Pieter J. Stuyfzand

Analysis of a deep well recharge experiment by calibrating a reactive transport model with field data.

This paper describes the modeling of the hydrogeochemical effects of deep well recharge of oxic water into an anoxic pyrite-bearing aquifer. Kinetic expressions have been used for mineral dissolution-precipitation rates and organic matter oxidation. Hydrological and chemical parameters of the model were calibrated to field measurements. The results showed that oxidation of pyrite (FeS(2)) and, to a lesser extent, organic matter dominate the changes in quality of the recharged water during its passage through the aquifer. The recharge leads to the consumption of oxygen and nitrate and the formation of sulfate and ferrihydrite. Complexation reactions, cation exchange and precipitation and dissolution of calcite, siderite and rhodochrosite were also identified through the modeling. Despite problems of non-uniqueness of the calibrated parameters, the model was used successfully to depict the geochemical processes occurring in the aquifer. Non-uniqueness can be avoided by constraining the model as much as possible to measurements and/or data from literature, although they cannot be considered always as fixed values and should be considered as stochastic variables instead.

The fate of organic micropollutants during long-term/long-distance river bank filtration

The fate of organic micropollutants during long-term/long-distance river bank filtration (RBF) at a temporal scale of several years was investigated along a row of monitoring wells perpendicular to the Lek River (the Netherlands). Out of 247 compounds, which were irregularly analyzed in the period 1999-2013, only 15 were detected in both the river and river bank observation wells. Out of these, 10 compounds (1,4-dioxan, 1,5-naphthalene disulfonate (1,5-NDS), 2-amino-1,5-NDS, 3-amino-1,5-NDS, AOX, carbamazepine, EDTA, MTBE, toluene and triphenylphosphine oxide) showed fully persistent behavior (showing no concentration decrease at all), even after 3.6 years transit time. The remaining 5 compounds (1,3,5-naphthalene trisulfonate (1,3,5-NTS), 1,3,6-NTS, diglyme, iopamidol, triglyme) were partially removed. Their reactive transport parameters (removal rate constants/half-lives, retardation coefficients) were inferred from numerical modeling. In addition, maximum half-lives for 14 of the fully removed compounds, for which the data availability was sufficient to deduce 100% removal during sub-surface passage, were approximated based on travel times to the nearest well. The study is one of very few reporting on the long-term field-scale behavior of organic micropollutants. It highlights the efficiency of RBF for water quality improvement as a pre-treatment step for drinking water production. However, it also shows the very persistent behavior of various compounds in groundwater.

Observations and analytical modeling of freshwater and rainwater lenses in coastal dune systems

Observations are reported on (i) groundwater recharge rates under various types of vegetation as measured with megalysimeters in the dunes, (ii) freshwater lenses along the Dutch North Sea coast in the early 1900s, and (iii) rainwater lenses that develop on top of laterally migrating, artificially recharged riverwater. Subsequently analytical methods are presented to estimate annual natural groundwater recharge as function of rainfall and vegetation, and to calculate the size, shape and transition zone of freshwater lenses on saline groundwater and rainwater lenses on infiltrated riverwater. An empirical correction factor, based on the hydraulic resistance of an aquitard within the freshwater lens, is proposed to account for the frequently observed reduction of the Ghyben-Herzberg ratio of 40. This factor raises the groundwater table, reduces the depth of the fresh/salt interface and increases the lens formation time. The suite of methods offers a tool box for knowledge based water management of dune systems, by rapidly predicting: (i) more or less autonomous changes due to sealevel rise, climate change and vegetation development; and (ii) the potential (side) effects of interventions. Knowing what happened or will happen to the fresh water lens or a rainwater lens is important, because changes impact on important natural habitat parameters such as salinity, depth to groundwater table, decalcification rate (and thus on pH, Ca/Al, PO4, NH4) and nutrient availability, and on drinking water supply. The analytical models are applied to predict effects of sealevel rise, coastal progradation, vegetation changes, and increased temperature of coastal air and river water to be infiltrated.

