Journal of Hydrologic Engineering

The Dresden-Hosterwitz water utility works operates five open managed aquifer recharge (MAR) basins supplied with pretreated Elbe River water and is located on the floodplain of the Elbe River. The quaternary aquifer consists of sand and gravel, and has a thickness of 9–14 m. The infiltration of untreated river floodwater and its effects on basin clogging and pathogen removal were investigated within a risk-based framework. Column experiments were performed to simulate infiltration in the basin during a flood event. Three vertical profiles were evaluated, which ranged from coarse filter sand to sandy gravel, and installed in sequence in an experimental flow-through column setup. Turbidity, total coliform, and E. coli were measured in the inflow and outflow at infiltration rates from $0.8–3.6\mathrm{m}/\mathrm{day}$. Experimental data was supplemented with clogging and pathogen measurements at the basin during a flood event, and infiltration of untreated river floodwater. The hydraulic conductivity of the filter sand decreased exponentially due to external clogging by two orders of magnitude from 126 to $4\mathrm{m}/\mathrm{day}$. It was $0.4\mathrm{m}/\mathrm{day}$ in field measurements during the flood event due to a higher particle load in the flood water. Total coliforms and E. coli numbers were reduced by $>4$ and $>3{\log }_{10}$ units after 2.5–4.4 m total travel distance. E. coli removal was greater than total coliform removal but E. coli counts in the river water were one order of magnitude lower than total coliform count. Mean turbidity was reduced from 14–0.9 nephelometric turbidity units after passing only 0.6–0.9 m filter sand. In field measurements, no breakthrough of turbidity and E.coli was observed during flooding and infiltration of untreated river water into the aquifer. In conclusion, a flood event of the infiltration basin and the subsequent infiltration of untreated river water into the aquifer poses no serious threat to water quality in terms of E. coli, total coliform, and turbidity at the abstraction wells at 70-m flow distance (4–5 days travel time), provided that the well heads are made flood-proof.

There are 26 managed aquifer recharge (MAR) plants in Finland. The main objective of MAR in Finland is the removal of natural organic matter from surface waters. A typical Finnish MAR plant consists of infiltration of lake water in an esker or other glaciofluvial sand and gravel formation and withdrawal of water from wells downstream. Basin infiltration is used most often. Some MAR plants use sprinkling infiltration and well infiltration. Sprinkling and well infiltration can be attractive for areas not suitable for the construction of basins, e.g., eskers with slopes and forest areas having recreational values with restrictions of tree cutting. Tavase Ltd. aims to construct a MAR plant (capacity $\mathrm{70,000}{\mathrm{m}}^3/\mathrm{d}$) in the neighborhood of the city of Tampere, Finland to provide potable water. Extensive on-site tests were conducted. The objective of this paper was to study the applicability of sprinkling infiltration and well infiltration in an infiltration site planned for potable water production of $\mathrm{20,000}{\mathrm{m}}^3/\mathrm{d}$. Groundwater was pumped from two wells 1.6 km upstream to infiltration using a temporary, aboveground pipeline. Infiltrated water flowed underground back to the wells for recirculation. The tests lasted for 308 days and the maximum infiltration rate was $\mathrm{7,000}{\mathrm{m}}^3/\mathrm{d}$. Two separate sprinkling rakes were built with altogether more than 100 valves to ensure an even distribution of water. Three infiltration wells (diameter 400 mm, depths 35–50 m) were constructed. Sprinkling infiltration worked well with all hydraulic loadings [$0.03–0.17{\mathrm{m}}^3/({\mathrm{m}}^2\mathrm{h})$]. The design loading [$0.07{\mathrm{m}}^3/({\mathrm{m}}^2\mathrm{h})$] could easily be doubled. Well infiltration worked well in all wells. Infiltration rate of $\mathrm{7,000}{\mathrm{m}}^3/\mathrm{d}$ could be achieved in each well. Both sprinkling and well infiltration were found suitable for MAR.

