Borehole Logs and New Borehole Drilling

Lithologic and geophysical logs from more than 650 borings, test holes, and wells were used in the framework analysis (fig. 2). During the period from 2019 to 2021, eighteen new boreholes were drilled for this study using the mud-rotary method (fig. 1; table 2). Core samples were obtained using split-spoon samplers at 20-ft intervals, when possible, except in selected intervals of the Cretaceous Raritan Formation, where 10-ft cores were collected. Core samples were examined and sample color was described using a Munsell color chart (Kollmorgan Instruments Corporation, 1994). Geophysical logs were collected in the mud-filled boreholes following installation of polyvinyl chloride (PVC) casing.

Additional lithologic data used in the analysis were obtained from previous publications and legacy lithology logs, including deep observation wells with geophysical and lithologic logs that were collected during previous U.S. Geological Survey (USGS) studies (fig. 2; Chu and Stumm, 1995; Stumm, 2001; Cartwright, 2002; Stumm and others, 2002, 2004). Data from 50 deep boreholes that were cored and gamma-ray (gamma) logged by private contractors as part of a groundwater-plume investigation in southeastern Nassau County were also used in the analysis (fig. 2). Legacy lithologic logs utilized for the framework analysis were from New York Resources Commission publications (Leggette, 1938a, b; Leggette and Brashears, 1944; Roberts and Jaster, 1946, 1947; Johnson and others, 1958), from the Works Progress Administration (WPA) (Fluhr, 1937a, b, c), and geologic data from test borings taken for the Verrazzano-Narrows Bridge (Fluhr, 1954; fig. 2). The lithologic logs, which include descriptions of texture (gravel, sand, silt, and clay) and other characteristics, such as sorting, mineralogy, color, and the presence of lignite, as reported by drillers and geologists from drill cuttings and split-spoon core samples, are available in Finkelstein and others (2020) and the USGS GeoLog Locator (U.S. Geological Survey, 2022b).

A study by deLaguna (1948) highlighted the incomplete and vague descriptions of the subsurface in Kings and Queens Counties from the first half of the 20th century. Investigations of the hydrologic framework of Kings and Queens Counties have often produced conflicting results regarding the mapped extents and thicknesses of the major hydrogeologic units (deLaguna, 1948; Suter and others, 1949; Soren, 1971, 1978; Buxton and others, 1981; Buxton and Shernoff, 1999; Cartwright, 2002). This study uses the most complete dataset of subsurface geology available for western Long Island, including legacy lithologic descriptions, WPA descriptions of cores taken during well drilling, borehole geophysical logs, offshore seismic-reflection surveys, horizontal-to-vertical seismic surveys, and newly drilled and cored boreholes, to map the hydrogeologic framework of Kings, Queens, and Nassau Counties.

Geophysical logs collected during this and previous studies were analyzed to aid in mapping the hydrogeological framework in the study area. Geophysical logs are continuous digital records that reflect the physical properties of the sediments surrounding a borehole, fluids in the borehole and the surrounding formation, or both. Geophysical logs provide high-resolution information that may not be obtained by drill cores or water quality sampling alone. Borehole data varied from a single geophysical log (typically gamma) to a more complete log suite that included electric, electromagnetic induction (EM), temperature, and caliper logs. The geophysical logs used in these analyses are available for download from USGS GeoLog Locator (U.S. Geological Survey, 2022b).

For boreholes drilled for this study, gamma, spontaneous-potential (SP), single-point resistance (SPR), short- and long-normal resistivity (R), temperature, caliper, and EM logs were collected in mud-filled open boreholes at each site prior to the installation of PVC casing and screen (fig. 1; table 2) with the exception of K 3617, where only gamma and EM logs were collected. Only the deepest cased wells were logged using EM at each site. Nuclear magnetic resonance (NMR) logs were collected from selected wells through PVC casing (fig. 3; table 2). A typical example of a full suite of geophysical logs is presented in figure 4, and the different types of geophysical logs are described below.

Gamma logs provide a record of the total gamma radiation emitted by naturally occurring uranium, thorium, and potassium in a formation surrounding a borehole (Keys, 1990). Clays generally are more radioactive than the quartz sand that forms the bulk of the sediments on Long Island (fig. 4). Gamma logs were used for lithologic and stratigraphic correlation, and bed thickness.

SP logs provide a record of the electrical potential, or voltage, that develops at the contact between clays and sands penetrated by a borehole (Keys, 1990). SP differs from one formation to the next and is measured in millivolts (Serra, 1984). SP is a function of the chemical activity of the borehole fluid, water in the adjacent sediments, water temperature, and the type and quantity of clay. SP logs were useful in determining lithology, bed thickness, and salinity (fig. 4; Keys, 1990).

SPR logs provide a measure of the resistance in ohms between an electrode in the borehole and an electrode at land surface. The volume of surrounding material to which the SPR probe is sensitive is spherical and only 5 to 10 times the electrode diameter, and is affected by the borehole fluid (Keys, 1990). SPR logs were used to obtain high-resolution lithologic information (fig. 4).

Normal-R logs measure apparent resistivity in ohm-meters and are used to interpret lithology and water salinity (fig. 4; Keys, 1990). This technique uses four electrodes typically spaced 8 to 64 inches apart in the borehole, called short- and long-normal logs, respectively. The volume of surrounding material to which normal-R probes are sensitive is spherical, with a diameter about twice the electrode spacing (Serra, 1984; Keys, 1990).

Temperature logs are the continuous measurement of the temperature of the fluid column as a function of depth in degrees centigrade and can provide information on the materials immediately surrounding a borehole (Keys and MacCary, 1971; Keys, 1990). Because the temperature sensor only records the temperature of the water or air in the immediate vicinity of the sensor, measured temperatures may indicate the temperature of adjacent geologic materials and their contained fluids. Rock temperature may be indicated if no flow exists in the borehole and if equilibrium exists between the temperature of the fluid in the well and the temperature of the adjacent geologic materials. If there is no flow in or adjacent to a borehole, then the temperature gradually will increase with depth as a normal function of geothermal gradient (Keys and MacCary, 1971; Keys, 1990).

EM logs measure formation conductivity and are used in conjunction with gamma logs for distinguishing between conductive fluids and conductive clays (fig. 4). This technique uses an electromagnetic emitter coil that induces current loops within the surrounding formation to generate a secondary electromagnetic field. The intensity of the secondary field received by the receiver coil is proportional to the formation conductivity (fig. 4; Keys and MacCary, 1971; Serra, 1984; McNeill, 1986; Taylor and others, 1989; Keys, 1990). EM logs are measured in units of millisiemens per meter (mS/m) and are inversely related to the ohm-meter value of normal-R logs (Keys and MacCary, 1971; Serra, 1984; Keys, 1990).

The normally low conductivity of geologic sediments underlying Long Island is favorable for induction logging to delineate highly conductive fluids such as saltwater or leachate. EM logging has been used on Long Island to delineate the saltwater interface (Stumm, 1993, 1994, 2001; Stumm and Lange, 1994, 1996; Chu and Stumm, 1995; Stumm and others, 2002). Metzger and Izbicki (2013) and Prinos and Valderrama (2016) correlated EM log responses to chloride concentrations and produced log-based equations to estimate concentrations of chloride from EM logs. A linear relationship between chloride concentration and the response of EM logs in aquifers on Long Island was determined by Stumm and Como (2017). Chloride concentrations can be estimated from this linear relationship using EM logs for aquifers penetrated by a PVC-cased well.

The least-squares regression equation of Stumm and Como (2017) that is used to relate changes in EM conductivity to changes in chloride concentration in groundwater penetrated by the logged wells is:

NMR logs measure the induced magnetic moment of hydrogen protons contained within the fluid-filled pore spaces of porous media. Unlike other geophysical logs that respond to the rock matrix and fluid properties and are strongly dependent on mineralogy, NMR logs respond to the presence of hydrogen protons in the formation fluid to determine water content or fraction (equivalent to porosity in the saturated zone) and pore-size distribution (bound versus mobile water; Dlubac and others, 2013). NMR logs can be used to estimate vertically continuous porosity and hydraulic conductivity in an adjacent formation using equations and compilations of regressions determined from published collocated hydraulic tests of unconsolidated aquifers (Walsh and others, 2013; Knight and others, 2016; Kendrick and others, 2021).

NMR logs were collected from five monitoring wells in Nassau County (fig. 3; table 2) and were analyzed to determine clay-bound, capillary-bound, and mobile water content or fraction, and to estimate hydraulic conductivity of aquifers and confining units.

During 2019, NMR logs were collected from monitoring wells N 12321, N 12523, and N 14421 using the NMRSA (NMR Services Australia) BMR90 logging tool (fig. 3; table 2). During the period from 2021–22, NMR logs were collected using a Vista Clara JPY300 logging tool from monitoring wells N 12321, RW8 MW01D3, and DEC HC05 VPB02. Selected intervals of well N 14421 had NMR logs collected using a Vista Clara JPY300 logging tool and were described and analyzed by Stumm and Williams (2022). As shown in figure 5, NMR logs collected with the BMR90 and JPY300 tools produced similar estimates of total porosity and clay-bound, capillary-bound, and mobile water content.

Hydraulic conductivity was estimated from NMR logs collected using the JPY300 tool by using the sum of squared echoes (SOE) equation (Walsh and others, 2013). The SOE equation is

In the present application, a constant of 1.38×10⁴ was used based on compilations of regressions determined from published colocated hydraulic tests of unconsolidated aquifers (Walsh and others, 2013; Knight and others, 2016; Kendrick and others, 2021). Hydraulic conductivity was estimated from the NMR logs collected with the BMR90 tool using the Schlumberger-Doll Research (SDR) equation (Walsh and others, 2013). The SDR equation is

In the present application, a=2, b=1, and C_SDR was visually scaled so that the hydraulic conductivity estimates from the SDR equation best matched those from the SOE equation for borehole intervals where NMR logs were collected with both the JPY300 and BMR90 tools (fig. 5).

Horizontal-to-Vertical Spectral Ratio (HVSR) Soundings

HVSR soundings were made at 22 sites as part of a previous study of the depth to bedrock in New York City (fig. 3; DeMott and others, 2023b) These data were used to inform the hydrogeologic framework of western Long Island. HVSR soundings are a passive seismic method that is used to estimate the thickness of unconsolidated sediments and depth to bedrock (Lane and others, 2008). This passive seismic method uses a single broad-band three-component (two horizontal and one vertical) seismometer to record ambient seismic noise. The seismic noise induces resonance at frequencies that range from about 0.3 to 40 hertz in areas that have a strong acoustic contrast between the bedrock and overlying sediments. The ratio of the average horizontal-to-vertical spectrums produces a spectral-ratio curve that can be used with an average shear-wave velocity or a power-law regression equation to estimate sediment thickness and depth to bedrock.

