Since 1995, the Coupled Model Intercomparison Project (CMIP) has coordinated climate model experiments involving multiple international modelling teams. This has led to a better understanding of past, present, and future climate, and CMIP model experiments have routinely been the basis for future climate change assessments made by the Intergovernmental Panel on Climate Change (IPCC).
The sixth phase of the CMIP process (CMIP6) consists of a handful of common experiments, the DECK (Diagnostic, Evaluation and Characterisation of Klima) and CMIP6 historical simulations (1850-near present), and an ensemble of CMIP-Endorsed Model Intercomparison Projects (MIPs). Together the CMIP6 experiments address three broad questions:
- How does the Earth system respond to forcing?
- What are the origins and consequences of systematic model biases?
- How can we assess future climate changes given internal climate variability, predictability, and uncertainties in scenarios?
CMIP6 consists of 21 Endorsed MIPs and 244 separate experiments. Descriptions of all the CMIP6 experiments can be found in the academic literature, in summary tables produced by PCMDI and also in the CMIP6 data request. So why use the Common Information Model (CIM) to document the CMIP6 experiments?
CIM provides the formalism to collect detailed information in a standard way across the 17 MIPs that are contributing experiments to CMIP6. The scope of the CIM is such that we can capture and render information gathered from all the sources mentioned above and more. As such the ES-DOC metadata and services provide a unified source for CMIP6 information which is especially important given the federated nature of the CMIP6 project.
The experiments in the ES-DOC database consist of lists of forcings, model configuration requirements, information about building the ensembles, links to citations and contact information of the principle investigators as well as text descriptions and information about the rationale behind each experiment. The experiment information was gathered from statements about the experiments found in the academic literature, the MIP submissions to the WCRP, the PCMDI summary tables and correspondence with the principle investigators for each MIP. The database of CMIP6 experiments has been collated using spreadsheets which are archived in the ES-DOC github repository and then rendered on the ES-DOC website.
The workflow for collecting CIM metadata for the CMIP6 experiments is outlined in the figure below.
Recent updates to the CIM allow for the same forcing constraint to be referred to by more than one experiment, the same is true for many experiment attributes, and as such we are now able to ask questions such as:
- Which MIPs require experiment A?
- Which experiments use forcing constraint B?
- Which experiments have experiment C as their control?
An advantage of allowing forcing constraints (and other attributes) to be referred to by more than one experiment is that we can streamline the process of collecting information about how the modelling groups set up their models to meet the specifications of each experiment. The information that the modelling groups provide to ES-DOC will be used to generate model forcing summary tables such as table 12.1 in the IPCC AR5 WGI report.