The main objectives of the METAFOR project were to develop and promulgate an ipso-facto standard for describing climate models and associated data.
This standard has been formalized and named the Common Information Model (CIM).
Adoption of the CIM standard will allow the climate science community to nurture an eco-system of CIM compliant tools and services to be integrated into the day to day activities of climate research institutes worldwide.
Such an eco-system is essential to encouraging adoption of the CIM by the global climate modelling community.
The benefits of adopting a internationally recognized information standard such as the CIM are many:
- Institutes can share information regarding models, experiments, simulations, data, grids ... etc, in a standardized format;
- Documentation tools can render CIM compliant information in multiple formats such as html, pdf ...etc;
- Inter-operability issues can be resolved by adapting existing software to leverage the CIM;
- Search tools can provide a unified metadata access by ingesting CIM compliant metadata from multiple sources;
- Differencing engines can attempt to compare climate models.
The CIM is an ontology, i.e. an informational model describing a particular domain (i.e. climate science).
Such a model is formed using a construct known as a class (e.g. simulation).
Classes form relationships with other classes (e.g. a simulation has data).
Related classes are grouped into packages.
The CIM is formally defined using the Unified Modelling Language and is organised into the following packages:
- Activity - classes describing the experiments in support of which software is run.
- Data - classes describing actual archived data emitted during the course of climate simulations.
- Grids - classes describing the geographic grids that data is mapped onto, software adheres to, and activities constrain.
- Quality - classes describing the quality of metadata or the data/models that metadata describes.
- Shared - general classes used throughout the CIM
- Software - classes describing the software that produced data during the course of a simulation.
Built ontop of the CIM standard are a set of community developed tools & web services collectively referred to as the CIM eco-system.
Such an eco-system acts as an infrastructure via which adoption of the CIM by the international climate modelling community is promoted.
Eco-systems tend to evolve in non-predictive ways, and the CIM eco-system is already being leveraged in surprising ways.
For example metadata entered into the CMIP5 questionnaire is appearing in web-sites run by the impacts community.
The University of Manchester is developing a Fortran code parser that generates CIM compliant metadata thereby auto-documenting climate model source code.
Both examples leverage the CIM eco-system.
Within this website you will find links to learning resources such as webcasts, publications & presentations.
You can read about the to which the CIM is targeted.
You can view the different packages found within the CIM ontology outlined above.
Alternatively you may wish to search the CIM Repository in order to view CIM related metadata generated by projects such as CMIP5.
You can also use the CIM tools to view & validate CIM metadata you may yourself have generated.
Metafor Open Dissemination Workshop - Part 1
Metafor Open Dissemination Workshop - Part 2
Metafor: managing metadata for climate models
Sarah Callaghan, Reinhard Budich, Gerry Devine, Eric Guilyardi, Bryan Lawrence and Sophie Valcke
The CMIP5 model and simulation documentation: a new standard for climate modelling metadata
Eric Guilyardi, V. Balaji, Sarah Callaghan, Cecelia DeLuca, Gerry Devine, Sebastien Denvil, Rupert Ford, Charlotte Pascoe, Michael Lautenschlager, Bryan Lawrence, Lois Steenman-Clark, Sophie Valcke.
The METAFOR project: preserving data through metadata standards for climate models and simulations.
Callaghan et al.
The CMIP5 questionnaire: web-based metadata collection for climate modelling.
Gerard Devine, Bryan Lawrence, Charlotte Pascoe, Rupert Ford, Paul Slavin, and Metafor Team.
Supporting the climate community by providing common metadata for climate modelling digital repositories: the METAFOR project.
Sarah Callaghan, Mark Morgan, Eric Guilyardi, Sophie Valcke, Charlotte Pascoe, Bryan Lawrence and the METAFOR Project Team.
A Common Information Model paired with scientific Controlled Vocabularies for Climate Models and Statistical Downscaling.
Charlotte Pascoe, Marie-Pierre Moine, Allyn Treshansky, Gerard Devine, Sebastien Denvil, Michel Kolasinski, and Rupert Ford.
A Common Information Model (CIM) for the climate modeling process.
Allyn Treshansky, Gerard Devine, and the METAFOR Team.
The CMIP5 questionnaire: web-based metadata collection for climate modelling.
Gerard Devine, Bryan Lawrence, Charlotte Pascoe, Rupert Ford, Paul Slavin, and Metafor Team.
The CMIP5 questionnaire: web-based metadata collection for climate modelling.
Gerard Devine, Bryan Lawrence, Charlotte Pascoe, Rupert Ford, Paul Slavin, and Metafor Team.
The METAFOR project: providing community metadata standards for climate models, simulations and CMIP5.
Sarah Callaghan, Eric Guilyardi and the Metafor Team.
Standard controlled vocabulary for climate models.
Marie-Pierre Moine, Charlotte Pascoe, Eric Guilyardi, Rupert Ford and the METAFOR Team.
The CMIP5 Model Documentation Questionnaire: Development of a Metadata Retrieval System for the METAFOR Common Information Model.
Charlotte Pascoe, Bryan Lawrence, Marie-Pierre Moine, Rupert Ford, and Gerry Devine.
WDCC Metadata Generation with GeoNetwork.
Hans Ramthun, Michael Lautenschlager, and Hans-Hermann Winter.
Models, Metadata and Metafor.
B. Lawrence, E. Guilyardi, A. Treshansky, and S. Valcke.
Developing a Common Information Model for climate models and data.
S. Valcke, V. Balaji, P. Bentley, E. Guilyardi, B. Lawrence, C. Pascoe, L. Steenman-Clark, F. Toussaint, and A. Treshansky.
Home - Home
