Critical appraisal of BOD and DO models applied to a highly polluted river in India

Most commonly used biochemical oxygen demand (BOD) and dissolved oxygen (DO) models have been tested for their applicability in the River Kali, which is one of the most polluted rivers in India. A total of 732 field data sets were generated during field survey from March 1999 to February 2000. The modelling of BOD and DO in the River Kali involves derivation and solution of the governing equations that describe concentration change with time and space brought on by advective, decay, settling and loading functions. However, due to continuous discharges (e.g. from wastewater treatment plants) and steady-state flow conditions in the River Kali, the dispersion effects are found to be insignificant. In the analysis, the model parameters used in BOD-DO models were optimized using the Newton-Raphson technique and the performance of different models was evaluated using correlation statistics (r(2)) and error estimation, viz. standard error (SE) and mean multiplicative error (MME). The results indicate that the BOD-DO models developed after Camp (1963) yielded the best agreement with the observed values as compared with several other approaches.