Li Shichun, Dang Yuanyuan, Zhang Penghui, Hua Dengxin, Gao Yingchun, Di Huige, Xin Wenhui
Appl Opt. 2024 Feb 10;63(5):1210-1216. doi: 10.1364/AO.509724.
Aimed at the stability of calibration coefficients in a general non-orthogonal retrieval algorithm (NRA) of pure rotational Raman lidars (PRRLs), an orthogonal retrieval algorithm (ORA) of atmospheric temperature profiles based on the orthogonal basis function is proposed. This algorithm eliminates the correlation between the calibration coefficients in the NRA to reduce the influence of the number of calibration points and the selection scheme on the calibration coefficients. In this paper, the stabilities of calibration coefficients in the NRA and ORA are compared and analyzed, and the data analysis for atmospheric temperature profiles with a time resolution of minute-level are given, based on the developed Cloud Precipitation Potential Evaluation (CPPV) lidar data and the parallel radiosonde temperature data. The analysis results show that coefficients of variation (CVs) of ORA calibration coefficients are one order of magnitude smaller than those of NRA coefficients. The mean deviation of the ORA retrieval results is roughly reduced by 16.1% compared with the NRA, and the root-mean-square deviation is roughly reduced by 15.0% compared with the NRA. Therefore, the temperature retrieval performance of the ORA is better than that of the NRA.
针对纯转动拉曼激光雷达（PRRL）通用非正交反演算法（NRA）中校准系数的稳定性问题，提出了一种基于正交基函数的大气温度廓线正交反演算法（ORA）。该算法消除了NRA中校准系数之间的相关性，以减少校准点数和选择方案对校准系数的影响。本文基于研制的云降水潜力评估（CPPV）激光雷达数据和平行探空仪温度数据，比较分析了NRA和ORA中校准系数的稳定性，并给出了分钟级时间分辨率的大气温度廓线数据分析。分析结果表明，ORA校准系数的变异系数（CV）比NRA系数小一个数量级。与NRA相比，ORA反演结果的平均偏差大致降低了16.1%，均方根偏差大致降低了15.0%。因此，ORA的温度反演性能优于NRA。
