Geology topics

Groundwater & aquifers

Find USGS groundwater and aquifer publications. Search an aquifer name or place, compare source-supplied dates and open reports for hydrogeology, recharge and water-level studies.

Follow this topic with RSS

At least 163 records · Page 9Linked to original sources

USGS2024

Status and understanding of groundwater quality in the Mojave Basin Domestic-Supply Aquifer study unit, 2018—California GAMA Priority Basin Project

Groundwater quality in the western part of the Mojave Desert in San Bernardino County, California, was investigated in 2018 as part of the California State Water Resources Control Board Groundwater Ambient Monitoring and Assessment Program Priority Basin Project. The Mojave Basin Domestic-Supply Aquifer study unit (MOBS) region was divided into two study areas—floodplain and regional—to assess differences between the two major aquifers used for drinking water supply in the area. This assessment characterized the quality of ambient groundwater and not the quality of treated drinking water. The study included three components: (1) a status assessment, which characterized the quality of groundwater resources used for domestic drinking-water supply in the floodplain and regional study areas; (2) a brief understanding assessment, which evaluated factors that could potentially affect the quality of groundwater used by domestic wells in the region; and (3) a comparative assessment between the groundwater resources used by domestic wells and public-supply wells in the two study areas. The domestic-well assessment was based on data collected by the U.S. Geological Survey from 48 domestic wells in January–May 2018. The public-supply assessment was based on data for samples from 322 public-supply wells in 2008–18, either collected by the U.S. Geological Survey or compiled from the California State Water Resources Control Boards Division of Drinking Water publicly available database. Concentrations of water-quality constituents in ambient groundwater were compared to regulatory and non-regulatory benchmarks typically used by the State of California and Federal agencies as health-based or aesthetic standards for public drinking water. Relative concentrations, defined as the measured concentration divided by the benchmark concentration, were classified as high (greater than 1.0), moderate (greater than 0.5 for inorganic constituents or 0.1 for organic and special-interest constituents, and not high), or low (concentrations lower than moderate). The floodplain and regional study areas were divided into 15 and 35 grid cells, respectively, and grid-based methods were used to compute the areal proportions of the two study areas with high, moderate, or low relative concentrations of individual constituents and classes of constituents. For the domestic-supply assessment, one or more inorganic constituents with health-based benchmarks were detected at high relative concentrations in 58 percent of the regional study area and 13 percent of the floodplain study area. The inorganic constituents with health-based benchmarks detected at high relative concentrations in the regional study area were arsenic, chromium and hexavalent chromium, fluoride, adjusted gross alpha particle activity, uranium, molybdenum, strontium, and nitrate; only arsenic was detected at high relative concentrations in the floodplain study area. One or more inorganic constituents with secondary maximum contaminant level benchmarks were detected at high concentrations in 15 and 6.7 percent of the regional and floodplain study areas, respectively. The constituents detected at high relative concentrations in the regional study area were total dissolved solids, chloride, sulfate, and iron; only total dissolved solids and sulfate were detected at high relative concentrations in the floodplain study area. Organic constituents were not detected at moderate or high relative concentrations in either the regional or floodplain study areas. Volatile organic compounds were detected at low relative concentrations in 21 and 27 percent of the regional and floodplain study areas, respectively, and pesticides were detected at low relative concentrations in 9.1 and 20 percent of the regional and floodplain study areas, respectively. The only individual organic constituent detected in more than 10 percent of either study area was the trihalomethane trichloromethane. Total coliform bacteria were detected in 15 and 27 percent of the grid wells in the regional and floodplain study areas, respectively. The greater prevalence of high relative concentrations of many inorganic constituents in the regional study area compared to the floodplain area likely indicates the greater diversity of geologic material at depth in aquifer material and generally finer-grained alluvium compared to the floodplain study area combined with generally older groundwater that has had more contact time with aquifer materials. In general, trace element concentrations (1) increased with increasing groundwater age, (2) increased with distance from recharge sources in the mountains, and (3) increased with closer proximity to some types of geological units. In general, groundwater from domestic wells in the floodplain study area is young, with most samples containing a component of modern groundwater based on tritium and unadjusted carbon-14 activities, whereas groundwater from domestic wells in the regional study area generally is old, with most samples having unadjusted carbon-14 ages of 5,000–40,000 years. Public-supply wells in MOBS generally were deeper than domestic wells and presumably are in contact with older, more weathered alluvium that may have more mobile trace elements, such as arsenic or uranium. However, only 26 percent of the public-supply regional study area had high relative concentrations of inorganic constituents, compared to 58 percent for the domestic regional study area. The percentages of the public-supply and domestic floodplain study areas with high relative concentrations of inorganic constituents were 11 and 13 percent, respectively. The ages of groundwater used by public-supply and domestic wells in each study area were similar, which was not expected given the greater depth of the public-supply wells. Three potential factors may contribute to these results: (1) greater spatial footprint of domestic well network, which may result in domestic wells pumping groundwater from fractured bedrock or mineralized areas not used by public-supply wells; (2) greater pumping rates in public-supply wells, resulting in more water being withdrawn from coarse-grained, heterogeneous alluvium than finer-grained layers, which may have higher concentrations of (or more mobile) inorganic constituents; and (3) a greater degree of well management with public-supply wells, which may include pausing use of or decommissioning wells if treating or blending water is not feasible to lower constituent concentrations.

