Open AccessDelineation of water body from Sentinel 2 MSI imagery – A comparative study
Author(s) -
P. Y. Sherjah,
N. Sajikumar
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1114/1/012029
Subject(s) - pixel , turbidity , remote sensing , water body , environmental science , surface water , atmospheric correction , geology , reflectivity , computer science , artificial intelligence , oceanography , physics , environmental engineering , optics
Remote sensing gives a synoptic view of the extent of water area and also assists in understanding the spatial and temporal variations in the water surface coverage. Several methods have been adopted for extracting water pixels from an imagery. Delineation of water bodies has been attempted based on threshold values at a single band, band ratios or normalized difference indices. The accuracy of a method depends on the geophysical and spectral characteristics of the area of study. The present study attempts to compare the performance of widely used water extraction indices NDWI, MNDWI, SNN with the water pixels extracted by atmospheric correction processors C2RCC and Acolite. The Sentinel 2 imageries of the northern region of Vembanad Lake has been used for the study. The comparison on these methods indicated that C2RCC could extract even the finer streams that were not identified by most of the other methods. The presence of turbidity may be a reason for these pixels to be unidentified by the other methods. The temporal variations in the water surface coverage was also attempted here, by comparing the water area in March 2019, October 2019, December 2019.