Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
119 APPENDIX C-1 Centrifuge Experiment 1. Overview The primary purpose of the centrifuge experiment was to create a case history of strong ground shaking at a liquefiable site. The secondary purpose was to replicate a suite of three recorded Wildlife Liquefaction Array (WLA) case histories in a centrifuge experiment. An attempt was made to achieve both purposes in the same centrifuge experiment by designing it to resemble, as closely as possible, the upper 7.0 meters of the WLA site. The soil profile at the WLA site consists of an approximately 3.1 m thick clay1 layer (mostly CL and CH) underlain by an approximately 3.9 m thick layer of silty sand (SM). The WLA site was subject to strong ground shaking in the recent past. At the time of these past seismic events, the groundwater table was approximately 1.5 m below the ground surface (b.g.s.). The 35 g (i.e., the 37.8-g with adjustment for soil settlement) experiment was performed at the University of New Hampshire (UNH) centrifuge facility. The centrifuge, shake table, mounted laminar box, and an oblique view of the laminar box are shown in Figure C-1(a) through C-1(d), respectively. Figure C-1. Key Components of the UNH Centrifuge. 1 Plasticity Index (PI) of the WLA clay changes with depth and across the site. Depending on the sample recovery location, it was classified as Lean Clay (CL), sandy silt (ML), and Fat Clay (CH) in the past. Fat clay (CH) appears to be the dominant material in the profile and was used for the centrifuge experiment.
120 The model of the top 7.0 m of the WLA site (clay and sand layers) was built in the UNH laminar box. The box was shaken, in succession, by a suite of five ground motions. The first part of the centrifuge test, when excess pore pressure mainly accumulates, is considered a surrogate2 of an undrained Cyclic Direct Simple Shear (CyDSS) test and is used, as such, for comparisons. Relevant information about the centrifuge experiment is provided in the balance of this Appendix. Measurements, unless noted otherwise, are in the prototype scale (e.g., before shaking, the WLA soil profile is 7.0 m thick while the corresponding specimen in the laminar box is 241 mm thick). 2. Material Properties, Sample Preparation, Instrumentation, and Input Motions 2.1. Target Material Properties and Sample Preparation The specimen for the centrifuge experiment was prepared in a laminar box. This specimen mimics the WLA site profile with a target groundwater depth of 1.5 m below surface and a 3.1- m thick layer of (remolded) clay placed over a 3.9-m thick layer of (remolded) silty sand. A schematic representation of specimen is shown in Figure C-2. Also shown in Figure C-2 is a range of shear wave velocity at the site (target value was 100 m/s for clay and 125 m/s to 175 m/s for silty sand) and the results of shear wave velocity measurement on remolded clay and silty sand specimens in the CyDSS apparatus (220 m/s for clay and 207 m/s for silty sand; using the same specimen preparation technique as in the laminar box). Review of Figure C-2 reveals that tested specimen of clay and silty sand were stiffer than target, i.e., stiffer than soil within the WLA site profile. The specimen for centrifuge experiment was prepared under â1-gâ condition, i.e., it was prepared without spinning. The specimen was, however, spun after the saturation stage was completed, as explained below. The achieved shear wave velocity (i.e., shear wave velocity after spinning) was likely higher than specified values and likely closer to values measured in CyDSS by Bender Elements (BE), as explained below and acknowledged in footnotes of Table C- 1. The silty sand layer was prepared using a âdryâ pluviation method. The dry sand was poured by means of a crane and hopper system. Layers of silty sand were poured in 28 mm loose lifts, i.e., in thickness that allows for the placement of accelerometers and pore water pressure (PWP) transducers at target depths. After completion of placement of the silty sand layer, the container was filled with de-aired water (from the bottom of the laminar container; slow process). A thin layer of gravel, placed below the silty sand layer, allowed the water to permeate uniformly through the soil. The saturation process was terminated after the water 2 In a dynamic centrifuge test on saturated cohesionless soils, seismic loading is applied rapidly while the duration of strong shaking is relatively short. Therefore, one may expect that these tests are âtrulyâ undrained. However, dissipation of excess PWP occurs even during shaking, continues after cessation of shaking, and may even revert back to zero at the end of experiment. Dissipation occurs mainly through overlying soil (there are no side drains in the model). One may assess if the test is âtrulyâ undrained by reviewing the PWP records.
