OVERVIEW OF THE COUPLED MODEL INTERCOMPARISON PROJECT

The Coupled Model Intercomparison Project (CMIP) involves study and intercomparison of multi-model simulations of present and future climate. The simulations of the future use idealized forcing in increase is compounded which CO₂ 1% yr−1until it doubles (near year 70) with global coupled models that contain, typically, components representing atmosphere, ocean, sea ice, and land surface. Results from CMIP diagnostic subprojects were presented at the Second CMIP Workshop held at the Max Planck Institute for Meteorology in Hamburg, Germany, in September 2003. Significant progress in diagnosing and understanding results from global coupled models has been made since the time of the First CMIP Workshop in Melbourne, Australia, in 1998. For example, the issue of flux adjustment is slowly fading as more and more models obtain stable multi-century surface climates without them. El Niño variability, usually about half the observed amplitude in the previous generation of coupled models, is now more accurately simulated in the present generation of global coupled models, though there are still biases in simulating the patterns of maximum variability. Typical resolutions of atmospheric component models contained in coupled models are now usually around 2.5° latitude –longitude, with the ocean components often having about twice the atmospheric model resolution, with even higher resolution in the equatorial Tropics. Some new-generation coupled models have atmospheric resolutions of around 1.5° latitude–longitude. Modeling groups now routinely run the CMIP control and 1% CO₂ simulations in addition to twentieth- and twenty-first-century climate simulations with a variety of forcings [e.g., volcanoes, solar variability, anthropogenic sulfate aerosols, ozone, and greenhouse gases, with the anthropogenic forcings for future climate as well]. However, persistent systematic errors noted in previous generations of global coupled models are still present in the current generation (e.g., overextensive equatorial Pacific cold tongue, double ITCZ). This points to the next challenge for the global coupled climate modeling community. Planning and commencement of the Intergovernmental Panel on Climate Change Fourth Assessment Report (AR4) has prompted rapid coupled model development, which is leading to an expanded CMIP-like activity to collect and analyze results for the control, 1% CO₂, and twentieth-, twenty-first, and twenty-second-century simulations performed for the AR4. The