Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Fundamentals of The Effects of Earthquakes On The Ground and Structures

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

Paper No.

TS-3-2

FUNDAMENTALS OF THE EFFECTS OF EARTHQUAKES ON THE


GROUND AND STRUCTURES
Abel ORDOEZ1
SUMMARY
Important mining projects are being developed in our country, which require a structural design not only
for their buildings, process plants and various facilities, but also for their earth and mining waste
structures, where the concepts of performance-based seismic design of structures are being implemented.
The effects of an earthquake on the ground and on structures are based on the existing knowledge of
seismology, geotechnics, geophysics, soils dynamics and structural behavior. To the author, it is clear that
this is the current situation in many countries. The inertial model which aims at estimating seismic
forces, the basis of structural design, has been questioned in the past few years since the way it represents
the behavior of structures during an earthquake seems to be very limited, particularly when the structure
collapses on soft soil, where damages increase significantly causing sometimes large numbers of victims.
The work submitted herein is founded on the study of ground motion during an earthquake, considering
not only the acceleration but also the displacements generated by the earthquake. Two case studies are
presented based on experience acquired in the past years in the mining industry and alternative, different
structure failure mechanisms are proposed with a recommendation to review and complement the current
seismic design concepts.
INTRODUCTION
Performance of the foundation soil and soil structures under seismic loads has been discussed since 1987
by the International Society of Soils Mechanics and Geotechnical Engineering (ISSMGE) and its recent
contributions were summarized during the last International Conference of Performance-Based Design in
Geotechnical Earthquake Engineering, Kokusho [1].The knowledge of ground motion and deformations
induced by an earthquake is the most important stage of performance-based design, particularly for soft
soil. A quick review of the ground motion and structure performance during the earthquakes that occurred
in the world indicates that structures under a conventional seismic design are incapable of withstanding
large deformations, especially when founded on soft soil, figure 1 and 2. Performance-based design is
based on the criterion of the tolerable ground motion in accordance with the structures design criteria and
the application of appropriate methodologies to determine them.
Ground Motion Caused by an Earthquake
According to Newmark [2], the effect of an earthquake on structures, the following should be
considered:
a) All of the earthquakes characteristics, being the peak acceleration not necessarily the main factor of
seismic response.

National University of Engineering, Lima-Per Email: ohabel@yahoo.com

Figure 1, 2: Evidence of excessive displacement on very superficial concrete slab foundation (the design
was based on the inertial model and horizontal force of the Peruvian Technical Standard E0.30 for
Earthquake Resistant Design, and did not consider the effect of the horizontal and vertical displacements
during the earthquake). Earthquake 8.15.07 in Pisco, Peru.

b) The effect of the earthquakes velocity, displacement and duration may be of equal or greater
importance to determine the seismic response.
c) All of the earthquakes characteristics, being the peak acceleration not necessarily the main factor of
seismic response.
d) The effect of the earthquakes velocity, displacement and duration may be of equal or greater
importance to determine the seismic response.
Velocity (cinematic energy) is the most important variable to study motion as well as displacement
(potential energy). Both are indicators of the intensity of the earthquake and therefore of potential damage
to structures. During the occurrence of an earthquake and the resulting ground motion, there are moments
during which the velocity is zero and there is maximum displacement. The earthquakes intensity (energy)
should be assessed compared with its maximum velocity or maximum displacement (Fig. 3). From this
viewpoint, acceleration is not representative of the earthquakes intensity (energy) and can therefore not
necessarily be the basis for the seismic design of the structures.
It should be highlighted that large accelerations recorded in recent earthquakes did not necessarily result in
large structural or geotechnical damage. For instance, Kokusho [3] indicated that during the earthquakes
in San Fernando (1971) and Northridge (1994) in the USA, the peak ground accelerations (PGA) of
1.0G and 1.8G did not cause major structural damage in the surrounding areas. During the 2004
earthquake in Nigata, Japan, the 1.7G acceleration in Tokamachi did not cause the major damage as had
been anticipated either, and many similar cases have been recorded. This indicates that acceleration may
not be a parameter governing damage to structures, compared to other parameters such as displacement or
velocity, as Kokusho stresses [3].
The inertial model has its origin in the rigid body mechanics and is intended to estimate seismic forces in
the current design methods which are earthquake resistant; this is why it has been questioned in recent
years, Villarreal [5], as it is said to represent the behavior of structures during an earthquake in a very
limited way, particularly on soft soil.
The study of ground motion caused by an earthquake, where seismic waves experience the phenomenon
of amplification in passing from bedrock to less dense soil, indicates the following:
a) The energy which is generated by the earthquake and which propagates through seismic waves, does
so through elasto-plastic behavior.

