See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/261831405

Spectral and Amplification Characteristics in

San Salvador City (El Salvador) for Upper-
Crustal and Subduction Earthquakes

Conference Paper · January 2002

CITATIONS READS

6 43

1 author:

Walter Salazar
Catholic University of El Salvador
30 PUBLICATIONS 144 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

REAKT WP7 - Strategic Applications and Capacity Building View project

All content following this page was uploaded by Walter Salazar on 24 April 2014.

The user has requested enhancement of the downloaded file.

 SPECTRAL AND AMPLIFICATION CHARACTERISTICS
IN SAN SALVADOR CITY (EL SALVADOR) FOR UPPER-CRUSTAL
AND SUBDUCTION EARTHQUAKES
Walter SALAZAR (PhD student, Tokyo Institute of Technology, Japan, [email protected])
Kazuoh SEO (Professor, Tokyo Institute of Technology, Japan, [email protected])

El Salvador is a small country situated in the Central America Isthmus. Four destructive earthquakes have occurred since 1982
associated with the upper-crustal and subduction seismic activity in the region, two from each source. A very high vulnerability has
been observed in the capital City - San Salvador- where many buildings have collapsed or they still have severe damage. Adobe and
Bahareque houses have always collapsed and many landslides have occurred in the volcanic ash, especially in the January 13th 2001
event in a large scale. It has been noticed that seismic microzonation studies have not been taken into account in the city planning.
Despite that there was some available seismic microzanation proposal by the Italian mission after the earthquake disaster of October
10th 1986, no governmental and private institution use it to mitigate the seismic risk in the area. Also, no seismic microzonation study
has been included in the actual seismic code.
In this article an attempt is made to study the strong motion characteristics for the upper-crustal and subduction earthquakes,
developing horizontal elastic response spectra for 5% damping and getting the amplification characteristics from soil profiles
available, in order to confirm the basic features of the seismic microzonation. For this purpose, the strong motion records from the
four destructive earthquakes have been collected and processed and the soil characteristics as shear wave velocity (VS ) and damping
are taken from the excellent job of the Italian Mission in 1988. To get the transfer function a 1D SH wave propagation model is used.
Besides, the predicted spectral ordinates from available attenuation relationships for Central America are compared with the response
spectra of observed records.
The San Salvador strata are products of distinct eruptions of active Quaternary volcanoes. It includes mainly lava flows and volcanic
ashes. A very important characteristic is that these deposits can be intercalated each other. Ilopango Lake deposited the youngest
pyroclastic flow which is called Tierra Blanca. Tuffs are generally presented underlying the Tierra Blanca. Generally the shear wave
velocity distributes from 225 m/s (Tierra Blanca) to 500 m/s (Tuffs). A stiff lava flow is found with a shear wave velocity of 2,100
m/s.
The seismic microzoning proposed looks very reasonable for the subduction and upper-crustal earthquakes when the response spectra
are compared. But the strong motion records are not so many and it is difficult to have a definitive opinion. Using the borehole data it
is noticed that the layers which contribute to the amplification are the ones which are next to the stiff lava flow. This might be due to
the high impedance ratio since high contrast in the shear wave velocity is found between the lava and the volcanic ash. It is very
difficult to conclude about the amplification characteristics in some zones of the microzonation using only the borehole data, since
only a few meters (30-60 m) are available, because the hole was made reaching the first stiff lava flow. Some soil profiles from water
wells (depth at least of 100 m) reveal the existence of sediments below the lava. Very interesting features are observed in the
amplification characteristics at these sites. For example, some times the second mode of vibration has bigger amplification factor
than the first mode or similar amplification factors are obtained in the first modes due to the special configuration of the intercalated
materials. Sometimes, the shape of the amplification function is similar to those of the response spectra. Generally, amplification
factors yields less than 10, in the period range of 0.1 to 1.0 s. The cumulative thickness of lava flow seems to be an important factor
in the amplification characteristics in the area, and it might be an important aspect in the microtremors characteristics. It is desirable
to have more detail information of dynamic properties for deeper soil structures and to install new strong motion instruments in some
areas where there is a lack of data.
For upper-crustal and subduction events, the predicted spectral values from the attenuation relationships for Central America always
underestimate the observed ones. It is worth mentioning that these equations incorporate a site term in the regression analysis to
distinguish a soil site or rock condition in the prediction, but they do not separate the subduction from the upper-crustal events. The
Mexican earthquakes in the regression analysis were included, where the upper clay sediments with VS of 50 m/s are the responsible
for the amplification in long period components, which is the opposite case from San Salvador City. A new effort should be done to
have new attenuation relationships for the Central America area.
Special attention should be focused in the buildings with a fundamental period less than 1.0s. Despite that some buildings suffered
heavy damage by the last earthquake of October 10th 1986 in the capital area, the differences in the observed spectral ordinates
partially explain why the damage in the capital for 4-6 story buildings was not so severe during the January 13th 2001 event.

