Geology Reports⌕ Search

Geology topics

Keith J. Halford

Publications and source records attributed to Keith J. Halford.

At least 37 records · Page 2Linked to original sources
USGS2009

Revised multi-node well (MNW2) package for MODFLOW ground-water flow model

Wells that are open to multiple aquifers can provide preferential pathways to flow and solute transport that short-circuit normal fluid flowlines. Representing these features in a regional flow model can produce a more realistic and reliable simulation model. This report describes modifications to the Multi-Node Well (MNW) Package of the U.S. Geological Survey (USGS) three-dimensional ground-water flow model (MODFLOW). The modifications build on a previous version and add several new features, processes, and input and output options. The input structure of the revised MNW (MNW2) is more well-centered than the original verion of MNW (MNW1) and allows the user to easily define hydraulic characteristics of each multi-node well. MNW2 also allows calculations of additional head changes due to partial penetration effects, flow into a borehole through a seepage face, changes in well discharge related to changes in lift for a given pump, and intraborehole flows with a pump intake located at any specified depth within the well. MNW2 also offers an improved capability to simulate nonvertical wells. A new output option allows selected multi-node wells to be designated as 'observation wells' for which changes in selected variables with time will be written to separate output files to facilitate postprocessing. MNW2 is compatible with the MODFLOW-2000 and MODFLOW-2005 versions of MODFLOW and with the version of MODFLOW that includes the Ground-Water Transport process (MODFLOW-GWT).

Techniques and Methods↗
USGS2008

Permeameter data verify new turbulence process for MODFLOW

A sample of Key Largo Limestone from southern Florida exhibited turbulent flow behavior along three orthogonal axes as reported in recently published permeameter experiments. The limestone sample was a cube measuring 0.2 m on edge. The published nonlinear relation between hydraulic gradient and discharge was simulated using the turbulent flow approximation applied in the Conduit Flow Process (CFP) for MODFLOW-2005 mode 2, CFPM2. The good agreement between the experimental data and the simulated results verifies the utility of the approach used to simulate the effects of turbulent flow on head distributions and flux in the CFPM2 module of MODFLOW-2005.

Ground Water↗
USGS2008

Source, Distribution, and Management of Arsenic in Water from Wells, Eastern San Joaquin Ground-Water Subbasin, California

Between 1974 and 2001 water from as many as one-third of wells in the Eastern San Joaquin Ground Water Subbasin, about 80 miles east of San Francisco, had arsenic concentrations greater than the U.S. Environmental Protection Agency Maximum Contaminant Level (MCL) for arsenic of 10 micrograms per liter (ug/L). Water from some wells had arsenic concentrations greater than 60 ug/L. The sources of arsenic in the study area include (1) weathering of arsenic bearing minerals, (2) desorption of arsenic associated with iron and manganese oxide coatings on the surfaces of mineral grains at pH's greater than 7.6, and (3) release of arsenic through reductive dissolution of iron and manganese oxide coatings in the absence of oxygen. Reductive dissolution is responsible for arsenic concentrations greater than the MCL. The distribution of arsenic varied areally and with depth. Concentrations were lower near ground-water recharge areas along the foothills of the Sierra Nevada; whereas, concentrations were higher in deeper wells at the downgradient end of long flow paths near the margin of the San Joaquin Delta (fig. 1). Management opportunities to control high arsenic concentrations are present because water from the surface discharge of wells is a mixture of water from the different depths penetrated by wells. On the basis of well-bore flow and depth-dependent water-quality data collected as part of this study, the screened interval of a public-supply well having arsenic concentrations that occasionally exceed the MCL was modified to reduce arsenic concentrations in the surface discharge of the well. Arsenic concentrations from the modified well were about 7 ug/L. Simulations of ground-water flow to the well showed that although upward movement of high-arsenic water from depth within the aquifer occurred, arsenic concentrations from the well are expected to remain below the MCL.

