From Research to Practice in Geotechnical Engineering

In perhaps one of his less famous papers, Schmertmann (1969) introduced the concept of Swell Sensitivity as the ratio of the swelling behavior of a remoulded soil to the swelling behavior of the same soil in the undisturbed state. In effect, this Technical Note, which appeared in Geotechnique, essentially established the importance of inherent soil structure and natural bonding on the swelling potential of fine-grained soils. Observations made by Schmertmann and presented in the Note showed that some remoulded soils which have lost their natural structural bonds through simple mechanical remoulding have a higher swelling potential than the undisturbed soil. In the author's opinion, this paper is an important but lesser appreciated paper, in part because of the success of Schmertmann's other more popular papers. Swell Sensitivity has great practical importance to practicing geotechnical engineers, one of the main themes of all of Schmertmann's work. In the current paper, the author reexamines Swell Sensitivity as defined and described by Schmertmann (1969) and illustrates the behavior for a number of expansive soils from around the U.S.

The paper presents the results of extensive centrifuge tests modeling rigid circular piled rafts laying on a bed of loose very fine silica sand. The tests were aimed at investigating the behavior of rafts on settlement reducing piles. The testing program included: an unpiled raft, rafts on 1, 3, 7 and 13 piles. In each test, some model piles were instrumented with load cells to determine the distribution of load along the shaft. Beneath the rafts, two types of model piles, close-ended and free headed were installed: quasi displacement (QD) and quasi-non displacement (QND) piles. The obtained results permitted figuring out the role of piles in terms of their effectiveness as settlement reducers and to quantify the load sharing mechanism between piles and the raft-soil contact. The tests were aimed at investigating the load transfer mechanisms adopted in the design approach, and in particular at validating a numerical code which can be used in engineering practice. The paper describes the details of experiments undertaken, the adopted procedures and some of the results; where not specified, all the experimental data given in this paper referred to model scale. The adopted numerical code is briefly described and its validation, via test results on quasi-non displacement piles, is also reported.

The analysis of the 1959 landslide at Furre, Norway, had given an effective friction angle φ' of 7°, whereas large-scale in-situ shear box tests were interpreted to yield an angle of 9°. Triaxial tests had given values of φ' in the range 21–25°. It was concluded at the 1967 Oslo conference that the slide mechanism at work was that of retrogressive sliding, i.e. the slide went out piece by piece rather than moving out as a single flake. In the present paper it is postulated that the extremely low friction angles are not the real friction angles. The various failures, whether in the field or in the laboratory, whether drained or undrained and whether in the direction of initial shear or in the opposite direction, all occur when the real friction is fully mobilized along some plane in the failing clay. At this stage, if it is not physically possible for failure to take place along such a plane, i.e. if conventional sliding failure is hindered by boundary restraints, the clay will display a discontinuity appearing as a relatively sharp reduction of the gradient of the stress-strain curve.

Although normally associated with ground-shaking events such as earthquakes, liquefaction also is indicated to occur during dynamic installation of Rammed Aggregate Pier^TM foundation elements in susceptible soils. Such piers nevertheless give a satisfactory performance, suggesting likelihood that drainage is contemporaneous with ramming. This was investigated by measuring lateral in-situ stresses at various depths and distances from the piers that are made by ramming graded aggregate in layers into open borings. Liquefaction is indicated by an undiminished transfer of radial stress for a distance outward into the soil. Total stress in the liquefied zone then is reduced by drainage that is accelerated by the hydraulic gradient caused by ramming. Simultaneous seepage forces act to compress and compact the soil; hence the coined term "liquepaction," Radial and tangential in situ stress measurements indicate that radial tension cracks occur in a peripheral elastic zone regardless of whether the soil is saturated or liquefied. In saturated soil the cracks act as a drainage gallery that may be extended by hydraulic fracturing. Stress measurements, which were made with the K_o Stepped Blade, also indicate that the radial cracks are propped open by soil invading from the liquefied zone.

