The Green Data Acquisition Framework establishes the information architectural foundations and standardized data models required to build and run the GreenInCities Data Marketplace. Modern smart-city projects are often held back by "data silos" where different sensors, digital twins, and municipal databases cannot communicate.

D4.1 resolves this by defining a unified Domain Model and translating it into a set of interoperable, machine-readable NGSI-LD data models. These models ensure that every real-world urban greening activity, sensor reading, key performance indicator (KPI), and community co-creation activity shares a common technical language.

To enable open, secure, and scalable real-time context management, the framework combines FIWARE context-management components (such as the Orion-LD Context Broker) with Sensative’s Yggio horizontal IoT middleware. This hybrid model unifies highly heterogeneous on-site devices and legacy networks while maintaining strict data sovereignty and fine-grained access control for each pilot municipality.

Primary Audience:

• Planners, researchers, and technologists

• GreenInCities Consortium Technical Partners

Secondary Audience:

• Urban Data Managers, Smart City Officers and Municipal IT departments

• The 112 Climate-Neutral & Smart Cities network

The deliverable outlines a rigorous methodological path to translate abstract urban planning concepts into implementation-ready code schemas:

• High-level domain modeling: Identifying the key components, abstract behaviors, and interrelationships within the project’s scope without considering technical database restrictions.

• Standards review & deconstruction: Conducting an in-depth review of EU smart-city guidelines, FIWARE background standards, smart data ontologies, and Next Generation Service Interface - Linked Data (NGSI-LD) specifications developed by ETSI.

• Entity-relationship mapping: Using Unified Modeling Language (UML) diagrams to systematically map how high-level domain entities structurally link to one another.

• Information model translation: Encoding the mapped associations into formal NGSI-LD schemas, defining precise attributes, mandatory cardinalities, data types, and inheritance rules.

• Linked-data validation: Writing concrete JSON-LD context files and URN schemas (utilizing unique Uniform Resource Names and timestamps) to allow automated verification of live data feeds.

The framework operates on a dual-layer structure to translate abstract urban planning concepts into structured, machine-readable code:

• The Conceptual Domain Model: Places the Intervention entity at the absolute center of the data flow. It structurally links urban areas, timed planning tasks, evaluation phases (ex-ante/ex-post baselines), and KPIs. Crucially, it introduces the Nature Stakeholder entity, formally integrating non-human elements (local fauna/flora) directly into the smart-city data architecture.

• The NGSI-LD Information Model: Translates these conceptual blocks into ETSI-standardized Next Generation Service Interface - Linked Data (NGSI-LD) schemas. D4.1 formally registers nine open-source models, including AssessmentTool (sensor metadata), KeyPerformanceIndicator (KPI metrics), Dataset (storage formats), and Area (geospatial polygons).

• Eliminates data silos: Provides a unified technical language that allows diverse IoT sensors, 3D digital twins, and municipal platforms to exchange data.

• Guarantees data sovereignty: Leverages the open-source FIWARE broker and SENSATIVE’s Yggio middleware to enforce fine-grained, secure access control for each municipality.

• Enables replication: Standardizing around European ETSI smart-city ontologies ensures that our tools can be effortlessly adopted and deployed by the 112 Climate-Neutral and Smart Cities network.

• Figure 3: Domain Model of GreenInCities

• Figure 1 : NGSI-LD Information Model UML

• Table 4: Assessment Tool Data Model

• Table 5: Key Performance Indicator Data Model

• Table 7: Intervention Data Model