Intercomparison and Evaluation of Cumulus Parameterizations under Summertime Midlatitude Continental Conditions

Conference · Thu May 17 04:00:00 EDT 2001

OSTI ID:15013152

Xie, S; Cederwall, R T; Yio, J; Xu, K-M

Parameterization of cumulus convection in general circulation model (GCM) has been recognized as one of the most important and complex issues in the model physical parameterizations. In earlier studies, most cumulus parameterizations were developed and evaluated using data observed over tropical oceans, such as the GATE (the Global Atmospheric Research Program's Atlantic Tropical Experiment) data. This is partly due to inadequate field measurements in the midlatitudes. In this study, we compare and evaluate a total of eight types of the state-of-the-art cumulus parameterizations used in fifteen Single-Column Models (SCM) under the summertime midlatitude continental conditions using the Atmospheric Radiation Measurement (ARM) Southern Great Plains (SGP) summer 1997 Intensive Operational Period (IOP) data, which covers several continental convection events. The purpose is to systematically compare and evaluate the performance of these cumulus parameterizations under summertime midlatitude continental conditions. Through the study we hope to identify strengths and weaknesses of these cumulus parameterizations that will lead to further improvements. Here, we briefly present our most interesting results. A full description of this study can be seen in Xie et al. (2001). The authors conclude that: (1) The SCM simulation errors are closely related to problems with model cumulus parameterizations. The schemes with triggering based on CAPE generally produce more active cumulus convection than the schemes with triggering based on local parcel buoyancy over land surface at midlatitudes since CAPE is usually large and is mainly determined by the strong solar diurnal heating there. The use of positive CAPE to trigger model convection can lead to an overestimation of convection during the daytime. This results in warmer/drier atmospheres in the former and cooler/more moist atmospheres in the latter. (2) A non-penetrative convection scheme can underestimate the depth of conditional instability for deep midlatitude continental convection, and therefore results in a cooler atmosphere in the upper troposphere. (3) The updraft mass fluxes in the SCMs agree quantitatively well with those in the CRMs. In contrast, most SCMs produce very weak downdraft mass fluxes compared with those diagnosed from the CRMs. These discrepancies are primary attributable to the neglect of mesoscale circulations in the SCMs, uncertainties in the diagnosed mass fluxes from the CRMs, and deficiencies in the SCM parameterizations of mass fluxes.

• Research Organization: Lawrence Livermore National Laboratory (LLNL), Livermore, CA
• Sponsoring Organization: USDOE
• DOE Contract Number: W-7405-ENG-48
• OSTI ID: 15013152
• Report Number(s): UCRL-JC-142040-REV-1
• Country of Publication: United States
• Language: English

Related Subjects

54 ENVIRONMENTAL SCIENCES
CONVECTION
EVALUATION
GENERAL CIRCULATION MODELS
HEAT EXCHANGERS
HEATING
INSTABILITY
PERFORMANCE
RADIATIONS
SIMULATION
TROPOSPHERE