b)
c)
d)

e)

The (non-recoverable) inelastic deformation component is the dissipated seismic energy, which is
called ground damping.
Softer, less dense soils dissipate larger amounts of energy since they suffer larger unrecoverable
deformations than hard soils.
The energy is transmitted more slowly in soft soils compared to hard soil. Superficial seismic waves
in soft soil, for example, can travel at velocities between 50 and 200 m/s, while in hard rock they
propagate at a velocity exceeding 2000 m/s, which is 10-40 times faster.
The impedance of the physical medium associated with the difference in density between soil and
bedrock transforms the bedrocks kinematic type energy into energy of the potentially soft soil
deforming type.

Seismic Ground Amplification


The analytic methods used to study seismic amplification of the site in order to rigorously evaluate the
effect of an earthquake consider the following stages:
1) Establishment of an appropriate time-history record of bedrock accelerations to model the dynamic
response.
2) Modeling of the unidimensional dynamic response using the results of the previous stage and
applying dynamic modeling using the SHAKE program, Schnabel [4].
Time-History Record of Accelerations in Bedrock
The records used resulted from the processing records from 17 earthquakes occurred in Peru (Table 1).
From the assessment performed, processed records of seismic behavior accelerations in the country were
obtained. Based on the assessment of the predominant period, 06 records were selected as representative
of the subduction and continental sources seismicity (Table 2).
Unidimensional Dynamic Response Modeling
In the unidimensional dynamic analysis model, it is assumed that earthquake accelerations will occur in
the bedrock, making up a unidimensional ground column. From the bedrock, horizontal shear waves
propagate vertically and reflect on the deposit surface. The dynamic properties of the materials that make
up the profile were estimated from the results of measurement tests of surface shear waves, Vs, Japanese
Society of Civil Engineers [5], Kramer [6] and Towhata [7]. Dynamic properties, such as shear modulus
and damping, were estimated from the existing technical literature (Hardin, 1972; Seed and Idriss, 1970;
Seed et al., 1984; Vucetic and Dobry, 1991).
Dynamic Parameters
The dynamic behavior was assessed unidimensionally through geophysical profiles, which have provided
average values for surface shear wave velocities Vs(30), in accordance with the IBC. Measurements of
velocity profiles for surface shear waves were made through multichannel analysis of surface waves
(MASW) Park et al., 1999; Xia et al. 1999; Miller et al., 1999.
Case 1: The effect of an earthquake on gravel soil
The soil is composed of gravel silty (GM, GW-GM, GP-GM and SM) with angular to sub-angular
particles and 4 - 7 diameter blocks. S.P.T. tests ranged from 10 to 35. Density varies from loose to
medium dense up to a depth greater than 50 meters. Four (04) geophysical profiles were evaluated
resulting in average surface shear wave velocity values Vs(30) of 340, 380, 400, 440 and 780 m/s
respectively. The seismic response analysis was carried out using the short period accelerations record
(0.1 to 1 sec.) and long period accelerations (up to 1 and 3 sec) of the Pisco earthquake (2007) and
Yurimaguas earthquake (2005), respectively; the accelerations record was scaled to 0.22 g. The results of
the seismic amplification analysis at ground surface level are shown in Tables 3 and 4.

