A CONVECTIVE WAKE PARAMETERIZATION SCHEME FOR USE IN GENERAL-CIRCULATION MODELS

Citation
Ly. Qian et al., A CONVECTIVE WAKE PARAMETERIZATION SCHEME FOR USE IN GENERAL-CIRCULATION MODELS, Monthly weather review, 126(2), 1998, pp. 456-469
Citations number
41
Categorie Soggetti
Metereology & Atmospheric Sciences
Journal title
ISSN journal
00270644
Volume
126
Issue
2
Year of publication
1998
Pages
456 - 469
Database
ISI
SICI code
0027-0644(1998)126:2<456:ACWPSF>2.0.ZU;2-3
Abstract
In the atmosphere, the cold, dry, precipitation-driven downdrafts from deep convection spread laterally after striking the surface, inhibiti ng new convection in the disturbed wake (cold pool) area. In contrast, the leading edge of this cold air lifts unstable environmental air he lping to trigger new convection. A GCM cannot resolve the disturbed an d undisturbed regions explicitly, so some parameterizations of these c ritical mesoscale phenomena are needed. The simplest approaches are to either instantly mix downdraft air with the environment, or instantly recover the downdraft air. The instant-mixing approach tends to lead to unrealistic pulsing of convection in environments that would otherw ise be able to support long-lived mesoscale convective systems while t he instant recovery approach usually overestimates surface energy flux es. By replacing these simplistic approaches with a physically based c onvective wake-gust front model, these problems are substantially reme died. The model produced realistic parameterized wakes that closely re semble those observed in the Global Atmospheric Research Program's Atl antic Tropical Experiment and Tropical Ocean and Global Atmosphere Cou pled Ocean-Atmosphere Response Experiment when given reasonable inputs based on observations taken during these experiments. For realistic d owndraft characteristics, wake recovery time is on the order of hours, which is significantly different from the instant recovery or instant mixing assumed in previous parameterizations. A preliminary test in m idlatitude continental conditions also produced reasonable wake charac teristics. Sensitivity tests show the model sensitivities to variation s in downdraft mass Aux, downdraft thermodynamic characteristics, and surface wind/downdraft traveling velocity. Prognostic studies using a simple coupled cloud model successfully simulated the convective termi nation due to stabilization of the boundary layer by precipitation-dri ven downdrafts, the initiation of convection after the boundary layer recovery by surface fluxes, and the phenomenon of surface flux enhance ment during the convective phase.