Climate and Forecast (CF) metadata

Climate and Forecast (CF) metadata
Name	Climate and Forecast (CF) metadata
Acronym	CF
Domain	Meteorology; Oceanography; Atmospheric Sciences
First published	2003
Latest release	ongoing
License	Open specification
Repository	Community-maintained

Climate and Forecast (CF) metadata The Climate and Forecast (CF) metadata conventions are a community-driven specification for describing gridded, point, and observational geophysical data in machine-readable files. The conventions enable interoperability among software engines, data centers, and research projects by prescribing standardized attributes and coordinate constructs for datasets produced by institutions such as National Oceanic and Atmospheric Administration, European Centre for Medium-Range Weather Forecasts, National Aeronautics and Space Administration, World Meteorological Organization, and Intergovernmental Panel on Climate Change. CF supports exchange among formats and tools used by Met Office, NOAA National Centers for Environmental Prediction, Scripps Institution of Oceanography, NASA Jet Propulsion Laboratory, and European Space Agency projects.

Overview

CF defines a controlled vocabulary and structural rules for embedding metadata in array-oriented formats like Network Common Data Form, NetCDF Classic, and Hierarchical Data Format. The conventions ensure that variables representing geophysical quantities are annotated with attributes enabling semantic discovery, algorithmic processing, and visualisation by software such as Panoply (software), ncview, CDO (Climate Data Operators), NCL (NCAR Command Language), and xarray (Python) ecosystems. CF is widely used in initiatives including Coupled Model Intercomparison Project, World Climate Research Programme, Global Climate Observing System, Copernicus Programme, and regional centers like Hadley Centre.

History and development

Work on CF began in the early 2000s as community groups from National Center for Atmospheric Research, NOAA, Met Office, and ECMWF sought to standardize metadata practices across modeling and observational communities. The original specification drew on precedents from GRIB, DODS/OPeNDAP, and ISO 19115 while integrating requirements from programs such as Argo (oceanography), Global Ocean Observing System, and Ocean Observatories Initiative. Governance evolved through steering groups, community workshops at venues like AGU Fall Meeting and EGU General Assembly, and contributions from projects funded by agencies including National Science Foundation and European Research Council.

Key concepts and conventions

CF centers on concepts such as standard_name, units, long_name, and attributes specifying missing values, cell methods, and bounds. The standard_name table formalizes names for quantities measured in atmospheres, oceans, and cryosphere fields used by IPCC Assessment Report authors and intercomparison projects like AMIP and CMIP6. Conventions define how to represent time axes in compliance with calendars used by ISO 8601-based models, how to annotate vertical coordinates for applications in International Hydrographic Organization workflows, and how to declare provenance metadata useful to archives such as British Atmospheric Data Centre and Earth System Grid Federation.

Coordinate systems and grid mappings

CF prescribes representations for geographic coordinates (latitude, longitude), vertical coordinates (pressure, height, depth), and time coordinates, enabling interoperable descriptions of structured, unstructured, curvilinear, and hybrid grids. Grid mapping attributes permit encoding of map projections used by Lambert Conformal Conic, Mercator projection, Polar Stereographic, and Rotated Pole systems, facilitating transformation by libraries like PROJ and Geospatial Data Abstraction Library. CF also supports mosaic and composite grids relevant to regional modeling systems such as WRF and ocean models like ROMS and HYCOM.

Metadata attributes and standard names

The conventions enumerate required and optional global and variable-level attributes including conventions, title, institution, source, history, references, comment, standard_name, units, and ancillary_variables. The standard_name table is maintained collaboratively and maps lexical names to detailed definitions and canonical units used by communities around NOAA GFDL, Météo-France, CSIRO, PBL Netherlands Environmental Assessment Agency, and Japanese Meteorological Agency. Classifiers such as cell_methods capture aggregation procedures (mean, sum, maximum) used in product suites from Copernicus Climate Change Service and reanalysis datasets like ERA5 and NCEP/NCAR Reanalysis.

Tools, implementations, and adoption

Many libraries and tools implement CF conventions: readers and validators exist in CDO (Climate Data Operators), NCO (NetCDF Operators), NetCDF-Java, Python netCDF4, and cf-python. Validation tools and metadata editors are provided by organizations like UK Met Office and communities around Earth System Documentation (ES-DOC). Major data archives and portals including ESGF, NOAA National Centers for Environmental Information, Copernicus Data Store, and PANGAEA (data publisher) publish CF-compliant products to ensure discoverability and downstream usability.

Use cases and community governance

CF metadata underpin use cases from multi-model intercomparison and climate impact assessments used by IPCC authors to operational forecasting and satellite product calibration performed by EUMETSAT and NASA Earth Science. Community governance combines stewardship by a CF Executive Committee, public mailing lists, and collaborative maintenance of standard_name tables via version-controlled repositories and meetings at conferences such as AGU and EGU. Adoption is reinforced through training activities run by World Climate Research Programme and interoperability testing in projects like AERIS, GEOSS, and national research infrastructures.

Category:Climate data standards