0
 SPECTRAL AND AMPLIFICATION CHARACTERISITICS
IN SAN SALVADOR CITY (EL SALVADOR) FOR UPPER-CRUSTAL
AND SUBDUCTION EARTHQUAKES

Walter SALAZAR1 and Kazuoh SEO2

1
PhD student, Tokyo Institute of Technology, Japan, [email protected]
2
Professor, Tokyo Institute of Technology, Japan, [email protected]

ABSTRACT: An attempt is made in order to study the strong motion characteristics for upper-crustal and subduction
destructive earthquakes in San Salvador City. Horizontal elastic response spectra for 5% damping and the site
amplification factors are obtained to confirm the seismic microzonation, which has been proposed by an Italian mission
in 1988, after the earthquake disaster of October 10th 1986. Besides, the predicted spectral ordinates from available
attenuation relationships for Central America are compared with the spectra of observed records at the specific sites. A
brief discussion is made regarding with some microtremors measurements at strong motion stations.

Key Words: seismic microzonation, response spectra, site amplification factor, lava flow, volcanic ash.

INTRODUCTION analogue records were processed using a baseline correction

and with a band pass filter of 0.20-25 Hz. The records from
Four destructive earthquakes have occurred in El Salvador digital instruments were corrected with a band pass filter of
since 1982 associated with the subduction and upper-crustal 0.05-50 Hz. Representative accelerograms for the four
seismic activity in the region (Fig. 1). The subduction event of earthquakes studied are presented in Fig. 2 and the locations of
June 19th 1982 is located in front of El Salvador coast in the the strong motion stations are drawn in Fig. 3.
Pacific Ocean with a magnitude of 7.3 MS and 80 km in depth.
On January 13th 2001, an earthquake occurred in front of El
Salvador coast with a magnitude 7.8 MS and depth of 60 km.
Normal fault are reported for both intraplate shocks in the
Cocos plate (USGS). The upper-crustal earthquake of October
10th 1986 had the epicenter in San Salvador City with a depth
of 10 km. Despite of its moderate magnitude (5.4 MS), a lot of
damage was observed in the epicentral area in many RC frame
structures. On February 13th an upper-crustal shock occurred at
the east of San Salvador City with a magnitude of 6.5 MS and
depth of 7 km. These last two events had a strike slip
mechanism. Many landslides have been occurred in the
volcanic ash for all the events, especially on January 13th 2001
in a large scale. Adobe and Bahareque constructions have
always collapsed and it is noticed that the seismic
microzonation studies have not been taken into account in the
city planning (Salazar & Seo, 200110). This article is focusing
the spectral and amplification characteristics based on a new
strong motion data set and the available seismic microzonation. Fig. 1 Epicenters of destructive earthquakes in El Salvador
since 1982. I.L: Ilopango Lake

500 500
STRONG MOTION RECORDS AND SEISMIC OBS (Zone 2) 1982/06/19 CIG (Zone 2) 1986/10/10
250 250
MICROZONATION
Acceleration (cm/s/s)

0 0

-250 -250
Strong motion records N-S PGA: 169.1 cm/s/s
180 deg.(Az.)PGA: 409.9 cm/s/s
-500 -500

The strong motion records used in this work are coming from 500 EXT (Zone 1B) 2001/01/13 500 EXT (Zone 1B) 2001/02/13
Central America University UCA (2001)13 and the Ministry of 250 250
Environment and Natural Resources of El Salvador MARN. In 0 0

the case of the event of October 10th 1986, the uncorrected -250 N-S PGA: 295.3 cm/s/s -250 N-S PGA: 121.3 cm/s/s
digital records (coming from analogue instruments) did not -500 -500
0 20 40 60 0 20 40 60
have equal time interval between acceleration values. A high- time(s)
order polynomial interpolation for each 10 points was made to
have equal time interval of 0.005 s (Press et al., 19926). The Fig. 2 Accelerograms for destructive earthquakes.
1
A total of 27 records in 16 stations for the four earthquakes are RESPONSE SPECTRA AND SITE AMPLIFICATION
analyzed in the whole area of San Salvador. FACTORS

