ARM data-oriented metrics and diagnostics facilitates use of field data in climate model evaluation

 

Submitter:

Zhang, Chengzhu Jill — Lawrence Livermore National Laboratory
Xie, Shaocheng — Lawrence Livermore National Laboratory

Area of research:

General Circulation and Single Column Models/Parameterizations

Journal Reference:

Zhang C, S Xie, C Tao, S Tang, T Emmenegger, J Neelin, K Schiro, W Lin, and Z Shaheen. 2020. "The ARM Data-oriented Metrics and Diagnostics Package for Climate Models - A New Tool for Evaluating Climate Models with Field Data." Bulletin of the American Meteorological Society, 101(10), 10.1175/BAMS-D-19-0282.1.

Science

Climate model developers often use a set of standard metrics and diagnostics as a way of routinely assessing general model performance or judging the performance of new physical parameterizations. ARM data can help to address a range of issues, including diagnosing summertime warm bias, the metrics of convective onset, precipitation distribution, and the diurnal cycle of both cloud fraction and precipitation.

Impact

Climate model developers routinely use satellite remote sensing to calibrate and tune models. But employing detailed, high-frequency ground measurements with a tool like ARM-DIAGS provides a complementary test in evaluating models. To date, ARM data have not been extensively used in model development workflows. With growing interest in improving parameterization using process-oriented metrics and diagnostics, ARM observations should play a more important role, especially in the way cloud and precipitation processes are represented in climate models. Inclusion of ARM-DIAGS in evaluation workflow allows climate modelers to compare their models with ARM data and supplemented CMIP data sets.

Summary

ARM’s high-frequency and long-term data sets are unique and invaluable in developing and improving climate models. A metrics and diagnostics package called ARM-DIAGS has been developed to further facilitate the use of ground-based ARM measurements in evaluating climate models. It includes both ARM data sets, CMIP data, and a Python-based analysis toolkit for computation and visualization. The package can serve as an easy entry point for climate modelers to compare their models with ARM data and supplemented CMIP data sets.