121 table reached the surface of the silty sand layer. Both the achieved dry and saturated unit weights matched the target values of 13.4 kN/m3 and 18.0 kN/m3, respectively. Figure C-2. Target Shear Wave Velocity Profiles compared to Measurements on Remolded Specimen (Black Line). BE and sCPT-1 to sCPT-5: This Study; A5 (OY O), sCPT-31, sCPT-38 and sCPT-50: Y oud et al. (2004); SASW, Crosshole-1, and Crosshole-2: Bierschwale and Stokoe (1984). The fat clay layer was prepared in a mold by the âwetâ tamping method. The target moist unit weight was 18.4 kN/m3. The target thickness of the clay layer in the mold was 89 mm. Following the tamping, the clay layer was cut from the mold to the dimensions of the laminar container and placed on the surface of the underlying, saturated, silty sand. Pocket penetrometer and vane shear tests were performed on excess trimmings from the lean clay layer. The values of the pocket penetrometer tests varied between 0.5 to 1.0 kg/cm2 with an average of 0.74 kg/cm2. Pocket vane shear values varied between 0.19 to 0.25 kg/cm2 with an average of 0.23 kg/cm2. For comparison, the same tests, performed using the same equipment, but on âintactâ samples of the same soil varied between 0.25 to 1.5 kg/cm2 with an average of 0.73 kg/cm2 (pocket penetrometer) and between 0.13 to 0.50 kg/cm2 with an average of 0.24 kg/cm2 (pocket vane shear).
122 Upon saturation of silty sand and the initial spin-up of the specimen in the centrifuge, specimen surface uniformly settled by approximately 15 mm or 8.1% (with respect to the initial height of 185 mm). The causes of the settlement are both an increase in unit weight (in a 3.9 m thick layer of silty sand) and consolidation that occurred during the initial spin (in a 3.1 m thick layer of fat clay). Simple calculations of primary consolidation, with the coefficient of consolidation estimated using the Skempton (1944) correlation, indicate that consolidation settlement equals 6.75 mm. The remaining 8.25 mm is due to an increase in self-weight. Due to settlement, the saturated unit of silty sand likely increased from its target value of 18.0 kN/m3 to 19.5 kN/m3 while the moist unit weight of fat clay increased from 18.4 kN/m3 to 19.9 kN/m3. Table C-1. Comparison of Target and Achieved Soil Parameters Soil 1) Shear Wave Velocity (m/s) Unit Weight (kN/m 3) Pocket Penetrometer (kg/cm2) Pocket Vane Shear (kg/cm2) Target(1) Achieved(1) Target Achieved Target Achieved Target Achieved Clay 100 < 220 18.4(2) 19.9(2),(5) 0.73(4) 0.74(4) 0.24(4) 0.23(4) Silty Sand A 125 < 207 18.0 (3) 19.5(3),(5) N/A N/A N/A N/A Silty Sand B 175 < 207 18.0 (3) 19.5(3),(5) N/A N/A N/A N/A (1) See range of target values in Figure C-1. Value measured by BE in CyDSS device is tentatively listed as achieved. Actual achieved values were not measured in laminar box but are likely close to their counterparts measured in the CyDSS device with BEs. (2) Moist unit weight. (3) Saturated unit weight. (4) Average of tested values. Measured on samples recovered from the laminar box. (5) Values after spinning in the centrifuge. These values account for soil settlement. N/A = Not Applicable. 2.2. Instrumentation To better simulate free-field site conditions (no structure in the profile), UNH used the laminar box (i.e., laminar container) that is shown in Figure C-1 and is further schematically represented in Figure C-3. Also shown in Figure C-3 is the test layout and instrumentation. The instrumentation was inserted within the silty sand layer and on top of the fat clay layer prior to the experiment (no instrumentation was placed within the fat clay layer). Instrumentation included accelerometers (i.e., Strong Motion, SM, instruments), PWP transducers (PWP sensors), and Linear Variable Differential Transformers (LVDTs).