Open-File Report↗
USGS2008

Technical Review of Water-Resources Investigations of the Tule Desert, Lincoln County, Southern Nevada

The Nevada State Engineer in Ruling No. 5181 required Lincoln County and Vidler Water Company, Inc., to provide results from additional water-resources studies of Tule Desert in southern Nevada to support water-rights application 64692. As outlined by the ruling, the additional studies were to include the determination of the amount of ground water available from the Tule Desert basin, ground-water recharge to the Tule Desert, and the direction of ground-water flow. Results of these additional studies were published in five reports prepared for Lincoln County and Vidler Water Company, Inc. The National Park Service formally requested that the U.S. Geological Survey provide technical reviews of these five reports. The Nevada State Engineer in Ruling No. 5181 required Lincoln County and Vidler Water Company, Inc., to provide results from additional water-resources studies of Tule Desert in southern Nevada to support water-rights application 64692. As outlined by the ruling, the additional studies were to include the determination of the amount of ground water available from the Tule Desert basin, ground-water recharge to the Tule Desert, and the direction of ground-water flow. Results of these additional studies were published in five reports prepared for Lincoln County and Vidler Water Company, Inc. The National Park Service formally requested that the U.S. Geological Survey provide technical reviews of these five reports.

Open-File Report↗
USGS2008

Predevelopment Water-Level Contours for Aquifers in the Rainier Mesa and Shoshone Mountain area of the Nevada Test Site, Nye County, Nevada

Contaminants introduced into the subsurface of the Nevada Test Site at Rainier Mesa and Shoshone Mountain by underground nuclear testing are of concern to the U.S. Department of Energy and regulators responsible for protecting human health and safety. Although contaminants were introduced into low-permeability rocks above the regional flow system, the potential for contaminant movement away from the underground test areas and into the accessible environment is greatest by ground-water transport. The primary hydrologic control on this transport is evaluated and examined through a series of contour maps developed to represent the water-level distribution within each of the major aquifers underlying the area. Aquifers were identified and their extents delineated by merging and analyzing multiple hydrostratigraphic framework models developed by other investigators from existing geologic information. The contoured water-level distribution in each major aquifer was developed from a detailed evaluation and assessment of available water-level measurements. Multiple spreadsheets that accompany this report provide pertinent water-level and geologic data by well or drill hole. Aquifers are mapped, presented, and discussed in general terms as being one of three aquifer types?volcanic aquifer, upper carbonate aquifer, or lower carbonate aquifer. Each of these aquifer types was subdivided and mapped as independent continuous and isolated aquifers, based on the continuity of its component rock. Ground-water flow directions, as related to the transport of test-generated contaminants, were developed from water-level contours and are presented and discussed for each of the continuous aquifers. Contoured water-level altitudes vary across the study area and range from more than 5,000 feet in the volcanic aquifer beneath a recharge area in the northern part of the study area to less than 2,450 feet in the lower carbonate aquifer in the southern part of the study area. Variations in water-level altitudes within any single continuous aquifer range from a few hundred feet in a lower carbonate aquifer to just more than 1,100 feet in a volcanic aquifer. Flow directions throughout the study area are dominantly southward with minor eastward or westward deviations. Primary exceptions are westward flow in the northern part of the volcanic aquifer and eastward flow in the eastern part of the lower carbonate aquifer. Northward flow in the upper and lower carbonate aquifers in the northern part of the study area is possible but cannot be substantiated because data are lacking. Interflow between continuous aquifers is evaluated and mapped to define major flow paths. These flow paths delineate tributary flow systems, which converge to form the regional ground-water flow system. The implications of these tributary flow paths in controlling transport away from the underground test areas at Rainier Mesa and Shoshone Mountain are discussed. The obvious data gaps contributing to uncertainties in the delineation of aquifers and development of water-level contours are identified and evaluated.

Scientific Investigations Report↗
USGS2008

Portable chamber measurements of evapotranspiration at the Amargosa Desert Research Site near Beatty, Nye County, Nevada, 2003-06

Portable chamber measurements of evapotranspiration (ET) were made at the U.S. Geological Survey's Amargosa Desert Research Site in southern Nevada to help quantify component- and landscape-scale contributions to ET in an arid environment. Evapotranspiration data were collected approximately every 3 months from 2003 to 2006. Chamber measurements of ET were partitioned into bare-soil evaporation and mixed-species transpiration components. The component-scale ET fluxes from native shrubs typically surpassed those from bare soil by as much as a factor of four. Component-scale ET fluxes were extrapolated to landscape-scale ET using a one-layer, multi-component canopy model. Landscape-scale ET fluxes predominantly were controlled by bare-soil evaporation. Bare soil covered 94 percent of the landscape on average and contributed about 70 percent of the landscape-scale vapor flux. Creosote bush, an evergreen shrub, accounted for about 90 percent of transpiration on average due to its dominance across the landscape (80 percent of the 6 percent shrub cover) and evergreen character.