This paper discusses the development of an up-to-date computerized CPT (Cone Penetration Test) based soil engineering classification system to provide geotechnical engineers with a handy tool for their daily design activities. Five CPT soil engineering classification systems are incorporated in this effort. They include the probabilistic region estimation and fuzzy classification methods, both developed by Zhang and Tumay, the Schmertmann, the Douglas and Olsen, and the Robertson et al. methods. In the probabilistic region estimation method, a conformal transformation is used to determine the soil classification index, U, from CPT cone tip resistance and friction ratio. A statistical correlation is established between U and the compositional soil type given by the Unified Soil Classification System (USCS). The soil classification index, U, provides a soil profile over depth with the probability of belonging to different soil types, which more realistically and continuously reflects the in-situ soil characterization, which includes the spatial variation of soil types. The CPT fuzzy classification on the other hand emphasizes the certainty of soil behavior. The advantage of combining these two classification methods is realized through implementing them into visual basic software with three other CPT soil classification methods for friendly use by geotechnical engineers. Three sites in Louisiana were selected for this study. For each site, CPT tests and the corresponding soil boring results were correlated. The soil classification results obtained using the probabilistic region estimation and fuzzy classification methods are cross-correlated with conventional soil classification from borings logs and three other established CPT soil classification methods.

This study addresses the small-scale testing of Mechanically Stabilized Earth (MSE) walls reinforced with geogrids. Two High Density Polyethylene (HDPE) and two Polyethylene Terephthalate (PET) geogrid reinforced small-scale (1:5.5) MSE walls were tested. Additionally, a prototype full-scale MSE wall was tested, with one half of the wall reinforced with HDPE geogrids, and the other half with PET geogrids. Panel movements, strains in the reinforcement, and settlements were measured. The analytical work consisted of finite difference modeling of the small-scale MSE walls with Fast Lagrangian Analysis of Continua (FLAC) software. The results of both the HDPE and PET walls testing showed that the deformations and strains encountered in the geogrids were very small, even at loads above normal working conditions. It was also observed that there were large differences in the values between the small-scale testing, and both the FLAC modeling and the full-scale MSE wall prototype testing. This is a consequence of the material property discrepancies and geogrid distortion for the FLAC modeling, and the lack of gravity simulation of full scale wall behavior. Therefore, although small scale testing can be useful for obtaining preliminary results, computer modeling and/or full prototype testing are still necessary to accurately determine the behavior of MSE walls.

The seismic dilatometer (SDMT) is the combination of the standard flat dilatometer (DMT) with a seismic module for measuring the shear wave velocity V_S. This paper summarizes the experience gained from SDMT tests performed at 34 sites. In particular' the paper presents an overview of the SDMT equipment, comparisons of V_S measured by SDMT and by other methods, interrelationships G₀, E_D, M_DMT based on 2000 data points and a selection of significant SDMT results and related comments. The paper also illustrates the major issues of present research on use and applications of the SDMT, mostly focused on the development of methods for deriving the in situ G-γ decay curves and for evaluating the liquefaction resistance of sands based on SDMT results.

Construction of deep foundations causes changes to the original geotechnical condition of the surrounding soils and rock. The nature, extent, and effects of these alterations, and subsequent time-dependent modifications, on the long-term load bearing capacity of the foundation elements are mainly functions of the characteristics of the supporting geo-materials and type of installation method. Post-installation Dynamic Load Testing (DLT) is commonly used for capacity evaluation of driven and drilled deep foundation elements and is performed under impacts of the pile driving hammer or a special large drop weight loading device. Use of a large drop weight loading device is usually appropriate for cast-in-place shafts and driven piles in high-setup soils to activate the full bearing capacity. The DLT procedure consists of applying a specific number of blows (typically 2 to 10 blows) and measuring pile/shaft strain and motion time-records, and the resulting displacement. This paper presents discussions on the analysis of post-installation DLT data for capacity evaluation of driven and drilled deep foundations. Technical aspects related to the proper test record selection for analysis considering pile/shaft set per blow, changes of capacity from blow to blow with subsequent impacts, energy levels, and total penetration during the test are delineated. Analytical procedure and engineering evaluation of results are discussed with recommendations for practical use. Test records from actual case histories are utilized to demonstrate data characteristics and illustrate proposed numerical methods and recommended analytical procedures for rational evaluation of DLT results for engineering use.