Soil characterisitics and seismic microzonation Response spectra

The San Salvador strata are products of distinct eruptions of the Since there is no records in zone 1A, the attention is focused in
active Quaternary volcanoes in the region. It includes mainly zone 1B and 2. For the October 10th 1986, June 19th 1982 and
lava flows and volcanic ashes. Generally these deposits are January 13th 2001 earthquakes, generally in Zone 2 the
intercalated each other. The youngest pyroclastic flows were predominant period looks longer that in Zone 1B (Fig. 6 and 8).
deposited in successive eruptions from Ilopango Lake (Fig. 1) This can be related to the definitions above, the cumulative
and this volcanic ash is called Tierra Blanca (white soil) and thickness of lava flow in Zone 2 is less than Zone 1B. For the
has been dated at about 260 A.D.. Generally tuffs are presented February 13th 2001 earthquake this tendency is not observed
underlying the Tierra Blanca and they were deposited by San clearly unlike the other earthquakes (Fig. 7).
Salvador volcano together with the lava flows (Schmidt- VS (m/s) VS (m/s) VS (m/s)
Thome, 197511). Five soil profiles with shear wave velocity 0 200 400 600 800 0 200 400 600 800 0 200 400 600 800
(VS) are available at strong motion stations (Fig. 3-4). 0 0 0

Generally the VS distributes from 225 m/s (Tierra Blanca) to 10 10 10

scoria
500 m/s (tuffs). A very stiff lava flow is found with VS of 2100
20 20 scoria
2,100 m/s. Besides, a scoriaceous lava can be found also with 20
2100
a VS of 280-470 m/s but with a very high P-wave velocity 30 30 scoria 30
around 2,600 m/s as the case of UCA and HSH station.

H (m)
40 40 2100 40
The characteristics of the Zones in the seismic
microzonation (Faccioli et al., 19883) are briefly discussed as 50 50 50
follows (see Fig. 3): 60 60
60
-Zone 1: At least 65 m total thickness of lava flows within HCR (Zone 1B) UCA (Zone 1B) HSH (Zone 1B)
the first 100 m, and it is divided in 1A: Lava flows are exposed 70 70 70

or within 5 m from the surface. 1B: Lava flows are found at VS (m/s) VS (m/s)
0 200 400 600 800 0 200 400 600 800
minimum depth ranging between 5 m and 35m. 0 0
-Zone 2: less than 35 m of total thickness of lava flows
10
within the first 100 m. It is expected that in Zone 2 more longer 10
2100
period components appears that in the Zone 1. 20 20

30 30
H (m)

40 40

50 50

60 60

IGN (Zone 2) 70 CIG (Zone 2)

70

Fig. 4 Shear wave velocity profiles at strong motion stations.

ELS2-59 8N9 P2 9N8 2N10

(Zone 1B) (Zone 2) (Zone 2) Boundary Outside of
Zone 1B-2 the study
area
6 13 3
15
18 22.8
12.5
Fig. 3 Seismic microzonation for the south part of the city of 30.5 48 67
18.5
San Salvador (Faccioli et al, 19883). Projection in the
surface of rupture fault for the event of October 10th
1986 is based on aftershock distribution and the study 95
11
of Harlow et al. (1993)8. 43
76.8 86.9
43
The seismic microzonation does not cover the whole area of the 100 m
San Salvador city, only the south part of it. 3
Some available soil profiles from water wells (from
Administración Nacional de Acueductos y Alcantarillados
ANDA) with a depth more than 100 m reveal that there are 39.6 96
sediments underlaying the lava. Four representative soil
profiles have been chosen inside the study area (Fig. 3 and Fig.
5). One soil profile is outside of the microzonation.

Fig. 5 Soil profiles. The thickness of volcanic ash (white) and

lava (black) are written in the right side of the profiles
in meters. See locations in Figure 3.

