Hu, CM, Muller-Karger, FE, Andrefouet, S, Carder, KL (2001). Atmospheric correction and cross-calibration of LANDSAT-7/ETM+ imagery over aquatic environments: A multiplatform approach using SeaWiFS/MODIS. REMOTE SENSING OF ENVIRONMENT, 78(2-Jan), 99-107.
Atmospheric correction of Landsat/TM and Landsat-7/Enhanced Thematic Mapper (ETM+) over aquatic environments is generally more demanding than over land because the signal from the water column is small. Because the Sea-viewing Wide Field-of-view Sensor (SeaWiFS; Orbview-II satellite) and the Moderate Resolution Imaging Spectroradiometer (MODIS; Terra satellite) provide highly improved radiometric calibration, sensitivity, and spectral bands specifically designed for estimating aerosol radiance and its spectral quality, we attempted a multiplatform assessment of the path radiance and diffuse transmittance. Using SeaWiFS and ancillary data, we estimated the ETM+ path radiance and found that for a typically clear atmosphere, without knowledge of aerosol type (difficult to estimate with ETM+ data alone), the errors in the estimated aerosol radiance in Bands 1 and 2 can be a few counts (1 count corresponds to 0.0786 and 0.0817 mW cm(-2) mum(-1) sr(-1) for Bands 1 and 2, respectively), comparable to errors in the estimated Rayleigh radiance by ignoring polarization correction. The same method can also be used to cross-calibrate the ETM+ over clear water where the target radiance (water-leaving radiance) is known. For a windless day (28 July 1999, wind < 2 in s(-1)), ETM+ Bands 2 and 3 agreed with the SeaWiFS-predicted values to within 0.5 count (similar to 1.5-3.3% of the total signal), while Band 1 was a few counts higher (similar to 5%) than predictions, possibly due to polarization or calibration effects. For a high-wind day (5 February 2000, wind speed similar to 10 in s(-1)), the agreement is less satisfactory due to uncertainties in estimating the whitecap contribution. However, the overall trend for all the bands remains the same over the period: the difference for Band 1 is consistently larger than for Bands 2 and 3. Based on these results, we propose to use Rayleigh and aerosol data estimated with SeaWiFS and/or MODIS for atmospheric correction of ETM+ over aquatic environments. We also propose to use SeaWiFS/MODIS for monitoring the long-term stability of the Landsat-7/ETM+ calibration as the mission progresses. (C) 2001 Elsevier Science Inc. All rights reserved.