Support Vector Regression (SVR) to Investigate Diurnal Fluctuation in Chromium Concentration in River Amba

In Pali, Maharashtra, petrochemical factories have caused chromium contamination in River Amba, which can cause ulcers, liver damage, and cancer to locals. This study uses Support Vector Regression (SVR) to investigate the non-linear diurnal variation in chromium concentration in River Amba in a 24-hour cycle and its relation to river temperature and river speed. The chromium contamination in River Amba was below the maximum limit of 50µg/L only from 9 p.m. to 6 a.m. Higher chromium concentrations occurred at daylight hours, peaking at 59.5µg/L at 10 a.m., and lower chromium concentrations at night, the minimum being 45.8µg/L at 5 a.m. Similarly, the river temperature was highest (27.6°C) at 10 a.m. and the lowest (20.6°C) at 4 a.m. The diurnal fluctuation in chromium concentration was correlated to the diurnal fluctuation in the river speed, as a positive moderate linear relationship between chromium concentration and river speed with an r2 value of 0.522 emerged. The diurnal variation in river speed, in turn, is likely caused by the fluctuations in the viscosity of heavy oil in River Amba with river temperature, which is influenced diurnally by the River's and the Sun's thermal dynamics.