The Coastal Plain in the southern Chesapeake Bay area is becoming increasingly susceptible to nuisance flooding as a result of the combination of sea-level rise and land subsidence associated with aquifer compaction from excessive groundwater pumping. Detailed time-series of cumulative compaction data (land subsidence) from the three U.S. Geological Survey deployed extensometers in the regions, along with cyclical piezometer data, reflect the nature of the complex multi-aquifer/aquitard system in the Coastal Plain. Franklin, Virginia and Suffolk, Virginia extensometers were deactivated in 1995 and were reactivated in 2016 along with the addition of a high-sensitivity borehole extensometer in Nansemond, Virginia in collaboration with the Hampton Roads Sanitation District as a part of the Sustainable Water Initiative for Tomorrow (SWIFT). Yearly compaction rates estimated from the reactivated extensometers are -3.3 mm/year, 15.6 mm/year, and -20.7 mm/year in Franklin, Suffolk, and Nansemond, Virginia respectively. One-dimensional vertical compaction modeling is utilized to estimate the total compaction and differentiate which fine-grained confining units or aquifer interbeds are contributing most to total compaction historically and presently. Additionally, properties of the system can be estimated including the elastic specific storage of the aquitards and aquifers and the inelastic storage of the aquitards. The total cumulative change in aquifer system thickness estimated by the MODFLOW subsidence package can be compared to the observed total cumulative change in aquifer system thickness at each site for validation of hypothesis about the dynamics of the aquifer system to known changes in stress. Subsidence rates and aquifer/aquitard properties can be useful for managing and modeling the groundwater in the Coastal Plain of Virginia. / Master of Science / The Coastal Plain in the southern Chesapeake Bay area is becoming increasingly susceptible to flooding at high tides in low lying areas as a result of the combination of sea-level rise and sinking of the land surface (land subsidence) associated with aquifer compaction from excessive groundwater pumping from buried aquifers. Detailed time-series of land subsidence data from the three U.S. Geological Survey deployed extensometers in the region, along with water level data from nearby wells, reflect the nature of the complex multi-aquifer/aquitard system in the Coastal Plain. Franklin, Virginia and Suffolk, Virginia extensometers were deactivated in 1995 and were reactivated in 2016 along with the addition of a high-sensitivity borehole extensometer in Nansemond, Virginia in collaboration with the Hampton Roads Sanitation District as a part of the Sustainable Water Initiative for Tomorrow (SWIFT). Yearly land subsidence rates estimated from the reactivated extensometers are -3.3 mm/year, 15.6 mm/year, and -20.7 mm/year in Franklin, Suffolk, and Nansemond, Virginia respectively. One-dimensional vertical compaction modeling is utilized to estimate the total sinking of the land surface as well as to differentiate which fine-grained confining units or aquifer interbeds are contributing most to total subsidence historically and presently. Additionally, properties of the system can be estimated including the elastic specific storage of the aquitards and aquifers and the inelastic storage of the aquitards. The total cumulative change in aquifer system thickness estimated by the MODFLOW subsidence package can be compared to the observed total cumulative change in aquifer system thickness at each site for validation of hypothesis about the dynamic changes of the aquifer system with known changes in stress. Subsidence rates, understanding the dynamics of the aquifer system, and aquifer/aquitard properties can be useful for managing groundwater and modeling the aquifer system in the Coastal Plain of Virginia.

<p>When protection zones for wells are delineated, it is important to acquire good knowledge about possible travel time from different points in the catchment area to the well. Often, simple analytical methods are used for estimating travel times and the assumption is made that the hydraulic conductivity is relatively homogenous within the aquifer. Nevertheless, many aquifers are strongly heterogeneous which may lead to differences between estimates and actual travel times. As a part of the process to develop improved methods for delineating protection zones for groundwater supply wells, a tracer experiment was performed in a glaciofluvial esker formation in Järlåsa. On the basis of the experiment, a numerical flow model was created for the test site.</p><p>The purpose of this master’s thesis was to apply the flow model to an aquifer where the hydraulic conductivity shows great variability and should be described by a stochastic distribution. The purpose was also to determine the statistical properties of the hydraulic conductivity and to simulate the transport times and their variation from different locations in the aquifer to the pumping well.</p><p>The hydraulic conductivity was estimated from grain size distributions in soil samples that were taken at various locations within the test site. The analysis of the hydraulic conductivity showed a large variation and confirmed the hypothesis that the aquifer is heterogeneous. Using the statistics, a large number of stochastic conductivity fields were created and flow simulation was performed for each realization. From the simulation result, frequency distributions of the transport times were produced describing the probability for the transit time for a water particle between a certain location in the aquifer and the pumping well. A comparison with the tracer experiment shows higher simulated transport times implying the need for improved model calibration. The conclusion is that the method used in this project is suitable for glaciofluvial esker aquifers. When protection zones are delineated, stochastic modeling can be used to express the zone boundaries in statistical terms.</p> / <p>När skyddsområden för grundvattentäkter skapas är det viktigt att ha god kännedom om vattnets transporttider från olika delar i tillrinningsområdet till uttagsplatsen av grundvattnet. Metoderna för att bestämma dessa tider är ofta enkla och vanligen görs antagandet att områdets hydrauliska konduktivitet är relativt homogen. Många akviferer är dock kraftigt heterogena och de verkliga transporttiderna kan då skilja sig från de uppskattade. Som ett led i metodutvecklingen för bättre avgränsning av skyddsområden genomfördes ett spårämnesförsök i en isälvsavlagring i Järlåsa. Med utgångspunkt från försöket har en numerisk flödesmodell konstruerats över försöksområdet.</p><p>Syftet med examensarbetet var att tillämpa flödesmodellen i en akvifer där den hydrauliska konduktiviteten visar sådana variationer att den beskrivs bäst av en stokastisk fördelning. Vidare var syftet att bestämma den hydrauliska konduktivitetens statistiska egenskaper och att simulera transporttiderna och deras variation från olika punkter i akviferen till pumpbrunnen.</p><p>Den hydrauliska konduktiviteten uppskattades utifrån kornstorleksfördelningar i jordprover som togs på en mängd platser i försöksområdet. Analysen av den hydrauliska konduktiviteten visar stora variationer i området vilket bekräftar att akviferen är heterogen.</p><p>Utifrån konduktivitetens statistik genererades ett stort antal stokastiska konduktivitetsfält och transporttiderna beräknades för varje realisering. Resultatet från simuleringarna gav frekvensfördelningar för transporttiderna som beskriver sannolikheten för hur lång uppehållstid en vattenpartikel har i marken mellan en startpunkt och pumpbrunnen. Jämfört med spårämnesförsöket blev de simulerade transporttiderna något större vilket tyder på att flödesmodellen kräver en bättre kalibrering mot fältmätningar. Slutsatsen är att metodiken är lämplig för att studera vattnets transporttider i isälvsavlagringen och när ett skyddsområde skapas för den här typen av akvifer kan stokastisk modellering användas för att beskriva skyddszoner i form av statistiska termer.</p>

A contaminação da água subterrânea por nitrato em áreas urbanas é um problema muito comum no Brasil. Geralmente muito dos estudos restringem-se em descrever somente os problemas associados às porções mais superficiais dos aquíferos. Mais recentemente, alguns trabalhos têm detectado a presença de nitrato nas partes mais profundas de aquíferos livres no Estado de São Paulo. Então, o principal objetivo desse trabalho foi o de avaliar os impactos potenciais do saneamento urbano no Aquífero Adamantina (livre, de porosidade primária, com 120 m de espessura saturada) na cidade de Urânia, simulando diferentes cenários de usos da terra, usando modelos numéricos de fluxo e transporte, com os softwares MODFLOW e MT3DMS. A cidade de Urania é abastecida por água subterrânea e a rede de esgoto cobre quase a totalidade da área urbana, em um sistema operado pela SABESP (Companhia de Saneamento Básico do Estado de São Paulo). Poços privados e sistemas de saneamento in situ são práticas bastante comuns na cidade. Os resultados da modelagem de fluxo permitiram identificar que o tempo de trânsito das águas subterrâneas entre sua área de recarga e a porção inferior do Aquífero Adamantina (120 m da superfície), sem a operação de poços de produção, é de aproximadamente 60 anos. Este tempo diminui para menos de 40 anos quando estes poços estão bombeando. Caso a contaminação por nitrato seja de uma fonte constante em toda a área urbana, os resultados da modelação mostram que o aquífero tem capacidade de diluir apenas 30% da contaminação inicial. Considerando a construção de uma rede de esgoto totalmente eficiente, o modelo indicou que o tempo de recuperação das porções rasa e intermediária do aquífero é de 10 anos. Já na porção mais profunda, a contaminação perduraria por um período de 90 anos após o encerramento da fonte, com uma fração de 11% de sua concentração inicial, após 10 anos. Assim, conclui-se que a efetividade de estender a rede de esgoto no município em um período de 10 anos é uma medida interessante para mitigação do aquífero frente à contaminação por nitrato. A capacidade máxima de diluição do contaminante pelo aquífero na área urbana de Urânia corresponde a fontes de contaminação de até 100 mg/L de nitrato oriundas de áreas de 65 m x 65 m na zona urbana do município, com um espaçamento de 130 m x 130 m entre elas. A alternativa de contenção da contaminação para porções inferiores do aquífero por meio da existência de poços de produção com seção filtrante inserida na porção rasa do aquífero, visando à utilização desta água bombeada para usos menos nobres, enquanto os poços de produção destinados ao consumo humano, cujos filtros estariam posicionados na porção profunda, permaneceriam protegidos, somente se mostrou interessante quando as concentrações originais de nitrato forem inferior a 56 mg/L, pois, desse modo, a capacidade de diluição do aquífero faria com que a concentração em água da porção profunda fosse inferior ao limite de potabilidade. / Groundwater contamination by nitrate in urban areas is a very common problem in Brazil. Until now, many studies just describe the problem associated to shallowest part of aquifers. More recently, some works have detected the presence of nitrate in the deeper portion of some important unconfined aquifers in Sao Paulo state. Considering this potential problem, the main objective of this work was to evaluate the potential impacts of urban sanitation into the Adamantina Aquifer (primary porosity, unconfined aquifer with 120 m of saturated zone thickness) in the city of Urânia, simulating different land occupation scenarios using flow and transport mathematical models, respectively MODFLOW and MT3DMS codes. The city of Urânia is totally supplied by groundwater and the sewage mains cover almost all the urban area. Both water and sewage mains are operated by the SABESP (State of Sao Paulo water company), but it is necessary to recognize an important number of private wells and also septic tanks and pit latrines operating in the city. The numerical model results could estimate the groundwater transit time from the recharge to the Adamantina Aquifer bottom (120 mbs) in 60 years. When the water wells are pumping, this time is reduced to 40 years. A big impact is expected when the city is not served by any sewage mains and all effluent is infiltrated direct into the soil. In this case, considering a concentration of 100 mg/L, the nitrate will reach the whole aquifer (including its bottom part) with 70% of the original concentration, in 60 years. Taking into the consideration that all domestic effluent is drained by the public sewage mains, the time for recovery is about 10 years for the shallower and intermediate parts of the aquifer. For the deeper part, it takes more than 90 years, but the expected concentration is < 11% of the original one, after 10 years. This permits to conclude that the construction of an efficient public sewage mains network could be a good solution in a time frame of 10 years. It is also possible to define that to dilute a constant load of nitrate of 100 mg/L in an area of 65x65 m, it is necessary to have an area of 130 m x 130 m to keep the water below the Brazilian potable standards.

A contaminação da água subterrânea por nitrato em áreas urbanas é um problema muito comum no Brasil. Geralmente muito dos estudos restringem-se em descrever somente os problemas associados às porções mais superficiais dos aquíferos. Mais recentemente, alguns trabalhos têm detectado a presença de nitrato nas partes mais profundas de aquíferos livres no Estado de São Paulo. Então, o principal objetivo desse trabalho foi o de avaliar os impactos potenciais do saneamento urbano no Aquífero Adamantina (livre, de porosidade primária, com 120 m de espessura saturada) na cidade de Urânia, simulando diferentes cenários de usos da terra, usando modelos numéricos de fluxo e transporte, com os softwares MODFLOW e MT3DMS. A cidade de Urania é abastecida por água subterrânea e a rede de esgoto cobre quase a totalidade da área urbana, em um sistema operado pela SABESP (Companhia de Saneamento Básico do Estado de São Paulo). Poços privados e sistemas de saneamento in situ são práticas bastante comuns na cidade. Os resultados da modelagem de fluxo permitiram identificar que o tempo de trânsito das águas subterrâneas entre sua área de recarga e a porção inferior do Aquífero Adamantina (120 m da superfície), sem a operação de poços de produção, é de aproximadamente 60 anos. Este tempo diminui para menos de 40 anos quando estes poços estão bombeando. Caso a contaminação por nitrato seja de uma fonte constante em toda a área urbana, os resultados da modelação mostram que o aquífero tem capacidade de diluir apenas 30% da contaminação inicial. Considerando a construção de uma rede de esgoto totalmente eficiente, o modelo indicou que o tempo de recuperação das porções rasa e intermediária do aquífero é de 10 anos. Já na porção mais profunda, a contaminação perduraria por um período de 90 anos após o encerramento da fonte, com uma fração de 11% de sua concentração inicial, após 10 anos. Assim, conclui-se que a efetividade de estender a rede de esgoto no município em um período de 10 anos é uma medida interessante para mitigação do aquífero frente à contaminação por nitrato. A capacidade máxima de diluição do contaminante pelo aquífero na área urbana de Urânia corresponde a fontes de contaminação de até 100 mg/L de nitrato oriundas de áreas de 65 m x 65 m na zona urbana do município, com um espaçamento de 130 m x 130 m entre elas. A alternativa de contenção da contaminação para porções inferiores do aquífero por meio da existência de poços de produção com seção filtrante inserida na porção rasa do aquífero, visando à utilização desta água bombeada para usos menos nobres, enquanto os poços de produção destinados ao consumo humano, cujos filtros estariam posicionados na porção profunda, permaneceriam protegidos, somente se mostrou interessante quando as concentrações originais de nitrato forem inferior a 56 mg/L, pois, desse modo, a capacidade de diluição do aquífero faria com que a concentração em água da porção profunda fosse inferior ao limite de potabilidade. / Groundwater contamination by nitrate in urban areas is a very common problem in Brazil. Until now, many studies just describe the problem associated to shallowest part of aquifers. More recently, some works have detected the presence of nitrate in the deeper portion of some important unconfined aquifers in Sao Paulo state. Considering this potential problem, the main objective of this work was to evaluate the potential impacts of urban sanitation into the Adamantina Aquifer (primary porosity, unconfined aquifer with 120 m of saturated zone thickness) in the city of Urânia, simulating different land occupation scenarios using flow and transport mathematical models, respectively MODFLOW and MT3DMS codes. The city of Urânia is totally supplied by groundwater and the sewage mains cover almost all the urban area. Both water and sewage mains are operated by the SABESP (State of Sao Paulo water company), but it is necessary to recognize an important number of private wells and also septic tanks and pit latrines operating in the city. The numerical model results could estimate the groundwater transit time from the recharge to the Adamantina Aquifer bottom (120 mbs) in 60 years. When the water wells are pumping, this time is reduced to 40 years. A big impact is expected when the city is not served by any sewage mains and all effluent is infiltrated direct into the soil. In this case, considering a concentration of 100 mg/L, the nitrate will reach the whole aquifer (including its bottom part) with 70% of the original concentration, in 60 years. Taking into the consideration that all domestic effluent is drained by the public sewage mains, the time for recovery is about 10 years for the shallower and intermediate parts of the aquifer. For the deeper part, it takes more than 90 years, but the expected concentration is < 11% of the original one, after 10 years. This permits to conclude that the construction of an efficient public sewage mains network could be a good solution in a time frame of 10 years. It is also possible to define that to dilute a constant load of nitrate of 100 mg/L in an area of 65x65 m, it is necessary to have an area of 130 m x 130 m to keep the water below the Brazilian potable standards.

Située au Nord-Ouest de la Tunisie centrale, le bassin de Foussana du gouvernorat de Kasserine constitue une zone de transition entre la région tellienne pluvieuse au Nord et la région saharienne au Sud. Il est bordé par une série de reliefs de formes et de directions variées. Dans cette région aride à semi-aride, l’eau souterraine représente la principale réserve et ressource de production d’eau. Du point de vue tectonique, le bassin de Foussana est caractérisé par la présence de multiples failles. Les failles de direction NW-SE, dont la plus importante est celle qui s’aligne parallèlement à l’Oued Hatab (Faille de Foussana). Les failles qui bordent le fossé du côté est et du côté Ouest se placent dans la famille directionnelle NW-SE. Les failles de direction NE-SW, dont la plus importante est celle qui borde Jebel Chambi du côté sud, montrent une activité décrochante senestre à composante normale. Les failles de direction EW constituent la troisième famille de fracture ayant contrôlé l’évolution de cette structure effondrée. Le bassin d’effondrement de Foussana formé par le jeu de failles de direction NW-SE constitue une fosse subsidente comblée de dépôts MioPlioQuaternaires à caractère continental et renferment plusieurs niveaux aquifères avec des intercommunications latérales entre les grés miocènes et les niveaux Plio-Quaternaires le long des failles de bordure et par drainance le long de l’axe de la fosse matérialisé par l’oued El Hatab. Le Quaternaire couvre toute la cuvette. Sa constitution varie selon les régions, en bordure, nous retrouvons les éboulis de pente formés de gros éléments de calcaire sub- arrondis, parfois bien cimentés par de l’argile rouge. Ces dépôts laissent la place au centre de la cuvette de l’Oued Hatab à des dépôts plus fins constitués de sol sableux ou argilo - sableux avec une certaine proportion de limon. Dans les régions de bordure, les dépôts du PlioQuaternaire très grossiers sont directement superposés aux grès miocènes. Les aquifères profond du Pliocène (PQ2) et superficiel du Quaternaires (PQ1) sont susceptibles de constituer un aquifère unique, puisque leurs résistivités sont presque semblables, leurs salinités sont voisines, et leurs piézométries sont confondues. Les grès Miocènes varient d’une région à l’autre selon qu’on est dans les zones de bordures ou dans des fosses. Ils sont libres au Nord et au Sud de la plaine séparés par les deux grands accidents tectoniques. Les grès en affleurement sont non seulement plus grossiers mais aussi plus homogènes, ce qui attribue aux aquifères libres une grande perméabilité, contrairement aux aquifères des grès en charge où les alternances fréquentes d’argiles et de marnes réduisent énormément ce paramètre. Cette homogénéité croit également en profondeur. Dans l’optique de caractériser le fonctionnement du système hydrogéologique multicouche et de calculer un bilan hydrologique pour le système aquifère, une méthodologie multi techniques couplant l’ hydrochimie, l’isotopie et l’hydrodynamisme a été déployée. L’analyse détaillée des données hydrochimiques recueillies a mis en évidence la grande variabilité du fond géochimique local, directement fonction de la stratigraphie. Cette complexité stratigraphique induit une vision hydrogéologique complexe de nombreux corps aquifères. L’évolution de la chimie des eaux révèle la prépondérance du phénomène de dissolution des minéraux évaporitiques et la précipitation de la calcite comme principal mécanisme de contrôle de l’hydrochimie devant les phénomènes d’échange de base et par la pollution anthropique induit par la présence des nitrates. En conséquence il n’a pas été possible de déduire de l’hydrochimie des eaux les preuves d’une recharge par drainance ascendante depuis l’aquifère du Miocene vers les aquifères superficiels, bien que la similarité des faciès hydrochimiques tend à consolider l’hypothèse d’une origine unique des eaux, sans toutefois permettre une identification des chemins d’écoulement. / The study area lies in the northern part of central Tunisia and extends over about 650 km2. It is located betweenthe rainy Tellian region in the North and the Saharan region in the South. The Foussana Basin is a typical closedinland basin surrounded by high mountains where the precipitation varies over space and time as a result oftopography and seasonality.The main outcrop that marks the watershed of Foussana basin is the quaternary alluvium, which is surrounded byCretaceous reef of deposits covering folds Mountains bounding. The Foussana basin is composed of two aquifers,the PlioQuaternary and the Miocene aquifers. The PlioQuaternary aquifer consists of clayey sand, coarse sand andsandstone deposits. This aquifer is composed of two permeable levels PlioQuaternary Aquifer (PQ1) and PlioceneAquifer (PQ2), which are separated by interbedded sandy marl. The Miocene aquifer consists of sandstone andcoarse sands. This aquifer constitutes an important water source in the study area because of its relativelysignificant thickness and its water quality. It‘s separated to the PlioQuaternary aquifer by a clay layer. The differentaquifers communicate through faults.The groundwater flows occurs from the borders to the center of the basin,i.e. from the surrounding hills to the depression area. The main source of aquifer recharge is infiltration of rainfalland dam water.Groundwater samples with low TDS values, which typically correspond to outcrops of the Miocene aquifer andsome samples of the first PlioQuaternary aquifer PQ1, are typically of SO4-Ca-HCO3-Na water type. In contrast,groundwater samples with high TDS values, most of which were collected in the PlioQuaternary aquifer (PQ1 andPQ2) and groundwater of the Confined Miocene, are typically of SO4-Ca-Na-Mg water type. The main sources ofthe water mineralization in the Foussana basin are the water-rock interaction processes i.e. dissolution ofevaporates and cation-exchange process.The isotopic signatures permit to classify the studied groundwaters into different groups. The PlioQuaternaryaquifer groundwater was classified into two groups. These are the non-evaporated groundwater, which ischaracterized by depleted δ18O and δ2H contents highlighting the importance of modern recharge at higher altitude,and the mixing process with the deep aquifer of the Miocene. The evaporated groundwater that exhibits enrichedstable isotope contents, these enriched values could be related to the evaporation process, which occurs possiblyin the upper part of the unsaturated zone of the PlioQuaternary aquifer (depression area).The Miocene aquifersgroundwaters are classified into two groups. The first group is characterized by relatively depleted isotope contentscorresponding to outcropping Miocene in the border indicating the altitude effect. The second group isdistinguished by relatively depleted contents of stable isotopes corresponding to the confined Miocene reflectinga palaeoclimatic origin. Tritium data permit to identify recent groundwaters originating from a mixture ofcontemporaneous and post-nuclear recharge; and ancient groundwaters deriving from pre-nuclear recharge.Radiocarbon activities decrease from 80 pmc in the recharge area to less than 3 pmc in the confined aquiferproviding ages from present day to 30 000 years BP.In conclusion, a conceptual model has been produced to describe the functioning of the aquifers. The modelling isused as a tool to synthesise the data and the functioning hypothesis. It gives a visualization mean and can put inevidence the aspects, which should be deepened in next works.

Det borras en tunnel under Stockholm av Svenska kraftnät som ska bli 13,4 km lång. Närtunneln är klar ska den förstärka och ersätta vissa delar av Stockholms elnät. Grundvattennivånhar inte sjunkit lika mycket som förväntat när inläckget till tunneln var högt.När grundvatten sjunker är det var och hur mycket grundvattennivån sjunker som ärproblemet inte själva inläckaget. Syftet med rapporten var att undersöka verktyg för attprognostisera avsänkning av grundvatten samt försöka förstå vilka parametrar som är viktigaför att skapa en bra modell. Det undersöktes även varför grundvattennivån i moränenovanför bergtunneln sjunker över ett stort område. Det sattes upp en modell med fleralager i jorden och berget med olika konduktivitet. I berget lades det även in svagare zoneri form av krosszoner för att längs dem lokalt ge högre konduktivitet. Modellen gavs ävenflera områden med speciella egenskaper för att efterlikna verkligheten. Först modelleradesen grundvattenyta i en referensmodell som kalibrerades med hjälp av grundvattenröri området. Sedan lades tunneln in i modellen med hjälp av uppmätta inläckage för att beräknaen ny grundvattenyta. Grundvattenytan från referensmodellen jämfördes med denfrån modellen med tunneln för att få fram grundvattensänkningen. Modellen visade enstörre skillnad i grundvattennivå där det inte var krosszoner som korsade tunneln medandär den korsades var avsänkningen mer utspridd längs krosszonen.Det finns flera sätt att angripa ett problem. För att kunna använda analytiska ekvationerbehöver det göras stora förenklingar vilket kan vara bra i ett inledningsskede för att fåen bättre bild av problemet förhållandevis enkelt. För att få en mer detaljerad bild avproblemet behövs däremot andra modeller. I en modell behövs det många parametrar ochvissa kan vara viktigare än andra. I just det här arbetet bedömdes vissa delar inom geologinvara viktigast som var det finns mäktiga jordlager, den hydrauliska konduktivitetenför berget och var svaghetszonerna är i berget. Anledningen att grundvattennivån runtSvenska kraftnäts tunnel sjunker över ett stort område kan vara svaghetszonerna i berget. / Svenska kraftnät excavates a 13.4 km long tunnel in the rock below Stockholm. Whencompleted, it will strengthen and replace parts of Stockholm’s electricity grid. When theleakage to the tunnel was high, the groundwater level did not drop as much as expected.When the groundwater drops, it is where and how much that is the problem and notthe actual leakage to the tunnel. The purpose of the report was to investigate toolsfor forecasting groundwater subsidence and to try to understand which the importantparameters are for creating a reliable model. It was also investigated why the groundwaterlevel in the moraine above the rock tunnel drops over a large area. A model was set upwith several layers in the soil and rock with different conductivity. Weaker zones werealso added to the bedrock to provide higher conductivity locally along them. The modelwas also given several areas with special features to mimic reality. A groundwater surfacewas first modeled in a reference model that was calibrated using groundwater wells in thearea. The tunnel was then added to the model using measured leakage to calculate a newgroundwater surface. The groundwater surface from the reference model was comparedwith that from the model with the tunnel to obtain the groundwater lowering. The modelshowed a large local reduction of the groundwater level where there was more stable rock,while where weaker zones crossed the tunnel the reduction was more spread out along theweaker zone.There are several ways to adress problems with. To be able to use analytical equations,large simplifications are required, which can be good in an initial stage to get a betterpicture of the problem relatively easily. However, to get a more detailed view of theproblem, other models are required. In a model, many parameters are required and somemay be more important than others. In this report, certain parts of the geology werejudged to be most important, such as where there are thick soil layers, the conductivity ofthe rock and where the weaker zones are in the bedrock. The reason why the groundwaterlevel around Svenska kraftnät’s tunnel falls over a large area may be the zones of weaknessin the rock.

No description available.

No description available.

