Evaluation of Cloud Fraction Simulated by Seven SCMs against the ARM Observations at the SGP Site*

• DOI: 10.1175/jcli-d-13-00555.1
• Publication Date: 2014-06-27T19:06:16Z
• Authors: Hua Song; Wuyin Lin; Yanluan Lin; Audrey B. Wolf; Leo J. Donner; Anthony D. Del Genio; Roel Neggers; Satoshi Endo; Yangang Liu
• Keywords: Cloud fraction; Fraction (chemistry); Liquid water content; Atmospheric models; Cloud albedo

Abstract This study evaluates the performances of seven single-column models (SCMs) by comparing simulated cloud fraction with observations at Atmospheric Radiation Measurement Program (ARM) Southern Great Plains (SGP) site from January 1999 to December 2001. Compared 3-yr mean observational fraction, ECMWF SCM underestimates all levels and GISS below 200 hPa. The two GFDL SCMs underestimate lower-to-middle level but overestimate upper-level fraction. three Community Atmosphere Model (CAM) produce lower-level similar as a result compensating overproduction convective underproduction stratiform Besides, CAM3 CAM5 both midlevel whereas CAM4 underestimates. frequency partitioning analyses show large discrepancy among SCMs: Contributions nonstratiform processes production are mainly in cloudy events over cover range 10%–80% prognostic schemes 15%–50% diagnostic schemes. Further analysis reveals different relationships between relative humidity (RH) observations. underestimation most is due larger threshold RH used models. overestimation CAM primarily plus more occurrences around 40%–80%, respectively.