123 Figure C-3. Cross Section through Laminar Box â Schematic of Instrumentation Dimensions are in Meters (Prototype Scale). To provide supplemental insight into the WLA site response to strong ground shaking, four accelerometers were installed within the silty sand layer (no accelerometers were installed within this layer in 1982 and two were installed in 2004). Number of PWP sensors installed within the same layer (four) is less than number of PWP sensors installed in 1982 (five) and number of PWP sensors installed in 2004 (eight). The LVDTs were installed at the top of the soil profile to record surface settlement during shaking and along the side of the laminar box to record the displacement profile. Following the preparation of the specimen, the laminar box (container) was placed on an in- flight shake table. The loaded container was first spun at the calculated g-level of 37.8 g (at the soil profile midpoint), so the centrifuge experiment is representative of a soil profile that is 7 m thick. This g-level was based upon the observed settlement of silty sand and lean clay during saturation and spun up in the centrifuge. After reaching the target g-level, the groundwater table was lowered to the target depth of 1.5 m. This was achieved by opening and closing valves and ports to allow water to drain from the specimen. After reaching the target groundwater table depth and achieving hydrostatic conditions, the self-weight induced settlement was 14.9 mm in the model scale or 0.55 m in the prototype scale, resulting in a specimen thickness of 6.45 m. A suite of five ground motions, calibrated and modified for reproducibility in the hydraulic actuator, was applied to the base of the laminar container. 2.3. Target and Achieved Input Motions The centrifuge experiment was performed with five motions. There was a 150 min (in prototype scale) delay between subsequent application of ground motions to allow for dissipation of excess PWP. The first three motions were recorded at the WLA site. They are
124 from a series of shallow crustal events and were applied in the same sequence they were recorded. The second motion in the sequence, the NS component of the Superstition Hills Earthquake, liquefied silty sand at the site. The PWP records from this event are available for comparison. The fourth and fifth motion, the 1977 M 7.68 Alaska Earthquake (Sitka Magnetic Observatory, MO, NS record) and the 1949 M 7.1 Western Washington Earthquake (Olympia, Washington Highway Testing Laboratory, WHTL, EW record) are representative of large- magnitude events. Relevant information about input (i.e., target) motions, and motions achieved at the base of laminar box, is provided in Table C-2 and its footnotes. Table C-2. Input Motions - Comparison of Target and Achieved Parameters No(1) Motion(2) Motion ID(3) Peak Ground Acceleration Arias Intensity Target(4) Achieved(5) Target Achieved 1 Elmore Ranch (NS) ELMR 0.078 g 0.079 g 0.10 m/s 0.11m/s 2 Superstition Hills (NS) SUPH 0.172 g 0.22 g 0.90 m/s 1.11 m/s 3 Hovley (EW) HOV 0.239 g 0.17 g 0.07 m/s 0.14 m/s 4 Sitka MO (NS) SIT 0.68 g 0.65 g 6.37 m/s 2.37 m/s 5 Olympia WHTL (EW) WW 0.68 g 0.57 g 2.79 m/s 7.51 m/s (1) Motions were applied at the base of the laminar container in this sequence (1 â 5). (2) An attempt was made to simulate the amplitude and frequency of these motions by a hydraulic actuator. (3) Identification label for both target and achieved ground motions. (4) For the ELMR and SUPH motions, this is Peak Ground Acceleration (PGA) recorded in-hole at 7.5 m b.g.s., (accelerometer was embedded 0.5 m into clay underlying silty sand). The target PGA for the SIT and WW motions equals to the highest PGA that can be applied in a 35-g experiment by the UNH hydraulic actuator (0.68 g). (5) Refers to PGA achieved at the base of the UNH laminar box. PGA = Peak Ground Acceleration (at the base of the laminar container); AI = Arias Intensity (index of motion energy; the inherent property of ground motion). Target motions are applied to the base of laminar box by means of hydraulic actuator. Without corrections, motion at the base of hydraulic actuator differs from its specified counterpart. To minimize the difference, modifications applied in an iterative process are required. The process is lengthy and tedious. Seismic hazard parameters of target and achieved motions, such as PGA and Arias Intensity (AI) are compared in Table C-2. Plots of AI (Husid Plots) are compared in Figure C-4. Figure C-4. Husid Plots of the Motions at the Base of Laminar Box (â Achieved Motionsâ ).
