Proceeding of the NCEER workshop on evaluation of liquefaction resistance of soils

Over the past twenty-five years, a procedure, termed the "simplified procedure" has evolved for evaluating liquefaction resistance of soils. This procedure has become the standard of practice in North American and trougthout much of the world. The purpose of the 1996 workshop, sponsored by the National Center for Earthquake Engineering Research (NCEER) was to convene a group of experts to review developments and gain consensus for furher augmentations to the procedure. To keep the workshop focused and the content tractable, the scope was limited to evaluation of liquifaction resistance. Post-liquefaction phenomena, such as soil deformation and ground failure, although equally or more important, were beyond the scope of this workshop. The participants developed consensus recommendations on the following topics: (1) use of the standard and cone penetration tests for evaluation of liquefaction resistance, (2) use of shear ware velocity measurements for evaluation of liquefaction resistance, (3) use of the Becker penetration test for gravelly soils, (4) magnitude scaling factors, (5) correction factors K and K, and (6) evaluation of seismic factors required for the evaluation procedure. Probabilistic analysis and seismic energy considerations were also reviewed. Seismic energy conceptss were judged to be insufficiently developed to make recommendations for engineering practice. Probabilistic methods have been used in some risk analyses, but are still outside the mainstream of standard practice. (AU)

Magnitude scaling factors

In developing the ùSimplified procedure for evalualing liquefaction resistance, Seed and Idriss (1982) compiled a sizable database from sites where liquefaction did or did not occur during earthquake with magnitudes near 7.5. From this database, these investigators defined a conservative deterministic bound separating data indicative of liquefaction from data indicative of nonliquefaction. The workshop gained consensus that a range of MSF values should be suggested for engineering practice. Practitioners could then select MSF from within this range, depending on the degree of risk they or their clients are willing to accept for various applicactions. The lower bound for that recommended range is the MSF proposed by Idriss (Column 3, Table 1) and the upper bound for the range is the factors proposed by Andrus and Stokoe (Column 7, Table 1). for magnitudes greater than 7.5 the revised factors of Idriss should be used. These factors are smaller than the original Seed and Idriss (1982) factors, and hance lead to an increase of calculated liquefaction hazard compared to the old factors. The workshop participants agreed, however, that the original factors by Seed and Idriss (1982) may not have been adequately conservative for large magnitude earthquakes. (AU)

Seismic factors for use in evaluating liquefaction resistance

Application of the simplified produrre for evaluation of liquefaction resistance requires estimates of earthquake magnitude and peak acceleration sa input seismic parameters. These factors characterize duration and intensity, respectively, of earthquake shaking at a site. The following comments summarize consensus statements developed at the workshop for determining these seismic factors.-The simple correlation between magnitude and duration incorporated within is adequate and conservative for typical liquefaction hazard analyses. THis relationship is adequate for application in the eastern U.S. as well as the western U.S.- Magnitudes based on the moment magnitudes scale, M, are the preferred for calculation of liquefaction resistance. Where estimates of M are not available, magnitudes from other scales may be substituted within the following limits: M<6 m<7.5, and 6

Liquefaction criteria based on statistical and probabilistic analyses

Probabilistic procedures for evaluating liquefaction resistance have the advantage of allowing an acceptable level of risk to be specified by the user. Liao and his colleagues used a logistic procedure to develop probabilistic CRR curves. The original Seed and Idriss magnitude scaling factors, however, we ussed to correct for magnitude. Youd and Noble (herein) use the logistic procedure to analyze liquefaction resistance with a magnitude added as an independent variable. New case history data and (N) were added to enlarge the case history data set. Primary conclusions from the study are: The probabilistic procedure allows direct incorporations of an appropriate probability, or risk factor in liquefaction hazard analyses.- The analyses by Liao and his collegegues indicate, for clean sands, that the standard criteria from the simplified procedure provides a probability of occurrence of about 20

for corrected blow counts (N) between 11 and 28. Below an (N) of 11 the original simplifiesd base curve is characterized by a probability of liquefaction smaller than 20

. Above an)N) of 28, the curves of Liao et al. indicate a probability of liquefaction greater than 20

. The curves in the upper par of the range, however, aare near the limit of liquefaction occurrences and are not well constrained by empirical data. - The analyses by Youd and Noble include magnitude as an independent variable eliminating the need for magnitude scaling factors in the analysis. The Youd and Nobel results are more conservative than those of Liao et al. for (N) lees than 20 and characteriza the simplified base curve bby probabilities ranging from 20

to 50

. (AU)

Liquefaction criteria based on energy content of seismograms

Because energy is a more fundamental property or measure of earthquake excitation than peak acceleration and magnitude, several investigators have suggested that the amount of energy generated at specific points in a soil layer un response to earthquake shaking may correlate better with the development of liquefaction than with cyclic stress ratio. Kayen (1993) reviewed the literature on this topic and investigated yhe use of accelerogram energy, expressed in terms of Arias intensity as a parameter for use in evaluating liquefaction resistance of soils. What procedures should be used, and how accurately can Arias intensity be estimated for application at field sites where strong motion records are not avaible? has sufficient case history data and experience been compiled and analyzed to provide adequate verification of the procedure? The workshop encourages continued research to answer these questions and further develop energy procedures for use in engineering practice. (AU)

