Abstract: Objective　The limb imaging spectrometer enables limb observation of oxygen (O2) A-band (759 – 767 nm) airglow. To evaluate the system design parameters of the spectrometer, accurate modeling of the spectral line intensity distributions and spatial variations of this airglow emission is essential. In this study, an O2 A-band airglow radiative transfer model was developed based on four key components: the radiative transfer process, volume emissivity at different altitudes and solar zenith angles, emission spectral lines, and response spectral lines of spectrometer. The performance of the developed model was systematically evaluated in simulating the intensity distribution and variations of airglow spectral lines. Methods　The model was constructed in four sequential stages. Firstly, the radiative transfer process of O2 A-band airglow was established based on the imaging spectrometer system parameters and limb-observation geometry. Secondly, the volume emissivity at different altitudes and solar zenith angles was derived using a photochemical model of O2 A-band airglow. Thirdly, O2 airglow spectral lines at various limb-observation heights were generated by combining the radiative transfer processes with atmospheric parameters, including O2 A-band line intensity, atmospheric oxygen density, and airglow volume emissivity. And these spectral lines are subsequently convolved with the instrument functions to obtain the emission spectral lines at the instrument resolution. Finally, the spectrometer's response spectra were obtained by combining the convolved airglow spectral lines with the parameters of the limb imaging spectrometer. Results and Discussion　The results demonstrate that the established O2 A-band radiative transfer model provides robust theoretical support for optimizing optical system design parameters of spaceborne limb imaging spectrometers. Compared to other ranges, the volume emissivity of dayglow exhibits higher values at altitudes of 40 to 100 km which is influenced by multiple factors. The first one is that solar resonance scattering contributes to the volume emissivity across the 30–120 km altitude range. The second one is that photolysis of oxygen and ozone significantly impacts the volume emissivity of dayglow, and oxygen photolysis occurs primarily at approximately 100 km, while ozone photolysis occurs mainly at around 40 km. Additionally, the Barth mechanism dominates at the altitude range of 80–100 km, although it accounts for a relatively small proportion of total emissivity. Below an altitude of 30 km, the volume emissivity is affected by atmospheric oxygen optical thickness, and the larger optical thickness makes it difficult for solar radiation to reach this altitude near the Earth's surface. Above an altitude of 120 km, despite low optical thickness, the sharp decline in reactant concentration leads to a rapid decrease in volume emissivity. During space-based limb observations, the solar zenith angle (SZA) varies with the detection region, thereby affecting the volume emissivity. When the SZA is small, the solar radiation reaching the atmosphere is strong, triggering strong photochemical reactions and resulting in a relatively high volume emissivity. As the SZA increases, the intensity of solar radiation reaching the atmosphere decreases, leading to a decrease in volume emissivity. Notably, at an altitude of around 90 km, the change of SZA has the least impact on volume emissivity, the reason of which is that the dayglow at this altitude primarily originates from the Barth mechanism rather than solar radiation. Based on the calculated volume emissivity of the O2 A-band, the HITRAN database, and the established radiative transfer model, it is shown that atmospheric O2 has a significant contributes to the absorption effect of airglow, and this effect of atmospheric O2 must therefore be included in the simulation. The simulated airglow intensity varies with altitude, it is strong near the altitude of 80 km but weak below the altitude of 70 km, primarily due to the strong self-absorption by atmospheric O2 below the altitude of 70 km. And above the altitude of 90 km, the intensity of airglow decreases, which is mainly due to the thin atmospheric conditions and low concentrations of substances involved in photochemical reactions. Based on the simulation results of the airglow response of the limb imaging spectrometer, the signal-to-noise ratio and other parameters of the system can be analyzed, providing a basis for evaluating the detection accuracy of the system and the inversion precision of airglow data. Conclusions　This study successfully developed a comprehensive radiative transfer model for space-based limb observations of O2 A-band airglow emissions. Based on this model, the characteristics of the O2 A-band airglow spectral structures can be quantitatively analyzed, including line-intensity profiles across various limb cut-off heights, the effect of atmospheric O2 self-absorption on the airglow spectra, and the variation of airglow intensity with altitude. Furthermore, the high-resolution O2 A-band airglow spectra based on this model simulation can be applied to the parameter evaluation of spaceborne limb imaging spectrometers, such as spectral resolution, optical system parameters, and response characteristics, etc. In future research, this model can also be used as a forward modeling tool for airglow, combined with optimal estimation, least squares and other algorithms to be applied to airglow inversion.

Key words: oxygen A-band, limb detection, photochemical model, volume emissivity, airglow spectral line