Ashok Kumar Roya, Shelu Kumarib*, Ravi Ranjan Kumarc, Ishitac & Abhinav Chauhand
aDepartment of Botany, P. C. Vigyaan Mahavidyalaya, J. P. University, Chapra, Bihar, India
bDepartment of Botany, J. D Women’s College, Patliputra University, Patna, Bihar, India
cDepartment of Botany, T.P.S. College, Patliputra University, Patna, Bihar, India
dUniversity Department of Botany, Patliputra University, Patna, Bihar, India
Received : 25th May, 2024 ; Revised : 25th June, 2024
DOI:-https://doi.org/10.5281/zenodo.15106344
Abstract– Water quality parameters are the most important indicators of water quality in inland water systems. Maintaining systems for monitoring physicochemical parameters is time-consuming and cost-intensive, as developing appropriate river management plans requires in-situ water quality data with high spatial and temporal resolution. In this study, we used 10-m Sentinel-2 pictures to map the spatial changes in the Punpun River’s water quality. We used spectral predictors obtained from the satellite pictures to train one machine learning algorithm, Random Forest (RF), to predict concentrations of pH, DO, chlorophyll, BOD, COD, TSS, and turbidity. In addition, we computed a number of metrics to evaluate the accuracy of the water quality maps and the performance of the models, such as Mean Squared Error (MSE), Coefficient of determination (R2), and Root Mean Squared Error (RMSE). The modelled and measured concentrations of pH, DO, chlorophyll, BOD, COD, TSS, and turbidity exhibited good agreement with minor residual errors ranging between 0.201 mg/L and 0.241 mg/L, according to our results. Additionally, bands 5 (B5, vegetation red edge) and 8 (B8, NIR) were found to be significant predictors of parameter concentrations, and RF was found to be a dependable and efficient algorithm for doing so. The Punpun River had good to exceptional concentrations of pH, DO, chlorophyll, BOD, COD, TSS, and turbidity. The Punpun River’s current state of water quality and the effectiveness of the management measures implemented to control and prevent eutrophic issues have been spatially illuminated by our findings.
Be First to Comment