Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Performance of MODIS high-resolution MAIAC aerosol algorithm in China: Characterization and limitation
https://orcid.org/0000-0002-2528-2943
https://orcid.org/0000-0002-7334-0490
Abstract The MODIS Multiple Angle Implication of Atmospheric Correction (MAIAC) algorithm enables simultaneous retrieval of aerosol and bidirectional surface reflectance at high resolution of 1 km. Taking advantage of multi-angle and image-based information, the MAIAC algorithm has great potential for improving retrieval of aerosols over both dark and bright surfaces. Here, by comparing MAIAC aerosol products with the ground-based observations at 9 typical sites spread out in China, we gain the insights regarding the performance of MAIAC algorithm, for the first time, over Asia that has complicated surface types, diverse aerosol sources, and heavy loading of aerosols in the atmosphere. While aerosol products from MAIAC show similar spatial distribution as that from MODIS Dark-Target (DT) and Deep-Blue (DB) algorithms, they are superior to reveal numerous hotspots of high AOD values in fine scales due to their higher resolution at 1 km. Moreover, since MAIAC algorithm for cloud screening uses time series of observations, it shows higher effectiveness to mask cloudy pixels as well as the pixels of the melting and aging ice/snow surfaces. While MAIAC and ground-observed AOD values show high correlation coefficient of ∼0.94 in two AERONET sites of Beijing and Xianghe, considerable bias is prevalent in other regions of China. Systematic underestimation is found over the deserts in western China likely due to the high bias of single scattering properties of aerosol model prescribed in MAIAC algorithm. In eastern China, the distinct positive bias is found in conditions with low-moderate AOD values and likely results from errors in regression coefficients in the surface reflectance model. Given its advantages in cloud and snow/ice screening and retrieval in fine spatial resolution, MAIAC algorithm can be improved by further refinement of regional aerosol and surface properties.
Highlights First comprehensive insight into performance of MAIAC aerosol algorithm in complicated background in China. Accuracy of MAIAC retrievals exhibits distinct spatial variations and prevalent bias. Aerosol and surface assumptions of MAIAC algorithm needs to be improved by considering regional variations.
Performance of MODIS high-resolution MAIAC aerosol algorithm in China: Characterization and limitation
https://orcid.org/0000-0002-2528-2943
https://orcid.org/0000-0002-7334-0490
Abstract The MODIS Multiple Angle Implication of Atmospheric Correction (MAIAC) algorithm enables simultaneous retrieval of aerosol and bidirectional surface reflectance at high resolution of 1 km. Taking advantage of multi-angle and image-based information, the MAIAC algorithm has great potential for improving retrieval of aerosols over both dark and bright surfaces. Here, by comparing MAIAC aerosol products with the ground-based observations at 9 typical sites spread out in China, we gain the insights regarding the performance of MAIAC algorithm, for the first time, over Asia that has complicated surface types, diverse aerosol sources, and heavy loading of aerosols in the atmosphere. While aerosol products from MAIAC show similar spatial distribution as that from MODIS Dark-Target (DT) and Deep-Blue (DB) algorithms, they are superior to reveal numerous hotspots of high AOD values in fine scales due to their higher resolution at 1 km. Moreover, since MAIAC algorithm for cloud screening uses time series of observations, it shows higher effectiveness to mask cloudy pixels as well as the pixels of the melting and aging ice/snow surfaces. While MAIAC and ground-observed AOD values show high correlation coefficient of ∼0.94 in two AERONET sites of Beijing and Xianghe, considerable bias is prevalent in other regions of China. Systematic underestimation is found over the deserts in western China likely due to the high bias of single scattering properties of aerosol model prescribed in MAIAC algorithm. In eastern China, the distinct positive bias is found in conditions with low-moderate AOD values and likely results from errors in regression coefficients in the surface reflectance model. Given its advantages in cloud and snow/ice screening and retrieval in fine spatial resolution, MAIAC algorithm can be improved by further refinement of regional aerosol and surface properties.
Highlights First comprehensive insight into performance of MAIAC aerosol algorithm in complicated background in China. Accuracy of MAIAC retrievals exhibits distinct spatial variations and prevalent bias. Aerosol and surface assumptions of MAIAC algorithm needs to be improved by considering regional variations.
Performance of MODIS high-resolution MAIAC aerosol algorithm in China: Characterization and limitation
https://orcid.org/0000-0002-2528-2943
https://orcid.org/0000-0002-7334-0490
Abstract The MODIS Multiple Angle Implication of Atmospheric Correction (MAIAC) algorithm enables simultaneous retrieval of aerosol and bidirectional surface reflectance at high resolution of 1 km. Taking advantage of multi-angle and image-based information, the MAIAC algorithm has great potential for improving retrieval of aerosols over both dark and bright surfaces. Here, by comparing MAIAC aerosol products with the ground-based observations at 9 typical sites spread out in China, we gain the insights regarding the performance of MAIAC algorithm, for the first time, over Asia that has complicated surface types, diverse aerosol sources, and heavy loading of aerosols in the atmosphere. While aerosol products from MAIAC show similar spatial distribution as that from MODIS Dark-Target (DT) and Deep-Blue (DB) algorithms, they are superior to reveal numerous hotspots of high AOD values in fine scales due to their higher resolution at 1 km. Moreover, since MAIAC algorithm for cloud screening uses time series of observations, it shows higher effectiveness to mask cloudy pixels as well as the pixels of the melting and aging ice/snow surfaces. While MAIAC and ground-observed AOD values show high correlation coefficient of ∼0.94 in two AERONET sites of Beijing and Xianghe, considerable bias is prevalent in other regions of China. Systematic underestimation is found over the deserts in western China likely due to the high bias of single scattering properties of aerosol model prescribed in MAIAC algorithm. In eastern China, the distinct positive bias is found in conditions with low-moderate AOD values and likely results from errors in regression coefficients in the surface reflectance model. Given its advantages in cloud and snow/ice screening and retrieval in fine spatial resolution, MAIAC algorithm can be improved by further refinement of regional aerosol and surface properties.
Highlights First comprehensive insight into performance of MAIAC aerosol algorithm in complicated background in China. Accuracy of MAIAC retrievals exhibits distinct spatial variations and prevalent bias. Aerosol and surface assumptions of MAIAC algorithm needs to be improved by considering regional variations.
Performance of MODIS high-resolution MAIAC aerosol algorithm in China: Characterization and limitation
Tao, Minghui (Autor:in) / Wang, Jun (Autor:in) / Li, Rong (Autor:in) / Wang, Lili (Autor:in) / Wang, Lunche (Autor:in) / Wang, Zifeng (Autor:in) / Tao, Jinhua (Autor:in) / Che, Huizheng (Autor:in) / Chen, Liangfu (Autor:in)
Atmospheric Environment ; 213 ; 159-169
01.06.2019
11 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Multi-angle implementation of atmospheric correction for MODIS (MAIAC): 3. Atmospheric correction
| Online Contents | 2012