Nevada↗
USGS2006

MODOPTIM: A general optimization program for ground-water flow model calibration and ground-water management with MODFLOW

MODOPTIM is a non-linear ground-water model calibration and management tool that simulates flow with MODFLOW-96 as a subroutine. A weighted sum-of-squares objective function defines optimal solutions for calibration and management problems. Water levels, discharges, water quality, subsidence, and pumping-lift costs are the five direct observation types that can be compared in MODOPTIM. Differences between direct observations of the same type can be compared to fit temporal changes and spatial gradients. Water levels in pumping wells, wellbore storage in the observation wells, and rotational translation of observation wells also can be compared. Negative and positive residuals can be weighted unequally so inequality constraints such as maximum chloride concentrations or minimum water levels can be incorporated in the objective function. Optimization parameters are defined with zones and parameter-weight matrices. Parameter change is estimated iteratively with a quasi-Newton algorithm and is constrained to a user-defined maximum parameter change per iteration. Parameters that are less sensitive than a user-defined threshold are not estimated. MODOPTIM facilitates testing more conceptual models by expediting calibration of each conceptual model. Examples of applying MODOPTIM to aquifer-test analysis, ground-water management, and parameter estimation problems are presented.

Scientific Investigations Report↗
USGS2006

Documentation of a spreadsheet for time-series analysis and drawdown estimation

Drawdowns during aquifer tests can be obscured by barometric pressure changes, earth tides, regional pumping, and recharge events in the water-level record. These stresses can create water-level fluctuations that should be removed from observed water levels prior to estimating drawdowns. Simple models have been developed for estimating unpumped water levels during aquifer tests that are referred to as synthetic water levels. These models sum multiple time series such as barometric pressure, tidal potential, and background water levels to simulate non-pumping water levels. The amplitude and phase of each time series are adjusted so that synthetic water levels match measured water levels during periods unaffected by an aquifer test. Differences between synthetic and measured water levels are minimized with a sum-of-squares objective function. Root-mean-square errors during fitting and prediction periods were compared multiple times at four geographically diverse sites. Prediction error equaled fitting error when fitting periods were greater than or equal to four times prediction periods. The proposed drawdown estimation approach has been implemented in a spreadsheet application. Measured time series are independent so that collection frequencies can differ and sampling times can be asynchronous. Time series can be viewed selectively and magnified easily. Fitting and prediction periods can be defined graphically or entered directly. Synthetic water levels for each observation well are created with earth tides, measured time series, moving averages of time series, and differences between measured and moving averages of time series. Selected series and fitting parameters for synthetic water levels are stored and drawdowns are estimated for prediction periods. Drawdowns can be viewed independently and adjusted visually if an anomaly skews initial drawdowns away from 0. The number of observations in a drawdown time series can be reduced by averaging across user-defined periods. Raw or reduced drawdown estimates can be copied from the spreadsheet application or written to tab-delimited ASCII files.

Scientific Investigations Report↗
USGS2005

Hydraulic characterization of overpressured tuffs in central Yucca Flat, Nevada Test Site, Nye County, Nevada

A sequence of buried, bedded, air-fall tuffs has been used extensively as a host medium for underground nuclear tests detonated in the central part of Yucca Flat at the Nevada Test Site. Water levels within these bedded tuffs have been elevated hundreds of meters in areas where underground nuclear tests were detonated below the water table. Changes in the ground-water levels within these tuffs and changes in the rate and distribution of land-surface subsidence above these tuffs indicate that pore-fluid pressures have been slowly depressurizing since the cessation of nuclear testing in 1992. Declines in ground-water levels concurrent with regional land subsidence are explained by poroelastic deformation accompanying ground-water flow as fluids pressurized by underground nuclear detonations drain from the host tuffs into the overlying water table and underlying regional carbonate aquifer. A hydraulic conductivity of about 3 x 10-6 m/d and a specific storage of 9 x 10-6 m-1 are estimated using ground-water flow models. Cross-sectional and three-dimensional ground-water flow models were calibrated to measured water levels and to land-subsidence rates measured using Interferometric Synthetic Aperture Radar. Model results are consistent and indicate that about 2 million m3 of ground water flowed from the tuffs to the carbonate rock as a result of pressurization caused by underground nuclear testing. The annual rate of inflow into the carbonate rock averaged about 0.008 m/yr between 1962 and 2005, and declined from 0.005 m/yr in 2005 to 0.0005 m/yr by 2300.

