Soil Dynamics and Earthquake Engineering 48 (2013) 252–266

Contents lists available at SciVerse ScienceDirect

Soil Dynamics and Earthquake Engineering

journal homepage: www.elsevier.com/locate/soildyn

Seismic microzonation for the northeast Texcoco lake area, Mexico

Luis Osorio, Juan M. Mayoral n
Instituto de Ingenierı́a, Universidad Nacional Autónoma de México, Mexico City, Mexico

a r t i c l e i n f o a b s t r a c t

Article history: The Texcoco Lake area is currently undergoing a rapid urbanization. This has required designing and
Received 22 October 2012 building strategic infrastructure. However, this zone has been poorly explored, and thus, a building
Received in revised form code design response spectrum is not available. This paper presents the exploration and laboratory
19 January 2013
works, as well as the analytical studies conducted for obtaining a design response spectrum for the
Accepted 21 January 2013
northeast Texcoco Lake area. From geotechnical information gathered previously for other projects in
the studied zone, and applying the ordinary kriging technique, the necessary dynamic soil parameters
for site response analysis were determined. A total of 2501 virtual soil profiles were generated.
The seismic environment was characterized throughout a uniform hazard spectrum obtained in a
nearby rock outcrop. Probabilistic site response analyses were performed using the uniform hazard
spectrum computed in rock as input motion for all 2501 virtual profiles. Finally, a design response
spectrum for the studied area was proposed.
& 2013 Elsevier Ltd. All rights reserved.

1. Introduction
Driven by the extensive damage observed during the 1985
Michoacan earthquake, several studies were launched to reduce
the seismic risk in Mexico City valley, which included dynamic
soil property determination (e.g. [1–5]), global seismicity evalua-
tions (e.g. [6–9]), and amplification studies (e.g. [10,2,11–13,3]).
However, most of these studies have been focused on downtown
Mexico City, where the damage was more notorious. Even con-
sidering the large amount of information gathered from this
effort, the so-called virgin Texcoco Lake, towards the north east
of downtown, is still limitedly explored, as stated in the Mexico
City building code [14], Fig. 1. Therefore, seismic parameters are
not available for engineering design in this zone. This paper
presents the exploration and laboratory works, and analytical
analyses carried out for obtaining a design response spectrum for
a 150 km2 area located at the northeast Texcoco Lake region.
Based on available geotechnical information and geo-statistical
techniques (e.g. [15–17]), the parameters that affect the dynamic
response of a soil deposit, such as the shear wave velocity,
thickness layer, and normalized modulus degradation and damp-
ing curves were characterized in a grid of points spaced 250 m, to
developed 2501 virtual soil profiles. The seismic environment was
characterized in terms of a uniform hazard spectrum derived for a
nearby rock outcrop, UHSrock. Finally, throughout 2501 probabil-
istic site response analyses were performed using the UHSrock as
input motion, a suit of response spectra were developed for the
area, which, in turn, were nicely enveloped by a proposed design
spectrum for the studied area.
2. Description of the studied area
The studied area is located at a side of the former Texcoco
Lake, commonly the soil profile at this zone presents a desiccated
crust of clay at the top extending up to a depth of 1.0 m, which is
underlain by a soft clay layer approximately 25.0 m thick, with
interbedded lenses of sandy silts and silty sands. The plasticity
index varied from 139 to 265%. Underlying the clay there is a
4.0 m thick layer of very dense sandy silt, which rests on top of a
stiff clay layer which goes up to a 60.0 m of depth. The plasticity
index varied from approximately 59 to 106%. Underneath this
elevation a competent layer of very dense sandy silt is found.
Fig. 1 shows the location of the studied area, whose length and
width are 15.00 km  10.00 km, respectively, and has a total area
of 150.00 km2. The distance from the National University of
Mexico, UNAM, to the polygon center that encloses the studied
area is approximately 31.70 km. The region studied is instrumen-
ted with four seismic stations, TXSO, TXS1, TXS2 and TXCH. A fifth
station used in the analysis, TXRC, is located to the east, on a
rocky outcrop, about 18.70 km away from the studied site.
3. Available data

n
Corresponding author. Tel.: þ525 55 5622 35x8469; fax: þ 510 642 7476. Data obtained from different projects developed in the area,
E-mail address: jmayoralv@iingen.unam.mx (J.M. Mayoral). including a power plant, an electric network, and a new airport,

0267-7261/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.soildyn.2013.01.013
 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266 253

19.60

LATITUDE
TXSO TRANSITION
LAKE ZONE
ZONE
"CARACOL" HILLS

15.00 km
TEXCOCO ZONE
19.55 STUDIED
AREA TXCR

19.50 18.70 km
X.
- TE
KE ÉX.
LA OP. M
O T
OC AU
XC
TE

O
19.45

OC
10.000 km

XC
INTERNATIONAL

TE
AIRPORT TXCH

S-
YE
RE
19.40
-99.10 -99.05 -99.00 -98.95 -98.90 -98.85 -98.80 -98.75
LONGITUDE

Zone I Zone IIIb This zone will be considered as II (transition)

when using the complementary Technical
Zone II Zone IIIc Norms for foundation design
This region are no sufficiently investigated
Zone IIIa Zone IIId therefore the proposed zoning is only indicative
Scale
0 1 2.5 5 10 15 20 km

Fig. 1. Location of the area of study with respect to the seismic zoning proposed by the Mexico City building code.

Fig. 2. Location of explored sites and virtual soil profiles.

combined with site specific exploration, were used as input for
the geo-estatistical analyses. A total of 28 Standard Penetration
Tests, SPT, combined with selected undisturbed sampling recov-
ery, and 67 Cone Penetration Tests, CPT, were available. The
locations of the existing borings are presented in Fig. 2, and
Table 1 shows their corresponding UTM and geographic coordi-
nates. In addition, four piezocone tests, PZC, were carried out to
characterize the in situ pore water pressure distribution, and four
PS suspension loggings were used to determine in situ values of
shear wave velocity. The information gathered allowed identify-
ing the general layering sequence and thicknesses, shear wave
velocities distribution with depth at some points, and index and
mechanical soil properties. Estimations of shear wave velocities
for clays and silts were carried out through the application of the
expression proposed by Ovando and Romo [2] in terms of the tip
cone penetration resistance, qc, whereas the relationship pro-
posed by Seed et al. [18] was used for sands, based on the SPT
blow counts, as described in Mayoral et al. [5]. Thus, 69 shear
254 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266

Table 1
UTM and geographic coordinates of the sites with available data.

