The surficial geology of the study area was mapped through analysis of (1) borehole lithologic logs and field observation from Kappel and others (1996); (2) borehole and test-pit lithologic logs, field observations, and electromagnetic and seismic data collected during this study; (3) light detection and ranging (lidar) imagery; (4) data from the digital Soil Survey Geographic Database (U.S. Department of Agriculture, 2019); and (5) previous surficial mapping by Pair and others (2007a, b).

The multichannel analysis of surface waves seismic method was applied to determine the shear-wave velocity of the shallow sediments (Park and others, 1999). Shear-wave velocity is a measure of the rigidity or stiffness of a sediment and is related to the grain-size distribution, density, and geotechnical properties (such as liquefaction potential) of that sediment (Hussien and Karray, 2015). For example, soft, liquefiable clayey silt has low shear-wave velocities, and stiff, sandy gravel has high shear-wave velocities. Shear-wave velocity helped distinguish between the effects of lithologic texture and salinity on subsurface electrical conductivity. Shear-wave velocity data were collected on November 16–18, 2021, at 18 sites along the proposed stream-relocation paths and at 3 sites in the offline settling basin area (fig. 4).

The shear-wave velocity data were collected using a single channel seismograph and an array of as many as 14 Atom vertical-component geophones. At each data-collection site, the geophones were spaced in a line 2 meters (m) apart for an initial survey length of 26 m. At the end of each line, strike plates were placed at an offset of 8 m from the first and last geophone. Anthropogenic sources of noise were minimized by waiting for vehicle traffic to pass before starting data collection and standing still during data collection. Starting at the near end of the array, a sledgehammer was used to strike the plate at least three times. Each sledgehammer impact (shot) was separated by at least 5 seconds. Shots were stacked in postprocessing. Shots were repeated at the far end of the line. A roll-along geometry was used to extend the survey by relocating the first two geophones and attached seismographs to the far end of the line with the same 2-m spacing. With this geometry, two new sets of shots were taken by maintaining the 8-m offset from the new positions of the first and last geophones. The roll-along process was repeated a total of eight times for all seismic lines for a total profile length of 58 m.

Raw data from the Atom units were processed by using the Seisimager/SW software suite version 1.27.0 from Geometrics, Inc. The common midpoint cross correlation method (Hayashi and Suzuki, 2004) was used to group correlated traces that have a common midpoint from each shot gather within the Pickwin software module version 7.1.0.0 of Seisimager/SW. Each common midpoint cross correlation gather was combined to generate a phase-velocity diagram where surface-wave phase velocities were plotted against their corresponding frequencies. For each wave frequency, there are multiple possible velocities. The slowest velocity, called the fundamental mode, was identified for each phase-velocity diagram by using automated picking in the Pickwin module. The picks were then manually corrected, cleaned, and exported to the WaveEQ module version 7.1.0.0 of Seisimager/SW as a fundamental mode dispersion curve.

Data were inverted by using the least squares method with five to eight iterations, guided by a horizontal regularization constant of 0.3. Higher regularization values cause more horizontal smoothing in the final model, and the value of 0.3 was chosen to balance smoothing with horizontal variations in shear-wave velocity. Final model plots represent the modeled shear-wave velocity structure across the length of each line. Data were recorded at a 4-m horizontal resolution and varying vertical resolution, from about 0.5 m in the shallow parts to a 5-m vertical resolution in the deep parts of the model. The shear-wave velocity data and models are available from the associated USGS data release (Terry and others, 2026).