The depth to bedrock was estimated from HVSR soundings using an empirical equation established between measured resonance frequencies and reported depths to bedrock at nearby wells for 15 sites in New York City, and for reported depths to bedrock that are less than 600 ft on Long Island (DeMott and others, 2023b). The power-law regression equation with a r² of 0.837 is as follows:

For depths to bedrock greater than 600 ft (DeMott and others, 2023b), a power-law regression equation with a r² of 0.948:

Time-Domain Electromagnetic (TDEM) Soundings

TDEM soundings made at 12 sites (fig. 3) from 2017 to 2019 were used in this study. The TDEM sounding method uses a transmitter to drive an electrical current through a square loop of insulated cable on the ground and a receiver to measure the current induced in the subsurface. The time-variant nature of the primary electromagnetic field creates a secondary electromagnetic field in the ground beneath the loop that generally mirrors the primary field of the transmitter loop (Christiansen and others, 2006; North Carolina Division of Water Resources, 2006). This secondary field immediately begins to decay, generating additional eddy currents that propagate downward and outward into the subsurface (U.S. Army Corps of Engineers, 1995).

The signal strength of the decaying currents at specific times and depths is controlled by the bulk conductivity of the subsurface, including the conductivity of subsurface rock and sediment units and their contained fluids (Fitterman and Stewart, 1986; Stewart and Gay, 1986; McNeill, 1994; Auken and others, 2008). The depth investigated in a study depends upon the ability of the equipment used to monitor the early and late arrival times of the transient signal.

TDEM sounding locations were selected based on several factors, including locations where saltwater was suspected to be in the subsurface, site access, amount of open space to accommodate wire loop sizes, and the need to minimize electromagnetic interference. Soundings made at sites near power lines and buried utilities have a higher likelihood of electromagnetic interference due to these anthropogenic sources of signal “noise.” The soundings were made using 20-, 40-, or 100-meter square transmitter loops. Generally, the largest possible loop size was used at any location to increase the likelihood of obtaining the deepest aquifer measurements (Payne and Teeple, 2011). The data from each sounding were used to develop layered- and smoothed-earth resistivity models for each location and to calculate apparent resistivity to depths of about 375–1,100 ft below land surface (Como and others, 2020; Stumm and others, 2020).

Marine Seismic-Reflection Surveys

Marine seismic-reflection surveys were conducted near Great Neck, Manhasset Neck, and Oyster Bay along the northern coast of Nassau County during 1999–2003 (Stumm, 2001; Stumm and others, 2002, 2004), and in the East River from the southern part of Roosevelt Island to the northeastern part of New York Harbor during 1999 (DeMott and others, 2023a). A marine seismic reflection survey is conducted by using a boat-towed sound source that creates signals in the form of sound energy. These signals travel through the water column and partially penetrate the sediments at and underlying the sea floor. The reflected signal returns are measured by boat-towed hydrophones. Changes in the acoustic impedance (a product of the density and acoustic velocity of each medium) indicate where the seabed and stratigraphic interfaces are located (Haeni, 1986, 1988; Robinson and Çoruh, 1988). These seismic surveys were used to better define the bedrock surface and characterize the unconsolidated sediments in offshore areas of the study area (Stumm, 2001; Stumm and others, 2002, 2004; DeMott and others, 2023a, b, c). Applications of this geophysical technique for characterization of unconsolidated sediments in the subsurface and delineation of the bedrock surface for hydrogeologic and water-resource studies have been described by Haeni (1986, 1988), Reynolds and Williams (1988), and Stumm (2001). These marine seismic-reflection data were incorporated into this hydrogeologic framework analysis (fig. 3).

Hydraulic Tests

Hydraulic properties of the underlying Kings, Queens, and Nassau Counties hydrogeologic units were compiled from published single- and multiple-well aquifer tests and specific-capacity tests. Additional specific-capacity tests were also collected and analyzed for selected wells installed as part of this investigation. Estimates of hydraulic properties associated with specific wells are stored in the USGS National Water Information System (NWIS) database (U.S. Geological Survey, 2022c). In the NWIS, hydraulic test results can be searched for by using various parameters, including site ID, test type, aquifer, and publication; then viewed; and can be downloaded using the Aquifer Test Locator graphical user interface and mapper (U.S. Geological Survey, 2022a).

McClymonds and Franke (1972) presented estimates of the transmissivity and hydraulic conductivity of aquifer sediments at 57 sites in Queens and Nassau Counties. The analyses include both single and multiple well tests to determine hydraulic conductivity in the North Shore, Magothy, and Lloyd aquifers using methods discussed in Cooper and Jacob (1946) and Theis and others (1954). In addition, aquifer properties of transmissivity, hydraulic conductivity, and storage were estimated at seven sites in the Magothy aquifer using numerical radial-flow models (Lindner and Reilly, 1982; Aronson and others, 1983; Prince and Schneider, 1989; Cartwright, 1997).

Williams and others (2020) estimated the transmissivity of aquifer sediments in the upper glacial, North Shore, Magothy, and Lloyd aquifers from specific-capacity tests performed at 447 production-well sites in Kings, Queens, and Nassau Counties. Transmissivities were estimated using the method described by Bradbury and Rothschild (1985). As part of this investigation, transmissivity estimated from the specific-capacity tests of the unconfined upper glacial aquifer was corrected for partial penetration using the method described by Brons and Marting (1961).

Hydraulic conductivity was estimated by assuming that the top of an aquifer was at the prepumping water level and that the bottom of the aquifer was at the base of the screen. An unconfined storage coefficient of 0.25 was assumed. The transmissivity estimated from specific-capacity tests of the Jameco aquifer was corrected for partial penetration, and the hydraulic conductivity was estimated by assuming the top of aquifer to be the bottom of the overlying Gardiners confining unit (Finkelstein and others, 2024), and the bottom of the aquifer to be at the base of the screen. A storage coefficient of 2×10^-5 was assumed for the confined aquifer.

Specific-capacity tests were conducted at 13 monitoring wells installed between 2019 and 2022 as part of this investigation. These wells were screened in the upper glacial, North Shore, Magothy, and Lloyd aquifers in Kings, Queens, and Nassau Counties (fig. 1, table 2). A submersible pump was used to pump the wells at a constant rate; pumping rates varied between well sites from 6.8 to 30 gallons per minute. Static and pumping water levels were manually measured with an electric tape. A Level TROLL 700 30 PSI transducer manufactured by In-Situ that was programmed to record water levels in logarithmic mode with a maximum time interval of 1 minute was used at some sites. The specific-capacity tests were analyzed following the method of Bradbury and Rothschild (1985) to estimate transmissivity. Partial penetration of the wells was accounted for, and hydraulic conductivity estimated for the North Shore and Lloyd aquifers where aquifer thickness could be determined from the geophysical and core logs. A confined storage coefficient of 2×10^-5 was assumed for the confined aquifers.

Compilation and Synthesis of Lithologic Data

Lithologic and geophysical logs from new and existing boreholes were analyzed, and hydrogeologic information was inferred based on driller’s logs and geologist’s descriptions, geophysical logs, or both. The borehole data guided the delineation of contacts between different hydrogeologic units, thus allowing identification of the surface elevation of each unit. However, differentiation between hydrologic units is not completely implied, and knowledge of the geologic framework must be used to determine what sediment types are expected to be within each unit. The lithologic data were used to populate a database in a geographic information system (GIS). All geoprocessing was done by using ArcGIS Desktop 10.8.1 (Esri, 2020).

The hydrogeologic surfaces for Kings, Queens, and Nassau Counties were interpolated using the data sources as described in previous sections. The hydrogeological surfaces point data and associated vertical data were used to define the three-dimensional surfaces of the major hydrogeologic surfaces in Kings, Queens, and Nassau Counties. The surfaces for the following hydrogeological units were interpolated and mapped from geophysical logs: upper glacial aquifer, glacial semiconfined aquifer, Gardiners confining unit, Jameco aquifer, North Shore confining unit, North Shore aquifer, Magothy aquifer, upper Raritan aquifer, Raritan confining unit, and Lloyd aquifer. The bottom of the hydrogeologic framework is the bedrock surface. The surface of the upper glacial aquifer, which is the uppermost unit, is land surface as estimated from high-resolution light detection and ranging (lidar) data (USGS, 2016).

About 650 lithologic, geophysical, or combined lithological and geophysical logs were examined, and more than 1,700 unique elevation points representing distinct geologic units were defined when summing the points for all the surfaces. Contour lines and extents were drawn based on the assigned elevation point data for each surface and were digitized at a 50-ft interval, representing the surface of each geological unit. The points, contour lines, and extents were used as inputs to create an interpolated surface using the Topo to Raster tool in ArcMap (Esri, 2020). The Topo to Raster tool was used because it can incorporate points, contour lines, and boundary extents as inputs for interpolations. All surfaces were interpolated on coincident grids with a uniform discretization of 100 ft. A 1-ft minimum thickness was imposed on each unit (where present) to ensure a continuous hydrogeologic framework in three dimensions and to facilitate its later use in a groundwater model of the system. Where a unit is not present, it is implied to have a 0-ft thickness. The only exception to the interpolation methodology is the bedrock surface elevation for Kings and Queens Counties as described in DeMott and others (2023c).

Limitations

Development of the hydrogeologic framework surfaces for Kings, Queens, and Nassau Counties came with some challenges. Due to data gaps and data availability, additional checking and postprocessing of the interpolated hydrogeologic surfaces were performed.

The hydrogeologic surfaces were interpolated independently of one another, which can result in interpolation artifacts in areas where the surface elevations of some units are not in stratigraphic or depositional order, or where units are very thin. As an example, some interpolated grid cells within the Magothy aquifer surface had an elevation greater than the corresponding grid cells for the Gardiners confining unit surface. The Gardiners confining unit overlies the Magothy aquifer, so the interpolated grid cells for the Magothy aquifer that were higher than the Gardiners confining unit resulted in a negative thickness for the Gardiners confining unit. This situation mostly occurred in areas with sparse data and away from point and contour data. These occurrences were expected because of data gaps and the number of interpolated hydrogeologic surfaces in the framework.

Interpolation artifacts were adjusted through use of the Raster Calculator tool (Esri, 2020) by either raising or lowering the elevation of the surfaces at locations where there was a conflict with the elevation of another surface. In the example given above for the interpolated cells from the Magothy aquifer, the problematic grid cells for the Magothy aquifer surface were lowered so that they were 1 ft less than and below the Gardiners confining unit. The surface elevations published in the accompanying data release (Finkelstein and others, 2024) accounted for all of the changes that were needed to keep the elevations of each unit in the correct depositional order with a 1-ft minimum thickness for each unit.

Mapping the erosional extents for each hydrogeologic unit was simplified by avoiding the contouring of each unit pinching out at its extent boundary. Trying to represent the erosional surfaces would have been difficult due to a lack of lithologic data present along the unit extents. This mostly occurred along the buried valley edges where the North Shore units are in contact with the Cretaceous units. It was determined that ending each surface abruptly, creating cliff-like features along the extents, was easier for the GIS to interpolate. Incorporating the contours necessary to represent the erosion yielded more complex surfaces with interpolation artifacts that were not desired.