Amax=200cm /s (T=0.2-0.5seg)
Registro de Aceleracin - Sismo de Pisco 100-110 seg
0.25
0.20

Acceleration (g)

0.15
0.10
0.05
0.00
-0.05
-0.10
-0.15
-0.20
-0.25
100

101

102

Vmax=30cm/s (T=1-1.5seg)

103

104

105

106

107

108

109

110

107

108

109

110

107

108

109

110

Time (sec)
Registro de Velocidad - Sismo de Pisco 100-110 seg

40
30

Velocity (cm/sec)

20
10
0
-10
-20
-30
-40
100

101

102

103

104

105

106

Time (sec)

Dmax=15cm (T=2-4seg)

Registro de Desplazamiento - Sismo de Pisco 100-110 seg


15

Displacement (cm)

10

-5

-10

-15
100

101

102

103

104

105

106

Time (sec)

Figure 3: Acceleration, velocity and maximum displacement (by integration in surface soft soil), during
the 10 most intense seconds (100-110) of the Earthquake 8.15.07 in Pisco.

Station

Table 1: Calibrated and Evaluated Earthquake Records


Earthquake
Date
Magnitude
Strike

U.N.S.A. Arequipa
Tarapac
Characato-Arequipa
Tarapac
Characato-Arequipa
Pisco
Callao
Pisco
C.D. Lima-CIP
Pisco
CISMID-UNI
Yurimaguas
CISMID-UNI
Pisco
La Molina-Lima
Yurimaguas
La Molina-Lima
Pisco
G.R. Moquegua
Arequipa
G.R. Moquegua
Tarapac
Moquegua
Tarapac
Moquegua
Tarapac
Moyobamba-San Martn Yurimaguas
U.N. Jorge Basadre-Tacna Arequipa
U.N. Jorge Basadre-Tacna Tarapac
U. Privada Tacna
Tarapac

Jun-13-05
ML=7.2, Mw=7.8
Jun-13-05
ML=7.2, Mw=7.8
Aug-15-07 ML=7.0, Ms=7.9, Mw= 8
Aug-15-07 ML=7.0, Ms=7.9, Mw= 8
Aug-15-07 ML=7.0, Ms=7.9, Mw=8
Sep-25-05
ML=7
Aug-15-07 ML=7.0, Ms=7.9, Mw=8
Sep-25-05
ML=7
Aug-15-07 ML=7.0, Ms=7.9, Mw=8
Jun-23-01
Mb=6.9, Mw=8.3
Jun-13-05
ML=7.2, Mw=7.8
Jun-13-05
ML=7.2, Mw=7.8
Jun-13-05
ML=7.2, Mw=7.8
Sep-25-05
ML=7
Jun-23-01
Jun-13-05
ML=7.2, Mw=7.8
Jun-13-05
ML=7.2, Mw=7.8

E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W
E-W

Table 2: Representative Earthquake Records


Earthquake (recorded)
acceleration
displacement
recorded
calculated
g
Tarapac 2005 (Arequipa)
0.13
3.0
Pisco 2007 (Lima)
0.05
2.5
Arequipa 2001 (Moquegua)
0.30
5.0
Tarapac 2005 (Moquegua)
0.05
0.3
Yurimaguas 2005 (Moyobamba)
0.14
7.0
Tarapac 2005 (Tacna)
0.12
2.8
Table 3: Case 1: Acceleration Amplification Factors
Profile
Vs(30)
Pisco (2007) Yurimaguas (2005)
1
340
1.5
1.6
2
380
2.1
3.5
3
400
2.1
3.1
4
440
1.8
2.1
5
780
2.0
1.5
Table 4: Case 1: Displacement Amplification Factors
Profile
Vs(30)
Pisco (2007) Yurimaguas (2005)
1
340
1.1
1.3
2
380
1.1
1.2
3
400
1.0
1.2
4
440
1.1
1.2
5
780
1.0
1.1

