Joint Retrieval of Aerosol Optical Depth and Surface Reflectance Over Land Using Geostationary Satellite Data
Yanqing Xie,Yong Xue,C. Fan,Xihua Yang,Ying Li,Y. Che,J. Guang,L. She,John F. Leys
DOI: https://doi.org/10.1109/TGRS.2018.2867000
IF: 8.2
2019-03-01
IEEE Transactions on Geoscience and Remote Sensing
Abstract:The advanced Himawari imager (AHI) aboard the Himawari-8 geostationary satellite provides high-frequency observations with broad coverage, multiple spectral channels, and high spatial resolution. In this paper, AHI data were used to develop an algorithm for joint retrieval of aerosol optical depth (AOD) over land and land surface bidirectional reflectance. Instead of performing surface reflectance estimation before calculating AOD, the AOD and surface bidirectional reflectance were retrieved simultaneously using an optimal estimation method. The algorithm uses an atmospheric radiative transfer model coupled with a surface bidirectional reflectance factor (BRF) model. Based on the assumption that the surface bidirectional reflective properties are invariant during a short time period (i.e., a day), multiple temporal AHI observations were combined to calculate the AOD and surface BRF. The algorithm was tested over East Asia for year 2016, and the AOD retrieval results were validated against the aerosol robotic network (AERONET) sites observation and compared with the Moderate Resolution Imaging Spectroradiometer Collection 6.0 AOD product. The validation of the retrieved AOD with AERONET measurements using 14 713 colocation points in 2016 over East Asia shows a high correlation coefficient: $R = 0.88$ , root-mean-square error = 0.17, and approximately 69.9% AOD retrieval results within the expected error of $\pm 0.2\cdot {\mathrm {AOD}}_{\mathrm {AERONET}}\pm 0.05$ . A brief comparison between our retrieval and AOD product provided by Japan Meteorological Agency is also presented. The comparison and validation demonstrates that the algorithm has the ability to estimate AOD with considerable accuracy over land.
Environmental Science,Computer Science