An Improved SWIR Atmospheric Correction Model: A Cross-Calibration-Based Model

Jun Chen,Tingwei Cui,Changsong Lin

DOI: https://doi.org/10.1109/tgrs.2013.2278340

IF: 8.2

2013-01-01

IEEE Transactions on Geoscience and Remote Sensing

Abstract:The objectives of this study are to validate the applicability of a shortwave infrared atmospheric correction model (SWIR-based model) in deriving remote sensing reflectance from turbid Case II waters and to improve the model using an improved SWIR model (ISWIR-based model). In the ISWIR-based model, the aerosol reflectance and aerosol type at near-infrared wavelengths are determined from SWIR wavelengths using a 5 x 5 box-based cross-calibration method initialized in Case I waters. The remote sensing reflectance predicted by these models was compared to remote sensing reflectance measured from three different coastal waters. The results indicate that the ISWIR-based model provides a superior performance in atmospheric correction than the SWIR-based model. Using a SWIR-based model decreases MRE values from the SWIR-based model by 14.21%-42.7%. These findings indicate that the cross-calibration method can remove noise errors in MODIS SWIR data and improves the accuracy of remote sensing reflectance retrievals using MODIS SWIR wavelengths.