The catalytic effect of aqueous Fe(II) (Fe2+aq) on the transformation of Fe(oxyhydr)oxides has been extensively studied in the laboratory. It involves the transfer of electrons between Fe2+aq and Fe-(oxyhydr)oxides, rapid atomic exchange of Fe between the two states, and recrystallization of the Fe-oxides into more stable Fe-(oxyhydr)oxides. The potential occurrence of these reactions in natural soils and sediments can have an important impact on biogeochemical cycling of iron, carbon, and phosphorus. We investigated the possible isotopic exchange between Fe2+aq and sedimentary Fe(III) in Fe-Si-C-rich lake sediments. 57Fe Mössbauer spectroscopy was used to evaluate Fe mineral speciation in unaltered lake sediments. Unaltered and oxidized sediment laboratory incubations were coupled with a classical kinetic approach that allows a quantitative description of the reactivity of assemblages of Fe-(oxyhydr)oxides found in sediments. Specifically, unaltered and oxidized sediment samples were separately incubated with an 55Fe2+aq-enriched solution and exchange was observed between 55Fe2+aq and sedimentary Fe(III), highest in the top of the sediment and decreasing with depth with the 55Fe2+aq tracer distributed within the bulk of the sedimentary Fe(III) phase. Our results indicate that atomic exchange between Fe2+aq and sedimentary Fe(III) occurs in natural sediments with electrons transferred from the Fe(III)-particle to Fe(III)-particle via Fe2+aq intermediates.

The relation between arsenic groundwater concentrations and hydrogeological processes was investigated in the proximal part of the Red River delta, Vietnam, west of the depression cone formed by the exploitation of groundwater in Hanoi. Flow paths in the Quaternary aquifers were modeled based on previously interpreted geological structure and hydrogeological data gathered during field work in 2014 & ndash;2017. Sedimentary structures and simulated flow patterns were compared with the spatial distribution of the groundwater arsenic concentration. The regression of the sea in the area started 4 ka BP in the Holocene. The low tectonic subsidence rate of the Red River delta led to intensive erosion and replacement of fine grained sediments of the sea level high stand by sandy channel belts, resulting in hydraulic connections between the Pleistocene and Holocene aquifers. The Pleistocene aquifer is recharged by both regional flow paths and naturally occurring vertical recharge through Holocene sand and clay layers. Young groundwater (< 40 a) in the shallow Holocene aquifer generally discharges to surface water bodies. The shallow flow system is also seasonally recharged with surface water, as indicated by delta O-18 enrichment of groundwater and oscillating groundwater ages in wells in the vicinity of water channels. The deeper flow system discharges into the Red River and Day River or flows parallel to the rivers, toward the sea. The overall pattern of arsenic groundwater concentrations (decreasing with increasing sediment age) is modified by groundwater flow. The arsenic contamination of the Pleistocene aquifer of the Red River delta is not only caused by the intensive groundwater abstraction in Hanoi, as reported previously, but also by the natural flow of high arsenic groundwater from Holocene to Pleistocene aquifers in areas located outside of the depression cone. Groundwater with < 50 mu g L-1 arsenic is found in the Pleistocene aquifer close to the recharge zone in the mountains bordering the Red River delta and in the Holocene and Pleistocene aquifers where clay deposits were eroded. Close to the recent Red River channel, recharge of older Holocene and Pleistocene sediments occurs partially by arsenic-contaminated groundwater from the youngest Holocene aquifers, and here arsenic concentrations exceed 50 mu g L-1. A high arsenic concentration is also present in the early Holocene-Pleistocene aquifer, beneath thick clay layers, indicating a limited extent of flushing and the inflow of fresh organic matter.

