Return to Colloquia & Seminar listing
The Madden-Julian Oscillation and convectively coupled waves in a Global Climate Model with a simple multicloud parametrization
PDE and Applied Math SeminarSpeaker: | Boualem Khouider, University of Victoria |
Location: | 1147 MSB |
Start time: | Tue, Apr 27 2010, 4:10PM |
Convection in the tropics is organized into a hierarchy of cloud clusters and super-clusters raging from a few 100 km's to the planetary scale (10,000 to 20,000 km). The superclusters, that occur at the intermediate-synoptic scale of a few 1000 km's, are recognized as the moist-equivalents of the equatorially trapped waves (wave form solutions of the shallow water equations in the vicinity of the equator that travel in both directions along the equator), thus known as convectively coupled waves. The mesoscale clusters, on the 100 km scale, often travel in the direction of the low-level background shear-flow and are known as squall-lines resembling some of the weather systems found here in midlatitudes. The planetary scale disturbances known as the Madden-Julian oscillation or MJO (after its discoverers) appears as an envelope of the mesoscale and/or the synoptic-scale disturbances and it dominates the tropical variability on this scale. It occurs on the intra-seasonal time scale of 40-60 days while the super-clusters and clusters have a life time and periods ranging from a few hours to a few days. The MJO is believed to affect and somewhat regulate all the weather and climate system in the tropics. Things like ENSO (El-Nino), tropical storms, and monsoons are all believed to interact with the MJO. The MJO is also believed to affect and interact with extra-tropical weather patterns such as the north Atlantic oscillation (NAO) that regulates the north Atlantic and western Europe's weather systems and the Pacific north American oscillations that influences the rainfall in the North West. Floods in Washington and British Columbia are believed to be connected to the MJO. Despite all this, current general circulation models (GCMs) used for climate and long-range weather predictions, predict poorly the MJO and the associated convectively coupled waves in general due to the non-adequate treatment of organized convection by the underlying sub-grid cumulus parameterizations.
Recently, K. and Majda introduced and analyzed a simple multicloud model that captures all the convectively coupled waves that appear as cloud super-clusters on satellite measurement of outgoing long-wave radiation. The multicloud models capture the right phase speed and the right dynamical and morphological structure of these waves, as seen in reanalysis data.
The multi-cloud model is coupled to the next-generation GCM of the National Centre for Atmospheric Research (NCAR), in lieu of a more complex recipe to parametrize convection as it is traditionally done. The coupled multicloud-GCM model is run on an all-sea planet with a coarse resolution of roughly 160km with convection restricted to the equatorial belt between -30 and 30 degrees. The results of three different simulations, corresponding to different values of key parameters that control the amounts of moisture and high level clouds. These three cases explore various behaviors of the tropical atmosphere ranging from a regime with a prominent MJO with a realistic phase speed and dynamical and physical structure, i.e, resembling observations, to a regime of interacting convectively coupled waves, imbedded in each other and moving in both directions along the equator.