Distributed Computing Model of Multispectral Time Series Data Analysis for Chlorophyll Concentration Determination Using Ocean Color Monitor-2 Data

Abstract

Abstract In general, the ocean color monitor sensor (OCM) is used to identify and monitor the phytoplankton bloom and fishing zones. In the ocean, there is a food chain among plants and fish, and by calculating the normalized difference vegetation index (NDVI), we can identify chlorophyll concentration of the particular ocean area. NDVI is an indication of the presence of chlorophyll concentration. In this study, importance is given to the NDVI of ocean area, because our Earth is mostly occupied by the ocean in the sense of water. OCM sensor data are used to identify the chlorophyll, which in turn indicates the presence of phytoplankton. This is the primary production in the basic food chain, and also for the fishes. Remote sensing methodology is used to find and understand the spatial allocation of oceanwater constituents. Usually, the phytoplankton pigment emits a greenish color in the seawater, which is the visible region in the ocean. It enables plant objects to be identified from among the other suspended matter on the oceanwater. NDVI is mainly used in land applications to identify vegetation and forestation, and it is used to identify the chlorophyll pigment concentration of the ocean surface. Therefore, NDVI can be used to map chlorophyll-determined zones through which possible fishing zones can be generated. This study also shows that the NDVI generation technique is used to discover the resources of seawater for mapping the fishing zones. However, the limitation on large-scale computation for the entire earth surface leaves challenges toward raising the technological solution. Hence, an attempt is made to integrate a distributed computational model to cover the larger spatial data. Therefore, a grid-based satellite image processing system is designed to discover the chlorophyll pigment concentration on the ocean through NDVI generation. With multiple computing nodes of the configured grid, the spatial coverage on the oceanic surface is widened and computational speed is also improved to yield a promising outcome.