Innovations and Applications in Geotechnical Site Characterization

The Mid-America earthquake region is recognized as containing significant seismic hazards from historically-large events that were centered near New Madrid, MO in 1811 and 1812 and Charleston, SC in 1886. Methods for evaluating ground hazards as a result of soil liquefaction and site amplification are needed in order to properly assess risks and consequences of the next seismic event in these areas. In-situ tests provide quick, economical, and practical means for these purposes. For this research effort, seismic piezocone penetration tests (SCPTu) have been performed at a number of sites in the heart of the Mid-America earthquake regions. Many of these sites have already been associated with liquefaction features such as sand dikes, sand boils, or subsidence, observed during geologic and paleoseismic studies. Data collected at these sites have been analyzed under current methodologies to assess the validity and internal consistency of empirical relations developed for Chinese, Japanese, and Californian interplate earthquakes when applied to historical Mid-American earthquakes. Validation of extrapolation of cyclic resistance curves to high cyclic stress ratio values will be considered. Lower bound estimates of moment magnitude (M_w) from parametric studies on SCPTu data indicate an earthquake event of magnitude greater than 7.0 would have been necessary to induce soil liquefaction at the sites studied.

The NCEER liquefaction workshop document presented the Standard Penetration Test (SPT), and the newer Cone Penetration Test (CPT) methods to undertake liquefaction analyses, and discussed available data that supported the validity of their use. The workshop encouraged verification of CPT soil classification by site specific SPTs. It also stated that the two methods appear to correlate well in terms of predicting the occurrence of liquefaction. However, examination of recent case histories suggest that cyclic resistance ratios (CRR), and in turn factors of safety against liquefaction, may differ depending on the method used; in some cases by as much as ±30%. This paper examines those differences and poses the possibility that in some cases either the CPT or SPT base curves may not be conservative enough, and thus could lead to erroneous conclusions. On the issue of CPT interpretations this paper presents laboratory test results of fines content and Chinese criteria parameters of side-by-side probes for comparison with published CPT correlations.

Dynamic in situ penetration tests and crosshole shear wave velocity measurements were performed in deep alluvial gravel deposits at Seward, Alaska that were shaken and reportedly liquefied by the March 27, 1964 Alaska Earthquake. Two sizes each of split spoon and dynamic cone penetrometers were driven into the gravels near the mouth of the Resurrection River that had exhibited settlement and lateral spreading consequent to earthquake shaking. Two safety hammers of different size were employed, and the energy delivered using various hammer and penetrometer combinations was measured throughout all tests, with the exception of two duplicate tests of the larger sampler and dynamic cone. Limited measurements were also made of hammer velocity, using a radar system developed for that purpose, to allow for kinetic energy determination. Disturbed soil samples were evaluated for classification and to determine stratigraphy. This paper summarizes the procedures used at the Seward site, as well as preliminary correlation of penetration resistance, measured by alternate means and normalized for overburden stress and driving energy, and shear wave velocity in the coarse-grained deposits. Comparisons are also made to penetration resistance measurements made by the Alaska Highway Department immediately following the 1964 earthquake.

In this study, existing and newly developed SPT- and CPT-based simplified methods for liquefaction potential evaluation are calibrated with field performance databases. The calibration is carried out through probabilistic analyses of all cases in the associated database, which results in risk-based methods for liquefaction potential evaluation. The risk-based approach is applied to the SPT- and CPT-based databases from the 1989 Loma Prieta earthquake published recently by the U.S. Geological Survey.

The traditional CPT and CPTU tests are very good tools for stratigraphic soil logging. Using empirical techniques based on tip resistance, local friction and excess pore pressure, a continuous soil profile can be established which is more detailed than that obtained by a conventional SPT test and sampling. However, because tip resistances are controlled by the mechanical soil properties averaged over at least 5–10 cone diameters of elevation, and because the local friction is measured over a 13 cm distance, even the CPT cannot easily discern relatively thin lenses and seams in the stratigraphy. Such lenses may nevertheless control the hydrologic and mechanical behavior of a site. The vision cone penetrometer (VisCPT) collects a continuous stream of magnified images of the soil stratigraphy and thus, every lens and seam is identified. The present paper demonstrates that textural image indices show very high correlation to soil stratigraphy, as confirmed by high quality continuous borehole samples, while the CPT alone often misses or misinterprets the constituents of thin lenses and seams.

