首页 >  2020, Vol. 24, Issue (11) : 1363-1378

摘要

全文摘要次数: 1116 全文下载次数: 950
引用本文:

DOI:

10.11834/jrs.20208412

收稿日期:

2018-10-23

修改日期:

PDF Free   HTML   EndNote   BibTeX
O2-O2云反演算法及其在TROPOMI上的应用
张文强1,刘诚2,3,4,5,郝楠6,Gimeno Garcia Sebastian6,邢成志3,张成歆3,苏文静3,刘建国2
1.中国科学技术大学 环境科学与光电技术学院, 合肥 230026;2.中国科学院合肥物质科学研究院安徽光学精密机械研究所 环境光学与技术重点实验室, 合肥 230031;3.中国科学技术大学 地球和空间科学学院, 合肥 230026;4.中国科学院城市环境研究所 区域大气环境研究卓越创新中心, 厦门 361021;5.中国科学技术大学 极地环境与全球变化安徽省重点实验室, 合肥 230026;6.欧洲气象卫星应用组织, 达姆施塔特 64295
摘要:

利用遥感技术对大气环境污染进行监测时,云是影响痕量气体反演精度的重要因素,因此在痕量气体反演中需要对云的影响进行校正,通常使用的云参数主要是有效云量和云压。本文基于O2-O2 477 nm吸收波段构建了O2-O2云反演算法:首先,根据有效云量和云高与连续反射率和O2-O2斜柱浓度之间的对应关系,结合假定的云模型利用VLIDORT辐射传输模型建立关于有效云量和云压的查找表;然后,通过差分吸收光谱技术拟合卫星载荷观测的大气层顶辐射,获得O2-O2斜柱浓度并计算连续反射率;最后,结合辅助数据,根据查找表进行插值反演获得有效云量和云压。通过将算法应用到OMI观测数据,将反演结果与OMCLDO2产品进行对比验证,有效云量和云压空间分布一致,相关系数R均超过0.97;并还将该算法应用于下一代大气成分监测仪器TROPOMI,与FRESCO+产品对比,有效云量和云压空间分布基本一致,当地表类型为海洋时,有效云量相关系数R大于0.97,云压相关系数R大于0.94,云压反演结果存在一定的区别;通过将O2-O2云反演算法和FRESCO+云压反演结果与CALIOP Cloud Layer产品进行对比,结果表明,在低云情况下,O2-O2云反演算法线性回归方程斜率为0.782,截距为198.0 hPa,相关系数R为0.850,算法表现优于FRESCO+。而在高云情况下,FRESCO+反演结果更接近CALIOP云压结果。在OMI和TROPOMI上的应用表明O2-O2云反演算法在大气云反演中具有较高的准确性和可行性,可以为大气痕量气体反演的校正提供云参数,并为中国同类型卫星载荷的云反演算法提供算法参考。

O2-O2 cloud retrieval algorithm and application to TROPOMI
Abstract:

The cloud covers more than 50% of the Earth, which plays a major role in radiation budget of Earth climate system and hydrological cycle through their strong impact on radiation process. Cloud is an important factor to affect the retrieval accuracy of trace gases during the measurement of air pollution based on remote sensing method. The effective cloud fraction and cloud pressure should be used in the process of correcting cloud effects. In this paper, O2-O2 cloud retrieval algorithm based on O2-O2 absorption band at 477 nm will be described.The O2-O2 cloud retrieval algorithm is developed based on the Look-Up Table (LUT) method. Effective cloud fraction and cloud pressure LUTs were generated using VLIDORT radiation transfer model based on the relationships of effective cloud fraction, cloud pressure, continuum reflectance and O2-O2 slant column density. Then, the Differential Optical Absorption Spectroscopy (DOAS) is used to fit the radiance of top-of-atmosphere measured by the satellite payload, to obtain O2-O2 slant column density and the continuum reflectance. Finally, combined with the auxiliary data, the effective cloud fraction and cloud pressure are retrieved by the interpolation based on the LUTs.We have a validation between the O2-O2 cloud retrieval algorithm results based on OMI data and OMCLDO2 products. The spatial distribution of the effective cloud fraction and cloud pressure show great consistency, and the correlation coefficients (R) between them are greater than 0.97. Then the O2-O2 cloud retrieval algorithm was applied to the new generation of atmospheric sounding instrument TROPOMI. The cloud retrieval results also show high correlation compared with the FRESCO+ results, and R of effective cloud fraction and cloud pressure between above two results are greater than 0.97 and 0.95, respectively, when the surface type is ocean. The time series analysis of three months in Beijing from the two algorithms shows the good consistency of the retrieval results. Moreover, we compared with CALIOP Cloud Layer products using all the retrieved cloud pressure results of FRESCO+ and O2-O2 cloud retrieval algorithm. O2-O2 cloud retrieval algorithm performed better than FRESCO+ under low cloud condition. However, the FRESCO+ retrieval results are closer to CALIOP cloud pressure than O2-O2 cloud retrieval algorithm under high cloud condition.The O2-O2 cloud retrieval algorithm was used during the cloud retrieval of OMI and TROPOMI, and it shows high accuracy and feasibility. This kind of algorithm can provide information of effective cloud fraction and cloud pressure in the process of atmospheric trace gases retrieval. The more important is that O2-O2 cloud retrieval algorithm could provide reference for the development of cloud retrieval algorithm applied to the same type of satellite payload of China.

本文暂时没有被引用!

欢迎关注学报微信

遥感学报交流群