Assessment of Water Quality Parameters for Some Inland Water Bodies of Western India Using Landsat 8 Data

Rapid assessment of water quality at a regular interval is important to monitor changes in water quality caused by manmade and natural interventions. However, large expanse of water bodies cannot be frequently monitored by sample collection and analysed due to economical and practical reasons. In the present study, water samples were collected during Sept 2017-May 2018 from inland water bodies: Narmada River and Reserviors (Bargi, Ukai, and Ujjani), and analysed for various water quality parameters. Two optically active water quality parameters viz., Chl-a and turbidity were correlated with Landsat 8 remote sensing data (LASRC). Surface reflectance values obtained from Band 1(0.435-0.451 mu m), Band 2 (0.452-0.512 mu m), Band 3 (0.533-0.590 mu m), Band 4 (0.636-0.673 mu m), Band 5 (0.851-0.879 mu m) were used to derive various single and multiple band algorithms to predict these two parameters. Multiple linear regression using B3, B4, and B5 provided good correlation between predicted and actual turbidity data (R-2=0.62) with root mean squared error (RMSE) 27.85%. In case of Chl-a, ratio of B1/B3; and B2/B3 correlated well with actual Chl-a data with R-2=0.62 and RMSE=34.70% and R-2=0.72 and RMSE=23.69% for Chl-a values between 1 and 7 mg/m(3), and 5 and 25 mg/m(3), respectively. The study revealed that remote sensing data can be used to predict fairly accurate values of turbidity and Chl-a temporally in inland water bodies. Values derived using remote sensing data can be confirmed by sporadic collection and analysis of water samples enabling rapid and economical monitoring of vast water bodies.