Aquifer storage and recovery (ASR) of freshwater surpluses can reduce freshwater shortages in coastal areas during periods of prolonged droughts. However, ASR is troublesome in saline coastal aquifers as buoyancy effects generally cause a significant loss of injected freshwater. The use of a pair of parallel, superimposed horizontal wells is proposed to combine shallow ASR with deep interception of underlying saltwater. A shallow, fresh groundwater lens can thereby be enlarged and protected. This freshmaker setup was successfully placed in a coastal aquifer in The Netherlands using horizontal directional drilling to install 70-m-long horizontal directional drilled wells (HDDWs). The freshmaker prototype aims to inject a specific volume of freshwater and abstract the same volume of water (consisting of injected water and ambient native groundwater) within the targeted water quality. Groundwater transport modeling preceding ASR operation demonstrates that this set up is able to abstract a water volume of $\mathrm{4,200}{\mathrm{m}}^3$ equal to the injected freshwater volume without exceeding strict salinity limits, which would be unattainable with conventional ASR. This is the first study to demonstrate the potential benefits of HDDWs for a field ASR application. The model outcomes indicate that the feasibility perspectives of ASR in coastal aquifers worldwide require revision thanks to recent developments in hydrologic engineering.

Arid regions frequently have competing demands for scarce water mainly for urban supply and irrigation. This paper presents a decision support system (DSS) as an instrument to find the optimal scheme for management of water resources from multiple supplies using managed aquifer recharge (MAR). The DSS capabilities are demonstrated through its application to the Sonora River basin. The watershed is located in Northern Mexico and has a wide variety of agricultural zones and many small towns along the river. The most important urban center is the city of Hermosillo with a concentrated water demand for urban and industrial uses. The DSS includes the hydrological modeling of the entire system, stream flow in rivers, the groundwater-flow model including MAR, and the characterization of the water quality from various sources. The result from this study establishes that it will be difficult to satisfy the increasing demand for water using the current supply-and-demand-allocation schemes. Alternative sources are thus needed to mitigate the future difference between supply and demand. The MAR with treated wastewater represents an option that could sustain development of urban water supply and irrigation within this basin. It is concluded that the best strategy (least groundwater-storage depletion) is to restrict irrigation and allocate all the effluent from the wastewater treatment plant (WWTP) to recharge the aquifer through infiltration lagoons. However, if there is not a limitation to irrigation, urban demand increases at 1.7% per year and without artificial recharge, the aquifers will be depleted in 25 years. Alternatively, if the WWTP effluent is shared between agriculture and MAR, it is possible to allow irrigation of 2,800 ha with small groundwater depletion. The DDS also allows for studying the potential improvements in water quality and resilience to global climate change.

Setting of critical control and monitoring point trigger limits for water quality in managed aquifer recharge (MAR) schemes ensures human health, environmental, and operational risks are controlled. Application of a transferrable methodology for selection of appropriate critical limits and setting of alert limits for MAR systems is demonstrated for a storm water aquifer storage and recovery scheme used for nonpotable applications in Parafield, South Australia. This is the first demonstrated application of an approach incorporating hazard analysis and critical control point principles for setting water quality targets in a storm water MAR system consistent with Australian Water Recycling Guidelines and Australian and international drinking water guidelines. Salinity and Escherichia coli (E. coli) are used as example parameters but others follow the same principles. A decision tree was developed for systematic selection of relevant targets and setting of critical and alert limits. Up to 10 years of operational water quality data across the system were used for the analyses. Comparison of injected and recovered water using a stochastic model showed the aquifer could be expected to provide a mean E. coli removal of approximately $2.0\log$. A mean E. coli reduction of 39% in the distribution system following a change in operational procedures was predicted using the probabilistic model. Long-term data were used to set alert limits for critical control and quality monitoring points in the system. Management actions following occasions where limits were not met are discussed.

According to the experiments performed by the Tragsa Group and in the framework of the research and development Depth Investigation of New Areas for Managed Aquifer Recharge (DINA-MAR) Project, which relates to the management of aquifer recharge in the context of sustainable development, researchers studied clogging in managed aquifer recharge (MAR) facilities in two demonstration sites within the Guadiana Basin in central Spain. The author focused on the detection and distribution of physical (including air), chemical, and biological clogging processes in 25 wells and bores. Several of these sites have been in operation since 1988. The most remarkable detected processes were algae colonies and the combination of carbonate and iron. Synergistic combinations among the different elements have been studied by means of clogging and water sampling, field and laboratory analysis, field photographs, and down-hole videos. These studies were performed at two experimental sites in a permanent development in which new designs and their infiltration capacity and evolution are being investigated. Several outcomes relating to the infiltration capacity of these wells are presented in this paper. Results from both sites were used to implement a practical methodology that improves current well designs and utilizes two devices to clean and maintain injection wells and bores. This paper also outlines several practical recommendations and protocols for the management of MAR wells. These recommendations establish parameters for MAR wells considering various processes, such as single groundwater extraction, pretreatment filters, backwashing frequency, and cleaning methods.