Distribution of ground displacement sand strains induced by lateral spread during the 1964 Niigata earthquake

The 1964 Niigata, Japan earthquake caused widespread liquefaction and lateral spread in the Niigata, Japan urban area. Hamada et al. mapped vectors of lateral ground displacement over much of that area, providing an exceptional data set. This data set was used to determine areal extents of lateral spreads, distributions of ground displacements and magnitudes of ground deformations or strains. This study provides guidance to designers and risk managers of pipelines and other lifeline facilities as to sizes, magnitudes and distributions of lateral spread displacements.(AU)

Liquefaction-induced ground surface disruption

Although liquefaction is a major cause of earthquake damage, little harm occurs unless liquefaction generates some form of ground surface disruption or ground failure. Thus, ability to accurately predict potential for ground surface disruption is a major concern for geotechnologists charged with safe siting of constructed works. Ishihara presented preliminary empirical criteria for assessing the potential for ground surface disruption at liquefaction sites. Those criteria are based on relationships between the thickness of liquefiable sediment beneath a site and the corresponding thickness of overlying non-liquefiable soil. The purpose of this study is to further evaluate and verify these criteria by testing them against thicknesses calculated from a wide range of earthquake and site conditions. The newly developed data lead to the following conclusions: (1)For sites not susceptible to ground oscillation or lateral spread, the thickness bounds proposed by Ishihara appear to be valid. (2) For sites susceptible to ground oscillation or lateral spread, the bounds suggested by Ishihara are not sufficient for predicting ground surface disruption.(AU)

Magnitude scaling factors for analysis of liquefaction

The "simplified procedure" developed by Seed and Idriss is widely used in the United States to evaluate liquefaction hazard. Empirical evidence suggests that magnitude scaling factors (MSF) required in this procedure are very conservative for moderate-sized earthquakes. We compiled soil and site data for several earthquakes and locaties where surface effects of liquefaction did or did not occur. We statistically analyzed these data using logistic regression to develop MSF that have about the same conservatism as other factors in the simplified procedure. Regressed MSF values for magnitude 5.5,6, and 6.75 earthquakes are 4.5,2.8, and 1.6, respectively, compared to values of 1.43, 1.32 and 1.13, respectively, as listed by Seed and Idriss. Use of the regressed MSF values may safely reduce calculated liquefaction hazard for moderate-sized earthquakes.(AU)

Prediction of liquefaction - induced ground displacement near bridges

Liquefaction - induced ground displacement resulting from lateral spread is a major causes of bridge damage during large earthquakes. Most damage of this type occurs at river crossings where bridges are founded on thick, liquefiable deposits of floodplain alluvium. An estimate of the probable ground displacement at potentially liquefiable sites in an important parameter for engineers to design safe bridges and to assess potential losses for lateral spread. Tothis and, case histories of lateral spread displacement resulting from several major earthquakes were analyzed using multiple linear regression (MIR) techniques. This paper discusses the type of lateral spread damage to bridges and presents and empirical model that can be used to estimate liquefaction-induced horizontal ground displacement near bridge crossings (AU)

Empirical prediction of lateral spread displacement

Data compiled from case histories of liquefaction-induced lateral spread are used to develop an empirical model for predicting the amount of horizontal ground displacement at potentially liquefiable sites. Earthquake, topographical, geological, and soil factors associated with lateralspreads from eight major earthquakes are analyzed. Multiple linear regression (MLR) is used to determine which factors are most strongly correlated with horizontal ground displacement and an empirical model is developed from those factors. The perfomance of the MLR model is evaluated by comparing the displacements measured at the case history sites with those predicted by the model.(AU)

Empirical analysis of horizontal ground displacement generated by liquefaction - induced lateral spreads

Liquefaction-induced ground failure is responsible for considerable damage to engineered structures during major earthquakes. Presently, few empirical techniques exist for estimating the amount of horizontal ground dusplacement resulting from liquefaction-induced lateral spread. None of these techniques fully addresses all the earthquake and site conditions known to influence ground displacement. This study compiles earthquake, geological, topographical, and soil factors that affect ground displacement and develops empirical models from these factors. Case histories of lateral spread are gathered from the 1906 San Francisco, 1964 Alaska, 1964 Niigata, 1971 San Fernando, 1979 Imperial Valley, 1983 Nihonkai-Chubu, 1983 Borah Peak, Idaho, and 1987 Superstition Hills earthquakes. Multiple linear regression (MLR) is used to develop empirical models from the compiled data. Two general models are derived herein, one for free face failures and one for ground slope failures. The predictive performance of the proposed empirical models is determined by comparing predicted displacements with those actually measured at the case history sites (AU)