13th World Conference on Earthquake Engineering Vancouver, B.C., CanadaAugust 1-6, 2004

MODELS OF SEISMIC MICROZONING FOR APSHERON
PENINSULA AND BAKU METROPOLIS, AZERBAIJAN
Oktay BABAZADE¹, Gulam BABAYEV², Nigyar BABAZADE3
SUMMARY
The basic point in seismic microzoning study is to provide strong seismic motions over the areas and
vulnerability to destructive earthquakes. Seismic microzoning maps allow conducting estimations of
the seismic hazard, evaluating probable damages to infrastructures due to earthquake disasters. For
carrying out these aspects successfully, it is fundamentally necessary to determine careful
investigation of a researching area in terms of not only seismicity, but also surface and base geology,
geomorphology, topography and soil conditions. In this study, Apsheron Peninsula, including Baku
City-the capital of Azerbaijan Republic was researched by seismicity, target earthquakes, attenuation,
soil and rock properties, geological crossection, the boring data of measured shear-wave velocity,
ground modeling, amplifications and base rock and surface ground motions. The process of research
was done using visualization techniques applying SHAKE program (Japanese version) and MapInfo
Professional 4.5 (USA). Defining an amplification factor of the each geological unit through shear
wave velocity was important and necessary in order to apply attenuation formula for calculating Peak
Ground Acceleration (PGA) at base rock, because calculating PGA at surface seismic motion was
proceeded by multiplication PGA at the base rock with amplification parameter of each surface layers
all the way through 3800 meters. The result of data processing, modeling, mapping, interpreting made
it possible to identify certain regularities in the attenuation of the earthquake intensities, to classify soil
conditions, to determine peak ground acceleration at base rock and ground surface caused by target
earthquakes. In this approach, Quaternary type of deposits were examined in detail, because for the
areas with living objects, villages, small towns Quaternary outcrops represent the fact of risk with
increasing PGA value at the surface. In this paper, there was an attempt to apply the relationship
between PGA values with MSK scale of seismic intensity. Finally, seismic microzoning numerical
models of PGA value was drawn which gives the possibility for estimating a level of a seismic motion,
the visual seismic intensity picture of the researched area for countermeasures plans, for future
assessment and for mitigating the level of future disasters.

A PERFORMANCE TEST STUDY ON CHINESE G4 LEAD RUBBER
BEARINGS
Demin FENG1, Cliff CHEN2, Wenguang LIU3, Kiyoshi TANAKA 4
SUMMARY
In this paper, the soft G4 type Chinese lead rubber bearings are studied in detail. Test contents covered
fundamental properties including ultimate properties, various dependence properties and durability
properties. In the fundamental property tests, results of 102 products were compiled statistically. Average
values and standard deviations of the ratio between a test result and its catalog value were used as
evaluating parameters. For the compressive stiffness, post-yield stiffness and yield load of the lead plug,
the values of the parameters were obtained as 1.05 and 0.06, 1.04 and 0.05, 1.03 and 0.06, respectively.
The products are proved to have steady quality. Tensile yielding load was confirmed to be 1.12MPa by a
φ600 specimen at the shear strain of 100%. Ultimate property diagram was confirmed by several
specimens. For specimens with S2 greater than 5, φ600, φ1000 and φ1100 specimens were confirmed
exceeding the shear strain of 400% under the design compressive force. Two φ700 specimens and one
φ600 specimen with S2 smaller than 5, were tested until buckling occurred to confirm the validity of the
proposed ultimate property diagram. The dependence tests were performed for factors such as shear strain,
compressive force and temperature. For shear strain and compressive force dependence property tests,
φ600 ~ φ1100 products were used. The durability property tests consisting of degradation and creep tests
were conducted to make sure the bearing performs well after sixty years service. From the degradation test
of a φ300 specimen, it was found that the post yield stiffness increased by +4% and the yield load
increased by +2%. The creep was estimated to be 4.4% for a φ300 specimen.