Nevada↗
USGS2004

Plume-scale testing of a simplified method for detecting tritium contamination in plants and soil

Cost-effective methods are needed to detect contamination near radioactive-waste and other contaminated sites. Such methods should be capable of providing an early warning of contaminant releases and should be accurate and robust enough for assessing the long-term performance of waste-isolation facilities and remediation measures. Recently, a simplified method for detecting tritium contamination in plants and soil was developed (1). The method includes solar distillation of plant water from foliage, followed by filtration and adsorption of scintillation-interfering constituents on a graphite-based solid-phase-extraction column prior to direct-scintillation counting. The objectives of the in-progress study described here are to (i) test the simplified contamination-detection method for collection and analysis of plume-scale tritium data and (ii) gain insight into tritium migration pathways and processes.

Nevada↗
USGS2003

Estimated Ground-Water Withdrawals from the Death Valley Regional Flow System, Nevada and California, 1913-98

Ground-water withdrawals from 1913 through 1998 from the Death Valley regional flow system have been compiled to support a regional, three-dimensional, transient ground-water flow model. Withdrawal locations and depths of production intervals were estimated and associated errors were reported for 9,300 wells. Withdrawals were grouped into three categories: mining, public-supply, and commercial water use; domestic water use; and irrigation water use. In this report, groupings were based on the method used to estimate pumpage. Cumulative ground-water withdrawals from 1913 through 1998 totaled 3 million acre-feet, most of which was used to irrigate alfalfa. Annual withdrawal for irrigation ranged from 80 to almost 100 percent of the total pumpage. About 75,000 acre-feet was withdrawn for irrigation in 1998. Annual irrigation withdrawals generally were estimated as the product of irrigated acreage and application rate. About 320 fields totaling 11,000 acres were identified in six hydrographic areas. Annual application rates for high water-use crops ranged from 5 feet in Penoyer Valley to 9 feet in Pahrump Valley. The uncertainty in the estimates of ground-water withdrawals was attributed primarily to the uncertainty of application rate estimates. Annual ground-water withdrawal was estimated at about 90,000 acre-feet in 1998 with an assigned uncertainty bounded by 60,000 to 130,000 acre-feet.

Water-Resources Investigations Report↗
USGS2002

Documentation of spreadsheets for the analysis of aquifer-test and slug-test data

Several spreadsheets have been developed for the analysis of aquifer-test and slug-test data. Each spreadsheet incorporates analytical solution(s) of the partial differential equation for ground-water flow to a well for a specific type of condition or aquifer. The derivations of the analytical solutions were previously published. Thus, this report abbreviates the theoretical discussion, but includes practical information about each method and the important assumptions for the applications of each method. These spreadsheets were written in Microsoft Excel 9.0 (use of trade names does not constitute endorsement by the USGS). Storage properties should not be estimated with many of the spreadsheets because most are for analyzing single-well tests. Estimation of storage properties from single-well tests is generally discouraged because single-well tests are affected by wellbore storage and by well construction. These non-ideal effects frequently cause estimates of storage to be erroneous by orders of magnitude. Additionally, single-well tests are not sensitive to aquifer-storage properties. Single-well tests include all slug tests (Bouwer and Rice Method, Cooper, Bredehoeft, Papadopulos Method, and van der Kamp Method), the Cooper-Jacob straight-line Method, Theis recovery-data analysis, Jacob-Lohman method for flowing wells in a confined aquifer, and the step-drawdown test. Multi-well test spreadsheets included in this report are; Hantush-Jacob Leaky Aquifer Method and Distance-Drawdown Methods. The distance-drawdown method is an equilibrium or steady-state method, thus storage cannot be estimated.