Frequency-domain electromagnetic surveys provide a nearly continuous measurement of electrical conductivity in the shallow subsurface. The major factors affecting electrical conductivity in sediments are their silt and clay content, their saturated porosity, and the salinity of the groundwater. Frequency-domain electromagnetic surveys were conducted by using GEM–2 and DUALEM–421 surface-geophysical instruments. The GEM–2 is a multiband electromagnetic instrument consisting of a Global Positioning System (GPS) receiver, electronics, and a battery mounted to a rigid ski about 6 ft in length in which a transmitter and receiver coil are embedded at a fixed offset (5.45 ft). The transmitter emits an electromagnetic field at multiple frequencies that induces eddy currents in subsurface conductors, such as groundwater with elevated salinity, which produces secondary electromagnetic fields that are measured by the receiver. The instrument was carried flat (horizontal coplanar) at waist height (about 3.3 ft) at a comfortable walking speed (about 3 miles per hour). Data were logged in real time every 0.1 second and uploaded via Bluetooth connection to a Trimble Nomad by using Geophex WinGEM2 software. Raw binary data were exported to comma separated value format by using the Geophex EMExport software. About 57 linear mi of GEM–2 data were collected in the study area during May 24, 2018, September 27–30, 2021, and November 15–17, 2021 (fig. 5). The GEM–2 data and inversions are available from the associated USGS data release (Terry and others, 2026).

The DUALEM–421 is a multioffset electromagnetic instrument consisting of electronics and batteries mounted in a tubular mast with a total length of about 15 ft. In this study, the instrument was carried by two people to maintain a constant ground offset and orientation. This system consists of three receivers offset from the transmitter by 4 m, 2 m, and 1 m, each with two receiver coil orientations (in plane with the transmitter coil and perpendicular to the transmitter coil) that operate at a fixed frequency of 9 kilohertz. Data and GPS information were logged at a comfortable walking speed (about 3 miles per hour) with a Bluetooth-connected Android mobile device (Android LG Phoenix 5) running the DUALEM application. Raw data were exported to comma separated value format by using the application. About 19 linear mi of DUALEM–421 data were collected in the study area during November 15–18, 2021 (fig. 6). The DUALEM–421 data and inversions are available from the associated USGS data release (Terry and others, 2026).

Raw electromagnetic geophysical data are sensitive to subsurface bulk electrical conductivity, and the different frequencies and transmitter-receiver separations and orientations effectively sense different volumes of the subsurface; however, data points are not uniquely associated with a specific depth. To obtain depth-dependent electrical conductivity images, the process of geophysical inversion (iterative, numerical model-fitting approach) is required. The electromagnetic data were inverted in Aarhus Workbench version 6.5.1.0 by using appropriate import configuration and processing settings for each instrument. Terry and others (2026) provide further details of the import, processing, and inversion settings used and output as georeferenced models of subsurface earth conductivity to a depth of 49 ft bls. Due to the inherent physical limitations of the instruments, the electrical conductivity contrasts within the subsurface itself, and necessary constraints applied in the inversion, these conductivity models are blurry (smooth) and only interpretable to some fraction of the total 49 ft bls. Workbench performs a sensitivity analysis during the inversion to provide an estimate of this maximum interpretable depth at each location, known as the depth of investigation. In general, the depth of investigation was about 15 to 25 ft for the DUALEM–421 instrument, and about 10 to 20 ft for the GEM–2 instrument. Inversion results from the DUALEM–421 instrument are presented in this report as specific depth-slice maps and vertical sections of electrical conductivity. Owing to the shallow depth of investigation of the GEM–2 survey, only the cumulative instrument response (measured sum of quadrature values) is presented in this report. The GEM–2 survey provided an independent check of the shallow DUALEM–421 response and an increased areal coverage for surficial geologic mapping purposes.

Well drilling, sediment sampling, and standard penetration tests (SPTs) were completed at eight sites during April 11–14, 2022, to help characterize subsurface lithologic texture and stiffness of the sediments in the study area (fig. 4). Seven of the sites were along or near the proposed stream-relocation paths, and the eighth site was at the northwestern edge of the proposed offline settling basin area. The wells were drilled with 8-inch (in) diameter hollow-stem augers. Sediment sampling and SPTs were completed in accordance with standard procedures and equipment described by ASTM International (2022): a 2-in diameter, 1.5-ft long split-spoon sampler and 140-pound automatic hammer are used to obtain nearly continuous sediment samples and measure the resistance of the sediments to the penetration of the sampler. During each test, the number of hammer blows were recorded for three 0.5 ft intervals. The standard penetration test number of blows (SPT–N) value was calculated as the number of hammer blows for the second and third 0.5 ft intervals of penetration. SPT–N values were used to classify fine-grained sediments as very soft (< 2), soft (2–4), medium stiff (5–8), stiff (9–15), very stiff (16–30), or hard (>30) (ASTM International, 2022) and coarse-grained sediments as very loose (<4), loose (4–10), compact (10–30), and dense (>30) (Meyerhof, 1956). In very soft sediments, sampler penetration commonly required no hammer blows, and the sampler advanced under the weight of the hammer.