2
Because these response spectra contain the influence of the components (Seo, 199712).
source, path and site effects, an attempt is made in order to
separate the site amplification factors using the soil information 6 6
shown before. 5 5
4

SA/PGA

SA/PGA
UCA (1B) 4 CIG (2)
Site amplification 3 HCR (1B) 3 IGN (2)
2 2
1 1
In order to confirm the amplification characteristics, a 1-D
0 0
model for SH wave propagation has been used. For the
0.1 1 10 0.1 1 10
volcanic ashes, the quality factor Q is calculated as 1/2h where Period (s) Period (s)
h is the damping. For the stiff lava the empirical formula
Q=Vs/12 is used (AIJ, 19931). No data is available for the 6
damping of the scarious lava flow, then a value of 10% is 5
4

SA/PGA
assumed.
3 HSH (1B)
The transfer functions are shown in Fig. 9 for the five
2
strong motion stations. The one at station IGN located in Zone
1
2 does not have a higher peak in the longer period components
0
as it was expected. It is worth mentioning that the soil profile in 0.1 1 10
IGN does not reach a stiff lava flow of 2,100 m/s (Fig. 4). The Period (s)
CIG station in Zone 2 shows very short predominant period,
despite it is located in Zone 2. This might be attributed to the Fig. 6 Horizontal acceleration response spectra (SA) for 5%
fact that only the upper part of the soil structure is being used. damping for the upper-crustal earthquake of October
Notice that the two stations are located just inside the 10th, 1986. The largest horizontal component has been
projection of the rupture area for the event of October 10th, used and divided by PGA (see locations in Fig. 3). See
1986, the result motion must have a very big influence of the in parenthesis the Zone in Figure 3.
source (Fig. 3). For the Zone 1B the predominant periods are
between 0.13-0.4s and the amplification factors are between 8- 6 6
10. 5 SJM (1B) 5
For the soil profiles shown in Figure 5, the transfer OBS (2)
SA/PGA

4 4

SA/PGA
UCA (1B)
functions have been calculated using the average of the shear 3 3
wave velocity mentioned before. We simply assume that Tierra 2 2
Blanca (VS=225 m/s) is in the upper part and that the more 1 1
consolidated tuffs (VS=500 m/s) underlay the stiff lava flow 0 0
(VS=2,100 m/s) when they are intercalated each other. This is 0.1 1 10 0.1 1 10
consistent with the geological formation of the San Salvador Period (s) Period (s)
valley. For Zone 2 a very interesting characteristics are
6
observed. For the soil profile P2 the biggest amplification 5 EXT (1B)
corresponds to the second mode (Figure 10), not to the first
SA/PGA

4 REX - top (1B)

mode (1.0 s) as can be expected usually. It is noticed that 3
similar shape is observed in the spectra of CIG station in Zone 2
2 (Fig. 6). For the soil profile 8N9 similar amplification factors 1
are observed in the first modes as the case of the profile ELS2- 0
59 in Zone 1B. It is noticed that a similar shape is observed in 0.1 1 10
the spectra for OBS (Feb. 13th/01) and REX station (Fig. 7 and Period (s)
8). We do not have to forget the limitations of these
calculations since average values of VS were used. Fig. 7 Horizontal acceleration response spectra (SA) for 5%
Two special cases are analyzed also. The first one is the damping for the upper-crustal earthquake of February
profile 9N8 that is located just in the boundary of two Zones 13th 2001. The largest horizontal component has been
(Fig. 3 and 5). It does not correspond to the classification used and divided by PGA (see locations in Fig. 3).
Zones of the microzonation, since cummulative lava flow in the .
6 6
first 100 m is 48 m. Similar characteristics are observed with REX -top (1B)
5 5 OBS (2)
the soil profiles located in Zone 2 (Fig. 11). The second case is EXT (1B)
SA/PGA

