High-resolution temperature profiles retrieved from bichromatic stellar scintillation measurements by GOMOS/Envisat

In this paper, we describe the inversion algorithm for retrievals of high vertical resolution temperature profiles (HRTPs) using bichromatic stellar scintillation measurements in the occultation geometry. This retrieval algorithm has been improved with respect to nominal ESA processing and applied to the measurements by Global Ozone Monitoring by Occultation of Stars (GOMOS) operated on board Envisat in 2002-2012. The retrieval method exploits the chromatic refraction in the Earth's atmosphere. The bichromatic scintillations allow the determination of the refractive angle, which is proportional to the time delay between the photometer signals. The paper discusses the basic principle and detailed inversion algorithm for reconstruction of high-resolution density, pressure and temperature profiles in the stratosphere from scintillation measurements. The HRTPs are retrieved with a very good vertical resolution of similar to 200m and high precision (random uncertainty) of similar to 1-3K for altitudes of 15-32 km and with a global coverage. The best accuracy is achieved for in-orbital-plane occultations, and the precision weakly depends on star brightness. The whole GOMOS dataset has been processed with the improved HRTP inversion algorithm using the FMI's scientific processor; and the dataset (HRTP FSP v1) is in open access. The validation of small-scale fluctuations in the retrieved HRTPs is performed via comparison of vertical wavenumber spectra of temperature fluctuations in HRTPs and in collocated radiosonde data. We found that the spectral features of temperature fluctuations are very similar in HRTPs and collocated radiosonde temperature profiles. HRTPs can be assimilated into atmospheric models, used in studies of stratospheric clouds and used for the analysis of internal gravity waves' activity. As an example of geophysical applications, gravity wave potential energy has been estimated using the HRTP dataset. The obtained spatiotemporal distributions of gravity wave energy are in good agreement with the previous analyses using other measurements.