Atmospheric Precipitation-Cloud-Radiative Biases and their Impacts on Fully Coupled GCMs

Jui-Lin (Frank) Li
July 21st 10:30 am in room 2155

Significant systematic clouds, radiation and moist biases are found in the CMIP3/CMIP5 models ensemble average in sub- and tropical Pacific regions in conjunctions with biases of upward motion, eastward low-level wind with moist convergence north of SPCZ and south of ITCZ – referred to as the V-Shaped regions.

We characterize the impacts of precipitating snow and radiation effects, which are not included in most of the CMIP3/CMIP5 models, by using the National Center for Atmospheric Research (NCAR)-coupled GCM that the model includes diagnostic snow for radiation calculation.  A number of differences associated with snow-radiation effect off are found consistent with the biases in CMIP3/CMIP5 simulations.  With snow-radiation off, there excessive longwave radiative cooling at the top of atmosphere while shortwave radiative warming at the surface resulting in net radiative cooling near the cloud top, triggering compensating upward motion in the heavily precipitating regions of the tropics such as ITCZ, SPCZ and warm pool. This leads to local subsidence at the north edge of SPCZ and south edge of ITCZ, generating low-level eastward/southeastward wind (i.e., weakening mean northwestward trade wind) and moist convergence from the warm pool, SPCZ and ITCZ and resulting increase total precipitable water and column of moisture.  This further leads to the model’s biases of local and remote areas of sea surface temperatures (SSTs), sea surface height, and surface wind stress and large-scale circulations.