4 4
SA/PGA

the profile 2N10 which is outside of the area, but its features fix 3 3
with the Zone 1A, where not discussion have been made since 2 2
the lack of data. Predominant period looks not so short, around 1 1
0.38 s, and the amplification about 2 is due to the volcanic ash 0 0
that is trapped between the lava (Fig. 11). 0.1 1 10 0.1 1 10
In Fig. 12 different structural models using the borehole Period (s) Period (s)
information of HCR station are examined to confirm the
contribution of each layer in them. It can be noticed that the Fig. 8 Horizontal acceleration response spectra (SA) for 5%
amplification characteristics are due to the sediments near the damping for subduction earthquakes. Left: January 13th
stiff lava flow. This might be due to the high impedance ratio 2001, right: June 19th 1982. The largest horizontal
since there is a very high contrast in the shear wave velocity component has been used and divided by PGA (see
between the lava and the volcanic ash. This is the opposite case locations in Fig. 3).
of Mexico City where the upper clay sediments with a VS of 50
m/s are the responsible for the amplification in long period Also the influence of the scoriaceous lava was checked
3
removing it from the analysis in the case of UCA station (see
profile in Fig. 4). It is noticed that when only one layer of this
material is removed (thickness of 3 m), the predominant period
and the amplifications remain almost the same. When also the
16 m of scoria next to the stiff lava flow are removed, the
fundamental period changes from 0.4 s to 0.3 s and the
amplification factor change from 8 to 10. Since we assumed a
damping of 10%, another analysis was made with a damping of
5%, but the results remain the same.
.
100.0
UCA (Zone 1B) 100.0
CIG (Zone 2)
Amplification

HCR (Zone 1B)

Amplification

10.0 IGN (Zone 2)

10.0
1.0
1.0

0.1
0.1
0.1 1.0 10.0
0.1 1.0 10.0
Period (s)
Period (s)
100.0
Amplification

10.0 HSH (Zone 1B)

1.0

0.1
0.1 1.0 10.0
Period (s)

Fig. 9 Site amplification using 1D model SH wave propagation

for borehole data at strong motion. See locations in Fig.
3. Fig. 12 HCR station (Zone 1B). Different structural models
are examined to confirm the contribution of each
100.0 100.0 layer in them. H: thickness in m, VS: shear wave
velocity (m/s), ρ: density (g/cm3), Q: quality factor.
Amplification

P2 (2)
Amplification

ELS2-59 (1B)
10.0 10.0

At the station (REX) there is an instrument in the top and in the

1.0 1.0
bottom of a borehole of 12.5m depth. Spectral ratio suggests
an amplification factor between 3-4 in the period range of 0.2
0.1 0.1
0.1 1.0 10.0
to 0.3 s (Fig. 13).
0.1 1.0 10.0
Period (s) Period (s)
5
100.0 Ratio (top/bottom)
Amplification

4 Jan.13/01
Amplification

8N9 (2)
10.0 Ratio(top/bottom)
3
Feb.13/01
1.0 2

0.1
1
0.1 1.0 10.0 0
Period (s)
0.1 1 10
Fig.10 Site amplification using 1D model SH wave propagation Period (s)
for soil profiles from water wells. See locations in Fig.
3. Fig. 13 REX station (Zone 1B). Ratio between top and bottom
record in a borehole of 12.5 m depth.
100.0
2N10 (outside of zones)
Previous jobs in the area considering non-linearity behavior in
Amplification

9N8 (in the boundary)

10.0 the soil (Salazar, 19999), suggests that weak motions are more
amplified than strong motion due to the damping effect. As it is
1.0 expected, fundamental period increase in a strong shaking.
Then amplification characteristics must be different depending
0.1 on magnitude and distance from the source. The last
0.1 1.0 10.0 earthquakes of January 13th and February 13th 2001 were
Period (s) recorded in San Salvador city outside the microzonation (see
Fig. 14). It can be noticed that the predominant period is
Fig.11 Site amplification using 1D model SH wave propagation always less than 1.0 s (Fig. 15). In some spectra similar shapes
for two special cases. See locations in Fig. 3. are observed with the previous ones. The H/V ratio (Nakamura,
4
1989)5 from microtremor measurement in three strong motion Predicted and observed response spectra
stations are presented in Fig. 16. At ESJO station, the
predominant period is around 0.4 s with an amplification of 6. Spectral attenuation relationships (5% damped) for Central
For the case of HSRF station at Santa Tecla City, the America have been proposed by Dahle et al (1995)2. A total of
predominant period is similar with the record of Feb 13th/01. It 280 records from 72 earthquakes in around 100 stations of El
is noticed that the source of this earthquake is far from this site Salvador, Nicaragua, Costa Rica and Mexico were used to get
and that the level of shaking is very low in comparison with the equations. The Mexican earthquakes were incorporated in
subduction event of January 13th/01. For CSBR station no the analysis to complete the distribution of magnitude-distance
agreement is observed at all. for big earthquakes. In the regression analysis the upper-crustal
and subduction events are not separated. A coefficient for the
site response to distinguish soil and rock site conditions is
included.
The model used is:

ln A = c1 + c2 M W + c3 ln R + c4 R + c5 S (1)

