摘要
基于Fernald方法和Klett方法,推导出一个具有明确物理意义的确定气溶胶消光系数边界值的表达式,该式比目前用于确定边界值的Collis斜率法表达式增加了两项:空气分子消光系数项和后向散射项,这两项与Collis斜率法的值符号相反。空气分子消光系数项较小,但后向散射项为后向散射系数的倒数和导数的乘积,绝对值能达到Collis斜率法的75.2%。分析表明,考虑了新增两项反演的大气气溶胶光学厚度(AOD)与实测更接近,所以增加这两项是合理的、必要的。不同标高反演2007年9月20日的AOD在0.20～0.25之间,变化范围较小;反演的AOD方差为0.003,相对较小。说明新方法对标高的依赖较小且较稳定。分析424个时次晴空资料的反演结果可知,反演比实测大7.4%,反演与实测的相关系数为93.2%,相对误差和绝对误差分别为10.9%和0.03,反演的AOD方差为0.02,AOD小于0.45(占到资料总数的91.7%)时,反演结果较好。
Based on Fernald＇s and Klett＇s method, a new expression of the atmospheric extinction coefficient boundary value was presented. This expression was a formula that has two more terms than Collis＇ method. The absolute value of the two terms may account for 78.2% of the term in Collis＇ method and have opposite sign. Taking the two terms into consideration, aerosol optical depth （AOD）derived by lidax is closer to the observed one by sun-photometer. With the change of reference height, the derived AOD ranging from 0.20 to 0.25 and the variance is about 0.003, showing that this approach is independent from reference height and relatively stable. 424 clear sky data were selected in the inversion. AOD derived by lidar is 7.4% bigger than the observed one. The correlation coefficient is 0.932, the relative error is 10.9%, the absolute error is 0.03 and the variance is 0.02. The approach has a better performance when AOD is less than 0.45.
出处
《量子电子学报》
CAS
CSCD
北大核心
2013年第1期57-65,共9页
Chinese Journal of Quantum Electronics
基金
国家自然科学基金(41075104)
国家重点基础研究发展规划(2010CB428604)资助项目
关键词
大气遥感
激光雷达
消光系数
边界值
标定高度
光学厚度
remote sensing
lidar
extinction coefficient
boundary value
reference height
aerosol opticaldepth
作者简介
田鹏飞（1987-），甘肃人，研究生，主要从事激光雷达反演大气气溶胶和边界层高度研究。E-mail：tianpf07@lzu．edu．cn
导师简介：张镭（2960-），山西人，博士，教授，从事大气物理与大气环境，大气气溶胶及其气候效应，大气污染及其控制方面的教学和研究。E-mail：zhanglei@lzu．edu．cn.
