Development of a Practical Atmospheric Correction Algorithm for Inland and Nearshore Coastal Waters

Abstract

A practical Atmospheric Correction algorithm for inLand and Nearshore Coastal waters (ACLANC) is proposed in this study. The ACLANC algorithm uses interpolated aerosol optical depth (AOD) products (AOD <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\mathrm {interp}}$ </tex-math></inline-formula> ) from nearby land surfaces and simulates the corresponding aerosol reflectance spectrum using a combination of the continental model in the satellite signal in the solar spectrum-vector (6SV) radiative transfer code and an approximate aerosol model (r85f20) in the Sea-viewing Wide Field-of view Sensor (SeaWiFS) Data Analysis System (SeaDAS). Validations with worldwide <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> measurements show that the ACLANC-derived remote-sensing reflectance ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\mathrm {rs}}$ </tex-math></inline-formula> ) for nine Moderate Resolution Imaging Spectroradiometer (MODIS) bands agreed well with the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> datasets, where the mean <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R^{2}$ </tex-math></inline-formula> was 0.77 ± 0.09 and the mean unbiased percent difference was 28.7% ± 9.8%. ACLANC outperformed the existing atmospheric correction algorithms in not only the accuracy of the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\mathrm {rs}}$ </tex-math></inline-formula> retrievals but also data coverage. Vicarious calibration over the ACLANC algorithm showed minor improvement in the derived <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\mathrm {rs}}$ </tex-math></inline-formula> products. Error budget analysis revealed that the uncertainties in AOD <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">interp</sub> represent >50% of the errors for ACLANC and that this proportion increases with decreasing AOD. Further efforts can also be applied to improve the aerosol models, especially for turbid aerosol environments, where the fixed aerosol model in SeaDAS contributes up to 30% of the error budget. The ACLANC algorithm can potentially be implemented in ocean color missions other than MODIS to obtain <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\mathrm {rs}}$ </tex-math></inline-formula> with high accuracy and wide coverage for global inland and nearshore coastal waters.