A is the largest horizontal component (pseudo velocity, PSV in

m/s), MW is the moment magnitude, R=(r2+rh2)1/2, where r is
hypocentral distance in km and rh is a constant (fictitious depth)
introduced to ensure constant ground motion in the very near
field. S is zero for rock sites and 1 for soil sites; c1 to c5 are
constants obtained in the regression analysis. In Fig. 17 the
comparison between observed and predicted spectra is
presented. It can be observed that the prediction underestimates
the observed spectral ordinates for subduction and upper-
crustal events.

10 10
EXT HSRF
H/V Ratio

1 1

Fig. 14 New Metropolitan Region by Planning Office of San

Salvador City OPAMSS (streets are shown) and 0.1
strong motion stations outside of the microzonation 0.1 0.2 0.5 1 0.1 0.1 0.2 0.5 1
(triangles). 10
5 CSBR
H/V Ratio

Jan 13/01 HSRF 5

4 Feb 13/01 CSBR
4
Feb 13/01 HSRF Jan 13/01 CSBR
SA/PGA

3 1
SA/PGA

3
2
2
1
1
0
0
0.1 0.1 0.2 0.5 1
0.1 1 10
Period (s) 0.1 1 10
Period (s) Period (s)
5 Oct 10/86 IVU
Jan 13/01 UTON 5 Fig. 16 H/V ratio at three strong motion stations. See their
4 Feb 13/01 UTON Amplification locations in Figure 3 and 14.
Amplification

4
at VDUA
SA/PGA

3 3
2 2

1
1 CONCLUSIONS
0
0
0.1 1 10 Predominant period appears less than 1.0 s. The seismic
0.1 1 10 Period (s)
Period (s) microzonation proposed looks very reasonable for the
subduction and upper-crustal earthquakes. But the strong
Fig. 15 Selected response spectra for 5% damping and motion records and soil profile data are not so many and it is
amplification factor in VDUA (Feb. 13th) at strong difficult to have a definitive opinion. One bad thing is that no
motion stations in San Salvador, outside of the governmental institution is using this kind of job for risk
microzonation (see Fig. 15). In VDUA, there is an mitigation and land use in San Salvador city, not only for
instrument in the top and bottom in a borehole with amplification of motion, also for landslide hazard.
depth about 23 m. The “cumulative stiff lava flow “ seems to be a very
important factor in the amplification characteristics in the area.
Using the soil profiles of water wells from governmental and
private institutions, the microtremors measurements could be
5
 done in these sites in order to confirm the amplification assistance in geographical coordinates conversion in the maps.
characteristics in the whole area of San Salvador. But it seems The GIS digital information covers are coming from MARN
that microtremors might be related only with the upper (2000)4 and the microzonation map was digitized by PRISMA
sediments above the stiff lava. (1997)7. Mario Lungo and Marisol Corcio from Planning
Dahle et al
Office of San Salvador provided us all the GIS data bank for
1600 Dahle et al 1600
(1995)
(1995) San Salvador area. Katsumi Kurita provided us a great help
1200 REX-top (1B) with the source FORTRAN code to get the transfer functions.

Acc. (cm/s/s)
1200
Acc. (cm/s/s)

HCR (1B ) 2001/02/13

1986/10/10 We want to express our thanks for their great help.
800 800

400 400 REFERENCES

0 0
1) AIJ: The Architectural Institute of Japan (1993)
0.1 1 10 0.1 1 10
Earthquake motion and ground conditions.
Period (s) Period (s)
2) Dahle, A., A. Climent, W. Taylor, H. Bungum, P. Santos,
1600 Dahle et al M. Ciudad Real, C. Lindholm, W. Strauch & F. Segura
Dahle et al 1600
(1995)
(1995) (1995) New spectral strong motion attenuation models for
Acc. (cm/s/s)

1200 1200 OBS (2)

EXT (1B) Central America. Proceedings, Fifth International
Acc. (cm/s/s)