Open-File Report↗
USGS2002

User guide for the drawdown-limited, multi-node well (MNW) package for the U.S. Geological Survey's modular three-dimensional finite-difference ground-water flow model, versions MODFLOW-96 and MODFLOW-2000

A computer program called the drawdown-limited, Multi-Node Well (MNW) Package was developed for the U.S. Geological Survey three-dimensional finite-difference modular ground-water flow model, commonly referred to as MODFLOW. The MNW Package allows MODFLOW users to simulate wells that extend beyond a single model node. Multi-node wells can simulate wells that are completed in multiple aquifers or in a single heterogeneous aquifer, partially penetrating wells, and horizontal wells. Multi-aquifer wells dynamically distribute flow between nodes under pumping, recharging, or unpumped conditions. Variations in intraborehole flow can be simulated with the MNW Package, which is limited by how finely an aquifer system has been discretized vertically. Simulated discharge from single-node and multi-node wells also can be drawdown limited, which is user specified for pumping or recharging conditions. The MNW Package also has the ability to track potential mixes of a water-quality attribute. Simulated wellbore flow can be compared with measured wellbore flow, which provides another constraint for model calibration.

Open-File Report↗
USGS2002

Delineation of the Troy Bedrock Valley and particle-tracking analysis of ground-water flow underlying Belvidere, Illinois

The U.S. Geological Survey, U.S. Environmental Protection Agency, and Illinois Environmental Protection Agency began a study of the hydrogeology, flow system, and distribution of contaminants in the aquifers underlying Belvidere, Ill., and vicinity in 1992. As part of the study, the ancestral Troy Bedrock Valley, located about 1.5 miles west of Belvidere, was identified as an important part of the ground-water-flow system. In the deepest parts of the valley, the basal Glenwood confining unit may be absent; thick deposits of sand and gravel that infill part of the valley may directly overlie the sandstone St. Peter aquifer, a regionally important source of water for public supply. With few deep wells open to the St. Peter aquifer present in the valley to provide necessary geologic information, tritium and other water-chemistry data were collected from eight wells to possibly delineate areas where the confining unit may be absent; the data also provide baseline water-quality information for an area expecting changes in land use and increases in water withdrawal. Also as part of the study, particle-tracking analysis was done using an available flow model to (1) identify possible discharge locations of ground water and contaminants and (2) delineate areas contributing recharge to the Belvidere municipal wells. This report presents and interprets water-chemistry data collected during December 2000 and presents results of particle-tracking analysis. Ground water in samples from two of four wells open to the St. Peter aquifer appears to have recharged after 1954, suggesting that the Glenwood confining unit may be absent near the wells. Other hydrogeologic and water-chemistry data, however, were inconclusive or contradictory. Concentrations of iron, manganese, and lead exceeded maximum contaminant levels in five or less samples, but materials associated with the water-distribution systems appear to contribute to the elevated concentrations above natural levels. Particle-tracking analysis indicates that most ground-water flow beneath possible contaminant-source areas discharges from the glacial drift aquifer to the Kishwaukee River. Most of the source areas are in or near Belvidere and are within 1,500 feet of the river. The analysis also indicates the possibility that in parts of the study area, some ground water does not discharge to the river, but flows beneath the Kishwaukee River in the underlying carbonate Galena-Platteville aquifer. Ground water that discharges to the one municipal well open to the glacial drift aquifer is estimated to travel over 1 mile in less than 25 years. Simulated residence (travel) times of ground water from the base of the glacial drift aquifer to the six municipal wells open, in part, to the Galena-Platteville aquifer, are estimated at less than about 40 years. Because fractures in this aquifer are unaccounted for in the flow model, actual areas contributing recharge are likely larger and travel times faster than those simulated for most of the municipal wells. Tritium data indicate that, in general, travel times from the land surface to the deepest parts of the Galena-Platteville aquifer are less than 46 years. Methyl tertiary-butyl ether data indicate that travel times to the upper part of the aquifer may be less than 16 years. The water-quality-based estimates of travel time generally are consistent with the estimates from particle-tracking analysis.

Illinois↗
USGS2001

Hydrogeology, water quality, and simulated effects of ground-water withdrawals from the Floridan aquifer system, Seminole County and vicinity, Florida

