CHAPTER 2
Review of Related Literature
2.1 Foreign Literatures 2.1.1 The Simplified Method The Simplified Method is the most common procedure for evaluating the liquefaction potential of soil. Developed by H.B. Seed and I.M. Idriss in 1971, this method uses two fundamental terms: (1) Cyclic Stress Ratio (CSR), and (2) Cyclic Resistance Ratio (CSR). The CRR characterizes the liquefaction resistance of granular soils and is typically acquired from in situ tests. It can also be estimated by cyclic tests on reconstituted soil samples. According to a summary report by Youd and Idriss (2001), the 1996 NCEER and 1998 NCEER/ NSF Workshops reviewed the modern method of analysing the potential liquefiable layers of soil. They recommended revised criteria for routine evaluation of CRR from various tests such as the Cone Penetration Test (CPT) and the Standard Penetration Test (SPT). Robertson and Wride (1998) informed that CRR evaluated by CPT may be adequate for low-risk, small-scale projects. But for medium to high-risk projects, there may be more than one method that should be used to determine the CRR. Idriss and Boulanger (2004) commented that the quality of the site characterization is the greatest factor that could affect any liquefaction evaluation. The workshops concluded that it might be much more specific when two or more tests should be used for a more reliable value of CRR. As for the evaluation of CRR from laboratory tests, the chief hindrance is the acquiring of in situ samples. Porcino and Ghionna (2002) said that reconstituted samples that seem to have not veered from its in situ condition is still doubtable, unless sampling techniques such as ground freezing are used. The CSR, on the other hand, is evaluated by four parameters: (1) the depth of soil, (2) the unit weight of soil, (3) pore water pressure, and (4) the maximum horizontal acceleration of the earthquake considered at the soil surface. After the determination of the values of CRR and CSR, the liquefaction resistance against an earthquake is then commonly expressed in terms of a factor of safety defined as:
FSL=CRRCSR
In the usual computation of the liquefaction resistance, the maximum horizontal acceleration of the earthquake (amax) is usually that associated with a single return period (e.g., 475 years). The soil is proved to resist the earthquake-induced liquefaction of a certain moment magnitude when the factor of safety against liquefaction exceeds unity. In other studies by Martin and Lew (1999), the soil deposit is said to have adequate resistance if the FSL computed with the aforementioned parameters exceeds some minimum value (e.g., 1.1-1.3).

Figure 2.1 – Contribution of Various Magnitudes and Distances to the Seismic Hazard (Moriwaki et al., 1994)