2001/01/13 1982/06/19
800 800 Conference on Seismic Zonation, Nice, 17-19 October,
400 400
vol. II, 1005-1012.
3) Faccioli, E., C. Battistella, P. Alemani & Tibaldi (1988)
0 0
0.1 1 10
Seismic microzoning investigations in the Metropolitan
Period (s) 0.1 1 10
Period (s) area of San Salvador, following the destructive earthquake
of October 10, 1986. Proceedings, Int. Seminar on
Fig. 17 Comparison of predicted and observed response spectra Earthquake Engineering, Universitat Innsbruck, Austria,
for upper crustal earthquakes (top) and subduction
28-65.
earthquakes (bottom).
4) MARN (2000) Colección de CD’s Medio Ambiente 2000.
It is desirable to have more detail information of dynamic El Salvador C.A. Ministerio de Medio Ambiente y
properties for deeper soil structures, especially in the Recursos Naturales.
accelerographic stations and in the new developed areas. New 5) Nakamura, Y. (1989) A method for dynamic
strong motion instruments should be installed in Zone 1A. We characteristics estimation of subsurface using microtremor
have noticed that the records do not contain absolute time, then on ground surface. QR of RTI, Vol. 30, No. 1.
a GPS system will be suitable in a new network. Where a lava 6) Press, S. Teukolsky, W. Vetterling & B. Flannery (1992)
flow is found in the surface, we must be very careful to Numerical Recipes in Fortran 77. Second Edition.
consider it as a rock site, since volcanic ash with a considerable Cambridge University Press.
thickness can be trapped between the lava. Santa Tecla City 7) PRISMA (1997) Perfil ambiental de la Región
was severe damaged by the earthquake of January 13th/01, Metropolitana de San Salvador. Sponsored by USAID,
despite that most of the City is on a lava tongue. San Salvador, El Salvador. Programa Salvadoreño de
Special attention should be focused in the buildings with a Investigación sobre Desarrollo y Medio Ambiente.
fundamental period less than 1.0 s. Many RC buildings suffered
8) Harlow, D., R. White, M. Rymer & S. Alvarez (1993) The
heavy damage in the earthquake of October 10th 1986 in
downtown area. They must be repaired as quickly as possible. San Salvador earthquake of 10 October 1986 and its
The differences in the observed spectral ordinates (Figure 17) historical context. Bull. Seism. Soc. Ame. Vol. 83, pp.
partially explain why the damage in the 4-6 story buildings was 1115-1142.
not so severe for the January 13th 2001 event in the capital area. 9) Salazar, W. (1999) Ground motions characteristics in San
A new effort should be done in order to have new attenuation Salvador City, El Salvador (Central America).
relationships for the Central America. The differences between International Institute of Seismology and Earthquake
prediction and observed spectral ordinates are very big. The Engineering, Ministry of Construction, Tsukuba, Japan.
inclusion of the Mexican earthquakes might be an important Vol. 35, pp. 213-225.
factor in the site effect term in the equation. An optimum 10) Salazar, W. & Seo (2001) Earthquakes disasters of January
characterization of strong ground motion can be done 13th and February 13th, 2001. Proceedings of the Annual
proposing bedrock motion estimation together with meeting of the Japanese Seismological Society,
amplification factors, using the data bank from several Kagoshima, JAPAN.
countries of Central America. 11) Schmidt-Thome, M. (1975) The geology in the San
Salvador area (El Salvador, Central America) : a basis for
city development and planning. Geologisches Jarbuch, 13,
ACKNOWLEDGMENT pp. 207-228.
12) Seo, K (1997) JICA research and development program
Central America University UCA provided strong motion on earthquake disaster prevention. The Japan Building
records through INTERNET at http:\\www.uca.edu.sv. We Disaster Prevention Association, Japanese International
would like to thank to Manuel Díaz, Daniel Hernandez and Cooperation Agency (JICA).
Carlos Pullinger from Ministry of Enviroment and Natural 13) UCA (2001) Registros de terremotos de Enero y Febrero
Resources MARN to provide us all data used for the del 2001. Grupo de Desastres Naturales.
government network. Arturo Escalante from UCA and Katia http://www.uca.edu.sv.
Madrid from Centro Nacional de Registros (CNR) help us with
6

View publication stats