The hydrogeology and ground-water quality of Seminole County in east-central Florida was evaluated. A ground-water flow model was developed to simulate the effects of both present day (September 1996 through August 1997) and projected 2020 ground-water withdrawals on the water levels in the surficial aquifer system and the potentiometric surface of the Upper and Lower Floridan aquifers in Seminole County and vicinity. The Floridan aquifer system is the major source of ground water in the study area. In 1965, ground-water withdrawals from the Floridan aquifer system in Seminole County were about 11 million gallons per day. In 1995, withdrawals totaled about 69 million gallons per day. Of the total ground water used in 1995, 74 percent was for public supply, 12 percent for domestic self-supplied, 10 percent for agriculture self-supplied, and 4 percent for recreational irrigation. The principal water-bearing units in Seminole County are the surficial aquifer system and the Floridan aquifer system. The two aquifer systems are separated by the intermediate confining unit, which contains beds of lower permeability sediments that confine the water in the Floridan aquifer system. The Floridan aquifer system has two major water-bearing zones (the Upper Floridan aquifer and the Lower Floridan aquifer), which are separated by a less-permeable semiconfining unit. Upper Floridan aquifer water levels and spring flows have been affected by ground-water development. Long-term hydrographs of four wells tapping the Upper Floridan aquifer show a general downward trend from the early 1950's until 1990. The declines in water levels are caused predominantly by increased pumpage and below average annual rainfall. From 1991 to 1998, water levels rose slightly, a trend that can be explained by an increase in average annual rainfall. Long-term declines in the potentiometric surface varied throughout the area, ranging from about 3 to 12 feet. Decreases in spring discharge also have been observed in a few springs with long-term record. Chloride concentrations in water from the Upper Floridan aquifer in Seminole County range areally from 6.2 to 5,300 milligrams per liter. Chloride concentrations are lowest in the recharge areas of the Floridan aquifer system in the western part of Seminole County and near Geneva. The most highly mineralized water occurs adjacent to the Wekiva River in northwestern Seminole County, around the eastern part of Lake Jesup, and along the St. Johns River in eastern Seminole County. Analysis of limited long-term water-quality data indicates that the chloride concentrations in water for most wells in the Floridan aquifer system in Seminole County have not changed significantly in the 20-year period from 1976 to 1996, and probably not since the mid 1950's. Analysis of water samples collected from some Upper Floridan aquifer springs, however, indicates that the water has become more mineralized during recent years. Increases in specific conductance and concentrations of major cations and anions were observed at several of the springs within the study area where long-term water-quality data were available. Associated with these increases in the mineralization of spring water has been an increase in total nitrate-plus- nitrite as nitrogen concentration. A three-dimensional model was developed to simulate ground-water flow in the surficial and Floridan aquifer systems. The steady-state ground-water flow model was calibrated to water-level data that was averaged over a 1-year period from September 1996 through August 1997. The calibrated flow model generally produced simulated water levels in reasonably close agreement with measured water levels. As a result, the calibrated model was used to simulate the effects of expected increases in ground-water withdrawals on the water levels in the surficial aquifer system and on the potentiometric surface of the Upper and Lower Floridan aquifers in Seminole County. The ca

Water-Resources Investigations Report↗
USGS1999

Simulated effects of projected ground-water withdrawals in the Floridan aquifer system, greater Orlando metropolitan area, east-central Florida

Ground-water levels in the Floridan aquifer system within the greater Orlando metropolitan area are expected to decline because of a projected increase in the average pumpage rate from 410 million gallons per day in 1995 to 576 million gallons per day in 2020. The potential decline in ground-water levels and spring discharge within the area was investigated with a calibrated, steady-state, ground-water flow model. A wetter-than-average condition scenario and a drought-condition scenario were simulated to bracket the range of water-levels and springflow that may occur in 2020 under average rainfall conditions. Pumpage used to represent the drought-condition scenario totaled 865 million gallons per day, about 50 percent greater than the projected average pumpage rate in 2020. Relative to average 1995 steady-state conditions, drawdowns simulated in the Upper Floridan aquifer exceeded 10 and 25 feet for wet and dry conditions, respectively, in parts of central and southwest Orange County and in north Osceola County. In Seminole County, drawdowns of up to 20 feet were simulated for dry conditions, compared with 5 to 10 feet simulated for wet conditions. Computed springflow was reduced by 10 percent for wet conditions and by 38 percent for dry conditions, with the largest reductions (28 and 76 percent) occurring at the Sanlando Springs group. In the Lower Floridan aquifer, drawdowns simulated in southwest Orange County exceeded 20 and 40 feet for wet and dry conditions, respectively.