California↗

USGS2020

Machine-learning models to map pH and redox conditions in groundwater in a layered aquifer system, Northern Atlantic Coastal Plain, eastern USA

Study region The study was conducted in the Northern Atlantic Coastal Plain aquifer system, in the eastern USA. Study focus Groundwater pH and redox conditions are fundamental chemical characteristics controlling the distribution of many contaminants of concern for drinking water or the ecological health of receiving waters. In this study, pH and redox conditions were modeled and mapped in a complex, layered aquifer system. Machine-learning methods (boosted regression trees) were applied to data from 3000 to 5000 wells. Predicted pH and the probability of anoxic conditions, defined by three thresholds of dissolved oxygen (0.5, 1, and 2 mg/L), were mapped at the 1-km 2 scale for each of 10 regional aquifer layers. New Hydrological Insights for the Region Maps depict the extent of acidic groundwater and oxic conditions in the shallow, unconfined surficial aquifer and in unconfined, recharge-proximal areas of underlying aquifers, in contrast to alkaline and anoxic groundwater elsewhere. Geographic patterns and influential predictors–including elevation, overlying confining-units thickness, and simulated groundwater age and flux–are consistent with prior understanding of the processes controlling pH and redox in the aquifer system. The model-based maps support robust estimates of aquifer proportions, either areal or volumetric, likely to contain groundwater of a specified quality or be vulnerable to specific pH- or redox-sensitive contaminants. The machine-learning methods were an effective tool to map groundwater quality at the regional scale.

New Jersey, Maryland, Delaware, Virginia↗

USGS2013

Geochemical, isotopic, and dissolved gas characteristics of groundwater in a fractured crystalline-rock aquifer, Savage Municipal Well Superfund site, Milford, New Hampshire, 2011

Tetrachloroethylene (PCE), a volatile organic compound, was detected in groundwater from deep (more than (>) 300 feet (ft) below land surface) fractures in monitoring wells tapping a crystalline-rock aquifer beneath operable unit 1 (OU1) of the Savage Municipal Well Superfund site (Weston, Inc., 2010). Operable units define remedial areas of contaminant concern. PCE contamination within the fractured-rock aquifer has been designated as a separate operable unit, operable unit 3 (OU3; Weston, Inc., 2010). PCE contamination was previously detected in the overlying glacial sand and gravel deposits and basal till, hereafter termed the Milford-Souhegan glacial-drift (MSGD) aquifer (Harte, 2004, 2006). Operable units 1 and 2 encompass areas within the MSGD aquifer, whereas the extent of the underlying OU3 has yet to be defined. The primary original source of contamination has been identified as a former manufacturing facility—the OK Tool manufacturing facility; hence OU1 sometimes has been referred to as the OK Tool Source Area (New Hampshire Department of Environmental Services, undated). A residential neighborhood of 30 to 40 houses is located in close proximity (one-quarter of a mile) from the PCE-contaminated monitoring wells. Each house has its own water-supply well installed in similar rocks as those of the monitoring wells, as indicated by the New Hampshire State geologic map (Lyons and others, 1997). An investigation was initiated in 2010 by the U.S. Environmental Protection Agency (USEPA) region 1, and the New Hampshire Department of Environmental Services (NHDES) to assess the potential for PCE transport from known contaminant locations (monitoring wells) to the residential wells. The U.S. Geological Survey (USGS) and the NHDES entered into a cooperative agreement in 2011 to assist in the evaluation of PCE transport in the fractured-rock aquifer. Periodic sampling over the last decade by the USEPA and NHDES has yet to detect PCE in groundwater from the residential-supply wells (as of 2012). However, part of assessing the potential for PCE transport involves understanding the origin of the groundwater in the monitoring and residential wells. One of the tools in delineating the movement of groundwater to wells, particularly in complex, highly heterogeneous fractured-rock aquifers, is the understanding of the geochemical and isotopic composition of groundwater (Lipfert and Reeve, 2004; Harte and others, 2012). This report summarizes findings from analyses of geochemical, isotopic, and dissolved gas characteristics of groundwater. Samples of groundwater were collected in 2011 from monitoring wells and nearby residential-supply wells in proximity to OU1.

New Hampshire↗

USGS2012

Groundwater quality and occurrence and distribution of selected constituents in the Aquia and Upper Patapsco aquifers, Naval Air Station Patuxent River, St. Mary's County, Maryland, July 2008

The Naval Air Station Patuxent River in southern Maryland has continued to expand in the first decade of the 21st century, contributing to rapid population growth in the surrounding area. The increase in population has caused State and County water managers and others to be concerned about the impact of population growth on the quantity and quality of groundwater supplies. The U.S. Geological Survey has been investigating the groundwater resources of the air station since 1998. As part of that ongoing investigation, groundwater was sampled in 2008 in six wells in the Aquia aquifer and two wells in the Upper Patapsco aquifer in the vicinity of Naval Air Station Patuxent River and Webster Outlying Field. Groundwater samples were analyzed for basic chemistry (field parameters, major ions, and nutrients) as well as several water-quality issues of concern including the occurrence of arsenic and tungsten, and saltwater intrusion. The results of the 2008 groundwater-quality sampling indicate that the overall quality of groundwater in the Aquia aquifer has not changed since 1943; data are too limited to determine if groundwater quality has changed in the Upper Patapsco aquifer. At one well in the Aquia aquifer, the arsenic concentration exceeded the U.S. Environmental Protection Agency standard for drinking water. Arsenic was not detected in samples from the Upper Patapsco aquifer. Tungsten concentrations were detected at low concentrations near the laboratory reporting level in all eight samples. There was no evidence of saltwater intrusion in any of the wells.

Maryl↗

USGS2012

Numerical simulation of groundwater movement and managed aquifer recharge from Sand Hollow Reservoir, Hurricane Bench area, Washington County, Utah