Project Boring Coordinates

UTM Geographic

X Y Longitude Latitude

ELECTRIC NETWORK LINE 400 AND 230 kV L400-T12-SPT 504,817.944 2,168,612.44  98.954053 19.6125932
L400-T13-CPT 504,763.715 2,168,310.64  98.9545709 19.6098659
L400-T15-CPT 503,955.022 2,168,279.92  98.9622831 19.6095901
L400-T18-CPT 503,290.581 2,168,272.59  98.9686195 19.609525
L400-T19-CPT 503,071.455 2,167,887.14  98.9707098 19.606042
L400-T20-CPT 502,843.298 2,167,972.3  98.9728855 19.6068119
L400-T22-CPT 502,665.594 2,167,495.36  98.9745808 19.602502
L400-T23-SM 502,522.902 2,167,112.39  98.975942 19.5990413
L400-T25-CPT 502,245.524 2,166,312.28  98.978588 19.591811
L400-T27-CPT 501,923.168 2,165,560.27  98.9816626 19.5850153
L400-T29-CPT 501,319.294 2,165,628.89  98.9874205 19.5856359
L400-T31-CPT 501,252.734 2,165,027.63  98.9880555 19.5802023
L400-T32-CPT 501,009.331 2,165,024.73  98.9903763 19.5801762
L400-T35-SM 500,228.632 2,165,467.52  98.99782 19.584178
L400-T39-CPT 499,064.618 2,165,375.08  99.0089188 19.5833425
L400-T42-CPT 498,262.264 2,164,705.45  99.0165685 19.5772904
L230-P1-CPT 498,165.272 2,164,511.52  99.0174931 19.5755377
L230-T6-CPT 498,601.701 2,163,426.51  99.0133312 19.5657328
L230-T14-CPT 498,652.811 2,161,415.25  99.0128425 19.5475567
L230-T23-CPT 498,648.369 2,159,139.9  99.0128832 19.5269941
L230-T24-CPT 498,547.46 2,159,036.87  99.013845 19.5260629
L230-T33-CPT 498,533.284 2,156,741.62  99.0139783 19.5053203
L230-T41-CPT 498,492.279 2,154,787.1  99.0143675 19.4876569
L230-T42-CPT 498,281.089 2,154,623.46  99.0163799 19.4861779
L230-T46-CPT1 497,241.512 2,154,470.35  99.026286 19.484793
L230-T47-CPT 497,077.68 2,154,357.73  99.027847 19.483775
L230-T51-CPT-1 496,381.013 2,154,686.35  99.0344862 19.4867436
L230-P55-CPT2 495,452.172 2,155,056.73  99.0433383 19.490089
L230-T60-CPT2 494,543.825 2,155,495.77  99.0519956 19.4940544
NEW AIRPORT SCE-1 500,557.705 2,157,077.57  98.9946848 19.5083568
SCE-2 502,482.281 2,156,944.57  98.9763428 19.5071533
SCE-3 503,976.899 2,156,878.07  98.9620985 19.5065499
SCE-4 505,491.991 2,156,791.1  98.9476594 19.5057604
SCE-5 500,921.106 2,158,154.38  98.9912209 19.5180879
SCE-6 505,865.629 2,157,878.14  98.9440951 19.5155831
SCE-7 501,310.116 2,159,254.21  98.9875124 19.5280271
SCE-8 503,244.929 2,159,172.36  98.9690706 19.5272852
SCE-9 504,749.784 2,159,075.17  98.9547271 19.5264038
SCE-10 506,259.757 2,158,967.74  98.940335 19.5254288
SCE-11 501,699.126 2,160,365.75  98.9838035 19.5380719
SCE-12 503,613.465 2,160,273.67  98.9655557 19.5372372
SCE-13 506,633.412 2,160,074.16  98.9367697 19.5354265
SCE-14 502,072.78 2,161,414.43  98.9802406 19.5475486
SCE-15 503,024.831 2,161,358.16  98.9711649 19.5470389
SCE-16 503,971.764 2,161,312.12  98.9621381 19.5466212
SCE-19 502,886.63 2,158,077.64  98.9724874 19.5173926
SCE-22 503,505.975 2,161,327.46  98.9665784 19.5467607
SCE-28 504,493.856 2,159,310.48  98.9571659 19.528531
SCE-31 502,573.951 2,155,168.53  98.9754715 19.4911028
SCE-32 505,635.31 2,155,650.35  98.9462969 19.4954508
SCE-33 508,081.069 2,162,407.95  98.9229605 19.5565119
SCE-34 505,558.515 2,162,447.76  98.9470087 19.5568803
SCE-35 503,306.352 2,162,442.64  98.9684794 19.5568391
SCE-36 501,985.765 2,162,427.29  98.981069 19.5567022
SCE-37 505,926.603 2,154,943.45  98.9435231 19.4890616
POWER PLANT CPT-1 498,218.696 2,164,609.7  99.0169839 19.576425
CPT-2 498,264.552 2,164,630.81  99.0165467 19.5766158
CPT-3 498,227.526 2,164,559.99  99.0168996 19.5759758
CPT-4 498,172.84 2,164,588.58  99.017421 19.5762341
CPT-5 498,209.867 2,164,659.4  99.0170681 19.5768741
SCE-6 498,191.354 2,164,623.98  99.0172446 19.5765541
SCE-7 498,232.522 2,164,636.14  99.0168521 19.576664
SCE-8 498,237.074 2,164,600.08  99.0168086 19.5763381
SCE-9 498,204.87 2,164,583.25  99.0171157 19.5761859
SCE-10 498,318.673 2,164,611.44  99.0160306 19.5764408
ZONATION TXSO 498,037.161 2,164,894.98  99.018715 19.579003
TXS1 502,669.995 2,155,490.15  98.9745558 19.4940093
TXS2 502,683.001 2,155,067.71  98.9744325 19.4901916
TXCH 505,253.726 2,148,537.45  98.949953 19.431171
 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266 255

wave velocity profiles were generated. In addition, series of
resonant column and triaxial tests were carried out in twin
samples to determine modulus degradation and damping curves,
in specimens recovered at the studied site, to develop normalized
modulus degradation and damping curves from small to large
strains. Each sample was tested with three different confining
pressures, ranging from 1.5 to 2 times the in situ effective vertical
stress, to reproduce, in a practical manner, the field conditions
prevailing at the site, and other two loading scenarios.
and CPT tests, and is in good agreement with observations made
by other researchers (e.g. [22–24]). The interpolation mesh was
already depicted in Fig. 2. To obtain a continuous shear wave
velocity profile from the surface to a depth of 50 m, interpolations
were carried out at each meter, in each virtual soil boring. Fig. 5a
depicts the shear wave velocity distribution at 1 m of depth,
and Fig. 5b shows the shear wave velocity profiles along soil
column 1221.
4.3. Modulus degradation and damping curves

4. Geostatistical determination of soil parameters
A mesh of 41 columns and 61 rows was built to establish the
location of the virtual soil profiles. Fig. 2 shows the location of the
sites were geotechnical information is available (dots), and the
interpolation points (circles) selected for the geostatistical ana-
lyses. This array leads to 2501 virtual soil profiles, in addition to
the layering sequence and thickness it was necessary to infer with
the ordinary kriging technique [19], the shear wave velocity
distribution with depth and modulus degradation (G/Gmax), and
(l–g) damping curves for each material, which were characterized
throughout the Darendeli and Stokoe’s model [20], as a function
of the plasticity index and other soil parameters, as will be
described later.
4.3.1. Darendeli and Stokoe model (2001)
Modulus degradation and damping curves can be obtained
through out laboratory testing in undisturbed soil samples, or
throughout empirical models function of soil type and other
variables. Darendeli and Stokoe [20], developed an empirical
model from that proposed by Hardin and Drnevich [25], to
generate modulus degradation and damping curves, which take
into account confining pressure effects, s0 0, plasticity index, PI,
overconsolidation ratio, OCR, the frequency of loading, f, and the
number of loading cycles, N. This model is given by the following
relationships:
G 1
¼  a ð1Þ
Gmax 1 þ gg
r

