Geo-Congress 2019: Engineering Geology, Site Characterization, and Geophysics

Clogging is ubiquitous and inevitable in long-term recharge and has significant influence on the groundwater flow. In this study, a mathematical model of groundwater flow is established to estimate the seepage field induced by single-well recharge. In this model, well clogging is considered under the assumption that the permeability coefficient is exponentially decayed. Moreover, the variable recharge rate and hydraulic head are assumed to be exponentially increasing and are implemented by setting dynamic hydraulic head and flux boundary. The model is solved using integral-transform method. Furtherly, the influences of the permeability degradation, recharge flux, and applied pressure on the groundwater flow and recharge ability are analyzed. Results show that the permeability degradation has considerable influence on the operation capacity of the recharge well and the distribution of seepage field. The influence tends to be more significant with the increase of the operation time. The adverse effect of permeability degradation on recharge can be eliminated to a certain extent by improving the recharge flow and pressure.

A field exploration study was performed on an inactive coal refuse impoundment in the Appalachian region (USA) to investigate in situ characteristics of fine coal refuse (FCR). A series of standard penetration tests (SPT) were conducted from the crest of the impoundment and relatively undisturbed Shelby tube samples were obtained from various depths within the impoundment to a depth of about 60 meters (m). The SPT blow counts (N₆₀-values) in the FCR ranged from 7 to 15 blows per foot suggesting medium to stiff material consistency. The split-spoon samples revealed intermittent layering over short intervals, with thin layers of coarser FCR separated by layers of the finer very soft FCR. Particle size analyses identified two groupings of material, a “sandy” FCR and a “silty” FCR, with no apparent trend with depth. Index testing identified a few locations where the liquidity index was greater than one. It appears thin layers of coarser FCR material produce blow counts that obscure the presence of the soft layers of FCR. This suggests the SPT is not an effective means to investigate these slurry placed materials, if the intent is to identify flowable, under-consolidated materials.

The push to incorporate green infrastructure (GI) practices in large metropolitan areas has become a major component of many city vision plans. Due to the need to capture and minimize runoff resulting from continued development, organizations such as the Philadelphia Water Department and New York City Department of Environmental Protection (NYCDEP) have been conducting geotechnical investigations at parks, schools, crosswalks, and alongside public roads to repurpose paved and impervious surfaces for improved stormwater management. In general, as part of each city’s green infrastructure program, a geotechnical investigation consisting of soil borings, field permeability, and laboratory index property testing is performed at prospective locations. In many cases, the approval to implement GI at a location is based on adequate field permeability as well as minimal fines content. Based on observation of hundreds of field-permeability tests conducted by our firm, however, it was noticed in many cases that, although there was a percent finer exceeding the prescribed criteria, the in situ field permeability test indicated a level of performance with passing permeability. As both city programs are based on these two acceptance criteria, it is the intent of this paper to use the data collected to-date to help identify that additional qualifiers should be considered to understand the behavior of the soil and long-term performance of GI features. This paper presents research behind an attempt to prepare a refined predictor of permeability and acceptance criteria for regulatory agencies rather current model of only evaluating fine-grained fraction of soil.

Typical earthwork quality assurance/quality control (QA/QC) procedures rely on in situ testing to ensure that the soil being placed meets required project specifications. This process commonly involves the use of a nuclear density gauge to measure the dry unit weight and moisture content of the in situ soil. The determination of soil moisture content is important during earthwork QA/QC, as moisture content has a strong influence on the compaction behavior of a soil, as well is its resulting strength and stiffness. In this study, a low-radiation hybrid nuclear density gauge was utilized, which employs a dielectric based moisture probe to measure soil moisture content. Moisture content measurements made by this device were compared against representative laboratory moisture content test results to develop a moisture content calibration model for the gauge. A calibration approach was utilized that employed an exhaustive testing approach to determine the best possible combination of samples to develop the calibration model. This approach was used to calibrate the gauge for moisture content, and the gauge test results were compared against measurements of moisture content made using drive cylinder testing. The results from this study showed that the exhaustive testing approach to calibration yielded superior end results, as it was able to yield moisture contents that had a relative error of less than 1% versus the drive cylinder’s laboratory determined moisture contents.

