Vous êtes sur la page 1sur 13

13

th
World Conference on Earthquake Engineering
Vancouver, B.C., Canada
August 1-6, !!"
#a$er %o. 1&'
HOW PERUVIAN SEISMIC CODE GREATLY IMPROVED BUILDING RESPONSE TO REAL
EARTHQUAKES
Javier PIQUE
1
, Peer MARTEL
!
SUMMARY
Actual response of school buildings, with almost identical architecture and configuration, subjected to a
moderate to severe shaking was analyzed. A group of these was designed with 1977 eruvian !tandard
while another was designed with 1997 updated !tandard. "hose designed with the former 1977 #ode
suffered damage while the ones designed with modern 1997 !tandard did not e$perience damage
whatsoever. "his evidences eruvian practice has achieved real protection of essential and other building
structures due to an strategy based on drastically limiting displacements. Although allowable displacements
are similar to worldwide practice, computed displacements are large based on large % values.
INTRODUCTION
After &' years, in 1997 the eruvian "echnical #ommittee for !eismic %esistant (esign issued a new
standard which profited from e$periences from earth)uakes worldwide between 19*+ and 199,. -ne of its
key features was the increase on computed lateral displacements .over / fold0. 1uilding structures so
designed became much more rigid than those obtained using specifications from the former 1977 code.
"his paper presents a comparison of the seismic response of actual state school buildings to the 2une &1
st
,
&''1 Atico .Are)uipa0 earth)uake .3w4*,50 in southern eru. "his structures are located at the same
region thus assuming were subjected to the same shaking, with identical architectural distribution .& story0
in which the only variable is the seismic design code and more precisely the procedure for calculating
lateral displacements. 6ield information was collected after the earth)uake, dynamic analysis were
performed. 7t could be observed that buildings designed using the higher displacement re)uirements from
the 1997 code evidence no damage al all whereas those designed using the former code, with re)uirements
similar to world standards had short column failures almost widespread in spite of being separated from
partition walls. "his strongly suggest the current eruvian !tandards produces buildings that fully respond
to the re)uirements for essential buildings, that is to maintain operation after a strong earth)uake, limiting
damage and precluding collapse.
EARTHQUAKE MOTION
1
rofessor of #ivil 8ngineering, eru 2apan #enter for 8arth)uake 8ngineering %esearch and (isaster
3itigation, #7!37(, 6aculty of #ivil 8ngineering, 9ational :niversity of 8ngineering, ;ima, eru.
resident of the eruvian !eismic #ode #ommittee. 8mail< jpi)ue=uni.edu.pe
&
%esearch assistant, eru 2apan #enter for 8arth)uake 8ngineering %esearch and (isaster 3itigation,
#7!37(, 6aculty of #ivil 8ngineering, 9ational :niversity of 8ngineering, ;ima, eru. 8mail<
petermargas=hotmail.com
Ear"#$a%e i& S'$"er& Per$ (Are#$i)a Re*i'&+ '& J$&e !,r- !..1
>"he earth)uake occurred in Are)uipa on 2une &/rd &''1 at 1+ hours // minutes local time caused
important damages in (epartments of Are)uipa, 3o)uegua and "acna and to the cities of Arica and
7)ui)ue in #hile. 9ational 7nstitute for #ivil (efense .79(8#70 indicates the earth)uake produced 75 dead
people, in addition to ,5 dissapeared, &,,*9 injured and &17,59+ homeless in the whole affected region.
eople disappeared due to the effects of a tsunami caused by the earth)uake and was responsible for the
destruction of around &''' housing units at beaches of #amana city?. @amudio A1B
Ceophysical 7nstitute of eru .7C0 located the epicenter of this earth)uake at coordinates ituto CeofDsico
del erE .7C0 1,.&'F !, 7/.7+F GH that is at *& km 9G of the town of -coIa, closer to Atico, as shown in
6igure 1.
/i*$re 10 /'1a2 Me1"a&i34 a&- 3re33 'rie&ai'&0 Ear"#$a%e '5 J$&e !,
r-
, !..10 Are#$i)a, Per$0
9ational !eismic 9etwork registered a total of 1/5 aftershocks in the first &5 hours, many of them
producing intensities between 777 and J in 3odified mercallu !cale .330 in the city of Are)uioa, which is
the largest city in southern eru. .1 ''' ''' inhabitants0.
I&e&3iie3 a&- 3ei34i1i6 '5 "e re*i'&
eru 2apan #enter for 8arth)uake 8ngineering %esearch and (isaster 3itigation .#7!37(0 of the 6aculty
of #ivil 8ngineering of the 9ational :niversity of 8ngineering .:970 maintains a network of
accelerographs having registered the main event in the 3o)uegua !tation, .a city south of Are)uipa, around
5''km away from the epicenter0. eak acceleration value obtained for the 8KG direction was &9+./ gals, in
the 9K! direction &&' gals and 1,'., gals in the vertical component. 7n 6igures & and / the registered
accelerographs are plotted.
>(amage produced by this seismic event showed once again that structures built which do not follow
technical standards and are built without technical supervision, with poor workmanship and low )uality
materials will have a high level of seismic vulnerability?, #7!37( A&B.
/i*$re !0 A11e2erai'& Re1'r- i& M'#$e*$a, Per$0 E7W C'4)'&e&0 Ear"#$a%e '5 J$&e !,r-, !..1
Are#$i)a0 (M89:0;+ PGA9 !<=0, *a230
/i*$re ,0 A11e2erai'& Re1'r- i& M'#$e*$a, Per$0 N7S C'4)'&e&0 Ear"#$a%e '5 J$&e !,r-, !..1
Are#$i)a0 (M89:0;+ PGA9 !!.0. *a230
COMPARISON O/ PERUVIAN 1<>> AND 1<<> SEISMIC STANDARDS
A22'8a?2e Di3)2a1e4e&3
-ne of the important changes between 1977 and 1997 was displacement control. "his was done in two
ways. "hrough a substantial increment of computed displacements and through a reduction of the
allowable drift. 7n 1997 separate limitations were introduced for different structural materials.
8n la 1977 standard the ma$imum interstory displacement was '.'1 .1L1''0 of story height, in the case
where nonstructural elements could be damaged. 7t that was not the case and additional +'M was allowed,
rising up to '.'1+ .1L,,0.
8n "able 1 a comparison of drift control in both standards is presented. 7t can be observed the 1977
!tandard allowed a greater fle$ibility. Although limits allowed in 1997 !tandard are not far from current
practice worldwide, it meant a significant improvement in terms of separate limits as a function of material.
6or most buildings which in eru are made of reinforced concrete the increased was 5/M. ;imits for
masonry structures was more stringent based on local e$perience since brick walls crack al very low
distortions .1L1'''0.
Ta?2e 1 Dri5 2i4iai'&3 i& 1<>> a&- 1<<> Per$via& Sei34i1 Sa&-ar-3
Ba3e S"ear
7n 1977 !tandard base shear was computed through the classical e$pression<
P
R
C S U Z
H
d