Reclaimed water has been widely utilized because of water shortages in Beijing, China. Approximately $18\mathrm{million}{\mathrm{m}}^3/\mathrm{year}$ of reclaimed water has been used for scenic water in the dry Chaobai River, and over $12\mathrm{million}{\mathrm{m}}^3/\mathrm{year}$ of such water has percolated from the river bed to recharge the underlying aquifers. Twenty-seven monitoring wells were built into 30, 50, and 80-m depths to quantify changes of groundwater level and quality in the multiaquifer system beneath the river and adjoining areas. The monitored data were utilized to develop a groundwater flow model and a numerical solute transport model to project future changes in the multiaquifer system. It was deduced from the monitored data that the hydrodynamic field in the 30-m depth aquifer was affected by reclaimed water recharge, whereas the 50 and 80-m depth aquifers were relatively slightly affected. The 80-m depth aquifer was the mainly exploited layer in the study area. Chloride ion was utilized to track the constituent movement of reclaimed water. The monitored data showed that groundwater quality in the 30 and 50-m depth aquifers was substantially influenced. Model prediction showed that groundwater level of 80-m depth aquifers over the next 20 years would decline by 22 m for water source A and B, and 12 m for water source C. At the same time, chloride concentration in three different aquifers would be substantially influenced by reclaimed water recharge, and the impacted area for these three aquifers was 22.2, 21.9, and $15.0{\mathrm{km}}^2$, respectively.

This paper argues that sequential in-channel reservoirs in the Syrdarya River Basin of Central Asia are not sufficient to meet the needs of both upstream and downstream water users. Some of the consequences of the use of in-channel reservoirs include fresh water discharges into the saline depression and flooding of irrigated land in winter and water shortages in summer, which causes low water productivity. This study examines the feasibility of two alternative management strategies (1) building additional off-channel reservoirs, and (2) managed aquifer recharge (MAR). First, the potential for MAR in the upstream Fergana Valley is estimated through a regional assessment of MAR potential, field-scale pilot experiments, and simulation modeling. Second, the potential for MAR is analyzed for lift canal irrigation areas, which cover 70% of the irrigated highlands and the foothills in the river midstream, with a focus on energy use intensity and water and energy productivity. Third, the potential for building off-channel reservoirs and MAR is compared in the downstream portion of the river. Specifically for the Makhtaral Irrigation Project (MIP), a scenario of shifting from lift canal irrigation and groundwater extraction for drainage purposes is compared with groundwater irrigation and MAR. The MIP, located in the tail-end of the transboundary Dustyk Canal, faces waterlogging and salinity issues because of the shallow water table. The study concludes that transboundary cooperation in transboundary water transfers for MAR may increase the amount of water that could be stored underground. Overall, it is shown that there is a high potential for MAR in the Syrdarya River Basin with benefits at both the state and farm levels.

In Jinan, China, which is named the Spring City of China, karst groundwater is an important water source for urban supply and maintaining spring flow. However, rapid urbanization in Jinan, China, has caused overexploitation of groundwater and an increase in surface runoff. Consequently, roof runoff harvesting and recycling via the aquifer is being considered to augment the recharge for this area. A pilot-managed aquifer recharge scheme with urban roof runoff was investigated as part of this case study to supplement groundwater recharge. This case study is intended as a pilot for the wider application of managed aquifer recharge at the University of Jinan. The pilot-scale system is composed of the roof catchment, first-flush removal device, regulating tank, filter tank, and recharge well; dimensions and operating parameters of each component were designed by laboratory experiment and hydraulic calculations. The quality of roof runoff is better than road runoff; however, it is still contains a number of hazards. The injected water quality must meet the quality standard for groundwater for the protection of potable groundwater quality. Barriers, or measures to protect water quality, within the treatment train include first-flush removal (first 6 mm rainfall) followed by sedimentation and filtration in a zeolite filter. Water quality monitoring of each barrier and geochemical simulation were used to evaluate the potential influences on groundwater recharge and quality. The rainfall, recharge quantities, and groundwater levels as well as water quality through the recharge process were monitored continuously in June to November 2012. All indexes of treated roof runoff were below the quality standard for groundwater, with the exception of turbidity. Geochemical modeling indicates there is some calcite dissolution after injection but this does not pose a risk to the operational lifetime of the recharge well. However, pretreatment of the source water still needs further improvement to decrease the turbidity to acceptable levels. In addition, water quality risk management is also required for sustainable long-term operation of the scheme.