Because no single, high-resolution land-surface raster for Long Island exists, lidar data from multiple sources were merged to create a seamless top surface for Long Island and the surrounding waterways. For most of the study area, a topobathymetric elevation model of New England that covers all of Long Island except for the westernmost part of Kings County was used (U.S. Geological Survey, 2016). The section of Kings County that was not covered by the topobathymetric elevation model was supplemented by National Elevation Data (U.S. Geological Survey, 2013), and data for offshore areas surrounding Kings County were obtained from the U.S. Geological Survey (2015). All lidar data were resampled to match the 100-ft grid discretization being used for the hydrogeologic framework data surfaces.

For this report, most of the chloride data used was obtained from the USGS NWIS database (U.S. Geological Survey, 2022c). Supplementary chloride data was obtained from USGS records consisting of hand-written field sheets, and while the accuracy and field methods used for obtaining these data are unknown, repeated field visitations were carefully documented by date, time, and location, with a corresponding chloride concentration recorded in mg/L. These data appear to correlate well with nearby chloride samples that were collected and within the USGS NWIS database. Some chloride data were estimated from geophysical logs using equation 1. The authors were able to analyze and archive the historical and geophysical data in a uniform chloride concentration database, which was published in an accompanying data release (Finkelstein and others, 2024).

Bedrock

The bedrock underlying the study area has been mapped previously as a structurally complex assemblage of high-grade metamorphic rocks, including gneiss, schist, amphibolite, marble, granite, and serpentinite. The bedrock assemblage is Precambrian to Paleozoic, and is part of the Hartland Formation (Veatch and others, 1906; Fuller, 1914; Sanford, 1938; Suter and others, 1949; Merguerian and Baskerville, 1987; Baskerville and Mose, 1989; Baskerville, 1992; Merguerian and Sanders, 1993; Brock and Brock, 2001). Small bedrock outcrops are present as isolated exposures in the northwesternmost part of Queens County.

The bedrock surface generally consists of a saprolite or weathered rock layer that is absent in the northern and northwestern parts of the study area where it was removed by glacial erosion and solid bedrock is encountered, to an average of about 50 ft thick south of the moraine (fig. 1). The weathered zone generally appears as a transition from clay above to partially weathered bedrock below, and consists mostly of white, yellow, and gray clay with quartz fragments, and shows relict foliation in places. This weathered zone was previously interpreted in many of the references mentioned in the previous paragraph to be clay or clayey gravel above the bedrock surface. In the northern parts of Kings and Queens Counties, narrow, buried, U-shaped valleys and isolated knobs and depressions are present, indicating localized erosion of the bedrock surface (fig. 7). The bedrock forms a relatively impermeable boundary for the groundwater flow system on Long Island (McClymonds and Franke, 1972).

The bedrock surface was mapped to include the saprolite zone where it is present. In general, the bedrock slopes toward the southeast, except in northwestern Queens and Kings Counties and in northern Nassau County, where glacial erosion has removed the saprolite and mostly solid bedrock remains (figs. 7, 8, 9; Stumm, 2001; Stumm and others, 2002, 2004). The highest elevation of the bedrock surface is about 16 ft above NAVD 88 in these areas. The lowest elevation is in the southeasternmost corner of the study area just offshore of the Nassau and Suffolk County border, where the elevation is more than 2,000 ft below NAVD 88 (fig. 7).

Glacial erosion and alluvial scouring of weathered bedrock and overlying Cretaceous sediments, and possibly Pleistocene sediments, formed several valleys that were subsequently buried (Stumm 1994; Stumm and others, 2002, 2004). The weathered bedrock at the base of these valleys has been partly or totally removed. Drill-core data from observation wells in the northernmost parts of Nassau, Queens, and Kings Counties indicate that the bedrock surface is relatively flat and does not follow the regional southeastward dip (figs. 7, 8, 9). The timing of the removal of the weathered bedrock is unknown, but it could have occurred during successive glacial advances during the Pleistocene (Stone and Borns, 1986; Stanford, 2010; Stanford and others, 2021). In the study area, drill-core data indicate that below in-place Cretaceous sediments, there is a saprolite zone above the bedrock surface, and where Cretaceous sediments are absent, typically there is no saprolite zone (Stumm 2001; Stumm and others, 2002, 2004). Marine seismic-reflection data indicate that the bedrock surface beneath the embayments of Long Island Sound is uneven and undulates (Stumm 2001; Stumm and others, 2002, 2004). This finding is consistent with observations of Tagg and Uchupi (1967), who found that the bedrock surface beneath most of Long Island Sound was severely eroded during the Pleistocene, with deep, flat-bottomed, U-shaped troughs that were formed and extended over 600 ft below the sea surface within the sediment.

Lloyd Aquifer

The Lloyd Sand Member of the Cretaceous Raritan Formation is an upward-fining sequence of white and pale yellow (2.5Y 7/3) sand and gravel with white clay and silt lenses, and these sediments form the Lloyd aquifer on Long Island (Veatch and others, 1906; Suter and others, 1949; Garber, 1986). The Lloyd aquifer is confined by bedrock below, and generally is confined above by the Raritan confining unit, which is formed by the unnamed clay member of the Raritan Formation. The base of the Lloyd aquifer typically consists of a 10- to 50-ft-thick gravel layer. The Lloyd aquifer dips to the southeast and is present throughout western Long Island except where the sediments that compose the Lloyd aquifer were removed by glacial erosion (figs. 8, 9, 10; Stumm 2001; Stumm and others, 2002, 2004). The upper surface of the Lloyd aquifer ranges from about 38 ft above NAVD 88 in northwestern Queens County to 1,455 ft below NAVD 88 in southeasternmost Nassau County (figs. 8, 9, 10).

The Lloyd aquifer, where present, ranges in thickness from less than 1 ft to about 660 ft in southeasternmost Nassau County. The Lloyd aquifer was removed by glacial erosion in the northwestern parts of Queens County, much of Kings County, and the northern part of Nassau County (figs. 8, 9, 10; Stumm 1994; Chu and Stumm, 1995; Stumm and others, 2002, 2004). Within the buried valleys onshore and offshore, the Lloyd aquifer was not penetrated during drilling and was not interpreted as being present in marine seismic-reflection surveys (figs. 8, 9, 10). Fourteen of the eighteen newly drilled wells penetrated the Lloyd aquifer: K 3616, K 3617, Q 4112, Q 4114, Q 4115, N 14392, N 14394, N 14417T, N 14418, N 14420, N 14421, N 14430, N 14438, and N 14477 (figs. 1, 8, 9, 10). Even though K 3617 was drilled at the base of a buried valley in Kings County, it also penetrated the Lloyd aquifer (figs. 9A, 10).

The Lloyd aquifer is a major source of water supply along the northern and southern coastal areas of Nassau County. McClymonds and Franke (1972) estimated an average hydraulic conductivity of about 50 ft/d for the Lloyd aquifer. According to Neuman (1982), the geometric mean provides a good approximation of hydraulic conductivity within three-dimensional aquifers. The geometric mean hydraulic conductivity of the Lloyd aquifer estimated from 25 pumping tests in western Long Island that were analyzed for this study was 55 ft/d (table 4). The geometric mean hydraulic conductivity estimated from the NMR log of the Lloyd aquifer penetrated by monitoring well N 12523 in north-central Nassau County was 80 ft/d, and ranged from less than 1 ft/d for fine-grained beds to more than 1,000 ft/d for coarse-grained beds (fig. 11). The hydraulic conductivity of the Lloyd aquifer estimated from a pumping test of a nearby production well (N 1328) was 66 ft/d. The geometric mean hydraulic conductivity estimated from the NMR log of the Lloyd aquifer in monitoring well N 14421 in southwestern Nassau County was 56 ft/d, and ranged from about 1 ft/d for fine-grained beds to more than 500 ft/d for coarse-grained beds (figs. 5, 12). The hydraulic conductivity of the Lloyd aquifer estimated from a pumping test of the screen zone at N 14421 was 47 ft/d (Stumm and Williams, 2022).

The mobile water content of the Lloyd aquifer estimated from the NMR logs in wells N 12523 and N 14421 averaged 0.22 and 0.20, respectively (table 5). The estimated capillary water content of the aquifer in wells N 12523 and N 14421 averaged 0.06 and 0.09, respectively. The clay-bound water content for the aquifer in the wells averaged less than 0.01.

Raritan Confining Unit

The unnamed clay member of the Cretaceous Raritan Formation forms the Raritan confining unit, which underlies most of the study area and dips to the southeast (fig. 13; Suter and others, 1949). The Raritan confining unit consists of dense clays and silts that are multicolored, including gray (2.5Y 4/1), white (2.5Y 9/1), red (10R 4/8), or tan (2.5YR 4/8), and are commonly variegated (Suter and others, 1949; Stumm, 2001; Stumm and others, 2002, 2004). The top of the Raritan confining unit is based on core samples and gamma logs, and is defined here as a unit consisting of dense clay and silt that is at least 20 feet thick. The Raritan confining unit ranges in thickness from less than 20 ft where it pinches out in the northwestern part of the study area to more than 310 ft in the southeasternmost part of Nassau County (figs. 8, 9, 13). The upper surface elevation of the Raritan confining unit ranges from about sea level in northwestern Queens County to about 1,230 ft below NAVD 88 in southeasternmost Nassau County (fig. 13). In many of the northern and western parts of the study area, the Raritan confining unit was severely eroded or completely removed by glacial scouring, creating an unconformity. The depositional extent of the Raritan confining unit is truncated by buried valleys that trend from the northwest to southeast that are interpreted as having been formed by glacial erosion (figs. 8, 9, 13; Stumm, 2001; Stumm and others, 2002, 2004). Eleven separate buried valleys were mapped in western Long Island during this study: four are offshore in the embayments of northern Nassau County, four are onshore in northern Nassau County (Stumm, 2001; Stumm and others, 2002, 2004), one was previously mapped in northern Queens County, and two new valleys are delineated in northern and western Kings County as part of this study (figs. 8, 9, 13; Veatch and others, 1906; Soren, 1971, 1978; Buxton and Shernoff, 1999).

The Raritan confining unit is the most extensive confining layer on Long Island, and restricts flow between the Lloyd aquifer and overlying aquifers. The geometric mean hydraulic conductivity of the clay and silt of the Raritan confining unit estimated from the NMR log of well N 12523 was 0.30 ft/d (fig. 11; table 4). The geometric mean hydraulic conductivities estimated from the NMR log from well N 14421 for a portion of the Raritan confining unit dominated by a silt and sand interval was 9.9 ft/d, and was 0.12 ft/d for a clay and silt interval (fig. 5). The mobile, capillary, and clay-bound water contents of the silt and sand interval of the Raritan confining unit estimated from the NMR logs in well N 14421 averaged 0.15, 0.15, and 0.02, respectively (table 5). The mobile, capillary, and clay-bound water contents of the clay and silt interval of the Raritan confining unit estimated from the NMR logs in wells N 12523 and N 14421 averaged 0.02, 0.18, and 0.08 and 0.01, 0.17, and 0.1, respectively.

