The Japanese Meteorological Agency (JMA) geostational meteorological satellite, Himawari-8, carrying Advanced Himawari Imager (AHI) has been in regular operation since July 7, 2015. Before the AHI, geostational meteorological sensors hardly observed ocean color due to only one broad band in visible wavelengths and relatively large radiometric noise. However, AHI has six spectral bands from visible to shortwave infrared wavelengths (470 nm, 510 nm, 640 nm, 856 nm, 1610 nm, and 2257 nm) in addition to ten thermal infrared bands, and their radiometric noise can be reduced by temporal average since AHI observes the full disk by every 10 minutes. This study investigates the possibility of ocean color estimation (remote sensing reflectance, Rrs, and chlorophyll-a concentration, Chla) from AHI observations. The estimation sensitivity was tested using in-situ data, a simple in-water optical model, and the spectral response of AHI bands. We found the error characteristics and limitation of the estimation by AHI as follows. Chla can be estimated by the traditional scheme, blue (470 nm) green (510 nm) ratio in Chla <3 mg/m3. The estimation in Chla > 5 mg/m3 can be improved by using the green (510 nm) red (640 nm) bands. The random noise can be reduced, by averaging original 10-minute images over an hour. Good estimates are obtained in the summer hemisphere; however, retrieved imagery becomes noisy in the mid and high latitudes (e.g., > 35°) of winter hemisphere due to a long path of the solar light.