Penetration-test data collected during investigations in the 1990s were also included in the analysis. Although a similar method to that described by ASTM International (2022) was used, the data do not meet the ASTM International standard for calculation of SPT–N values. The 1990s penetration test results, which are reported in blow counts per foot of penetration, were collected by using a rotating cathead and rope to raise and drop a weight rather than an automatic hammer. The data are more varied but are roughly comparable to the SPT results. In this report, data collected for this study in accordance with ASTM International (2022) are recorded as SPT–N values and historical data collected during investigations in the 1990s are referred to as “hammer blows per foot.”

For depth intervals not sampled on a continuous basis, split-spoon samples were collected when the auger drilling indicated a probable change in lithology. The split-spoon samples were described in the field by texture (very fine to very coarse), sorting, color, and presence of laminations and moisture. Split-spoon samples collected from wells OD3221 and OD3226 at depths of 30 to 32 ft bls were placed in plastic bags and submitted to Kenney Geotechnical Services of Syracuse, New York, for laboratory analysis of moisture content and grain size. Two Shelby-tube samples were collected from well OD3219 at depths from 12.5 to 16 ft bls and submitted for laboratory analysis of moisture content and unconfined compressive strength in accordance with procedures and equipment described by ASTM International (2015; 2016). Very soft soils have unconfined compressive strength values less than 500 pounds per square foot (lb/ft²); soft soils have unconfined compressive strength values from 500 to 1,000 lb/ft²; and stiff soils more than 1,000 lb/ft² (Singh, 2020). Split-spoon sample descriptions and drilling notes, SPT data, and soil geotechnical properties from the wells from this study and previous studies are available from the associated USGS data release (Terry and others, 2026) and USGS GeoLog Locator (USGS, 2024a).

On May 5, 2022, test pits OD1998 and OD1999 were excavated with a backhoe to depths of 10 ft bls for field observations of sediment types and for sediment sampling in the proposed offline settling basin area (fig. 4). Sediment samples were collected in plastic bags from both test pits at a depth of 9.5 ft bls and submitted for laboratory analysis of grain size and moisture content. Sediment samples were collected in plastic bags from test pit OD1998 at a depth of 2.5 ft bls and test pit OD1999 at a depth of 5.5 ft bls and submitted for laboratory analysis of moisture content and Atterberg Limits. Atterberg Limits establish the moisture content at which fine-grained sediments transition between solid, semisolid, plastic, and liquid states (ASTM International, 2018b; Andrews and Martin, 2000; O’Kelly, 2021). Fine-grained sediments are considered liquifiable where they contain less than 15 percent clay, have a liquid limit of less than 35 percent and plasticity index (liquid limit minus plastic limit) less than 12, and their water content is more than 90 percent of the liquid limit (Seed and Idriss, 1982; Bray and Sancio, 2006; and Bennett and others, 2011). The liquidity index, which is the moisture content minus the plastic limit divided by plasticity index, can be used to classify the consistency of fine-grained sediments as very brittle (<0.00), brittle (0.00–0.25), medium soft (0.26–0.50), soft (0.51–0.75), very soft (0.76–1.00), and liquid (>1.00) (Venkatramaiah, 2006). Descriptions and geotechnical properties of the sediment samples from the test pits are available from the associated USGS data release (Terry and others, 2026).

Four piezometers and eight wells were installed as part of the study (fig. 4; table 1). The piezometers were installed by the drive-point method (Waller, 1994) along the southern reach of the proposed far east stream relocation path and provided access for groundwater level and specific conductance measurements (fig. 2). Piezometer OD3227 was installed on October 1, 2020, and piezometers OD3228, OD3229, and OD3230 were installed on November 18, 2021. The piezometers were constructed with 3/4-in diameter steel casing and 1-ft long screens.