Transport of bacteriophage MS2 and PRD1 in saturated dune sand under suboxic conditions

Soil passage of (pretreated) surface water to remove pathogenic microorganisms is a highly efficient process under oxic conditions, reducing microorganism concentrations about 8 log10 within tens of meters. However, under anoxic conditions, it has been shown that removal of microorganisms can be limited very much. Setback distances for adequate protection of natural groundwater may, therefore, be too short if anoxic conditions apply. Because removal of microorganisms under suboxic conditions is unknown, this research investigated removal of bacteriophage MS2 and PRD1 by soil passage under suboxic conditions at field scale. At the field location (dune area), one injection well and six monitoring wells were installed at different depths along three suboxic flow lines, where oxygen concentrations ranged from 0.4 to 1.7 mg/l and nitrate concentrations ranged from 13 to 16 mg/L. PRD1 and MS2 were injected directly at the corresponding depths and their removal in each flow line was determined. The highest bacteriophage removal was observed in the top layer, with about 9 log removal of MS2, and 7 log removal of PRD1 after 16 meters of aquifer transport. Less removal was observed at 12 m below surface, probably due to a higher groundwater velocity in this coarser grained layer. MS2 was removed more effectively than PRD1 under all conditions. Due to short travel times, inactivation of the phages was limited and the reported log removal was mainly associated with attachment of phages to the aquifer matrix. This study shows that attachment of MS2 and PRD1 is similar for oxic and suboxic sandy aquifers, and, therefore, setback distances used for sandy aquifers under oxic and suboxic conditions provide a similar level of safety. Sticking efficiency and the attachment rate coefficient, as measures for virus attachment, were evaluated as a function of the physico-chemical conditions.

Effects of intake interruptions on dune infiltration systems in the Netherlands, their quantification and mitigation

In the coastal dunes of the Western Netherlands, managed aquifer recharge (MAR) is applied for drinking water supply since 1957. The MAR systems belong to the Aquifer Transfer Recovery (ATR) type, because recharge and recovery are operated without interruption. This makes these systems very vulnerable to intake interruptions, which are expected to increase in frequency and duration due to climate change. Such interruptions are problematic, because: (i) groundwater recovery from dunes needs to continue to supply fresh drinking water to the Western Netherlands; (ii) risks of salt water intrusion are high, and (iii) MAR bordering wet dune slacks with an EU Natura 2000 status cannot survive for long without MAR. In this paper, effects of intake stops are discussed and quantified. The hydrological effects consist of the decline of water tables, disappearance of flow-through dune lakes, reservoir depletion, salt water intrusion, disruption of rainwater lenses, and entrapped air hampering a rapid refill of the groundwater reservoir. Water quality effects include changes in (i) redox environment of the flushed aquifer, impacting the behavior of nutrients, calcium, sulfate and organic micro-pollutants, and (ii) the mixing ratio of water types. The main ecological impacts comprise the dying of organisms in recharge ponds and dune lakes, and a decline of biodiversity. Effects of very long intake interruptions (years) are predicted via historical observations during the long overexploitation period (1900-1957) prior to MAR. A closed form analytical solution for safe yield of a semiconfined aquifer is proposed, together with a related upconing risk index. Both also apply to the pumping from any fresh water lens without MAR. Some mitigation strategies are discussed, such as a dual intake, raising the storage capacity, earlier mud removal, and accelerated refilling of the reservoir. A magnitude scale for intake stops (MIS) is proposed.

Simulation of saltwater intrusion in a poorly karstified coastal aquifer in Lebanon (Eastern Mediterranean)