4.1. Layering sequence and thickness gr ¼ ðf1 þ f2 PIðOCRÞf3 Þs0f4 ð2Þ

 0:1
Based on the information regarding layering sequence, includ- G
l ¼ lmin þb lmsg ð3Þ
ing soil type and thickness gathered at each exploration boring, Gmax
and using the ordinary kriging technique implemented in the
program GSþ Gama Design Software, V.9, 2008, a total of 2501 lmsg ¼ c1 þ lmsg,a ¼ 1:0 þ c2 lmsg,a ¼ 1:0 2 þ c3 lmsg,a ¼ 1:0 3 ð4Þ
virtual soil profiles were constructed. The soil thicknesses were 2   3
g þ gr
established considering a linear distribution among interpolation 100 4 ggr ln gr
points, accounting for the general geological setting of the lmsg,a ¼ 1:0 ¼ 4 25 ð5Þ
p g2
g þ gr
Texcoco Lake region [21]. In particular, Fig. 3 shows the layer
thickness distribution of stratum number five, which corresponds where a ¼0.9190 (curvature coefficient), b¼0.620 (scaling
to clay. coefficient), g is the shear strain (%), gr is a reference shear strain,
s0 ¼ effective confining stress (atm), lmin is damping ratio at small
4.2. Virtual shear wave velocity profiles deformations, lmsg is the massing damping, and lmsg,a ¼ 1.0 is the
damping for a ¼1.0. The model coefficients are given by
Applying again the ordinary kriging technique, 2501 virtual
shear wave velocity profiles were developed considering a log-
normal distribution for this parameter. This assumption was c1 ¼ 1.1143na2 þ1.8618na þ0.2523
corroborated by the probability density functions computed base c2 ¼0.0805na2  0.0710na  0.0095
on shear wave velocities, presented in Fig. 4, derived from the SPT c3 ¼ 0.0005na2 þ0.0002na þ0.0003

Fig. 3. Thickness distribution prediction of the fifth layer (clay) using the ordinary kriging technique.
256 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266

Fig. 4. Probability density functions of estimated shear wave velocities, Vs.

Shear wave velocity, Vs (m/s)

0 200 400 600 800 1000
0

10

20
Vs at 1 m
Depth (m)

30

40

50

60

Tridimensional Model Bidimensional Model

Fig. 5. Prediction of shear wave velocity distribution using ordinary kriging technique at 1 m of depth (a) and shear wave velocity profiles of column 1221 (b).

f1 ¼0.0352, f2 ¼ 0.001, f3 ¼ 0.3246, f4 ¼0.3483, f6 ¼0.801,
f7 ¼0.0129, f8 ¼  0.107,
f9 ¼ 0.289
For this research work the curves proposed by Darendeli and
Stokoe were deemed appropriated because they can be used for
sands and clays, taking into account explicitly the most important
factors that can influence the dynamic soil behavior. To obtain the
modulus degradation and damping curves, the over consolidation
ratio, OCR, was taken as one, considering that the studied zone is
located in the virgin former Texcoco lake, and that the over-
consolidation of the soil for desiccation occurred only in the first
 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266 257

couple of meter. Thus, changes in the seismic response of the soil Darendeli’s model follow closely the experimental G/Gmax–g
deposit due to this fact are expected to be negligible. In this curve. The Darendeli and Stokoe [20] model used a fix value of
manner, the definition of the modulus degradation and damping the reference shear strain, gr, corresponding to 50% of the
curves results in only a function of plasticity index, PI. As an degradation of the stiffness model. Regarding the damping curve
example, Fig. 6 shows the modulus degradation and damping l–g, although the general data trend is captured by the model,
curves, for a sample of clay recovered at 51.90 m of depth this consistently over predicts the measured response for shear
and PI of 51%, retrieved from site TXS1. It can be seen that strains larger than 0.1%. Thus, spectral accelerations computed

1 20
Site: TXS1
Depth: 51.90 m
PI=51 %

Measured
max

0.8
15 Darendeli and Stokoe
Normalized shear modulus, G/G

model (2001)
Romo model (1995)

Damping ratio, λ (%)

0.6

10

0.4
Site: TXS1
Depth: 51.90 m
PI=51 % 5
0.2
Measured
Darendeli and Stokoe
model (2001)
Romo model (1995)
0 0
-6 -5 -6 -5
10 10 0.0001 0.001 0.01 0.1 1 10 10 10 0.0001 0.001 0.01 0.1 1 10
Shear strain, γ (%) Shear strain, γ (%)

Fig. 6. Comparison of the G/Gmax–g and l–g curves obtained with the Darendeli and Stokoe [20] and Romo [3], with the experimental data.

Fig. 7. Probability density functions of measured plasticity indexes.

258 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266

PI at 2 m

Tridimensional Model Bidimensional Model

Fig. 8. Plasticity index distribution at 2 m of depth determined using the ordinary kriging interpolation technique.

Shear wave velocity, VS (m/s)

Shear wave velocity, VS (m/s)
0 200 400 600 800 1000
0 0 200 400 600 800 1000
0
Suspension logging
Suspension logging
P-55 (Measured)
T-42 (Measured)
Virtual point PV7
Virtual point PV310
(Estimated)
(Estimated)

10
10

20 20
Depth (m)

Depth (m)

30 30

40 40

50 50

60 60

Fig. 9. Comparison of measured Vs at exporation boring P-55, and estimated Fig. 10. Comparison of measured Vs at exporation boring T-42, and estimated
values at virtual profile PV7. values at virtual profile PV310.
 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266 259

above this strain may be slightly underpredicted. However this
falls outside the range of shear strains computed in clays during
the site response analyses in most cases. In this figure were also
included the curves determined by Romo’s [3] model, which is
commonly used in Mexico City, and which lead to a similar soil
behavior.
parameter was estimated in the virtual soil profiles. Again, the
ordinary kriging interpolation technique was used to infer the
values of plasticity index in each virtual point, assuming a normal
probability distribution. This assumption agrees with the prob-
ability density functions obtained from measured data, as shown
in Fig. 7. The spatial distribution of the plasticity index at 2 m of
depth is depicted in Fig. 8.

4.3.2. Virtual profiles of plasticity index

To determine the modulus degradation and damping curves
that define the dynamic behavior of the geomaterials found at the
site, it is necessary to obtain the plasticity index, as established
previously. The plasticity index is a parameter that strongly
affects the dynamic properties of clayey soils [3,26]. In this sense,
it can be seen that the linear behavior of the soils found in the
studied zone increases when the plasticity index increases, lead-
ing to small degradation of the shear modulus even for strains as
large as 10%. From the information gathered in the laboratory
regarding plasticity indices, and geo-statistical analyses, this
Shear wave velocity, VS (m/s)
0 200 400 600 800 1000
0 can earthquake, and compare both responses. As can be seen in
Suspension logging
SEL1 (Measured)
Virtual point PV350
(Estimated)
4.3.3. Assessment of geostatistical model
The geostatistical model that lead to the virtual soil profiles
was validated in two stages. Initially, the appropriateness of
predicted shear wave velocity profiles was tested comparing the
in situ measured response at points P-55, T-420 and SEL1, which
of course were not taken into account during the kriging inter-
polation procedure, with the closest virtual shear wave profiles.
The comparisons are presented in Figs. 9–11. As can be noticed
the predicted values are in good agreement with the measured
data. To assess the validity of the rest of the parameters, an
indirect approach was followed, which consisted on computing
the seismic response of the soil deposit using the uniform hazard
response spectrum derived for a nearby rock outcrop for a
moment magnitude Mw ¼8.2, as described in the following sec-
tions, at the closest virtual soil profile to seismic station TXSO,
which recorded the ground movement during the 1985 Michoa-
Fig. 12, the computed response agrees well, from a practical
standpoint, to the measured response, considering that the 1985
Michoacan earthquake had a moment magnitude of Mw ¼ 8.1.