125 Plots of target and achieved acceleration histories and plots of target and achieved acceleration response spectra (5% damping) are compared in Table C-3. Table C-3. Comparison of Input and Achieved Motions. Spectra are 5% Damped ELMR â Acceleration Histories ELMR - Spectra SUPH â Acceleration Histories SUPH - Spectra HOV â Acceleration Histories HOV - Spectra SIT â Acceleration Histories SIT - Spectra
126 WW â Acceleration Histories WW - Spectra In an effort to compare the in-situ recorded response of WLA to its centrifuge counterpart, special attention was devoted to matching of the ELMR, SUPH, and HOV acceleration histories. As shown in Tables C-2 and C-3, the low-intensity ELMR motion is matched the best. The SUPH is a higher intensity motion which is more difficult to match. With exception of the most prominent spectral peak, match is relatively good. The HOV motion is matched poorly, however. This is because the HOV motion is a âpulseâ motion, i.e., a motion of the type that is very difficult to match in a centrifuge. 3. Centrifuge Experiment Results - ELMR Motion 3.1. Acceleration and PWP Response Acceleration histories recorded at the base of the laminar box, within silty sand, and at the top of fat clay are presented in Table C-4. Also presented in Table C-4 are excess PWP histories recorded within silty sand. The corresponding instrument depths were corrected to account for settlement that occurred during the soil saturation effort and further during the initial spin-up of the box. An accelerometer at a depth of 3.05 m b.g.s. malfunctioned during the centrifuge experiment (see acceleration history reported for depth of 3.05 m in Table C-4).
127 Table C-4. Acceleration and PWP Response - ELMR. Model Surface Base (Input) ru = / â² where is the Excess PWP; â² = the Initial Effective Stress at Depth. PWP record at depth = 5.33 m affected by ânoiseâ generated by the limitations of the 64-bit data acquisition board.
128 Figure C-5 is a supplemental presentation of recorded acceleration response with depth. It reveals amplification of input motion from PGA = 0.078 g at a depth of 7.0 m below the model surface to approximately PGA = 0.15 g at the model surface. The corresponding PGA amplification factor (FPGA) is approximately 1.9. ( a) ( b ) Figure C-5. (a) PGA Profile; (b) FPGA Profile - ELMR. 3.2. Settlement of Model Surface Settlement history of the laminar box surface response in vertical direction (i.e., settlement) is shown in Figure C-6. Figure C-6. Surface Settlement - ELMR. 3.3. Lateral Displacement The profile of recorded maximum lateral displacement, as measured along the right side of the laminar container, is shown in Figure C-7. When excited with the ELMR motion, the top of the container displaced for approximately 12 mm. The LVDT installed at a depth of about 5.05 m b.g.s. (i.e., bottom LVDT in Figure C-7) malfunctioned.