Undisturbed sampling of sands is practically difficult or even, at times, completely unfeasible. Hence, the determinations of in situ void ratios and/or relative densities of uncemented sands are generally achieved using correlations with the cone penetration test (CPT) and the standard penetration test (SPT). Different sands of the same relative density may have dissimilar CPT and SPT resistances because of their compressibility differences. Accordingly, the effect of compressibility should be considered while assessing the results of these tests. In this paper, an attempt is made to quantitatively incorporate the sand compressibility into the void ratio and/or relative density correlations. The presented approach utilizes compressibility indicators to quantify the effect of sand compressibility. The proposed relationships are formulated based on a published worldwide sand database that was inferred from the testing of high-quality undisturbed samples. It is demonstrated that these relationships can be utilized to reliably assess the void ratios of sands using the CPT and SPT. Two case studies are analyzed, and the estimated void ratios are found to be in a close agreement with the values inferred from undisturbed samples and other tests.

The legacy of mining in the Salt Lake Valley has left a large mine tailings impoundment centrally located in the Salt Lake City urban corridor, where high real estate prices and prime location couple to make development on mine tailings worth considering. A conventional foundation design approach would have involved installing deep foundations through the variable and highly compressible tailings. However, costs and a complex environmental cap made deep foundations less desirable. So, to design structures for these unique subsurface conditions a detailed test program augmented with full-scale footing load tests was performed. These tests proved that buildings could be constructed on the tailings, without resorting to deep foundations. This paper presents the results of the footing load tests, discusses the potential causes of the unusual settlement behavior observed, and explains the logistics involved in conducting footing load tests and accurately measuring the very small settlement rates observed.

Different data reduction formulas are used with drained, partially drained, and undrained piezocone test (CPTu) data. In practice, drainage conditions are inferred based on a characterized soil behavior type (SBT) from the CPTu profile. Alternatively, drainage conditions around the cone probe can be confirmed from CPTu pore-pressure dissipation (PPD) tests. This alternative is time-consuming and only provides confirmation of the drainage conditions at a discrete point in the profile. This paper reports on the relationship between SBT indices and drainage conditions over a dataset of 2792 PPD tests. We evaluated the predictive capabilities of the SBT indices to infer undrained behavior using a receiver-operator characteristic (ROC) curve. The area under the ROC curve was 75–76%, in-between a perfect metric (100%) and an uninformative metric (50%).

Nearshore areas in front of river inlets feature significant spatial and temporal variations in sediment distribution, transport, and deposition processes. This makes subaqueous sediment characterization of such locations essential for engineering applications and for the fundamental understanding of regional geomorphodynamics. Site investigation was conducted in the nearshore area off the Situk River Inlet, Yakutat, Alaska. This region is characterized by highly variable sediments and energetic wave conditions. In situ tests were conducted along ten cross shore transects with 64 locations in total using a portable free fall penetrometer. The results showed the presence of a loose sediment top layer of 3.8–6.9 cm in thickness overlying a stiffer stratum with a maximum quasi-static bearing capacity ranging from 30–106 kPa at sediment depths of less than 15 cm. More variability in sediment strength was observed with decreasing water depths (approaching the shore), and sediment strength generally somewhat decreased seaward. For transects closer to the inlet, a more distinct trend of decreasing sediment strength with distance to the shore was observed. As expected, the impact of waves on the sediments appeared significant in shallow water depths. The results suggested an influence of hydrodynamics and the presence of the inlet on the topmost sediment layers, and consequently on the geotechnical characteristics.

A number of sinkhole events have emerged in East Central Florida (ECF) and the consequences of its occurrence lead to loss of property and/or human lives. Increasing concerns for the potential risk of sinkholes have raised the necessity of developing a sinkhole hazard map for the region. A sinkhole hazard can be assessed as the probability of occurrence accounting for time, location, and magnitude. This paper presents the development of 1) a probability-based sinkhole hazard model and 2) a spatio-magnitude sinkhole hazard map for ECF within a geographic information system (GIS) environment. Logistic regression (LR) method was used for the spatial probability analysis and a cumulative distribution function (CDF) method was used for the magnitude probability analysis. The developed sinkhole hazard model was validated by existing data. The probabilistic sinkhole map for ECF would provide useful information on future sinkhole hazard in order for decision makers and land-use planners to build suitable strategies for long-term development.

The behavior of soil under loading is strongly influenced, and in some cases governed, by the relative percentage of coarse and fine particles. There are many intermediate graded soils whose response simply cannot be explained by “granular (non-cohesive)” versus “clayey (cohesive)” soil behavior. These soils have been commonly referred to as marginal fills. The limiting fines content when soil behavior changes from coarse-grained to fine-grained-dominated (i.e. fines fraction corresponding to a given behavioral threshold) is critical for developing a better understanding and more effective design decisions. Soil classification systems are beneficial in grouping soils into similar behavioral categories. Yet, there are limitations/drawbacks in currently utilized classification systems, when it comes to classification of marginal fills, since they fail to capture the dominant role of fines on the mechanical and hydraulic properties of soils. In this study, a total of ten sandy soil mixtures with 0%, 12%, 20%, 30%, and 40% fines, and plasticity indexes of 5% and 15% were prepared in the laboratory by mixing different proportions of coarse and fine sands, non-plastic silt, kaolin, and bentonite. All index properties including grain size distribution curves, Atterberg limits, maximum and minimum void ratios, maximum dry density, and optimum water content were determined. The mixtures were classified according to USCS, AASHTO, and revised soil classification system. Comparisons of different classification systems and their limitations were demonstrated for these marginal soils.