10 5. Seismic environment

The seismic environment at the studied site was characterized

through a probabilistic seismic hazard analysis, PSHA. In general, the
steps needed to conduct a probabilistic seismic hazard analysis
(e.g. [27–30]) are: (1) identification of all earthquake sources capable
of producing damaging ground motions, (2) characterization of the
20
earthquake recurrence model, (3) definition of the attenuation
relationship, (4) calculation of seismic risk and uniform hazard
response spectrum, and (5) computation of deaggregation of the
probabilistic seismic risk. Fig. 13 shows schematically these steps.
Depth (m)

30
0.6
Mean measured
Envelope measured
NS component
0.5 EW component
Spectral acceleration, Sa (g)

PV350 Estimated
(Mw=8.2)
40 0.4
5 % damping

0.3

0.2
50

0.1

0
0 1 2 3 4 5
60 Period (s)
Fig. 11. Comparison of measured Vs at exporation boring SEL1, and estimated Fig. 12. Response spectrum computed at virtual soil profile 350 and measured
values at virtual profile PV350. responses at TXSO station.
260 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266

Step 4

Seismic hazard curves for diferent periods (T)

Exceedance probability,λ (Sa)

Period (TA) Period (TB) Period (TC) Period (TD) Period (TE)

Return Return
period period

Sa (TA) Sa (TB) Sa (TC) Sa (TD) Sa (TE)

Spectral acceleration, Sa

Uniform hazard spectrum (UHS)

for a return period given reference
Spectral acceleration, Sa

Sa (TC)
Sa (TB)
Sa (TD)

Sa (TA)
Sa (TE)

(TA) (TB) (TC) (TD) (TE)

Period (s)

Fig. 13. Steps required for performing a probabilistic seismic hazard analysis (modified from Klügel [29]).
 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266 261

5.1. Identification of earthquake zones capable of producing Table 3

damaging ground motions Exceedance rate parameters for zones of occurrence of characteristics earthquakes
with M 47.0 (after Ordaz and Reyes [32]).

All seismogenic zones considered in this study were estab- Zone Name Exceedence rate,
lished according to the zonation that Nishenko and Singh [31] l(7) (1/year)
carried out for the mexican subduction zone (Fig. 14) because
these corresponds to where the more damaging earthquakes for 1 Chiapas 0.0369
2 Tehuantepec Gap 0.03344
Mexico City have been generated. Tables 2 and 3 include the 3 Oaxaca east 0.02793
corresponding source parameters. 4 Oaxaca center I 0.01898
As described by Ordaz and Reyes [32], for each source, each 5 Oaxaca center II 0.01339
parameter was obtained from the Mexican catalog of earthquakes 6 Oaxaca west 0.01116
7 Ometepec 0.02899
prepared by Zuñiga and Guzmán [33], in which bayesian statistics
8 San Marcos 0.01116
procedures described by Rosenblueth and Ordaz [34], and Arbo- 9 Gerrero 0.02232
leda and Ordaz [35] were used. 10 Petatlán 0.01563
11 Michoacán 0.03356
12 Colima 0.01786
13 Brecha de Colima 0.01675
14 Jalisco 0.04566
5.2. Earthquake recurrence models

The recurrence models allow for determining of the average 5.2.1. Gutenberg–Richter model
time that has to pass to have an earthquake with equal char-
acteristics, in a given site. Selection of an appropriate model is logðlm Þ ¼ abm ð6Þ
essential considering that the variability of the seismic periodicity
Where lm, is the exceedance rate of earthquakes with larger or
can be very significant, reaching up to 40% in the average
equal magnitude than m, a and b are constants, which are
recurrence [36]. Several recurrence models of earthquakes are
estimated using statistical analysis of the historical observations
available in the technical literature, among the more used are:
of earthquakes and the data obtained from the geologic evidence.
The value of a indicates the total rate of earthquakes in the region,
and b the ratio of earthquakes of low intensity to large magnitude.
Usually, expression 6 is expressed in complementary accumula-
tive exponential form as
(
l0 ebðMM0 Þ ,if M0 r M r Mu
lðMÞ ¼ ð7Þ
Latitude

1S1 0 ,if M 4M u
1 where l0, is the magnitude exceedance rate for M¼M0, M0 is the
minimum reference magnitude, b, which is a parameter control-
1S2
2 ling the relative frequencies from large to small events, is equal to
bln(10), and Mu, is the maximum possible magnitude at the
source.
Longitude
5.2.2. Gutenberg-Richter modified model [37]
( bM bMu
l0 eebM0e
ebMu
, if M 0 rM rM u
lðMÞ ¼ ð8Þ
0 , if M 4 M u
Latitude

This model is very similar to the previous one for magnitudes

close to Mo, but it shows a smooth transition to l(M) ¼0 at M¼ Mu,
instead of a quit drop.

5.2.3. Characteristic earthquake models

An alternative model is the so called characteristic earthquake
Longitude model. In this model, the maximum earthquake that is generated
in a fault, or in a fault zone, is presented with a larger frequency of
Fig. 14. Seismogenic zones studied (a) earthquake of small to moderate intensity, what is deduced from the Gutenberg and Richter model. This
and (b) characteristics earthquakes. recurrence model has been studied by a number of researchers in
the past decades (e.g. [38–47]). A characteristic earthquake
releases all the seismic potential storage in the system, which
Table 2 has to be recovered during a given period of time. It is common to
Seismic sources parameters for small and moderate events, in all cases M0 ¼ 4.5 be found in the technical literature studies of seismic risk that
and Mu ¼ 7.0 (after Ordaz and Reyes [32]).
adopt models that combine aspects of the characteristic model
Zone Exceedance rate, l0 (1/year) b and the Gutenberg–Richter model (e.g. [38]). For any adopted
model, the parameters needed to define the seismicity in each
1S1 2.014 1.827 zone or fault, for the computation of seismic risk, are the
1 4.792 1.547 maximum magnitude, Mu; minimum magnitude, M0; the rate of
1S2 6.717 1.847
2 18.938 2.059
earthquakes above the minimum magnitude l(M0), and the
coefficient b.
262 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266

