Application and Evaluation of an Explicit Prognostic Cloud‐Cover Scheme in GRAPES Global Forecast System

Abstract

Abstract An explicit prognostic cloud‐cover scheme (PROGCS) is implemented into the Global/Regional Assimilation and Prediction System (GRAPES) for global middle‐range numerical weather predication system (GRAPES_GFS) to improve the model performance in simulating cloud cover and radiation. Unlike the previous diagnostic cloud‐cover scheme (DIAGCS), PROGCS considers the formation and dissipation of cloud cover by physically connecting it to the cumulus convection and large‐scale stratiform condensation processes. Our simulation results show that clouds in mid‐high latitudes arise mainly from large‐scale stratiform condensation processes, while cumulus convection and large‐scale condensation processes jointly determine cloud cover in low latitudes. Compared with DIAGCS, PROGCS captures more consistent vertical distributions of cloud cover with the observations from Atmospheric Radiation Measurements (ARM) program at the Southern Great Plains (SGP) site and simulates more realistic diurnal cycle of marine stratocumulus with the ERA‐Interim reanalysis data. The low, high, and total cloud covers that are determined via PROGCS appear to be more realistic than those simulated via DIAGCS when both are compared with satellite retrievals though the former maintains slight negative biases. In addition, the simulations of outgoing longwave radiation (OLR) at the top of the atmosphere (TOA) from PROGCS runs have been considerably improved as well, resulting in less biases in radiative heating rates at heights below 850 hPa and above 400 hPa of GRAPES_GFS. Our results indicate that a prognostic method of cloud‐cover calculation has significant advantage over the conventional diagnostic one, and it should be adopted in both weather and climate simulation and forecast.