SIMULATIONS BY THE ATMOSPHERIC MODEL INTERCOMPARISON PROJECT OF ATMOSPHERIC CIRCULATION OVER SOUTHERN AFRICA

Authors
Citation
Am. Joubert, SIMULATIONS BY THE ATMOSPHERIC MODEL INTERCOMPARISON PROJECT OF ATMOSPHERIC CIRCULATION OVER SOUTHERN AFRICA, International journal of climatology, 17(11), 1997, pp. 1129-1154
Citations number
66
Categorie Soggetti
Metereology & Atmospheric Sciences
ISSN journal
08998418
Volume
17
Issue
11
Year of publication
1997
Pages
1129 - 1154
Database
ISI
SICI code
0899-8418(1997)17:11<1129:SBTAMI>2.0.ZU;2-A
Abstract
Seven simulations by the Atmospheric Model Intercomparison Project (AM IP) of mean climate and interannual variability over southern Africa d uring the 1979-1988 decade are evaluated. The focus of the study is th e representation in the models of both regional and hemispheric circul ation features which influence the climate of the subcontinent. The AM IP integrations provide an improved simulation of mean circulation aro und southern Africa when compared with earlier atmospheric general cir culation models. Improvements are noted in the representation of the p osition and intensity of the subtropical anticyclones, the circumpolar trough and the strength of the mid-latitude westerlies. The represent ation of tropospheric circulation features throughout the Southern Hem isphere, including zonal asymmetries and planetary wave 1 at 500 hPa, appear to be dependent on the spatial resolution of the models. No sin gle model simulates all features of southern African circulation accur ately, and model performance varies with the seasons. Several of the A MIP models are capable of simulating the observed features of interann ual variability in circulation over southern Africa. Simulated interan nual variability in both summer rainfall and circulation is weaker tha n observed. Simulated circulation adjustments in the subtropics and mi d-latitudes are in good agreement with observed adjustments associated with periods of above and below average summer rainfall over the subc ontinent. It is recognized that considerable uncertainty is associated with the single integrations using observed sea-surface temperature f orcing considered in the present analysis. Ensemble approaches are req uired to provide a more comprehensive assessment of the simulated resp onse to imposed forcing. (C) 1997 by the Royal Meteorological Society.