To date, there has been no agreement on the best way to simulate saltwater intrusion (SWI) in karst aquifers. An equivalent porous medium (EPM) is usually assumed without justification of its applicability. In this paper, SWI in a poorly karstified aquifer in Lebanon is simulated in various ways and compared to measurements. Time series analysis of rainfall and aquifer response is recommended to decide whether quickflow through conduits can be safely ignored. This aids in justifying the selection of the exemplified EPM model. To examine the improvement of SWI representation when discrete features (DFs) are embedded in the model domain, the results of a coupled discrete-continuum (CDC) approach (a hybrid EPM-DF approach) are compared to the EPM model. The two approaches yielded reasonable patterns of hydraulic head and groundwater salinity, which seem trustworthy enough for management purposes. The CDC model also reproduced some local anomalous chloride patterns, being more adaptable with respect to the measurements. It improved the overall accuracy of salinity predictions at wells and better represented the fresh–brackish water interface. Therefore, the CDC approach can be beneficial in modeling SWI in poorly karstified aquifers, and should be compared with the results of the EPM method to decide whether the differences in the outcome at local scale warrant its (more complicated) application. The simulation utilized the SEAWAT code since it is density dependent and public domain, and it enjoys widespread application. Including DFs necessitated manual handling because the selected code has no built-in option for such features.RésuméA ce jour, il n’y a pas eu d’accord sur la meilleure façon de simuler l’intrusion d’eau salée (IES) dans les aquifères karstiques. Un milieu poreux équivalent (MPE) est généralement supposé sans justification de son applicabilité. Dans cet article, IES dans un aquifère peu karstifié au Liban est simulé de différentes manières et comparé aux mesures. Il est recommandé d’effectuer une analyze des séries chronologiques des précipitations et de la réponse de l’aquifère pour décider si un écoulement rapide au sein des conduits peut être négligé en toute sécurité. Cela aide dans la justification de la sélection du modèle MPE donné en exemple. Pour examiner l’amélioration de la représentation de l’IES lorsque des caractéristiques discrètes (CD) sont incorporées dans le domaine du modèle, les résultats d’une approche couplée discrète-continu (CDC) (approche hybride MPE-CD) sont comparés à ceux du modèle MPE. Les deux approches ont fourni des représentations raisonnables de la charge hydraulique et de la salinité des eaux souterraines, qui semblent suffisamment fiables pour des objectifs de gestion. Le modèle CDC a également reproduit certaines anomalies locales en chlorure, étant plus ajustable par rapport aux mesures. Cela a amélioré la précision globale des prédictions de salinité aux puits et a permis de mieux représenter l’interface eau douce–eau saumâtre. Par conséquent, l’approche CDC peut être bénéfique dans la modélisation de l’IES d’aquifères peu karstifiés, et doit être comparé aux résultats de la méthode MPE pour décider si les différences dans les résultats à l’échelle local justifient son application (plus complexe). La simulation a utilisé le code SEAWAT car il dépend de la densité et est disponible dans le domaine public, et il bénéficie d’une application étendue. L’intégration de CD a nécessité une manipulation manuelle car le code sélectionné n’a pas d’options intégrées pour de telles fonctionnalités.ResumenHasta la fecha, no ha habido acuerdo sobre la mejor manera de simular la intrusión de agua salada (SWI) en los acuíferos kársticos. Por lo general, se asume un medio poroso equivalente (EPM) sin justificación de su aplicabilidad. En este trabajo, la SWI en un acuífero pobremente karstificado en el Líbano se simula de varias maneras y se lo compara con mediciones. Se recomienda el análisis de series de tiempo de la lluvia y la respuesta del acuífero para decidir si el flujo rápido a través de los conductos se puede ignorar de manera segura. Esto ayuda a justificar la selección del modelo EPM ejemplificado. Para examinar la mejora de la representación de la SWI cuando las características discretas (DFs) están integradas en el dominio del modelo, los resultados de un enfoque de continuo discreto (CDC) acoplado (un enfoque híbrido EPM-DF) se comparan con el modelo EPM. Los dos enfoques arrojaron patrones razonables de carga hidráulica y de salinidad del agua subterránea, que parecen lo suficientemente confiables para los fines de gestión. El modelo de CDC también reprodujo algunos patrones de cloruro anómalos locales, siendo más adaptable con respecto a las mediciones. Mejoró la precisión general de las predicciones de salinidad en los pozos y representó mejor la interfaz de agua dulce–salobre. Por lo tanto, el enfoque CDC puede ser beneficioso para modelar SWI en acuíferos pobremente karstificados, y debe compararse con los resultados del método EPM para decidir si las diferencias en el resultado a escala local justifican su aplicación (más complicada). La simulación utilizó el código SEAWAT ya que depende de la densidad y es de dominio público, y goza de una amplia aplicación. La inclusión de los DF requiere manejo manual porque el código seleccionado no tiene una opción incorporada para tales características.摘要迄今为止,人们一直还没有统一意见认为哪种方法是模拟岩溶含水层海水入侵的最好方法。通常假定为等量孔隙介质,而 不去过多地考虑其适用性。本文中,通过不同方式模拟了黎巴嫩一个未充分岩溶化含水层的海水入侵情况,并与测量的结果进行了对比。建议进行降雨及含水层的响应的时序分析以决定通过管道的快流是否能安全地被忽略。这有助于合理选择例证的等量孔隙介质模型。当离散特点嵌入模型域时,为了检验海水入侵展示的改善情况,一个耦合离散-连续统方法(混合的等量介质-离散特点方法)的结果与等量孔隙介质模型进行了对比。两种方法都出现了水头和地下水含盐度合理的模式,似乎用于管理足够可靠。耦合离散连续统模型还再生出一些局部异常氯化钠模式,在测量结果方面更加适合。它能提高井中含盐度的整体精度,更好地展示了淡水-微咸水界面。因此,耦合离散-连续统方法利于模拟未充分岩溶化含水层海水入侵,应当与等量孔隙介质方法进行比较,以决定局部尺度上结果差异是否能保证其(更复杂)的应用。模拟利用了SEAWAT编码,因为其依靠密度,并且是公共域,受到了广泛的应用。包含离散特点需要人工处理,因为对于这样的特点选择的编码没有内置的选择。ResumoAté à data, não há acordo sobre a melhor maneira de simular a intrusão de água salina (IAS) em aquíferos cársticos. Um meio poroso equivalente (MPE) é assumido geralmente sem justificativa de sua aplicação. Neste artigo, a IAS em um aquífero pobremente carstificado no Líbano foi simulada de várias maneiras e comparada às medições. A análise de séries temporais de precipitação e respostas do aquífero é recomendada para decidir se o fluxo rápido através dos condutos pode ser ignorado com segurança. Isso ajuda a justificar a seleção do modelo de MPE exemplificado. Para examinar a melhoria da representação da IAS quando as características discretas (CDs) são incorporadas no domínio do modelo, os resultados de uma abordagem de contínuo-discreto acoplado (CDA) (abordagem híbrida MPE-CD) são comparados ao modelo de MPE. As duas abordagens renderam padrões razoáveis de carga hidráulica e salinidade das águas subterrâneas, os quais parecem ser confiáveis o suficiente para fins de gerenciamento. O modelo de CDA também reproduziu alguns padrões locais anômalos de cloretos, sendo mais adaptável em relação às medições. O modelo melhorou a precisão geral das previsões de salinidade em poços e representou melhor a interface água doce–salobra. Portanto, a abordagem de CDA pode ser benéfica na modelagem da IAS em aquíferos pobremente carstificados, e deve ser comparada com os resultados do método de MPE para decidir se as diferenças no resultado em escala local garantem sua aplicação (mais complicada). A simulação utilizou o código SEAWAT, uma vez que depende de densidade e domínio público, e goza de uma aplicação generalizada. A inclusão de CDs exigiu um manuseio manual, pois o código selecionado não possuía opção interna para tais recursos.