Water-Resources Investigations Report↗
USGS1997

Hydrogeologic framework and simulation of ground-water flow and travel time in the shallow aquifer system in the area of Naval Support Activity Memphis, Millington, Tennessee

Naval Support Activity (NSA) Memphis is a Department of the Navy facility located at the City of Millington, Tennessee, about 5 miles north of Memphis. Contaminants have been detected in surface-water, sediment, and ground-water samples collected at the facility. As part of the Installation Restoration Program, the Navy is considering remedial-action options to prevent or lessen the effect of ground-water contamination at the facility and to control the movement and discharge of contaminants. A numerical model of the ground-water-flow system in the area of NSA Memphis was constructed and calibrated so that quantifiable estimates could be made of ground-water-flow rates, direction, and time-of-travel. The sediments beneath NSA Memphis, to a depth of about 200 feet, form a shallow aquifer system. From youngest to oldest, the stratigraphic units that form the shallow aquifer system are alluvium, loess, fluvial deposits, and the Cockfield and Cook Mountain Formations. The shallow aquifer system is organized into five hydrogeologic units: (1) a confining unit composed of the relatively low permeability sediments of the upper alluvium and the loess; (2) the A1 aquifer comprising sand and gravel of the lower alluvium and the fluvial deposits, and sand lenses in the upper part of the preserved section of the Cockfield Formation; (3) a confining unit composed of clay and silt within the upper part of the Cockfield Formation; (4) the Cockfield aquifer comprising sand lenses within the lower part of the preserved section of the Cockfield Formation; and (5) a confining unit formed by low permeability sediments of the Cook Mountain Formation that composes the upper confining unit for the Memphis aquifer. Thicknesses of individual units vary considerably across the facility. Structural and depositional features that affect the occurrence of ground water in the shallow aquifer system include faulting, an erosional scarp, and 'windows' in the confining units. Underlying the shallow aquifer system is the Memphis aquifer, the primary source of water for NSA Memphis and the City of Memphis, Tennessee. Analyses of sediment cores, aquifer and well specific-capacity tests, and numerical modeling were used to estimate the hydraulic characteristics of units of the shallow aquifer system. The vertical hydraulic conductivity of core samples of the alluvium-loess confining unit ranged from about 8.5 x 10-5 to 1.6 x 10-2 feet per day, and the total porosity of the samples ranged from about 35 to 48 percent. The results of the aquifer test were used to estimate a horizontal hydraulic conductivity of about 5 feet per day for the alluvial-fluvial deposits aquifer. The total porosity of core samples of the alluvial-fluvial deposits aquifer ranged from about 22 to 39 percent. The vertical hydraulic conductivity of core samples of the Cockfield confining unit ranged from about 4.5 x 10-5 to 2.5 x 10-3 feet per day, and the total porosity ranged from about 41 to 55 percent. Well specific-capacity tests indicate that the horizontal hydraulic conductivity of sand units that compose the Cockfield aquifer range from about 0.5 to 3 feet per day. The vertical hydraulic conductivity of core samples of the Cook Mountain confining unit ranged from about 5.0 x 10-6 to 9.9 x 10-4 feet per day. Total porosity of core samples of the Cook Mountain confining unit ranged from about 30 to 42 percent. Ground-water flow and time-of-travel in the shallow aquifer system were simulated using the MODFLOW finite-difference model and the -particle-tracking program MODPATH. A three-layer, steady-state model of the shallow aquifer system was constructed and calibrated to the potentiometric surface of the A1 aquifer. Results of numerical modeling support the proposed conceptual hydrogeologic model of the shallow aquifer system. Ground-water time-of-travel in the A1 aquifer was simulated using an assumed effective porosity of 25 percent. Typical ground-water-flow velocities were on the order of 15 to 25 feet per year in the layer representing the Al aquifer in the model. The average residence time of particles seeded in this layer was about 800 years. Ground-water travel times were simulated at three sites within the Al aquifer: (1) the former N-6 hangar location, (2) the "grassy" area near Solid Waste Management Unit 7, and (3) at Solid Waste Management Unit 2. Results indicate close agreement between the particle-tracking simulations and the measured extent of contaminant plumes at the former N-6 hangar area and the grassy area near Solid Waste Management Unit 7. Based on the results of particle-tracking analyses of ground-water flow and the estimated locations of contaminant plumes at these two sites, the potential for contaminants to reach the Memphis aquifer in the next 100 years is negligible. However, particle-tracking analysis of ground-water flow at Solid Waste Management Unit 2 suggests that the time-of-travel of contaminants to Big Creek Drainage Canal could be less than 30 years.

Tennessee↗