On the Estimation of Uncertainties of Wave Activity in the Upper Troposphere and Lower Stratosphere and its Application to a Relative Temperature Variance Climatology

A. DE LA TORRE1, P. LLAMEDO1, P. ALEXANDER1, T. SCHMIDT2AND J. WICKERT2

Montreal, Canada

Congreso; IAMAS MOCA 09; 2009

In a recent work, from CHAMP (CHAllenging Minisatellite Payload) GPS radio occultation data a comparison was made between a Gaussian filter applied to the "complete" temperature profile and to the "separate" tropospheric and stratospheric height intervals, for atmospheric wave analyses. The specific mean potential wave energy as well as the mean relative temperature variance, represent an accepted measure of atmospheric wave activity. In that work, it was found that the separate filtering method considerably reduces the wave activity artificial enhancement near the tropopause (see discussion in Schmidt et al in this session).         We propose now a simple approach to estimate the committed error in the calculation of the mean wave relative temperature variance, when a band-pass filter is applied to the measured temperature profiles for the complete altitude range, independently of the experimental origin of the data. This approach was developed through a statistical simulation from synthetic wave perturbations and temperature profiles from reanalyses. A systematic bias between the true and calculated mean relative variance content was clearly detected and its variation with latitude and season is underlined. These results are finally applied to adjust a long-term relative variance climatology of the lower and middle atmosphere, derived from CHAMP GPS radio occultation data and filtered over each complete temperature profile, between 2001 and 2009.