatives analysis and implementation of hazard mitigation (Art. 9(2)(a))	AUBAC opinion; prior alternatives analysis and implementation of hazard mitigation (Art. 10(2)(a))	Regional pathway for infrastructure capable of affecting slope stability (Art. 11(3)(g); see note)

Note: “No specific PAI opinion” means that the NTA do not require a specific determination under the Plan; technical assessments and approvals required under urban planning, building, landscape, seismic and sectoral legislation remain applicable. All examples remain subject to the conditions and qualifications of the cited provisions. Under Landslide Article 11(3), where regional law does not require a separate opinion, the permitting authority verifies the filing of the required technical documentation and may examine its merits or request further information. Source: prepared by the author based on the final NTA of the District Landslide PAI [43] (Arts. 9–11).

Table A7. Examples of graded admissibility and required opinions by hazard class in the District Landslide PAI.

Type of Intervention	P4	P3	P2/P1
New construction of public buildings or buildings of public interest, such as hospitals, schools, barracks and elevated parking structures	Prohibited (Art. 9(1)(c))	AUBAC + regional opinions; no alternative solution and assessment or implementation of hazard mitigation (Art. 10(3)(b))	Regional pathway where exposure or occupancy load increases significantly (Art. 11(3)(c); see note)
Urban redevelopment, subject in P3 to prior reduction of hazard or risk	Prohibited (Art. 9(1)(b))	AUBAC opinion + regional opinion (Art. 10(3)(a))	Regional pathway where Art. 11(3) applies; otherwise no specific PAI opinion if all Art. 11(4) conditions are met
New road, railway and port infrastructure	AUBAC opinion; prior alternatives analysis and implementation of hazard mitigation (Art. 9(2)(a))	AUBAC opinion; prior alternatives analysis and implementation of hazard mitigation (Art. 10(2)(a))	Regional pathway for infrastructure capable of affecting slope stability (Art. 11(3)(g); see note)

Note: “No specific PAI opinion” means that the NTA do not require a specific determination under the Plan; technical assessments and approvals required under urban planning, building, landscape, seismic and sectoral legislation remain applicable. All examples remain subject to the conditions and qualifications of the cited provisions. Under Landslide Article 11(3), where regional law does not require a separate opinion, the permitting authority verifies the filing of the required technical documentation and may examine its merits or request further information. Source: prepared by the author based on the final NTA of the District Landslide PAI [43] (Arts. 9–11).

Appendix B.4. Procedural Instruments and Distribution of Functions

The NTA distinguish the technical documents and determinations involved in each review. In the Flood PAI, the hydrological–hydraulic study is prepared by the proponent; the hydraulic compatibility verification demonstrates that the intervention is compatible with the flood scenarios and Plan criteria; the hydraulic compatibility assessment is the review conducted by the competent Hydraulic Authority; and the AUBAC opinion concerns compatibility at the river basin scale. Where both determinations are required, the authority responsible for the procedure submits the applications simultaneously or obtains the determinations within a services conference (conferenza di servizi), avoiding a mandatory sequence between administrations [42].

In the Landslide PAI, the Technical Verification (“Verifica Tecnica”) is the principal document demonstrating compatibility with existing instability and risk conditions. Only the interventions specified in Article 8(2), without significant excavation or earthworks, foundation or consolidation works, or increased occupancy or urban-development load, may instead use a certified statement of no aggravation (“Attestazione asseverata di non aggravamento”). Where opinions of both AUBAC and the Region or delegated authority are required, applications are submitted simultaneously or addressed through a services conference [43].

The provisions are coordinated with the Consolidated Building Act, including Article 89 of Presidential Decree No. 380/2001 on opinions concerning urban planning instruments [68]; the administrative procedure law, Law No. 241/1990 [69]; and sectoral hydraulic legislation, Royal Decree No. 523/1904 [70]. In the Flood PAI, Article 3(2) treats Areas of Potential Significant Flood Risk (APSFR) as elements of the knowledge base that do not by themselves introduce prohibitions or requirements; Annex 2 sets the minimum content, scenarios, modeling assumptions and pre- and post-project comparisons for hydrological–hydraulic studies. In the Landslide PAI, the annexes specify the documentation required for interventions subject to an AUBAC opinion, mitigation works and map updates, and introduce a unified landslide survey form; the persistent scatterer interferometric synthetic aperture radar (PS-InSAR) guidelines remain a non-prescriptive technical support document [42,43].

Incorporation into urban planning instruments is not the source of the legal effect of the PAI provisions but the process through which local plans, maps and procedures are brought into conformity with them. Table A8 summarizes the distribution of functions among the main institutional levels [42,43].

Table A8. Distribution of functions in the district PAI framework.

Institution	Main Functions
AUBAC	Ensures coherence at the river basin scale; defines hazard classes and minimum protection thresholds; issues opinions within its remit; governs the separate procedures for map updates and regulatory amendments
Regions	Regulate incorporation at the territorial scale; exercise or delegate the functions of the competent Hydraulic Authority; issue the opinions assigned by the Landslide PAI; organize information on the secondary river network; submit proposals and datasets for updates
Municipalities and other procedure authorities	Incorporate delineations into urban planning instruments; assess interventions through procedures within their remit; contribute to updating by transmitting place-based knowledge

Source: prepared by the author based on the final NTA of the District Flood and Landslide PAIs [42,43].

Table A8. Distribution of functions in the district PAI framework.

Institution	Main Functions
AUBAC	Ensures coherence at the river basin scale; defines hazard classes and minimum protection thresholds; issues opinions within its remit; governs the separate procedures for map updates and regulatory amendments
Regions	Regulate incorporation at the territorial scale; exercise or delegate the functions of the competent Hydraulic Authority; issue the opinions assigned by the Landslide PAI; organize information on the secondary river network; submit proposals and datasets for updates
Municipalities and other procedure authorities	Incorporate delineations into urban planning instruments; assess interventions through procedures within their remit; contribute to updating by transmitting place-based knowledge

Source: prepared by the author based on the final NTA of the District Flood and Landslide PAIs [42,43].