where %d intended to represent ductility factor, and where


5 . ' 1, . ' ,
1
*' . '

,
_

+
C
T
T
C
s
7n 1997 !tandard a similar e$pression was used, but this time % was considered to be only a %esponse
reduction factor.
P
R
C S U Z
V

where the spectral amplification function + . & + . &


&+ . 1

,
_

T
T
C
p
!ince changes were introduced in the @ and # coefficients to reflect international tendencies. @ was to
reflect effective peak ground acceleration of the zone and # was to reflect response amplification
7n "able & a comparison of seismic parameter for standard buildings and firm soil is presented.
Ta?2e ! C'4)ari3'&3 ?e8ee& ?a3e 3"ear 5a1'r3 i& 1<>> a&- 1<<> Sa&-ar-3
Seismic Standard 1977 1997
Z factor 1 0.4
U factor 1 1
S factor 1 1
C factor (for short periods) 0.4 2.5
ZUCS 0.4 1
7t can be observed that if % values did not change it would result in a force larger for a factor 1L'.54&,+.
Nowever an important consideration decided by the !tandards #ommittee in 1997 was to maintain the
force level at the same values of the 1977 !tandard. As can be seen from "able &, change in these
coefficients meant a &,+ fold increased in the forces. 1ecause of that the #ommittee decided to multiply the
% factor by &.+ in order to maintain base shear at the same level. %d was renamed, %, %esponse %eduction
6actor, and including the ductility and overstrength factors. A comparison of these values is shown in "able
/.
As it can be seen % factors appear to be rather large, but it must be kept in mind that they include as
adjusting factor and do not represent e$pected ductility only.
Ta?2e ,0 Re3)'&3e Re-$1i'& /a1'r i& a&- 1<<> Per$via& Sei34i1 Sa&-ar-3
EA()*+,A-E (E./.)A%)
.)(,C),(A0 .1.)E2
1334 .tandard
(es$onse
(eduction
Coefficient (
1344 .tandard
5uctilit6 7actor
(d
1997 Standard 1977 Standard Increased limitation
!"#$%I&'&(
%'("!I')
,
_


