The Use of Available Potential Energy to Evaluate the Impact of Satellite Data on Numerical Model Analysis during FGGE

Abstract

Available potential energy (APE) calculations are used to evaluate the influence of the FGGE satellite observing system during the first Special Observing Period (SOP-1). Two datasets consisting of the Goddard Laboratory for Atmospheres analyses are used: the complete FGGE IIIb set which has incorporated satellite soundings and a NOSAT set which incorporates only conventional data. The major portion of the study uses the exact (isentropic) formulation of APE; however, some introductory comparisons are made with the approximate (isobaric) form. Time series of the daily total APE values show the NOSAT set yielding slightly larger values in both APE formulations, with the approximate form zonal component exhibiting the largest difference, An examination of the average eddy APE (exact form) contributions from 4° latitude rings reveals that in the Southern Hemisphere middle latitudes the NOSAT values are larger than the FFGE values; however in the Northern Hemisphere the differences are negligible. Analyses of the vertically integrated gridpoint contributions to the eddy APE reveal that the exact form of the APE clearly defines cyclone scale features. The gridpoint distributions show only minor differences between the Northern Hemisphere FGGE and NOSAT analyses, but substantial differences exist in the Southern Hemisphere where the satellite soundings apparently add detail to the FGGE set by locating the trough and ridge systems associated with cyclones more accurately. However, the weaker thermal gradients obtained from satellite soundings tend to yield smaller eddy APE values. Temporal standard deviations of the gridpoint contributions to the eddy APE are closely related to extratropical cyclone tracks. Again there is essentially no difference in the two datasets in the Northern Hemisphere, but differences exist in the Southern Hemisphere. Finally an analysis of the FGGE-NOSAT temperature differences in vertical cross sections along 54°N and 54°S show larger differences in the Southern Hemisphere. Some of the differences clearly propagate eastward during a three-day period. In some areas large differences of one sign overlay large differences of the opposite sign implying significant differences in static stability between the two sets.