Application of risk-based assessment and management to riverbank filtration sites in India.

This is the first reported study of a riverbank filtration (RBF) scheme to be assessed following the Australian Guidelines for Managed Aquifer Recharge. A comprehensive staged approach to assess the risks from 12 hazards to human health and the environment has been undertaken. Highest risks from untreated ground and Ganga River water were identified with pathogens, turbidity, iron, manganese, total dissolved solids and total hardness. Recovered water meets the guideline values for inorganic chemicals and salinity but exceeds limits for thermotolerant coliforms frequently. A quantitative microbial risk assessment undertaken on the water recovered from the aquifer indicated that the residual risks of 0.00165 disability-adjusted life years (DALYs) posed by the reference bacteria Escherichia coli O157:H7 were below the national diarrhoeal incidence of 0.027 DALYs and meet the health target in this study of 0.005 DALYs per person per year, which corresponds to the World Health Organization (WHO) regional diarrhoeal incidence in South-East Asia. Monsoon season was a major contributor to the calculated burden of disease and final DALYs were strongly dependent on RBF and disinfection pathogen removal capabilities. Finally, a water safety plan was developed with potential risk management procedures to minimize residual risks related to pathogens.

Interaction of Aquifer and River-Canal Network near Well Field.

The article presents semi-analytical mathematical models to asses (1) enhancements of seepage from a canal and (2) induced flow from a partially penetrating river in an unconfined aquifer consequent to groundwater withdrawal in a well field in the vicinity of the river and canal. The nonlinear exponential relation between seepage from a canal reach and hydraulic head in the aquifer beneath the canal reach is used for quantifying seepage from the canal reach. Hantush's (1967) basic solution for water table rise due to recharge from a rectangular spreading basin in absence of pumping well is used for generating unit pulse response function coefficients for water table rise in the aquifer. Duhamel's convolution theory and method of superposition are applied to obtain water table position due to pumping and recharge from different canal reaches. Hunt's (1999) basic solution for river depletion due to constant pumping from a well in the vicinity of a partially penetrating river is used to generate unit pulse response function coefficients. Applying convolution technique and superposition, treating the recharge from canal reaches as recharge through conceptual injection wells, river depletion consequent to variable pumping and recharge is quantified. The integrated model is applied to a case study in Haridwar (India). The well field consists of 22 pumping wells located in the vicinity of a perennial river and a canal network. The river bank filtrate portion consequent to pumping is quantified.