California guidelines for indirect potable recycled wastewater reuse projects currently require groundwater tracers to demonstrate subsurface residence time for pathogenic microorganism control. Residence times over 6 months from infiltration to drinking water extraction are required. Two prospective tracers were evaluated in this case study: boron-10 (as ${}^{10}\mathrm{B}$-enriched boric acid) and heat (with recharging water $\sim 10°\mathrm{C}$ warmer than native groundwater). Bromide (${\mathrm{Br}}^{-}$) was also released as a control. ${}^{10}\mathrm{B}$ is attractive as a deliberate tracer because (1) reasonably accurate and affordable measurements can be made on an inductively coupled plasma mass spectrometer (ICP-MS) system, and (2) isotopic tracers require significantly less mass to tag an equivalent water volume than concentration-based salt tracers like ${\mathrm{Br}}^{-}$. ${}^{10}\mathrm{B}$ and ${\mathrm{Br}}^{-}$ tracer breakthroughs were observed at seven of nine monitoring wells, although at one well the detection of ${}^{10}\mathrm{B}$ was barely observable and may have resulted from a slight change in source water composition. ${}^{10}\mathrm{B}$ arrived 25% later than ${\mathrm{Br}}^{-}$ on average, showing retardation through exchange with clay surfaces. Heat flow, requiring no artificial input, was interpreted from temperature changes recorded hourly at well loggers. Residence times to all wells were successfully determined from temperature changes with the longest flow path of 6 months. This implies that dilution of ${\mathrm{Br}}^{-}$ and ${}^{10}\mathrm{B}$ is a limitation the geochemical tracer experiment.

Fresh water resources are scarce in rural communities in the southern deltaic plains of Bangladesh where both shallow and deep groundwater is frequently brackish, and fresh water ponds have been increasingly salinized by inundation during storm surges and brackish-water aquaculture. Low-cost aquifer storage and recovery (ASR) schemes were constructed at 13 villages in three coastal districts by developing storage in shallow confined fine to medium sand aquifers overlain by variable thicknesses of silt and clay. A typical ASR scheme consisted of a double-chambered graded sand filtration tank with a volume of $19.5{\mathrm{m}}^3$ that feeds filtered pond water to four to six large diameter ($d=30.5$ or 56 cm) infiltration wells through PVC pipes fitted with stop valves and flow meters. The infiltration wells were completed at 18–31 m below ground and filled with well-sorted gravel capped with a thin layer of fine sand that acts as a second stage filter. Infiltration rates at 13 sites averaged $3{\mathrm{m}}^3/\mathrm{day}$ (range: $3–6{\mathrm{m}}^3/\mathrm{day}$) over one year of operation. At 11 sites where water was abstracted, the recovery rate ranged from 5 to 40%. The source pond source water frequently had turbidity values of $\ge 100\mathrm{NTU}$. After sand filtration, the turbidity is typically 5 NTU. Despite this, clogging management involving frequent (monthly to weekly) manual washing to remove fine materials deposited in the sand filtration tank and the infiltration wells is found to be necessary and effective, with post-manual-washing operational infiltration rates restored to annual average values. E. coli counts in recovered water are greatly reduced compared to raw pond water, although E. coli is still detected in about half of the samples. Arsenic in recovered water was detected to be at level of $>100\mathrm{\mu }\mathrm{g}/\mathrm{L}$ repeatedly at three sites, suggesting that As risks must be carefully managed and require further investigation.