瑞利散射是生活中重要而又常见的自然现象之一，瑞利光学厚度是衡量瑞利散射强度的重要指标。通过对大气散射理论和瑞利光学厚度理论的梳理，总结了现有瑞利光学厚度两类模拟模型的优缺点。随着全球气候变化中CO₂浓度已突破400 ppm，近似数值模型因受到大气温度和CO₂浓度为300 ppm的背景条件的限制会导致部分模型误差的增加；而理论离散模型虽然有明确的物理意义，对CO₂浓度也具有自适应性，模拟结果理论上可信可靠，但各相关输入物理参数求解复杂。为获得满足CO₂浓度为400 ppm的近似数值模型，通过对不同高度和纬度的九个试验地点，以理论离散模型为基础，模拟特定大气条件下（P₀=1 atm， T=15 ℃， CO₂=400 ppm）的瑞利光学厚度。通过拟合分析得出，瑞利散射强度与波长的4.529次方成反比，且在紫外—蓝波段CO₂浓度对瑞利光学厚度的贡献在10^-4—10^-3数量级。因此，在CO₂浓度发生改变的情况下，以理论离散模型为主要算法模拟瑞利光学厚度将能更好的提高模型的自适应性并减少模型本身带来的误差；并通过该模拟结果可进一步获得该大气条件下的计算简单方便的数值模拟模型。

Rayleigh scattering is one of the most important and common natural phenomena in life. Moreover, Rayleigh optical depth (ROD) is a significant index for measuring Rayleigh scattering intensity. By combining the theories of atmospheric scattering and ROD, we summarize in this paper the advantages and disadvantages of existing ROD simulation models. We find that, first, given that the CO₂ concentration in global climate change has exceeded 400 ppm, some modeling errors will arise due to the parametric limitations of the atmospheric temperature and the background conditions of 300 ppm CO₂ concentration in the approximate numerical model. Second, although the theoretical discrete model has a clear physical meaning and self-adaptability to the CO₂ concentration index, and presuming that the simulation results are reliable in theory, deriving the solutions of various relevant input physical parameters is complicated. In obtaining an approximate numerical model with 400 ppm CO₂ concentration, we first simulated ROD under specific atmospheric conditions (P₀=1 atm, T=15℃, CO₂=400 ppm) based on a theoretical discrete model for nine test sites with different heights and latitudes. Then, we analyzed and fitted the ROD as a function of wavelength and altitude. The Rayleigh scattering intensity was set to be inversely proportional to the square of 4.529 times of the wavelength. Furthermore, the contribution of CO₂ concentration in the ultraviolet-blue band to Rayleigh optical depth was in the order of 10^-4 to 10^-3. In the case of changing CO₂ concentrations, we suggest that the theoretical discrete model be used as the main algorithm to simulate ROD. In this manner, the adaptability of the model can be improved and the errors resulting from the modeling itself can be reduced. Furthermore, on the basis of the simulation results, a simple and convenient numerical simulation model for atmospheric conditions can be obtained.