10:45 AM - 11:00 AM
[AOS12-06] A study on optimization method of atmospheric correction in coastal areas using GCOM-C/SGLI
Keywords:Optical satellites, Ocean color remote sensing, Atmospheric correction
Water quality estimation using optical satellites is a suitable observation method for understanding the marine environment spatially. In particular, GCOM-C/SGLI, which has a spatial resolution of 250 m and observes every 2 or 3 days, is expected to be utilized for detailed and sustainable monitoring of water environments. Atmospheric scattered light accounts for most of satellite observed light, so it is necessary to remove atmospheric effects accurately when using satellite data. For SGLI, Remote Sensing Reflectance (Rrs) is retrieved through an atmospheric correction process using standard atmospheric correction algorithm, and then water quality is estimated. However, water quality estimation in coastal areas is difficult due to the underestimation of atmospheric correction outputs at visible short wavelength. This is because anthropogenic absorptive aerosols are not considered in this standard correction algorithm, and aerosol reflectances at visible wavelengths are estimated by extrapolation using the ratio of aerosol optical thickness at two near-infrared wavelengths (673.5 nm and 868.5 nm).
In this study, an optimization method for atmospheric correction results in coastal areas using in-water model was developed. In this algorithm, Rrs retrieved by standard atmospheric correction is re-estimated by recalculating aerosol reflectance and using bio-optical model. Developed algorithm was compared and verified with field observations, and it was suggested that underestimated Rrs could be optimized. Sensitivity analysis showed that light absorption coefficients of phytoplankton and those of detritus and CDOM were highly sensitive to the model. Improved spatial distributions of Rrs in coastal areas indicated the effectiveness of this algorithm.
In this study, an optimization method for atmospheric correction results in coastal areas using in-water model was developed. In this algorithm, Rrs retrieved by standard atmospheric correction is re-estimated by recalculating aerosol reflectance and using bio-optical model. Developed algorithm was compared and verified with field observations, and it was suggested that underestimated Rrs could be optimized. Sensitivity analysis showed that light absorption coefficients of phytoplankton and those of detritus and CDOM were highly sensitive to the model. Improved spatial distributions of Rrs in coastal areas indicated the effectiveness of this algorithm.