The Hurricane Bench area of Washington County, Utah, is a 70 square-mile area extending south from the Virgin River and encompassing Sand Hollow basin. Sand Hollow Reservoir, located on Hurricane Bench, was completed in March 2002 and is operated primarily as a managed aquifer recharge project by the Washington County Water Conservancy District. The reservoir is situated on a thick sequence of the Navajo Sandstone and Kayenta Formation. Total recharge to the underlying Navajo aquifer from the reservoir was about 86,000 acre-feet from 2002 to 2009. Natural recharge as infiltration of precipitation was approximately 2,100 acre-feet per year for the same period. Discharge occurs as seepage to the Virgin River, municipal and irrigation well withdrawals, and seepage to drains at the base of reservoir dams. Within the Hurricane Bench area, unconfined groundwater-flow conditions generally exist throughout the Navajo Sandstone. Navajo Sandstone hydraulic-conductivity values from regional aquifer testing range from 0.8 to 32 feet per day. The large variability in hydraulic conductivity is attributed to bedrock fractures that trend north-northeast across the study area. A numerical groundwater-flow model was developed to simulate groundwater movement in the Hurricane Bench area and to simulate the movement of managed aquifer recharge from Sand Hollow Reservoir through the groundwater system. The model was calibrated to combined steady- and transient-state conditions. The steady-state portion of the simulation was developed and calibrated by using hydrologic data that represented average conditions for 1975. The transient-state portion of the simulation was developed and calibrated by using hydrologic data collected from 1976 to 2009. Areally, the model grid was 98 rows by 76 columns with a variable cell size ranging from about 1.5 to 25 acres. Smaller cells were used to represent the reservoir to accurately simulate the reservoir bathymetry and nearby monitoring wells; larger cells were used in the northern and southern portions of the model where water-level data were limited. Vertically, the aquifer system was divided into 10 layers, which incorporated the Navajo Sandstone and Kayenta Formation. The model simulated recharge to the groundwater system as natural infiltration of precipitation and as infiltration of managed aquifer recharge from Sand Hollow Reservoir. Groundwater discharge was simulated as well withdrawals, shallow drains at the base of reservoir dams, and seepage to the Virgin River. During calibration, variables were adjusted within probable ranges to minimize differences among model-simulated and observed water levels, groundwater travel times, drain discharges, and monthly estimated reservoir recharge.

Utah↗

USGS2010

Quality of groundwater at and near an aquifer storage and recovery site, Bexar, Atascosa, and Wilson Counties, Texas, June 2004-August 2008

The U.S. Geological Survey, in cooperation with the San Antonio Water System, did a study during 2004–08 to characterize the quality of native groundwater from the Edwards aquifer and pre- and post-injection water from the Carrizo aquifer at and near an aquifer storage and recovery (ASR) site in Bexar, Atascosa, and Wilson Counties, Texas. Groundwater samples were collected and analyzed for selected physical properties and constituents to characterize the quality of native groundwater from the Edwards aquifer and pre- and post-injection water from the Carrizo aquifer at and near the ASR site. Geochemical and isotope data indicated no substantial changes in major-ion, trace-element, and isotope chemistry occurred as the water from the Edwards aquifer was transferred through a 38-mile pipeline to the aquifer storage and recovery site. The samples collected from the four ASR recovery wells were similar in major-ion and stable isotope chemistry compared to the samples collected from the Edwards aquifer source wells and the ASR injection well. The similarity could indicate that as Edwards aquifer water was injected, it displaced native Carrizo aquifer water, or, alternatively, if mixing of Edwards and Carrizo aquifer waters was occurring, the major-ion and stable isotope signatures for the Carrizo aquifer water might have been obscured by the signatures of the injected Edwards aquifer water. Differences in the dissolved iron and dissolved manganese concentrations indicate that either minor amounts of mixing occurred between the waters from the two aquifers, or as Edwards aquifer water displaced Carrizo aquifer water it dissolved the iron and manganese directly from the Carrizo Sand. Concentrations of radium-226 in the samples collected at the ASR recovery wells were smaller than the concentrations in samples collected from the Edwards aquifer source wells and from the ASR injection well. The smaller radium-226 concentrations in the samples collected from the ASR recovery wells likely indicate some degree of mixing of the two waters occurred rather than continued decay of radium-226 in the injected water. Geochemical and isotope data measured in samples collected in May 2005 from two Carrizo aquifer monitoring wells and in July 2008 from the three ASR production-only wells in the northern section of the ASR site indicate that injected Edwards aquifer water had not migrated to these five sites. Geochemical and isotope data measured in samples collected from Carrizo aquifer wells in 2004, 2005, and 2008 were graphically analyzed to determine if changes in chemistry could be detected. Major-ion, trace element, and isotope chemistry varied spatially in the samples collected from the Carrizo aquifer. With the exception of a few samples, major-ion concentrations measured in samples collected in Carrizo aquifer wells in 2004, 2005, and 2008 were similar. A slightly larger sulfate concentration and a slightly smaller bicarbonate concentration were measured in samples collected in 2005 and 2008 from well NC1 compared to samples collected at well NC1 in 2004. Larger sodium concentrations and smaller calcium, magnesium, bicarbonate, and sulfate concentrations were measured in samples collected in 2008 from well WC1 than in samples collected at this well in 2004 and 2005. Larger calcium and magnesium concentrations and a smaller sodium concentration were measured in the samples collected in 2008 at well EC2 compared to samples collected at this well in 2004 and 2005. While in some cases the computed percent differences (compared to concentrations from June 2004) in dissolved iron and dissolved manganese concentrations in 11 wells sampled in the Carrizo aquifer in 2005 and 2008 were quite large, no trends that might have been caused by migration of injected Edwards aquifer water were observed. Because of the natural variation in geochemical data in the Carrizo aquifer and the small data set collected for this study, differences in major-ion and trace element data among the samples collected in 2004, 2005 and 2008 cannot be directly attributed to the ASR site operations. When the data were analyzed graphically, no appreciable differences in isotope concentrations were observed between samples collected in 2004 and 2008 from Carrizo aquifer wells, indicating that the Edwards aquifer source water might not have affected the isotope chemistry of the native Carrizo aquifer water near the sampled Carrizo wells by July 2008.