129 Figure C-7. Profile of Recorded Maximum Lateral Displacement - ELMR. 4. Centrifuge Experiment Results - SUPH 4.1. Acceleration and PWP Response Acceleration histories recorded at the base of the laminar box, within silty sand, and at the top of fat clay are presented along with excess PWP histories recorded within silty sand in Table C- 5. The corresponding instrument depths were corrected to account for settlement that occurred during the soil saturation effort and further during the initial spin-up of the box. The accelerometer at a depth of 3.05 m b.g.s. malfunctioned during the experiment. Table C-5. Acceleration and PWP Response - SUPH. Model Surface Instrument Malfunctioned
130 Base (Input) ru = / â² where is the Excess PWP and â² is the Initial Effective Stress at Depth. Figure C-8 is a supplemental presentation of recorded acceleration response with depth. It reveals amplification of input motion from PGA = 0.216 g at depth of 7.0 m below model surface to approximately PGA = 0.288 g at the model surface. The corresponding PGA amplification factor (FPGA) is approximately 1.3. ( a) ( b ) Figure C-8. (a) PGA Profile; (b) FPGA Profile - SUPH. 4.2. Settlement of Model Surface Surface settlement history of the laminar box surface response in vertical direction (i.e., settlement) is shown in Figure C-9.
131 Figure C-9. Surface Settlement History - SUPH. 4.3. Lateral Displacement The profile of recorded maximum lateral displacement, as measured along the right side of the laminar container, is shown in Figure C-10. When excited with the SUPH motion, the top of the container displaced for approximately 44 mm. The bottom LVDT malfunctioned. Figure C-10. Profile of Recorded Maximum Lateral Displacement - SUPH. 5. Centrifuge Experiment Results - HOV 5.1. Acceleration and PWP Response Acceleration histories recorded at the base of the laminar box, within silty sand, and at the top of fat clay are presented along with excess PWP histories recorded within silty sand in Table C-6. The corresponding instrument depths were corrected to account for settlement that occurred during the soil saturation effort and further during the initial spin-up of the box. The accelerometer at a depth of 3.05 m b.g.s. malfunctioned during the experiment.
132 Table C-6. Acceleration and PWP Response - HOV. Model Surface Instrument Malfunctioned Base (Input) ru = / â² where is the Excess PWP and â² is the Initial Effective Stress at Depth.
133 Figure C-11 is a supplemental presentation of recorded acceleration response with depth. It reveals amplification of input motion from PGA = 0.168 g at depth of 7.0 m below model surface to approximately PGA = 0.226 g at the model surface. The corresponding PGA amplification factor (FPGA) is approximately 1.3. ( a) ( b ) Figure C-11. (a) PGA Profile; (b) FPGA Profile - HOV. 5.2. Settlement of Model Surface Settlement history of the laminar box surface response in vertical direction (i.e., settlement) is shown in Figure C-12. Figure C-12. Surface Settlement History - HOV. 5.3. Lateral Displacement The profile of recorded maximum lateral displacement, as measured along the right side of the laminar container, is shown in Figure C-13. When excited with the HOV motion, the top of the container was displaced by approximately 41 mm. Two bottom LVDTs malfunctioned.
134 Figure C-13. Profile of Recorded Maximum Lateral Displacement - HOV. 6. Centrifuge Experiment Results - SIT 6.1. Acceleration and PWP Response Acceleration histories recorded at the base of the laminar box, within silty sand, and at the top of fat clay are presented along with excess PWP histories recorded within silty sand in Table C- 7. The corresponding instrument depths were corrected to account for settlement that occurred during the soil saturation effort and further during the initial spin-up of the box. The accelerometer at a depth of 3.05 m b.g.s. malfunctioned during the experiment. Table C-7. Acceleration and PWP Response - SIT. Model Surface Instrument Malfunctioned
135 Base (Input) ru = / â² where is the Excess PWP and â² is the Initial Effective Stress at Depth. Figure C-14 is supplemental presentation of recorded acceleration response with depth. It reveals amplification of input motion from PGA = 0.547 g at depth of 7.0 m below model surface to approximately PGA = 0.247 g at the model surface. The corresponding PGA amplification factor (FPGA) is approximately 0.45 g. ( a) ( b ) Figure C-14. (a) PGA Profile; (b) FPGA Profile - SIT.