Dist.
Hypoc.
422
414
520
200
200
706
200
713
200
340
324
323
320
146
220
231
232

S. Fourier
period
<0.1
0.3-1.5
0.4-1.0
0.1
0.3-4.0
0.2-0.5

Case 2: The effect of an earthquake on sandy soil


The soil is composed of sand and silt (SM, ML, SP-SM), of recent formation. S.P.T. tests ranged from 4
to 30. Density ranges from loose to medium dense to a depth greater than 30 meters; there is no ground
water table. Three (03) geophysical profiles were assessed which resulted in average surface shear wave
velocity values Vs(30) of 200, 240 and 270 m/s, respectively. The seismic response analysis was carried
out using the accelerations record of the earthquakes in Pisco and Yurimaguas; the acceleration record
was scaled at 0.40 g. Tables 5 and 6 shown he results of the seismic amplification analysis.
Table 5: Case 2: Acceleration Amplification Factors
Profile
Vs(30)
Pisco (2007) Yurimaguas (2005)
6
200
1.1
1.1
7
240
1.4
1.6
8
270
1.8
1.7
Table 6: Case 2: Displacement Amplification Factors
Profile
Vs(30)
Pisco (2007) Yurimaguas (2005)
6
200
1.3
2.0
7
240
1.3
2.2
8
270
1.2
2.0
The real case study of seismic wave propagation in the ground shows the following:
a) The predominant vibration or seismic motion period of the ground is associated with the soil density
and rigidity.
b) The predominant vibration period of the bedrock ranges from 0.1 to 1 second. The predominant
vibration period in soft soil may reach values from 1 to 3 seconds.
c) Energy damping or dissipation is larger in soft soils.
d) Degradation or reduction of rigidity in soft soil is very significant, reaching up to 10-20% of the
initial rigidity for 1% deformation.
e) Strong, short-period earthquakes (between 0.1 and 1 second) in hard soil cause accelerations above
1.0G, associated with the acceleration amplification phenomenon. Strong, long-period earthquakes
(between 1 and 3 seconds) in soft soil cause deformations exceeding 1%, generating loss of rigidity,
loss of strength and excessive increase of displacements, from 10 to 25 cm, associated with the
displacement amplification phenomenon due to the seismic resonance.
f) The inertial model cannot be applied to structure behavior in soft soil. Instead, ground displacement
should be considered during an earthquake, or a combination of both. In the proposed kinematic
model, apart from the inertial model, other failure mechanisms should be evaluated, where
horizontal shear stresses are not necessarily dominant.
Seismic Amplification Study in Lima
According to Predes, an earthquake with a magnitude of Mw 8.0, similar to the one that occurred in
Pisco, with an acceleration of 0.35-0.40g, is expected to occur in Lima in the next years. In Table 7,
CISMID-UNI [8], proposes the following seismic amplification factors for accelerations, to be applied in
accordance with the earthquake resistant standard based on the inertial approach. In 2009, WAPMEER
(World Agency for Planetary Monitoring and Earthquake Risk Reduction) made preliminary estimates
of losses due to possible earthquakes in Lima, and concluded that if 50% of the people are located
inside buildings at the time of an earthquake, between 7,000 and 30,000 people will die. In the worst
case scenario, which is at night, with an 80% occupation of buildings, the number of fatalities would range
from 10,000 to 50,000, Predes [9].

Table 7: Seismic Amplification in Lima - CISMID-UNI (2003)


