Bloom of the dinoflagellate, Noctiluca scintillans, appears in the form of a green tide in the Northern Arabian Sea covers a large area of the basin between west coast of India and Oman. It is known to occur during winter-spring and is reasonably persistent on time scale (January- March). Moreover, it is found that the bloom is not mono-species one and more exactly is concurrent with diatom. These blooms have been sampled at various stages by the Indian research vessels from 2003 onwards. Noctiluca scintillans can be identified by SeaWiFS or MODIS chlorophyll images due to high chlorophyll pattern associated with it. However, generalized algorithm for chlorophyll retrieval breaks down in presence of the algal bloom and provides unrealistic estimates of chlorophyll. Besides this, the chlorophyll images do not reveal phytoplankton species/group level information. To enable this, an approach was developed and validated for detection of the bloom forming algae Noctiluca scintillans and its discrimination from diatom in a mixed species environment using ocean colour data of MODIS-AQUA. Spectral shapes of the reflectance spectra were used for the species identification. This paper highlights detection of onset (evolution) of the bloom and study of its spatial and temporal variations at species/group level using remote sensing. Capability of optical detection of the species from space has been used in generating a time-series images for this purpose. Green Noctiluca was found associated with extremely high levels of phytoplankton, which influenced its abundance. In addition, observations from the time-series also pointed towards feeding Available online at: www.mbai.org.in" target="_BLANK" title="External site that opens in a new window "/a> doi: 10.6024/jmbai.2016.58.2.1920-02 behaviour of Noctiluca scintillans and its coupling with diatom. The two were found to be out of phase and revealed a time lag, Noctiluca scintillans being successor of diatom.