The author explored the importance of momentum in pile driving in a paper published in GSP No. 125 (see Crapps, 2004). Parts I through III of the GSP No. 125 paper presented the pertinent theory and discussed the role of momentum in pile force, penetration and pile specifications. The entire hammer-pile-soil system must be considered for successful pile installation. This paper presents relationships to estimate the pile force and ram momentum required for pile penetration for specification and hammer selection purposes. These relationships and principles presented herein may be used to assist with pile design, to evaluate the effects of site variability, to assist with hammer selection, to evaluate the effects of hammer performance variability and to reduce and/or settle pile claims.

A design of the foundation for a wind turbine is presented employing a single pile and a companion design employs a pile group. Modern principles, using concepts of soil-structure interaction, are recommended in the designs. An important criterion in design is to limit the rotation at the foundation to within a tolerance of 0.001 radians in order to restrict the movement at the turbine, which is typically 50 m to 80 m above the ground. The method of design, based on the use of p-y curves and t-z curves for taking soil-structure interaction into account, is rational and practical for wind-turbine foundations.

This paper reports on the behavior of large diameter cylinder piles under both driving and static conditions. Of special interest is the understanding of the tip resistance and the associated soil column within the pile under static and dynamic conditions. An analysis of driving forces, shows that inertia on the soil plug rapidly exceeds the static skin friction between the pile and soil on the inside of the cylinder. The analysis of the soil column for static conditions was undertaken with an Eulerian finite element viscous analysis, ADINA-F. The static analysis revealed that even when the cylinder "cookie cut" under driving it behaved "plugged" under static conditions for a multitude of soil strengths and wall thicknesses. Next, based on a database of 35 load tests (22 concrete & 13 steel) unit side shear and end bearing for cylinder piles as a function of pile material (i.e., steel and concrete) and insitu SPT N were developed and compared successfully to Schmertmann's original values for small diameter piles. Finally, LRFD resistance factors, φ, for the proposed cylinder pile design were determined.

A case history of repeated dynamic and static loading tests in Alberta on two pipe piles during dissipation of driving-induced pore pressures is presented together with three reanalyzed published case histories involving similar records. The four case histories demonstrate that, for each case, the same effective-stress proportionality coefficients, beta-coefficients, fit the capacities at different degrees of dissipation of excess pore pressures. For two of the test sites, the beta-coefficients back-calculated from the tests differed considerable from the values determined from the soil plasticity relation, while for two, the agreement is good. For one case, the back-calculated shaft resistance agreed well with the values of vane shear strength, while a less good agreement was found for the other tests. Neither case showed good agreement was found for methods combining undrained shear strength and effective overburden stress. Capacity calculations for two cases employing methods based on CPT soundings gave excellent agreement with one test and a poor agreement with the other. The increase of capacity due to aging after dissipation of excess pore pressures did not agree with cited recommendation for calculations of aging effect.

This paper addresses the problem of wave forces on bridge decks created by water flowing against and over the decks such as in the case of surges due to hurricanes. Such waves develop horizontal and vertical loads on the bridge decks. These decks in turn may fail by horizontal sliding since they are not tied to the piers. One solution is to tie the decks down to the piers but then the foundation has to resist the additional load. This paper takes a first look at the magnitude of such forces and at the impact they are likely to create on the foundation of the piers.

The I-10 Blackwater Bridge (parallel bridges) experienced settlements since its completion in 1968 of sufficient magnitude to create a rough ride. Florida Department of Transportation maintenance crews releveled the bridge a number of times by jacking and shimming the superstructure. In 1995 the FDOT noted additional settlements after Hurricanes Erin and Opal. Total settlements had reached up to about 100 mm (4 inches) at that time. The FDOT requested and was granted emergency funds from the Federal Highway Administration to replace the Blackwater Bridge. Vibrations, especially from heavy truck traffic, were suspected as the likely cause of the settlements. Driving piles for the replacement bridge would create additional vibrations of a magnitude that would likely induce settlement. This paper describes the study made at the Blackwater Bridge site using test piles at two locations to document the magnitude of settlement induced in the existing bridge by shear waves from pile driving. Crosshole shear wave velocities were made at each test site location and down-hole measurements were made of the soil vibrations induced by pile driving. No settlement was detected in the nearby piers. This paper considers possible explanations of the no-settlement including "soil aging", a favorite topic of Dr. Schmertmann (see Schmertmann (1991)). This paper also discusses the complication of detecting shear wave signatures from pile driving. Our field exercise provided confirmation that the vertical angle between the source and the receiver has a major impact on the interpretation of shear wave arrivals and amplitudes.