Seven wells were installed along and near the proposed creek-relocation paths on April 11–14, 2022 (table 1). A single well (OD3219) was installed at the northwestern edge of the offline settling basin area on April 13, 2022. The wells, which provided access for geophysical logging, hydraulic-slug testing, and groundwater monitoring, were constructed with 2-in diameter polyvinyl chloride (PVC) pipe, 5- to 10-ft long screens, and as much as 10-ft long sumps. The annular space surrounding the screens were filled with sand, and the space above the screens was filled with bentonite.

Gamma and electromagnetic-induction (EMI) logs were collected from the eight wells on May 24, 2022, to provide information on grain size and groundwater salinity. The gamma and EMI logs of the wells are available in Log ASCII Standard format (LAS 2.0) from USGS GeoLog Locator (USGS, 2024a). Gamma logs measure the total gamma emission produced by naturally occurring isotopes of potassium, thorium, and uranium in sediments surrounding the well. Gamma logs were collected from the wells with a Mount Sopris Instruments gamma probe by using logging procedures described by ASTM International (2018a). Gamma activity is generally assumed to be related to grain size: gamma emissions (recorded in counts per second) increase with clay content. However, sediments that contain sand- to clay-sized grains that are dominated by quartz mineralogy with negligible clay alteration minerals, as is the case in the study area, have a reduced range of gamma response and overlap for the different grain sizes (Crow and others, 2017).

EMI logging provides a nearly continuous vertical profile of electrical conductivity of the sediments and pore water surrounding the well. The major factors affecting electrical conductivity in unconsolidated deposits is their silt and clay content, their saturated porosity, and the salinity of the groundwater. A W&R instruments three-coil induction probe that has intercoil spacings of 0.5 m and 0.8 m and operates at a frequency of 100 kilohertz was used for EMI log-data collection. The conductivity readings from the two intercoil spacings tracked each other and only that from the 0.5 m is presented. EMI logs were collected from the wells by using logging and calibration procedures as described by ASTM International (2017). The EMI probe was lowered to the bottom of the well and the probe electronics were allowed to equilibrate with the groundwater temperature (10 to 20 minutes depending on the outside air temperature). After equilibration, EMI data were collected at 0.1 ft deep intervals with the probe moving in an upward direction at a rate of 10 ft/min. Probe calibration checks were made after logging and removal of the probe from the well, during which the temperature of the probe electronics was assumed to be close to the groundwater temperature. Probe readings were recorded while holding the probe away from any metal in free air with an assumed electrical conductivity of 0 millisiemens per meter (mS/m) and while a plastic ring with an embedded wire loop that has a factory-calibrated electrical conductivity of 100 mS/m was placed at the midpoint of the intercoil spacing. Check measurements were within the accuracy range of the probe, which is ±5 mS/m.

Hydraulic-slug tests were completed in the eight wells on May 24, 2022, to estimate the horizontal hydraulic conductivity of the screened sediments. Slug-test design, procedures, and equipment are described by Cunningham and Schalk (2011). A Midwest Geoscience Solid H(o) Slug was used for the study tests. The slug has tapered ends to minimize the pressure wave and is designed to displace 1 ft of water in a 2-in diameter well. The slug was inserted for slug-in tests and removed for slug-out tests rapidly and smoothly. A Solinst levelogger nonvented pressure transducer with a 33-ft range was used to record water-level displacement at 1-second intervals. Three sets of manual measurements were made with a Solinst water-level meter referenced to top of casing to ensure static water-level conditions prior to slug insertion, to check transducer readings during the tests, and to determine when the tests could be ended.

The slug-test datasets were analyzed by using the Bouwer and Rice solution method (Bouwer and Rice, 1976; Bouwer, 1989) in AQTESOLV software (Duffield, 2007) to estimate hydraulic conductivity of the tested sediments. A 10:1 ratio of horizontal to vertical hydraulic conductivity was assumed for the analysis. The thickness of the tested aquifer was approximated as the difference between the top of screen depth and the water-level depth. The slug test for well OD3219 provides a representative example of the displacement data and match fit of the Bouwer and Rice analytical solution (fig. 7).

