CAPT (http://www-pcmdi.llnl.gov/capt), a joint project of the U.S. Department of Energy’s Climate Change Prediction (CCPP) and Atmospheric Radiation Measurement (ARM) Programs, exemplifies a new approach for evaluating parameterizations in a global climate model. After realistically initializing the model from analyses of the observed global atmospheric state, a series of short-range weather forecasts are generated, and subsequently are verified globally against six-hourly analyses and locally against high-frequency field observations such as those recorded at ARM sites. Diagnosis of systematic forecast errors then can provide insights on the deficiencies of the climate model's parameterizations and also may foster improvements in these schemes.

We will report on selected results of applying the CAPT diagnostic protocol to the Community Atmosphere Model (CAM). These include 1) analyses of model errors in forecasting atmospheric moisture and temperature budgets at the ARM Southern Great Plains (SGP) site; 2) comparisons of model forecasts of moist processes with observations at ARM sites; and 3) improvements in the CAM forecasts which result from changes in the model's convective triggering mechanism.

* This work was performed under the auspices of the U.S. Department of Energy by the University of California, Lawrence Livermore National Laboratory under Contract W-7405-Eng-48.

+ The National Center for Atmospheric Research is sponsored by the National Science Foundation.