Evaluation of RSM-Simulated Precipitation During CEOP

• Published in: Journal of the Meteorological Society of Japan Vol. 85A; pp. 145 - 166
• Main Authors: MEINKE, Insa, ROADS, J., KANAMITSU, M.
• Format: Journal Article
• Language: English
• Published: Meteorological Society of Japan 2007
• Subjects: Southeast Asia; AS, Atlantic, Intertropical Convergence Zone
• ISSN: 0026-1165; 2186-9057
• DOI: 10.2151/jmsj.85A.145

Precipitation simulated using the Regional Spectral Model (RSM) during the Coordinated Enhanced Observing Period (CEOP; July 1, 2001 to December 31, 2004) is evaluated by transferring the RSM to seven different regions of the globe and comparing the simulations with observations. These regions cover the eight Continental-Scale Experiments (CSEs) of the Global Energy and Water-cycle EXperiment (GEWEX) and encompass a broad variety of physical and dynamical meteorological processes. Gridded observations of the Global Precipitation Climatology Project (GPCP) and the Global Precipitation Climatology Center (GPCC), as well as CEOP reference site precipitation observations are compared with the RSM simulated precipitation for the first half of the CEOP Enhanced Observation Period (EOP) III (October 2002 to March 2003). After estimating the uncertainty ranges of both the model and the observations, model deficiencies were obtained for almost all model domains in terms of the amount of simulated precipitation. Although the RSM is able to accurately simulate the seasonal evolution and spatial distribution of precipitation, the RSM has an almost uniform positive bias (i.e., RSM values are greater than observed values) over almost all the domains. Most of the positive bias is associated with convection in the Intertropical Convergence Zone (ITCZ) or monsoonal convection in Southeast Asia. Predicted stratiform precipitation is also excessive over areas ofelevated topography. As the control simulations used a Relaxed Arakawa-Schubert scheme (RAS), sensitivity tests with three additional convection schemes were then carried out to assess whether the simulations could be improved. The additional convection schemes were: 1) the Simplified Arakawa-Schubert scheme (SAS); 2) the Kain-Fritsch scheme (KF); and 3) the National Centers for Atmospheric Research (NCAR) Community Climate Model (CCM) scheme. The precipitation simulation was significantly improved for almost all domains when using either the KF scheme or the SAS scheme. The best simulations of ITCZ convective precipitation and Southeast Asian monsoon convective precipitation were achieved using the SAS convection scheme.