Winston Chao (PI)
NASA Goddard Space Flight Center
winston.c.chao@nasa.gov

Understanding and Simulation of Tropical Large-Scale Rainfall and Circulation Using the GMAO GCM and GMAO Reanalysis

This proposal aims at the goal stated in part (a) of subsection B.4 of the MAP research announcement: ''Reduce uncertainties in model results and predictions by understanding the processes that …''.    The current atmospheric GCMs, including the one used at GMAO, have considerable difficulty in simulating the climatological precipitation in the tropics, such as the intensity and the location of the ITCZ and the principal components of the precipitation variations such as the annual cycle of the ITCZ.  These difficulties are an important contributor to poor climate forecast,  including El Nino forecast, when the atmospheric GCMs are coupled to ocean GCMs.  This is because that the intensity and the location of the ITCZ determine the surface wind distribution, a crucial factor in the coupling between ocean and atmosphere.  Model improvements require the understanding of the basic physical mechanisms.  Therefore, there is an urgent need to understand the tropical climatological precipitation, including its annual cycle.  In the attempt to obtain such understand the current GCMs, though imperfect, can provide a very useful tool.  The PI has done some work (Chao 2000, Chao and Chen 2001, Chao and Chen 2004) trying to understand the latitudinal location of the ITCZ in an aqua-planet setting using an early version (GEOS 2) of the Goddard GCM.  The results show the importance of earth’s rotation, interaction between convection and radiation and interaction between convection and surface fluxes.  This effort has been just the initial steps toward the ultimate goals of understanding and simulation of tropical climatological precipitation and it should be continued.  Our immediate next steps will include the understanding of the intensity of the ITCZ under the aqua-planet setting.  Our previous work hinted that both the interaction between convection and radiation and interaction between convection and surface fluxes are importance factors governing the intensity of the ITCZ.   Our working hypothesis has been laid out in Chao and Chen (2004).  In short, the ITCZ’s latitudinal position in this setting is determined by the balance of two kinds of forcing.  The first kind pulls the ITCZ toward the equator, due to inertial stability.  The second kind pulls the ITCZ poleward due to the modifying effect of earth’s rotation on the convective circulation, which makes the surface converging winds stronger by developing a tangential wind component and thereby affects surface evaporation and thus vertical stability.  If the first kind dominates, the ITCZ is at the equator.  If the second kind dominates, the ITCZ is not at the poles but is pulled away from the equator by only about 14 degrees due to the nonlinear nature of the latitudinal dependency of the two forcings.  Continuing in this direction, we will introduce land and realistic SST distribution one at a time into the model and study their impacts on ITCZ location and intensity.  After we have sufficient grasp of the atmospheric aspects of the problem, we will then use a coupled ocean-atmosphere model in study the role of the coupled interaction in the climatological tropical precipitation.  The second component of the study is to apply the understand that we have gained to the improvement of the simulation of tropical climatological precipitation and its variations, such as annual cycle, El Nino, and MJO simulations.  The basic guideline here is to first find out the sensitivity of the assorted phenomena related to tropical precipitation to various interactions in the atmosphere and between ocean and atmosphere.  The most sensitive interaction, within the realm of reasonable variations, points to the aspect of the model that requires modification or tuning.  In all phases of our proposed work the GCM will be our main tool.  We intend to use the latest version of the GMAO GCM.   GMAO reanalysis data will be used for comparison with model results.  The requested budget will be used to cover the salary of one research associate and part of PI’s salary (under full cost accounting).

+ Back to Participant Listing