The PIs will produce a four-dimensional internally consistent gridded data set for the global atmosphere for the period May 1982 to November 1983. They will do so by assembling the most complete set of atmospheric and oceanic observations made during that period and by using a state-of-the-art multivariate optimal interpolation scheme for atmospheric data assimilation. The primary motivation for this work is to demonstrate the feasibility of reanalyzing past atmospheric observations using a state-of-the-art global model and a global data assimilation system. Such a feasibility study is needed before making plans for reanalysis of long term (more than 10 years) historical data sets for climate research. The data assimilation system to be used will include the COLA general circulation model, the forecast of which will provide the first guess, and the most accurate optimum interpolation algorithm available. The observational data set will be a combination of level II-data archived in real time during the period and delayed mode data, including the comprehensive Ocean-Atmosphere Data Set (COADS), which have been assembled at NCAR. The critical evaluations of the analyses will include general circulation statistics; global and local energy spectra; global energetic diagnoses including generation and dissipation of energy; a detailed examination of the global hydrologic cycle including regional budgets for the tropical Pacific, North America, the Amazon basin, and sub-Saharan Africa; and computations of surface wind stress and total heat flux to force a global ocean model. The PIs will archive the various measures of analysis error obtained during the assimilation cycle to study the quality of the analyses and to provide confidence estimates and error bars. This study will be extremely useful to support the scientific objectives of national and international research programs such as the Tropical Oceans Global Atmospheric (TOGA), World Ocean Circulation Experiment (WOCE), Global Energy and Water Cycle Experiment (GEWEX) and International Geosphere Biosphere Program (IGBP). The PIs have chosen the 1982-83 period for this study because it witnessed the most outstanding example of ocean-atmosphere interaction as manifested by the E1 Nino- Southern Oscillation phenomenon and its related world-wide effects.
