In view of the increasing reliance on general circulation models for simulation of potential global climate change, there is a compelling need to improve the performance of these models in predicting the present climate. To this end, we are developing the CCPP-ARM Parameterization Testbed (CAPT) as a joint effort of the U.S. Department of Energy's Climate Change Prediction Program (CCPP) and Atmospheric Radiation Measurement (ARM) Program.

CAPT (see website at http://www-pcmdi.llnl.gov/capt) exemplifies an innovative approach for testing and improving climate model parameterizations that borrows from techniques originally developed in numerical weather prediction (NWP). First, an NWP reanalysis of the weather on a particular day is used to initialize the state variables of a climate model. Next, errors in short-range forecasts by the climate model, many of which are attributable to parameterization deficiencies, are identified from concurrent reanalysis data and observations of physical forcings such as provided by ARM. The efficacy of changing selected parameterizations in attempting to improve the climate model's performance then can be provisionally evaluated in the same way.

In this presentation, we will elaborate the CAPT diagnostic protocol and show examples of its application to the Community Atmosphere Model (CAM).

This work was performed under the the auspices of the U.S. Department of Energy by the University of California, Lawrence Livermore National Laboratory under Contract W-7405-Eng-48, as well as by staff at the National Center for Atmospheric Research, which is sponsored by the National Science Foundation.