Texas↗

USGS2019

Groundwater-level change for the periods 2002–8, 2008–12, and 2008–16 in the Santa Fe Group aquifer system in the Albuquerque area, central New Mexico

The U.S. Geological Survey, in cooperation with the Albuquerque Bernalillo County Water Utility Authority (ABCWUA), has developed a series of maps and associated reports, beginning in 2002, that document groundwater levels in the production zone of the Santa Fe Group aquifer system beneath a large area of the City of Albuquerque, New Mexico (hereafter called the study area). Herein, we document the construction of groundwater-level change maps for representative conditions during three periods: 2002–8, 2008–12, and 2008–16. Groundwater-elevation changes correspond to water use by the ABCWUA, with declines occurring prior to 2008 and accelerating recovery after 2008. Prior to 2008, the ABCWUA relied exclusively on groundwater from the Santa Fe Group aquifer system for municipal water supply. For the period 2002–8, near the end of the period of exclusive groundwater use, groundwater elevations in the production zone of the Santa Fe Group aquifer system declined as much as 20 to 30 feet. The largest 2002–8 groundwater-elevation declines were observed near the southeast corner of the study area and to the west of the Rio Grande. Since the ABCWUA implemented the San Juan-Chama Drinking Water Project in 2008, the proportion of municipal water supply sourced directly from surface water has increased to approximately two-thirds of the total water supply in 2016. Following initiation of this change in supply in 2008, groundwater elevations in the production zone of the Santa Fe Group aquifer system cumulatively rose as much as 20 to 30 feet by 2012 and 30 to 40 feet by 2016. The largest groundwater-elevation rises were observed near the northeast and southeast corners of the study area and to the west of the Rio Grande, whereas groundwater-elevation declines since 2008 were restricted to a localized area on the eastern margin of the study area. The area beneath the pre-flood-control-era (1971) flood plain of the Rio Grande underwent the least amount of groundwater-level change during any period, with minimal change prior to 2008 and small groundwater-elevation rises of less than 10 feet since 2008.

New Mexico↗

USGS2021

Update of the groundwater flow model for the Great Miami buried-valley aquifer in the vicinity of Wright-Patterson Air Force Base near Dayton, Ohio

A previously constructed numerical model simulating the regional groundwater flow system in the vicinity of the Wright-Patterson Air Force Base near Dayton, Ohio, was updated to incorporate current hydrologic stresses and conditions and improve the usefulness of the model for water-supply planning and protection. The original model, which simulated conditions from 1997 to 2001, was reconstructed with the most recently available U.S. Geological Survey groundwater modeling software and recalibrated to represent average groundwater flow conditions for the period of October 2018. The steady-state, three-dimensional, three-layer MODFLOW model of the aquifer encompasses about 241 square miles in Montgomery, Greene, and Clark Counties. The Great Miami buried-valley aquifer consists of glacial sands and gravels in a buried bedrock valley. The shale bedrock in the area is poorly permeable, but the glacial deposits can yield as much as 2,000 gallons per minute to wells. As groundwater is the primary source of drinking water in the heavily populated study area, groundwater pumping from the buried-valley aquifer represents the largest time-varying stress in the groundwater flow model. The model simulated 228 pumped wells. Hydraulic conductivities in the model ranged from less than 1 foot per day to 450 feet per day. Simulated recharge rates ranged from 6 inches per year to 12.2 inches per year. Boundary conditions and aquifer properties were unchanged from the previous model. Model grid spacing and orientation also were not modified from the previous model. Parameter estimation software was used to optimize model input parameters by matching simulated values to observed (estimated or measured) values. Calibrated parameters included horizontal hydraulic conductivity, vertical hydraulic conductivity, riverbed conductance, and recharge. Model calibration used measured water levels (hydraulic heads) from 124 observation wells, and streamflow gain/loss measurements from select reaches of the Mad River and its tributaries were compared with simulated streamflow gain/loss. Performance of the updated model is similar to previous studies. Eighty-one percent of simulated hydraulic heads were within 10 feet of the measured hydraulic heads, but comparison of the simulated streamflow gain/loss with the measured gain/loss indicates that streamflow gain/loss is not well represented by the updated model. The particle tracking program MODPATH was used to calculate groundwater flow paths from recharge areas to selected existing and proposed groundwater withdrawal sites that service Wright-Patterson Air Force Base. Areas contributing groundwater to withdrawal sites were delineated based on 1-, 5-, and 10-year groundwater travel times. In addition, groundwater flow paths were calculated to simulate a groundwater release at eight sites near Wright-Patterson Air Force Base.

Scientific Investigations Report↗

USGS2026

Hydrogeologic framework and conceptual groundwater-flow model of the panhandle and northwest parts of the High Plains (Ogallala) aquifer in Oklahoma, 1998–2022

This study was conducted by the U.S. Geological Survey, in cooperation with the Oklahoma Water Resources Board, to update the hydrogeologic framework and conceptual flow model for the panhandle and northwest parts of the High Plains (Ogallala) aquifer in Oklahoma, which together compose the Ogallala aquifer focus area. The study included the construction of a potentiometric surface, and available geologic and hydrologic data were used to evaluate saturated thickness of the aquifer. The water budget for the updated conceptual groundwater-flow model was based on estimated inflows and outflows for the 1998–2022 study period. Saturated thickness of the Ogallala aquifer averaged 127 and 116 feet for the panhandle and northwest parts, respectively. Groundwater withdrawals from the Ogallala aquifer for 1998–2022 averaged 422,054 and 39,645 acre-feet per year (acre-ft/yr) for the panhandle and northwest parts, respectively. Recharge, the primary inflow, was estimated at 0.63 inch per year for the 1998–2022 study period, with the panhandle part of the Ogallala aquifer receiving 175,068 acre-ft/yr and the northwest part of the Ogallala aquifer receiving 49,376 acre-ft/yr. Additional inflows included irrigation return flows, estimated at 8,111 and 642 acre-ft/yr for the panhandle and northwest parts, respectively, of the Ogallala aquifer. Net lateral groundwater flows, considered to be aquifer outflows, were estimated to account for 31,908 acre-ft/yr for the Ogallala aquifer focus area. Streambed seepage, which was an outflow of 5,535 acre-ft/yr, was only present in the northwest part of the Ogallala aquifer. Vertical leakage and saturated-zone evapotranspiration were considered negligible outflows. These findings provide a revised conceptual groundwater-flow model water budget for the Ogallala aquifer focus area in Oklahoma.