Upper Raritan Aquifer

The name “upper Raritan aquifer” is introduced here to represent an aquifer contained in a unit of sediments that are part of the Cretaceous Raritan Formation that overlies the clay and silt of the Raritan confining unit and lies below the basal gravel bed of the Magothy aquifer in parts of western Long Island (fig. 14). The upper Raritan aquifer was initially identified in this study through analyses of high-resolution (5-ft intervals) core descriptions and gamma logs from a dense network of about 50 monitoring wells that were drilled into the Raritan confining unit in southeastern Nassau County as part of a contaminant-plume study (fig. 2). These analyses indicated a substantial deviation of the top surface of the Raritan confining unit as compared to that surface as defined in previously published maps. Dense clay beds were absent in the upper part of the Raritan confining unit in this area where the monitoring wells were drilled, indicating a possible facies or depositional change from fine- to coarse-grained lithologies in the upper part of the Raritan Formation within the study area. The newly mapped upper part of the Raritan Formation is generally characterized by fine to coarse sand with silt and minor clay lenses below the basal gravel bed of the Magothy Formation (figs. 8, 9, 14), whereas the Raritan confining unit has been previously described as consisting of dense clay and silt with small sand interbeds (Suter and others, 1949, Smolensky and others, 1990).

Further analysis of hundreds of core descriptions and gamma logs in Kings, Queens, and Nassau Counties indicated the upper Raritan aquifer extends throughout the southernmost parts of these counties (figs. 8, 9, 14). The aquifer also appears to extend throughout the central part of Nassau County, and may extend eastward into Suffolk County, Long Island. According to regional hydrogeologic framework investigations, including pollen studies of deep cores in the Raritan Formation of Suffolk County, the Raritan Formation between the Magothy Formation and Lloyd Sand Member appears to correlate chronologically with the South Amboy Fire Clay, Sayreville Sand, and Woodbridge Clay Members of the Raritan Formation in New Jersey (Zapecza, 1984; Sirkin, 1986). In the Potomac-Raritan-Magothy aquifer system underlying northeastern New Jersey, the Sayreville Sand Member of the Raritan Formation contains coarse-grained facies that are considered to be part of the upper aquifer in this system, and fine-grained facies that are considered to be part of the confining unit between the middle and upper aquifers (Farlekas, 1979; Zapecza, 1984).

The upper Raritan aquifer ranges in thickness from less than 1 ft where it pinches out to more than 200 ft at well N 12894 in southeastern Nassau County, corresponding to about one-third to two-thirds of the thickness of what had been previously mapped as the Raritan confining unit at that location (figs. 8, 9, 14; Suter and others, 1949; Perlmutter and Geraghty, 1963; Soren, 1971, 1978; Buxton and others, 1981; Buxton and Shernoff, 1999). Multicolored clays make up most of the Raritan confining unit, but are only found in lenses less than 20 ft in thickness in the upper Raritan aquifer, which contains more silt and sand than clay.

The geometric mean hydraulic conductivities estimated from the NMR logs of the upper Raritan aquifer penetrated by monitoring wells N 14421, DEC HC05 VPB02, and RW8 MW01D3 were 110, 100, and 40 ft/d, respectively, suggesting that the aquifer has a similar hydraulic conductivity to that of the upper parts of the Magothy aquifer (tables 4 and 5; figs. 2, 5, 15, 16). The geometric mean hydraulic conductivities ranged from less than 2 ft/d for fine-grained intervals to more than 800 ft/d for coarse-grained intervals. The mobile and capillary water contents of the upper Raritan aquifer estimated from the NMR logs in wells N 14421 averaged 0.26 and 0.08, respectively (table 5). The clay-bound water content for the aquifer in the well was less than 0.01.

Magothy Aquifer

The Magothy aquifer is present in Cretaceous sediments that consist of fine, micaceous, interbedded sand, silt, and clay (Suter and others, 1949; Smolensky and others, 1990). These sediments represent an upward-fining sequence consisting of gray, white, or pinkish sand, silt, and clay of the undifferentiated Matawan Group, or of the Magothy Formation (Smolensky and others, 1990). In some areas, there are lignite and iron-oxide concretions, and there is a nearly ubiquitous basal gravel bed that overlies the Raritan Formation. The Magothy aquifer overlies the Raritan confining unit or the upper Raritan aquifer (Suter and others, 1949).

The Magothy aquifer is absent in most of Kings County, the northern half of Queens County, and the northernmost parts of Nassau County, and is interpretated as having been eroded away by glacial erosion (figs. 8, 9, 17). Drill-core samples indicate that the Magothy aquifer consists of gray, subrounded, silty quartz sand and silt (5Y 6/1) and sand and silt-sized chemically stable, opaque heavy minerals. The top of the Magothy aquifer ranges from about 150 ft above NAVD 88 in north-central Nassau County to more than 310 ft below NAVD 88 in southeastern Nassau County (figs. 8, 9, 17). Glacial erosion produced a steep, wavy, north-facing scarp along the northern extent of the aquifer in south-central Queens County and north-central Nassau County (figs. 8, 9, 17). The Magothy aquifer ranges in thickness from less than 1 ft along the northern extent of the aquifer to about 930 ft in southeastern Nassau County.

The Magothy aquifer is the major water supply source for western Long Island. The geometric mean hydraulic conductivity of the Magothy aquifer as estimated from 35 pumping tests was 120 ft/d (table 4). The geometric mean hydraulic conductivities estimated from the NMR log of the Magothy aquifer penetrated by wells were 140 ft/d (N 12523), 81 ft/d (N 14421), 88 ft/d (DEC HC05 VPB02), and 110 ft/d (RW8 MW01D3) (figs. 2, 5, 11, 15, 16). In the Magothy aquifer, the geometric mean hydraulic conductivities ranged from a low of 0.33 ft/d for fine-grained intervals to a high of 590 ft/d for coarse-grained intervals (figs. 2, 5, 11, 15, 16; table 4). The mobile, water contents of the Magothy aquifer estimated from the NMR logs in the four wells averaged 0.26, 0.24, 0.25, and 0.28, respectively (table 5). Capillary water contents estimated for the aquifer in the four wells averaged 0.06 ,0.12, and 0.1, respectively. Estimated clay-bound water contents averaged 0.01 or less.

North Shore Aquifer

The North Shore aquifer is Pleistocene age fine to coarse sand, and is confined by Pleistocene clay and silt above and Precambrian to Cambrian bedrock below (Stumm, 2001, Stumm and others, 2002, 2004). Laterally, it is hydraulically connected with the Lloyd aquifer (Stumm, 2001; Stumm and others, 2002, 2004). The North Shore aquifer generally is present beneath the North Shore peninsulas and estuaries where Cretaceous sediments are absent due to glacial erosion (figs. 8, 9, 18). The North Shore aquifer was identified by Chu and Stumm (1995) in northwestern Queens, and by Stumm (2001) and Stumm and others (2002, 2004) in northern Nassau County, indicating that a unifying nomenclature for this aquifer unit along the northern shore of Long Island was needed. These aquifer sediments were called the Jameco Gravel by Swarzenski (1964) and the Port Washington aquifer by Kilburn (1979). The name “North Shore aquifer” is used in preference to the names “Jameco aquifer” or “Jameco Gravel” because the previously used names imply correlation with the Jameco aquifer or Jameco Gravel in southern Queens County. Also, the sand units that were mapped as the Port Washington aquifer were later found to be where marine seismic-reflection surveys indicated the presence of valleys that were hundreds of feet thick and filled with clay, indicating incorrect correlation and designations (Stumm, 2001; Stumm and others, 2002, 2004). Similar sediments (massive clay and varved clay units) have been identified in Suffolk County (Krulikas and Koszalka, 1983; Schubert and others, 2004). The North Shore aquifer is a distinct hydrogeologic unit that lies directly above bedrock and is overlain by a thick unit of clay and silt of the North Shore confining unit (figs. 8, 9, 18; Stumm, 2001; Stumm and others, 2002, 2004).

The North Shore aquifer is present in the northernmost part of Nassau County, the northwesternmost part of Queens Counties, the northern and parts of central Kings County, and extends southward to near Jamaica Bay in Kings County. This aquifer consists of moderately sorted, stratified, glacial-drift sediments filling in low-lying areas created after the removal of Cretaceous sediments and parts of the bedrock saprolitic zone by glacial erosion, fluvial erosion, or both. The aquifer is capped by clay and silt of Pleistocene age, which is the North Shore confining unit. Subsequent glacial advances eroded northwest–southeast-trending valleys that reached the bedrock surface and removed older sediments, possibly including those of the North Shore aquifer. These valleys were subsequently filled with hundreds of feet of clay and silt of the North Shore confining unit (figs. 8, 9, 18). The buried-valley sediments likely are associated with the most recent glacial advance in the area; however, the presence of the Jameco Gravel (also referred to as the Jameco aquifer) above some of these clay and silt sediments in two buried valleys in Kings County suggest that these sediments may not be the same age as the North Shore aquifer sediments in other parts of the study area. Therefore, the sediments in the buried valleys are grouped with the undifferentiated Pleistocene sediments of the North Shore confining unit.

Stanford (2010) and Stanford and others (2021) described early pre-Illinoian (about 230,000 years before present [ybp]) drainage in the New Jersey–New York area as including a diversion of the Hudson River to the East River and then eastward through Long Island, creating valleys that were subsequently filled. Veatch and others (1906) indicated potential Pleistocene age drainage patterns in Kings County through the coastal plain and southward offshore. Soren (1978), Buxton and Shernoff (1999), and Stanford and others (2021) proposed that the buried Flushing valley in north-central Queens County may be a buried valley related to pre-Illinoian drainage; however, detailed mapping presented in this study indicates that this valley is most likely glacial in origin, was not filled with sand, and did not have a southern extent offshore as had been proposed. The results of drilling in southern Queens County for this study indicate that the valley is filled with varved clay of the North Shore confining unit and does not extend to the southern coast.

One or two of the buried valleys mapped in Kings County may have been part of this early drainage system and may have ultimately continued offshore southward. Stanford (2010) and Stanford and others (2021) indicated that these buried valleys represent river channels in Kings County, and may have been modified, deepened, and filled during the most recent Wisconsinan glaciation during the Pleistocene. Glacial scouring directions in New York County (Manhattan) and Richmond County (Staten Island) indicate that glacial lobe flows were subparallel to the mapped buried valleys in this report (Stanford, 2010). Swarzenski (1964) interpreted sediments in northernmost Great Neck as being Sangamon age (130,000 to 80,000 ybp), which is an interglacial period between the Wisconsinan (about 80,000 to 17,000 ybp) and older Illinoian glaciations, or possibly of Illinoian age (about 191,000 to 130,000 ybp), and thus the oldest glacial deposits in the area.