136 6.2. Settlement of Model Surface Settlement history of the laminar box surface response in vertical direction (i.e., settlement) is shown in Figure C-15. Figure C-15. Surface Settlement History - SIT. 6.3. Lateral Displacement The profile of recorded maximum lateral displacement, as measured along the right side of the laminar container, is shown in Figure C-16. When excited with the SIT motion, the top of the container displaced for approximately 47 mm. Two bottom LVDTs malfunctioned. Figure C-16. Profile of Recorded Maximum Lateral Displacement - SIT.
137 7. Centrifuge Experiment Results - WW In this section similar set of results for the tests under WW motion is presented. 7.1. Acceleration and PWP Response Acceleration histories recorded at the base of the laminar box, within silty sand, and at the top of fat clay are presented along with excess PWP histories recorded within silty sand in Table C- 8. The corresponding instrument depths were corrected to account for settlement that occurred during the soil saturation effort and further during the initial spin-up of the box. The accelerometer at a depth of 3.05 m b.g.s. malfunctioned during the experiment. Table C-8. Acceleration and PWP Response - WW. Model Surface Instrument Malfunctioned
138 Base (Input) ru = / â² where is the Excess PWP and â² is the Initial Effective Stress at Depth. Figure C-17 is a supplemental presentation of recorded acceleration response with depth. It reveals the amplification of input motion from PGA = 0.569 g at depth of 7.0 m below model surface to approximately PGA = 0.349 g at the model surface. The corresponding PGA amplification factor (FPGA) is approximately 0.61. ( a) ( b ) Figure C-17. (a) PGA Profile; (b)FPGA Profile - WW. 7.2. Settlement of Model Surface Settlement history of the laminar box surface response in vertical direction (i.e., settlement) is shown in Figure C-18.
139 Figure C-18. Surface Settlement History - WW 7.3. Lateral Displacement The profile of recorded maximum lateral displacement, as measured along the right side of the laminar container, is shown in Figure C-19. When excited with the WW motion, only the top functioned and recorded a 74-mm displacement. Figure C-19. Profile of Recorded Maximum Lateral Displacement - WW. 8. Cumulative Settlement of Laminar Box The response of laminar box at the end of shaking is summarized in Table C-9. Reported settlement values are in the prototype scale. Values of relevant ground motion parameters are provided for reference. Table C-9. Cumulative Surface Settlement of Laminar Box No Event / Motion Motion ID) Motion Parameter Surface Settlement PGA AI End of Event Cumulative 1 Elmore Ranch (NS) ELMR 0.079 g 0.11m/s 4.1 mm 4.1 mm 2 Superstition Hills (NS) SUPH 0.22 g 1.11 m/s 19.4 mm 23.5 mm 3 Hovley (EW) HOV 0.17 g 0.14 m/s 3.8 mm 27.3 mm 4 Sitka MO (NS) SIT 0.65 g 2.37 m/s 42.2 mm 69.5 mm 5 Olympia WHTL (EW) WW 0.57 g 7.51 m/s 110.3 mm 179.8 mm PGA = Peak Ground Acceleration (at the base of the laminar container); AI = Arias Intensity (index of motion energy; the inherent property of ground motion).