In this research work, the recurrence law for the characteristic Table 4
earthquake proposed by Ordaz and Reyes [32] was adopted, in Values of coefficients ai and s for the geometric mean (MG) of the spectral
acceleration (Reyes [49]).
which it is assumed that the behavior of the characteristic earth-
quake observed by Singh et al. [48] in the Mexican subduction T (s) E00 [a1] E00 [a2] E00 [a3] E00 [a4] E00 [a5] E00 [s]
zone follows a Gaussian distribution, and is defined for each
source with the following expression: 0 5.8929 1.2457  0.09757  0.5  0.00632 0.41983
   0.1 6.0831 1.1954  0.09668  0.5  0.00643 0.43341
MEM 0.2 6.7942 1.0675  0.09858  0.5  0.00732 0.43005
lðMÞ ¼ lð7Þ 1F , if M 47 ð9Þ
sM 0.3 6.9623 1.1303  0.10357  0.5  0.00768 0.38868
0.4 6.7632 1.2513  0.09682  0.5  0.00727 0.44214
where, l(7) is the exceedance rate for M4 7, EM and sM are the 0.5 6.9039 1.2236  0.08753  0.5  0.00753 0.4171
mean and standard deviation, respectively, of the magnitude, and 0.6 6.5941 1.2748  0.06768  0.5  0.00693 0.43516
F(U) is the normal cumulative distribution function. 0.7 6.7755 1.3445  0.04662  0.5  0.0076 0.45236
0.8 6.5941 1.3676  0.03662  0.5  0.00705 0.4477
0.9 6.4534 1.347  0.0244  0.5  0.00648 0.42867
5.2.4. Parameters that define the seismicity in the seismogenic zones 1 6.5638 1.3387  0.05429  0.5  0.00665 0.42775
For this study, as suggested by Reyes [49], the seismicity of the 1.1 6.6701 1.3186  0.05696  0.5  0.00703 0.43626
seismogenic zones was modeled with the modified Gutenberg– 1.2 6.6903 1.3167  0.05225  0.5  0.00723 0.44576
1.3 6.6186 1.3183  0.04406  0.5  0.00723 0.45722
Richter recurrence law, to define the exceedance rate of earthquakes
1.4 6.4825 1.3203  0.06347  0.5  0.00662 0.45886
with magnitudes larger than 4.5 and lower of 7. In addition, to define 1.5 6.3741 1.3742  0.08896  0.5  0.00616 0.47824
the exceedance rate of earthquakes with magnitude larger than 7, it 1.6 6.4614 1.4268  0.10542  0.5  0.00632 0.48386
used the recurrence law defined by the expression 9, corresponding 1.7 6.3949 1.4291  0.10135  0.5  0.00604 0.50109
to characteristics earthquakes. The parameters that define the seis- 1.8 6.0912 1.4088  0.09393  0.5  0.00516 0.52712
1.9 5.9378 1.3967  0.07854  0.5  0.00494 0.52748
micity of the seismogenic zones studied for the Gutenberg–Richter 2 5.8698 1.3854  0.05267  0.5  0.00505 0.53709
modified model, are presented in Table 2. Table 3 compiles the model 2.1 5.8057 1.3938  0.03657  0.5  0.00522 0.5694
parameters that define the seismicity of the seismogenic zones 2.2 5.8367 1.4032  0.04392  0.5  0.00547 0.59354
studied for the characteristic earthquake model. 2.3 5.8408 1.4162  0.05743  0.5  0.00561 0.59214
2.4 5.838 1.4032  0.07922  0.5  0.00559 0.58237
2.5 5.8323 1.3937  0.08121  0.5  0.0057 0.584
5.3. Attenuation relationship 2.6 5.858 1.3951  0.06917  0.5  0.00601 0.58756
2.7 5.754 1.3905  0.06887  0.5  0.00589 0.57904
The attenuation relationship relates the ground movement in a 2.8 5.6616 1.3937  0.0717  0.5  0.00583 0.57934
given site, through a generic parameter of the motion, Y, which 2.9 5.5518 1.4126  0.07024  0.5  0.00579 0.58407
3 5.4214 1.4344  0.0608  0.5  0.00568 0.59134
usually is an acceleration or velocity, with the parameter that 3.1 5.323 1.4555  0.06183  0.5  0.0056 0.59488
established the size of the earthquake in the source (e.g. magni- 3.2 5.1785 1.4662  0.06368  0.5  0.00537 0.59129
tude, M), distance from the site to the zone of energy release (R), 3.3 5.0476 1.4684  0.06438  0.5  0.00519 0.58594
and a measure of the model dispersion. Some models include 3.4 4.8327 1.4731  0.06784  0.5  0.00477 0.57814
3.5 4.6026 1.4899  0.06365  0.5  0.00438 0.57545
other terms such as the soil type factor, the fault type that
3.6 4.4198 1.5125  0.06782  0.5  0.00412 0.56937
generates the earthquake, or the earthquake type as a function 3.7 4.2003 1.5357  0.06716  0.5  0.00377 0.5649
of its hypocentral location (e.g. [50–59]). The same cited authors 3.8 3.9923 1.5517  0.0638  0.5  0.00344 0.56863
have proposed several attenuation models to estimate the peak 3.9 3.8123 1.5584  0.06047  0.5  0.00314 0.5817
ground acceleration, PGA, the spectral acceleration corresponding 4 3.6367 1.5601  0.05951  0.5  0.00283 0.59661
4.1 3.4877 1.5489  0.0618  0.5  0.00254 0.60756
to a given specific frequency, Sa (T), or the maximum ground 4.2 3.3429 1.5375  0.06438  0.5  0.00225 0.61156
velocity, PGV, for a given site. For this research it was used the 4.3 3.2848 1.5117  0.05615  0.5  0.00219 0.6209
attenuation law proposed by Reyes [49], which allows the 4.4 3.1955 1.4952  0.05207  0.5  0.00203 0.62776
estimation of the spectral acceleration, Sa, in rock outcrops in 4.5 3.098 1.4864  0.05698  0.5  0.00187 0.63315
4.6 3.0614 1.4894  0.06039  0.5  0.00196 0.63549
Mexico City, and that has the following functional form:
4.7 3.0962 1.5011  0.06049  0.5  0.00232 0.63683
ln YðTÞ ¼ ln SaðTÞ ¼ a1 ðTÞ þ a2 ðTÞðM6Þ 4.8 3.1583 1.5143  0.0536  0.5  0.00276 0.64459
4.9 3.2384 1.5218  0.04547  0.5  0.00325 0.65558
þ a3 ðTÞðM6Þ2 þ a4 ðTÞln Rþ a5 ðTÞRþ eðTÞ ð10Þ
5 3.2887 1.5282  0.03953  0.5  0.00364 0.66443
where Y represents the maximum ground acceleration in one of 5.1 3.2792 1.5237  0.03394  0.5  0.00379 0.67003
5.2 3.213 1.5208  0.0312  0.5  0.00375 0.67315
the orthogonal directions, or its geometric mean, MG, in cm/s2; M
5.3 3.1213 1.5266  0.03198  0.5  0.00364 0.67495
is the seismic moment magnitude; R is the minimum distance 5.4 3.0212 1.5268  0.03054  0.5  0.0035 0.67766
from the site to the rupture area, in km; ai are coefficients 5.5 2.9345 1.523  0.0244  0.5  0.0034 0.68002
presented in Table 4 for the geometric mean, and in Reyes [49] 5.6 2.8914 1.5329  0.01676  0.5  0.00348 0.68336
5.7 2.9014 1.5408  0.00898  0.5  0.00371 0.68717
for the EW and NS components; e is the error made when
5.8 2.9438 1.5417 0.00253  0.5  0.00403 0.69477
estimating the response spectrum with the attenuation law. This 5.9 3.0208 1.5413 0.01408  0.5  0.00444 0.70459
functional is the same as that proposed by Joyner and Boore [60]. 6 3.0975 1.5468 0.02315  0.5  0.00484 0.71082

5.3.1. Validation of the attenuation law for rock outcrops

Seeking to corroborate the appropriateness of the attenuation
law used, the response spectrum for the 1995 September earth-
quake was obtained in two sites located in rock, CU (UNAM), and
TXCR. The earthquake had a moment magnitude, Mw ¼7.3, and an
epicentral distance of 280 km, with epicenter in the subduction
zone. Fig. 15 depicts the response spectra estimated with Eq. (10)
for the geometric mean and EW component, and those obtained
with the records measured for sites CU (UNAM) and TXCR,
showing a good agreement overall between the predicted and
the measured data, in both seismic stations. Based on these
results, the model was considered representative of the ground
motions observed in Mexico City in rock outcrops.
5.4. Solution of the integral of seismic risk
Once established the earthquakes recurrence models, the
seismogenic zones where they occur, and the attenuation law
for the studied site, the integral of seismic risk needs to be solved
 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266 263

40 12
EW model EW model

Spectral acceleration, Sa (cm/s )

Geometric mean

Spectral acceleration, Sa (cm/s )

35 Geometric mean

2
model 10 model
EW measured EW measured
30
8
25 5 % damping
5 % damping
20 6

15
4
10
2
5

0 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6
Period (s) Period (s)

Fig. 15. Spectral acceleration in sites CU and TXCR for the September 14 earthquake 1995, Mw ¼ 7.3.