The cone penetration test (CPT) is routinely used when characterizing cohesionless soil deposits; however, there is limited understanding regarding its performance in soils with increasing coarse-grained content. A series of centrifuge tests on coarse-grained soil mixtures was conducted at the UC Davis Center for Geotechnical Modeling to study the effects of soil gradation on CPT tip resistance (q_c). Four uniformly-graded soils with mean grain diameters (D₅₀) ranging from 0.18 to 2.58 mm were used to construct three soil mixtures: two uniformly-graded and one well-graded. All soils are of alluvial origin and were not crushed during processing; thus, any variations in minerology or particle shape are consistent with natural deposition. CPT tip resistance was dependent on probe diameter (D_CPT) and relative density (D_R), with lower q_c values typically occurring for the cone with the low D_CPT at high D_R. Additionally, CPT q_c was not uniquely related to relative density. Both particle size and packing efficiency effects resulted in different tip resistance values for each soil mixture.

Sinkholes extensively occur in China and all over the world. The sinkholes are paroxysmal, concealed, and often cause huge economic loss and human fatalities. Disintegration of the red clay is one of the major reasons contributing to sinkholes in the karst areas of southern China. Previous studies mainly focused on the mechanism of disintegration, while effects of the acid rain on the disintegration of the red clay were not fully studied. This paper investigated the effects of the acid rain on the disintegration of the red clay using a new approach. By comparing the pH values, concentrations of the oxide, specific surface areas, and other parameters for the red clay before and after the disintegration, the mechanisms of disintegration of the red clay were explored. Test results showed that degree of the red clay disintegration increased with the decrease of the pH value of the acid rain. This trend is especially dramatic when the red clay is unsaturated.

A 270-acre Superfund site that is underlain by soft, compressible mine tailings is centrally located in the Salt Lake City urban corridor where high real estate prices and prime location make commercial development of the site worth considering. Grading of the tailings at capping and closure of the site resulted in a series of drainage swales with grade changes of 10 to 16 feet. In developing the site, a conventional grading approach would balance cut and fill from ridges and valleys of these swales. However, conventional settlement analyses suggest long-term settlement from valley fills and building loads would be unacceptably high for most structures. A series of large test fills were performed to provide a better characterization of the compressibility of the soils at the site and to evaluate the feasibility of developing the site from a settlement perspective. The results of 12 test fills—some monitored over periods as long as 12 years—show how the settlement characteristics of the site have changed over time. Furthermore, a series of load-unload-reload tests resulted in a 95% reduction of settlement for the reload test, which suggests surcharging may be used to reduce long-term settlement from valley fills and building loads to acceptable tolerances. However, the amount and duration of secondary compression remaining after surcharging may require reduced grade raises to make this a viable approach.

The Delaware Department of Transportation (DelDOT) has acquired a CCS900 compaction monitoring system that allows for collection of continuous compaction control (CCC) data. CCC equipment allows contractors to collect near continuous compaction effort data during the compaction of soils. This is advantageous to the earthwork QC/QA process, as CCC data can be collected over nearly 100% of the evaluation area relative to traditional in place “spot” testing, which typically results in less than 1% of the evaluation area being tested. This study investigated the findings from preliminary data collection during active earthwork construction utilizing a compaction roller installed with a CCS900 system that collected two types of CCC readings, compactometer value (CMV), and machine drive power (MDP). Stiffness based in situ measurements, including those taken from a dynamic cone penetrometer (DCP) and light weight deflectometer (LWD), were also collected to investigate the effectiveness of compaction at various locations. Results from this study indicated that the CCC reading CMV was able to identify the presence of underlying soft soil, a discovery which was validated using the DCP and LWD. The results from this study provide valuable insight into the behavior and potential benefits of utilizing CCC monitoring equipment during active construction.