Colorado, Kansas, Oklahoma, Texas↗

USGS2020

Predicted pH of groundwater in the Mississippi River Valley alluvial and Claiborne aquifers, south-central United States

Regional aquifers in the Mississippi embayment are the principal sources of water used for public and domestic supply, irrigation, and industrial uses throughout the region. An understanding of how water quality varies spatially, temporally, and with depth are critical aspects to ensuring long-term sustainable use of these resources. A boosted regression tree (BRT) model was used by the U.S. Geological Survey (USGS) to map water quality in the three regional aquifers with the largest groundwater withdrawals in the embayment: the Mississippi River Valley alluvial (MRVA) aquifer, middle Claiborne aquifer (MCAQ), and lower Claiborne aquifer (LCAQ). The BRT model was used to predict pH to 1-kilometer raster grid cells for seven aquifer layers (one MRVA, four MCAQ, two LCAQ) following the hydrogeologic framework of the Mississippi embayment aquifer system regional MODFLOW model. The methods and approach used for pH predictions are the same as those used recently by the USGS to predict specific conductance and chloride in the aquifers. Explanatory variables for the BRT models included variables describing well location and construction, surficial variables such as soil properties and land use, and variables extracted from the groundwater flow model, such as groundwater levels and ages. The primary source of pH data was the USGS National Water Information System database. Additional data from State ambient groundwater monitoring programs and the Safe Drinking Water Information System also were used. For wells sampled multiple times, the most recent sample was used. Because groundwater residence times are long (greater than 100 years) throughout much of the study area, the possible effects of changes in water quality over time were considered small compared to the improvement in overall model accuracy by using available historical data. Values of pH from 3,362 wells for samples collected between 1960 and 2018 were used as training data for the BRT model. An additional 839 samples were used as holdout data to evaluate model performance. The predictive performance of the pH model is lower than for the training dataset, as indicated by an r-squared value of 0.89 for the training data and an r-squared of 0.71 for the holdout data. The root mean squared errors for the training and holdout data are 0.32 and 0.50 standard pH units, respectively. Data generated during this study and the model output are available from the companion data release.

Alabama, Arkansas, Louisiana, Mississippi, Missour↗

USGS2016

Simulating groundwater flow in karst aquifers with distributed parameter models—Comparison of porous-equivalent media and hybrid flow approaches

Understanding karst aquifers, for purposes of their management and protection, poses unique challenges. Karst aquifers are characterized by groundwater flow through conduits (tertiary porosity), and (or) layers with interconnected pores (secondary porosity) and through intergranular porosity (primary or matrix porosity). Since the late 1960s, advances have been made in the development of numerical computer codes and the use of mathematical model applications towards the understanding of dual (primary [matrix] and secondary [fractures and conduits]) porosity groundwater flow processes, as well as characterization and management of karst aquifers. The Floridan aquifer system (FAS) in Florida and parts of Alabama, Georgia, and South Carolina is composed of a thick sequence of predominantly carbonate rocks. Karst features are present over much of its area, especially in Florida where more than 30 first-magnitude springs occur, numerous sinkholes and submerged conduits have been mapped, and numerous circular lakes within sinkhole depressions are present. Different types of mathematical models have been applied for simulation of the FAS. Most of these models are distributed parameter models based on the assumption that, like a sponge, water flows through connected pores within the aquifer system and can be simulated with the same mathematical methods applied to flow through sand and gravel aquifers; these models are usually referred to as porous-equivalent media models. The partial differential equation solved for groundwater flow is the potential flow equation of fluid mechanics, which is used when flow is dominated by potential energy and has been applied for many fluid problems in which kinetic energy terms are dropped from the differential equation solved. In many groundwater model codes (basic MODFLOW), it is assumed that the water has a constant temperature and density and that flow is laminar, such that kinetic energy has minimal impact on flow. Some models have been developed that incorporate the submerged conduits as a one-dimensional pipe network within the aquifer rather than as discrete, extremely transmissive features in a porous-equivalent medium; these submerged conduit models are usually referred to as hybrid models and may include the capability to simulate both laminar and turbulent flow in the one-dimensional pipe network. Comparisons of the application of a porous-equivalent media model with and without turbulence (MODFLOW-Conduit Flow Process mode 2 and basic MODFLOW, respectively) and a hybrid (MODFLOW-Conduit Flow Process mode 1) model to the Woodville Karst Plain near Tallahassee, Florida, indicated that for annual, monthly, or seasonal average hydrologic conditions, all methods met calibration criteria (matched observed groundwater levels and average flows). Thus, the increased effort required, such as the collection of data on conduit location, to develop a hybrid model and its increased computational burden, is not necessary for simulation of average hydrologic conditions (non-laminar flow effects on simulated head and spring discharge were minimal). However, simulation of a large storm event in the Woodville Karst Plain with daily stress periods indicated that turbulence is important for matching daily springflow hydrographs. Thus, if matching streamflow hydrographs over a storm event is required, the simulation of non-laminar flow and the location of conduits are required. The main challenge in application of the methods and approaches for developing hybrid models relates to the difficulty of mapping conduit networks or having high-quality datasets to calibrate these models. Additionally, hybrid models have long simulation times, which can preclude the use of parameter estimation for calibration. Simulation of contaminant transport that does not account for preferential flow through conduits or extremely permeable zones in any approach is ill-advised. Simulation results in other karst aquifers or other parts of the FAS may differ from the comparison demonstrated herein.

