The work performed so far includes data gathering and processing, the design of a pilot reanalysis system, and running and analysis of a number of pilot runs. Data used and to be used that are additional to those of the GR are precipitation -- ConUS and Mexican rain gauge and CMAP pentad outside; TOVS-1b radiances, profilers, VAD winds, GOES radiances, and land surface temperature, wind and moisture data.
The pilot runs performed were in most cases for January-July 1998. For economy, they were done using the 80 km/38 layer resolution. This is still considerably greater than the T62 resolution of the GR. Pilots performed include "control", assimilation of radiances, and assimilation of precipitation, with the last one redone a number of times using various enhancements of precipitation data and/or alterations of the processing strategy.
Examination of the results of the pilot runs was done primarily for January and July 1998. Geopotential height rms fits to raobs, both in January and July, were overall considerably better than those of the GR. RR rms differences prior to entering the 3D-Var were also examined ("first guess" vs observations). They also tended to be smaller for the RR than for the GR. Monthly precipitation sums over the contiguous United States for both January and July, control, were overall in better agreement with observations than those of the GR. Precipitation assimilation runs resulted in additional considerable improvements.
Additional pilots are in progress. Results are and will be posted for evaluation by the expected user community as they are becoming available (http://www.emc.ncep.noaa.gov/mmb/rreanl). At the time of this writing, production phase is expected to begin in the last quarter of 2001.