EXPERIMENTAL AND NUMERICAL MODELING OF THE LATERAL
RESPONSE OF A PILE BURIED IN LIQUEFIED SAND
Jonathan RIVERA DUNGCA1, Jiro KUWANO2 and Akihiro TAKAHASHI3
SUMMARY
This paper presents experimental and numerical modeling of the lateral resistance of the pile subjected to
liquefaction-induced lateral flow. The main objective of this study is to model the behavior of the soil
surrounding the pile when a large relative displacement between the pile and the soil is induced.
A series of shaking table tests were undertaken to observe the soil surrounding the pile during
liquefaction. The pile was modeled as a buried cylinder that corresponded to a sectional model of a
prototype pile at a certain depth in the subsoil. In order to create a realistic stress condition in the model
ground, the model was prepared in a sealed container and the overburden pressure was applied to the
ground surface by a rubber pressure bag. The model pile was actuated back and forth through rods
attached on each side by an electro-hydraulic actuator.
To verify the results gathered in the experimental modeling, a numerical modeling was prepared using a
finite element program created exclusively for this purpose.
Test results show that a larger resistance is mobilized as the loading rate becomes higher. When the
loading rate is higher, the cylinder displacement required for the lateral resistance becomes smaller. This
behavior of the soil surrounding the pile is further verified using numerical modeling.

A STATISTICAL PROCEDURE FOR THE ASSESSMENT OF SEISMIC
PERFORMANCE OF EXISTING REINFORCED CONCRETE BUILDINGS
IN TURKEY
Ahmet YAKUT1, Guney OZCEBE2, and M. Semih YUCEMEN2
SUMMARY
A procedure is developed to assess the seismic performance of low- to mid-rise reinforced concrete
buildings using a statistical approach and the seismic damage database compiled from recent earthquakes
in Turkey. The database is based on the post-earthquake damage evaluations conducted after the 1999
earthquakes in Turkey and contain detailed information on some selected buildings that suffered various
degrees of damage. The detailed information collected for each building were processed and analyzed in
order to investigate and establish correlations between the building attributes that are believed to affect the
seismic performance and the observed damage. In this procedure, the building to be evaluated is required
to be visited to collect data on its structural layout, plan and vertical member dimensions, number of
stories and architectural features such as soft stories and overhangs. A damage score, obtained from a
discriminant function utilizing the building-specific information, is compared with a cutoff value to arrive
at a decision regarding its performance.
Seismic performance evaluation is carried out in two stages, first for life safety and then for immediate
occupancy performance levels. The building is then classified as safe or unsafe depending on the results
of these evaluations. Since the cutoff values were derived based on the seismic damage database compiled
from a particular earthquake and at a particular site, further endeavor was devoted to the improvements
and adjustments that would allow the application of the proposed procedure to other regions. For this
purpose the variability of ground motion with respect to the soil properties and the distance to source were
incorporated into the analysis. The sites are classified according to Turkish Seismic Code’s definitions
based on the shear wave velocity. The procedure is verified through application to a number of buildings
that were subjected to some of the recent earthquakes in Turkey.

DEVELOPMENT OF THE INELASTIC DEMAND SPECTRA
CONSIDERING HYSTERETIC CHARACTERISTICS AND
SOIL CONDITIONS
Hyun-Ho LEE1, Kee-Tae HWANG2
SUMMARY
The inelastic demand caused by earthquake load is strongly related to the natural period, displacement
ductility ratio, hysteretic characteristics of structures and soil conditions. The objective of this study is a
determination method proposition for the inelastic demand of earthquake hazards for the using in capacity
spectrum method. Five different hysteretic models are used which are elasto-perfectly plastic, bilinear,
strength deterioration, stiffness degradation and pinching models. For a given target ductility ratio,
inelastic demand is evaluated by using nonlinear time-history analysis with recorded earthquake ground
motions for three types of soil condition. Used earthquake records were selected by calibration parameters
of PGA, PGV, EPA and EPV. To determine the ductility factor of rock site, regression analysis is used by
ductility factor and period characteristics of elasto- perfectly plastic response. By two-stage analysis, a
functional equation for the inelastic demand spectra was derived from ductility factor and hysteretic
characteristics. By the same procedure of rock site, functional equations for stiff and soft soil sites were
also derived. From the compatibility analysis and statistical studies of nonlinear time-history analysis,
developed method of the inelastic demand in capacity spectrum can be used to find the performance point
in seismic performance of structures.

