Retrieval of thermosphere temperature based on O(¹S) and O(¹D) airglow observations from MIGHTI/ICON

Thermospheric Doppler temperature profile is of considerable significance in studying the physical processes of the upper atmosphere and the space environment. In this study, a joint retrieval method based on atomic oxygen O(¹S) and O(¹D) airglow emission signals is proposed to construct the global thermospheric temperature profile. First, the temperature retrieval theoretical model is established based on the radiation mechanism of O(¹S) and O(¹D) airglow and Doppler broadening theory of spectral lines, combined with the observation principle of a Doppler heterodyne interferometer. Subsequently, interference fringe information in the target altitude layer is extracted using the O(¹S) and O(¹D) airglow data observed by the Michelson Interferometer for global high-resolution thermospheric imaging (MIGHTI), combined with the "onion peeling" algorithm and Hilbert-Huang transform. The temperature profiles are retrieved by calculating the Doppler broadening of the airglow. Finally, the results are validated through comparison with the Mass Spectrometer Incoherent Scatter radar empirical model 2.1, atmospheric chemistry experiment-Fourier transform spectrometer, and the official MIGHTI O₂ A-band temperature product. Results show that the O(¹S) and O(¹D) airglow emissions are applicable to temperature retrievals within the 90-170 km and 160-250 km altitude ranges, respectively. Overall temperature deviations are controlled within 50 K, with optimal uncertainties of ±3 K and ±5 K, effectively covering the key region of the thermosphere from 90 km to 250 km. The error analysis results show that the temperature retrieval error of MIGHTI based on atomic oxygen airglow can be effectively controlled within 20 K, with the minimum temperature error being approximately 5 K. This method demonstrates high accuracy and reliability, providing an effective approach for thermospheric temperature retrieval.