Arsenic (As) is highly toxic and over 100 million people living on the floodplains of Asia are exposed to excessive groundwater As. A very large spatial variability over small distances has been observed in the groundwater As concentrations. Advances in the prediction of the As distribution in aquifers would support drinking water management. The application of remote sensing of geomorphic paleo river features combined with geological, geophysical and archeological data and available groundwater As measurements may be used to predict groundwater As levels in rural areas, as shown by the example from the Red River delta, Vietnam. Groundwater in sediments deposited in the marine environment is low in As, probably due to the precipitation of As in sulfide minerals under anoxic conditions. Groundwater As levels in freshwater alluvial deposits in undisturbed floodplain areas are slightly increased and the highest As concentrations are associated with meander belts. The meander belts remain clearly visible in remote sensing and may well reflect the youngest preserved alluvial sediments. High As levels in the meander belt aquifers are probably related to the availability of highly reactive organic matter and consequent reduction of iron oxyhydroxides and As release. Furthermore, given similar hydrogeological conditions, the extent of flushing of As from the youngest alluvial sands is limited compared to the older Pleistocene sands. Even within abandoned meander belts a high spatial variability of As concentrations was observed. The younger channel belts (<1 ka BP) and old Holocene aquifers below undisturbed floodplain environments deposited during a period with high sea level host groundwater enriched in As. Low As groundwater is found in sandy channel belts deposited during the regression of the sea and in Pleistocene islands preserved within the floodplain. The decisive influence of the depositional environment of the aquifer sediments on groundwater As content is revealed.

The estimation of groundwater-driven lacustrine dissloved inorganic phosphorus (DIP) loading is a challenging task associated with uncertainties related to simplifications used in the empirical and numerical methods. A multi-disciplinary approach can result in a better understanding of a system and therefore the improved estimate of external DIP loading into lakes. Groundwater discharge and DIP loading to a naturally eutrophic lake were estimated using three methods based on field measurements: (1) the lake segment approach, (2) threedimensional (3D) flow and transport modelling, and (3) integrating mapped water chemistry with the 3D flow model. Field measurements and 3D flow modelling show the groundwater contribution to the lake water budget 75-81%. The in situ measured spatial distribution of the groundwater discharge rate, based on seepage meters and a conservative tracer (Cl-) budget approach, is well reproduced by the 3D flow model. DIP fluxes between the aquifer and the surface water vary between 220 and 351 kg yr(-1). The integrated approach is associated with the lowest degree of uncertainty and therefore predicts the spatial distribution of the mass loading into the lake with the highest degree of accuracy. 3D hydrological information is crucial to estimate the magnitude of DIP transport. Despite the high measured and simulated DIP concentration in groundwater beneath gyttja covering the lake bottom, the actual contribution of this area into annual groundwater-driven DIP loading is low due to low groundwater discharge rates. DIP loading is concentrated instead in the narrow littoral high discharge zones. Low release rate of DIP within the old lake bottom (0.14 mu g L-1) predicted with the use of the zero-order release term in MT3DMS results in a high average DIP concentration in the groundwater (60 mu g L-1) and is one of the major factors responsible for the long-term lake eutrophication.

Arsenic contamination in groundwater is commonly found in alluvial plains of major river basins, in which the Red river delta has also been reported to be contaminated with high levels of arsenic. In this study, groundwater from 50 household wells was collected to study the spatial distribution of arsenic in northwestern Hanoi. The results showed that arsenic concentration in groundwater varied in a wide range of less than 5 to 334 μg/l, of which up 62% of the wells exceeded the WHO guideline value of 10 μg/l for arsenic content in drinking water. Arsenic groundwater in this area is unevenly distributed throughout the area, high arsenic concentrations are found in a narrow band between Red river and Day river. This pattern of arsenic distribution is strongly related to the sediment age, sedimentary processes, and it is also modified by local groundwater flow parts and the occurrence of hydraulic connection between aquifers, which are observed in the study area. Arsenic is released into the groundwater during the reductive dissolution of arsenic-bearing minerals under the presence of organic matter.

Awareness of groundwater-borne dissolved inorganic phosphorus (DIP) loadings into lakes and its role in lake eutrophication is increasing, albeit DIP of natural origin is often ignored. Release of geogenic DIP from an adjacent aquifer and its transport with groundwater into a eutrophic lake is described by combining hydrogeochemical data collected in this study (sampling of piezometers, hydrogeochemical profiles, and seepage meters) with groundwater flow and discharge rates from earlier studies. The major part of the DIP that entered the lake with discharging groundwater was mobilized from iron hydroxides reduced by organic matter buried in the sediments of the old lake bottom. This is indicated by the correlation between DIP and ferrous iron (Fe2+) concentrations, with a DIP/Fe(2+)molar ratio of 0.06, and an increase in pH. One-dimensional reactive transport modeling indicated that high discharge rates (> 0.1 m d(-1)) of anoxic groundwater upwelling in areas adjacent to the lakeshore prevent downward diffusion of oxygen into the aquifer and do not leave enough time for DIP to become rebounding to the mineral phases at the sediment-water interface. The groundwater-controlled DIP input into the lake calculated along a two-dimensional cross-section averaged 0.01 mol DIP m(-2)yr(-1). A 2 m wide offshore high discharge zone delivered approximately 13% of the DIP into the lake. The continuous, external loading of geogenic DIP sustains lake eutrophication and explains the failure of two previous lake restoration attempts.