A previously-developed SedImaging test quickly determines the particle size distribution (PSD) of coarse-grained soils. The test involves photographing and analyzing a soil following sedimentation through water to sort its particles by size. Generating a PSD by SedImaging takes a fraction of the time required by traditional sieving. Due to its large size and weight, the original SedImaging device is suitable only for laboratory settings. Therefore, a new FieldSed system has been designed as a field-adaptable version of the original SedImaging system. While it utilizes the same Harr wavelet image analysis methods as the original SedImaging test, FieldSed uses smaller-scale, lightweight hardware to make it practical for use in any testing environment. FieldSed also enables many parallel tests to be performed, thus greatly reducing testing time. The FieldSed-produced PSDs match sieving results very well.

This paper presents a site variability analysis algorithm that takes cone penetration test CPT data as input and quantifies both horizontal and vertical site variability using variability indices. The vertical variability is quantified via the vertical variability index VVI which considers both the inter-layer and intra-layer variability as inferred from CPT sounding data. The horizontal variability is quantified via the horizontal variability index HVI which is calculated considering the spacing between soundings, the cross correlation of tip resistances, and the difference in absolute magnitude of tip resistance between all pairs of soundings within test site. The application of the algorithm to three sites with different in situ variabilities is presented. Results show that the algorithm can adequately differentiate and quantify different levels of in situ vertical and horizontal variability observed at the sites.

Over 40,000 bridges across the United States lack critical information regarding their foundation system, including their geometry, and foundation type. Nondestructive testing (NDT) and geophysical methods have gained attention over the past two decades to characterize such foundations. Despite the surge in popularity of NDT and geophysics for unknown foundations, the multichannel analysis of surface waves (MASW) method has been largely neglected for this purpose. Using a pure numerical framework, the objective of the present study is therefore to compare the effectiveness of MASW in its traditional dispersion-based approach with full waveform inversion (FWI). FWI determines high-resolution velocity models by minimizing the difference between observed and modeled seismic waveforms. In this study, numerical simulation was performed on a model representing an in-service foundation system that consists of a deep foundation system with a pile cap buried below the ground surface. Results indicated that the traditional dispersion-based MASW approach suffers from limitations when evaluating relatively small and slender anomalies embedded in the subsurface. This method cannot see the depth and extent of the buried piles nor can it provide concrete evidence regarding the existence of the pile cap. On the other hand, the FWI approach significantly outperformed traditional MASW in detecting the location and dimensions of the pile cap.

The Kallang formation alluvial member (Ka) in Singapore was mainly deposited in river channels during middle or late Holocene. It carries information regarding the changes in the distribution of river channels. Identifying the distribution of rivers is important as the presence of rivers are usually associated with weak or complicated geotechnical properties of the soil deposits below. However, some ancient rivers are buried beneath the superficial formation layers and cannot be directly observed in the contemporary era. The objectives of this paper are to present a method to identify the distribution of rivers using a 3D geological model and use it to investigate the distribution of ancient rivers in Singapore. A 3D geological model in the Rochor River area of Singapore was firstly established using borehole data. The procedure adopted for the 3D geological modelling involved in geological data preparation, cross-section, and fence diagram creation, and 3D model construction. The 3D model was then used to identify the distribution of the ancient river channels with the validation of the deposited Ka. The interpretation results indicate that there are ancient rivers existed underneath the Rochor River area as well as a comparison of rivers in the ancient and present time is made.

Below-grade construction in Boston is generally complex. Sites often have to deal with some combination of historic fill, obstructions, compressible organic soils, variable complex marine and glacial geology, shallow groundwater, and urban settings. However, when a development contains all the above, the design, collaboration, and construction become exponentially more difficult. This paper provides the rationale behind the geotechnical design of a specific building within a larger overall site that contains all these conditions. The geotechnical aspects of the design include a 12.2 m (40 ft) deep mass excavation for a basement founded on a soil bearing mat. An atypically and extensive exploration and testing program to obtain site-specific data on the complexity of site conditions and provide appropriate design recommendations was undertaken and is described below. The information collected was used to provide recommendations including foundation design parameters, temporary support of excavation system, construction dewatering/depressurization, permanent subslab (hydrostatic) depressurization, and site surcharging. This paper focuses on the myriad of testing performed and their objectives rather than the actual values obtained and provides a brief discussion on the verification of the basis of design.

Land deformation, such as land subsidence and landslides, is conventionally monitored using extensometers and inclinometers. Distributed fiber optic sensing is an emerging technology that transforms common telecommunication fiber optic cables into sensors capable of making distributed strain measurements at kilometer scales and beyond. Recently, this technology has been extended to monitor subsurface deformation. This paper introduces the principle of the distributed fiber optic sensing technology with a particular emphasis on Brillouin optical time-domain reflectometry and explores its application in two cases (groundwater-related land subsidence in the southern Yangtze Delta and reservoir-induced landslide in the Three Georges Reservoir region) in China. Initial conclusions suggest that this technology can complement with conventional methods for improved monitoring of land deformation. The ability to obtain a complete subsurface strain profile with a single fiber optic cable could improve our ability to constrain analyses and forecast future land deformation.

