The electrical resistivity of the vadose zone is highly dependent on soil moisture content (VWC), making ERT a potential tool for non-intrusive infiltration monitoring. There’s a long history of using Archie’s law to convert resistivity values to VWC, often under the assumption that the pore-fluid conductivity is constant. Analysis of two years of soil conductivity and VWC data from sensors buried in a bioswale in Philadelphia, Pennsylvania showed that the characteristic curve of VWC versus resistivity exhibited hysteresis. The same value of resistivity was associated with different VWC during imbibition and drying. Accurate VWC could be estimated using Archie’s law only after incorporating changes in pore-fluid conductivity. Additionally, pore-fluid conductivity varied seasonally, with the highest values recorded during winter from road salt runoff. Archie exponents differed for sensors within 25 cm of each other despite both the uniform nature of the bioswale soil and fits to Archie’s Law for individual sensors, indicating considerable heterogeneity. This pattern was confirmed at two additional sites. Assuming the conductivity of the pore-fluid remains constant during infiltration not only produces erroneous soil moisture estimates, it can even reverse the apparent soil moisture trend. We conclude that to use ERT to monitor infiltration it is essential to independently record changes in pore-fluid conductivity, and that multiple sensors are required because changes in pore-fluid conductivity and Archie parameters can vary on a sub-meter scale for even apparently homogeneous sites. Additionally, long-term changes in Archie parameters, notably m, are possible, and may indicate long-term changes in soil fabric.

Mineral specific surface area (SSA) increases as primary minerals weather and restructure into secondary phyllosilicate, oxide, and oxyhydroxide minerals. SSA is a measurable property that captures cumulative effects of many physical and chemical weathering processes in a single measurement and has meaningful implications for many soil processes, including water-holding capacity and nutrient availability. Here we report our measurements of SSA and mineralogy of two 21 m deep SSA profiles at two landscape positions, in which the emergence of a very small mass percent (<0.1 %) of secondary oxide generated 36 %–81 % of the total SSA in both drill cores. The SSA transition occurred near 3 m at both locations and did not coincide with the boundary of soil to weathered rock. The 3 m boundary in each weathering profile coincides with the depth extent of secondary iron oxide minerals and secondary phyllosilicates. Although elemental depletions in both profiles extend to 7 and 10 m depth, the mineralogical changes did not result in SSA increase until 3 m depth. The emergence of secondary oxide minerals at 3 m suggests that this boundary may be the depth extent of oxidation weathering reactions. Our results suggest that oxidation weathering reactions may be the primary limitation in the coevolution of both secondary silicate and secondary oxide minerals. We value element depletion profiles to understand weathering, but our finding of nested weathering fronts driven by different chemical processes (e.g., oxidation to 3 m and acid dissolution to 10 m) warrants the recognition that element depletion profiles are not able to identify the full set of processes that occur in weathering profiles.

As part of the reconstruction of Interstate-95 in Philadelphia, a series of vegetated stormwater infiltration basins (bioswales) have been installed to manage highway runoff. To assess these bioswales, we used simulated runoff tests (SRTs), where the bioswale is flooded from a fire hydrant to simulate a major storm event. SRT monitoring relies on point measurements of inflow, outflow, and soil moisture to determine the volume of stormwater the bioswale can handle and the time to recovery. To provide better spatial coverage for site assessment, we tested the use of time-lapse electrical resistivity tomography (ERT) during a SRT. Using an onsite geophysical monitoring station, we performed ERT surveys every 4 h before, during, and after the SRT test. Inflow and outflow measurements taken during the SRT found that a majority of the water did not exit the bioswale via the outlet box. The time-lapse inversion results indicated that runoff uniformly spread throughout the basin before infiltrating into the heterogeneous urban soil below. The ERT demonstrated that the underlying native soil contributed to the overall performance of the bioswale, a contribution which was previously assumed to be minimal. Recovery to pretest soil moisture levels took roughly 2-3 days, according to both soil moisture sensors and time-lapse geophysical data. The SRT results were consistent with natural storm events recorded by our geophysical monitoring station, which have been monitored continuously for over a year. Use of ERT during SRTs characterize the pattern of infiltration and recovery rate of the soil beneath a stormwater control. Additionally, the heterogeneous infiltration observed during the SRT suggested that ERT surveys preconstruction may improve future planning of stormwater controls by guiding the location of infiltration measurements.

Designing green stormwater infrastructure (GSI) requires an accurate estimate of the contributing drainage area and a model for runoff generation. We examined some factors that add to the uncertainty associated with these two design steps in the urban environment. Delineated drainage areas at five GSI sites in southeastern Pennsylvania (PA) were compared for digital elevation model (DEM) resolutions (grid cell sizes) ranging from 8 to 300 cm. The findings point to an optimal DEM resolution range of 30-60 cm, with up to 100 cm resolution providing acceptable results for some sites. The delineated areas were validated with the observed flow and rainfall records at three sites by examining curve number (CN) values calculated for individual storms. The calculated CNs decreased with increasing rainfall volume, which supports a recommendation to consider a range of CNs in the GSI design process. The variation in calculated CNs was higher for the overestimated drainage areas derived from coarser DEM resolutions. We hypothesize that the observed continued decrease of CNs at high rainfall is the result of inlet bypass, a potentially significant factor in urban hydrology. The findings from this study provide insight into the variability in expected delineated drainage areas using standard methods in GSI design.