The North Shore aquifer is a unit of moderately to poorly sorted, dark olive-brown (2.5Y 3/3) and olive-gray (5Y 6/2) gravel, sand, and silt. The sediments consist of subangular to subrounded quartz, rock fragments, chemically unstable opaque minerals, and much biotite and muscovite. The surface of the North Shore aquifer ranges from 30 ft below NAVD 88 in Kings County to about 480 ft below NAVD 88 in a buried valley at well N 12855 in northeastern Nassau County (figs. 8, 9, 18), and appears to be relatively flat. The North Shore aquifer is about 340 ft thick at well N 12855 but is less than 1-ft thick near its extents where it pinches out (figs. 8, 9, 18).

The geometric mean hydraulic conductivity of the North Shore aquifer as estimated from five aquifer tests was 31 ft/d (table 4). The geometric mean hydraulic conductivity estimated from the NMR log of the North Shore aquifer penetrated by monitoring well N 12321 was 11 ft/d, reflecting the high silt content in the aquifer at that location (fig. 19). The geometric mean hydraulic conductivity estimates ranged from 0.11 ft/d for fine-grained intervals to 310 ft/d for coarse-grained intervals. The mobile, capillary, and clay-bound water contents of the aquifer penetrated by the well averaged 0.16, 0.14 and 0.02, respectively (table 5)

North Shore Confining Unit

The North Shore confining unit is a sequence of Pleistocene-age clay and silt that occurs locally along the northern shore of Nassau County (Stumm and Lange, 1994, 1996; Stumm, 2001; Stumm and others, 2002, 2004), Kings and Queens Counties (Chu and Stumm, 1995), and Suffolk County (figs. 8, 9, 20; Soren, 1971). Oliver and Drake (1951), Grim and others (1970), Williams (1981), Lewis and Stone (1991), Stanford (2010), and Stanford and others (2021) interpreted these sediments as part of the stratified glacial-lake clay and silt that underlie Long Island Sound and the northern embayments in the study area (fig. 20).

Offshore seismic reflection surveys, borehole geophysical logs, and drill-core samples were used to correlate these Pleistocene clay and silt sediments with buried valleys that extend across the northern peninsulas of Nassau County and the surrounding embayments of Long Island Sound of Little Neck Bay, Manhasset Bay, Hempstead Harbor, and Cold Spring Harbor (Stumm and Lange, 1994, 1996; Stumm, 2001; Stumm and others, 2002, 2004). The North Shore confining unit consists of sediments from at least two separate depositional sequences—a brackish-water to marine clay and silt sequence, and a proglacial freshwater lake clay and silt sequence.

Core samples from well N 12013 and well N 12465 (fig. 2) contain olive-gray clay (5Y 4/2) with oyster shells, suggesting deposition in a shallow, brackish-water or marine environment. Amino-acid dating of oyster shells from well N 10101, which is adjacent to well N 12465, indicates an age of about 225,000 years before present (Ricketts, 1986). This suggests a Yarmouth age for these fine-grained marine sediments that make up the North Shore confining unit in this part of Great Neck along the north shore of Nassau County (Stumm and others, 2002). The presence of Cretaceous sediments beneath wells N 12013 and N 12465 in Nassau County and the presence of the Jameco aquifer above similar sediments at well K 3617 within a buried valley in southern Kings County, suggest that sediments in these areas could represent older Pleistocene deposition. Cores from many of the wells in these buried valleys contain varved gray clay and silt sequences without shell material, suggesting deposition in a more recent proglacial lake. Cretaceous sediments typically are thin or not present in many of these glacially eroded areas. Investigation of the stratigraphy and geologic age of these Pleistocene sediments was beyond the scope of this study. The marine clay in central Great Neck, and the gray clay below the Jameco aquifer within the buried valley in southern Kings County, are placed with the younger glacial-lake sediments as an undifferentiated Pleistocene unit of clay and silt that constitutes the North Shore confining unit, although these clays may represent an erosional remnant of an older interglacial environment.

Stanford (2010) and Stanford and others (2021) indicated that the maximum southern extent of ice in the New York City metropolitan area occurred about 21,000 years ago, and based upon the glacial varve count in the southern basin of Lake Passaic, the ice remained at a terminal moraine in central Kings and Queens Counties for 750–1,000 years, then rapidly retreated. The varves represent a minimum of 200 to 300 years of elapsed time from deglaciation to deposition of the organic sediment, assuming that these are thick proximal varves (Stanford, 2010; Stanford and others, 2021). Tagg and Uchupi (1967) conducted seismic profiles in Long Island Sound and found that the bedrock surface contains deep, flat-bottomed, U-shaped troughs that extend to more than 600 ft below NGVD 29. Tagg and Uchupi (1967) interpreted these troughs or valleys to be glacially eroded but were unable to discern if these features extended beneath Long Island because nearshore seismic data were lacking. Reeds (1927) described varved clay sediments found in the New York City area, and suggested that the depositional environment was in postglacial freshwater lakes occupying low-lying basins in southern New York. Reeds (1927) also proposed that there were a series of interconnected lakes called Glacial Lake Flushing within parts of Long Island Sound and northern Queens County (fig. 1), and interpreted these glacial lakes as containing extensive layers of varved clays and silt. Lewis and Stone (1991) described similar silt and clay sediments infilling extensive parts of Long Island Sound. Stanford (2010) and Stanford and others (2021) mapped similar sediments and interpreted several glaciations in the New York City metropolitan area. Offshore seismic-reflection profiles collected by the USGS during previous studies in Little Neck Bay, Long Island Sound, Manhasset Bay, Hempstead Bay, and Cold Spring Harbor display subhorizontal, parallel reflectors that are indicative of draped bedding, which is characteristic of silt and clay (Stumm, 2001; Stumm and others, 2002, 2004). These sediments fill several sharply defined buried valleys that truncate the surrounding geologic deposits (figs. 8, 9, 20). These valleys are typically asymmetrical, steep-sided, 2- to 4-miles (mi) long and about 1 mi wide, and trend northwest to southeast. The valleys extend to bedrock, which in places is deeper than 600 ft below sea level (figs. 8, 9, 20; Stumm and Lange, 1994, 1996; Stumm, 2001; Stumm and others, 2002, 2004).

Analyses of lithologic and geophysical logs from legacy and newly drilled wells indicate that at least 11 northwest-to-southeast-trending buried valleys underlie the northern parts of Kings, Queens, and Nassau Counties. Four of these valleys underly offshore embayments, four are onshore in northern Nassau County, one is in northern Queens County, and two were recently delineated in Kings County (Stumm, 2001; Stumm and others, 2002, 2004). One buried valley near Flushing in Queens County appears to have been mapped by previous studies (figs. 8, 9, 20; Soren, 1971, 1978; Buxton and others, 1981; Buxton and Shernoff, 1999). A major difference between the mapping for this study and previously published hydrogeologic studies of Kings and Queens Counties is that by using geophysical logs and core sample analysis of newly drilled wells, this study has confirmed that the buried valley near Flushing is likely of glacial origin and filled with varved glacial-lake clay and silt rather than fluvial sand and gravel sediments of the upper glacial aquifer. Another major difference is that this study finds no hydraulic connection within this buried valley between the upper glacial aquifer and the Lloyd aquifer as noted by Soren (1978) and Buxton and Shernoff (1999). There appears to be no gaps within the clay and silt sediments of the North Shore confining unit and that of the surrounding Raritan confining unit within the Flushing buried valley (figs. 8, 9, 20). The clay and silt of the North Shore confining unit restricts vertical flow between the North Shore aquifer and the upper glacial aquifer and also the Jameco and Magothy aquifers in this buried valley. The buried valley appears to be typical of those mapped along the northern part of Nassau County, where glacially eroded valleys are filled with varved clay and silt indicative of the North Shore confining unit that confines water in the underlying North Shore aquifer (Stumm, 2001; Stumm and others, 2002, 2004). This valley does not extend into southern Queens County as indicated by recent drilling with sediment samples and geophysical logs from this study, showing that Cretaceous sediments are present in this area.

In contrast, a similar buried valley in north-central Manhasset Neck contains both the North Shore aquifer and confining unit. In the southernmost part of this Manhasset Neck valley, there are areas of potential openings where the upper glacial aquifer or Magothy aquifer may be in hydraulic connection with the North Shore aquifer because of a gap between the Raritan confining unit and the North Shore confining unit (figs. 8, 9E, 18, 20; Stumm and others, 2002). There also appears to be a complex structural feature, possibly a fold in the North Shore confining unit, at the southernmost area of this buried valley. This narrow fold extends through the North Shore aquifer and is in contact with the bedrock surface, creating a localized, lateral hydraulic separation of the North Shore aquifer (figs. 8, 9E, 18). Hydrographs of water levels in the Lloyd and North Shore aquifers from wells in Manhasset Neck and a from a potentiometric surface map in 1997 (Stumm and others, 2002) indicate that this possible fold in the North Shore confining unit forms a low-permeability barrier that appears to isolate the western and eastern parts of the Lloyd and North Shore aquifers in this part of Manhasset Neck.

Two previously unmapped buried valleys were delineated in Kings County (figs. 8, 9, 18, 20). It is possible that one or both of these buried valleys were eroded by the ancestral Hudson or East Rivers, and were then deepened and widened by glacial erosion like those mapped in northern Queens and Nassau Counties. Veatch and others (1906) described at least one possible ancestral Hudson River channel eroding out a buried valley in Kings County and extending to the south off Long Island. Subsequent mapping by other researchers did not show any indications of buried valleys within Kings County; however, this may be due to a lack of new wells drilled with cores in that area (deLaguna, 1948, Buxton and others, 1981; Buxton and Shernoff, 1999).

Similar buried-valley sediments were found in New York County (Manhattan) northwest of the buried valleys in Kings County, suggesting a possible northwest extension of the Kings County valleys (Stumm, 2005; Stumm and Como, 2017). Core samples and geophysical logs from test wells drilled and installed in southern Manhattan Island indicate a deep northwest-trending valley that is 150 to 200 ft in depth that is filled with sand and gravel with diabase rock fragments, suggesting a possible fluvial origin (Stumm, 2005; Stumm and Como, 2017).