The groundwater chemistry of the semi-arid volcanic island of Porto Santo, part of the Madeira archipelago, Atlantic Ocean, was investigated. Generally, the groundwater was brackish, containing 2-10 mol % seawater. Groundwater with up to 20 mM alkalinity and a Na enrichment of up to 30 mM, as compared to the Na concentration predicted by the seawater Na/Cl ratio, was found in the main aquifer. Also notable are the high concentrations of F (up to 0.3 mM), B (up to 0.55 mM), As (up to 0.35 mu M), all in excess of WHO recommendations, as well as up to 6 mu M V. Geochemical modeling, using the PHREEQC code, was used to explore different scenarios that could explain the genesis of the observed bulk groundwater chemistry. First, a model for aquifer freshening with the displacement of resident seawater from the aquifer by infiltrating freshwater, was tested. This scenario leads to the development of NaHCO3 waters as observed in many coastal aquifers. However, the measured alkalinity concentration in the groundwater was far higher than the concentration predicted by the freshening model. In addition, the behavior of modelled pH and P(CO2)were at variance with their distributions in the field data. The second model explored the possible effect of volcanic glass leaching on the groundwater chemistry. Using insight derived from studies of volcanic glass surface alteration as well as experimental work on water-volcanic glass interactions, a geochemical model was developed in which the exchange of H+ for Na+ on the volcanic glass surface is the main mechanism but the exchange of other cations on the volcanic glass surface is also included. The uptake of H+ by the glass surface causes the dissociation of carbonic acid, generating bicarbonate. This model is consistent with the local geology and the field data. It requires, however, volcanic glass leaching to occur in the unsaturated zone where there is an unlimited supply of CO2. The exchange reaction of H+ for Na+ is confined to the surface layer of volcanic glass as otherwise the process becomes limited by slow solid state diffusion of H+ into the glass and Na+ out of the glass. Therefore, volcanic ash deposits, with their high volcanic glass surface areas and matrix flow, are the aquifers where this type of high NaHCO3 waters can be expected, rather than in basalts, which predominantly feature fracture flow. The trace components F, B, As and V are believed to originate from hyaloclastites, consisting of predominantly (90%) of trachy-rhyolite volcanic glass. Although stratigraphically older than the main calcarenite aquifer, topographically they are often located at higher altitudes, above the phreatic level and located along the main recharge flow path. In addition, the semiarid climate conditions provide a long groundwater residence time for the reactions as well as limited aquifer flushing.

To assess the contribution of reactive nitrogen from groundwater to surface waters, we need more knowledge on how reactive nitrogen behaves in the glacial till systems underlying many agricultural fields. Groundwater sampled from suction cups and piezometers placed in the glacial till underlying a winter wheat field shows the nitrate concentration in water leaching to deeper than 2 m below ground surface (mbg) is ~ 60 mg L-1. Within 5 mbg, all of the nitrate is removed and this appears to take place within a redox zone rather than at a sharp redox front. Ammonium released from the till is negligible. A 2D dataset reveals that the depth to the redox zone undulates between 3 and 5 mbg, perhaps a result of local variations in infiltration. It appears that the nitrate is generally reduced by the oxidation of pyrite and locally by organic matter in lenses within the till.

The adsorption of As(III) onto reduced sediment from Holocene aquifers in the Red River floodplain, Vietnam, was studied in batch experiments under strict anoxic conditions using sandy sediments from four different sites with a sediment burial age between 600 and 3500 years. Sediments containing the highest content of Fe-oxide showed the highest As(III) adsorption capacity. Removal of the Fe-oxides from the sediment by reductive dissolution significantly reduced the As(III) adsorption capacity of the residual sediment. Normalizing the adsorption of As(III) to the amount of Fe-oxides in the untreated sediments resulted in almost identical adsorption isotherms. We therefore conclude that Fe-oxides are the main adsorbents for As(III) in these reduced aquifer sediments. However, the properties of the sedimentary Fe-oxides are different from those of synthetic Fe-oxides since we found that adsorbed As(III) is not displaced by phosphate and neither is the adsorption of As(III) influenced by changes in pH between 6.2 and 7.5. In the absence of competition for surface sites, As(III) adsorption onto aquifer sediment can be described by a Langmuir isotherm normalized to the amount of Fe-oxides present in the sediment. The finding that the As(III) adsorption capacity depends on the Fe-oxide content of the sediment implies that the As(III) sorption capacity will decrease over time because of reductive dissolution of Fe-oxides, which again is coupled to the burial age of the sediment. In consequence, reactive transport models need to include how aquifer properties, like sorption capacity, change as a function of the extent of chemical reaction. (C) 2018 Elsevier Ltd. All rights reserved.