Landslides are a globally occurring geologic hazard that affect infrastructure such as roadways constructed in steep mountainous terrain. In the Ozark Plateau portion of Arkansas, the combination of sloping bedrock, decomposition of shale into high PI clays, and high rainfall contribute to the formation of a significant number of landslides that impact transportation infrastructure. There are currently 175 active landslides affecting roadways in Arkansas. This case history employed electrical resistivity tomography (ERT) across a recurring landslide zone within this portion of Arkansas to characterize the general stratigraphic architecture of the area. ERT surveys conducted across the slide delineated the interface between the shale bedrock and clay layer, and identified highly saturated regions above the clay/bedrock interface. Overall, the addition of ERT data to landslide investigations provides a faster and more cost effective means to develop a detailed subsurface image of the landslide compared to traditional geotechnical instrumentation alone.

Natural Bridge, in Rockbridge County, Virginia, is a geological arch carrying U.S. Route 11 over Cedar Creek. The area has significant historical and cultural importance, and is listed on the National Register of Historic Places. Until 2015, the arch and area below were privately owned and operated, with only the pavement held by the Virginia Department of Transportation. Since then, the arch and area below have been leased to the Virginia Department of Conservation and Recreation, transferring liability to the Commonwealth. As part of the Commonwealth’s due diligence and to help ensure that the arch is preserved for future generations, the Department of Transportation, in partnership with Radford University, completed a comprehensive, non-invasive geological and geotechnical investigation in 2017 and 2018. A complementary variety of geophysical, laser, optical, seismic, and traditional geological methods of study were used to allow for integrated data analysis. The investigation allowed planning for protection of the environment, cultural resources, and local economies, while maintaining the functionality of the transportation corridor.

Karst describes a landscape caused by the natural dissolving of soluble bedrock, with features including variable depths to bedrock, settlement, and sinkhole development. Karst impacts civil engineering projects as there is an increased risk for settlement and sinkhole development during and after construction, complicated structure foundation design and installation due to variable rock depth and conditions, issues during construction including differing condition claims, schedule delays, and material overruns because of highly variable subsurface conditions. Enhanced site characterizations and karst-specific design and construction practices can be applied to reduce the impacts and risk from karst features. We present case histories of projects in karst where design and construction considerations were recommended or implemented to reduce the risk of settlement, sinkhole development, and cost overruns. Two case histories are presented from projects in Pennsylvania, including a major regional airport and one of the largest shopping malls in the country.

Shale is the most frequently encountered material in civil infrastructure works. It constitutes about 70% of subsurface materials around the world. These materials cause problems because they tend to degrade with time due to weathering. Degradation results in the shale deteriorating from a hard rock-like material with high frictional resistance to a soft fine-grained soil mass with lower shear strength and high deformability. Therefore, common problems associated with embankments constructed with compacted shales include excessive settlement, and local and global instabilities. This paper presents the results of consolidated undrained triaxial tests performed on compacted weathered shale at different compaction energies. The triaxial tests analysis showed that compacted shale can behave as normally consolidated material, slightly over consolidated material, or heavily over consolidated material depending on the energy of compaction and the effective consolidation pressure. Also, for this work characteristics of the stress paths were analyzed in order to evaluate the shear strength parameters obtained from various interpretations of shear strength.

Catastrophic tailings dam failures have resulted in significant social, economic, and environmental losses. The geotechnical properties of coal tailings have been investigated in the past using reconstituted samples or relatively undisturbed samples. To realistically assess the stability of coal tailings dams under dynamic loadings, there is a need to determine the in situ properties of coal tailings using non-destructive field testing. In this study, seismic refraction survey and multichannel analysis of surface waves method are performed in two coal tailings impoundments in the Appalachian coalfields in the USA. Shear wave and compressional wave velocities of coal tailings are determined based on the field data. Young’s modulus, shear modulus, and Poisson’s ratio that are critical in understanding the stress-strain relationship of coal tailings are estimated. The Poisson’s ratio of the unsaturated fine coal tailings is determined to be 0.39. The Young’s modulus of the soft fine coal tailings for the inactive site is higher than that for the active site (i.e., 65 MPa versus 35 MPa). The small-strain shear modulus of the fine coal tailings is determined to be in the range of 10 to 50 MPa. The parameters derived from these seismic investigations contribute to the scarce database of geotechnical properties of coal tailings and can be applied to future stability assessment of coal tailings dams.

