CONTENTS

Introduction Objective Rationale Filtration Processes Attenuation Processes During RBF Filtration Mechanisms Concept of Bank Filtration Factors affecting RBF Performance Site suitability for bank filtration Removal of DBPs and DBP precursors Technology Benefits Applications Wastewater Applications Case Study RBF sites in India

Introduction
Riverbank filtration (RBF) is a cost-effective, natural pretreatment technology that uses Mother Natures geologyinstead of chemicalsto pretreat surface water and groundwater supplies. As energy costs for conventional treatment technologies escalate and translate into increases in user rates, alternative pretreatment technologies are being used more widely. RBF technology is at the forefront as a method for not only pretreating existing raw surface water supplies but also for developing new and sustainable water supplies. RBF uses the bed of a reservoir, lake or river and an adjacent sand and gravel aquifer as a natural filter. The technology can be applied directly to existing surface water reservoirs, streams, lakes and rivers, and now it is often a guiding factor in the hydrogeologic investigation of new source supplies. Riverbank filtration (RBF) operates by extracting water from wells located near rivers (~20 to 200 m away). If engineered correctly, most of the extracted RBF water originates from the river. As this river water passes through the riverbed sediments, contaminants are removed by overlapping biological, physical, and chemical processes. Although simple in concept, the performance of an RBF system is dependent on local conditions, such as permeability of the riverbank, river level and sediment transport variability, and the type and load of contamination. These RBF site conditions are typically assessed for geographically narrowly defined sites, e.g. individual cities. Our approach to RBF seeks implementation over a much larger area (watershed), by employing modern spatial data referencing methods, i.e. geographical information systems (GIS) and geographic positioning systems (GPS).

Objective
To provide villagers in rural areas with safe, affordable, and reliable drinking water using a low-cost, easily replicated approach to treating polluted surface water resources with riverbank filtration wells.

Rationale
About 21% of communicable diseases in India are water related and polluted water kills over 1,600 people every day. Indias ability to sustain its rapidly expanding economy therefore depends heavily on improving its drinking water quality and infrastructure. Most major water suppliers however are inefficient or do not serve all areas, while private wells often fail prematurely. Hence, local communities need to have affordable alternatives to sustain themselves with clean drinking water at predictable quantities, qualities, and cost.

Filtration Processes
Bank-ltration water purication is based on two processes: rst, biological self-purication and second, geochemical and hydrogeological processes during groundwater passage (Gunkel & Hoffmann 2009). Biological selfpurication is realised by an adapted biocoenosis with high turnover rates, whereas geochemical and hydrogeological processes are characterised by low turnover rates. However, the transit time during bank ltration of about 50 days guarantees water purication by absorption and metabolisation. The zone with increased bioactivity and its structural components within the interstices, the biolm, amounts only to a few metres of the bank ltration zone. This observation is of great interest for water treatment by bank ltration, especially when considering shock loads and physical and chemical disturbances such as temperature increases. The purication process within the pore water is a redox-dependent biodegradation of organic material, and turnover rates are highly dependent on local redox conditions as well as on temperature. Up to now, little has been discovered about these processes during groundwater recharge. An assessment of redox processes and temperature effects for bank ltration is required for a better understanding of the regulatory mechanisms of biological self-purication and physical-chemical redox processes, as well as their effects on pore water quality in natural and induced bank ltration sites. Degradation of organic material leads to consumption of oxygen, and the resultant use of alternative electron acceptors such as nitrate, manganese, iron oxide, hydroxide and sulphate. This leads to the formation of distinct redox zones along the ow direction. Temperature effects on bank ltration are highly signicant and have become the focus of further research activities with special regard to climate change and water temperature all over the world, as well as implementation of bank ltration techniques in some subtropical and tropical countries. Temperature change scenarios in Europe vary regionally but show a clear trend toward warming. For example, regions of the Iberian Peninsula and East Germany are already characterised by low precipitation and little excess of water. The city of Berlin and the surrounding Brandenburg

region is one of the driest regions in Germany, with a negative climate water balance in summer. Therefore, this area is highly sensitive to climate change. In future, the mean discharge of the Havel and Spree Rivers, with extended bank ltration sites, will further decrease, and without this dilution effect, higher sewage and other contaminants will be increasingly concentrated, and will, thus, strongly inuence water quality following bank ltration. Lake water warming was registered at Lake Mu ggel, an enlargement of the Spree River in Berlin, and within 25 years the mean summer surface water temperature had risen about 2.38C, with consequences for oxygen and nutrient levels and phytoplankton response.

