This study aimed at characterizing the groundwater flow pattern in a semi-arid agricultural area in northern India crossed by an intermittent monsoon-controlled watercourse, the Najafgarh drain. More specifically, it focused on studying the impact of groundwater recharge from the riverbed to the regional aquifer using hydrogeochemical and isotopic data. Significant hydrogeochemical zonation was observed between the northern, central and southern sides of the drain, linked to different mineralization processes and mixings. Northward from the drain, groundwater was mainly brackish (4.1–23.4 mS/cm), due to dissolution of evaporites (halite and anhydrite). Southward from the drain, mostly fresh groundwater was found (from 0.5 to 2.3 mS/cm), revealing notable cation exchange processes. In the vicinity of the drain (central area), mineralization was intermediate (0.7–4 mS/cm) and groundwater showed low geochemical evolution, supposing a distinct origin. Stable isotopes of water (d18O, d2H) confirmed that central groundwater was not a simple mixing between northern and southern groundwater masses, but had a significant component of infiltrated surface water from the drain. Potentiometric data supported these findings and confirmed the contribution of the drain to the recharge of the aquifer, setting up a hydraulic barrier between north and south, despite surface water availability limited to the monsoon season and low hydraulic conductivity of the riverbed. This study demonstrates the value of the geochemical and isotopic analysis of groundwater to characterize groundwater flow pattern in peri-urban agricultural areas, especially surface water–groundwater interactions.

Emerging countries frequently afflicted by waterborne diseases require safe and cost-efficient production of drinking water, a task that is becoming more challenging as many rivers carry a high degree of pollution. A study was conducted on the banks of the Yamuna River, Delhi, India, to ascertain if riverbank filtration (RBF) can significantly improve the quality of the highly polluted surface water in terms of virus removal (coliphages, enteric viruses). Human adenoviruses and noroviruses, both present in the Yamuna River in the range of 10(5) genomes/100 mL, were undetectable after 50 m infiltration and approximately 119 days of underground passage. Indigenous somatic coliphages, used as surrogates of human pathogenic viruses, underwent approximately 5 log10 removal after only 3.8 m of RBF. The initial removal after 1 m was 3.3 log10, and the removal between 1 and 2.4 m and between 2.4 and 3.8 m was 0.7 log10 each. RBF is therefore an excellent candidate to improve the water situation in emerging countries with respect to virus removal.

Until around 2004, the term riverbank filtration (RBF) or simply bank filtration (BF, a unified term for river and lake bank / bed filtration) was not commonly used in context to drinking water supply in India. The abundant recharge of traditional dug wells (used for drinking and irrigation) located near surface water bodies (mainly rivers but also some lakes) by very low-turbidity water via natural bank filtration during and after the monsoon has been recognised in India for a very long time. Induced bank filtration has been suggested in the 1970s to address the growing agricultural irrigation demand in the alluvial plains along the Ganga River by inducing recharge from surface water bodies during and after the monsoon (Chaturvedi and Srivastava 1979). Documented evidence till date suggests that induced bank filtration has been used in India for at least 56 years, although even older BF systems may exist. In Nainital, bank filtrate has been abstracted from Nainital Lake since 1956 (Kimothi et al. 2012). BF supplements existing surface and groundwater abstraction for drinking water supply in the cities of Ahmedabad (by the Sabarmati River), Delhi and Mathura (Yamuna) and Nainital (Nainital Lake); on the other hand in Haridwar and Patna (Ganga), and Medinipur and Kharagpur (Kangsabati), BF is used as an alternative to surface water abstraction and to supplement groundwater abstraction (Sandhu et al. 2012). Considering the continuously growing demand for drinking water in sufficient quantities, the emphasis at many BF sites has traditionally been on maximising the volumes of raw water abstracted. Furthermore, the results of a fact-finding study (Ray and Ojha 2005) on the use of BF for drinking water production in India on one hand confirmed that a number of river-side communities have been already using BF for a long time, but that on the other hand only scarce information on the hydrogeological conditions and water quality of these BF sites existed. Holistic investigations on water quality aspects and sustainability (qualitative and quantitative) of these existing BF sites began only after 2004. Water quality investigations conducted at the BF sites of Srinagar by the Alaknanda river (Ronghang et al. 2011), Haridwar and Nainital (Dash et al. 2008, 2010; Sandhu et al. 2011a), Delhi (Sprenger et al. 2008; Lorenzen et al. 2010) and Mathura (Singh et al. 2010; Kumar et al. 2012) and Patna (Sandhu et al. 2011b) showed that the main advantage of using BF in comparison to direct surface water abstraction lies in the removal of pathogens and turbidity. The surface water concentration of trace organic contaminants and their removal at the investigated sites has not been widely investigated, but has shown to be high at sites in Delhi and Mathura (Sprenger et al. 2008; Singh et al. 2010). For conventional treatment, high concentrations of organic contaminants requires high (40–60 mg/L) doses of chlorine prior to flocculation thus creating a greater risk for formation of carcinogenic disinfection by-products, as reported in Mathura (Singh et al. 2010; Kumar et al. 2012). In such situations BF is advantageous as a pre-treatment in order to reduce the necessary doses of chlorine prior to flocculation. Additional advantages of BF may also be seen during the monsoon season principally in the removal of turbidity and pathogens, as well as in the removal of color and dissolved organic carbon (DOC), UV absorbance, turbidity, total and thermotolerant coliform counts, endocrine disruptor compounds and organochlorine pesticides (Dash et al. 2008, 2010; Sandhu et al. 2011a; Thakur et al. 2009a, 2009b; Sprenger et al. 2011; Mutiyar et al. 2011). BF, however, does not present an absolute barrier to other substances of concern (e.g. ammonium) and some inorganic trace elements may even be mobilized. This has been observed in Delhi which has poor surface water quality (Sprenger et al. 2008), at which extensive post-treatment is applied to remove high levels of ammonium. The objective of this deliverable is to provide an overview of known BF schemes in urban areas of India where the abstraction of bank filtrate is intentional. The main water quality issues of concern are highlighted. Related published and unpublished data, as well as new data collected since the commencement of the Saph Pani project in October 2011, is presented for the BF schemes in Haridwar, Nainital, Srinagar (by the Alaknanda river in Uttarakhand), Delhi Mathura and Satpuli (by the Eastern Nayar river in Uttarakhand).