Currently, CPT data obtained from the friction sleeve measurement (f_s) is less widely used than the tip measurement (q_c). The "underuse" of the friction sleeve data is related to its high variability and the common sentiment that the sleeve measurement is unreliable. Recent research has quantitatively demonstrated the dominant influence of surface roughness on interface strength. Laboratory experiments have indicated that a change in the surface roughness from the conventional "smooth" configuration can more than double the interface friction. This effect is significant and has yet to be incorporated into geotechnical engineering practice. A research program was undertaken to determine the effect of friction sleeve roughness on the measured sleeve resistance. A series of CPT soundings were performed using a set of roughened friction sleeves in addition to the conventional smooth friction sleeve. Results show the friction measurement to be heavily dependent on the surface roughness of the friction sleeve. The effect of surface roughness is material dependent, but in general the friction sleeve measurement obtained with a textured friction sleeve is about 1.6 to 2.0 times greater than a conventional "smooth" sleeve in granular soils. These results demonstrate the significant effect of surface roughness on f_s and the potential for modifying current in situ practice to account for this effect.

A recently developed in-situ testing method, called "The Sharp Cone Test" consists in pushing a low-angle truncated cone with 2 degrees taper into a smaller diameter prebored pilot hole. As the cone descends, it causes a continuous enlargement of the pilot hole, which, with a proper instrumentation, can be translated into a relationship between radial pressure and radial (or shear) strain, similarly as in a pressuremeter test. A first, preboring version of the sharp cone with three lateral pressure sensors, was tested in the field five years ago, and yielded positive results (Ladanyi et al. 1995). The present paper describes the use in the field of an improved, self-boring version of this instrument with four total-pressure transducers mounted on its lateral surface at different distances from its lower end. The new probe is able to furnish continuously four points of the pressure-expansion curve, which can be translated into a stress-strain relationship, using conventional pressuremeter data processing procedure. During 1998, the new instrument was tested in a thick layer of saturated clay at a site near Montréal. A comparison of the results with those obtained at the same site by some other types of tests, such as self-boring pressuremeter test and static cone test, was encouraging. The new testing method represents in fact a continuous and automated version of the pressuremeter test.

An electric downhole vibrocone has been designed for the site-specific evaluation of soil liquefaction susceptibility in high seismicity regions. The device induces localized cyclic pore pressures while concurrently measuring dynamic tip and friction resistances. Initial versions operated on a pneumatic impulse to generate dynamic forces and showed success in preliminary trials in historic liquefaction sites near Charleston, SC. The latest design couples a dual-element penetrometer with a piezo-actuator shaker and is completely electronic. This vibrocone offers a number of advantages over its predecessors including downhole vertical oscillation, adjustable dynamic displacement, control of excitation frequency from 0 to over 500 Hz, and the simultaneous measurement of porewater pressures at multiple positions including both midface and shoulder locations.

A miniature piezocone was developed for use in centrifuge environments, including the piezocone design and fabrication. Several penetration tests were performed in centrifuge soil models encompassing normally consolidated and moderately overconsolidated clays, and medium dense to dense sands. For preparation of clay specimens, including preconsolidation, and for performance of piezocone penetration tests (CPTU) in the centrifuge, an experimental apparatus was designed and built. This apparatus allows the performance of tests under different conditions, such as rate of penetration and in-flight test site positioning. This research has shown that piezocone testing in centrifuge models is feasible, producing consistent and reliable data.

A powerful method to quantitatively control and direct exploration that combines the uncertainty (or variance) of three dimensional (3D) site characteristics with sensitivity of the performance model through a Taylor series expansion is demonstrated. Fundamentally, the approach is made possible by directly differentiating the performance model to obtain its 3D sensitivity to changes in subsurface properties. A one-time only direct differentiation of a generic code for performance analysis allows sensitivity to be determined with a single model run for any subsurface geometry. Thus direct differentiation avoids 1000's of model runs necessary with parameter perturbation or Monte Carlo simulation. The integrated exploration approach will be demonstrated with three examples, each with very different performance objectives (settlement, groundwater flow, and contaminant transport). The examples will demonstrate the process of producing calculated performance variance from 3D model sensitivity and uncertainty associated with site layer geometry and properties. Exploration is directed to locations of maximum variance in project performance. Sampling at these points produces the maximum reduction in performance uncertainty. As directed exploration proceeds, sufficiency of characterization effort is quantitatively determined through a reliability index. This index combines calculated performance, variance in calculated performance, and required performance.

This paper presents the database of in situ test results (SPT, CPTU, PMT, DMT and Field Vane) from Recife soft clays. The geotechnical database was developed using tools of the information technology and presents all in situ tests results from two research sites. The database has shown to be a very good tool in research and projects, and it has also been used to help the geotechnical education of students in graduate and undergraduate geotechnical engineering courses.

A geostatistical sequential sample location selection method is developed for Expedited Site Characterization or accelerated site investigations. The method provides a measure of the uncertainty regarding contaminant concentrations at unsampled locations on the basis of the existing contaminant concentration data. Sample locations are selected within regions where the probability that contamination exceeds a specified threshold indicates a high level of uncertainty. Samples are spaced according to the range of influence determined from a variogram developed from measured contaminant concentrations. Sample collection ceases when the contaminant plume is bounded by a probability contour indicating low probability that a contaminant concentration will exceed the threshold at unsampled locations outside the bounded region. The method is demonstrated with trichloroethylene data collected during an Expedited Site Characterization project at the U.S. Department of Energy Savannah River Site in 1995.