Mitigation of liquefaction potential of soils with fines contents greater than 15 to 20 percent usually requires special adaptation of the dynamic densification methods — Deep Dynamic Compaction, Explosive Compaction, Vibro-Compaction, Compaction Piles — that are ordinarily used to improve loose, clean cohesionless soils. However, these methods can be used to attain the needed improvement in many soils if closely spaced vertical sand, gravel, or prefabricated (wick) drains are installed prior to construction. Soil-cement columns or walls installed by deep soil mixing or jet grouting can be used to provide both reinforcement and containment of liquefied soil. Removal and replacement of liquefiable silty and clayey soils using compaction, admixture stabilization or substitution by another 'material; e.g., roller compacted concrete, is also possible; however, the cost may be high, and consideration must be given to dewatering needs and the stability of open excavations. Mitigation of the liquefaction potential of high fines content sandy and silty soils using permeation grouting is not feasible because the low hydraulic conductivity of such soils.

This paper aims at providing engineers with understandable load (stress) movement curves whenever load tests are available. The idea is to allow correct interpretations of test results. When they are not available, the idea is to predict what would have been their results on basis of conventional field tests, the Standard Penetration Test, SPT, the Cone Penetration Test, CPT or the Menard Pressuremeter, PMT.

Field inspectors often observe a dramatic increase in penetration resistance between initial driving and restrikes of driven piles, a result of additional static "setup" capacity. Designers often ignore setup when specifying pile lengths and blow count acceptance criteria, and pile capacity design methods, calibrated against static load tests, already incorporate some setup that occurs during the test preparation delay. Research indicates that setup results mostly from an increase in side shear and occurs in all soil types as a result of combined pore pressure dissipation, consolidation, and mechanical aging effects. Repeated static tests provide a reliable measurement of setup, but also delay construction. Penetration resistance (set per blow) does not reliably measure setup during isolated restrikes as it varies per blow due to inconsistent driving system performance and changing pile capacity. Instrumented dynamic tests, using strain gages and accelerometers, provide an "easy button" to verify hammer energy and pile capacity during all phases of installation. This paper reviews pile setup, its causes, and related properties. It presents a simple method, with examples, to measure and analyze setup from dynamic test results. An engineer may use these results to reduce the length, size, or number of production piles, and adjust the required set per blow. Caveats include staged testing effects and consideration of the setup contribution from different soil strata.

Downdrag is pile settlement in response to settlement of the surrounding soil. The bearing capacity approach to downdrag leads to erroneous design calculations and the implementation of unnecessary construction mitigation techniques. Pile settlement calculations are required to properly evaluate downdrag, but geotechnical engineers have been reluctant to use settlement calculations in pile design. When viewed from a settlement perspective, it becomes clear that downdrag is a very common phenomenon that all pile foundations experience to some degree. In fact, downdrag can limit structure settlement by pre-compressing the pile and the soil at the pile tip prior to application of the structure load. Spreadsheet calculations using Florida Department of Transportation (FDOT) FB-Deep and FB-MultiPier computations for skin friction, end bearing and pile tip movement are presented to illustrate the effect of soil settlement on pile settlement, structural integrity and bearing capacity in numerical and graphical formats.

