Important Information for Analysts and Authors

Terms of use agreement

Proper citation and acknowledgement


How to access CMIP3 archive and getting started

Output availability at a glance

Check errata page frequently


List of simulation specifications

Precise definitions of variables

Official model names

Model documentation (including simulation forcing information)

Which control simulation (and time period) should each perturbation run be compared to?


IPCC AR4 historical details

Terms of use agreement

These data are for use in research projects only. A 'research project' is any project carried out by an individual or organized by a university, a scientific institute, or similar organization (private or public) for non-commercial research purposes only. Results based on these data must be submitted for publication in the open literature without any delay linked to commercial objectives.

An additional restriction is that the UK Met. Office / Hadley Centre requires strict adherence to the conditions set forth in their "License Statement", which applies to their model output:

"These data are licensed for use in Research Projects only. A 'Research Project' is any project organised by a university, a scientific institute, or similar organisation (private or public), for non-commercial research purposes only. A necessary condition of the recognition of non-commercial purposes is that all the results obtained are openly available at delivery costs only, without any delay linked to commercial objectives, and that the research itself is submitted for open publication.

Data provided by the UK Met. Office / Hadley Centre are expected to be acknowledged by:

(c) Crown copyright 2005, Data provided by the Met Office Hadley Centre"

Proper citation and acknowledgement

In first making reference to the data obtained from the CMIP3 archive, please include the phrase "the World Climate Research Programme's (WCRP's) Coupled Model Intercomparison Project phase 3 (CMIP3) multi-model dataset".  Subsequent references within the same publication might refer to the data with terms such as "CMIP3 data", "the CMIP3 multi-model dataset", "the CMIP3 archive", or the "CMIP3 dataset".

 Whenever you publish research based on model output from the CMIP3 database, you should include the following acknowledgement:

"We acknowledge the modeling groups, the Program for Climate Model Diagnosis and Intercomparison (PCMDI) and the WCRP's Working Group on Coupled Modelling (WGCM) for their roles in making available the WCRP CMIP3 multi-model dataset.  Support of this dataset is provided by the Office of Science, U.S. Department of Energy."

[Note that the acronyms, WCRP and CMIP3, should have already appeared in the main text of your article, where they were presumably defined. You may also have already defined "PCMDI" and "WGCM", which would allow you to simply refer to them here using the acronyms.]

You may also find it useful to reference the following article, which provides an overview and documentation of the CMIP3 modelling activity:

Meehl, G. A., C. Covey, T. Delworth, M. Latif, B. McAvaney, J. F. B. Mitchell, R. J. Stouffer, and K. E. Taylor, 2007: The WCRP CMIP3 multi-model dataset: A new era in climate change research,Bulletin of the American Meteorological Society, 88, 1383-1394.

How to access CMIP3 archive and getting started

Register to download output

Getting started and other information

Check errata page

In some cases data contributed to the database have been found to contain errors of various sorts. Certain files are then withdrawn or replaced and an entry is logged on the errata page:
 The errata page should be checked frequently to avoid analyzing "garbage", and to determine whether any of the files you have downloaded from the PCMDI site need to be updated or removed. If any of you find undetected errors in the data you are analyzing, please contact the responsible modeling group and copy

One general warning concerning the global attributes in each file -- the "title" and "comment" attributes may in some cases incorrectly identify the experiment, whereas the "experiment_id" will always be correct.  If there is a conflict, rely only on experiment_id.

List of Simulation Specifications

Click here for experiments and time periods requested from modeling groups.

Precise Definition of Variables

Click here for precise definitions of all model output fields collected.

Official model names

To minimize confusion and maximize uniformity across analyses, authors are urged to use the official CMIP3 model "I.D.'s" when referring to individual model results.  In addition it would usually be appropriate to include a table containing the names of the modeling groups along with citations to their relevant papers.

Model documentation

Click here for available documentation of models and their simulations, which have been contributed to the WCRP CMIP3 multi-model dataset. Note that for most models you can find information concerning the forcing that was applied in each simulation.

Which control simulation (and time period) should each perturbation run be compared to?

In order to compare each simulation to the corresponding time period of the control simulation, you will need the information contained in the following table, available as: