Relating groundwater to seasonal wetlands in southeastern Wisconsin, USA

Historically, drier types of wetlands have been difficult to characterize and are not well researched. Nonetheless, they are considered to reflect the precipitation history with little, if any, regard for possible relation to groundwater. Two seasonal coastal wetland types (wet prairie, sedge meadow) were investigated during three growing seasons at three sites in the Lake Michigan Basin, Wisconsin, USA. The six seasonal wetlands were characterized using standard soil and vegetation techniques and groundwater measurements from the shallow and deep systems. They all met wetland hydrology criteria (e.g., water within 30 cm of land surface for 5% of the growing season) during the early portion of the growing season despite the lack of appreciable regional groundwater discharge into the wetland root zones. Although root-zone duration analyses did not fit a lognormal distribution previously noted in groundwater-dominated wetlands, they were able to discriminate between the plant communities and showed that wet prairie communities had shorter durations of continuous soil saturation than sedge meadow communities. These results demonstrate that the relative rates of groundwater outflows can be important for wetland hydrology and resulting wetland type. Thus, regional stresses to the shallow groundwater system such as pumping or low Great Lake levels can be expected to affect even drier wetland types.     

Analysis of base flow trends in urban streams, northeastern Illinois, USA

Statistical analysis shows that mean annual base flows in three unregulated urban streams in northeastern Illinois do not display significant trends during periods of substantial watershed urbanization. However, statistically-significant upward trends in median annual base flow (probably emblematic of overall changes in the time-distribution of the base flow), characterized by increases in lower base flow rates, affected the flows in all three streams. The analysis reveals no trends in annual and monthly precipitation during these periods. These results are in contrast to results from studies of other urban streams that have shown decreases in base flow, but these results may be partially explained by the low permeability of the near-surface materials in the watersheds investigated. This study employs formal hypothesis-testing of Kendall tau-a trend statistics computed for monthly and annual base flow and precipitation rates to assess the impact of urban development on base flow rates.

---

Resumen El análisis estadístico muestra que el flujo base anual medio en tres arroyos urbanos no regulados en el noreste de Illinois no ostenta tendencias significativas durante los periodos de urbanización sustancial de la cuenca. Sin embargo, el flujo en los tres arroyos fue afectado por tendencias al incremento en el flujo base annual medio. Esto es probablemente emblemático en cambios globales en el tiempo-distribución del flujo base caracterizado por incrementos en las tasas de flujo base bajo y afecta a los tres arroyos. El análisis no revela ninguna tendencia en la precipitación annual y mensual durante estos periodos. Estos resultados contrastan con los resultados de estudios de otros arroyos urbanos que muestran un decrecimiento en el fujo base. Sin embargo, estos resultados pueden explicarse parcialmente por la baja permeabilidad de los materiales cercanos a la superficie en las cuencas investigadas. Este estudio emplea las estadísticas de tendencia tau-a de la hipótesis de prueba formal de Kendall - calculadas según los flujos base mensuales y anuales y las tasas de precipitación. El objetivo es evaluar los impactos del desarrollo urbano sobre las tasas de flujo base.

---

Résumé Une analyse statistique démontre que la moyenne annuelle de lécoulement de base dans trois rivières urbaines non réglementées du nord est de lIllinois ne montre pas de tendance significative lors de périodes durbanisation massive du bassin versant. Par contre, une tendance statistiquement significative à la hausse de lécoulement de base médian, probablement représentatif de lensemble des changements dans la distribution temporelle de lécoulement de base caractérisé par une augmentation des faibles taux découlement de base, affecte lécoulement dans les trois rivières. Lanalyse ne révèle pas de tendance dans les précipitations annuelles et mensuelles lors de ces périodes. Ces résultats contrastent avec les résultats dautres études de rivières en milieu urbain qui ont montré une diminution de lécoulement de base. Les résultats obtenus lors de cette étude peuvent partiellement être expliqués par la faible perméabilité des dépôts de surface dans le bassin versant étudié. Cette étude utilise le test dhypothèse tau de Kandall, une tendance statistique calculée pour lécoulement de base annuel et mensuel et des taux de précipitation dans le but de déterminer limpact du développement urbain sur les taux découlement de base.

     

Petrology,geochronology,and geophysical characterization of Mesoproterozoic rocks in central Illinois,USA

The Precambrian basement rocks of the Eastern Granite-Rhyolite Province(EGRP)in central Illinois(midcontinent region of North America)exhibit a complex history of early volcanism,granite emplacement,and intrusion of mafic rocks.A comprehensive suite of dedicated petrographic analyses,geophysical logs,and drill core from four basement-penetrating wells,two-dimensional and three-dimensional seismic reflection data,and U-Pb age data from the Illinois Basin–Decatur Project(IBDP)and Illinois Carbon Capture Storage(ICCS)Project site provide new constraints for interpreting the Precambrian basement of the Illinois Basin.These new data reveal the basement to be compositionally and structurally complex,having typical EGRP felsic volcanic rocks intruded by the first reported gabbro in the Precambrian basement in Illinois.Zircons(n?29)from rhyolite give a U-Pb weighted mean average age of 14679 Ma.Zircons(n?3)from a gabbro dike that intrudes the rhyolite yield a concordia age of 107312 Ma,which corresponds to Grenville-age extension and represents the first Grenville-age rock in Illinois and in the EGRP.A high-resolution three-dimensional seismic reflection volume,coincident with the four wells,provides a context for interpreting the petrological data and implies a high degree of heterogeneity for basement rocks at the IBDP–ICCS site,as also shown by the drill cores.The occurrence of Grenville-age gabbro is related to a prominent bowl-like structure observed on local two-dimensional seismic reflection profiles and the three-dimensional volume that is interpreted as a deep-seated mafic sill complex.Furthermore,heterogeneities such as the brecciated EGRP rhyolite and later gabbro intrusion observed in the basement lithology at the IBDP–ICCS may reflect previously unknown distal elements of the 1.1 Ga Midcontinent Rift in the EGRP and more likely Grenville-age extension.     

Heat transport in a coastal groundwater flow system near De Panne, Belgium

Temperature distribution and heat transport are studied in a coastal aquifer at De Panne in the western Belgian coastal plain. Field observations include temperature profiles of groundwater in the dunes and temperature measurements at the water table in a profile on the shore. Freshwater–saltwater distribution is known from previous studies. These are used to constrain a density-dependent model simulating the freshwater–saltwater distribution and heat transport using the SEAWAT code. The yearly fluctuation of the groundwater temperature in the phreatic aquifer under the dunes, shore and sea, and the influence of a tidal inlet in the dunes are simulated. The observations show that seawater temperature variations determine the temperature variations on the shore whereas atmospheric temperature changes determine this in the dunes. Yearly temperature fluctuations imposed at the water table propagate mainly vertically in the aquifer with only limited lateral influence. Heat transport is mainly convection dominated. Thickness of the surficial zone is determined by the amplitude of the groundwater temperature at the water table and the groundwater flow. Establishment of a tidal inlet in the dunes results in asymmetric temperature profiles under and in the vicinity of it.     

Noble gas tracing of groundwater/coalbed methane interaction in the San Juan Basin, USA

The San Juan Basin natural gas field, located in northwestern New Mexico and southwestern Colorado in the USA, is a case-type coalbed methane system. Groundwater is thought to play a key role in both biogenic methane generation and the CO₂ sequestration potential of coalbed systems. We show here how noble gases can be used to construct a physical model that describes the interaction between the groundwater system and the produced gas. We collected 28 gas samples from producing wells in the artesian overpressured high production region of the basin together with 8 gas samples from the underpressured low production zone as a control. Stable isotope and major species determination clearly characterize the gas in the high production region as dominantly biogenic in origin, and the underpressured low producing region as having a significant admix of thermogenic coal gas. ³He/⁴He ratios increase from 0.0836R_a at the basin margin to 0.318R_a towards the center, indicating a clear but small mantle He signature in all gases. Coherent fractionation of water-derived ²⁰Ne/³⁶Ar and crustal ⁴He/⁴⁰Ar* are explained by a simple Rayleigh fractionation model of open system groundwater degassing. Low ²⁰Ne concentrations compared to the model predicted values are accounted for by dilution of the groundwater-associated gas by desorbed coalbed methane. This Rayleigh fractionation and dilution model together with the gas production history allows us to quantify the amount of water involved in gas production at each well. The quantified water volumes in both underpressured and overpressured zones range from 1.7 × 10³ m³ to 4.2 × 10⁵ m³, with no clear distinction between over- and underpressured production zones. These results conclusively show that the volume of groundwater seen by coal does not play a role in determining the volume of methane produced by secondary biodegradation of these coalbeds. There is no requirement of continuous groundwater flow for renewing the microbes or nutrient components. We furthermore observe strong mass related isotopic fractionation of ²⁰Ne/²²Ne and ³⁸Ar/³⁶Ar isotopic ratios. This can be explained by a noble gas concentration gradient in the groundwater during gas production, which causes diffusive partial re-equilibration of the noble gas isotopes. It is important for the study of other systems in which extensive groundwater degassing may have occurred to recognize that severe isotopic fractionation of air-derived noble gases can occur when such concentration gradients are established during gas production. Excess air-derived Xe and Kr in our samples are shown to be related to the diluting coalbed methane and can only be accounted for if Xe and Kr are preferentially and volumetrically trapped within the coal matrix and released during biodegradation to form CH₄.     

Hydrogeochemistry and isotope analyses used to determine groundwater recharge and flow in low-gradient catchments of the province of Buenos Aires,Argentina

The region known as Pampa Plain, in Argentina, is a vast area characterized by slopes of less than 0.05%. The surface sediment is silty sand, mainly Aeolian, referred to as Pampean Loess, which is a phreatic aquifer unit of utmost importance for the water supply of the region. On account of the slight gradient, hydrogeological analyses using only hydraulic measurements are difficult to perform, often leading to confusing results. Thus, the study presented relies on hydrochemical modeling and isotopic determinations as well. The study area comprises three catchments in the inter-mountainous area corresponding to El Moro, Tamangueyú and Seco creeks, and covering 2,570 km² in the province of Buenos Aires. Measurements of piezometric levels and samples of groundwater, surface water and rainwater were carried out between January and March 2005. Major ion results were analyzed by means of hydrochemical graphs and hydrogeochemical modelling using NETPATH. The resulting data show that it is possible to identify local changes in recharge, flow direction and stream/groundwater relationship by using hydrochemical and isotopic information, which may become a useful and more precise tool for the study of particularly flat landscapes.     

Coal quality controls of the Danville Coal in Indiana (Illinois Basin, central USA)

The Danville Coal Member (Dugger Formation, upper Desmoinesian, Pennsylvanian) is a significant economic coal resource in the Illinois Basin, central USA. Deposition of the Danville Coal (peat) was in coastal environments, varying distances from the coastline and, in turn, variable influences from saline waters. The purpose of this study is to examine the coal quality and petrography of the Danville Coal; and to discuss their relationship with depositional environment as it relates to the final coal product. A medium sulfur (1.0–1.5 wt.%) Danville Coal reserve area (northern Indiana coalfield) was compared to a low sulfur (<1.0 wt.%) Danville Coal (central Indiana coalfield) reserve area, the two being approximately 70 km apart. The medium sulfur coal resulted from the peat being deposited in a near-marine environment less protected from the influence of saline waters, whereas the low sulfur coal resulted from fine-grained, clay-dominated sediment protecting the peat from the direct influence of saline waters. Within both areas, the coal quality, coal composition, and trace element concentrations vary as a function of the proximity of the coal to the overlying Busseron Sandstone Member (Pennsylvanian). Where the Busseron Sandstone rests near or directly on the coal, the sulfur content is significantly higher in the top third of the seam. Conversely, where there is a thick section (>3 m) of finer-grained clastic sediments atop the Danville, the sulfur and trace elements contents are significantly lower.     

Modeling of groundwater flow at depth in crystalline rock beneath a moving ice-sheet margin, exemplified by the Fennoscandian Shield, Sweden

On-going geological disposal programs for spent nuclear fuel have generated strong demands for investigation and characterization of deep-lying groundwater systems. Because of the long time scales for which radiological safety needs to be demonstrated in safety assessment applications, an analysis of the hydrogeological performance of the geosphere system during glacial climate conditions is needed. Groundwater flow at depth in crystalline rock during the passage of an ice-sheet margin is discussed based on performed groundwater-flow modeling of two bedrock sites, Forsmark and Laxemar, in the Fennoscandian Shield, Sweden. The modeled ice sheet mimics the Weichselian ice sheet during its last major advance and retreat over northern Europe. The paper elaborates and analyzes different choices of top boundary conditions at the ice sheet–subsurface interface (e.g. ice-sheet thickness and ice-margin velocity) and in the proglacial area (presence or lack of permafrost) and relates these choices to available groundwater-flow-model hydraulic output and prevailing conceptual hydrogeochemical models of the salinity evolution at the two sites. It is concluded that the choice of boundary conditions has a strong impact on results and that the studied sites behave differently for identical boundary conditions due to differences in their structural-hydraulic properties.     

Fracture flow and groundwater compartmentalization in the Rollins Sandstone, Lower Mesaverde Group, Colorado, USA

 This paper presents a site-specific conceptual model of groundwater flow in fractured damage zones associated with faulting in a package of sedimentary rocks. The model is based on the results of field and laboratory investigations. Groundwater and methane gas inflows from fault-fracture systems in the West Elk coal mine, Colorado, USA, have occurred with increasing severity. Inflows of 6, 160 and 500 L s⁻¹ discharged almost instantaneously from three separate faults encountered in mine workings about 460 m below ground level. The faults are about 600 m apart. The δ ²H and δ ¹⁸O compositions of the fault-related inflow waters and the hydrodynamic responses of each fault inflow indicate that the groundwaters discharge from hydraulically isolated systems. ¹⁴C data indicate that the groundwaters are as much as 10,500 years old. Discharge temperatures are geothermal (≈30°C), which could indicate upwelling from depth. However, calculations of geothermal gradients, analysis of solute compositions of groundwater in potential host reservoirs, geothermometer calculations, and results of packer testing indicate that the fractured groundwater reservoir is the Rollins Sandstone (120 m thick) directly beneath the coal seams. The packer test also demonstrates that the methane gas is contained in the coal seams. A geothermal gradient of 70–80°C km⁻¹, related to an underlying intrusion, is probably responsible for the slightly elevated discharge temperatures. Large discharge volumes, as great as 8.2×10⁵m³ from the 14 South East Headgate fault (14 SEHG), rapid declines in discharge rates, and vertical and horizontal permeability (matrix permeability generally <0.006 Darcy) indicate fracture flow. An in-mine pumping test demonstrates that the 14 SEHG fault has excellent hydraulic communication with fractures 50 m from the fault. Aeromagnetic data indicate that the faults are tectonically related to an igneous body that is several thousand meters below the coal seams. Exploratory drilling has confirmed a fourth fault, and two additional faults are projected, based on the aeromagnetic data. The conceptual model describes a series of parallel, hydraulically separate groundwater systems associated with fault-specific damage zones. The faults are about 600 m apart. Groundwater stored in fractured sandstone is confined above and below by clayey layers. Received March 1999 / Revised, November 1999 / Accepted, December 1999     

Internal architecture of a till sheet,Tiskilwa Formation,north‐central Illinois,USA: Application of architectural element analysis

Thick till sheets deposited during the Quaternary form significant aquitards in many areas of North America. However, the detailed sedimentary heterogeneity and architecture and depositional history of till units are not well understood. This study utilizes architectural element analysis to delineate the internal sedimentary architecture of the Tiskilwa Formation exposed at two outcrop sections in north‐central Illinois, USA. Architectural element analysis facilitates systematic delineation of sedimentary architecture based on the nature of facies contacts and change in facies associations, delineation of unit geometries and understanding of depositional processes at different scales of resolution; making architectural element analysis suitable for the sedimentological analysis and palaeoenvironmental reconstruction of subglacial deposits. Eleven facies types are identified in this study, including sand, gravel and diamict facies that record a suite of subglacial depositional processes. Detailed analysis of facies contacts (bounding surface hierarchy) and change in facies associations allows the delineation of five architectural elements, including coarse‐grained lens, coarse‐grained sheet, mixed zone, diamict lens and diamict sheet elements. The spatial arrangement and genetic interpretation of elements, and their spatial relationship with fifth‐order bounding surfaces, allows the delineation of five larger scale architectural units (‘element associations’), which can be mapped in the local study area and record at least three stacked successions of meltwater accumulation and till deposition. The results of this study can be utilized for architectural analysis of till sheets and provide insight to groundwater flow pathways through till in the study area and elsewhere.     

Heterogeneous redox conditions, arsenic mobility, and groundwater flow in a fractured-rock aquifer near a waste repository site in New Hampshire, USA

Anthropogenic sources of carbon from landfill or waste leachate can promote reductive dissolution of in situ arsenic (As) and enhance the mobility of As in groundwater. Groundwater from residential-supply wells in a fractured crystalline-rock aquifer adjacent to a Superfund site in Raymond, New Hampshire, USA, showed evidence of locally enhanced As mobilization in relatively reducing (mixed oxic-anoxic to anoxic) conditions as determined by redox classification and other lines of evidence. Redox classification was determined from geochemical indicators based on threshold concentrations of dissolved oxygen (DO), nitrate (NO ₃ ^– ), iron (Fe²⁺), manganese (Mn²⁺), and sulfate (SO ₄ ^2– ). Redox conditions were evaluated also based on methane (CH₄), excess nitrogen gas (N₂) from denitrification, the oxidation state of dissolved As speciation (As(III) and As(V)), and several stable isotope ratios. Samples from the residential-supply wells primarily exhibit mixed redox conditions, as most have long open boreholes (typically 50–100?m) that receive water from multiple discrete fractures with contrasting groundwater chemistry and redox conditions. The methods employed in this study can be used at other sites to gauge redox conditions and the potential for As mobilization in complex fractured crystalline-rock aquifers where multiple lines of evidence are likely needed to understand As occurrence, mobility, and transport.     

Geologic structure and groundwater flow in the Pachuca–Zumpango sub-basin,central Mexico

This paper is a hydrogeologic characterisation of the Pachuca-Zumpango sub-basin in Central Mexico, mainly concerned with the shallow geologic structure of the region and its implications in groundwater flow. The structure is interpreted as a series of graben structures limited by normal faults of metrical surface displacement. The graben are filled by a sequence of approximately 2,000 m alluvial and pyroclastic material and lava flows. An upper-aquifer that consists of Quaternary-Tertiary lava flows and clastic material is intensively used to obtain water for municipal supply. Abstraction was carried out with no control or management policy and resulted in severe environmental impacts. Numerical modelling simulation scenarios with continuous increments in the abstraction rate of 0.660 m³/s in 5-year periods between 2000 and 2020 predict an overall increase in depth to potentiometric surface of 15 m (0.80 m/year). Soil consolidation computation in terms of calculated drawdown values was not an objective in this study; however, lowering of the potentiometric surface will eventually control soil subsidence, as observed elsewhere in Mexico City.     

Microbial and chemical characterization of a groundwater flow system in an intermontane basin of southern New Mexico,USA

Groundwater of the southern Jornada del Muerto Basin, an intermontane basin structure associated with the Rio Grande rift located in south-central New Mexico, USA, was analyzed chemically and microbially. A microbial phospholipid fatty acids (PLFA) analysis revealed a sparse microbial population consisting of relatively simple microorganisms with no major population changes along the flow system. A nucleic acid (DNA) analysis of the groundwater resulted in the identification of ten eubacterial and one archeal species. Chemical analyses revealed that sulfate along with calcium, magnesium, iron, and manganese is removed by about an order of magnitude in concentration from the recharge area to the discharge area. The removal of iron, manganese, magnesium, and to some extent calcium can be explained by oxidation reactions and the precipitation of dolomite. Sulfate and additional calcium are most likely removed by the precipitation of gypsum. Thiobacillus spp. are oxidizing metal sulfides that occur as subsurface sulfide mineral deposits to sulfuric acid, which subsequently reacts with calcium carbonate and water to precipitate gypsum. The presence of these sulfide deposits exposed to oxygenated water in the deep groundwater flow system significantly alters its chemical and bacteriological composition. Electronic Publication     

美国Sand Hills地区地下水数值模拟及水量平衡分析

利用地下水数值模型MODFLOW和非饱和带水平衡模型对处于半干旱半湿润沙丘地区(Sand Hills)地下水位进行了模拟,并分析了含水层补排水量,河流与地下水补排关系,以及区域水平衡过程。揭示了独特沙丘地形和土壤特性对地下水补排量的影响。模拟结果表明,入渗率大、非饱和带厚的沙丘有利于降水入渗补给,减少了地下水蒸散发损失。加上下覆含水层具有良好的地下水储水空间,是该地区储存丰富的地下水量,以维持河流稳定流量,供给众多湖泊和湿地的原因。该研究对我国地下水资源评价和生态环境脆弱地区水资源保护具有指导意义。     

Why conceptual groundwater flow models matter: a trans-boundary example from the arid Great Basin, western USA

Spring and Snake valleys, western USA, are scheduled for development and groundwater export to Las Vegas, Nevada (USA). New work, compared to published studies, illustrates the critical role of conceptual models to underpin water withdrawals in arid regions. Interbasin flow studies suggest that 30–55?% of recharge to Snake Valley arrives from adjacent Spring Valley. This study, however, suggest little or no interbasin flow; rather, Spring and Snake valleys comprise separate systems. Contrary to expectation, δD and δ¹⁸O contours are perpendicular to proposed interbasin flow paths. ¹⁴C age gradients up to 10?ka along interbasin flow paths indicate that old waters are not displaced by such fluxes. ¹⁴C and ³H patterns indicate local recharge occurs in adjacent mountain ranges and is transferred to basin-fill by losing streams, mountain front recharge, and upward leakage from carbonate bedrock beneath basins. The choice of conceptual models is critical for groundwater development. Simple analyses of water withdrawals indicate that monitoring discharges at desert springs is an inadequate protective measure. Once flows decline, recovery is lengthy even if pumping is stopped. The conceptual framework behind quantitative evaluations of sustainable yield is critical to determine the ability of a groundwater system to deliver sustained withdrawals.     

Regional groundwater flow in an area mapped as continuous permafrost, NE Alaska (USA)

Fundamental knowledge of groundwater systems in areas of permafrost is often lacking. The likelihood of finding good quality groundwater resources of acceptable quantities generally decreases as the areal coverage of permafrost increases. In areas of continuous permafrost, the probability of finding areas of groundwater recharge and discharge are minimal. Still, in northeastern Alaska (USA), the presence of numerous springs and associated downstream aufeis formations clearly indicates that there has to be a groundwater system with the required complementary areas of groundwater recharge and transmission. Recharge zones and transmission pathways in this area of extensive permafrost, however, are essentially unknown. This study shows that the recharge occurs on the south side of the Brooks Range in northeastern Alaska, where extensive limestone outcrops are found. The transmission zone is beneath the permafrost, with discharge occurring through the springs via taliks through the permafrost (where faults are present) and also likely at the northern edge of the permafrost along the Beaufort Sea coast.     

Groundwater response to serial stream stage fluctuations in shallow unconfined alluvial aquifers along a regulated stream (West Virginia,USA)

Groundwater response to stream stage fluctuations was studied in two unconfined alluvial aquifers using a year-long time series of stream stages from two pools along a regulated stream in West Virginia, USA. The purpose was to analyze spatial and temporal variations in groundwater/surface-water interaction and to estimate induced infiltration rate and cumulative bank storage during an annual cycle of stream stage fluctuation. A convolution-integral method was used to simulate aquifer head at different distances from the stream caused by stream stage fluctuations and to estimate fluxes across the stream–aquifer boundary. Aquifer diffusivities were estimated by wiggle-matching time and amplitude of modeled response to multiple observed storm events. The peak lag time between observed stream and aquifer stage peaks ranged between 14 and 95 hour. Transient modeled diffusivity ranged from 1,000 to 7,500 m²/day and deviated from the measured and calculated single-peak stage-ratio diffusivity by 14–82 %. Stream stage fluctuation displayed more primary control over groundwater levels than recharge, especially during high-flow periods. Dam operations locally altered groundwater flow paths and velocity. The aquifer is more prone to surface-water control in the upper reaches of the pools where stream stage fluctuations are more pronounced than in the lower reaches. This method could be a useful tool for quick assessment of induced infiltration rate and bank storage related to contamination investigations or well-field management.     

Evidence for deep groundwater flow and convective heat transport in mountainous terrain, Delta County, Colorado, USA

The Tongue Creek watershed lies on the south flank of Grand Mesa in western Colorado, USA and is a site with 1.5 km of topographic relief, heat flow of 100 mW/m², thermal conductivity of 3.3 W m^–1 °C^–1, hydraulic conductivity of 10^-8 m/s, a water table that closely follows surface topography, and groundwater temperatures 3–15°C above mean surface temperatures. These data suggest that convective heat transport by groundwater flow has modified the thermal regime of the site. Steady state three-dimensional numerical simulations of heat flow, groundwater flow, and convective transport were used to model these thermal and hydrological data. The simulations provided estimates for the scale of hydraulic conductivity and bedrock base flow discharge within the watershed. The numerical models show that (1) complex three-dimensional flow systems develop with a range of scales from tens of meters to tens of kilometers; (2) mapped springs are frequently found at locations where contours of hydraulic head indicate strong vertical flow at the water table, and; (3) the distribution of groundwater temperatures in water wells as a function of surface elevation is predicted by the model.