Groundwater flow in karst terrain (fissured soluble bedrock) often creates underground cavities due to a continuous internal soil erosion. Consequently, raveling of soil occurs and those subsurface cavities grow and reach to the surface, resulting in a sudden collapse. In this study, finite element (FE) numerical simulations were performed to construct sinkhole stability charts that can be practically used by engineers. Two sinkhole sites in central Florida were selected for the analysis, and cone penetration test (CPT) soundings were conducted for characterizing site subsurface condition. The CPT data were used to determine input modeling parameters in the FE simulations. Hardening soil model (HSM) was adopted as constitutive model and the constitutive soil parameters were determined based on published correlations. The shear strength reduction (SSR) method was employed to compute the factor of safety as the size of underground cavity increases. A series of scenario-based simulations were carried out to construct the stability charts. These charts are aimed at helping engineers as an initial check in quantifying the vulneratiblity of sinkhole in the form of factor of safety for a given subsurface condition, which can be obtainable from CPT tests.

The impact of moisture content on the strength of sediments at a sandy beach is explored. Surficial beach sediments with limited variability in size (0.34–0.37 mm median grain size) were tested for moisture contents and quasi-static bearing capacities. Moisture contents and grain sizes were found from push core samples and in situ sediment strength was assessed from deployments of a portable free-fall penetrometer. Moisture contents ranged from 4.5–21%, and estimates of quasi-static bearing capacities varied from 14.3–43.7 kPa over a single transect during one sampling period (~30 minutes). Over a survey period of three days, moisture contents varied from 4.7–19.9%, and quasi-static bearing capacities ranged from 13.0–38.1 kPa for one exemplary location. Higher moisture contents generally resulted in higher strengths in the intertidal zone. With other variables remaining constant, the results indicate a spatiotemporal relationship between sediment strength and moisture content of sandy foreshore sediments.

In heavily traveled New Jersey, a single rockfall event has the potential to cause serious injuries and costly damages. Within a landscape of budget constraints and competing project priorities, the NJDOT, working with the consultant community, has been successful in developing and implementing a proactive rockfall mitigation approach. NJDOT’s Engineering Geology unit and Project Management group have partnered in the development of a multi-year Rockfall Mitigation Program that advances projects for remediation that have been prioritized in NJDOT’s rockfall hazard management system. Every rock slope and situation is unique, and there are numerous possible rockfall mitigation techniques. In the concept development phase, NJDOT subject matter experts work closely with consulting engineering geology experts to evaluate the rockfall hazards and project constraints, and then build a consensus with the stakeholders regarding the preferred mitigation alternative. This collaborative approach helps streamline the subsequent project phases by reducing or eliminating re-work and facilitating an efficient path to final design and construction to stretch available project dollars.

The paper investigates the mechanical behavior and anisotropic yielding of resedimented Nile silty clay (RNSC) in Egypt. The study uses samples resedimented in a consolidometer, and tested in a series of drained, stress controlled triaxial tests, where those samples are isotropically consolidated. The yield points measured show remaining asymmetry in the yield surface despite the isotropic consolidation. An asymmetric yield function is fitted to the yield data points using a modified stress tensor which accounts for the contacts between particles. The paper delineates the interaction between the inherent anisotropy, resulting from the resedimentation and the isotropic consolidation typically used in laboratory triaxial testing. Results indicate that the isotropic consolidation does not cause full rotation of the yield surface back to a symmetric state. The concept of local changes in the yield surface geometry around the consolidation path can explain the inconsistency between observed yield surface and the undrained stress path.

Naturally occurring sinkholes in Florida are formed when the soluble limestone bedrock weathers and creates cavities at its interface with the overburden finer-grained soils. The overburden soil then naturally erodes into the limestone fissures, driven by concentrated vertical groundwater flow. For the case of many Florida sinkholes, soil overlaying the limestone can internally erode for years before any settlement or collapse is visible along the surface. When internally eroded soil anomalies are encountered during an initial site investigation, engineers are then faced with the complex task of assessing the current health of the soil stratigraphy against sinkhole formation subsequent collapse. In efforts to further assess the severity of internal soil erosion (commonly referred to as raveling), an array of in situ piezometers were installed at a project site in central Florida at which extensive subsurface exploration testing has suggested severe soil raveling has previously occurred. Through analyzing the site’s deviation of piezometric elevation over multiple years, potentiometric depressions were identified at several areas within the project site; suggesting a concentrated vertical flow of groundwater may be occurring. This paper presents an analysis technique implementing these hydraulic gradient deviations with cone penetration testing (CPT) and subsequent sinkhole risk indices, to better understanding of the current state of soil raveling and resulting sinkhole severity at a specific location. This analysis technique was directly applied through a case history of a site in central Florida in which initial subsurface exploration identified abundant karst anomalies (Wekiva Parkway SR46 connector).