The North Shore confining unit underlies most northern parts of the study area (figs. 8, 9, 20). The upper surface of the North Shore confining unit is relatively flat in most parts of the study area except where older Pleistocene clay sediments included in the unit were possibly thrust-faulted and pushed to higher elevations during successive glacial advances during the late Pleistocene (figs. 8, 9, 20; Stumm, 2001; Stumm and others, 2002, 2004). The surface of the North Shore confining unit ranges from more than 50 ft above sea level in southern Great Neck to greater than 220 ft below NAVD 88 at well N 12855 (figs. 8, 9, 20). Marine seismic-reflection surveys indicate that the unit extends offshore and underlies embayments surrounding the study area (figs. 8, 9, 20; Stumm, 2001; Stumm and others, 2002, 2004). The North Shore confining unit has a maximum thickness of about 500 ft in northeastern Nassau County (figs. 8, 9, 20).

The geometric mean hydraulic conductivity estimated from the NMR log of the North Shore confining unit penetrated by monitoring well N 12321 was 0.22 ft/d (fig. 19; table 4). The estimated mean clay-bound, capillary-bound, and mobile water fractions of the confining unit penetrated by the well were 0.15, 0.20, and 0.01 respectively (table 5), which was the highest mean clay-bound and capillary-bound water content and lowest mean mobile water content estimated from the NMR logs for any hydrogeologic unit measured for this study.

Jameco Aquifer

The Jameco aquifer is present mainly in Kings and Queens Counties and only extends into the southwesternmost part of Nassau County (figs. 8, 9, 21). The aquifer sediments consist of reddish-brown (10R 4/3) to brown (7.5YR 4/3) gravel with coarse sand and boulders of mixed mineralogy and lithology, including diabase, gneiss, schist, sandstone, shale, and granite, which makes the unit distinctive when compared to the typical quartz-rich gravel in the upper glacial and Magothy aquifers (Veatch and others, 1906; Suter and others, 1949; Soren, 1971). Pebbles in the Jameco aquifer are well-sorted and well-rounded, suggesting fluvial transport. Analyses of core samples from wells Q 4117, K 3617, and K 3614 indicate nearly identical mineralogy and rock types in pebbles sampled from within the Jameco aquifer. This observation suggests that the aquifer sediments likely were deposited throughout northwestern Kings and Queens Counties, and extend into the southern parts of both Counties and into parts of southern Nassau County (figs. 8, 9, 21). The Jameco aquifer ranges in thickness from less than 1 ft near its northwestern extent to about 200 ft in west-central Queens County. The elevation of the Jameco aquifer ranges from about 150 ft above NAVD 88 in north-central Queens County to about 310 ft below NAVD 88 in the Rockaway area of southeastern Queens County (figs. 1, 9, 21). The variable surface elevation of the unit indicates that following deposition of the gravel, there was subsequent erosion. The Jameco aquifer sediments appear to show draped deposition from northwest to southeast over the westernmost parts of the study area (figs. 8, 9, 21).

Suter and others (1949) considered the Jameco aquifer to be the oldest Pleistocene unit of Long Island. Stone and Borns (1986) suggested a younger age (late Illinoian) for the Jameco aquifer than previously thought. Data collected as part of this study show that the Jameco aquifer overlies the North Shore confining unit and North Shore aquifer in a newly delineated buried valley at well K 3617 in central Kings County, suggesting that parts of the North Shore confining unit and aquifer may be older than the Jameco aquifer (figs. 8, 9, 21). The Jameco aquifer is the most hydraulically conductive of all the aquifers in western Long Island due to its well-sorted and coarse-grained lithology. The geometric mean hydraulic conductivity of the Jameco aquifer estimated from 49 pumping tests was 500 ft/d (Williams and others, 2020; table 4). Suter and others (1949) indicated that most of the public-supply wells that pumped from the Jameco aquifer in southern Kings County experienced rapid saltwater intrusion due to the high transmissivity of the aquifer. Previous studies typically combined the Jameco aquifer with the underlying Magothy aquifer due to uncertainties of well-screen depths and lack of detailed lithologic logs (Soren, 1971; Buxton and others, 1981; Buxton and Shernoff, 1999; Como and others, 2015).

Gardiners Confining Unit

The Gardiners confining unit consists of marine sediments, mainly a greenish-gray or gray clay (5Y 5/1), with a minor amount of sand in some areas that contains diatoms, foraminifers, and shell fragments that were deposited along the southern coast of Long Island during a Pleistocene interglacial period (Suter and others, 1949). The Gardiners confining unit is present along the southernmost parts of Kings, Queens, and Nassau Counties (figs. 8, 9, 22). The confining unit has a gentle slope to the south and appears to thin and pinch out toward its northern extent, indicating a transgressive depositional setting during an interglacial period (Suter and others, 1949). The unit decreases in thickness from north to south and ranges in thickness from 190 ft in the southwesternmost part of Kings County to less than 1 ft near the northern extents of the unit (figs. 9, 22). The Gardiners confining unit ranges in elevation from about 2 ft above NAVD 88 in the area north of Long Beach to about 190 ft below NAVD 88 in the Rockaway Island area (figs. 1, 9, and 22).

The Gardiners confining unit generally overlies the Jameco aquifer in the southmost parts of the study area (figs. 8, 9, 22). Locally, the unit is a confining unit for the underlying Jameco and Magothy aquifers. Previous studies delineated and separated out a shallow Pleistocene clay layer called the 20-foot clay (Smolensky and others, 1990). However, examination of core samples, geologic logs, and gamma logs in this study suggest that there are no clear, widespread separations between these clay units. The 20-foot clay was indistinguishable from the Gardiners confining unit, with only thin sand separating them in some areas, and is considered to be part of the Gardiners confining unit in this study (figs. 9, 22).

Upper Glacial Aquifer and Semiconfined Glacial Aquifer

The upper glacial aquifer consists of ice contact (glacial moraine and outwash) sediments of Pleistocene age (figs. 1, 9, 23; Suter and others, 1949; Cadwell, 1989; Smolensky and others, 1990). These sediments consist of stratified fine- to coarse-grained sand and gravel, unstratified boulders, clay, and silty till. This unit generally consists of moderately to poorly sorted brown (7.5Y 4/6) sand and dark yellowish-brown (10YR 4/6) sand with quartz, opaque minerals, rock fragments, and biotite and muscovite. The upper glacial aquifer is unconfined, and the water table is within the unit in most areas (fig. 24).

In a few isolated areas, the North Shore confining unit overlies sediments of the upper glacial aquifer, potentially confining or semiconfining water locally. In these instances, the sediments were mapped as a semiconfined subdivision of the upper glacial aquifer (figs. 1, 9, 23). Only a few areas of the upper glacial aquifer are thought to be semiconfined. These pockets of potentially semiconfined glacial aquifer range in thickness from less than 5 ft in southwesternmost Kings County to 290 ft in northeastern Nassau County (figs. 1, 9, 23, 23). The elevation of the semiconfined glacial aquifer ranges from about 5 ft below NAVD 88 in northern Great Neck to about 220 ft below NAVD 88 in southwesternmost Kings County (figs. 1, 9, 23).

Most of the upper glacial aquifer is unconfined throughout the study area. Offshore seismic-reflection surveys and core samples indicate that the upper glacial aquifer extends well offshore along the southern coast, but is limited to nearshore areas along the northern parts of Nassau County (Stumm and others, 2002). The upper glacial aquifer historically was a major source of water supply in western Long Island. The upper glacial aquifer south of the glacial moraines within the study area (Harbor Hill and Ronkonkoma) is composed predominantly of well-sorted outwash and is more hydraulically conductive than sediments north of the moraines where the aquifer is comprised of variably silty ice-contact sediments (figs. 1, 9, and 23). The geometric mean hydraulic conductivity of the upper glacial aquifer as estimated from 121 pumping tests in the glacial outwash was 200 ft/d; north of the moraine, the mean was 120 ft/d (table 4) as estimated from 22 pumping tests. The total porosity of the upper-glacial outwash south of the moraine ranged from 30 to 40 percent based on laboratory analysis of several hundred samples from southern Long Island as reported by Veatch and others (1906). The geometric mean hydraulic conductivity of the upper glacial aquifer estimated from the NMR log of monitoring well N 12321 located in Manhasset Neck north of the moraine was 34 ft/d, reflecting the high silt content of the ice-contact sediments at the site (figs. 2, 19). The relatively low hydraulic conductivity for the upper glacial aquifer, and a likely presence of local clay lenses, may be responsible in part for the elevated mound (120 ft above NAVD 88) in the water table in this area of Manhasset Neck (Stumm and others, 2002). NMR logging of well N 12321 confirmed saturated conditions throughout sediments with low permeability below this well, ruling out perched water conditions. The mobile, capillary, and clay-bound water contents of the upper glacial aquifer estimated from the NMR log of well N 12321 averaged 0.23, 0.15, and less than 0.01, respectively (table 5).

Land surface defines the top of the unsaturated portion of the upper glacial aquifer, while the top of the saturation zone within the aquifer is the water table. Locally in north-central Nassau County, the glacial sediments of the upper glacial aquifer can be unsaturated, and the water table occurs in the uppermost part of the underlying Magothy aquifer.

Merged Aquifers Units

Each aquifer unit previously discussed is defined by very specific age and depositional environments and variations in hydraulic properties due to facies changes. However, many units are laterally adjacent to each other with no confinement between them (figs. 8, 9).

The upper glacial, Jameco, and Magothy aquifers in Kings, Queens, and southern Nassau Counties were combined into one hydrogeologic complex, referred to in this report as the upper glacial-Jameco-Magothy aquifer complex (UJM). The Jameco aquifer is absent in northeastern Queens and northern Nassau Counties, so in these areas, only the upper glacial and Magothy aquifers were combined into the UJM. These aquifers are in hydraulic connection, and are above the major confining units of the Raritan confining unit, the North Shore confining unit, or both. Below these major confining units, the deeper Lloyd and North Shore aquifers were also combined into a single hydrogeologic complex, referred to in this report as the NSL (North Shore and Lloyd aquifer complex). The NSL is confined by the Raritan confining unit and North Shore confining unit, and the individual aquifers are in hydraulic connection throughout the study area. The NSL as an aquifer is present only in the northern parts of the study area because the North Shore aquifer is absent in the southern parts of the study area.

Historical Saltwater Intrusion in the Upper Glacial-Jameco-Magothy (UJM) Aquifer Complex

By the 1930s, chloride concentrations in the UJM aquifer complex in Kings County had increased to more than 100 mg/L (Lusczynski, 1952; Perlmutter and Arnow, 1953; Buxton and others, 1981), and by the 1940s, chloride concentrations at some inland wells exceeded 250 mg/L. The elevated chloride concentrations in the UJM resulted in the shutdown of public-supply wells within the Flatbush pumping center in Kings County in 1947 (Lusczynski, 1952). Lusczynski and others (1956) produced the earliest map of saltwater intrusion in the UJM in northwestern Queens and southern Nassau and Queens Counties. Lusczynski and others (1956) mapped a wedge of saltwater in the UJM in northwestern Queens and western Kings Counties in 1956. Breakthrough curves of chloride concentration in a well versus time were plotted for selected wells that had large (hundreds of mg/L) changes in chloride concentration and historical records of sampling. (fig. 25). These curves indicate that chloride generally increases slowly, followed by a rapid rise or breakthrough as the saltwater interface enters the well screen (fig. 25).