i
I
he

,
_


i
I
he
1'' 1
97
77

,
_

!einforced concrete 0.007 0.010 4*+


Steel (,) 0.010 0.015 50+
%asonr- 0.005 0.010 100+
(im.er 0.010 0.015 50+
Steel frames 10 /
!einforced Concrete frames 10 5
#0al s-stems 10 4
!einforced Concrete Shear 1alls 7.5 *
!einforced masonr- / 2.5
(im.er constr0ction 7 4
Di3)2a1e4e& S)e1ra
6or displacement computation this change in % factor meant an important increase in calculated values.
1977 !tandard specified that actual displacements was to be estimated multiplying computed values times
O %d. -n the other hand the 1997 !tandard specified that actual displacements were to be computed
multiplying analysis values time % values. 1y considering a factor of 1% instead of /L5% an additional
increment was included. Actual displacement corresponds to an inelastic behavior
xR
analysis from computed actual

/i*$re ;0 A1$a2 -i3)2a1e4e& 1'4)$ai'&0
7n the 1997 !tandard, design displacement spectrum is as follows<
g
R
ZUSC
S S
a d & &
1 1


g
R
T ZUST
g
T
T
R
ZUS T
S
P
p
d &
7+ . ' &+ . 1
&+ . 1
&
&
, . 1
+ . &
5

1
1
]
1

,
_

7n the 1977 !tandard, design displacement spectrum was as follows<


g
R
ZUSC
S S
d
a d & &
1 1


g
T T R
T ZUST
g
T
T R
ZUS T
S
s d
S
s
d
d
0 .
&' . '
1
*' . '
5
&
&
&
&
+

1
1
1
1
1
]
1

,
_


7n 6igure + a comparisons of (isplacement !pectra corresponding to both !tandards 1977 and 1997 is
presented for the case of school buildings where importance factor :, has also been increased, from 1,/ to
1,+. 1oth graphs are for !oil rofile 7, !41.
8lastic 1ehavior
7nelastic 1ehavior
P earth)uake4
P4@:!#
P design 4
P4@:!#L%
actual
analysis
%
.;ateral
(isplacement0
P 4 1ase shear
Ta?2e ;0 Sei34i1 Para4eer3 5'r -i3)2a1e4e& 3)e1r$4 1'4)ari3'&, 1<>> a&- 1<<> Sa&-ar-3
/i*$re =0 C'4)ari3'& '5 Di3)2a1e4e& S)e1ra, 1<>> Sa&-ar- (N7>>+ a&- 1<<> Sa&-ar- (E.,.7<>+0
B$i2-i&*3 Cae*'r6 @BA
7n 6igure , a similar comparisons of (isplacement !pectra is presented but including the soil factor for the
three soil profiles considered in both standards. corresponding to both !tandards 1977 and 1997 is
presented for the case of school buildings where importance factor :, has also been increased, from 1,/ to
1,+. 1oth graphs are for !oil rofile 7, !41.
Seismic arameter C2'!'C("!IS(IC 1344 .tandard 1334 .tandard
Z$&" er03ian Coast 1 *
Zone 4actor (Z) 1.0 0.4
Importance 4actor or Use (U) Cate5or- 6 1.* 1.*
arameter de soil (S) !oc7s or 3er- ri5id soils 1.0 1.0
arameter de soil ((p) !oc7s or 3er- ri5id soils 0.* 0.4
/i*$re B0 I&52$e&1e '5 S'i2 Pr'5i2e i& Di3)2a1e4e& S)e1ra, 1<>> Sa&-ar- a&- 1<<> Sa&-ar-0
B$i2-i&*3 Cae*'r6 @BA0
7n 6igure 7 the ratio between spectral displacement in 1997 standard and 1977 !tandard is given for all
three soil profiles. 7t can be appreciated that displacement demand has increased more than / fold for short
period structures. "his comes from two sources. 7ncrement in the % factor, which as e$plained before was
multiply time &.+. And from the specification that actual displacements are obtained multiplying
displacements from structural analysis times full % instead of O. 1y comparison 1997 !tandard will
