PI: Paul Dirmeyer Center for Ocean-Land-Atmosphere Studies dirmeyer@cola.iges.org
Co-I: Ben Kirtman Center for Ocean-Land-Atmosphere Studies kirtman@cola.iges.org
Co-I: Charles J. Vörösmarty University of New Hampshire charles.vorosmarty@unh.edu
Collaborator: Humberto R. da RochaUniversidade São Paulo, Brazil humberto@model.iag.usp.br
Collaborator: José A. MarengoCPTEC, Brazil marengo@cptec.inpe.br

October 2000     www.iges.org/lba
 

Update

The model development phase of our project has come to a close. To summarize the
development efforts, carried out predominantly by V. Misra, simulations with the Regional Spectral
Model (RSM) with an atmospheric physics package similar to the COLA GCM were conducted at
80 km horizontal resolution over the austral summer months of Jan-Feb-Mar-Apr-May of 1997, 1998
and 1999 which are normal, warm and cold phases of ENSO.  The higher resolution of the RSM
produces large improvement in the representation of convective variability at high frequency and
intraseasonal scales over GCM or NCEP reanalysis (Figs 1 and 2).  One of the other significant features in the
regional climate simulations is the interannual variability of the low level jet (Fig 3).  This variability in the
jet is associated with interannual variability of precipitation over the Amazon Basin (Fig 5).

Since these simulations we have further developed the RSM by implementing the COLA
version of the simplified SiB scheme, which required some alterations to the turbulent mixing
scheme and model initialization procedures.  Experiments  are now underway to test the sensitivity
of the land surface processes and convection on the regional climate variability over South America.
Since the model development phase has been completed, we may now begin the scientific
investigations associated with our project.  Tantamount is the nesting of the RSM into COLA GCM
to address the issue of predictability of the regional climate of South America, as outlined in the table
of experiments in the original proposal.  With the tools in place to conduct the scientific
experiments, teamwork with our collaborators and research opportunities for one or more graduate
students will emerge.
 
 

Figure 1. The variance of outgoing longwave radiation during Jan-Feb-Mar 1998 at 3-30 days (high frequency) from a) RSM, b) Observations (Liebmann and Smith 1997) and c) NCEP reanalysis.  The units are in W2 m-4. RSM captures the high variability over the Pacific much better than NCEP reanalysis, as well as the band of low variability across the Northern Hemisphere subtropics.  Over South America, and the Atlantic, variability remains low.  Preliminary results from the inclusion of the SSiB land surface scheme suggests there is much improvement in the simulation of variability over land.

Figure 2. The Jan-Feb-Mar 1998 variance of outgoing longwave radiation at 30-60days (low frequency) from a) RSM, b) Observations and c) NCEP reanalysis.  The units are in W2m-4. The improvement of the RSM over the NCEP reanalysis is most vivid at seasonal time scales, and is clear in the other years and during Austral fall as well (not shown).  The improvement in the simulation of convective variability should aid studies of the surface hydrology as well as energy and water budgets over the LBA region.

Figure 3. The cross-section of the seasonally averaged (JFM) meridional wind through 20º S during a) 1997, b) 1998 and c) 1999.  The units are in ms-1.  The profile of the model topography is shown in green. The low-level jet is clearly evident in the regional model, and exhibits interannual variability in its strength and position (predominatly manifest as north-south shifts in the latitude of maximum amplitude).

Figure 4. RSM circulation at 850hPa superposed over the simulated rainfall anomaly (in mm day-1) from a) 1998 and b) 1999 simulation. The reversal between El Niño and La Niña is quite evident both over the Pacific, and over South America.  Given ony three years of integration, the anomalies based on this short period are artificially symmetric.