Trophic Status Mapping of Inland Water Bodies Using Normalized Difference Indices Derived from Sentinel 2 MSI Imagery

The trophic state index (TSI) is generally accepted as an indicator of the trophic status of inland water bodies. Regular monitoring of the trophic status of a water body is vital for understanding the health and functioning of its ecosystem. The band ratios (BR) and normalized difference (ND) indices are customarily used for classifying the land uses, but their capacity for quality monitoring of a water body has not been fully explored. The present study explores the possibility of using ND and BR of the bands of Sentinel 2 multispectral imagery (MSI) (S2) imagery for mapping the TSI of an inland water body. The data of Milford Lake, Kansas, were used to derive TSI. Three atmospherically corrected data sets obtained from Case 2 regional coast color processor (C2RCC), Acolite, and Sentinel 2 Correction (Sen2Cor) were used to derive ND and BR indices. The results of the correlation analyses of these indices with TSI reinforced the importance of atmospheric correction on any algorithm. TSI derived from turbidity measurements was observed to be well correlated to ND and BR indices of green (B3) and red bands (B4, B5, and B6) derived from the C2RCC data set (correlation coefficient > 0.8). Thresholds of BR and ND indices of green and red bands for delineating the hypereutrophic areas from the total area of interest were also identified. The regression relations and threshold criteria were validated with the data from five other lakes in the US and one lake in India to verify the compliance of these relations to other areas. TSI could be estimated with a mean absolute percentage error (MAPE) of 9% from these lakes by using ND34, ND35, and ND45. The threshold criteria for BR34, BR46, and BR56 for delineating hypereutrophic areas were also validated for these lakes. The validations confirm that these computationally simple indices could be utilized for quality monitoring in areas of different geographic and climatic characteristics. (C) 2022 American Society of Civil Engineers