Spectral Indexing of MODIS-Image Pixels to Reveal the Variability
in the Phytopigment Composition in the Sea under the Influence
of Mesoscale Water Dynamics
G. S. Karabashev*
Shirshov Institute of Oceanology, Russian Academy of Science, Moscow, Russia
Correspondence to: *e-mail: genkar@mail.ru
Received 16 September, 2020
Abstract—A method is proposed for visualizing the inconstancy of the phytopigment composition in the water layer of the formation of backscattered solar radiation recorded by the MODIS multispectral ocean color scanner. The method relies on the fact that each pixel of the MODIS raster image of the investigated water area is assigned a wavelength reflectance minimum (WRM) index. The latter is equal to the sum of the wavelengths of minimums in the spectrum of a pixel or index 100 if they are absent. Such minimums are possible only in the presence of phytoplankton as the only carrier of plant pigments in the open sea. Pixel attributes of a MODIS raster image of the surface of water, like pixel coordinates and standard medium definitions, serve as WRM estimates. This makes it possible to visualize the variability of manifestations of the spectral diversity of phytoplankton against the background of the variability of the given medium properties. The effectiveness of the method is tested using the example of MODIS images of a spiral-eddy structure in the Caspian Sea and mesoscale variability of the bio-optical characteristics of waters over the continental slope in the Southwest Atlantic. The spectral indexing of the pixels of MODIS images of the water surface makes it possible to reveal some forms of variability of phytopigments of the aquatic environment that are inaccessible to the standard methods of ocean-color data processing.
Keywords: MODIS, spectral indexing, Caspian Sea, Southwest Atlantic, mesoscale eddy, upwelling
DOI: 10.1134/S0001437021060242