TY - JOUR
T1 - An improved spectral optimization algorithm for atmospheric correction over turbid coastal waters
T2 - A case study from the Changjiang (Yangtze) estuary and the adjacent coast
AU - Pan, Yanqun
AU - Shen, Fang
AU - Verhoef, Wouter
N1 - Publisher Copyright:
© 2017 Elsevier Inc.
PY - 2017/3/15
Y1 - 2017/3/15
N2 - Remote sensing-based retrieval of the concentrations of water components relies largely on the accuracy of the atmospheric correction. Although a variety of atmospheric correction algorithms have been developed for turbid waters, the water-leaving reflectance is still underestimated in extremely turbid waters, such as in the Changjiang (Yangtze) estuary and the adjacent coast. To address this issue, this paper proposes an improved algorithm that is based on a spectral optimization algorithm (ESOA) with a coupled water-atmosphere model. The model combines an aerosol model that is constructed from Aerosol Robotic Network (AERONET) observation data and a simple semi-empirical radiative transfer (SERT) model (Shen et al. 2010) for water component retrieval. Four unknown parameters are involved in the coupled model: the relative humidity (RH), fine-mode fraction (FMF), aerosol optical thickness in the near-infrared (NIR) wavelength τa(λ0) and suspended particulate matter (SPM) concentration (Cspm). These parameters are estimated by a global optimization approach that is based on a genetic algorithm (GA) without any initial inputs. Validation results of the atmospherically corrected remote sensing reflectance Rrs(λ) from matchups between Geostationary Ocean Color Imager (GOCI) data and in situ data show that the algorithm has satisfactory accuracy. The root mean square error (RMSE) and the absolute percentage difference (APD) are 0.0089 and 35.12, respectively. By contrast, the Rrs(λ) values retrieved from the same matchups using the GOCI data processing system (GDPS) have higher RMSE and APD of 0.0104 and 69.15, respectively. The ESOA method can be implemented conveniently within the open source code of SeaDAS (v7.1) as an alternative and operational tool for atmospheric correction of ocean color data, including GOCI, MERIS and MODIS, over highly turbid estuarine and coastal regions, such as the Yangtze estuary, the Hangzhou Bay and most of the coastal ocean in Eastern China.
AB - Remote sensing-based retrieval of the concentrations of water components relies largely on the accuracy of the atmospheric correction. Although a variety of atmospheric correction algorithms have been developed for turbid waters, the water-leaving reflectance is still underestimated in extremely turbid waters, such as in the Changjiang (Yangtze) estuary and the adjacent coast. To address this issue, this paper proposes an improved algorithm that is based on a spectral optimization algorithm (ESOA) with a coupled water-atmosphere model. The model combines an aerosol model that is constructed from Aerosol Robotic Network (AERONET) observation data and a simple semi-empirical radiative transfer (SERT) model (Shen et al. 2010) for water component retrieval. Four unknown parameters are involved in the coupled model: the relative humidity (RH), fine-mode fraction (FMF), aerosol optical thickness in the near-infrared (NIR) wavelength τa(λ0) and suspended particulate matter (SPM) concentration (Cspm). These parameters are estimated by a global optimization approach that is based on a genetic algorithm (GA) without any initial inputs. Validation results of the atmospherically corrected remote sensing reflectance Rrs(λ) from matchups between Geostationary Ocean Color Imager (GOCI) data and in situ data show that the algorithm has satisfactory accuracy. The root mean square error (RMSE) and the absolute percentage difference (APD) are 0.0089 and 35.12, respectively. By contrast, the Rrs(λ) values retrieved from the same matchups using the GOCI data processing system (GDPS) have higher RMSE and APD of 0.0104 and 69.15, respectively. The ESOA method can be implemented conveniently within the open source code of SeaDAS (v7.1) as an alternative and operational tool for atmospheric correction of ocean color data, including GOCI, MERIS and MODIS, over highly turbid estuarine and coastal regions, such as the Yangtze estuary, the Hangzhou Bay and most of the coastal ocean in Eastern China.
KW - Atmospheric correction
KW - GOCI
KW - MERIS
KW - Ocean color
KW - Turbid coastal waters
UR - https://www.scopus.com/pages/publications/85010788201
U2 - 10.1016/j.rse.2017.01.013
DO - 10.1016/j.rse.2017.01.013
M3 - 文章
AN - SCOPUS:85010788201
SN - 0034-4257
VL - 191
SP - 197
EP - 214
JO - Remote Sensing of Environment
JF - Remote Sensing of Environment
ER -