The objective of the Coupled Model Intercomparison Project (CMIP) is to better understand past, present and future climate changes arising from natural, unforced variability or in response to changes in radiative forcing in a multi-model context. This understanding includes assessments of model performance during the historical period and quantifications of the causes of the spread in future projections. Idealized experiments are also used to increase understanding of the model responses. In addition to these long time scale responses, experiments are performed to investigate the predictability of the climate system on various time and space scales as well as making predictions from observed climate states. An important goal of CMIP is to make the multi-model output publically available in a standardized format.
CMIP began in 1995 under the auspices of the Working Group on Coupled Modelling (WGCM). The first set of common experiments involved comparing the model response to an idealized forcing - a constant rate of increase which was accomplished using a CO2 increase of 1% per year compounded. Since that time a number of CMIP experiments have been developed. The experiments continue to include integrations using idealized forcings to facilitate understanding. They now also include integrations forced with estimates of the changes in the historical radiative forcings as well as estimates of the future changes.
In addition to the major experiments, CMIP has also included a series of smaller model intercomparison efforts, called the Coordinated CMIP Experiments, designed to understand specific aspects of the model response. One experiment investigated the Atlantic Meridional Overturning Circulation (AMOC), or thermohaline circulation (THC) response to changes in the surface fluxes of heat and water. Another Coordinated Experiment documented the model response to external surface water flux in high latitudes of the North Atlantic, the so-called "water hosing experiments". A number of papers have been written using these datasets.
From the inception of CMIP, there has been a focused effort to make the model intercomparison data available to other scientists besides those who run the models. This effort requires the development of numerous collaborations to solve various problems, for both hardware (how does one store/search/find what you want in so large datasets?) and software (how do I get the ocean data for the area off the west coast of the US?). The most important of these collaborations is with the US Department of Energy (DOE) Program for Climate Model Diagnosis and Intercomparison (PCMDI). For more detailed information about the various CMIP efforts and datasets, see PCMDI CMIP.
All CMIP activities are overseen by the CMIP Panel which is a standing subcommittee of the WGCM. It is the responsibility of the CMIP Panel to follow the guidance of the WGCM committee. The CMIP Panel oversees the design of the various CMIP experiments, oversees the various CMIP datasets and helps to resolve problems that arise during the model integrations/data generation phase or in the use of the datasets.