Recently in the Lehigh Valley, PA, a 2.5-million square-foot distribution complex in karstic terrain was built in less than one year, thanks in part to the efficient use of multiple geo-techniques during design and construction. A combination of geophysics, karst mapping, and conventional geotechnical intrusive testing was used to characterize the undulating 200-acre agricultural site. Geophysical methods included the use of ground penetrating radar and two-dimensional resistivity to assess the consistency of the subsurface profile, and to map the top-of-rock profile and karst-related anomalies. This enabled the intrusive geotechnical exploration to strategically target areas where geophysical anomalies were detected, and to reduce the overall scope of the geotechnical program. Due to the challenging fine-grained soils, winter conditions, and a compressed schedule, the site development incorporated various techniques to permit construction to proceed, including lime stabilization; cement stabilization of the floor slab subgrades; optimization of heavy-duty asphalt pavements; as well as sinkhole stabilization.

Surface wave testing was performed as part of a sub-surface soil investigation for a bridge replacement project for WYDOT, over the Snake River near Jackson, Wyoming. From borehole logs, the site consists mainly of cobbles and gravels, a challenge for any site investigation. Active and passive surface wave data were collected and analyzed using multi-channel surface wave methods and microtremor array measurement methods, respectively. Passive data were collected using both an L-array and nested equilateral triangular arrays, where horizontal to vertical spectral ratio data was collected as well. The layering ratio method was employed to help analyze the collected data. The site classified as a Site Class D, with an average shear wave velocity over the top 30 m of 321 m/s (1,056 ft/s). This paper presents the data collection and analyses of the shear wave velocity (Vs) profiles, including some lessons learned from investigators new to surface wave testing.

A total of 150 multi-channel analysis of surface waves array setups using Rayleigh waves were performed along the centerline crest and landside toe of the Melvin Price Reach of the Wood River Levee. Reference shear wave velocity (V_S) curves for various soil types, proposed in previous studies, were utilized to resolve the soil type and soil stiffness for both the body and foundation of the levee and then the results were compared to the extensive boring log data to evaluate the accuracy of the reference V_S curves for near-surface site characterization. The results indicate that while the existing reference V_S curves are valuable for soil classification at deeper depths, they fail to accurately predict the soil type and soil stiffness for very shallow depths due to the minimum effect of confining stress on V_S. Modified versions of the reference V_S curves are developed for soft clay and loose sand for very shallow depths, where the current reference V_S curves fall short of correctly predicting soil type and soil stiffness.

A slow moving landslide in Colorado was monitored over a five-year term using periodic measurements from a tripod-mounted, ground-based interferometric radar (GBIR). Ground-based radar interferometry is a relatively new remote sensing technique for scanning large landscapes to measure line-of-sight (LOS) surficial movements. GBIR measurements are performed by collecting multiple radar scans of the region of interest at different times and combining the imagery to detect small phase changes in the reflected signals caused by surface movements. Measurements were performed annually over a five year period with each measurement event consisting of radar scans collected at 5 to 15-minute intervals over time spans ranging from 5 to 72-hrs. The radar scans were then processed to develop maps of displacement rates of the entire landslide mass and displacement time histories of select points over the one to two-day period. In this unique case history, displacement rates are presented at different stages of the landslide over the five year period, including before and after remediation measures were applied. The results illustrate the excellent displacement sensitivity, as well as the spatial and temporal resolution of GBIR measurements. Limitations of the methods are also presented and discussed.

The Hager Slough Levee in west-central Illinois has experienced issues with seepage and sand boils since the 1990s. In 2015, unusual winter storms in the Midwest again caused flooding at the Hager Slough Levee and upwards of 50 sand boils were identified during a site reconnaissance. There is currently limited information on how to assess the extent of damage from the seepage channels that caused these sand boils without extensive drilling and sampling, so many sand boils are left unrepaired and are simply monitored in subsequent flood events. Thus, there is a need for a rapid, nondestructive methodology to evaluate or map the voids caused by seepage channels so that repairs needed to a section of levee can be identified prior to the next flood. The objective of this study is to demonstrate that geophysical methods can be used to identify potential locations and sizes of seepage channels in a levee following a flood. An induced polarization (IP) and two electrical resistivity tomography (ERT) surveys were conducted along the Hager Slough Levee in an area where sand bags remained from the 2015 flooding and the location in which the levee crown had settled. Laboratory electrical resistivity tests on samples collected from the sand boils and nearby boring logs provide support for the field electrical resistivity interpretations. The results of this study indicate that ERT in combination with IP can be used to assess zones of damage from a flood event within and beneath a levee. If used in practice, these methods have the potential to improve remediation of levees by prioritizing where repairs should be conducted and reduce excessive repairs where no damage was noted.

