Coupled Model Intercomparison Project

Coupled Model Intercomparison Project
Name	Coupled Model Intercomparison Project
Abbreviation	CMIP
Established	1995
Discipline	Climate science
Scope	Global climate modeling intercomparison
Participants	Many modelling centres worldwide

Coupled Model Intercomparison Project is an international coordinated effort to compare and evaluate climate models developed by major research institutions such as National Aeronautics and Space Administration, National Oceanic and Atmospheric Administration, Met Office Hadley Centre, Max Planck Institute for Meteorology, and Institute Pierre-Simon Laplace. It provides standardized experimental protocols used by modeling groups including Geophysical Fluid Dynamics Laboratory, Laboratoire de Météorologie Dynamique, Canadian Centre for Climate Modelling and Analysis, Japan Agency for Marine-Earth Science and Technology, and Commonwealth Scientific and Industrial Research Organisation. Outputs from the project inform assessments by organizations such as the Intergovernmental Panel on Climate Change, United Nations Framework Convention on Climate Change, World Meteorological Organization, and national agencies including European Space Agency and Environment and Climate Change Canada.

Overview and Objectives

The project aims to coordinate multi-model experiments among centers like Princeton University, Columbia University, Massachusetts Institute of Technology, University of Oxford, and ETH Zurich to quantify uncertainties relevant to reports by Intergovernmental Panel on Climate Change, Global Climate Observing System, and World Climate Research Programme. Objectives include evaluating model fidelity against observations from Argo program, TOGA, Global Precipitation Climatology Project, NOAA Observing System, and NASA Earth Observing System; diagnosing intermodel spread among systems developed at NCAR, CSIRO, NIWA, Korea Meteorological Administration, and Beijing Climate Center; and producing projections used by policy bodies such as European Commission, United Nations Environment Programme, and International Energy Agency.

History and Phases (CMIP1–CMIP7)

Origins trace to collaborations among institutions including World Climate Research Programme, International Geosphere-Biosphere Programme, Hadley Centre, GFDL, and Max Planck Society in the 1990s, spawning CMIP phases used in successive Intergovernmental Panel on Climate Change Assessment Reports. CMIP1 and CMIP2 concentrated on model intercomparison among centers like UK Met Office, GFDL, and MPI-M; CMIP3 supported the IPCC Fourth Assessment Report; CMIP5 informed the IPCC Fifth Assessment Report with input from Potsdam Institute for Climate Impact Research and Scripps Institution of Oceanography; CMIP6 expanded scenarios and diagnostics with contributions from European Centre for Medium-Range Weather Forecasts, Lawrence Berkeley National Laboratory, and Princeton Plasma Physics Laboratory feeding into the IPCC Sixth Assessment Report. CMIP7 planning involves workshops hosted by WCRP, WMO, IPCC, and regional centers such as Asian Development Bank and African Union to address emergent needs.

Experimental Design and Protocols

Protocols prescribe standardized experiments like historical runs, abrupt greenhouse gas increases, and forcing scenarios tied to frameworks from Representative Concentration Pathways and Shared Socioeconomic Pathways, coordinated with institutions including IPCC, OECD, and United Nations. The design employs forcings derived from datasets produced by Paleoclimate Modelling Intercomparison Project, Aerosol Comparisons, Coupled Carbon Cycle Intercomparison Project, and observational archives maintained by NOAA National Centers for Environmental Information, ECMWF, and Hadley Centre Sea Ice and Sea Surface Temperature dataset. Model intercomparison protocols enable diagnostics used by research groups at Brown University, Yale University, University of California, Berkeley, University of Tokyo, and Australian National University.

Model Components and Coupling Frameworks

Models couple atmosphere, ocean, land surface, sea ice, and biogeochemical components developed by teams at NCAR, MPI-M, GFDL, CNRM, and ACCESS. Coupling frameworks and software infrastructures include platforms from Earth System Modeling Framework, ESMF, Flexible Modeling System, Common Infrastructure for Modelling the Earth (CIME), and tools maintained by University Corporation for Atmospheric Research and European Centre for Medium-Range Weather Forecasts. Component models integrate parameterizations drawn from research at Scripps Institution of Oceanography, Lamont–Doherty Earth Observatory, Jet Propulsion Laboratory, Oak Ridge National Laboratory, and Los Alamos National Laboratory.

Data Management, Metrics, and Evaluation

Data stewardship follows standards from World Meteorological Organization, Group on Earth Observations, DataCite, Pangeo, and community portals such as Earth System Grid Federation supported by Lawrence Berkeley National Laboratory, ESGF nodes at NASA Goddard, NCAR and UK Met Office. Evaluation metrics derive from observational networks maintained by Global Precipitation Measurement, NOAA Climate Prediction Center, US Geological Survey, COPERNICUS, and paleodata from PAGES and Neotoma Database. Model output archiving, provenance, and FAIR principles are coordinated with initiatives at Digital Object Identifier System, ORCID, and academic repositories at Zenodo and Dryad.

Scientific Contributions and Key Findings

The project has quantified transient climate response, equilibrium climate sensitivity bounds debated in literature from Nature, Science, Journal of Climate, and reports by IPCC, influenced studies at Stanford University, Harvard University, Imperial College London, University of Washington, and Princeton University. Key findings include attribution of recent warming consistent with analyses by James Hansen-led teams at GISS, detection and projection of regional changes relevant to Arctic Council and Small Island Developing States, and insights into feedbacks involving clouds, aerosols, and carbon cycle processes studied at Lawrence Livermore National Laboratory, Columbia University, and University of Colorado Boulder.

Challenges, Limitations, and Future Directions

Challenges include structural uncertainty, internal variability, model tuning practices critiqued by groups at University of Exeter and University of Reading, and computational limits addressed by supercomputing centers such as Oak Ridge Leadership Computing Facility, NERSC, PRACE, and national labs including Argonne National Laboratory. Future directions emphasize higher-resolution ensemble experiments, improved aerosol and cloud representations, coupling with socio-economic models at IIASA and International Institute for Applied Systems Analysis, expanded paleoclimate and emergent constraint diagnostics coordinated with PAGES and advances in machine learning from teams at Google DeepMind and Microsoft Research.

Category:Climate modeling