Bioinfiltration systems are an increasingly prevalent mechanism for urban stormwater mitigation. One major challenge for the sustainability of bioinfiltration systems is erosion and channelization due to high bed shear stresses developed during large storm events. Sedimentation within these systems could also impact their performance as fine sediment may clog pathways necessary for infiltration. Understanding the geomorphology, shear stress, and sediment flux in the system can help predict maintenance needs associated with erosion and deposition. The current study introduces a framework for addressing this problem by combining a sediment transport model, FaSTMECH, with the Green-Ampt infiltration model. A comparison of observed and predicted ponding depths shows very good agreement (median Nash–Sutcliffe efficiency coefficient = 0.93) and demonstrates the ability of this novel framework in predicting the hydraulics and morphology within a bioswale bioinfiltration system. The framework introduced in this study opens the door to understanding sediment transport dynamics within a bioswale, which has the potential to advance planning and design to minimize impacts due to excessive erosion or deposition within bioswale bioinfiltration systems.

Rayleigh waves employed in engineering geophysical applications are typically generated by vertical strikes on a metallic plate that serves as a coupler between an active input source (e.g., a sledgehammer) and the ground surface. Seismic surface waves can also be generated with horizontal impacts (i.e., Love waves) using specialized coupling plates. Survey design, including the source characteristics and data acquisition parameters, can influence the acquired signals and the information derived from their analysis. Despite a growing body of research on Love wave testing, much of the basic information regarding optimal Love wave field data acquisition including source effects, near-field limitations, and receiver spacing remains unquantified. The objective of this study was to improve understanding of Love survey parameters and their effect on resolution. Rayleigh and Love waveforms were collected with multiple active seismic sources at three sites and a systematic comparison was made between the two types of waves. Also, seismic wave propagation was simulated using the code SPECFEM2D to further investigate their differences. The results revealed essential information about the role of different source and survey parameters on surface wave data. For example, the minimum source offset to avoid near-field effects was comparable for Rayleigh and Love waves (0.3–0.4 of the maximum wavelength). At closer source offset locations, Rayleigh waves were more affected by near-field effects and showed an additional 10% underestimation of phase velocities. The results from this case study provide new insights about survey design and its effects on surface wave testing.

The Pennsylvania Department of Transportation (PennDOT) is expanding Interstate 95 near Philadelphia, one of the most heavily trafficked highways in the United States. As part of the expansion project, a series of vegetated stormwater infiltration basins (bioswales) are being installed along the I-95 corridor to manage highway runoff. To assess these bioswales, we use simulated runoff tests (SRTs), where the bioswale is flooded from a fire hydrant or water truck to simulate a major storm event. Typically, SRT monitoring relies on point measurements of inflow, outflow, and soil moisture to determine the volume of stormwater the bioswale can handle and the time to recovery. To improve site assessment, we tested the use of time-lapse electrical resistivity surveys during an SRT. Using an on-site geophysical monitoring station, we performed Wenner-Schlumberger resistivity surveys every four hours before, during, and up to a week after the SRT test. Timelapse inversion results indicate that within minutes the runoff uniformly infiltrates the basin fill, then water ponds in the basin before it infiltrates into the heterogeneous urban soil below. After a few hours, some discharge was recorded at the outlet, but most of the simulated runoff infiltrated; however, only a very small amount of infiltration was retained by the basin fill. Recovery to pre-test conditions in the subsurface took roughly 4-5 days. These results are consistent with real storm events recorded by our geophysical monitoring station, which have been collected continuously for over a year. Future work involves assessing the role of vegetation and evapotranspiration, as well as performing a full water balance analysis.

There are a significant number of bridges for which information regarding the foundation is missing or incomplete. It is extremely challenging to evaluate the performance of such unknown foundation bridges (UFB), particularly against scour or when their foundations are reused. For critical UFB, it is often necessary to estimate performance by developing an appropriate working model of subsurface foundation conditions. Typically, nondestructive testing (NDT) has served as the most viable alternative due to the costs and risk associated with excavation, coring, and probing. NDT can be quite difficult to perform and their results can contain significant uncertainty in highly urban settings. The primary objective of this study was therefore to compare performance of borehole NDT methods when evaluating the depth of two in-service unknown foundations (concrete-filled steel pipe piles and H-piles) in highly urban settings. The borehole magnetic, parallel seismic, borehole sonic, and borehole radar methods were implemented to determine the foundation bottom locations. Though uncertainty was present in all measurements, the borehole magnetometer and radar results proved the most conclusive. Parallel seismic testing did not yield any evidence of foundations due to issues with background noise and lack of direct access to the foundation. Likewise, borehole sonic testing was generally inconclusive due to issues with sensor directivity and attenuation. Borehole magnetometer estimated the depth to the foundation bottom as 8.6 m and 9.2 m at the two sites. Borehole radar estimates for the depth to foundation bottom compared favorably at 9.8 m and 8.0 m for the two sites. Given the borehole construction and depth to competent rock at the sites, these results for borehole magnetometer and radar were likely a minimum estimate for the location of the foundation bottoms. Such information can help evaluate long-term performance of this system as part of rehabilitation and reuse efforts.