Esta pesquisa objetivou mostrar a viabilidade de utilização do Sistema de Informação Geográfica (SIG) associado à modelagem matemática, como opção metodológica a ser aplicada no estudo das modificações induzidas no nível freático, após o enchimento de reservatórios. A área piloto escolhida foi a bacia de drenagem do rio Biritiba-Mirim, localizada na região do Alto Tietê, no Estado de São Paulo. O entendimento do funcionamento do sistema hidrogeológico, obtido por meio de dados de geologia, hidrogeologia, climatologia, poços existentes e sondagens executadas na área, subsidiou a elaboração do modelo conceitual. Na área, foram identificados dois tipos de aqüíferos: o Sedimentar, restrito às planícies fluviais e o Cristalino alterado, decorrente da alteração intempérica da rocha. Os fluxos locais da água subterrânea ocorrem em direção aos cursos d’água que constituem a malha de drenagem da bacia, enquanto que, regionalmente, as águas subterrâneas fluem na direção do rio Biritiba-Mirim. As informações do modelo conceitual foram, inicialmente, organizadas no Sistema de Informação Geográfica e, posteriormente, transferidas para o modelo matemático. Foi construído um modelo matemático para a bacia de drenagem do rio Biritiba-Mirim utilizando o programa tridimensional de diferenças finitas MODFLOW desenvolvido por McDONALD & HARBAUGH (1988). A calibração do modelo foi executada a partir de ajustes sucessivos, efetuados por meio do SIG e, posteriormente, aferidos no MODFLOW. Uma vez concluído o processo de calibração, procedeu-se à simulação do enchimento do reservatório, utilizando novamente o Sistema de Informação Geográfica para a entrada dos dados. A simulação, também realizada no MODFLOW, possibilitou delimitar as áreas nas quais os níveis das águas subterrâneas sofrerão maiores influências do reservatório, dando o subsidio necessário para a locação dos poços de monitoramento. Os dados coletados no período de um ano de monitoramento permitiram executar a verificação e eventuais correções do modelo. Com o modelo aferido, repetiu-se a simulação do enchimento do reservatório e, a partir do mapa potenciométrico resultante, foi elaborado o mapa de profundidades de níveis d’água subterrânea previstas para após o enchimento do lago, denominado de Mapa Previsional de Influência. Nesse mapa foram delimitadas duas zonas de influência (de 0-5 m e de 5-10 m) e uma zona sem influência do reservatório ou com profundidade final do nível freático superior a 10 m. Finalmente, definiu-se um programa de monitoramento potenciométrico do sistema aqüífero livre para execução em situação posterior ao enchimento do reservatório, a fim de avaliar o tempo e tipo de resposta do aqüífero livre ao processo de enchimento e efetuar a retroanálise das previsões. De acordo com programa definido, a primeira campanha de medição deverá coincidir com o início do enchimento do reservatório, a partir desta, o monitoramento deverá ser mensal até os três primeiros meses e, posteriormente, trimestral até completar um ano após o enchimento. A execução do presente estudo comprovou a viabilidade da aplicação do SIG, associado à modelagem matemática com fins previsionais, em escala de bacia de contribuição, para a avaliação das modificações induzidas no nível do aqüífero livre após o enchimento de reservatórios. Recomenda-se a metodologia aqui proposta como ferramenta a ser aplicada em outros locais que apresentem a mesma problemática. / The purpose of the current research was to demonstrate the viability of using Geographic Information System – GIS associated to mathematical models as a methodological option to be applied in the study of the water table elevation induced by dam reservoirs after their impoundment. The chosen area was the drainage basin of Biritiba-Mirim river, located in Alto Tietê area, in São Paulo State. The understanding of the hydrogeological system was acquired through geological, hydrogeological and climatological data. The surveys executed in the area, also allowed the construction of the conceptual model. In the area, two aquifers types were identified: the Sedimentary, restricted to the fluvial plains, and the altered Crystalline, due to the weathering. The regional groundwater flow was toward Biritiba-Mirim river, while the secondary drainage was a local groundwater discharge area. The conceptual model data was, initially, organized in a Geographic Information System and later transferred to the mathematical model. A mathematical model was built on the Biritiba-Mirim drainage basin using MODFLOW, a three-dimensional program of finite differences developed by McDONALD & HARBAUGH (1988). The model calibration was executed through successive trial-error process, using Geographic Information System as a input data and MODFLOW to evaluate results. After the conclution of the calibration process, the Biritiba-Mirim reservoir impoundment was simulated using Geographic Information System to input the new variable. The simulation, also accomplished in MODFLOW, made possible to limitate the main areas that will be impacted on their ground water levels. Using that map were located the monitoring wels. From the monitoring data, collected in a year period, it was possible to verify the model. The reservoir impoundment simulation was repeated. From the resulted potenciometric map, the groundwater final depths map was elaborated. In that map, named Predictional Influence Map, two influence zones were defined (0-5 m and 5-10 m), and a zone without the reservoir influence or with final water table level more than 10 m depth. Finally, an aquifer monitoring program was defined, in order to follow the water table impact after the reservoir impoundment. According to this program, two weeks after the impoundment should be made the first field measurement. The periodicity should be monthly until the first three months, and, after that, quarterly until the end of the first year. The present study showed the viability of using Geographic Information System associated to mathematical models to predict modifications induced on the water table levels after a reservoir impoundment. This methodology is recommended as an important tool to be applied in other cases with the same problem.