Seismic Amplification
Soil Type
Factor
Type I: Hard soil, old conglomerate, dense and consolidated.
1.0
Type II: Intermediate soil, La Perla, San Miguel, La Molina,
1.2
Los Olivos, San Juan de Miraflores, Surco. Sandy and silty
sediments of limited thickness, less than 10 meters.
Type III: Soft soil, San Juan de Miraflores, Chorrillos, Callao,
1.4
Villa El Salvador, Magdalena. Sandy and silty soil with a
thickness greater than 10 meters.
Type IV: Very soft soil, Ventanilla, Chorrillos, San Juan de
1.4
Miraflores, Villa El Salvador. Coastal sand, loose sand with
presence of a ground water table.
For the aforementioned estimates, a magnitude 8 earthquake was used. An amplification factor for strong
ground motion was considered based on the CISMID-UNI microzoning map. The results of these
studies suggest in the near future to review, complement the seismic microzoning studies, the earthquake
resistant methodologies, the Technical Standard for Soils and Foundations and for Earthquake Resistant
Design.
Failure Mechanism and Solutions Associated with the cinematic Model
In the cinematic model, based on the movement induced by an earthquake in the structures, not only
horizontal, but also gravitational (vertical) forces are produced because it is made possible by the ground
motion. Moreover, it has been determined that horizontal acceleration in soft soil is controlled (decreased)
by the impedance of the phenomenon of seismic wave propagation. The failure mechanism in soft soil is
associated with the gravitational forces that generate shear strengths in the vertical plane of any structure
(Figure 4,5,6), associated with the vertical and slow undulating motion of the ground (for a large
predominantly long period). It is estimated that the vertical shear strengths will reach magnitudes which
are much greater than those currently estimated, based on the inertial model; therefore, there is still a
large amount of damage current engineering cannot control, unless special structures are used as in the
case of the Latin American Tower located in the soft zone of Mexico City. This is a 44 story building
founded on piles, which was built in 1956. It withstood the M=7.7 earthquake on July 28, 1957; the
M=8.1 earthquake on September 19, 1985 and the M=7.9 earthquake on March 20, 2009. It is estimated
that shear strengths in conventional structures are of such magnitude that other alternatives will be
searched in order to reduce not only shear strengths through light and flexible structures, but also to
reduce the structure motion by replacing or reinforcing the foundation soil using non-conventional
materials such as geosynthetics, Tatsuoka [10] and The Japanese Geotechnical Society [11] or
considering deeper foundations.

Figure 4: Damage caused by vertical displacements and excessive differentials in the ground during an
earthquake.

Figure 5: Structure failure caused by vertical shear stress concentration.

Figure 6: Failure caused by turn over in structure founded on a surface concrete slab.
ACKNOWLEDGEMENTS
To Knight Piesold for to participate and to develop these studies in their most important mining projects,
and to OM Ingeniera y Laboratorio for their experience in performing MASW geophysical testing.
REFERENCES
Kokusho T., Tsukamoto Y., Yoshimine M. Performance-Based Design in Earthquake Geotechnical
Engineering from Case History to Practice. CRC Press, 2009.
2. Newmark, N.M. Effects of the earthquakes on dams and embankments. Geotechnique, Vol.15,
No. 2, page 139-160, 1965.
3. Kokusho T. PDB in Earthquake Geotechinical Engineering and Energy-Based Design; Proceedings
of the International Conference on Performance-Based Design in Earthquake Geotechnical
Engineering, 2009.
4. Schnabel P. B., Lysmer J., Seed H. B. SHAKE: a computer program for earthquake response
analysis of horizontally layered sites, University of California, 1972.
5. Japanese Society of Civil Engineers, Dynamic Analysis and Earthquake Resistant Design, Volume
3, Dams, Nuclear Power Plants, Electrical Transformers and Transmission Lines, Above ground
Storage Tanks and Piping. A.A. Balkema, 2001.
6. Kramer S.L. Geotechnical Earthquake Engineering, Prentice Hall, 1996.
7. Towhata I. Geotechnical Earthquake Engineering, Springer, 2008.
8. CISMID-UNI. Estudio de Vulnerabilidad y Riesgo Ssmico en 42 distritos de Lima y Callao,
CISMID/APESEGEVRLYC, 2004.
9. PREDES Diseo de escenario sobre el gran impacto de un sismo de gran magnitud en Lima
Metropolitana y Callao, Per, 2009.
10. Tatsuoka F. Recente practice and research of geosynthetic-reinforced earth structures in Japan.
Tokyo University of Science, Japan, 2011.
11. The Japanese Geotechnical Society. Geo-hazards during earthquakes and Mitigation measures.
July 2011.
1.

You might also like