ATTENUATION RELATIONS OF ACCELERATION RESPONSE
SPECTRA AT DAM FOUNDATIONS IN JAPAN
Norihisa MATSUMOTO , Dr. Eng.1, Takashi SASAKI2 , Kenji INAGAKI3, Tadashi ANNAKA4
SUMMARY
The evaluations of design earthquakes and characterization of earthquake ground motions are
important aspects of the seismic analysis and design of dams. Earthquake ground motions are affected by
earthquake source mechanisms, source-to-transmission path properties, and site conditions. There are
three basic approaches to developing site-specific earthquake motions: theoretical, semi-empirical and
empirical. Although the first two approaches, i.e. theoretical and semi-empirical approaches are rapidly
evolving, their results should still be examined based on the empirical relationships for design use. As for
the last approach, the attenuation relations for response spectra of ground motions developed by some
empirical ways contribute greatly to the rough estimation of possible ground motions.
The Ministry of Land, Infrastructure and Transport (former, the Ministry of Construction) has
conducted earthquake-motion observation at dam sites since 1957. With the Kobe Earthquake in 1995 as a
turning point, the Ministry decided to strengthen the seismograph installation at all of jurisdictional dams
for the intensification of the emergency management. The accumulation of earthquake records at dam
sites, that is very valuable for the evaluation of seismic motion at dam sites, have greatly advanced in this
decade.
We have developed attenuation relations on ground motions at dam sites by the statistical analysis
using the abovementioned records compilation[1]. The earthquakes used for the analysis have magnitude
larger than 5.0, epicentral distance less than 200km, and hypocentral depth less than 130km. All
accelerograms were recorded at the dams on rock foundations. They were recorded at 91 dam sites and for
63 earthquakes. The attenuation relations were developed for the three types of earthquakes, namely, the
shallow crustal, inter-plate, and deep intra-slab earthquakes.
In this paper, the influence of magnitude, distance measure, and the earthquake depth upon the
spectrum characteristis is analyzed. Acceleration time histories measured in the Tokachi-oki Earthquake
are compared with acceleration response spectrum presumed from attenuation relations .Based on theseattenuation relations, we discuss the characteristics of ground motions at rock foundations for existing
dams in Japan.

AN INVERSION METHOD TO DETERMINE STRESS DISTRIBUTION
ON A FAULT PLANE USING GROUND MOTIONS
Hisashi TANIYAMA1
SUMMARY
The rupture process of an earthquake has large effect on near-field ground motions. Because source-time
functions are affected by the stress drop process, it is important to understand the dynamic rupture process
of earthquake faults to estimate near-field ground motions. Kinematic inversion analyses have been
carried out to obtain spatiotemporal slip distributions, from which the stress distributions on the fault
plane have been inferred. In this study, an inversion method to infer stress drop process of an earthquake
fault is developed. To express a spatiotemporal stress distribution on an earthquake fault during an
earthquake, a fault is divided into subfaults and the shear stress on a subfault after the passage of the
rupture front is assumed to change repeatedly to its residual stress. Stress changes of subfaults are inferred
directly from the observed ground motions. Total rupture processes are composed of the stress change
histories of the subfaults. To examine the performance of the method, a simulation test was conducted. A
relatively simple model was used in the simulation and the computed synthetics were used as observed
seismograms. The inversion was stabilized by adding smoothing constraints and ABIC was employed to
objectively determine the optimal values of hyperparameters. The inferred time histories of stress changes
on subfaults fit generally well with the assumed rupture process.

