Oxygen A-band (O₂A, 762 nm) airglow is one of the strongest radiation features in the middle and upper atmosphere. O₂A airglow is evenly distributed across all latitudes and occurs at altitudes of 40‒200 km. Most ground-based measurements are limited due to the strong absorption of O₂ in the airglow. Therefore, it is necessary to study space-based limb detection methods to detect global airglow. Based on this, the global temperature profile can be obtained. This profile supports studies of atmospheric environmental change, atmospheric dynamics, and meteorological monitoring.

Using the temperature and ozone profile monitoring spectrometer (TOPS), we discuss space-based limb detection methods for O₂A airglow. TOPS was launched on January 29, 2025, to measure the global distribution of temperature, atmospheric ozone, and other atmospheric components. In this paper, the O₂A airglow radiation calculation method is analyzed. Using TOPS’s limb observation geometry, along with temperature, pressure, oxygen number density in each layer, and O₂ self-absorption data, we calculate high-spectral-resolution airglow emissions at different limb tangent heights. Then, TOPS’s detection mode and system parameters are analyzed, and spectral and radiometric calibrations are performed to obtain TOPS’s performance parameters. Next, the on-orbit airglow data detected by TOPS are preprocessed, and the optimal estimation algorithm is used to retrieve the temperature profile. We also discuss the selection method of the inversion wavelength. O₂A airglow lines are generally classified as strong, medium, or weak. The inversion wavelength is selected by analyzing each line’s radiation intensity, atmospheric transmittance, and temperature dependence. Finally, for an error analysis, the random noise error, smoothing error, and model error of the inversion results are discussed.

In the O₂A airglow radiation calculation, the pressure, temperature, and O₂ number density profiles are obtained from the MISI model, and the O₂ line intensity is obtained from the HITRAN database. We obtain the high spectral resolution O₂A airglow radiation from 60 to 100 km (Fig. 5). The results show that airglow radiation varies with tangent height, primarily due to changes in excited oxygen number density and O₂ self-absorption. The results also show a strong oxygen self-absorption effect at the tangent height of 60 km, and the effect decreases with increasing height. Due to the low emission intensity, TOPS adopts a limb scanning mode, with an instantaneous field of view of 200 km (horizontal)×2 km (vertical), a vertical scanning range of 10‒100 km, and a vertical resolution of 2 km (Fig. 6 and Fig. 7). TOPS’S spectral range is 498.1‒802.3 nm, spectral resolution is 1.46 nm, spectral calibration accuracy is 0.1 nm, and radiometric calibration accuracy is 3.6%. The signal-to-noise ratio is 140 at a tangent height of 80 km. Analysis results show that TOPS can accurately detect O₂A airglow. The in-orbit O₂A airglow data detected by TOPS are preprocessed, providing knowledge of the dependence of O₂A airglow radiation on tangent heights and solar zenith angles. Based on TOPS’s spectral resolution, we use the optimal estimation algorithm to invert the temperature profile. For the inversion wavelength selection, the weak lines have a strong positive response to temperature increases, while the strong lines are difficult to use for recovering temperatures at lower altitudes due to strong self-absorption. Combined with the spectral resolution of TOPS, the medium line at 765 nm is selected. The inversion results show that the average kernel function peak above 80 km is more than 0.6. As self-absorption increases, the average kernel function peak decreases to less than 0.1 below 80 km, indicating that the temperature inversion error increases. At the same time, it can also be seen that the weight function peak above 90 km and below 75 km decreases, but the average kernel function peak above 90 km basically does not decrease, indicating that the inversion accuracy in the region above 90 km is higher than that in the region below 75 km (Fig. 14). The error results show that errors above 80 km mainly come from measurement errors, whereas errors below 80 km mainly come from the temperature prior due to the significant contribution of the prior values to the inversion results. The error range is -6 to 6 K at 90‒100 km, -10 to 10 K at 80‒90 km, and -12 to 12 K at 64‒80 km.

The analysis of in-orbit airglow shows that TOPS can detect O₂A airglow effectively. The limb detection method, system, and performance parameters of TOPS serve as a reference for designing systems to monitor the middle and upper atmosphere. Based on O₂A airglow spectra, we initially obtain the temperature profile using the optimal estimation algorithm. This provides a foundation for analyzing long-period airglow data, optimizing inversion algorithms, and cross-comparing inversion results.