Why has GFDL run climate experiments using both the CM2.0 and CM2.1 models?

When originally developed, the expectation was that CM2.0 would be the GFDL workhorse deccen model for a couple years. However, a few months later, when development work at GFDL demonstrated that significant improvements in the simulation could be achieved by changing the atmospheric dynamical core, and computer resources could be saved via a time stepping change in the ocean, CM2.1 was created. Because CM2.0 and CM2.1 share the same model grids and much of the same code and configuration details, it was decided to name the newer version CM2.1, rather than considering the newer model the start of a distinct CM3 line of models.

Data from experiments conducted using both models will be made available. The CM2.0 experiment results will be completed, and their data made available, sooner than the CM2.1 experiment results.

Questions related to the GFDL CM2.x models may be directed to…