Florida↗

USGS2015

Groundwater and surface-water interaction and effects of pumping in a complex glacial-sediment aquifer, phase 2, east-central Massachusetts

The U.S. Geological Survey, in cooperation with the Town of Framingham, Massachusetts, has investigated the potential of proposed groundwater withdrawals at the Birch Road well site to affect nearby surface water bodies and wetlands, including Lake Cochituate, the Sudbury River, and the Great Meadows National Wildlife Refuge in east-central Massachusetts. In 2012, the U.S. Geological Survey developed a Phase 1 numerical groundwater model of a complex glacial-sediment aquifer to synthesize hydrogeologic information and simulate potential future pumping scenarios. The model was developed with MODFLOW-NWT, an updated version of a standard USGS numerical groundwater flow modeling program that improves solution of unconfined groundwater flow problems. The groundwater model and investigations of the aquifer improved understanding of groundwater–surface-water interaction and the effects of groundwater withdrawals on surface-water bodies and wetlands in the study area. The initial work also revealed a need for additional information and model refinements to better understand this complex aquifer system. In this second phase of the study, the original groundwater flow model was revised to improve representation of groundwater and surface-water hydrology, stabilize the model, and reduce model error. The model was simplified by reducing the number of layers from 5 to 3 and adding the MODFLOW lake package (LAK) to simulate Lake Cochituate and Pod Meadow Pond and better represent interaction between the lakes and the aquifer. Model revisions improved stability and shortened run times, allowing use of automated parameter estimation software (PEST) to further refine the model hydraulic parameters and reduce simulation errors. Model simulations indicate that under average base-flow conditions, the Birch Road wells have a small effect on flow in the Sudbury River during most months, even at the maximum pumping rate of 4.9 ft 3 /s (3.17 Mgal/d). Maximum percent streamflow depletion in the Sudbury River caused by simulated pumping takes place during simulated drought conditions, when streamflow decreased by as much as 21 percent under maximum continuous pumping. Simulations also indicate that groundwater withdrawals at the Birch Road site could be managed so that adverse streamflow impacts are substantially ameliorated. Under the most ecologically conservative simulated drought conditions, simulated streamflow depletion was reduced from 21 percent to 3 percent by pumping at the maximum rate for 6 months rather than for 12 months. Simulations that return 10 percent of the Birch Road well withdrawals to Pod Meadow Pond indicate a modest reduction in the Sudbury River streamflow depletion and provide a larger percentage increase to streamflow just downstream of the pond. The groundwater model also indicates that well locations can have a large effect on the sustainable pumping rate and so should be chosen carefully. The model provides a tool for evaluating alternative pumping rates and schedules not included in this analysis.

Massachusetts↗

USGS2018

The Ozark Plateaus Regional Aquifer Study—Documentation of a groundwater-flow model constructed to assess water availability in the Ozark Plateaus

Recent short-term drought conditions have emphasized the need to better understand the delicate balance between abundance, sustainability, and scarcity of groundwater in the Ozark Plateaus aquifer system. In 2014, the U.S. Geological Survey began construction of a groundwater-flow model as a tool for the assessment of groundwater availability in the Ozark Plateaus aquifer system. The model was developed to benefit concurrent and future investigations involving groundwater-pumping scenarios, optimization, particle transport, and groundwater-monitoring network analysis. The groundwater model simulates 116 years (1900–2015) of hydrologic conditions and the response of the groundwater system to changes in stress including changes in recharge and groundwater pumping for water supply. Semiseasonal stress periods were simulated from the later part of 1991 to 2015 and represent higher demand and lower recharge in the spring and summer months and lower demand and higher recharge in the fall and winter months. Groundwater pumping increases throughout the simulation period with a maximum rate of about 600 million gallons per day (Mgal/d). The process of matching historical hydrologic data for the Ozark Plateaus aquifer system model was accomplished by a combination of manual changes to parameter values and automated calibration methods. Observation data used in the development and evaluation of the model included 19,045 hydraulic-head observations from 6,683 wells within the model area. Observation data also included stream leakage estimates summed to calculate a net gain or net loss value for approximately 81 named streams. The majority (mean of over 95 percent) of the recharge component is discharged through streams simulated in the model. The total simulated discharge to streams fluctuates seasonally between 7,500 and 17,500 Mgal/d with a mean outflow of 11,500 Mgal/d. Much of the remaining balance between modeled recharge inflows and stream outflows is made up by water moving into or out of storage in the aquifer system resulting in changes in modeled groundwater levels. The goal of the model was to develop a model capable of suitable accuracy at regional scales. The intent was not to reproduce individual local-scale details, which are typically not possible given the uniform cell size of 1 square mile. Although the model may not represent each local-scale detail, the model can be applied for a better understanding of the regional flow system and to evaluate responses to changes in climate and groundwater pumping.

Ozark Plateaus aquifer system↗

USGS2016

Simulation of groundwater withdrawal scenarios for the Redwall-Muav and Coconino Aquifer Systems of northern and central Arizona