Soft soils are often improved through a grouting technique that injects grout at multiple ground locations. Soils surrounding the injected areas are displaced and compressed and the subsequent consolidation provides additional reinforcement within the soil structure. The overall soil stiffness is hence improved. Although various methods have been traditionally used to evaluate the improvement of the soil stiffness, for example, SPT, CPT, and pressuremeter test, they all provide local results highly dependent on test location. On the other hand, the seismic surface-wave (MASW) method can provide a macro-scale assessment of grouting effectiveness by analyzing wave propagation characteristics that respond to bulk elastic properties. In addition, an MASW survey can produce a cross section of subsurface shear-wave velocity (Vs), which is scientifically the most relevant indicator of material stiffness. Multiple MASW surveys were conducted over an area near Navegantes, Santa Catarina, Brazil, where the Federal highway was being duplicated and soft soils existing in the region were being improved by a special grouting technique locally developed. The survey results showed that, compared to the virgin soil, the Vs increase at grouting depths (2–8 meters below surface) is up to 200 m/sec, which corresponds to approximately 80 MPa and 240 MPa in shear modulus and Young's modulus, respectively. Some key aspects in data acquisition and processing steps are explained that included an active-passive combined survey followed by data analysis using a special algorithm for a full utilization of advantages from both type of MASW surveys.

This paper presents a 3D surface-based full waveform inversion (FWI) method for detection of voids in karst terrain. The FWI method is based on a finite-difference solution of 3D elastic wave equations and Gauss-Newton inversion method. The key advantage of the waveform tomography approach is the ability to generate all possible wave propagation modes of seismic wavefields (body and surface waves) that are then compared with measured data to infer complex subsurface properties. Seismic wavefields are acquired from geophysical testing using sensors and sources located in uniform 2D grids on the ground surface, and then inverted for the extraction of 3D subsurface wave velocity structures. Field experiment was conducted at a karst site in Newberry, Florida, which is a dry retention pond, approximately 1.6 ha in size. The site has been susceptible to sinkhole formation with a number of open chimneys as well as large sinkholes that have formed and been repaired. Field experimental data were collected using 48 vertical receivers and a propelled energy generator (PEG, 40 kg model) on the ground surface to induce seismic wave energy. The 3D seismic inversion results show that the waveform analysis could be able to characterize the highly variable subsurface soil layers and karstic limestone bedrock at high resolution (meter pixel). An identified void was confirmed by SPT results.

In this study, several hypothetical soil profiles were analytically analyzed to assess potential errors associated with two simple, straight-line ray path methods of downhole data analysis: the interval method and the slope method. Seismic velocities obtained from the interval method were often significantly different than the correct hypothetical values. These differences were most significant over the top several meters, but also appeared at greater depths for certain stratigraphy. Moreover, using smaller testing depth intervals to obtain better depth resolution could cause discrepancies at soft-to-stiff layer boundaries with the interval method. The presence of a shallow stiff layer causes the near-surface velocities to generally be underestimated, while a soft surface layer causes them to be overestimated. Results indicate that errors resulting from the interval method are lessened if the source offset distance is limited to no more than 1.5 m. Finally, the slope method is shown to yield reliable velocities that are comparable, and often equal, to the actual hypothetical values, despite being based on simple, straight-line ray paths. Consequently, we recommend using the slope method in lieu of the interval method for more reliable estimates of seismic velocities.

Natural Bridge is a naturally occurring stone archway and historical site located in Rockbridge County, Virginia. It is the only natural stone archway in the country that supports a state road that carries everything from motorcycles to tractor-trailers. The arch is part of a state park, and as a tourist attraction hundreds of pedestrians walk under the arch each week. In 1993 a large slab of rock fell from the bottom of the arch and killed a tourist. Though rock bolting was undertaken to mitigate the risk of future rock falls, no thorough geologic study has ever been conducted. This study was undertaken to supplement a larger study conducted by the Radford University Geohazards Institute as part of a Virginia Department of Transportation initiative to gather geologic information on the arch. This initiative prohibited drilling as a means of obtaining the geologic information. Therefore, multiple non-destructive means were employed in the study that included geophysical, laser, optical, and acoustic methods. The authors were involved in the collection and analysis of the geophysical data on the arch that included resistivity imaging, seismic refraction, multi-channel analysis of surface waves (MASW), and ground penetrating radar (GPR). This paper will present the results of each method and how the various methods were integrated, as well as the advantages and disadvantages of each.