Attenuation Processes During RBF

Four attenuation processes are involved in RBF: h y d r o d y n a m i c , m e c h a n i c a l , b i o l o g i c a l , a n d physicochemical. Hydrodynamic processes include convective-dispersive transport, and dilution. The aquifer acts as a filter for the temporal variation of the pollutant compounds in the river caused by accidental (or intentional) spills, which, due to the connection between the river and the aquifer, represent a risk of contamination to the groundwater. As a result, high frequency variations in the surface water quality are reduced in groundwater. Beyond smoothing fluctuations in river water quality, dilution takes place when the river water mixes with groundwater, which is usually of higher quality, further enhancing the quality of bank filtrate. The most important mechanical processes for the improvement of water quality are those involving the natural filtration of fine sediments, particulate organic matter, and pathogens, especially in the first few meters from the river to the well. A disadvantage of physical filtration is associated with the obstruction or the clogging of the porous media, as will be explained later. The biological processes that occur during RBF are directly dependent on the type of microorganisms that inhabit the aquifer. The metabolic processes of these microorganisms mainly determine the final quality of filtered water. Finally, physicochemical processes are associated with sorption, precipitation reactions, flocculation, coagulation, and redox reactions. All these processes govern the removal of particles from the porous media, affecting the concentration and the behavior of metals and other inorganic compounds, thus having implications for the chemical evolution of water.

Filtration Mechanisms
Three filtration mechanisms are possible. Physical filtration or straining takes places when suspended particulates are too large to pass through interstitial spaces between alluvial soil particles. Biological filtration occurs when soil microorganisms remove and digest dissolved or suspended organic material and chemical nutrients. Chemical filtration or ion exchange may take place when aquifer soils react with soluble chemicals in the water. Most 'normal' contaminants (microbial organisms and inorganic or organic pollutants) will be removed by bank filtration, either because they get filtered out by the sand/earth of the bank, or because the passage time (which may be days or potentially weeks) is sufficient to render them

inactive. Research has also shown that the removal efficiency depends not only on the contaminant, but also on the "hydraulic and chemical characteristics of the bottom sediment and the aquifer, the local recharge-discharge conditions, and biochemical processes". There have been indications that some pharmaceutical compounds (medical drug traces from human use) may not always be sufficiently removed by bank filtration, and that in areas with substantial contamination of this type, additional treatment may be needed.

Concept of Bank Filtration

The bank filtration technique takes advantage of existing geologic formations adjacent to water bodies to filter drinking water. Wells are dug in fine, sandy sediments next to water bodies and water is extracted from these wells. Water in the water bodies filters through the sediments removing contaminants. The water obtained is often of much higher quality than the raw surface water. The removal of contaminants in surface water is important because when the water is chlorinated these contaminants can be converted to harmful chemicals referred to as disinfection byproducts (DBPs). Chlorine must be added to most drinking water supplies to kill pathogens. The emphasis on controlling DBPs began when research showed that trihalomethanes (THMs) were produced by the reaction of chlorine with aquatic humic substances during the chlorination of drinking water. THMs include some known and suspected carcinogens (chloroform and dichloroacetic acid). National surveys by USEPA indicated the presence of THMs in virtually every chlorinated water supply tested. Research was initiated in the 1970's, and continues today, into identification of significant DBPs and methods to avoid their formation in the drinking water. Bank filtration wells can either be

horizontal or vertical depending upon the hydrogeologic setting, required production rate, and the utility's preference. Shallow alluvial deposits and a higher rate of pumping from a given location often favor horizontal wells, sometimes called Ranney wells or laterals. The laterals of the collector wells can all be directed towards the river or distributed in all directions. Figure 1 provides a schematic diagram of a bank filtration well.

In Europe, particularly in Germany, bank filtration as a water treatment process has been applied since the 1870's and the first research project on bank filtration started in 1898. In the United States there have been bank filtration wells operating for nearly half a century. Since Sandoz Pharmaceutical's accidental spill of herbicides into the Rhine River in 1986, a rigorous 3-year study involving 6 water works, five research institutes, and several universities was conducted by Prof. Heinrich Sontheimer at the University of Karlsruhe, Germany. The study focused on hydrology, removal of organic compounds that included natural organic matter (NOM) and DBP pre-cursors, removal of particles, and the simulation of ground water passing through test filters. Harmful contaminants often taint drinking water drawn directly from a river, but a low-cost natural filter may lie just beyond the banks. Johns Hopkins researchers have found that the soil alongside a river can remove dangerous microbes and organic material as water flows through it. The cleaner water is then pumped to the surface through wells drilled a short distance from the river. This technique, called riverbank filtration, has been used in Europe for more than 50 years to improve the taste and smell of drinking water and to remove some hazardous pollutants such as industrial solvents. But after studying these natural filtration processes for six years at three rivers in the Midwestern United States, Johns Hopkins researchers have determined that

passing river water through nearby sediment can produce other health benefits and may cut water treatment costs.

In riverbank filtration, water from a river passes through nearby soil before it is drawn up through wells. This process removes some contaminants.