according to the probabilistic approach (e.g. [61,27]). The risk is

evaluated as the probability of exceeding an upper value of the
parameter of the ground motion at the site, due to the activity of all
seismogenic zones that surround it, and which may contribute to the
TXCR
expected ground motion. The functional form of the integral of
seismic risk due to a set of N seismic sources is the following [28]:
XN Z Rmax Z Mmax

gðyÞ ¼ ni f Mi ðMÞf Ri ðRÞP i ½Y 4 yM,R dMdR ð11Þ
i¼1 Rmin Mmin

where the double integral has limits as the minimum and maximum
of the magnitude and distances from the source. g(y) represents the
annual rate of exceeding the level of motion, y, due to the occurrence
of earthquakes in the N sources, which is the sum of the annual rates
of exceedance g(y) in each of the sources (that have an annual rate of
occurrence of ni earthquakes). The term Pi[Y4y9M,R] gives the
probability of exceedance of y, conditioned to the variables M and
R. Finally, the functions fMi(M) and fRi(R) are probability density
TXCR
functions (PDF) of the magnitude and distance, respectively. In
particular in this research, it was adopted the particular simplified
case of point-source, thus fRi(R)¼18R, is equal to one for every point.

5.5. Probabilistic seismic hazard curves for the TXCR site

Probabilistic seismic hazard curves for the San Miguel Tlaixpan

TXCR site were determined for a range of seismic moment magni-
tudes varying from 4.5 to 8.2. Seismogenic zones with irregular Fig. 16. Seismogenic areas studied, earthquakes of small to moderate intensity
shapes were subdivided in regular constant areas (squares of (a) and characteristic earthquakes (b).

30  30 km), see Fig. 16, as suggested in Ordaz and Reyes [32]. This
subdivision allowed to assume that the forms generated are point
sources with all the tributary seismicity concentrated at its geome-
trical center [62], simplifying the probabilistic seismic hazard analy-
sis. For each point the distance from the source to the study site was
determined for each seismic zone to where they belong. With these
analyzes 61 seismic hazard curves for a range of periods from 0 to
6.0 s were obtained. In particular, Fig. 17 presents the seismic hazard
curves for periods of 0, 3.0 and 6.0 s.
5.6. Uniform hazard spectrum for the TXCR site
representation of the relationship between the natural vibration
period, T, and spectral acceleration, Sa, for a given exceedance
probability associated with a return period. A uniform hazard
spectrum for return period 125 years was obtained from the seismic
hazard curves (see Fig. 18). This figure also shows the uniform hazard
spectrum for CU site (UNAM) for the same return period, obtained by
Ordaz et al. [63]. Note that this spectrum defines the seismic
environment of the area. Therefore it was used as input motion
during the site response analyses.

The uniform hazard spectrum determined to characterize the
seismic environment was developed at the same location as the rock
station TXCR, at about 18.70 km from the site, to be able to compare
it directly with measured responses, if needed, for future earthquakes.
As it is well known, the uniform hazard spectra, UHS, is a
6. Site response analysis
The seismic response of each virtual soil profile was obtained
using random vibration theory. The algorithm used to compute
the probabilistic site response of horizontally stratified soil
264 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266

105 104 103

Exceedance Rates, λ (Sa) (1/year)

Exceedance Rates, λ (Sa) (1/year)

Exceedance Rates, λ (Sa) (1/year)

T= 0 s T=3 s T=6 s
103 1
10
102
101
10-1
-1 0
10 10
10-3
10-3
10-2
10-5
10-5

10-7 10-4 10-7

0.01 0.1 1 10 100 0.01 0.1 1 10 100 1000 0.01 0.1 1 10 100 1000
Spectral Acceleration, Sa (cm/s2) Spectral Acceleration, Sa (cm/s2) Spectral Acceleration, Sa (cm/s2)

Fig. 17. Seismic hazard curves for periods of 0, 3.0 and 6.0 s.

0.16
CU (Ordaz
0.14 et al., 2000)
TXCR
Spectral acceleratio, Sa (g)

(Osorio, 2012)
0.12

0.1
5 % damping
0.08

0.06

0.04

0.02
0 1 2 3 4 5 6
Period (s)

Fig. 18. Uniform hazard spectra for a return period of 125 years for TXCR and UC
sites, obtained with the parameters of the geometric mean of the attenuation
relationship proposed by Ordaz et al.

deposits subject to two dimensional SH waves propagating

vertically is formulated in the frequency domain, assuming Fig. 19. Response spectra obtained for the studied area.
equivalent linear soil properties. It uses the Thompson–Haskell
[64] solution. The input motion is defined in terms of a response
spectrum from which the equivalent power spectrum of the Table 5
excitation is computed. The accelerations, strains and response Parameter values needed to compute the proposed design spectra.
spectra in different points of the system are computed from the
corresponding power spectrum and the extreme value theory. c a0 Ta a Tba r

Thus, this random process is completely characterized by its 0.37 0.13 0.40 3.0 2.2
power spectra. Using the random vibration theory, the maximum
responses of a linear system excited by a stochastic process can be a
Period in second.
computed for a given confidence level. Physically this character-
ization of the seismic environment is equivalent to consider an  r
infinite number of acceleration time histories with the same Tb
Sa ¼ c ; if T 4 T b ð14Þ
mean frequency content but with randomly distributed phases. T
A total of 2501 site response analyses were carried out using the The parameters to be used in these expressions are summar-
UHSrock as input motion. Following the recommendation of ized in Table 5.
Rosenblueth et al. [65], a reduction of 40% was applied to the
enveloped of the maximum response spectra computed to obtain
the suggested response spectra for the studied area. Fig. 19 shows 6.1. Comparison with measured response
the response spectra obtained for all sites, as well as the proposed
response spectrum. Given by the following equations: Fig. 20 presents a comparison of the design response spectrum
proposed, with the medium, medium plus one standard deviation,
T
Sa ¼ a0 þðcaO Þ ; if T oT a ð12Þ s, and enveloped of the recording measured in sites TXSO
Ta
and TXCH, scaled to a Peak Ground Acceleration, PGA¼0.13 g.
The mean, mean þ1s and enveloped response spectrum of mea-
Sa ¼ c; if T a rT rT b ð13Þ sured responses in the sites TXSO and TXCH, were obtained
 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266 265

0.7 0.7
TXSO TXCH
Envelope Envelope
0.6 Mean + 1σ 0.6 Mean + 1σ

Spectral acceleration, Sa (g)

Spectral acceleration, Sa (g)
Mean Mean
Proposed Proposed
0.5 0.5
5 % damping 5 % damping

0.4 0.4

0.3 0.3

0.2 0.2

0.1 0.1

0 0
0 1 2 3 4 5 0 1 2 3 4 5
Period (s) Period (s)

Fig. 20. Mean, mean þ1s, and enveloped of recorded ground motions compared with the proposed response spectrum.

Table 6
Earthquakes used to obtain the mean, mean þ1s and the envelope measured at sites TXSO and TXCH.

Site Event Ms Epicentral distance (km) Soil type PGA (gal)

TXSO 25/OCT/81 7.3 311.97 Soft clay NS ¼  23.793; V ¼4.324; EW¼  28.320
21/SEP/85 7.60 381.1 Soft clay NS ¼38.561; V ¼ 0.000; EW ¼34.748
19/SEP/85 8.10 444.42 Soft clay NS ¼103.036; V ¼  25.529; EW¼  102.973
TXCH 07/JUN/82 (2) 7.0 353.74 Soft clay NS ¼  11.41; V ¼5.47; EW ¼ 12.19
07/JUN/82 (1) 6.9 365.94 Soft clay NS ¼  22.17; V ¼5.33; EW ¼ 14.30
14/MAR/79 Mb¼ 7.0 324.74 Soft clay NS ¼31.86; V ¼  12.49; EW¼  22.43

following the methodology proposed by Mayoral et al. [5], using obtained from the 2501 site response analyses, from which, in
the subduction earthquakes compiled in Table 6. turn, the proposed response spectrum was derived.