Mitigation of saltwater intrusion by ‘integrated fresh-keeper’ wells combined with high recovery reverse osmosis

Most countermeasures to mitigate saltwater intrusion in coastal, karstic or fractured aquifers are hindered by anisotropy, high transmissivities and complex dynamics. A coupled strategy is introduced here as a localized remedy to protect shallow freshwater reserves while utilizing the deeper intercepted brackish water. It is a double sourcing application where fresh-keeper wells are installed at the bottom of a deepened borehole of selected salinized wells, and then supported by high recovery RO desalination. The RO design has <1kWh/m3 energy consumption, and up to 96% recovery in addition to low scaling propensity without use of any anti-scalant. A feasibility study is presented as an example for a salinizing, brackish well (TDS ~1600mg/L) in the Damour coastal aquifer in Lebanon. The concept is expected to produce ca. 1000m3/d of freshwater from this well by pumping 250m3/d of fresh groundwater from the top well screen and 800m3/d of brackish groundwater (to be later desalinized) from the fresh-keeper well screen below. Cost analysis shows that the capital cost could be returned back in 1 to 4years depending on the choice of produced water (bottled or tap) and available market. As an alternative, water from the RO plant could be blended with lower quality water, for instance untreated brackish groundwater (if unpolluted), to supply 3 more volumes for domestic use. The usage of brackish groundwater from integrated fresh-keeper wells thus serves 3 purposes: production of high quality drinking water, financial gain and mitigation of water stress by overpumping.