Factors affecting RBF Performance

Clogging of the riverbed is still an important factor causing uncertainness in the planning stage of riverbank filtration plants. Several attempts have been made to develop tools, which are suitable to predict this process. But up till now, these tools are only a slight help for the engineering of riverbank filtration plants. On the other hand there exists a lot of experience about clogging from the operation of riverbank filtration plants. But to utilize this experience for a new plant, the hydrological and morphological aspects of the river and the aquifer have to be analyzed carefully to create a basis for the transfer of available knowledge. This paper deals with the relevant properties of rivers, concerning riverbank filtration: the runoff regime and the runoff dynamics, the river-aquifer interactions, the stream processes erosion, transport and deposition and the progress of the clogging process itself.

Site suitability for bank filtration

Geohydrologic conditions impact the effectiveness of bank filtration. The permeability of the sediment affects seepage velocity and often internal clogging is associated with sediments having low hydraulic conductivity and small vertical gradients. Sediments that have excessively high conductivity will not be efficient in removing contaminants.

Removal of DBPs and DBP precursors

Chlorine must be added to most drinking water supplies to keep them free of pathogens. Chlorination of water that contains certain naturally occurring organic compounds produces certain harmful chemicals referred to as disinfection byproducts (DBPs). The emphasis on controlling DBPs began in earnest when research showed that trihalomethanes (THMs) were produced by the reaction of chlorine with aquatic humic substances during the chlorination of drinking water. National surveys by USEPA indicated the presence of THM in virtually every

chlorinated water supply tested. Considerable research was initiated in the 1970's, and continues today, into identification of significant DBPs and methods to avoid their formation in finished water. Methods for avoiding the formation of DBPs fall into several categories including the removal of the DBP precursors prior to disinfection, removal of DBPs after formation, and use of alternative disinfectants that do not promote DBP formation. Removal of DBPs after their formation has been extensively studied, using techniques such as oxidation, sorption, and aeration. Although many of these techniques are effective, they are not economical. Alternative disinfection method such as ozonation and UV irradiation have problems. Ozone and its decomposition product, hydroxyl radical, react with most organic matter. Aldehydes are known by-products of ozonation, and hydroxyl radical reactions. In addition bromate is formed during the ozonation of bromide-containing waters, through both ozone- and hydroxyl radical-mediated pathways. Bromate formation currently represents a fairly serious barrier to the more widespread adoption of ozone as an alternative to chlorine in the U.S. Chlorine dioxide is not as good a disinfectant as chlorine or ozone, and may produce chlorite. Chlorine therefore remains the most common disinfectant for drinking water supplies, despite the DBP problem. Currently, THMs, halogenated acetic acids (HAAs), acetonitriles, chloral hydrate, and cyanogen chloride receive most attention as DBPs in chlorinated drinking water supplies. As a result of the deficiencies of alternative disinfectants, DBP precursor removal is considered the most promising solution and considerable research is currently being conducted to understand and improve removal by pre-treatment at the plant. Numerous studies have proven bank filtration's ability to remove DBP precursors, microbes, particles, some pesticides and trace organics. In addition to proposing maximum contaminant levels (MCLs) for total THMs and HAAs, the U.S. EPA's proposed Disinfection/Disinfection By-Product Rule, also focuses on DBP precursor removal. To be in compliance with the rule (finalized on December 16, 1998) conventional water treatment facilities treating surface water or ground water under the influence of surface water must achieve a specified percentage removal during pre-treatment, or have a total organic carbon (TOC) residual of <2 mg/L, prior to disinfection. Bank filtration facilities, with their pre-filtered feed water, will have an advantage over conventional treatment plants in meeting these new EPAs requirements.

Technology Benefits
Advantages of RBF include natural pretreatment through bank filtration, reduced chemical usage for pretreatment and resistance to contaminant threats. It is a natural treatment process which is effective in aerobic and anaerobic conditions. It has minimal color, odor, turbidity and algae, features a low profile and is aesthetically pleasing. It improves the microbiological quality of the water and can be easily integrated with existing technology. It requires little maintenance and is easy to operate. The technology decreases construction and operation costs, offering the lowest costs among supply options, and it provides maintenance cost savings (e.g., no leaf debris, which is common to surface water intakes).

