基于CO2测量数据的大气辐射传输模型LBLRTM优化

根据TIMED/SABER 2002—2018年的CO₂观测数据，分析CO₂浓度的变化特征.依据变化特征给出了CO₂浓度随时间、高度、纬度变化的月平均拟合公式，利用非线性最小二乘拟合法，对不同高度和不同纬度的CO₂浓度数据分别进行拟合，生成相应的拟合参数.然后，将所有拟合参数汇总并生成拟合参数文件，结合拟合公式构建全球CO₂浓度经验计算模块，并将该模块应用到大气辐射传输模型LBLRTM中，对该模型进行优化.将优化前与优化后的LBLRTM模型模拟结果分别与TIMED/SABER观测数据进行比较发现，优化前的LBLRTM模型模拟结果与观测值的均方根误差为15.4%，而优化后的LBLRTM模型模拟结果与观测值的均方根误差由15.4%下降至8.91%.结果表明该优化方法可以提高LBLRTM模型在红外波段的辐射模拟精度. 

Optimization of Atmospheric Radiative Transfer Model LBLRTM Based on Measured CO2 Data

According to the CO2 observation data of TIMED/SABER from 2002 to 2018, analyze the variation characteristics of CO2 concentration, gives the monthly average fitting formula of CO2 concentration varying with time, height and latitude according to the various characteristics, and uses the nonlinear least square fitting method to fit the CO2 concentration data of different heights and latitudes respectively to generate corresponding fitting parameters, Then, summarize all fitting parameters and generate the fitting parameter file. Combined with the fitting formula, the empirical calculation module of global CO2 concentration is constructed, and the module is applied to the atmospheric radiative transfer model LBLRTM to optimize LBLRTM. Comparing the simulation results of the optimized LBLRTM model with the TIMED/SABER observation data, the root mean square error between the simulation results of the not optimized LBLRTM model and the observation value is 15.4%, while the root mean square error between the simulation results of the optimized LBLRTM model and the observation value is reduced to 8.91%. The results show that this optimization method can further improve the radiation simulation accuracy of the LBLRTM model in the infrared band.