The resulting hydraulic-conductivity estimates for the wells are in table 1 and the time-displacement data are published in the associated USGS data release (Terry and others, 2026).

Groundwater-level and specific-conductance data were periodically collected from the piezometers and wells after their completion. Water levels in piezometers that were above the top of casing were measured by using clear plastic tubing connected to the casing and a calibrated measuring rod. Water levels below top of casing in piezometers and wells were measured by using a Solinst water-level meter. Water samples were collected directly from the piezometers with flowing artesian conditions for field measurement of specific conductance with a YSI Incorporated conductivity meter. A bailer was used to purge and collect water samples from the wells and nonflowing piezometers for field measurement of specific conductance with a YSI Incorporated conductivity meter (USGS, variously dated).

Groundwater levels were continuously monitored in the eight wells from April 23, 2022, to April 25, 2023. Specific conductance and temperature also were continuously monitored during the same period in the screen zones of wells OD3221, OD3222, and OD3223. Solinst Levelogger sensors were used to record pressure and temperature, and Solinst Levelogger LTC sensors were used to record pressure and specific conductance. A Solinst Barologger installed in well OD1932 in the southern end of the Tully Valley was used to compensate all water-level readings for atmospheric pressure fluctuations. Pressure, barometric-pressure, and specific-conductance data were recorded at 15-minute intervals. Periodic discrete water-level measurements were made in the wells and compared with that measured by the vendor-calibrated pressure transducers. The vendor calibration of the specific conductance sensors was checked with standards from the National Water Quality Laboratory after removal from the wells and corrected using standard procedures (USGS, 2019).

Water levels measured in wells OD400, OD401, OD404, OD407, and OD408 during 1995–2010 as part of a previous investigation provided some historical long-term information on groundwater conditions in the study area, including information from an extreme flooding event in January 1996 (Lumia, 1998). The discrete and continuous groundwater-level data collected as part of previous and current investigations are available from the USGS National Water Information System (USGS, 2024b). The specific-conductance and temperature data collected as part of the current investigations are published in the associated USGS data release (Terry and others, 2026).

Four proposed creek-relocation paths were targeted for investigation, each named for its location in relation to Onondaga Creek: far east (FE), near east (NE), near west (NW), and far west (FW) (fig. 13). The shallow hydrogeologic frameworks along each of the four proposed creek relocation paths are discussed on a reach-by-reach basis (fig. 14).

The proposed offline settling basin area is in the northern part of the Rattlesnake Gulf alluvial fan and would be about 4,500 ft long and 1,300 ft wide (figs. 2 and 8). The center of this rectangular area is about 3,300 ft north of Otisco Road, 1,700 ft south of Nichols Road, 1,700 ft east of Tully Farms Road, and 1,600 ft west of Onondaga Creek and New York State Route 11A. The southeastern corner of the area is about 1,500 ft north of the mudboil corridor and 500 ft west of where Rattlesnake Gulf and Rainbow Creek flow into Onondaga Creek.

Natural and human-made diversions of Rattlesnake Gulf have resulted in saturated conditions in the general area of the proposed offline settling basin (fig. 24). The Rattlesnake Gulf channel has aggraded during flooding raising the channel about 3 ft and causing the creek to flow off the aggraded channel to the south and north. The northern diversion flows along the southern and then eastern edge of a farm field and flows north to the area of the proposed settling basins. A human-made diversion of this flow channels the water toward Onondaga Creek just north of the proposed basin area, but varied topography allows some of the water to flow in an easterly direction across this area prior to this diversion. A recent attempt to rechannel Rattlesnake Gulf creek on the valley floor might help reduce the formerly diverted flow from the creek, but the aggradation process will likely continue with sand and gravel infilling the channel, and the flow returning to the north and south diversion channels unless the newly excavated channel is maintained.