140 At the end of shaking, the thickness of the WLA soil profile is reduced by approximately 2.6 %, i.e., from 7.0 m to approximately 6.8 m. Measured settlement correlates well with ground motion parameters. 9. Discussion The primary goal of the centrifuge experiment (i.e., create a supplemental case history of liquefied soil) has been achieved. The Olympia WHTL (labelled as WW in figures) record liquefied the specimen, the relevant information was recorded in the experiment, and gathered information is sufficient to create a case history of strong ground shaking at a liquefiable site. The secondary goal of the centrifuge experiment (i.e., to replicate three WLA site case histories) has been partially achieved. Relatively good agreement between recorded in the field and in the centrifuge has been achieved for the laminar box excitation with the Elmore Ranch motion (see Table C-3). However, the same physical model did not liquefy when subjected to the Superstition Hills motion, which liquefied the WLA site in 1987. The centrifuge model subject to the Superstition Hills motion did not liquefy because: (i) silty sand in the centrifuge model was approximately 8% denser than its 1987 in-situ counterpart (19.5 kN/m3 vs. 18.0 kN/m3 in-situ; see Section 2.1 above); (ii) silty sand was further densified by the preceding Elmore Ranch experiment; (iii) the input (i.e., base of the laminar box) motion was not matched very well (see Table C-3); and (iv) it is possible that sand was not fully saturated prior to the centrifuge experiment. If it was not fully saturated, the air bubbles would have increased the liquefaction resistance. In retrospect, despite difficulties associated with clouding of water that would prevent proper control of layer thickness which is important for proper placement of instrumentation within the laminar box, it would have been better to prepare the layer of silty sand with wet pluviation. The material should have been placed at an initial unit weight that is lower than the target. PWP response of the specimen to Hovley motion is 94 % lower than recorded in the field (recorded ru in the middle of the silty sand layer is 4% while the recorded counterpart in the field is ru = 65%). The explanation for the mismatch between measured and calculated is analogous to the explanation provided above for the lack of liquefaction in the 1987 Superstition Hills event. Throughout the experiment, three out of the four LVDTs set to record lateral deformation of the laminar box malfunctioned. This was likely due to the failure of glued contacts when they were subjected to relatively high shaking (PGA ⥠0.2 g) that can break the bonds. However, with the exception of the laminar box displacement profile, this instrument malfunction had no impact on the centrifuge data interpretation.
141 There are two options to saturate the soil specimen in a centrifuge experiment: (i) with de-aired water; and/or (ii) with metolose3. Given that, for instrumentation purposes, specimen has to be built in layers, and model described herein included layers of fat clay and silty sand, saturation was performed with de-aired water. The scaling factors for dynamic time and diffusion time in a centrifuge experiment on a soil sample saturated with water are different, hence the dissipation of the excess PWP would occur faster than in the field. Hydraulic conductivity of soil saturated with de-aired water is higher than hydraulic conductivity of soil saturated with metolose. Therefore, excess PWP in samples saturated with de-aired water dissipates faster than in the field. References Bierschwale, J.G. and Stokoe, K.H., II (1984), âAnalytical Evaluation of Liquefaction Potential of Sands Subjected to the 1981 Westmorland Earthquake,â Geotechnical Engineering Report GR 84-15, University of Texas, Austin, Texas. GLA (2021), âSite Exploration Report Wildlife Liquefaction Array Imperial Wildlife Area, Imperial County, California,â Technical Report, Geo-Logic Associates, Inc., Costa Mesa, California. Skempton, A. W. (1944), âNotes on the Compressibility of Clays,â Quarterly Journal of the Geological Society, 100(1-4), pp. 119-135. Youd, T. L., Bartholomew, H. A., and Proctor, J. S. (2004), âGeotechnical Logs and Data from Permanently Instrumented Field Sites: Garner Valley Downhole Array (GVDA) and Wildlife Liquefaction Array (WLA),â Technical Report, Deptartment of Civil and Environmental Engineering, Brigham Young University, Provo, Utah. 3 Metolose is a high-viscosity material. It is difficult to saturate cohesive and silty soil with liquid of such viscosity. Soils saturated with metolose cannot be re-used â another concern with this type of testing.