The role of preferential flow through unsaturated saprock (fractured bedrock with weathering restricted to fracture margins) in hillslope hydrology remains inadequately described. To address this issue, ground-penetrating radar (GPR), controlled infiltration, and high-frequency subsurface moisture monitoring were integrated to characterize saprock preferential flow (SPF) in the Susquehanna Shale Hills Critical Zone Observatory in Pennsylvania, U.S.A. In a planar hillslope with shallow fractured shale bedrock (starting at 0.1-0.3 m below ground), two pulses of water (79.5 L in total) were released followed by time-lapse GPR surveys. Differentiating GPR images collected before and after infiltration revealed lateral SPF in the direction of bedding plane fractures near the infiltration trench but with limited development of SPF down gradient along the hillslope. This was confirmed by soil and saprock moisture monitoring at a soil pit (0.2 m downslope of the GPR grid) where only one out of fifteen probes responded to the controlled infiltration. Lateral SPF frequently occurred at the GPR grid during a 24-day period with ten rainfall events, especially under the wet initial conditions. Additional infiltration experiments in a convex hillslope and a nearby bare slope with exposed saprock demonstrated the impact of fracture patterns on the routing of SPF. Three types of SPF in hillslope hydrology were identified, including (1) vertical percolation, (2) exfiltration from saprock to soil, and (3) stormwater transported downslope from planar and convex hillslopes to concave hillslopes. A 3D fill-and-spill model is proposed for the study site and similar areas that recognizes the importance of subsurface flow networks, with shallow saprock and concave hillslopes as essential controls of hillslope subsurface flow.

Core Ideas Lateral flow patterns revealed by dye tracing agreed with time‐lapse GPR data. Lateral flow downslope varied from <1 to 1 m at adjacent sites. 3D radar detected banding in the strike direction but dye fingering did not. Fractured saprock modifies the fill‐and‐spill model by facilitating vertical flow. Preferential flow (PF), which bypasses large portions of the soil or subsurface matrix, is critical in the transport of water and dissolved constituents in the unsaturated zone. To test the “fill‐and‐spill” model of hillslope hydrology that describes the generation and pattern of downslope lateral PF after storms, we used dye tracer and time‐lapse, ground‐penetrating radar (GPR) on a forested hillslope in the Susquehanna–Shale Hills Critical Zone Observatory. We injected 50 L of water mixed with Brilliant Blue dye (4 g L−1) into a shallow trench cut perpendicular to the slope and used GPR to monitor the tracer downslope across a 1.0‐ by 2.0‐m grid. The site was then excavated to the soil–saprock interface and photographed to document the dye pathways. We observed vertical dye fingering near the infiltration trench. Downslope lateral PF at the soil–saprock boundary was limited to ~0.40 m, which is evidence that the soil–saprock interface did not fill‐and‐spill. The extent, depth, and direction of the downslope PF indicated by GPR generally matched the dye staining patterns in the excavation, but the resolution of the 800‐MHz GPR antenna was insufficient to distinguish small fingers of dye. A revised fill‐and‐spill model was proposed for this site that incorporates the PF through fractured saprock before water encounters fresh bedrock surface. This study demonstrates that GPR integrated with dye tracer infiltration can provide a useful means of testing hillslope hydrological hypotheses and unraveling the complexity of PF at the hillslope scale in a field setting.

Preferential flow (PF), which bypasses large portions of the soil or subsurface matrix, is critical in the transport of water and dissolved constituents in the unsaturated zone. To test the “fill-and-spill” model of hillslope hydrology that describes the generation and pattern of downslope lateral PF after storms, we used dye tracer and time-lapse, ground-penetrating radar (GPR) on a forested hillslope in the Susquehanna-Shale Hills Critical Zone Observatory. We injected 50 L of water mixed with brilliant blue dye (4 g/L) into a shallow trench cut perpendicular to the slope and used GPR to monitor the tracer downslope across a 1.0 m × 2.0 m grid. The site was then excavated to the soil-saprock interface and photographed to document the dye pathways. We observed vertical dye fingering near the infiltration trench. Downslope lateral PF at the soil-saprock boundary was limited to ~0.40 m, which is evidence that the soil-saprock interface did not fill and spill. The extent, depth, and direction of the downslope PF indicated by GPR generally matched the dye staining patterns in the excavation, but the resolution of the 800-MHz GPR antenna was insufficient to distinguish small fingers of dye. A revised “fill-and-spill” model was proposed for this site that incorporates the PF through fractured saprock before water encounters fresh bedrock surface. This study demonstrates that GPR integrated with dye tracer infiltration can provide a useful means of testing hillslope hydrological hypothesis and unraveling the complexity of PF at the hillslope scale in the field setting.