Sewage-Borne Ammonium at a River Bank Filtration Site in Central Delhi, India: Simplified Flow and Reactive Transport Modeling to Support Decision-Making about Water Management Strategies

摘要

In the Indian metropolis of Delhi, the Yamuna River is highly influenced by sewage water, which has led to elevated ammonium (NH 4 + ) concentrations up to 20 mg/L in the river water during 2012–2013. Large drinking water production wells located in the alluvial aquifer draw high shares of bank filtrate. Due to the infiltrating river water, the raw water NH 4 + concentrations in some wells exceed the threshold value of 0.5 mg/L ammonia-N of the Indian drinking water specifications, making the water unfit for human consumption without prior treatment. However, to meet the city’s growing water demand, it might be advantageous to consider the long-term use of the well field. This requires the development of an adapted post-treatment unit in concert with an adjusted well field management. To better understand the groundwater dynamics and contamination and decontamination times at the well field, a theoretical modeling study has been conducted. The results of 2D numerical modeling reveal that the groundwater flux beneath the river is negligible because of the aquifer and river geometry, indicating that infiltrating river water is not diluted by the ambient groundwater. Increasing the water abstraction in the wells closest to the river would result in a larger share of bank filtrate and a decreasing groundwater table decline. Simplified 1D reactive transport models set up for a distance of 500 m (transect from the riverbank to the first production well) showed that the NH 4 + contamination will prevail for the coming decades. Different lithological units of the aquifer (sand and kankar—a sediment containing calcareous nodules) have a strong influence on the respective contamination and decontamination periods, as the retardation of NH 4 + is higher in the kankar than in the sand layer. Although this simplified approach does not allow for a quantification of processes, it can support decision-making about a possible future use of the well field and point to associated research needs.