7. Conclusions References

This paper describes the framework used to develop a design [1] Romo MP, Jaime A. Caracterı́sticas dinámicas de las arcillas del Valle de
México y análisis de respuesta de sitio, Reporte Interno, Instituto de
response spectrum for a 150 km2 area located within the Texcoco Ingenierı́a, UNAM; April 1986.
lake region to be used in future updates of the Mexico City [2] Ovando E, Romo MP. Estimación de la velocidad de ondas S en la arcilla de la
building code. This involved field investigation, laboratory testing, ciudad de México con ensayos de cono, Revista Sismodinámica; 1991. 2,
p. 107–123.
and analytical studies. Geo-statistical analyses were carried out to [3] Romo, MP. Clay behavior, soil response and soil structure interaction studies
characterize the main parameters that affect the seismic response in Mexico city. In: Proceedings of the third international conference on recent
at the studied site, using a grid of 2501 virtual soil profiles. In advances in geotechnical earthquake engineering and soil dynamics, vol 2.
San Luis Missouri, USA; 1995. p. 1039–1051.
particular, it was observed that the spatial variation of measured
[4] Mayoral JM, Romo MP Osorio L. Estudios de campo y laboratorio llevados a
shear wave velocity exhibited a lognormal probability distribu- cabo para la caracterización geo-sı́smica en la zona del Lago de Texcoco,
tion, whereas the measured plasticity index data followed a Informe Interno, Instituto de Ingenierı́a, UNAM; 2006.
normal probability distribution. Regarding modulus degradation [5] Mayoral JM, Romo MP, Osorio L. Seismic parameters characterization at
Texcoco lake, Mexico. Soil Dynamics and Earthquake Engineering
and damping curves, the Darendeli and Stokoe’s [20] model 2008;28(July 7):507–21.
seems to capture well the laboratory data, in particular for shear [6] Rosenblueth E. Modelos probabilı́sticos de la ocurrencia de temblores,
stiffness, under predicting the damping for shear strains larger Memorias del Simposio Generación, Propagación y Efectos de Temblores,
SMMS, SMIS, SMF, UGM, April 1987.
than 0.1% in the clayey materials. The geo-statistical model [7] Esteva L, Ordaz M. Riesgo sı́smico y espectros de diseño en la República
generated to infer the virtual soil profiles in the area, based on Mexicana, Informe Interno, Instituto de Ingenierı́a, UNAM; 1989.
the ordinary kriging interpolation technique was able to capture [8] Singh, SK Ordaz, M. Sismicidad y Movimientos Fuertes en México: Una Visión
Actual. CENAPRED; 1994.
fairly well good shear wave velocity distributions measured at [9] Pérez-Rocha, Sánchez-Sesma, Ordaz, Singh Reinoso. Strong ground motion
four exploration borings, as well as the dynamic response prediction at México City. In: Proceedings of 12WCEE, New Zealand, paper
recorded at the seismic station TXSO, located within the area, 2693; 2000.
[10] Romo MP. Soil–structure interaction in a random seismic environment, PhD
during the 1985 Michoacan earthquake. Regarding the uniform dissertation, University of California, Berkeley; 1976.
hazard spectrum computed at seismic station TXCR, located at [11] Seed HB, Sun JI. Implications of site effects in the Mexico City earthquake of
about 18.7 km for the site in a rock outcrop, it was observed an Sept. 19, 1995 for earthquake-resistant design criteria in the San Francisco
Bay area of California, Report No UCB/EERC-89/03; March 1989.
slightly difference, specially in the frequency content, with
[12] Rosenblueth E, Ovando E. Geotechnical lessons from Mexico City and other
respect to that determined by other researchers at the CU (UNAM) recent earthquakes. In: Proceedings of Second International Conference on
station, located at about 31.7 km, supporting the fact of ground Recent Advances in Geotechnical and Earth Engineering and Soil Dynamics,
motion modification in the different rock formations of Mexico vol II, 1799–1811, University of Missouri, Rolla; 1991.
[13] Lermo J, Chávez-Garcı́a FJ. Site effect evaluation at Mexico city: dominant
City, even for distances relatively small. The potential spatial period and relative amplification from strong motion and microtremor
variability observed in the area was captured by a single envelope records. Soil Dynamics and Earthquake Engineering 1994;13:413–23.
266 L. Osorio, J.M. Mayoral / Soil Dynamics and Earthquake Engineering 48 (2013) 252–266
[14] RCDF. Reglamento de Construcciones para el Distrito Federal, Administración
Pública del Distrito Federal, Jefatura de Gobierno, Normas Técnicas Comple-
mentarias para el Diseño por Sismos, México; 2004.
[15] Feston Gordon A. Estimation for stochastic soil models. ASCE Journal of
Geotechnical and Geoenvironmental Engineering 1999;125(6):470–85.
[16] Lenz Jennifer A, Baise Laurie G. Spatial variability of liquefaction potential in
regional mapping using CPT and SPT. Soil Dynamics and Earthquake
Engineering 2007;27(July 7).
[17] Maximilian Huber, Axel Moellmann, András Bárdossy and Vermeer Pieter A.,
2009, Contributions to probabilistic soil modeling. In: Proccedings of the 7th
international probabilistic workshop, Delft.
[18] Seed HB, Idriss MI, Arango I. Evaluation of liquefaction potential using field
performance data. Journal of the Geotechnical Engineering Division, ASCE
1983;109(3):458–82.
[19] Rouhani S, Srivastava M, Desbarats AJ, Cromer. MV, Johnson AI. Geostatistics
for environmental and geotechnical applications, ASTM STP 1283. American
Society for Testing and Materials 1996, http://dx.doi.org/10.1520/STP1283-EB.
[20] Darendeli MB, Stokoe KH., II Development of a new family of normalized
modulus reduction and material damping curves, Geotechnical Engineering
Report GD01-1, University of Texas at Austin, Austin; 2001.
[21] Bill WR, Lomnitz C, Flores H. Late resonant response at Texcoco, valley of
Mexico, during distant earthquakes. Soil Dynamics and Earthquake Engineering
2006;26:791–8.
[22] Electric power research institute (EPRI). Guidelines for determining design
basis ground motions, in volume 2: appendices for ground motion
estimation—Appendix 6.A: probabilistic model of soil profile variability,
EPRITR-102293; 1993.
[23] Bazzurro Paolo, Cornell CAllin. Ground motion amplification in nonlinear soil
sites with uncertain properties. Bulletin of the Seismological Society of
America 2004;94:2090–109.
[24] Holzer TL, Bennett MJ, Noce TE, and Tinsley III JC. Statistical distribution sand
depth dependence of shear wave velocity, earthquake spectra, vol. 21, No. 1,
p. 161–177, Earthquake engineering research institute; February 2005.
[25] Hardin BO, Drnevich VP. Shear modulus and damping in soils: design
equations and curves. Journal of the Soil Mechanics and Foundations
Division, ASCE 1972;98(June SM6):603–24.
[26] Vucetic M, Dobry R. Effect of soil plasticity on cyclic response. Journal of oteh.
Engineering, ASCE 1991;114(1).
