论文部分内容阅读
To acquire high-quality operational data products for Chinese in-orbit and scheduled ocean color sensors, the performances of two operational atmospheric correction(AC) algorithms(ESA MEGS 7.4.1 and NASA Sea DAS 6.1) were evaluated over the East China Seas(ECS) using MERIS data. The spectral remote sensing reflectance R_(rs)(λ), aerosol optical thickness(AOT), and ?ngstr?m exponent(α) retrieved using the two algorithms were validated using in situ measurements obtained between May 2002 and October 2009. Match-ups of R_(rs), AOT, and α between the in situ and MERIS data were obtained through strict exclusion criteria. Statistical analysis of R_(rs)(λ) showed a mean percentage difference(MPD) of 9%–13% in the 490–560 nm spectral range, and significant overestimation was observed at 413 nm(MPD>72%). The AOTs were overestimated(MPD>32%), and although the ESA algorithm outperformed the NASA algorithm in the blue-green bands, the situation was reversed in the red-near-infrared bands. The value of α was obviously underestimated by the ESA algorithm(MPD=41%) but not by the NASA algorithm(MPD=35%). To clarify why the NASA algorithm performed better in the retrieval of α, scatter plots of the α single scattering albedo(SSA) density were prepared. These α-SSA density scatter plots showed that the applicability of the aerosol models used by the NASA algorithm over the ECS is better than that used by the ESA algorithm, although neither aerosol model is suitable for the ECS region. The results of this study provide a reference to both data users and data agencies regarding the use of operational data products and the investigation into the improvement of current AC schemes over the ECS.
To acquire high-quality operational data products for Chinese in-orbit and scheduled ocean color sensors, the performances of two operational atmospheric correction (AC) algorithms (ESA MEGS 7.4.1 and NASA Sea DAS 6.1) were evaluated over the East China Seas ( ECS) using the spectral remote sensing reflectance R_ (rs) (λ), aerosol optical thickness (AOT), and? Ngstr? M exponent (α) using using the two algorithms were validated using in situ measurements obtained between May 2002 and October 2009. Match-ups of R_ (rs), AOT, and α between the in situ and MERIS data were obtained through strict exclusion criteria. Statistical analysis of R_ (rs) (λ) showed a mean percentage difference (MPD) of 9% -13% in the 490-560 nm spectral range, and significant overestimation was observed at 413 nm (MPD> 72%). The AOTs were overestimated (MPD> 32%), and although the ESA algorithm outperformed the NASA algorithm in the blue-green bands, the situation was reversed in the red-near-infrared bands. NASA algorithm performed better in the retrieval of α, scatter plots of the α single (MPD = 41%) but not by NASA algorithm (MPD = 35%) These α-SSA density scatter plots showed that the applicability of the aerosol models used by the NASA algorithm over the ECS is better than that used by the ESA algorithm, although neither an aerosol model is suitable for the ECS region. The results of this study provide a reference to both data users and data agencies regarding the use of operational data products and the investigation into the improvement of current AC schemes over the ECS.