Climate Variability in Southeastern South America Related to ENSO. A Numerical Study

sINTRODUCTION. Several studies had shown relations between interannual anomalies in the surface climate of south eastern South America, "SSA", and anomalies of the equatorial Pacific sea surface temperature, "SST". It is during austral spring that such relations are stronger, connecting positive SST anomalies in the equatorial central Pacific with greater probabilities of rainfall above normal in SSA. Several mechanicm have been proposed as explanations for such fenomena: modifications in the jet streams, development of Rossby wave patterns, anomalies in the convection. Most of this studies focus the extreme seasons (summer and winter), therefore they are not completely suitable for explaining the phenomena that occurs during the intermediate seasons (in particular in November). In this study, a version of the atmospheric general circulation model of the University of California at Los Angeles (UCLA AGCM) is used for simulating the effect on the atmospheric circulation of SST anomalies in the equatorial central Pacific during November. The UCLA AGCM is a finite diference model for the primitive equations of the atmosphere, with parameterizations of physical processes relevant to the global circulation. The version of the UCLA AGCM used in this work includes its most recent modifications. The spatial resolution used in our experiments is 4 degrees in latitude and 5 degrees in longitude. In the vertical direction, there are 16 layers, extending from the globe surface to the level of 1 mb. One of those layers is the PBL. DESCRIPTION OF THE EXPERIMENTS. Four control simulations were performed. Each one is extended from October 1st. to December 31th. All of them had the same boundary conditions, they only differ in the initial conditions. The SSt included in the boundary conditions set is the climatological one, for each day. Also four anomalous simulations were performed They are respectively analog to each one of the control simulations in every sense, except that a time independent SST anomaly was added to the climatological SST field prescibed before. This SST anomaly is the composite of the observed anomalies at November 1982, 1986 and 1991 (this years correspond to warm episodes). RESULTS AND DISCUSSION The most important results that appears in the Anomay - Control differences are: - There is an intense increase of precipitation over the precribed SST anomaly. The anomaly resulted to be due to convective precipitation. - There is also a precipitation anomaly close to SSA. This anomaly resulted to be related to the large scale precipitation, associated to the baroclinic processes. - The Anomaly - Control difference of circulation fields at 200 mb. shows a wave-like pattern, similar to those obtained from observed composites at warm episodes.

The statistical significances of the Anomaly-Control differences was computed using a t-Student test.

For learning about the physical nature of the Anomaly-Control differences found at 200 mb, we used two usual dynamic tools: stationary wave number as function of latitude and vorticity tendency budgets. The relevant terms of the vorticity tendency budgets are those associated with the typical Rossby wave mechanisms. They changes found in the high troposphere circulations would be consistent with an enhancement of the baroclinic processe