Radiatively Active Hydrometeor Frequencies From CloudSat-CALIPSO Data for Evaluating Cloud Fraction in Global Climate Models

This study derives radiatively active hydrometeor frequencies (HFs) from CloudSat-CALIPSO satellite data to evaluate modeled cloud fraction in recent historical period simulations by the fifth phase of the Coupled Model Intercomparison Project (CMIP5) models. Most CMIP5 models do not consider precipitating and convective core hydrometeors in radiative transfer calculation except for CESM1-CAM5 with precipitating ice (snow). They do not provide radiatively active cloud fractions for all precipitating hydrometeor types such as snow fraction while the satellite-retrieved HFs include mixed profiles of stratiform clouds and precipitation as well as convective cores. A filtering method is applied to produce estimates of nonprecipitating and nonconvective core HF (NPCHF) from the total HF (THF) excluding rain, which is defined here as the sum of NPCHF, precipitating ice (snow, graupel, and hail), and convective core HF. We determine the optimal reference HF data for model comparison that include estimates of liquid-phase NPCHF from CloudSat radar-only data (2B-CWC) and ice-phase THF from CloudSat-CALIPSO 2C-ICE combined radar/lidar data. The zonal-mean at selected heights and regionally averaged profiles of CMIP5 multi-model-mean cloud fraction show good agreements in comparison with NPCHF estimates with biases within 5%, whereas cloud fraction is significantly underestimated (up to 30%) against the THF estimates. The biases appear below the midtroposphere over the extratropics, in the upper troposphere over the midlatitude lands and a few tropical convective regions, where precipitating ice and convective core hydrometeors contribute greatly. This comparison reveals some underestimates of thick cloud HFs resulting from measurement limitations due probably to attenuation of radar/lidar signals, which cannot be qualified in this study.