Recently, the research team led by Researcher He Xianqiang from our laboratory and their collaborators have published a research paper entitled Adjacency effect on Rayleigh scattering radiance for satellite remote sensing of river waters in the well-known remote sensing journal IEEE Transactions on Geoscience and Remote Sensing. The first author is Zhao Yaqi, a joint-training doctoral candidate jointly cultivated by our laboratory and Zhejiang University, and the corresponding author is Researcher He Xianqiang. The co-authors include Researcher Bai Yan, Researcher Wang Difeng, Associate Researcher Li Teng, Senior Engineer Gong Fang, Assistant Researcher Jin Xuchen from our laboratory, as well as Dr. Xu Fei from the Shanghai Institute of Satellite Engineering.
In ocean color satellite remote sensing of coastal and inland waters, especially narrow river waters, the adjacency pixel effect caused by surrounding high-albedo land cannot be ignored. This effect leads to the overestimation of satellite-received radiance over water pixels. Without accurate quantification and correction, it is impossible to precisely retrieve required water-leaving radiance from satellite radiative signals, which further brings uncertainties to the inversion of subsequent ocean color parameters. The adjacency pixel effect exerts particularly prominent impacts on elongated river waters, yet comprehensive understanding of its influences on riverine water remote sensing is still lacking worldwide.
Fig.1 Schematic diagram of adjacency pixel effect (purple and orange represent radiative contributions from adjacent pixels)
To address the above issues, this study firstly developed an inhomogeneous underlying surface 3D coupled water-atmosphere vector radiative transfer model named NUS-MC applicable to river waters based on the Monte Carlo method. Verified against Rayleigh scattering lookup tables of SeaDAS, PCOART model and 6Sv model, the results prove that NUS-MC can accurately simulate atmospheric molecular Rayleigh scattering radiance under both homogeneous and inhomogeneous underlying surface conditions.
Based on the NUS-MC model, this study systematically clarified the influences of adjacency pixel effect on Rayleigh scattering radiance over river waters under varying solar incident angles, satellite viewing angles, river widths, offshore distances, river orientations and land albedo conditions. The results show that for large rivers with a width of 10 km, in the blue band at 443 nm, the influences of adjacency pixel effect on Rayleigh scattering radiance at river center can reach 3.97%, 18.19% and 33.37% when the land albedo is 0.1, 0.3 and 0.5 respectively. This indicates that adjacency effects must be taken into account for precise atmospheric correction of river water remote sensing.
Moreover, the adjacency pixel effect decreases with the increase of solar zenith angle, rises with the growth of satellite viewing zenith angle, and is sensitive to viewing azimuth. This study confirms that the adjacency pixel effect on Rayleigh scattering over river waters involves complex mechanisms affected by multiple factors, and its impacts on atmospheric correction are non-negligible. Future research will further explore its influences on aerosol scattering and develop targeted atmospheric correction methods for river waters considering adjacency pixel effects.
Fig.2 Flowchart of the NUS-MC model
Fig.3 Percentage influences of adjacency pixel effect on Rayleigh scattering at river center under different river widths and land albedo at 443 nm. Curves in different colors denote results under different viewing zenith angles (VZA), and the black line represents the average value.
This achievement marks another breakthrough made by Researcher He Xianqiang’s team following their previous development of the spherical coupled air-sea vector radiative transfer model considering Earth curvature effects. It lays a solid foundation for developing ocean color remote sensing atmospheric correction algorithms that incorporate adjacency pixel effects.
This research was financially supported by the National Key R&D Program of China (Grant No. 2023YFC3108101), Zhejiang Provincial Pioneer R&D Program (Grant No. 2033C03011) and the National Natural Science Foundation of China (Grant No. U23A2037).
Citation
Zhao, Y., He, X.*, Bai, Y., Xu, F., Jin, X., Li, T., Wang, D., Gong, F., 2024. Adjacency effect on Rayleigh scattering radiance for satellite remote sensing of river waters. IEEE Transactions on Geoscience and Remote Sensing, DOI: 10.1109/TGRS.2024.3487906