The Northern Arizona Regional Groundwater Flow Model was used to estimate the hydrologic changes, including water-level change and groundwater discharge to streams and springs, that may result from future changes in groundwater withdrawals in and near the Coconino Plateau Water Advisory Council study area, Coconino and Navajo Counties, Arizona. Three future groundwater withdrawal scenarios for tribal and nontribal uses were developed by the Coconino Plateau Water Advisory Council and were simulated for the period representing the years from 2006 through 2105. Scenario 1 assumes no major changes in groundwater use except for increased demand based on population projections. Scenario 2 assumes that a pipeline will provide a source of surface water from Lake Powell to areas near Cameron and Moenkopi that would replace local groundwater withdrawals. Scenario 3 assumes that the pipeline extends to the Flagstaff and Williams areas, and would replace groundwater demands for water in the area. The Coconino Plateau Water Advisory Council withdrawal scenarios primarily influence water levels and groundwater discharge in the Coconino Plateau basin, near the western margin of the Little Colorado River Plateau basin, and the Verde Valley subbasin. Simulated effects of the withdrawal scenarios are superimposed on effects of previous variations in groundwater withdrawals and artificial and incidental recharge. Pre-scenario variations include changes in water-levels in wells; groundwater storage; discharge to streams and springs; and evapotranspiration by plants that use groundwater. Future variations in groundwater discharge and water-levels in wells will continue to occur as a result of both the past and any future changes. Water-level variations resulting from post-2005 stresses, including groundwater withdrawals and incidental and artificial recharge, in the area of the withdrawal scenarios are primarily localized and superimposed on the regional changes caused by variations in stresses that occurred since the beginning of the initial stresses in the early 1900s through 2005. Withdrawal scenario 1 produced a broad region on the Coconino Plateau where water-levels declined 3–5 feet by 2105, and local areas with water-level declines of 100 feet or more where groundwater withdrawals are concentrated, near the City of Flagstaff Woody Mountain and Lake Mary well fields, and the towns of Tusayan, Williams, and Moenkopi. Water-level rises of 100 feet or more were simulated at areas of incidental recharge near wastewater treatment facilities near Flagstaff, Tusayan, Grand Canyon South Rim, Williams, and Munds Park. Simulated water-level change from 2006 through 2105 for scenarios 2 and 3 is mostly different from water-level change simulated for scenario 1 at the local level. For scenarios 2 and 3, water levels near Cameron in 2105 where 1–3 feet higher than simulated for scenario 1. Water levels at Moenkopi are more than 100 feet higher due to the elimination of a proposed withdrawal well that was simulated in scenario 1. Scenario 3 eliminates more groundwater withdrawals in the Flagstaff and Williams areas, simulates 1–3 feet less water-level decline than scenario 1 across much of the Coconino Plateau, and water levels that are as much as 50 feet higher than simulated by scenario 1 near withdrawal wells in the Williams and Flagstaff areas. Scenario 1 simulated the most change in groundwater discharge for the Little Colorado River below Cameron and for Oak Creek above Page Springs where declines in discharge of about 1.3 and 0.9 cubic feet per second (ft 3 /s), respectively, were simulated. Other simulated changes in discharge through 2105 in scenario 1 are losses of less than 0.4 ft 3 /s at the Upper Verde River, losses of less than 0.3 ft 3 /s at Havasu Creek and at Colorado River below Havasu Creek, losses of less than 0.1 ft 3 /s at Clear Creek, and increases in flow at the south rim springs and Chevelon Creek of less than 0.1 and 0.3 ft 3 /s, respectively. Simulated changes in discharge for scenarios 2 and 3 are less than for scenario 1 because of lower rates of groundwater withdrawal. Scenario 3 resulted in greater groundwater discharge than scenarios 1 and 2 at all major groundwater discharge features from 2006 through 2105 except for Clear and Chevelon Creeks, where the same groundwater discharge was simulated by each of the three scenarios. Changes in groundwater discharge are expected to occur after 2105 to all major surface features that discharge from the Redwall-Muav and Coconino aquifers because change in aquifer storage was occurring at the end of the simulation in 2105. The accuracy of simulated changes resulting from the Coconino Plateau Water Advisory Council groundwater withdrawal scenarios is dependent on the persistence of several hydrologic assumptions that are inherent in the Northern Arizona Regional Groundwater Flow Model including, but not limited to, the reasonably accurate simulation of (1) transmissivity distributions, (2) distributions of vertical hydraulic properties, (3) distributions of spatial rates of withdrawal and incidental recharge, (4) aquifer extents, and (5) hydrologic barriers and conduits.

Arizona↗

USGS2012

Hydrogeology and simulation of groundwater flow and land-surface subsidence in the northern part of the Gulf Coast aquifer system, Texas, 1891-2009

In cooperation with the Harris–Galveston Subsidence District, Fort Bend Subsidence District, and Lone Star Groundwater Conservation District, the U.S. Geological Survey developed and calibrated the Houston Area Groundwater Model (HAGM), which simulates groundwater flow and land-surface subsidence in the northern part of the Gulf Coast aquifer system in Texas from predevelopment (before 1891) through 2009. Withdrawal of groundwater since development of the aquifer system has resulted in potentiometric surface (hydraulic head, or head) declines in the Gulf Coast aquifer system and land-surface subsidence (primarily in the Houston area) from depressurization and compaction of clay layers interbedded in the aquifer sediments. The MODFLOW-2000 groundwater flow model described in this report comprises four layers, one for each of the hydrogeologic units of the aquifer system except the Catahoula confining system, the assumed no-flow base of the system. The HAGM is composed of 137 rows and 245 columns of 1-square-mile grid cells with lateral no-flow boundaries at the extent of each hydrogeologic unit to the northwest, at groundwater divides associated with large rivers to the southwest and northeast, and at the downdip limit of freshwater to the southeast. The model was calibrated within the specified criteria by using trial-and-error adjustment of selected model-input data in a series of transient simulations until the model output (potentiometric surfaces, land-surface subsidence, and selected water-budget components) acceptably reproduced field measured (or estimated) aquifer responses including water level and subsidence. The HAGM-simulated subsidence generally compared well to 26 Predictions Relating Effective Stress to Subsidence (PRESS) models in Harris, Galveston, and Fort Bend Counties. Simulated HAGM results indicate that as much as 10 feet (ft) of subsidence has occurred in southeastern Harris County. Measured subsidence and model results indicate that a larger geographic area encompassing this area of maximum subsidence and much of central to southeastern Harris County has subsided at least 6 ft. For the western part of the study area, the HAGM simulated as much as 3 ft of subsidence in Wharton, Jackson, and Matagorda Counties. For the eastern part of the study area, the HAGM simulated as much as 3 ft of subsidence at the boundary of Hardin and Jasper Counties. Additionally, in the southeastern part of the study area in Orange County, the HAGM simulated as much as 3 ft of subsidence. Measured subsidence for these areas in the western and eastern parts of the HAGM has not been documented.