2.1.2 Liquefaction Analysis from Standard Penetration Resistance The Standard Penetration Test (ASTM D-1586) started from the experiments of Charles R. Gow in 1902 when he introduced a method of dry sample recovery. He developed a sampling tube having a diameter of 1-in., driven with a 110-lb weight. Later in 1922, the Charles R. Gow Company was acquired by Raymond Concrete Pile where L. Hart and G.A. Fletcher devised a split-spoon sampler of 2-in. diameter. Following a series of tests, Fletcher and Harry A. Mohr standardized the penetration test using the split-spoon sampler, driven with a 140-lb weight, dropping 30 in. (76.2 cm). There were a lot of improvements that were made in the next years to ensure there were no losses of material once the sampling tube from the boring is removed. (Notes from Gordon A. Fletcher). The Standard Penetration Test (SPT) was originally used to provide data for design of deep foundations. Because of its widespread use in the U.S., Karl Terzaghi and Ralph B. Peck further extended the application to shallow foundations as well. As what was mentioned a while ago, the Standard Penetration Test is used to determine the value of CRR in the Simplified Method. This is defined by the corrected SPT blow counts. Another way of evaluating the liquefaction potential is the construction of correlation charts with the standard penetration resistance. Seed, Mori, and Chan (1977) presented a detailed collection of field data for liquefaction potential in a graphical form which is a plot of (τhσv) versus N’. N’ is the corrected standard penetration resistance for an effective overburden pressure of 95.76 kPa. The correlation chart then shows the lower bounds of the correlation curve where liquefaction occurred. Since the early correlation charts were not accurate because they do not take into the account the magnitude of the earthquake and the duration of shaking, Seed (1979) considered the results of the large-scale simple shear test conducted by DeAlba, Chan, and Seed in 1976. From their studies, a new correlation curve was traced which included the effect of the magnitude of the earthquake and other significant factors. 2.1.3. Liquefaction caused by Olancha M5.2 earthquake Olancha M5.2 Earthquake on October 3, 2009 caused an extensive liquefaction and a vast horizontal ground deformation. Such behavior of soil is barely observed on earthquakes with moment magnitude less than 5.2. Because reports of liquefaction at small earthquake magnitude are rare, the Simplified Procedure (Seed and Idriss, 1971) was used to test how well it would predict the Olancha liquefaction. The popular National Center for Earthquake Engineering Research (NCEER) version of the procedure for the analysis (Youd et al., 2001) was also used.
The most common field penetration test used to assess liquefaction potential is the Standard Penetration Test (SPT). In performing the SPT, a standardized soil sampler is hammered into the soil and the number of blows required to advance the sampler one foot is counted. The field blow count is normalized for stress and corrected for a variety of factors (Idriss and Boulanger, 2008). From residual vane resistance, blow counts were inferred. It was assumed that residual vane shear resistance in the sand approximates residual shear strength, and used a recent correlation of residual strength with SPT blow counts (Idriss and Boulanger, 2008). From the experiment the standard vane resistance at one standard deviation was 1.9 and 8.3 kPa which implies that blow counts ranges from 3.5 to 8.0 blows/ft. From experience, low blow counts simply means that the soil is weak. The Simplified Procedure is referenced to a M7.5 earthquake, and a magnitude scaling factor (MSF) is used to adjust for different magnitudes. M5.2 earthquake liquefaction was computed through boundary curve with both the lower and upper bound MSFs recommended by the NCEER version (Youd et al., 2001). The lower bound MSF produces a more conservative prediction than does the upper bound MSF, and is therefore more commonly used in engineering practice. The prediction of liquefaction potential at LIQ with the upper bound MSF is more ambiguous, but the proximity of the observed point to the boundary curve would prompt additional investigation in most applications of the simplified procedure. Liquefaction and lateral spreading during the Olancha M5.2 earthquake provided a rare opportunity to investigate the factors that contribute to liquefaction during small magnitude earthquakes. Liquefaction during small magnitude earthquakes presumably depends on the presence of either or both very susceptible soil and anomalously high ground motion. 2.1.4. Assessment of Liquefaction Evaluation at Christchurch
The study presented herein is to assess three commonly used CPT-based liquefaction evaluation procedures and three liquefaction severity index frameworks using data from the 2010–2011 Canterbury earthquake sequence. Liquefaction is a phenomenon in which the strength and stiffness of a soil is reduced by earthquake shaking or other rapid loading.
The records and observations and results from the geotechnical site investigation in the city of Christchurch and the surrounding towns are used herein. Unlike similar studies that used data from free-field sites, accelerogram characteristics at the strong motion stations (SMS) locations can be used to assess the performance of liquefaction evaluation procedures prior to their use in the computation of surficial manifestation severity indices. Accelerogram measures the acceleration, velocity, displacement during an earthquake. The magnitude of post-liquefaction consolidation settlement (S) has also been proposed as a proxy for the severity of surficial liquefaction manifestation.

; where εv is the volumetric strain due to post-liquefaction consolidation of soil layer i, Δzi is the thickness of layer i, and j is the number of soil layers. The results obtained from the study indicate that for cases with liquefaction triggering the accelerograms, the majority of liquefaction evaluation procedures generate correct predictions, regardless of whether surficial manifestation of liquefaction was noticeable or not. When all cases are used to determine the performance of liquefaction severity index frameworks, a poor correlation is shown between the observed severity of liquefaction surface manifestation and the calculated severity indices. 2.1.5. Simplified Procedures for Assessing Soil Liquefaction Simplified methods to assess the effects of saturated sandy soil liquefaction are necessary for reasonable earthquake resistant design of structures considering the surrounding soil liquefaction. Based on the proposed methods by authors Toshio Iwasaki, Tadashi Arakawa and Ken-Ichi Tokida in their International Journal of Soil Dynamics and Earthquake Engineering (1984), the liquefaction potential can be estimated simply by using the fundamental properties of soils, i.e. N-values of the standard penetration test, unit weights, mean article diameters, and maximum acceleration at ground surface. Two Simplified Methods are introduced such, it is with the use of the liquefaction resistance factor FL which was tackled in the first part, the Simplified Method, and a liquefaction potential index IL to assess the liquefaction potential of saturated sandy soils. Liquefaction susceptibility analysis is the first step in the soil liquefaction engineering assessment. Liquefaction potential analysis is carried out with two essential parts, the deterministic (Simplified Procedure); the first part discussed, and probabilistic (Probability Density Function) methods for comparison purposes. Both deterministic and probabilistic based liquefaction potential analyses are based on standardised corrected earthquake magnitude of 7.5. Quantitatively, the total number of liquefiable sites in the international criteria is in good agreement for both deterministic and probabilistic method. More and more consensus studies are underway to improve the current assessment criteria. Instead of clay fraction, researchers believed that plasticity index is more relevance in representing the basic characteristic of fines in altering soil liquefaction behaviour, especially the source of liquefiable sites. Therefore, it becomes a good effort to incorporate both the results of both deterministic and probabilistic findings as one assessment criteria in order to serve as better assessment criteria.