In 2003, four grain silos, each 104 ft (32 m) in. diameter and 100 ft (31 m) tall, were constructed at a waterfront storage facility in Mobile, Alabama, to replace a cold storage facility that had been demolished. Foundation plans for the replacement silos included auger-cast piles to support the perimeter steel shell walls on a concrete ring, auger-cast piles to support the buried conveyor tunnel used to unload grain and a soil-supported interior floating floor slab (isolated from the perimeter ring and conveyor tunnel) to support grain loads on the order of 5,000 psf (240 kN/m²). During the initial loading of Silo #1, as filling neared capacity, the structure suddenly shifted and began to lean toward the south. Large cracks appeared in the concrete foundation system and structural steel beams located at the top of the silo twisted. Because grain storage silos were in short supply in the aftermath of hurricane Katrina, the owners of the storage facility hired the authors to conduct a unique, full scale test loading program, by placing Silos #2, #3 and #4 into production. Using available historical data, along with supplemental data gathered specifically for settlement monitoring, the authors developed a full scale test loading protocol designed to help the owner temporarily store grain, while simultaneously evaluating foundation performance. This paper discusses geotechnical and structural assumptions used in design of the silo foundations. Then, using data collected from full scale loading of the silo foundations to impending failure, the authors describe lessons learned by comparing expectations of foundation performance to measured performance.

Seepage analysis serves as one of the most significant stages in the design process of an embankment dam. In two-dimensional (2D) seepage analysis of embankment dams, little or no attention is paid to the seepage through side abutments. Moreover the role of grout curtain extensions into the side abutments and abutment material properties are inevitably neglected when performing a 2D seepage analysis. In this paper, two and three-dimensional. (3D) models of a rockfill dam during operation state are generated and several unsteady and steady state seepage analyses are performed using finite element method (FEM). The results obtained from 2D and 3D seepage analyses were compared with measurements from the instrumentation system installed in the dam body and foundation during construction. It is concluded that for embankment dams to be constructed in narrow valleys, 2D seepage analysis results are far from reality and considering 3D modeling of such dam sites is vital. Several design graphs were developed to calculate 3D discharge rates and hydraulic gradients from those obtained from 2D seepage analysis.

The authors describe two aspects of geotechnical engineering; site characterization utilizing the CPT and recognition of aging as a factor affecting soil properties. These were pioneered by Professor Schmertmann and are practiced by the Bechtel Corporation in general and at the Savannah River Site in South Carolina. This paper describes a general subsurface exploration approach that we have developed. It consists of "phasing" the investigation, employing the principles of the observational method suggested by Peck (1969) and others. We have found that the recommendations proposed by Sowers in terms of borehole spacing and exploration cost are reasonable for developing an investigation program, recognizing that through continuous review the final investigation program will evolve. At the SRS a field investigation and laboratory testing program was devised to measure and account for aging as it relates to the cyclic resistance of the site soils. Recently, a panel of experts (Youd et al., 2001) has made recommendations regarding the liquefaction assessment of soils. This paper will address some of those recommendations in the context of re-assessing the liquefaction resistance of the soils at the SRS. It will be shown that, indeed, aging should be accounted for in liquefaction potential assessments for soils older than Holocene age.

The Florida Phosphate Industry has been constructing earthen reservoirs for water storage and clay settling/disposal for more than 70 years. During the first 35 years of operation, the perimeter dams for the clay settling areas were simply overburden spoil (primarily silty sand) cast up in windrows and shaped to provide a crest road. The spoil windrow was excavated down to natural ground at one of two locations to install an outlet pipe for the decant structure(s). Little or no engineering was performed. Between 1936 and 1969, there were 32 reservoir failures. The first rule regulating the design and construction of earthen dams used in phosphate mining and processing operations was adopted by the state of Florida in 1969 after a reservoir failure near Brandon, Florida turned the Alafia River a brownish-white 3.2 kilometers into the Gulf of Mexico. The rule was subsequently modified in 1972 after a reservoir failure occurred near Fort Meade and again in 1999 after a reservoir failure occurred near Hopewell. All of the reservoirs constructed after 1969 were engineered. The authors' firm has investigated, designed, monitored construction, and inspected more than 120 earthen reservoirs constructed in Florida after 1967. The paper describes the advances that have been made during the past 40 years in the investigation, analysis, design, construction and inspection of the embankment dams surrounding these reservoirs. The lessons learned in the phosphate industry have and should be applied to other earthen reservoirs constructed in Florida and elsewhere.