A case study is presented where preloading was applied to reduce tank settlements at a liquid natural gas (LNG) facility in Puerto Rico. Stone columns were installed to further reduce settlements and provide an adequate factor of safety against soil liquefaction in the case of a seismic event. Seismic piezocone penetration tests (SCPTu) were used to quantitatively assess the effects of soil improvement. Soil properties estimated from SCPTu data taken before and after both improvement methods are compared to assess the degree of soil improvement. Preconsolidation stresses were estimated using methods based solely on net cone tip resistance, pore pressure difference, shear wave velocity, and a multiple regression based on cone tip resistance and shear wave velocity. Preloading provides additional vertical stress, which is directly related to an increase in preconsolidation stress if time rate consolidation properties are considered. Stone columns increase the lateral stress locally within the zone of influence of the column. Soils initially appear to be normally consolidated to slightly overconsolidated, and a significant increase in preconsolidation stress was noticed in clay layers after preloading. The high clay content of the soils likely led to a reduced zone of influence for the stone columns, and little improvement was noticed immediately after installation. Crosshole test (CHT) shear wave velocities taken across stone columns showed little to no increase in stiffness when compared to downhole shear wave velocities of the soil.

This paper investigates the use of in situ test data to characterize the changes in stress and density induced by ground improvement. Testing comprised seismic cone penetration tests, full displacement pressuremeter tests and resistivity cone penetration tests. After ground treatment, changes were observed in tip resistance, pore pressure response, shear wave velocity, the characteristics of pressuremeter curves and bulk resistivity. Some of these changes can be caused by changes in lateral stress as well as by density increases. The results generally indicate that the use of a combination of in situ tests will improve our understanding of the changes in soil behaviour achieved by ground treatment. The improved understanding of soil behaviour obtained from combinations of tests may allow ground improvement specifications to be written in terms of a desired post-treatment stress-deformation response. However, further research is required to develop reliable methods of characterization of stress-deformation behaviour.

Static and dynamic soil properties were monitored around a closed-ended steel pipe pile driven in a deep deposit of San Francisco Young Bay Mud. Shear wave velocity measurements were taken prior to pile driving to establish the baseline condition and at selected times after pile driving to document the changes as a function of time. Additional instrumentation and monitoring activities included pore water pressure response and lateral soil deformations as a result of pile driving. A series of conventional laboratory tests were performed on samples collected prior to pile driving and 8 months afterwards. Field measurements after pile driving indicate a significant increase of shear wave velocity as a function of time accompanied by excess pore pressure dissipation. Laboratory tests show a significant increase in the strain to failure with slight increases in undrained strength.

This paper reports on the ground investigation carried out for each test pile for a series of nine 30MN full-scale load tests of piles founded on rock. The work was carried out to provide data for the design of over 20km of the Kowloon-Canton Railway Corporation West Rail Project Phase I by foundation contractors. The ground investigation techniques used included Goodman Jack, High Pressure Dilatometer, Self-boring Pressuremeter and Geophysics as well detailed logging of the rock cores and laboratory testing for presentation in Rock Mass Rating (RMR) format. The advantages and limitations of each of the testing techniques are discussed. The paper concludes that the use of published correlations between stiffness and rock mass rating parameters such as RMR and GSI would result in a fivefold overestimate of the stiffness. It is recommended that the stiffness of the rock may be determined using conventional RMR data with the water and joint orientation ratings set to zero. A new coefficient Modified Rock Mass Rating (RM²) is proposed for deep foundations on weathered rock.

All karst is not created equal. Variations in the age of the formation (e.g., Florida "limerock" versus the limestones of Wisconsin) result in strength variations and hence, performance. Variations in tectonic environment can yield flat carbonates (e.g., Kentucky's Mammoth Cave) versus the folded and faulted Appalachian carbonates (e.g., Virginia's Carlsbad Caverns) resulting in unusual patterns of solutioning as well as different failure mechanisms. Thus, geotechnical investigation and analyses must understand how the depositional and tectonic constraints of these different "rocks" can result in varying problems to investigators, analysts, designers and constructors. This paper addresses the differences in possible structural failure or contamination modes that develop in these different rocks in relation to man's works. The study procedures presented are multi-phased; particularly if the geotechnical work is first used as input for planning purposes. A review of aerial photography, satellite imagery and available geologic information, as well as a reconnaissance to "ground-truth" the aerial photography and regional maps, are all generally part of the first stage of an investigation in karst areas. Subsequent phases consist of developing a geologic model of the site, then testing and refining it through field investigations and construction inspection. Direct investigation tools include test pits and borings using specific techniques, with perhaps the use of certain indirect (geophysical) tools to cautiously correlate between known hard data points. The procedures are not unique, but represent an attempt to present economical investigative concepts that recognize the nature of different karst environments.