2.1.6. Calibration of a CRR Model for Assessing Soil Liquefaction Potential Derived from the lengthened SPT-based database of the Idriss and Boulanger (2010) and Xie (1984) case histories, an innovative calibrated empirical liquefaction triggering correlation equation with five parameters could be possibly determined by means of trial and error. According to Guoxing Chen et al. liquefaction, transition and non-liquefaction zones for clean sands are defined based on different probability contours, i.e., empirical correlations with different liquefaction probabilities. The Cone Penetration Test (CPT) and Standard Penetration Test (SPT) are the two most widely used indices for evaluating the liquefaction characteristics of soils. The SPT was used first in developing liquefaction correlations but the CPT has a number of advantages that have made it the primary site characterization tool in certain geologic settings. The CRR or Cyclic resistance ratio is correlated to CPT and SPT penetration resistances after application of procedural and overburden stress corrections. For SPTs, the various procedural corrections for arriving at a standardized, energy-corrected N60 value are summarized in Idriss and Boulanger (2008, 2010) and thus not repeated herein. The soil's CRR is dependent on the duration of shaking (which is expressed through an earthquake magnitude scaling factor, MSF) and effective overburden stress (expressed through a K factor). The correlation for CRR is therefore developed by adjusting the case history. The soil's CRR is further affected by the presence of sustained static shear stresses, such as may exist beneath foundations or within slopes. The effect of sustained static shear stresses, which may be expressed through a K factor, is generally small for nearly level ground conditions and is not included herein because the case history database is dominated by level or nearly level ground conditions. According to Idriss and Boulanger (2008), a liquefaction analysis procedure concerning CRR-based liquefaction triggering correlations within the above stress-based framework ideally has the following two attributes:

1. The liquefaction analysis procedure is applicable to the full range of conditions important to practice; e.g., shallow lateral spreads settlement of structures, stability failures or deformations in levees or earth dams. Practice often results in the need to extrapolate outside the range of the case history experiences, requiring the framework to be supported by sound experimental and theoretical bases for guiding such extrapolations. 2. The functional relationships used to describe fundamental aspects of soil behavior and site response are consistent with those used in developing companion correlations for the other in-situ tests; e.g., SPT blow count, CPT penetration resistance, and shear wave velocity, Vs. Consistency in these functional relationships facilitates the logical integration of information from multiple sources and provides a rational basis for the calibration of constitutive models for use in nonlinear dynamic analyses.

2.2 Local Literatures

2.2.1 Seismic Hazard Analysis of Laoag City using Probabilistic and Deterministic approach. With Philippines lying in the Pacific Ring of Fire, studies on the assessment of liquefaction potential on one of its cities, Laoag City, were performed. This study is focused on the seismic hazard analysis through probabilistic and deterministic approaches and on earthquakes which possess magnitude-distance combinations that can cause liquefaction in the study area as defined by the empirical relationship of Youd and Perkins (1978). The probabilistic approach shows the evaluation on size, location, and rate of occurrences of earthquakes. For this given approach, it has been said that there is a limit to the distance from an earthquake source to induce liquefaction. Youd and Perkins (1978) developed a magnitude distanced criterion based from the observation on the analysis of past earthquakes and most distant effects of liquefaction and arrived with a 150-km distance from the earthquake source. This distance was used in the study to determine those earthquakes that falls within the radius of the study area. The result from this approach was then compared to the previous earthquakes struck the study area. On the other hand, deterministic approach evaluates the most severe earthquake on the site regardless of its occurrence. Basing from the magnitude and distance combination, the earthquake studied in the analysis is the largest ground motion magnitude and the nearest to the study area Kramer (1996) and Gupta (2002). The study achieved the result of having an earthquake with a magnitude of Ms 5.2 to 5.8 to occur at distance of 0-10 km from the study area in probabilistic approach. It has been stated that previous earthquakes are within this magnitude range that cause liquefaction. This study determined the worst-case scenario earthquake with a magnitude of Ms 8.2 occurring 7 km from the study area based on the deterministic approach. 2.2.2 New approach to Liquefaction Hazard Zonation This study mainly focuses on providing a liquefaction hazard zonation map which are the primary tools in mitigating liquefaction. It requires two maps which are combined to produce an integrated liquefaction hazard zonation map. The first map is called the preliminary liquefaction susceptibility map. This map is obtained by analyzing the geomorphological unit of the study area. It has been observed by the earlier workers that deposits with uniform grain distribution and are in loose state are likely to liquefy when saturated. From the obtained data, this study ranked the susceptibility of each distinct units in the study area by its geomorphological make-up and by its geomorphological evolution. The second map, site classification map, requires the microtremor measurements. Microtremor provides the qualitative information of soil deposits, notably its density and thickness. This study used the HSVR technique by Nogoshi and Igarashi and revisited by Nakamura in approaching the microtremor study due to its ease of application. The study confirms the validity of the liquefaction hazard zonation map through the quantitative liquefaction potential assessment. Like the previous study in Laoag City, seismic hazard analysis is used, and arrived with the same results.This study also used Simplified Method by Seed and Idriss. Datas obtained from the boreholes were used in this method. The results were expected to have the factor of safety decrease as the degree of ground shaking is increased.