A dynamic critical relative humidity based on temperature in cloud parameterization to improve low cloud in an AGCM

Low clouds are essential to the energy budget and the hydrological cycle, but simulation of low clouds in most AGCMs (atmospheric general circulation models) remains a challenge. The critical relative humidity (RHc) has great significance for cloud parameterization. Based on diagnostic results of CloudSat/CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations) satellite data, we propose a fourth-order curve-fitting formula for RHc with respect to temperature (coefficient of determination ( R 2 ) = 0.9659). The method was implemented in CAM6 (Community Atmosphere Model, version 6). Compared with the original scheme, the dynamic RHc significantly reduces the negative bias of low clouds over mid- and low-latitude oceans, increases the low cloud fraction by 20 %. Furthermore, the dynamic RHc has an impact on the vertical distribution of cloud amount, significantly increasing the cloud fraction below 700 hPa and reducing it above 400 hPa. The increase in low clouds is accompanied by an increase in liquid water path, which helps reduce the shortwave cloud forcing bias in the subtropics. Besides, the change in cloud fraction caused by the dynamic RHc has an impact on the simulation of precipitation. Finally, the simulation results at 1° and 2° indicate that the method is insensitive to the choice of model resolution. • Based on CloudSat/CALIPSO satellite data diagnostics, we propose an improved critical relative humidity (RHc) formula. • Dynamic RHc increases low-cloud fraction in low- to mid-latitude oceans and reduces subtropical low-cloud bias. • Dynamic RHc modifies cloud-fraction, reduces shortwave cloud forcing bias and increases shallow convective precipitation.