Esta pesquisa objetivou mostrar a viabilidade de utilização do Sistema de Informação Geográfica (SIG) associado à modelagem matemática, como opção metodológica a ser aplicada no estudo das modificações induzidas no nível freático, após o enchimento de reservatórios. A área piloto escolhida foi a bacia de drenagem do rio Biritiba-Mirim, localizada na região do Alto Tietê, no Estado de São Paulo. O entendimento do funcionamento do sistema hidrogeológico, obtido por meio de dados de geologia, hidrogeologia, climatologia, poços existentes e sondagens executadas na área, subsidiou a elaboração do modelo conceitual. Na área, foram identificados dois tipos de aqüíferos: o Sedimentar, restrito às planícies fluviais e o Cristalino alterado, decorrente da alteração intempérica da rocha. Os fluxos locais da água subterrânea ocorrem em direção aos cursos d’água que constituem a malha de drenagem da bacia, enquanto que, regionalmente, as águas subterrâneas fluem na direção do rio Biritiba-Mirim. As informações do modelo conceitual foram, inicialmente, organizadas no Sistema de Informação Geográfica e, posteriormente, transferidas para o modelo matemático. Foi construído um modelo matemático para a bacia de drenagem do rio Biritiba-Mirim utilizando o programa tridimensional de diferenças finitas MODFLOW desenvolvido por McDONALD & HARBAUGH (1988). A calibração do modelo foi executada a partir de ajustes sucessivos, efetuados por meio do SIG e, posteriormente, aferidos no MODFLOW. Uma vez concluído o processo de calibração, procedeu-se à simulação do enchimento do reservatório, utilizando novamente o Sistema de Informação Geográfica para a entrada dos dados. A simulação, também realizada no MODFLOW, possibilitou delimitar as áreas nas quais os níveis das águas subterrâneas sofrerão maiores influências do reservatório, dando o subsidio necessário para a locação dos poços de monitoramento. Os dados coletados no período de um ano de monitoramento permitiram executar a verificação e eventuais correções do modelo. Com o modelo aferido, repetiu-se a simulação do enchimento do reservatório e, a partir do mapa potenciométrico resultante, foi elaborado o mapa de profundidades de níveis d’água subterrânea previstas para após o enchimento do lago, denominado de Mapa Previsional de Influência. Nesse mapa foram delimitadas duas zonas de influência (de 0-5 m e de 5-10 m) e uma zona sem influência do reservatório ou com profundidade final do nível freático superior a 10 m. Finalmente, definiu-se um programa de monitoramento potenciométrico do sistema aqüífero livre para execução em situação posterior ao enchimento do reservatório, a fim de avaliar o tempo e tipo de resposta do aqüífero livre ao processo de enchimento e efetuar a retroanálise das previsões. De acordo com programa definido, a primeira campanha de medição deverá coincidir com o início do enchimento do reservatório, a partir desta, o monitoramento deverá ser mensal até os três primeiros meses e, posteriormente, trimestral até completar um ano após o enchimento. A execução do presente estudo comprovou a viabilidade da aplicação do SIG, associado à modelagem matemática com fins previsionais, em escala de bacia de contribuição, para a avaliação das modificações induzidas no nível do aqüífero livre após o enchimento de reservatórios. Recomenda-se a metodologia aqui proposta como ferramenta a ser aplicada em outros locais que apresentem a mesma problemática. / The purpose of the current research was to demonstrate the viability of using Geographic Information System – GIS associated to mathematical models as a methodological option to be applied in the study of the water table elevation induced by dam reservoirs after their impoundment. The chosen area was the drainage basin of Biritiba-Mirim river, located in Alto Tietê area, in São Paulo State. The understanding of the hydrogeological system was acquired through geological, hydrogeological and climatological data. The surveys executed in the area, also allowed the construction of the conceptual model. In the area, two aquifers types were identified: the Sedimentary, restricted to the fluvial plains, and the altered Crystalline, due to the weathering. The regional groundwater flow was toward Biritiba-Mirim river, while the secondary drainage was a local groundwater discharge area. The conceptual model data was, initially, organized in a Geographic Information System and later transferred to the mathematical model. A mathematical model was built on the Biritiba-Mirim drainage basin using MODFLOW, a three-dimensional program of finite differences developed by McDONALD & HARBAUGH (1988). The model calibration was executed through successive trial-error process, using Geographic Information System as a input data and MODFLOW to evaluate results. After the conclution of the calibration process, the Biritiba-Mirim reservoir impoundment was simulated using Geographic Information System to input the new variable. The simulation, also accomplished in MODFLOW, made possible to limitate the main areas that will be impacted on their ground water levels. Using that map were located the monitoring wels. From the monitoring data, collected in a year period, it was possible to verify the model. The reservoir impoundment simulation was repeated. From the resulted potenciometric map, the groundwater final depths map was elaborated. In that map, named Predictional Influence Map, two influence zones were defined (0-5 m and 5-10 m), and a zone without the reservoir influence or with final water table level more than 10 m depth. Finally, an aquifer monitoring program was defined, in order to follow the water table impact after the reservoir impoundment. According to this program, two weeks after the impoundment should be made the first field measurement. The periodicity should be monthly until the first three months, and, after that, quarterly until the end of the first year. The present study showed the viability of using Geographic Information System associated to mathematical models to predict modifications induced on the water table levels after a reservoir impoundment. This methodology is recommended as an important tool to be applied in other cases with the same problem.