Combined geological, hydrogeological, and geochemical controls on the arsenic concentration of contaminated aquifers in SE Asia were explored by two-dimensional (2-D) reactive transport modeling of data sets from Bangladesh, Cambodia, and Vietnam. For each site, the field data are summarized and used to create a conceptual 2-D reactive transport model that elucidates characteristic features influencing the groundwater arsenic concentration. Comparison of models for Bangladesh and Vietnam indicates that fine-grained layers overlying young sandy aquifers generate shallow high arsenic groundwater because low vertical groundwater velocities allow sufficient time for kinetic As release from the sediment. The low vertical groundwater velocity below major river channels, predicted by the model, also creates long groundwater residence times, leading to high arsenic groundwater. Young aquifer sediments release more arsenic than older sediments, and alternating young and older sediments create complex patterns of high and low arsenic groundwater. Over time, floodplain basins will subside, and river channels migrate, causing sedimentation and erosion on the floodplain while creating local environments with evolving hydrogeology and groundwater geochemistry. We have developed a three-step model for the evolution of the Red River floodplain with sedimentation and shifting channels over the last 6000years. The results show comparable timescales between the dynamics of arsenic release and of river migration, causing complex groundwater As distributions, comprising geochemical palinopsia of long vanished rivers. Plain Language Summary More than 100 million people in Southeast Asia are affected by too high arsenic concentrations in their groundwater. One of the issues that makes this difficult to handle is that the concentration of arsenic varies enormously within very short distances, making it difficult to find and exploit groundwater low in arsenic. The paper shows how this can be a consequence of complex interactions between the release of arsenic from rust coatings on the sand and clay due to microbes obtaining energy from dissolving the rust using organic matter, the rebinding of arsenic to the rust, the groundwater flow, and the dynamically changing river delta system, leaving chemical afterimages of ancient rivers in the groundwater. To show this, a number of well-studied sites are analyzed using 2-D computer calculations that include all the mentioned processes including a geochemical mathematical description of the processes as they change over time as the reactivity of the organic matter and the rust decreases.

Arsenic contamination of groundwater is a major health problem and has been a growing concern in the last decade in several regions of the world, especially in South and Southeast Asia, including the Red River Delta, Vietnam. Regional groundwater studies have been carried out in the vicinity of Hanoi, on the banks of the Red River and its adjacent floodplains. In this study, the groundwater from a transect in the Northwestern area of Hanoi was examined. The results showed that 28.8% the wells of the B-B' transect exceed the WHO guideline value for arsenic concentration in drinking water. The arsenic concentrations varied in a wide range from point to point, with the highest concentration found at Van Phuc and the lowest one found at Cam Yen. They also varied accordingly to the depth. (c) 2018 Vietnam Academy of Science and Technology

High spatial variability of arsenic (As) concentration in geogenic As-contaminated groundwater has been commonly observed worldwide, but the underlying reasons remain not well understood. Selecting a sulfate-containing, As-affected aquifer at the Datong Basin, northern China as the study area and combining hydrogeochemical investigation and sediment extraction with reactive transport modeling, this work elucidated the roles of Fe-S-As interactions in regulating the vertical variation of As concentration in the groundwater. Dissolved As concentration varied between 0.05 and 18 mot/I, but generally increased in the depth of 20-25 m and then decreased in 25-30 m. The high-As groundwater contained low Fe(II) (<0.007 mmol/L) and up to 15 mu mol/L sulfide, in contrary to the S/SE Asian deltas/floodplains where high Fe(II) and As jointly occur in the groundwater devoid of sulfate reduction. The reductive dissolution of As bearing Fe(III) oxides coupled to the degradation of organic matter with an estimated maximum rate of 0.22 mmol C/L/yr, mainly accounted for the depth-dependent increase of As concentration in the upper part of the shallow aquifer (<25 m deep). However, the decreasing reactivity of Fe(III) oxides together with the increase of pH over depth rendered the majority of electrons being transferred to sulfate reduction. The Fe(II) sulfides formed as a consequence not only helped to restrict the build-up of Fe(II) in the groundwater but also probably co-precipitated As to prompt As decrease in the depth of 25-30 m. Arsenite adsorbed on remaining Fe(III) oxides and newly-formed Fe(II) sulfides is another important pool of As in the aquifer, which varies in response to the extents of Fe(III)-oxide and sulfate reduction and consequently alters As distribution coefficient between the solid and the aqueous phases. This study highlights the importance of coupled geochemical cycling of Fe, S and As for As mobilization and reveals how it regulates As partitioning between groundwater and sediments. (C) 2018 Elsevier B.V. All rights reserved.