Chennai is the largest city in South India located in the eastern coastal plains. Water supply to the Chennai city is met by reservoirs and by groundwater. Most of the groundwater is pumped to the city from the well fields located in the Araniyar and Korttalaiyar River (A-K River) catchment north of Chennai.

Chennai is the largest city in South India located in the eastern coastal plains. Water supply to the Chennai city is met by reservoirs and by groundwater. Most of the groundwater is pumped to the city from the well fields located in the Araniyar and Korttalaiyar River (A-K River) catchment north of Chennai.

Bank filtration (BF) is a well established and proven natural water treatment technology, where surface water is infiltrated to an aquifer through river or lake banks. Improvement of water quality is achieved by a series of chemical, biological and physical processes during subsurface passage. This paper aims at identifying climate sensitive factors affecting bank filtration performance and assesses their relevance based on hypothetical 'drought' and 'flood' climate scenarios. The climate sensitive factors influencing water quantity and quality also have influence on substance removal parameters such as redox conditions and travel time. Droughts are found to promote anaerobic conditions during bank filtration passage, while flood events can drastically shorten travel time and cause breakthrough of pathogens, metals, suspended solids, DOC and organic micropollutants. The study revealed that only BF systems comprising an oxic to anoxic redox sequence ensure maximum removal efficiency. The storage capacity of the banks and availability of two source waters renders BF for drinking water supply less vulnerable than surface water or groundwater abstraction alone. Overall, BF is vulnerable to climate change although anthropogenic impacts are at least as important.

Waterborne diseases are frequent and recurrent episodes in developing countries with deficient sanitary conditions affecting drinking water. Waterborne epidemics might affect thousands of persons, like the Hepatitis-E-epidemics of Kuntra (Naik et al. 1992) and Delhi (Ramalingaswami and Purcell, 1988) with 79,000 and 25,000 ill persons respectively. It is not by chance that both epidemics occurred after drinking water treatment suffered a failure, allowing contaminated drinking water to reach the consumers. In order to ameliorate the consequences of water scarcity and poor sanitary conditions, systems for obtaining drinking water are needed which are efficient, robust, and require only low-cost technology. River Bank Filtration (RBF) is a process during which surface water is induced to infiltrate into the subsurface, either due to a natural hydraulic gradient or the depression cone of an abstraction well. During infiltration and soil passage, the quality of the surface water is substantially improved thanks to a combination of physical, chemical, and biological processes such as filtration, dilution with genuine groundwater, sorption and biodegradation of pollutants Apart from pathogens, organic trace compounds are widespread pollutants in rivers and lakes. The capacity of RBF to effectively or even completely remove both, pathogens and many organic contaminants has been confirmed in numerous investigations (e.g., Matthess et al. 1988). Its low costs in technology and labor makes RBF a very suitable drinking water treatment tool for developing countries. The aim of this study was to ascertain if RBF kept its power to remove pollutants and pathogens even in highly polluted waters as they are encountered in many urban agglomerations in developing countries.