NEURAL NETWORK ANALYSIS OF SEISMIC INTENSITY FROM
INSTRUMENTAL RECORDS
Peter DAVENPORT1
SUMMARY
Seismic intensity provides useful information on the distribution of earthquake effects. The concept of felt
intensity has been used for a long time and is considered a method to classify the severity of the ground
motion at a given location on the basis of effects observed either during the earthquake or afterwards. The
Modified Mercalli Intensity (MMI) scale is the most common felt intensity scale in use and is used to
indicate seismic hazard and level of damage.
With the more widespread use of strong motion recorders, it has become possible to obtain engineering
parameters such as peak ground acceleration, velocity and displacement, spectral values and other
measures of instrumental intensity. Many studies have compared these parameters, obtained from strong
motion records, to the seismic felt intensity measured by damage, but found the correlation is usually poor
and the relationships are highly nonlinear.
Artificial neural network (ANN) based methodology is not new, but has not been extensively applied to
engineering seismology problems. This technique essentially uses large quantities of data to train a model
which can then be used to explore the relationship. The ANN models allow complex and nonlinear
behaviour to be tracked. In this paper, strong motion records from New Zealand have been analyzed using
ANN methods to seek out the parameters, derived from strong motion records, which are important for an
understanding of seismic felt intensity as measured by observed effects and damage.

An integrated framework for seismic performance of reinforced concrete
building
Z.P Wen,1 MT Gao, XJ Li, FX Zhao, HS Lu
SUMMARY
A procedure is established to integrate seismic environment and site condition into the framework of
earthquake damage assessment. The model has four stages: i) probabilistic seismic hazard assessment at
rock site, ii) site response analysis, iii) the seismic response estimation for reinforced concrete building,
and iv) structural damage assessment. The equivalent linear analysis method for site seismic response
analysis and the equivalent lateral force method for structure seismic analysis are incorporated to
investigate the site effects on building fragility information. A 15-story shear wall-structure is selected as
an application to demonstrate the procedure. By comparing the damage distribution under the acceleration
response spectra with 63%, 10%, and 2% probability of exceeding in 50 years at rock site and specified
site, it can observed that the site effect is very important, then seismic environment and site condition
should be considered at same time when seismic performance is estimated

RESPONSE OF STEEL BURIED PIPELINES TO THREE-DIMENSIONAL
FAULT MOVEMENTS BY CONSIDERING MATERIAL AND
GEOMETRICAL NON-LINEARITIES
Hamzeh SHAKIB1 and Reza ZIA-TOHIDI2
SUMMARY
The purpose of this study is to evaluate the effects of three dimensional fault movements on the nonlinear
response of steel buried pipelines. The steel pipeline is divided into a number of segments, and each
segment is modeled as a beam element. The beam element has six degrees of freedom at each of the
nodes. The beam elements are considered to have material and geometrical non-linearities. The
surrounding soil is modeled by using nonlinear springs in the axial, lateral, and vertical directions. The
fault is considered to be an oblique-slip, which has three-dimensional movements. By using some of the
important system parameters the non-linear analysis of pipe-fault system is assessed. The parameters used
in this study are: anchored length of pipeline, buried depth of pipe, frictional angle of soil-pipe system,
geometrical characteristics of pipe, fault displacement, crossing angle of pipe with the fault, and the fault
slope angle. The effects of these parameters are investigated on the response of buried pipeline crossing
the oblique-fault movements. The detail parametric study reveals that the bearing capacity of the steel
buried pipelines, subjected to the oblique-slip fault movements can be improved, if the effects of the
above parameters are taken into consideration while designing the pipeline.