Hurricane Katrina resulted in the single most catastrophic failure of a civil engineered system in the history of the United States — failure of the flood defense system for the greater New Orleans area. This paper summarizes results from several forensic studies that have examined the causes for failure of one of the most important components of the flood protection system — failure of the levee and floodwall on the 17^th street canal. This failure has been publicly cited as an `engineering failure' (Walsh 2006) that involved `unforeseen and unforeseeable' (Marshall 2006) conditions. This paper illustrates why the engineering failure was firmly rooted in a failure to translate research to practice. Geotechnical engineering aspects of the levee and floodwall failure are developed including description of the soil and geologic conditions, analyses of the loading conditions, and analyses of the soil-structure-loading performance characteristics. In addition, the human and organizational aspects that played key roles in development of this failure are detailed. It is concluded that this was a predictable failure whose causes were embedded in a dysfunctional Technology Delivery System.

The paper describes the application of geotechnical site characterization, material testing, test fill construction and settlement analysis to the closure design for two settling ponds at a former chemical plant in southwestern Virginia, USA. The ponds contained by-product material from ammonia soda ash manufacturing that had been slurry-deposited to thickness up to 20m over a total area of approximately 480,000m². The material is a man-made particulate exhibiting a unique structure with high moisture content, high compressibility, and unusual geotechnical properties. Closure construction was successfully completed in 2001. The paper demonstrates that although the by-product material exhibits unusual characteristics, the application of geotechnical engineering principles allowed the successful execution of pond closure.

In this case study of a cut for a motorway in the Alpine region of Austria it could be demonstrated that an inadequate pre-construction site investigation can lead to unexpected problems connected with a slope instability. In this instance a dormant fossil slide, discovered by a later geotechnical survey, was in part reactivated by the slope cutting process. A geological model of the slide was made which gave sufficient insight into the event to establish a multi-stage remedial-work scheme. The motorway axis was shifted, a toe load placed and bored piles were constructed within the slope, and in addition, an efficient drainage system was provided. Inclinometers were installed to monitor the deformations of the slope and piles. In case the structural deformation of the piles would have exceeded a set limit, further piles were foreseen as a second stage. The remedial measures taken proved to be successful as demonstrated by plotting the frequent deformation measurements on a time scale.

Traditionally, separate geotechnical and structural engineering firms contribute to design and construction of building projects. This paper discusses the unique opportunity that Gibble Norden Champion Brown Consulting Engineers, Inc., (GNCB) a structural engineering firm in Connecticut, took when it combined expertise from both services into one firm. This innovative, in-house collaborative practice adds value to its menu of client services and demonstrates the benefits of incorporating professional and experienced geotechnical engineering into a project, from concept to completion. The geotechnical engineering staff enhanced its presence on the design team with direct participation in early design and construction decisions, where knowledge of subsurface soil conditions are used to benefit the important early stages of project planning. The benefits, illustrated by three case studies, have been far reaching and include enhanced contract drawings, expansion into new services, reduced company risk, greater structural engineer's awareness of geotechnical services, and increased recognition of geotechnical engineering as a significant factor in project success.

The natural desire to improve the way we design and build deep foundations provides a fundamental motive for achieving excellence and high quality. Unfortunately, perceived high cost and the lack of rewards or incentives to make the effort that quality work demands, often leads us to complacency and mediocrity. This complacency can result in substantial reductions in bored pile capacity that dwarf the improvements derived from good quality engineering design. Examples from case histories illustrate how seemingly minor factors can cause capacity reductions of 80% or more. Faced with the daunting risks and uncertainties associated with deep foundations many engineers fall back on code-value engineering and from there to unnecessarily expensive foundations. A review of the impact of shaft diameter on measured side shear resistances leads to cautionary advice about how relying too much on intuition can affect the quality of our analytical work. The engineer and the contractor do have several options for improving and maintaining quality including but not limited to — better assessment of the high cost of risk and uncertainty, changing the low-bid approach to awarding foundation contracts, including a special bid item to encompass all of the QA/QC work on a project and generally making a greater effort to focus on the value that high quality brings to any deep foundation project.

Making research and practice as simple as you can has many advantages, and the writer uses this approach. This paper gives ten personal examples. They vary from demonstrations of engineering behavior, direct insitu test methods, making field measurements, the installation and testing of piles, to a method to improve technical writing.