West African Monsoon 

Modeling and Evaluation

  Forcing and   Boundary Data
  Output Data
  Data/Model Output Access
  Models GCM RCM

Experimental design:

1)  Initial conditions: NCEP-DOE Reanalysis II.  The model runs start from April 1, 2, 3, and 4 through October 31 for years 2000, 2003, 2004, and 2005.  Each group downloads the initial data sets from the Reanalysis II web site and interpolates them to the models’ vertical coordinates and horizontal resolutions.


Due to the importance of initial land conditions, it recommends that using something generated by another run of the same model for April 1 to reduce the spin up/down periods.  Please note that the simulation results for April will be disregarded (i.e., only May, June, July, August, September, October data are needed).


2) Surface boundary conditions: The SST and sea ice data are HadISST1 (Rayner et al. 2003). They are monthly data with 1 degree resolution. The monthly data are on the website. Please interpolate the monthly data to the monthly data in your model grid. A program for the Karl Taylor procedure is provided on the website, which adjusts your interpolated SST data to ensure that the monthly average of the values from your interpolation is the same as the original monthly mean. Then you could interpolate to daily data if it is required by your model. A readme file is on the website for your application. 


3)  The regional climate model’s (RCM) lateral boundary conditions (LBC) will be obtained from Reanalysis II and NCEP and UK Met GCM simulations.  The data for UK Met LBC are in NetCDF format.  RCM will conduct one run for each year’s LBC. April 1 of each year (i.e., April 1 2004) should be selected for RCM experiment. In the UK Met LBC data set, the LBC data of this ensemble member is label ‘afoif’.


The RCM domain should cover 35W-35E, 20S-35N with about 0.5 degree horizontal resolution.   Drs. Len Druyan and Matthew Fulakeza of NASA/GISS provide a topographic map for MM5.  Other RCMs could also use this map or use the topographic map in their own models. 


4) In addition to the control runs, some sensitivity runs will be conducted.

Based on the first stages’ control runs, we will suggest some adjustments to test sensitivity of the WAM to some parameterizations, which could provide insight into the mechanisms that control the West African regional climate as well as uncertainty in mode simulations. 


5) The requirement for model outputs are in a separate file.


6) We expect the model results be submitted before June 15, 2007.