Water-sediment interactions were investigated in arsenic contaminated Holocene aquifers of the Red River floodplain, Vietnam, in order to elucidate the origin of the spatial variability in the groundwater arsenic concentration. The investigated aquifers are spread over an 8 x 13 km field area with sediments that varied in burial age from <1 kyr to 11 kyr. The ground-water age ranged from less than 2 yr, up to a maximum near 90 yr. Groundwater As concentrations are between 0 and 6.5 mu M and there are no simple correlations between the As concentration and groundwater age or aquifer sediment burial age. The aquifers are anoxic with up to 2 mM CH4 and up to 0.5 mM DOC. The downward advective DOC flux is too small to support both methanogenesis and the reduction of As-containing Fe-oxides and sedimentary carbon is therefore considered the main carbon source for the redox processes. The groundwater H-2 concentration ranged between 0.1 and 4 nM. These values are intermediate between ranges characteristic for Fe-oxide reduction and methanogenesis and suggest that both processes take place simultaneously. The groundwater pe was calculated from the H-2/H+ and CH4/CO2 redox couples, giving almost similar results that apparently reflects the pe of the bulk groundwater. The pe calculated for the As(III)/As(V) redox couple was found in disequilibrium with the other redox couples. Using the pe calculated from the CH4/CO2 redox couple we show that the groundwater has a reducing potential towards Fe-oxides ranging from ferrihydrite to poorly crystalline goethite, but not for well crystalline goethite or hematite. Hematite and poorly crystalline goethite were identified as the Fe-oxides present in the sediments. Reductive dissolution experiments identify two phases releasing Fe(II); one rapidly dissolving that also contains As and a second releasing Fe(II) more slowly but without As. The initial release of Fe and As occurs at a near constant As/Fe ratio that varied from site to site between 1.2 and 0.1 mmol As/mol Fe. Siderite (FeCO3) is the main sink for Fe(II), based on saturation calculations as well as the identification of siderite in the sediment. Most of the carbonate incorporated in siderite originates from the dissolution of sedimentary CaCO3. Over time the CaCO3 content of the sediments diminishes and FeCO3 appears instead. No specific secondary phases that incorporate arsenite could be identified. Alternatively, the amount of arsenic mobilized during the dissolution of reactive phases can be contained in the pool of adsorbed arsenite. Combining groundwater age with aquifer sediment age allows the calculation of the total number of pore volumes flushed through the aquifer. Comparison with groundwater chemistry shows the highest arsenic concentration to be present within the first 200 pore volumes flushed through the aquifer. These results agree with reactive transport modeling combining a kinetic description of reductive dissolution of As-containing Fe-oxide with adsorption and desorption of arsenite. Understanding variability in groundwater arsenic concentration requires appreciating the coupling of the chemical processes to both sedimentary and hydrogeological cycling. (C) 2018 The Author(s). Published by Elsevier Ltd.

The arsenic content in groundwater of the Red River floodplain decreases with the burial age of the aquifer sediment over a 6000 year period. This decrease is caused by diminishing reactivities of both sedimentary organic carbon, Fe-oxides as well as CaCO3. Here we present a 1-D reactive transport model developed in PHREEQC-3 that quantifies the resulting changes in groundwater bulk chemistry as well as arsenic content over the last six millennia.