Suter (1937) documented a very large cone of depression referred to as “the crater” in the UJM that covered most of central Kings County during the mid-1930s, with water levels that were depressed more than 35 ft below sea level (NGVD 29) due to industrial and public-supply pumpage. Soren (1971) documented saltwater intrusion in the UJM in the early 1970s in the northwestern, north-central, and southwestern (Woodhaven Pumping Center) parts of Queens County. Lusczynski and Swarzenski (1966) delineated intermediate and deep UJM aquifer saltwater wedges in southern Nassau and southeastern Queens Counties by using chloride data from sampled observation wells, filter-press core samples, and geophysical logs.

A complex saltwater wedge documented within the UJM in southwestern Nassau County during 1958–61 was described as interlaced shallow, intermediate, and deep saltwater wedges within the freshwater aquifer (Lusczynski and Swarzenski, 1966). Salty groundwater was documented in southern Kings and Queens Counties in the UJM aquifer in 1983, but the saltwater interface was assumed to be several miles offshore to the south of Rockaway Island in southern Queens County (Buxton and Shernoff, 1999). Terraciano (1997) mapped the extent of saltwater intrusion in southwestern Nassau County and southeastern Queens County from 1988 to 1989 from existing observation wells. Analyses of historical chloride data from industrial supply wells and from observation wells sampled by the USGS from 1920 to 2022 provide a more detailed pattern of saltwater intrusion in the UJM aquifer than those in previous studies.

Kings and Queens Counties

During the period from 1920–39, chloride concentrations were more than 20,000 mg/L in western Kings County, more than 5,700 mg/L in southern Kings County and parts of southern Queens County (fig. 26A), and were as high as 14,000 mg/L in parts of northwestern Kings County and 560 mg/l in Queens County (fig. 26A). Saltwater intrusion affected the Flatbush pumping center in Kings County as pumping increased, with chloride concentrations as high as 800 mg/L in the UJM aquifer during the period from 1940–59 (fig. 26B). Lusczynski and others (1956) mapped a wedge of saltwater intrusion in northwestern Queens County and into western Kings County with concentrations as high as 5,000 mg/L, and a wedge of saltwater in the UJM aquifer in southern Kings and Queens Counties and southwestern Nassau County during 1956. During the period from 1960–79, chloride concentrations were as high as 2,000 mg/L in the Flatbush area of Kings County and 1,800 mg/L in southern Queens County (fig. 26C). During the period from 1980–99, the largest changes in the saltwater interface were seen in northern Queens County, where chloride concentrations were as high as 550 mg/L, and in southern Queens County, where concentrations were as high as 2,600 mg/L (fig. 26D). Reductions in pumpage in western Queens and Kings Counties may have caused the saltwater interface to stabilize during the period from 1980–99 (fig. 26D).

Nassau County

In southwestern Nassau County, no wells were sampled during the period from 1920–39. However, the 250 mg/L isochlor in the UJM aquifer is estimated to have been near the border of Nassau and Queens Counties based on data from well Q 286 (fig. 26A). From 1940–59 in the Inwood area of southwestern Nassau County, chloride concentrations were as high as 1,300 mg/L (fig. 26B). Chloride concentrations in the UJM increased to as much as 18,300 mg/L along the southwestern border of Nassau County during the period from 1940–59 (fig. 26B). Chloride concentrations also began to increase in the UJM aquifer at Long Beach in Nassau County during the period from 1940–59, with values exceeding 3,000 mg/L (fig. 26B). During the period from 1960–79, concentrations as high as 16,000 mg/L were measured in southwestern Nassau County, and as high as 8,700 mg/L at Long Beach (fig. 26C). From 1980–99, isochlors in southwestern Nassau County coalesced into a large wedge from eastern Long Beach to the Inwood area, with chloride concentrations as high as 12,250 mg/L at Long Beach and 17,000 mg/L in the Inwood area, (fig. 26D). From 2000 to 2022, chloride concentrations continued to increase in Inwood and the Long Beach areas of Nassau County, with recorded values greater than 23,000 mg/L (fig. 26E).

In eastern Manhasset Neck in northern Nassau County, chloride concentrations in the thousands of milligrams per liter were documented during 1960–79, likely because of sand mining in that area (fig. 26C; Stumm and others, 2002). During 1980–2022, chloride concentrations in the thousands and tens of thousands of milligrams per liter were observed in western and eastern Manhasset Neck and the Bayville areas of Nassau County (fig. 26D-F).

Examination of EM logs indicated a consistent slope in the saltwater–freshwater interface underlying Long Island as it relates to estimating chloride concentrations. Using these EM logs, the estimated thickness of freshwater within the UJM aquifer complex was determined at each well. The data were interpolated using the Natural Neighbors tool within ArcMap (Esri, 2020), and this analysis indicates a very sharp interface between freshwater and saltwater within the UJM aquifer complex. Figure 27 shows the thickness of the aquifer complex that is intruded with saltwater, represented as a percentage.

Historical Saltwater Intrusion in the Lloyd and North Shore Aquifers (NSL)

Previous studies on the extent of saltwater intrusion in the NSL in southwestern Long Island concluded that the saltwater interface in the Lloyd aquifer is several miles offshore beneath the Atlantic Ocean (Lusczynski and Swarzenski, 1966; Terraciano, 1997; Buxton and Shernoff, 1999). Elevated chloride concentrations in the Lloyd aquifer in the Rockaway area of Queens County were assumed to be the result of compromised well casings because the Lloyd aquifer was thought to contain only freshwater (fig. 9G; Perlmutter and Geraghty, 1963). Elevated chloride concentrations observed in 1983 in a few monitoring wells screened in the Lloyd aquifer in north-central Queens were dismissed as being caused by casing leakage; as a result, the saltwater interface was mapped several miles offshore (Buxton and Shernoff, 1999). Public-supply pumpage in the Lloyd aquifer during the 1950s only occurred in central and northeastern Queens County, Great Neck, Manhasset Neck, Oyster Bay, and Long Beach in Nassau County (Johnson and Waterman, 1952).

Analyses of historical chloride data for the period from 1890 to 2022 from industrial and observation wells indicate saltwater intrusion in parts the NSL aquifer (fig. 28A-G), which is the first documentation of saltwater intrusion of this aquifer in Kings and Queens Counties. The presence of high chloride concentrations early in the development of industrial and public-supply pumpage supports the hypothesis that the saltwater interface was at the southern coastline or onshore in Kings and Queens Counties during predevelopment conditions (fig. 28A, B).

Kings and Queens Counties

As early as 1890, chloride data from the first wells screened in the Lloyd aquifer indicate elevated chloride concentrations as high as 18,000 mg/L in western Kings County (fig. 28A). From 1890 to 1919, data from two wells indicate that a wedge of saltwater was likely onshore in western Kings County prior to the start of industrial pumping in the area (fig. 28A). Data from wells drilled between 1920 and 1939 indicate that the saltwater interface was probably onshore in northwestern Queens County and in western and southern Kings and Queens Counties during predevelopment conditions. Chloride concentrations during 1920–39 were as high as 20,000 mg/L in western Kings County, 2,000 mg/L in northwestern Queens County, and 240 mg/L in the Rockaway area of southern Queens County (fig. 28B). USGS groundwater-flow models (Kontis, 1999; Buxton and Shernoff, 1999) used the assumption that the saltwater interface was miles offshore and did not show any saltwater intrusion in these areas of Kings and Queens Counties, which does not agree with historical chloride data and current geophysical logs.

Lusczynski and Swarzenski (1966) described isolated increases in chloride concentrations in the Lloyd aquifer in the Rockaway area and interpreted these as possible downward leakages through the overlying Raritan confining unit. However, this current study incorporates previously unpublished chloride data (Finkelstein and others, 2024) collected by the USGS from 1890 to 1970 that allows for a refined interpretation of the predevelopment interface position prior to pumping. Examination of the chloride concentrations and depth of well screens with respect to Lloyd aquifer thickness indicates a classic lateral intrusion of a saltwater wedge at the base of a confined aquifer (fig. 9G). The Lloyd aquifer in the Rockaway area is about 250 ft thick. Legacy records indicate a chloride concentration of 240 mg/L was recorded in 1936 at well Q 123, a public-supply well in the Lloyd aquifer screened in the upper quarter of the aquifer at 835 ft below land surface in central Rockaway (fig. 9G). In contrast, Lloyd supply wells Q 290, Q 288, Q 287, and Q 1030 had chloride concentrations that ranged from 3 to 202 mg/L in the 1930s. These four nearby supply wells were screened more shallowly in the Lloyd aquifer than Q 123, and thus these chloride concentrations may represent a freshwater zone in the upper quarter of the Lloyd aquifer (fig. 9G). Bedrock is about 950 ft below land surface at Q 123 and indicating that water suppliers may have been avoiding the deeper and more transmissive parts of the Lloyd aquifer due to a potential wedge of saltwater residing at depth (fig. 9G). The legacy chloride data and shallow screen zones for the Lloyd production wells indicated that saltwater was present in the deepest parts of the confined aquifer at the southern coastline in Queens County as early as the 1930s (fig. 9G).

At well Q 1929, chloride concentrations were below 20 mg/L through the early 1960s, then began to rise rapidly at the end of the decade, and by 1973, the chloride concentrations reached 9,500 mg/L. Well Q 1929 is screened in the bottom quarter of the Lloyd aquifer in the easternmost part of Rockaway near the border with Nassau County. This suggests that a wedge of saltwater that existed within the lower part of the Lloyd aquifer west of this well began to expand and reach this area by the late 1960s.

Most historical supply wells in the vicinity of well Q 287 that were installed during the 1930s and early 1940s were screened in the shallower part of the Lloyd aquifer where chloride concentrations were generally below 200 mg/L. Wells screened in the bottom quarter of the aquifer had chloride concentrations greater than 200 mg/L. All of these supply wells had rapidly increasing chloride concentrations soon after installation and were shut down. Data suggest that the saltwater interface was probably at the coastline along the southern shore in Kings and Queens Counties prior to any public-supply pumping (predevelopment)(figs. 9G, 29) as determined from historical chloride data, recent water-quality sampling, and EM log analyses of wells Q 3655 and Q 4112.

By 1940–59, chloride concentrations in the Lloyd aquifer had increased to as high as 1,800 mg/L in northern Queens County. Saltwater was reported in well Q 1374 in the North Shore aquifer in northern Queens County during the 1950s (Perlmutter and Arnow, 1953; Lusczynski and others, 1956). During the period from 1940–59, chloride concentrations stabilized in western Kings County, probably because of the cessation of public-supply pumpage and reduction of industrial pumpage in that area (fig. 28C), but chloride in southern Kings County chloride concentrations increased to about 15,000 mg/L (fig. 28C).

