Improved algorithm of aerosol particle size distribution based on remote sensing data
Improved algorithm of aerosol particle size distribution based on remote sensing data
复制标题
基于遥感数据的气溶胶粒径分布改进算法
DOI：
10.1364/ao.58.008075
复制
发表时间：
2019-10-10
期刊：
影响因子：
1.9
通讯作者：
Hua, Dengxin
中科院分区：
文献类型：
--
作者：
Yan, Qing;Di, Huige;Hua, Dengxin
多波长光束的光学参数（消光系数或后向散射系数）可用于反演气溶胶粒子尺寸分布（APSD）。研究和讨论了仅利用多波长消光系数数据计算APSD和气溶胶微物理参数（AMP）的改进算法。将正则化算法和先验值相结合，用于APSD和AMP的检索。将基于最小偏差原理和平均过程的正则化算法用于精细模式APSD的反演，并提出了一种能够获得稳定输出的平均过程。选择并验证接近最小差异的1%平均结果。基于正则化算法的精细模态反演结果，采用具有先验值（模型半径）的对数正态分布对数据进行拟合，重构出APSD的粗模态。正则化算法和平均过程的综合应用提高了精细模式下反演的稳定性，先验值的使用拓宽了APSD的反演半径范围。这种方法不需要假设复折射率。对不同类型气溶胶的反演误差进行了分析和研究。通过多个典型的APSD值和不同污染日的粒径谱仪实测APSD值验证了算法的可靠性。算法的灵敏度分析也提供和讨论。该算法可以在较大的半径范围内获得可靠的APSD和AMP反演。(C)2019美国光学学会
The optical parameters (extinction or backscatter coefficients) of multi-wavelength beams can be used for the retrieval of the aerosol particle size distribution (APSD). An improved algorithm for APSD and aerosol micro-physical parameters (AMPs) is studied and discussed by using only multi-wavelength extinction coefficients data. The regularized algorithm and prior value are combined, for the retrieval of APSD and AMPs. The regularization algorithm, based on minimum discrepancy principle and averaging procedure, is used for the retrieval of fine-mode APSD and an averaging procedure that can achieve stable outputs is proposed. The 1% averaging result near the minimum of the discrepancy is selected and verified. Based on the inversion results of fine mode from the regularization algorithm, the lognormal distribution with a prior value (model radius) is applied to reconstruct the coarse mode of APSDs through fitting the data. The comprehensive application of the regularization algorithm and averaging process improves the stability of the inversion in the fine mode, and the use of the prior value broadens the inversion radius range of APSD. The complex refractive index need not be assumed for this method. The inversion error for different types of aerosols is analyzed and studied. The reliability of the algorithm is tested and verified by many typical APSDs and the measured APSDs by particle size spectrometer in different pollution days. The algorithm sensitivity analysis is also provided and discussed. The algorithm can obtain reliable inversion of APSD and AMPs with large radius range. (C) 2019 Optical Society of America