Use of RBF results in a reduced need for disinfection, less sludge generation, achievement of treatment removal credits used to meet the Long-Term Enhanced Surface Water Treatment Rule, easy maintenance and consistent water quality and temperatures. It also is not susceptible to invasive plant infestation and has no impact on fisheries. Thus, this technology offers the following advantages:

Natural pretreatment through bank filtration Reduce chemical usage for pretreatment Decrease construction and operation costs - lowest costs among supply options Consistent water quality and temperatures - no spikes Maintenance cost savings (no leaf debris which is common to surface water intakes) Resistance to contaminant threats Minimal color, odor, turbidity, algae and other microorganisms Reduction in disinfection by product precursors Reduced need for disinfection Ease of maintenance Not susceptible to invasive plant infestation Low profile and aesthetically pleasing - no scar to landscape or nearby recreational or environmental interests No impact to fisheries Reduction in sludge generation Achieves treatment removal credits (used to meet Long Term Enhanced Surface Water Treatment Rule)

Applications
Using alternative well technology (e.g., horizontal or angle wells) opens up numerous RBF applications depending on the site characteristics and soil conditions. These include:

Freshwater intakes beneath river and lake beds; Saltwater intakes beneath an ocean floor; Offsetting well head from wetlands, buildings and floodplains; High-capacity single- well pumping; Utilizing aquifers beneath rivers and lakes to provide prefiltration and enhance raw water quality; and Tapping aquifers from a distance where land constraints prevent drill rig access.

Wastewater Applications
Alluvial soils may also be used to purify waste-water being returned to a river after use. Treated wastewater is typically discharged into a percolation pond on the alluvial flood plain rather than flowing directly into the river. Chemical filtration will cease when all reactive sites within the flow pathway through aquifer soil have reacted. Physical filtration tends to

become more effective as trapped particulates block interstitial passages between alluvial soil particles; but may ultimately limit flow rates through the soil. Particulates on the river side of the soil interface may be periodically scoured away by turbulent flow. Wastewater percolation ponds may require maintenance to break up or remove a scum layer (Schmutzdecke) forming at the bottom of the pond. Biological filtration remains effective unless organic loading causes anaerobic conditions within the riverbed alluvium.

Case Study
RBF was used as a pretreatment technology for Manchester Water Works (MWW) in New Hampshire. MWW worked with its consultant, Wright-Pierce, to evaluate the application of RBF technology for its proposed 15-million-gal-per-day (mgd) supplemental drinking water supply at Merrimack River. Hydrogeologic investigation indicated that the technology was a potentially viable pretreatment technology and an alternative to a direct surface water intake. The RBF investigation involved aquifer exploration utilizing seismic reflection profiling, subsurface angle intake modeling, pumping tests, detailed water quality analysis and bank filtration modeling. Onshore and offshore geophysical and test well exploration studies identified a potential high-yield sand and gravel aquifer beneath the Merrimack River. In this study, a 21-degree angled well design was used to explore the possibility of developing a 12-mgd future RBF supply. The low-angle gravel well, 165 ft in length, was drilled beneath the river and pump tested for six days at 620 gal per minute. A detailed groundwater flow model with a rhodamine dye time-of-travel analysis was conducted, and a detailed water quality laboratory analysis was conducted. RBF was shown to be an excellent barrier against microorganisms. A log credit removal of at least 2 for bacteria, viruses and aerobic spore-forming bacteria could be assigned. It removed organic precursors by an average of 63%. Induced infiltration from the river was estimated to range from 53% to 64%. The design of a unique offshore well construction and pumping system allowed the utility to take advantage of the raw water source, positioning it to meet projected supply needs. By utilizing the technology, MWW may eliminate the need for filtration as well as reduce chemical, operational and treatment facility construction costs.

RBF sites in India

At certain sites in northern India, the efficacy of riverbank filtration (RBF) systems was tested. The first RBF site is located near Haridwar at the foothills of the Himalaya. Because at times a great numbers of pilgrims visit this site, the water demand is highly variable. The RBF is surrounded by the river Ganges and two canals originating from the river. Recently, the upstream reaches have witnessed increased sediment load because of deforestation and construction activities related to hydropower plants and residences. In view of these changes,

measurements of water quality in river water and bank filtrate are performed for one year to study the possible effects on the RBF system. A simple model is developed to relate influent (source water) and filtrate concentration. The model indicates that the filtration coefficient is linearly related to the natural logarithm of the influent concentration and other water quality parameters. The second site is located near Muzzafar Nagar, a city 100 km downstream of Haridwar. The river Kali receives a huge input of non-point pollution while flowing through the city and dissolved oxygen levels in this river are very low (many a times at zero level). For a period of six months water quality measurements were conducted in the river and in water abstracted from nearby wells. The data analysis focuses on the removal of different types of impurities including E. Coli and a model of the filtration coefficient as a function of influent concentration is also developed. The results from both RBF sites indicate that there is a lot to be done in refining the existing practices. For example, at the first site, an optimum abstraction of filtrate water and regulated release pattern of source water have to be implemented to achieve better filtrate quality. Further, RBF is a useful option to abstract water from the banks of a very polluted river, such as Kali, as evident from second site investigations. Overall, RBF promises to be a very cost effective approach of abstracting water in developing countries.