A Four-Season Impact Study of Rawinsonde, GOES, and POES Data in the Eta Data Assimilation System. Part I: The Total Contribution

Abstract

The impact of in situ rawinsonde (raob) data, remotely sensed Geostationary Operational Environmental Satellite (GOES), and Polar Operational Environmental Satellite (POES) data routinely used in NCEP?s Eta Data Assimilation/Forecast System (EDAS) is studied for extended-length time periods during four seasons. The work described in this paper is relevant for users of the Eta Model trying to compare and contrast the overall forecast impact of traditional, mostly land-based rawinsonde data with remotely sensed data that are available domainwide. The case studies chosen consist of 15-day periods during fall 2001, winter 2001/02, spring 2002, and summer 2002. During these periods, a 32-km/60-layer November 2001 version of the EDAS is run four times at both 0000 and 1200 UTC. The four runs include a control run, which utilizes all data types routinely used in the EDAS, and three experimental runs in which either all rawinsonde, GOES, or POES data are denied. Differences between the experimental and control runs are then accumulated over the 15-day periods and analyzed to demonstrate the 24- and 48-h forecast impact of these data types in the EDAS. Conventional meteorological terms evaluated include mean sea level pressure as well as temperature, both components of the wind, and relative humidity. Comparisons are made on seven pressure levels extending from near the earth?s surface to the lower stratosphere. The diagnostics are computed over both the entire horizontal model domain, and within a subsection covering the continental United States and adjacent coastal waters (extended CONUS). The 24-h domainwide results show that a positive forecast impact is achieved from all three data sources during all four seasons. Cumulatively, the rawinsonde data have the largest positive impact over both the entire model domain and extended CONUS. However, GOES data have the largest contribution for several fields, especially moisture during summer and fall 2001. In general, GOES data also provide larger forecast impacts than POES data, especially for the wind components. All three data types demonstrate comparable forecast impact in terms of relative humidity. Finally, raob and POES data display a ?spike? in positive forecast impact in the lower stratosphere during three of the four seasons. Two additional findings from this study are also important. The first is that the forecast impact of all data types drops by at least a factor of 2 during all seasons between 24 and 48 h. The second is that GOES data show a preference for providing nearly equal improvement to the 0000 and 1200 UTC forecast cycles, while rawinsonde and especially POES data provide consistently larger forecast impacts at 1200 than at 0000 UTC.