Reference
Look things up here once you know what you are after. The table pages and the code lists are generated from the schema that serves the data, so they change only when the model does. For the reasoning behind the tables, read The model at a glance first; for a task-shaped answer, the SQL guides are quicker.
- Code lists: every controlled value a coded column can hold, with its meaning.
- Open data file columns: every column of a downloaded or exported file, and every sheet after the first, each under its own name.
- Downloads: the lists a spreadsheet reader wants whole, as
.xlsxworkbooks regenerated whenever a list changes: geography (every region code, with its parent and the members of each group), categories, system boundaries, classification schemes, units and licence terms. Each opens with an About sheet that names the page explaining it. - Tables: one page per published table of the model, with every field, its type and what it means. Filed in that same group, and the page to read first, is
factors_flat: it is a view rather than a table, and its columns are what a query actually selects. - API reference: the factor API and the mapping endpoints, all under
/v2/, generated from the OpenAPI description, with a console beside each operation that sends a real request using your own key. Call the API is the narrative version of the factor half.
What is in the contract#
For a SQL reader the supported surface is a list, not a rule of thumb:
- the view
open_ef.factors_flat, which is what nearly every query on this site starts from; - every table with a page in this section, for auditing a number back to the rows it came from. That now includes
open_ef.license_terms,open_ef.price_indexandopen_ef.fx_annual: the plain answers behind alicense_code, and the two vendored World Bank series behind a restated monetary value, all three of which the guides join by name.
Schema evolution is additive is a promise about that list.
The rest of the open_ef schema is working state: run bookkeeping, the ledgers and indexes behind a served column, and the crosswalk out to consumer-side systems. Everything in them you are meant to read is already on factors_flat, which applies the release and quality predicates for you and carries the licence columns on every row. Two of them look readable and are not: factor_year_index is a copy, only as current as its last refresh, and activity_search is the index the search layer queries. Read the view instead. When it changes, the changelog says what changed and what to read instead.
In this section
- MCP toolsEvery tool on the open-climate.ai MCP server, with its purpose, its inputs with defaults and ranges, and the fields it returns.
- GlossaryPlain-English definitions of the terms used on factor pages, in factor search and across these docs, from unit basis and boundary to CO2e basis.
- ChangelogHow the served view, the data model and the libraries may change, what stays stable while they do, and every change so far.
- Code listsEvery closed code list in the open_ef model: system boundaries, classification schemes, units, GWP bases and licence tiers.
- Open data file columnsEvery column of an open data file or an exported list, with what it holds, the values a cell can take and where it is defined.
Tables
- activity_classification
- boundary_scope_term
- category_subject_kind
- classification_scheme
- data_quality_scheme
- emission_factor
- emission_factor_alias
- emission_factor_text
- emission_factor_value
- factors_flat
- fx_annual
- geography
- geography_member
- geography_resolution
- license_terms
- price_index
- publisher
- publisher_boundary_scope
- release
- source_row
- supplier
- supplier_resolution
- system_boundary
- system_boundary_scope
- text_term