Nassau County

The earliest indication of elevated chloride concentrations in the NSL within Nassau County was at supply well N 662, in eastern Manhasset Neck in 1939, with values as high as 165 mg/L (figs. 25, 28B). Chloride breakthrough curves for N 662 indicate a rapid saltwater intrusion during the 1940s and 1950s, with chloride concentrations reaching a peak of 440 mg/L in 1955 (Stumm and others, 2002). It appears that supply well N 656, adjacent to well N 662, is screened in the deeper part of the NSL and had chloride concentrations below 70 mg/L. This suggests a shallow wedge of saltwater intrusion within the NSL in this part of Manhasset Neck.

Supply well N 35 in western Manhasset Neck was affected by a wedge of saltwater by the late 1940s, with a peak chloride concentration in 1946 of 1,150 mg/L; the well was shut down soon thereafter (Stumm and others, 2002). Additional public-supply wells were installed in shallower parts of the NSL in western Manhasset Neck because continued development increased local water demands.

Chloride concentrations have been increasing slowly at well N 36 in northwestern Manhasset Neck since the late 1960s. Monitoring well N 12318, just northwest of well N 36, has chloride concentrations that have been slowly increasing since the 1990s. EM logging of N 12318 indicates much higher estimated chloride concentrations within the NSL about 40 ft above the screened zone. The estimated chloride concentrations from the EM logs are consistently above 1,000 mg/L.

Stumm and others (2002) delineated three wedges of saltwater intrusion in the NSL aquifer in Manhasset Neck in 1996—one in the northwestern part and two in the western part of Manhasset Neck. As a part of this study, these three were further delineated, plus a fourth was identified and delineated in the NSL in Manhasset Neck.

Analyses of chloride breakthrough curves for wells N 30, N 31, and N 1926 in Great Neck indicate rapid saltwater intrusion into the NSL within 20 years after the start of public-supply pumpage during the latter half of the 1960s (Stumm, 2001). Based on analyses of more than 50 years of public-supply pumping and chloride data, Stumm (2001) reported an apparent trend showing that once chloride concentrations reached a threshold of 50 mg/L, concentrations in the well remained above that threshold, regardless of whether pumping ceased. Chloride concentrations have been increasing at supply well N 3443 in central Great Neck since the 1960s, and chloride concentrations rose rapidly during the late 1980s to early 1990s at supply well N 687 in southwestern Great Neck (Stumm, 2001). Chloride concentrations in both wells have continued to increase. Chloride concentrations in well N 3443 reached a peak of 73 mg/L in 2019, and 130 mg/L in well N 687 in 2011.

Stumm (2001) delineated three wedges of saltwater in the NSL aquifer underling Great Neck in 1996 that were consolidated into two wedges as a part of this study.

In Bayville, at supply well N 10144, chloride concentrations increased rapidly within 10 years after the well installation in 1984, and reached about 250 mg/L in 1994 (Stumm and others, 2004). The well was subsequently shut down from production. North of well N 10144 at observation well N 12790 (located along the coast), estimated chloride concentration from EM logs reached a peak of about 5,000 mg/L in 2018. This saltwater wedge is thin and extends westward to well N 12870, where estimated chloride concentration from EM logs reached a peak of about 2,000 mg/L in 2018. This is the only saltwater wedge delineated within the NSL in the Bayville area.

Similar to the EM logs from the UJM, EM logs for the NSL also indicated a consistent slope in the saltwater-freshwater interface underlying Long Island based on estimated chloride concentrations. Using EM logs, the estimated thickness of freshwater within the NSL aquifer complex was determined at each well that penetrated the NSL in Nassau County. The data were interpolated using the Natural Neighbors tool within ArcMap (Esri, 2020). This analysis indicates a very sharp interface between freshwater and saltwater within the NSL aquifer complex. Figure 30 shows the thickness of the aquifer complex that is intruded with saltwater, represented as a percentage.

Current Saltwater Intrusion in the Upper Glacial, Jameco, and Magothy (UJM) Aquifers

Chloride concentration data from 2000–22 were used to map the current position of the saltwater interface in the UJM as part of this study (fig. 26E). Analyses of EM derived chloride data (equation 1) and water quality samples (U.S. Geological Survey, 2022c) from 2000–22 indicate that the 250 and 1,000 mg/L isochlors generally are close together in southwestern Nassau County, and a sharp interface exists at the brackish water to saltwater (fig. 26E) transition. The saltwater interface appears to be farther inland than previously mapped by Terraciano (1997). The 1,000 mg/L isochlor appears to indicate much more saltwater intrusion in southern Queens County than shown by previous mapping (fig. 26E; Buxton and Shernoff, 1999). Areas with salty groundwater exceeding 1,000 mg/L chloride were also mapped in northwestern, southern, and northern Queens County, indicating that significant saltwater intrusion has occurred in those areas (fig. 26E). Most of the UJM aquifer in Kings County has been intruded by saltwater (fig. 26E).

In northern Nassau County, a small area in southeastern Great Neck with about 500 mg/L chloride was observed in the UJM, whereas the rest of the aquifer on the Great Neck peninsula appears to be freshwater. An area on the eastern coast of Manhasset Neck has high chloride concentrations that are above 5,000 mg/L (fig. 26E), and localized saltwater intrusion also has been observed in northwestern Manhasset Neck. In the Bayville area, most of the shallow aquifers are intruded with chloride concentrations of more than 5,000 mg/L. EM logs also indicate that saltwater intrusion is occurring on the west side of the Manhasset Neck. EM logs from between 2007 and 2020 also indicate saltwater intrusion was occurring in most wells in southern Nassau County. Chloride concentrations exceed 5,000 mg/L throughout Long Beach and Inwood, along the southern coast of Nassau County (fig. 26E). The saltwater interface appears to be farther inland than had been previously mapped in this area, indicating that saltwater intrusion is occurring, probably due to current public-supply pumping.

The water table ranges in elevation from 82.8 ft above NAVD 88 at well N 9089 in north-central Nassau County to less than 5 ft above NAVD 88 along the coastline (fig. 24; Como and others, 2015). A localized mound of water has been mapped consistently in central Manhasset Neck, with elevations as high as 121.7 ft at well N 12451 (Stumm and others, 2002; Como and others, 2015). In Kings and Queens Counties, the elevations of the water table are similar to those of the potentiometric surfaces of the underlying Jameco and Magothy aquifers (Como and others, 2015). Higher elevations of the water table tend to coincide with high land-surface elevations such as those present along the glacial moraines (figs. 1, 23, 24; Como and others, 2015).

The Magothy aquifer water levels range from 80 ft above NAVD 88 at well N 11454 in north-central Nassau County to 4.9 ft below NAVD 88 at well N 11634 in southwestern Nassau County (Como and others, 2015). A mound of high water levels in the Magothy aquifer in north-central Nassau County coincides with the mound of high water levels in the underlying Lloyd aquifer (Como and others, 2015). Under public-supply pumping conditions, water levels in the Magothy aquifer can be as much as 15 ft below NAVD 88, with small cones of depression (Stumm, 1993).

Most of the monitoring wells in southern parts of Kings and Queens Counties are affected by saltwater intrusion and are not freshwater heads, and some wells in southwestern Nassau County are also similarly affected (Como and others, 2015). The highest water levels were measured in 2013 in the north-central part of Queens County at well Q 812 (Como and others, 2015). Due to the uncertainty of the screen depths at many of the south shore wells in Kings and Queens Counties, both the Jameco and Magothy aquifers are typically combined into a single potentiometric-surface map (Como and others, 2015).

Current Saltwater Intrusion in the Lloyd and North Shore Aquifers

Chloride concentrations from observation wells and estimated peak chloride concentrations derived from EM logs (equation 1) were used to delineate isochlors (fig. 28F). It should be noted that a portion of the thickness of the NSL aquifer may be fresh, even seaward of the isochlors. The chloride concentration in the Lloyd aquifer is normally 5 mg/L or less (Stumm, 2001; Stumm and others, 2002, 2004, 2020). Chloride breakthrough curves suggest that when public-supply wells screened in the Lloyd aquifer reach a chloride concentration of about 50 mg/L, saltwater intrusion may be essentially irreversible (Stumm, 2001). Chloride data from 2000–22 suggest that a 50 mg/L isochlor exists in the Lloyd aquifer in many parts of Long Beach, and possibly to the north of Long Beach in southern Nassau County (fig. 28F).

Hydraulic heads within the Lloyd aquifer measured in 2013 indicate a range of values from 49.5 ft above NAVD 88 at well N 11453 to 3.7 ft below NAVD 88 at well N 11730 (Como and others, 2015). A mound of high hydraulic head in the central part of northeast Nassau County had estimated elevations of 60 ft above NAVD 88 (Como and others, 2015). However, several cones of depression in the Lloyd aquifer caused by public-supply pumpage in northwestern Nassau County were below NAVD 88. Hydraulic heads were also depressed at wells N 13879 and N 13700 on the barrier island of Long Beach due to nearby public-supply pumpage from the Lloyd aquifer (Como and others, 2015). In Kings and Queens Counties, the highest elevation of the potentiometric surface occurred in the central part of Queens County at well Q 3627, which had a hydraulic head of 17.4 ft above NAVD 88, and the lowest elevations of the potentiometric surface were between 5 and 7 ft above NAVD 88 along the north and south coastal areas (Como and others, 2015). Under public-supply pumping conditions, the Lloyd aquifer can have water levels as low as 40 ft below NGVD 29, which is approximately 39 ft below NAVD 88 (Stumm, 1993; Stumm and others, 2002, 2004). Water levels in the Lloyd aquifer near the coast respond to tidal fluctuations (Stumm, 2001).

Due to the hydraulic interconnection of the North Shore aquifer and the Lloyd aquifer, the two aquifers have been combined into a single potentiometric-surface map (Stumm, 2001; Stumm and others, 2002, 2004; Como and others, 2015). In 2013, the North Shore aquifer potentiometric surface was highest in central Manhasset Neck at well N 12321 (22.5 ft above NAVD 88) and was lowest within a cone of depression in northernmost Manhasset Neck at well N 6282 (14.3 ft below NAVD 88) due to nearby public-supply pumpage (Como and others, 2015). In Great Neck, the North Shore aquifer is present in the northernmost part of the peninsula with water levels generally above NAVD 88 (Como and others, 2015). During the summer of 1992, public-supply pumpage resulted in water levels of the Great Neck peninsula to be dominated by multiple, large, interconnecting cones of depression that caused water levels to decline to several feet below NGVD 29 and NAVD 88 (Stumm, 2001). In 2013, the North Shore aquifer was measured at 8.8 ft above NAVD 88 at well K 3410 in Kings County, and was measured at 10.38 ft above NAVD 88 at well Q 577 in Queens County (Como and others, 2015).