A preliminary analysis of diffuse attenuation coefficients for downward irradiance in the KORUS-OC dataset and GOCI data

Abstract

Korea-US joint field campaign for ocean color research (KORUS-OC) was conducted around Korean seas May 20th through June 6th, 2016 concurrent with NASA. It has purposes that enhance the accuracy and performance of Geostationary Ocean Color Imager (GOCI) ocean environmental products and collect match-up in-situ measurements for GOCI. The diffuse attenuation coefficient for downward irradiance (Kd) was measured using the profiler-II, which also provides optical measurements of downward irradiances and upwelling radiances and thus remote-sensing reflectance (Rrs). The Kd is an important property for estimation of clarity of ocean, and it is an input parameter of visibility algorithm. The models for estimating Kd from ocean color image can be divided into an empirical (Austin and Petzold, 1981 Mueller and Trees, 1997 Mueller, 2000 Morel, 1988 Morel and Maritorena, 2001) and semianalytic algorithm (Sathyendranath and Platt, 1997 Lee et al., 2005). The Kd measurements from KORUS-OC campaign are analyzed to evaluate and improve GOCI-derived product. Especially this study focus on a highly turbid region (i.e. Total suspended matter concentration is over than 100 g/m3), where many existing algorithms tend to underestimate Kd. In this study, we compare three different Kd values of Korean seas obtained by both in situ Rrs derived and profiler-II measurements (truth), Kd derived from in situ Rrs and Kd derived from GOCI RImager (GOCI) ocean environmental products and collect match-up in-situ measurements for GOCI. The diffuse attenuation coefficient for downward irradiance (Kd) was measured using the profiler-II, which also provides optical measurements of downward irradiances and upwelling radiances and thus remote-sensing reflectance (Rrs). The Kd is an important property for estimation of clarity of ocean, and it is an input parameter of visibility algorithm. The models for estimating Kd from ocean color image can be divided into an empirical (Austin and Petzold, 1981 Mueller and Trees, 1997 Mueller, 2000 Morel, 1988 Morel and Maritorena, 2001) and semianalytic algorithm (Sathyendranath and Platt, 1997 Lee et al., 2005). The Kd measurements from KORUS-OC campaign are analyzed to evaluate and improve GOCI-derived product. Especially this study focus on a highly turbid region (i.e. Total suspended matter concentration is over than 100 g/m3), where many existing algorithms tend to underestimate Kd. In this study, we compare three different Kd values of Korean seas obtained by both in situ Rrs derived and profiler-II measurements (truth), Kd derived from in situ Rrs and Kd derived from GOCI R