Adsorption of As(III) to aquifer sediments from the Red River floodplain, Vietnam, has been studied with the aim of identifying controlling parameters for the adsorption. The adsorption capacity differs between similar sediments, indicating that it is not related to bulk components of the sediment but rather that certain properties of the sediments are important for the adsorption. Normalizing the adsorption of As(III) to the content of Fe-oxides in the sediments results in almost identical adsorption isotherms, indicating that Fe-oxides are the main adsorbents for As(III) in these three reduced aquifer sediments. This is further strengthened by adsorption experiments carried out with sediments leached for Fe-oxides, as this resulted in reduced adsorption capacity and despite the difference in adsorption capacity of the untreated sediments, the adsorption capacity became identical after Fe-oxide leaching. The importance of Fe-oxides for As(III) adsorption to reduced aquifer sediments has implications for the mobility of As(III) in the groundwater aquifers, as it will vary spatially and also over time, as the Fe-oxides in the sediment are reduced.

Twenty‐five years of groundwater quality monitoring in a sandy aquifer beneath agricultural fields showed large temporal and spatial variations in major ion groundwater chemistry, which were linked closely to the nitrate (NO3) content of agricultural recharge. Between 1988 and 2013, the NO3 content of water in the oxidized zone of the aquifer nearly halved, following implementation of action plans to reduce N leaching from agriculture. However, due to denitrification by pyrite oxidation in the aquifer, a plume of sulfate‐rich water migrates through the aquifer as a legacy of the historical NO3 loading. Agriculture thus is an important determinant of major ion groundwater chemistry. Temporal and spatial variations in the groundwater quality were simulated using a 2D reactive transport model, which combined effects of the historical NO3 leaching and denitrification, with dispersive mixing into the pristine groundwater residing deeper in the aquifer. Reactant‐to‐product ratios across reaction fronts are altered by dispersive mixing and transience in reactant input functions. Modelling therefore allowed a direct comparison of observed and simulated ratios of concentrations of NO3 (reactant) in the oxidized zone to those of SO4 (product) in the reduced zone, which aided a stoichiometric assessment of the mechanisms of denitrification. Denitrification by pyrite in the Rabis Creek aquifer results in oxidation of S−1 and Fe2+ in pyrite to S6+ in dissolved SO4 and Fe3+ in Fe‐oxide. Neither precipitation of elemental sulfur (S0), nor of jarosite, was supported by observations, and adsorption of sulfate was also dismissed.

Recharge of Red River water into arsenic-contaminated aquifers below Hanoi was investigated. The groundwater age at 40 m depth in the aquifer underlying the river was 1.3 ± 0.8 years, determined by tritium–helium dating. This corresponds to a vertical flow rate into the aquifer of 19 m/year. Electrical conductivity and partial pressure of CO2 (PCO2) indicate that water recharged from the river is present in both the sandy Holocene and gravelly Pleistocene aquifers and is also abstracted by the pumping station. Infiltrating river water becomes anoxic in the uppermost aquifer due to the oxidation of dissolved organic carbon. Further downward, sedimentary carbon oxidation causes the reduction of As-containing Fe-oxides. Because the release of arsenic by reduction of Fe-oxides is controlled by the reaction rate, arsenic entering the solution becomes highly diluted in the high water flux and contributes little to the groundwater arsenic concentration. Instead, the As concentration in the groundwater of up to 1 μM is due to equilibrium-controlled desorption of arsenic, adsorbed to the sediment before river water started to infiltrate due to municipal pumping. Calculations indicate that it will take several decades of river water infiltration to leach arsenic from the Holocene aquifer to below the World Health Organization limit of 10 μg/L.

Groundwater discharge into a seepage lake was investigated by combining flux measurements, hydrochemical tracers, geological information, and a telescopic modeling approach using first two-dimensional (2-D) regional then 2-D local flow and flow path models. Discharge measurements and hydrochemical tracers supplement each other. Discharge measurements yield flux estimates but rarely provide information about the origin and flow path of the water. Hydrochemical tracers may reveal the origin and flow path of the water but rarely provide any information about the flux. While aquifer interacting with the lake remained under seemingly steady state conditions across seasons, a high spatial and temporal heterogeneity in the discharge to the lake was observed. The results showed that part of the groundwater flowing from the west passes beneath the lake and discharges at the eastern shore, where groundwater springs and high discharge zones (HDZs) are observed at the lake bottom and at seepage faces adjacent to the lake. In the 2-D cross section, surface runoff from the seepage faces delivers 64% of the total groundwater inputs to the lake, and a 2 m wide offshore HDZ delivers 13%. Presence of HDZs may control nutrient fluxes to the lake.