Impact of the ozone monitoring instrument row anomaly on the long-term record of aerosol products
Impact of the ozone monitoring instrument row anomaly on the long-term record of aerosol products
复制标题
DOI：
10.5194/amt-11-2701-2018
复制
发表时间：
2017-12
影响因子：
3.8
通讯作者：
O. Torres;P. Bhartia;H. Jethva;H. Jethva;C. Ahn
中科院分区：
文献类型：
--
作者：
O. Torres;P. Bhartia;H. Jethva;H. Jethva;C. Ahn
抽象的。在EOS-Aura卫星上发射臭氧监测仪(OMI)大约三年后，传感器的观测能力一直受到据信是内部障碍的影响，这降低了OMI的空间覆盖范围。目前，它影响了该仪器60个观察位置中的大约一半。在这项工作中，我们利用2005-2007年行异常开始前三年的时间段进行了分析，以评估空间覆盖减少对反演气溶胶光学厚度(AOD)、单次散射反照率(SSA)和紫外线气溶胶指数(UVAI)月平均值的影响。将使用观察位置1到30计算的区域月平均值与使用位置31到60计算的类似获得的值进行比较，期望在两种计算之间找到接近一致的结果。正如预期的那样，利用OMI观测的这两个散射角相关子集获得的AOD和SSA的月平均值在碳质或硫酸盐气溶胶颗粒为主要气溶胶类型的区域一致。然而，在沙漠沙尘为主要气溶胶类型的干旱地区，两组计算的AOD区域平均值之间存在显著差异。结果表明，与散射角相关的观测子集之间的沙漠尘埃AOD的差异是由于沙漠尘埃散射相函数的不正确表示所致。利用辐射传输计算进行的灵敏度分析表明，观测到的气溶胶光学厚度偏差的来源是沙漠沙尘粒子的球形假设。根据UVAI进行的类似分析得出，两组计算在多云地区的月平均值有很大差异。相反，在云层最少的干旱地区，两组观测得到的UVAI月平均值非常接近。在多云条件下的差异被发现是由于云作为不透明的朗伯反射器的参数化造成的。当使用Mie理论适当地考虑云散射效应时，观测到的UVAI角偏差显著减小。这里讨论的分析揭示了与沙漠尘埃、气溶胶和云滴散射效应的角度相关性的模型表示有关的重要算法缺陷。由此产生的沙漠尘埃和云层散射处理方面的改进已被纳入OMAERUV算法的改进版本中。
Abstract. Since about three years after the launch the Ozone Monitoring Instrument (OMI) on the EOS-Aura satellite, the sensor's viewing capability has been affected by what is believed to be an internal obstruction that has reduced OMI's spatial coverage. It currently affects about half of the instrument's 60 viewing positions. In this work we carry out an analysis to assess the effect of the reduced spatial coverage on the monthly average values of retrieved aerosol optical depth (AOD), single scattering albedo (SSA) and the UV Aerosol Index (UVAI) using the 2005–2007 three-year period prior to the onset of the row anomaly. Regional monthly average values calculated using viewing positions 1 through 30 were compared to similarly obtained values using positions 31 through 60, with the expectation of finding close agreement between the two calculations. As expected, mean monthly values of AOD and SSA obtained with these two scattering-angle dependent subsets of OMI observations agreed over regions where carbonaceous or sulphate aerosol particles are the predominant aerosol type. However, over arid regions, where desert dust is the main aerosol type, significant differences between the two sets of calculated regional mean values of AOD were observed. As it turned out, the difference in retrieved desert dust AOD between the scattering-angle dependent observation subsets was due to the incorrect representation of desert dust scattering phase function. A sensitivity analysis using radiative transfer calculations demonstrated that the source of the observed AOD bias was the spherical shape assumption of desert dust particles. A similar analysis in terms of UVAI yielded large differences in the monthly mean values for the two sets of calculations over cloudy regions. On the contrary, in arid regions with minimum cloud presence, the resulting UVAI monthly average values for the two sets of observations were in very close agreement. The discrepancy under cloudy conditions was found to be caused by the parameterization of clouds as opaque Lambertian reflectors. When properly accounting for cloud scattering effects using Mie theory, the observed UVAI angular bias was significantly reduced. The analysis discussed here has uncovered important algorithmic deficiencies associated with the model representation of the angular dependence of scattering effects of desert dust aerosols and cloud droplets. The resulting improvements in the handling of desert dust and cloud scattering have been incorporated in an improved version of the OMAERUV algorithm.