Texas↗

USGS2016

Regional land subsidence caused by the compaction of susceptible aquifer systems accompanying groundwater extraction

Land subsidence includes both gentle downwarping and sudden sinking of segments of the land surface. Major anthropogenic causes of land subsidence are extraction of fluids including water, oil, and gas. Measurement and detec- tion of land subsidence include both ground-based and remotely sensed air- borne and space-based methods. Methods for measurement of subsidence at points include differential leveling, global positioning system surveys, and extensometers. Satellite-borne differential interferometric synthetic aperture radar and airborne LiDAR techniques can detect land-surface movement over wide areas of interest. Aquifer-system compaction and subsidence owing to groundwater extraction typically occurs in areas of unconsolidated alluvial or basin-fill aquifer systems comprising aquifers and aquitards. Approaches to analyzing and modeling deformation of aquifer systems follow from the basic relations between head, stress, compressibility, and groundwater flow. Analysis and simulation of aquifer-system compaction have been addressed primarily using either an approach based on conventional groundwater flow theory or an approach based on linear poroelasticity theory. Both approaches rely on the principle of effective stress outlined by Karl Terzaghi in 1925. In the approach based on conventional groundwater flow theory, an aquitard drainage model explains the compaction of fine grained material using the principle of effective stress and theory of hydrodynamic lag. Packages for the widely-used MODFLOW groundwater model are available to simulate aqui- fer-system compaction and land subsidence using the aquitard-drainage approach. Poroelasticity theory describes the more fully coupled processes of groundwater flow and three-dimensional deformation of aquifer systems. The general theory accounts for compressible fluid, porous matrix and solid grains. Simulation codes using the poroelastic theory include some commer- cial software products and a few research codes.

Book chapter↗

USGS2021

Time-domain electromagnetic soundings and passive-seismic measurements for delineation of saline groundwater in the Genesee Valley-fill aquifer system, western New York, 2016–17

The U.S. Geological Survey, in cooperation with the New York State Department of Environmental Conservation, used noninvasive surface geophysics in the investigation of the distribution of saline groundwater in the valley-fill aquifer system of the Genesee River Valley near the former Retsof salt mine in western New York. In 1994, the Retsof salt mine, the largest of its kind in the western hemisphere, underwent a catastrophic roof collapse that resulted in groundwater inflow from the valley-fill aquifer system and bedrock fracture zones into the mine through two bedrock-rubble chimneys and the subsequent dissolution and filling of the mine with saturated brine. Since the early 2000s, except for a period of remedial pumping in 2006 to 2013, high-salinity water has migrated upward through the rubble chimneys into the basal part the aquifer system. The extent of saline-water migration within the aquifer system had not been evaluated since the end of remedial pumping when all the monitoring wells were grouted shut and abandoned. Installation of a monitoring-well network would be expensive and difficult given the thickness and heterogeneous character of valley fill. An investigation of the current extent of saline water in the aquifer system was warranted because the basal part of the aquifer is shallow to the north and it is used for water supply. In fall 2016 and fall 2017, the U.S. Geological Survey collected time-domain electromagnetic soundings at 105 sites along 13 cross-valley transects north and south of the mine-collapse area, east of Piffard, and on the Fowlerville Moraine. The time-domain electromagnetic soundings were colocated with passive-seismic measurements to estimate the bedrock-surface elevation through use of a regression equation developed from measurements at well sites with reported bedrock depths in the study area. An integrated analysis of the time-domain electromagnetic soundings with the depth-to-bedrock estimates, well logs, and past chloride-monitoring data suggests the presence of a zone of high electrical conductivity associated with saline water in the confined lower part of the valley-fill aquifer system. This high-salinity zone delineated in the lower confined aquifer extends from the mine-collapse area northward for more than 2.5 miles (4.0 kilometers). The chloride concentration in groundwater within this high-conductivity zone may be about 20,000 milligrams per liter. Saline water flowing upward through the bedrock-rubble chimneys and mixing with northward groundwater flow in the lower confined aquifer likely is a major source of chlorides for this high-conductivity zone. The northern extent of the zone is unclear because of the presence of highly saline water zones that were delineated by time-domain electromagnetic soundings in the lower confined aquifer and uppermost bedrock and are probably associated with historic salt-solution wells in Piffard or possibly sourced from natural brine pools.

New York↗

USGS2023

Status and understanding of groundwater quality in the Redding–Red Bluff shallow aquifer study unit, 2019—California GAMA priority basin project

Groundwater quality in the north Sacramento Valley (NSV) was studied in the Redding–Red Bluff shallow aquifer study unit (referred to as the NSV shallow aquifer or NSV-SA) as part of the Priority Basin Project (PBP) of the California Groundwater Ambient Monitoring and Assessment (GAMA) Program. The study unit is in Shasta and Tehama Counties and included two physiographic study areas: (1) the Redding area to the north and (2) the Red Bluff area to the south. The study was focused on groundwater resources used for domestic drinking-water supply, which are mostly drawn from shallower parts of aquifer systems than those of groundwater resources used for public drinking-water supply in the same area. This assessment characterized the quality of ambient groundwater in the aquifer before filtration or treatment, rather than the quality of drinking water delivered to the tap. The water-quality evaluation in this study has three components: (1) a status assessment, which characterized the quality of the groundwater resources used for domestic supply for 2018–19, in reference to state and national benchmarks; (2) an understanding assessment, which evaluated the natural and human factors potentially affecting water quality in those resources; and (3) a comparison between the groundwater resources used for domestic supply and those used for public supply in the region. The status assessment was based on data collected from 50 sites sampled by the U.S. Geological Survey for the GAMA-PBP in 2018–19. To provide context for the measured concentrations of groundwater constituents compared to U.S. Environmental Protection Agency and California State Water Resources Control Board Division of Drinking Water regulatory and non-regulatory benchmarks for drinking-water quality, relative concentrations (RCs) of groundwater constituents were calculated as the concentration in a sample divided by the respective benchmark. Health-based benchmarks include regulatory and non-regulatory human-health benchmarks such as a maximum contaminant level, notification level, or health-based screening level. Aesthetic-based benchmarks are regulatory or non-regulatory non-health-based benchmarks that can affect the color or taste of water. A grid-based method was used to estimate the proportions of the groundwater resources used for domestic drinking wells that have water-quality constituents below (low), approaching (moderate, greater than half the benchmark), or above (high) benchmark concentrations. This method provides statistically unbiased results at the study-area scale and permits comparisons to other GAMA-PBP study areas.

California↗