[27] Cornell CAllin. Engineering seismic risk analysis. Bulletin of the Seismological
Society of America 1968;58(5):1583–606.
[28] McGuire K. Probabilistic seismic hazard analysis and design earthquakes:
closing the loop. Bulletin of the Seismological Society of America
1995;85:1275–84.
[29] Klügel Jens-Uwe. Seismic hazard analysis—Quo vadis? Earth-Science
Reviews0012-8252 2008;88(1–2):1–32 May.
[30] Baker JW. 2008. An introduction to probabilistic seismic hazard analysis
(PSHA), White Paper, Version 1.3, 72 pp.
[31] Nishenko SP, Singh SK. Conditional probabilities for the recurrence of large
and great interplate earthquakes along the Mexican subduction zone.
Bulletin of the Seismological Society of America 1987;77(6):2095–114.
[32] Ordaz M, Reyes C. Earthquake hazard in Mexico City: observations versus
computations. Bulletin of the Seismological Society of America 1999;89:
1379–83.
[33] Zúñiga R; Guzmán M. Main seismogenic source zones in Mexico, Technical
Report, Seismic Hazard Project, IPGH; 1994.
[34] Rosenblueth E, Ordaz M. Use of seismic data from similar regions. Earthquake
Engineering and Structural Dynamics 1987;15:619–34.
[35] Arboleda J Ordaz M. Un mejor uso de los datos estadı́sticos para estimación
de la sismicidad local. Memoria el X Congreso de Ingenierı́a Sı́smica, Puerto
Vallarta Jalisco, Méx; 1993. p. 21–27.
[36] Goes SDB. Irregular recurrence of large earthquakes: an analysis of historic
and paleoseismic catalogs. Journal of Geophysical Research
1996;101(B3):5739–49, http://dx.doi.org/10.1029/95JB03044.
[37] Cornell, CA, and Vanmarcke E. The major influences on seismic risk. In:
Proceedings of 4th world conference on earthquake engineering, Santiago de
Chile; 15–19 January 1969.
[38] Youngs R, Coppersmith Kevin J. Implications of fault slip rates and earth-
quake recurrence models to probabilistic seismic hazard estimates. Bulletin
of the Seismological Society of America 1985;75:939–64.
[39] Wesnousky SG. The Gutenberg richter or characteristic earthquake distribu-
tion, which is it? Bulletin of the Seismological Society of America
1994;84:1940–59.
[40] Sammis CG, Sornette D, Saleur H. Complexity and earthquake forecasting in
reduction and predictability in natural disasters. In: Rundie JB, Klein W,
Turcotte DL, editors. SFU Studies in the sciences of complexity, Vol X.X.V.
Reading, MA: Addison-Wesley; 1996. p. 143–56.
[41] Stirling MW, Wesnousky SG, Shimazaki K. Fault-trace complexity, cumulative slip,
and the shape of the magnitude-frequency distribution for strike-slip faults: a
global survey. Geophysical Journal International 1996;124:833–68.
[42] Ben-Zion Y. Stress, slip and earthquakes in models of complex single-
faultsystems incorporating brittle and creep deformations. Journal of Geo-
physical Research 1996;101:5677–706.
[43] Rüttener E, Egozcue JJ, Mayer-Rosa D, Mueller S. Bayesian estimation of
seismic hazard for two sites in Switzerland. Natural Hazards 1996;14:
165–78.
[44] Anderson JG, Wesnousky SG, Stirling MW. Earthquake size as a function of
fault slip rate. Bulletin of the Seismological Society of America 1996;86:
683–90.
[45] Yeats R, Sieh K, Allen C. The geology of earthquakes. New York: Oxford
University Press; 1997 567 pp.
[46] Huang Y, Saleur H, Sammis C, Sornette D. Precursors, aftershocks, criticality
and self-organized criticality. Europhysics Letters 1998;41:43–8.
[47] Dixon TH, Norabuena E, Hotaling L. Paleoseismology and global positioning
system: earthquake-cycle effects and geodetic versus geologic fault slip rates
in the Eastern California shear zone. Geology 2003;31:55–8.
[48] Singh SK, Rodriguez M, Esteva L. Statistics of small earthquakes and
frequency of occurrence of large earthquakes along the Mexican subduction
zone. Bulletin of the Seismological Society of America 1983;73:1779–96.
[49] Reyes, C. E1 estado limite de servicio en el diseño sismico de edificios, PhD
Thesis, School of Engineering, National Autonomous University of Mexico
(UNAM); 1999.
[50] Esteva, L.Seismic risk and seismic design. In: Hansen RJ, editor. Seismic
design for nuclear power plants, p. 142–182. The M.I.T. Press; 1970.
[51] McGuire RK. Seismic design spectra and mapping procedures using hazard
analysis based directly on oscillator response. Earthquake Engineering and
Structural Dynamics 1977;5:211–34.
[52] Blume, JA.Distance partitioning in attenuation studies. In: Proceedings of
seventh world conference on earthquake engineering, vol. 2, p. 403–410;
1980.
[53] Joyner WB, Boore DM. Peak horizontal acceleration and velocity from strong-
motion records including records from the 1979 Imperial Valley, California,
earthquake. Bulletin of the Seismological Society of America 1981;71(6):
2011–38.
[54] Bolt BA, Abrahamson NA. New attenuation relations for peak and expected
accelerations of strong ground motion. Bulletin of the Seismological Society
of America 1982;72(6):2307–21.
[55] Singh SK, Mena E, Castro R, Carmona C. Empirical prediction of ground
motion in Mexico City from coastal earthquakes. Bulletin of the Seismological
Society of America 1987;77(5):1862–7.
[56] Ambraseys NN. Uniform magnitude reevaluation of European earthquakes
associated with strong-motion records. Earthquake Engineering and Struc-
tural Dynamics 1990;19(1):1–20.
[57] Crouse CB, McGuire JW. Site response studies for purpose of revising NEHRP
seismic provisions. Earthquake Spectra 1996;12(3):407–39.
[58] Boore DM, Joyner WB, Fumal TE. Equations for estimating horizontal
response spectra and peak acceleration from western North American
earthquakes: a summary of recent work. Seismological Research Letters
1997;68(1):128–53.
[59] Campbell KW, & Bozorgnia Y. New empirical models for predicting near-
source horizontal, vertical, and V/H response spectra: implications for design.
In: Proceedings of the sixth international conference on seismic zonation;
2000.
[60] Joyner WB, & Boore DM. Measurement, characterization, and prediction of
strong ground motion. In: Proceedings of earthquake engineering & soil
dynamics II. Geotechnical Division, ASCE; 1988. p. 43–102.
[61] Esteva L. Criteria for the construction of spectra for seismic design, 3rd Pan-
American symposium of structures, Caracas, Venezuela, 3–8 July 1967.
[62] Ordaz M. Some integrals useful in probabilistic seismic hazard analysis.
Bulletin of the Seismological Society of America 2004;94:1510–6.
[63] Ordaz M, Miranda E Avilés J. Propuesta de espectros de diseño por sismo para
el D.F., Memorias, VI Simposio Nacional de Ingenierı́a Sı́smica, Querétaro,
Qro., México, septiembre; 2000. p. 52–66.
[64] Thomson-Haskell NA. The dispersion of surface waves on multilayered
media. Bulletin of the Seismological Society of America 1953;43:17–34.
[65] Rosenblueth E, Ordaz M, Sanchez-Sesma FJ, Singh SK. The Mexico earthquake
of September 19, 1985, design spectra for Mexico’s federal district. Earth-
quake Spectra 1989;5(1):273–91.