Atmospheric correction algorithms for ADEOS/OCTS ocean colar data: Performance comparison based on ship and buoy measurements Original Research Article

The paper aims at evaluating the performance of the atmospheric correction algorithm for the Ocean Color and Temperature Scanner (OCTS) visible band data used at Earth Observation Center (EOC) of National Space Development Agency of Japan (NASDA). The algorithm uses 10 candidate aerosol models including “Asian dust model” introduced in consideration of the unique feature of aerosols over the east Asian waters. Based on the observations at 670 and 865 nm bands where the reflectance of the water body can be discarded, the algorithm selects a pair of aerosol models that accounts best for the observed spectral reflectances to synthesize the aerosol reflectance in other bands. Three different schemes of the aerosol model selection are presented and their anticipated estimation error in terms of the retrieved water reflectance at 443 nm is analyzed. The results of our numerical simulation show that the standard deviation of the estimation error of the “weighted average with iteration” scheme is mostly within permissible level of ±0.002 in-water reflectance at 443 nm, reducing the error by roughly a factor of 2 compared to the other schemes. The paper also evaluates the performance of the algorithm by comparing the satellite estimates of “water-leaving radiance” nLW and chlorophyll-a concentration with selected buoy- and ship-measured data. In comparison with the old “CZCS-type” atmospheric correction, the OCTS algorithm records factor 2–3 less error in estimating nLW.