Climate Action: Official Release of the Anthropogenic Impact Accounting Ontology Suite
Understanding the challenge
Across climate and sustainability programs, data about human impacts on the environment is fragmented and inconsistently named. Projects use different schemas and vocabularies; even when the underlying methods align, the data often does not. The result is friction for integrators and auditors and limited verification of evidence across platforms.
The working group and the process
The Climate Action and Accounting Special Interest Group (CA2-SIG), originally launched under Hyperledger and now part of Linux Foundation Decentralized Trust (LFDT), was formed in 2021 as an open forum comprising domain experts, implementers, and researchers. Their objective was to leverage blockchain and other distributed ledger technologies to advance climate action, enhance transparency in carbon accounting, and support sustainable practices across industries.
Soon a Standards Working Group (WG) was established within the CA2-SIG. The mission was to distill an architecture-neutral, common language for impact accounting, of which climate accounting was one specific use case.
The group quickly realized that the key to developing a vocabulary that can serve such a broad industry and withstand the test of time is to root the vocabulary in a proper ontology. A simple premise anchored the design: "An agent engages in an activity that impacts an environment." From this triad (agent - activity - environment) they built up the ontology and added roles, parameters, instruments, and indicators, as well as the semantics of claims, attestations, and communication.
During their biweekly working sessions, the WG iterated from whiteboard models to formal OWL/RDF representations, testing against use cases from carbon accounting, environmental registries, and ESG reporting.
A notable chapter in the process was a mentorship program during the 2022–2023 Southern Hemisphere summer, where three South African ICT students contributed to the development process under guidance from SIG members. The students helped convert diagrams to OWL, authored Turtle examples, and prototyped a triple-store for testing. This combination of rigour and practical experimentation laid the foundations for a model that was both conceptually coherent and implementation ready.
The road to release
By late 2024, the group had a working ontology with dozens of classes and a set of core axioms that captured fundamental concepts of impact accounting. Examples of axioms included:
“An Activity has Input(s) and Output(s).“
“An Activity impacts an Environment.”
“An Agent enacts a Role(s) within the context of an Activity.”
“A Claim has a Claimant and can be supported by a Substantiation.”
The team validated the framework by mapping sample datasets from well-known programs (e.g., Gold Standard and CDM) to the ontology's concepts, and by tracing end-to-end scenarios from project activity to reported impact and independent attestation.
At the time, however, all the classes and axioms were still part of one, large ontology. This made the ontology difficult for users to grasp and navigate. The WG consequently decided to split the ontology up into a set of smaller ontologies according to the principle of separation of concerns. That effort eventually produced the Anthropogenic Impact Accounting Ontology (AIAO) and three companion ontologies: the Claim Ontology, the Impact Ontology, and the Information Communication Ontology.
The first version of the suite was finally released on September 16, 2025, and is now publicly available on W3ID as the following registered namespaces:
- claimont: https://w3id.org/claimont. Semantics of claims, claimants, evidence (substantiation), and attestations. It provides the vocabulary to state, support, and audit assertions of any nature (not just impact claims).
- infocomm: https://w3id.org/infocomm. Communication primitives for information exchange, including CommunicationEvent, Information, CommunicationParty, Medium, and Format. This can be used for any kind of information, not just impact claims.
- impactont: https://w3id.org/impactont. An ontology for the foundational concepts of impact (i.e., how events {general} impact things {general}). Several of the classes and properties of this ontology are extended by AIAO to describe anthropogenic impact specifically.
- aiao: https://w3id.org/aiao. This is the culmination of the suite into a framework for anthropogenic impact. It relies heavily on and extends many of the classes and properties of the other three ontologies.
All four ontologies are open source (Apache-2.0) and designed to be reused, extended, and mapped to existing standards. Together they provide a framework for describing who did what, with what consequence, how it was evidenced, and how those assertions can be verified and communicated. The framework's vocabulary was carefully chosen to be both human friendly and machine readable.
Why it matters
Standardizing the semantics of impact data is a prerequisite for trustworthy, decentralized impact MRV infrastructure. Without a shared vocabulary, ledgers, analytics and audit tools cannot reason over data in a comparable way. With the AIA suite:
- platforms can exchange and merge impact data
- impact evidence can be verified across platforms
- auditors can follow semantically explicit links from claims to evidence
- developers can reuse common classes and properties instead of inventing new ones
- communities can map legacy schemas into a common model rather than replacing them
Getting started with the AIA ontology suite
If you are interested in trying out the ontology suite, here are a few simple steps to help you get started:
- Pick a narrow use case relevant to your team (e.g., representing an emissions reduction claim for a single project).
- Model it using classes and properties from the AIA suite. Begin by identifying your agents, activities, and environments. Enumerate roles and define the indicators you calculate. Treat methods or protocols as controls on activities.
- Preserve stable IDs from your source systems; use IRIs that dereference to documentation where possible.
- Serialize your model and take it for a round trip through your pipeline. JSON-LD is a practical choice for application developers; OWL/Turtle is ideal for ontology work and reasoning.
- Add SHACL shapes to assert local constraints (e.g., minimum properties to be reported for an activity). Shapes can complement OWL without overconstraining the ontology.
- Share feedback, examples, and gaps with the Standards WG to inform the next releases!
Note: The ontologies are top-level ontologies. We encourage implementers to develop and/or propose extensions rather than overloading core classes and properties. Domain-specific modules (e.g., water use or biodiversity conservation) can live in separate namespaces while extending AIAO superclasses. This will help to keep the core stable and maximize reuse.
How to contribute
If your team is building climate ledgers, ESG reporting systems, or data exchange frameworks, consider piloting the AIA suite and telling us what works and what does not. We welcome issues, implementations, and mappings from collaborators and interested parties. A practical, community-hardened standard emerges only through real use.
Credits
Main contributors
Christiaan Pauw (https://www.linkedin.com/in/christiaan-pauw-a5034533/)
Alex Howard (https://www.linkedin.com/in/alexhoward1/)
Alfonso Govela (https://www.linkedin.com/in/alfonsogovela/)
Kit Blake (https://www.linkedin.com/in/kitblake/)
Additional contributors
Kyle Robinson (https://www.linkedin.com/in/kylegrobinson/)
Jeff Pribich (https://www.linkedin.com/in/jeff-pribich-9483/)
Abhi Sarvagyam (https://www.linkedin.com/in/abhi-sarvagyam/)
Elisabeth Green (https://www.linkedin.com/in/c-elisabeth-green-bb5369105/)
Lucia Pauw (https://www.linkedin.com/in/lucia-pauw/)
Pieter-Ruan Cronjé (https://www.linkedin.com/in/pieter-ruan-cronje/)
Wouter Dercksen (https://www.linkedin.com/in/alfred-wouter-alwo-dercksen-773a65374/)