The proposed offline settling basin would be excavated in, and the berms would be constructed on, alluvial-fan deposits and lacustrine fines. The electrical conductivity profile indicates that alluvial deposits are thick and saturated with freshwater below about 10 ft bls near Nichols Road (fig. 25). To the northeast of the section by Nichols Road, well OD435 penetrated brown alluvial gravel from 0 to 30 ft bls and massive red clay grading to silt from 30 to 104 ft bls. Eight split-spoon samples of the lacustrine fines collected from 30 to 46.9 ft bls had moisture contents from 26.8 to 44.6 percent, liquid limits from 28.8 to 45.3 percent, plastic limits from 16.2 to 27.4 percent, plasticity indexes from 10.5 to 23.4, and liquidity indexes from 0.8 to 1.2 (table 3).

The thickness of the alluvial deposits decreases to the south, and at the northern border of the proposed area, there is less than 10 ft of sand and gravel overlying the lacustrine fines. Well OD3219 at the northwestern corner of the proposed area penetrated sand and gravel with silt from 0 to 7 ft bls; silt with many sand stringers from 7 to 10 ft bls; and massive silt with some sand and gravel stringers from 10 to 17 ft bls. The SPT–N values at the well site indicate that the lacustrine fines were soft to very soft (fig. 26). The unconfined compressive strength of the sediments at 12.5 to 14 ft bls was 481.5 lb/ft² and that at 14.5 to 16 ft bls was 905.7 lb/ft² indicating very soft to soft soils. During 2022–23, the hydraulic head measured in the well, which was screened from 2 to 7 ft bls, ranged from 0.9 ft als to 1.5 bls. Maximum specific conductance of groundwater sampled from the well was 680 µS/cm. The estimated hydraulic conductivity of the screened sediments was 10 ft²/d.

The electrical conductivity along the western border of the proposed basin indicates that the alluvial deposits become thinner and finer grained between well OD3219 and test pit OD1999, and that the maximum depth to freshwater saturation is about 5 ft bls. Test pit OD1999 was excavated into topsoil from 0 to 2 ft bls; brown silty clay with small pebbles from 2 to 4 ft bls; wet brown silty clay with rounded pebbles and cobbles from 4 to 5 ft bls; gray silty clay from 5 to 7 ft bls; gray silty clay with trace pebbles from 7 to 9 ft bls; and gray clayey silt with very fine to fine sand (appears to be liquefiable) from 9 to 10 ft bls (fig. 27). A grab sample of the lacustrine fines from 5.5 ft bls had a moisture content of 52.3 percent, liquid limit of 51 percent, plastic limit of 32 percent, plasticity index of 19, and liquidity index of 1.1 (table 3).

The electrical conductivity along the middle part of the proposed area from test pit OD1999 to test pit OD1998 indicates (1) the presence of only very thin alluvial deposits, which are primarily fine grained, overlying the lacustrine fines and (2) the depth to freshwater saturation shallows from west to east. The localized electrical conductivity in this section is probably related to the pipeline and not due to the presence of brackish groundwater. Shear-wave velocities determined from seismic lines L21, L19, and L20 are consistent with the fine-grained character of the sediments. Test pit OD1999 was excavated into topsoil from 0 to 1 ft bls; brown clay from 1 to 6 ft bls; saturated gray silty clay and gravel with rounded small pebbles from 6 to 7 ft bls; gray silty clay with trace pebbles from 7 to 9 ft bls; and gray clayey silt with very fine to fine sand (appears liquefiable) from 9 to 10 ft bls (fig. 25). Grab samples of the fines from 2.5 ft bls had a moisture content of 17.5 percent, liquid limit of 34 percent, plastic limit of 22 percent, and plasticity index of 12. Well OD419 near New York State Route 11A penetrated silt from 0 to 7.5 ft bls; clay from 7.5 to 13.5 ft bls; silt from 13.5 to 19.5 ft bls; sand from 19.5 to 25 ft bls; gravel from 25 to 26.5 ft bls; sand from 26.5 to 31.5 ft bls; silt from 31.5 to 36 ft bls; and grayish bedrock from 36 to 42 ft bls. The well was difficult to grout up when it was decommissioned because of flowing artesian conditions.

Excavation, berm construction, and loading of the saturated, soft to very soft deposits underlying the proposed offline settling basin area may be problematic. Soil strengthening may be needed during the construction of the proposed basin.