produce displacements larger in &.+ $ 5L/4 /.// times. Additionally this is to be compared with
displacement limitations which , as shown in "able 1, have also increased by a minimum of 5/LM.
As a result structural designs based on the 1997 eruvian !tandard will produce much more rigid
structures than with former 1977 !tandard. Actual school buildings so designed have performed e$tremely
well in 2une &1 &''1 earth)uake with no damage while similar structures designed according to the former
#ode have suffered widespread damage. !hould this be the way to follow to give real protection to
essential structuresQ. "his e$perience seems to indicate this is the correct approach.
BUILDING RESPONSE
Sr$1$re3 i& 3$-6
7n the last decade of the RR century a large program of school building was carried out by the eruvian
Covernment. "housand of standardized pavilions were built all over eru. 3ost two and three story
buildings. 7n 1997 a new version of eruvian !eismic !tandards was released, after &' years, this meant
new more rigid designs were necessary due to increased re)uirements.
/i*$re > C'4)ari3'& i& Di3)2a1e4e& S)e1ra, 1<>> Sa&-ar- a&- 1<<> Sa&-ar-0 B$i2-i&*3
Cae*'r6 @BA0
7n order to compare the response to the same earth)uake of these structures which are similar in
architecture but with different in structure three schools designed with 1977 !tandards and two designed
with 1997 !tandard were chosen. All located in the same region. "his way, the main variable in the
response was the design standard. 3artel A,B
Ta?2e =0 B$i2-i&* 31"''23 -e3i*&e- 8i" 1<>> 3ei34i1 3a&-ar-
.chool 0ocation 1ear of
construction
Ch0carapi Isla-8 're90ipa 199*
:os; %. %orante $co<a8 're90ipa 1994
San '50st=n de 20nter 20nter8 're90ipa 1994
"wo school buildings built after the 1997 !eismic !tandard were chosen with similar architecture to the
older ones.
Ta?2e B0 B$i2-i&* 31"''23 -e3i*&e- 8i" 1<<> 3ei34i1 3a&-ar-
7n 6igure 7 location of these buildings in the Are)uipa %egion is shown.
.chool 0ocation 1ear of
construction
C. ". &> 40199 Soca.a-a8 're90ipa 1999
C. ". &> 40052 ? 60enos 'ires de Ca-ma Ca-ma8 're90ipa 1999
/i*$re > Are#$i)a Re*i'& i& 3'$"er& Per$ 8i" 2'1ai'& '5 31"''2 ?$i2-i&*3 i& "i3 3$-60
!tructures shown in 6igure * .1977 !tandard0 are built with reinforced concrete frames in the longitudinal
direction, although, as can be seen in the plan view, columns are oriented with the strong a$es in the
transverse direction, because of gravity load framing. "his is a fle$ible direction.
7n the transverse direction seismic resistance is provided by solid confined masonry, being this a very stiff
system.

/i*$re : E2evai'& a&- 5ir3 52''r )2a& '5 31"''2 ?$i2-i&*3
-e3i*&e- 8i" 1<>> Sa&-ar-0
!chools shown in 6igure 9 were designed with 1997 !tandard and columns have now small flanges in the
longitudinal direction giving the structure a higher stiffness. 7n the transverse direction the seismic
resistance system is the same based on masonry walls.
/i*$re < /ir3 a&- 3e1'&- 52''r )2a& '5 31"''2 ?$i2-i&*3 -e3i*&e- 8i" 1<<> Sa&-ar-0
N'e T 3"a)e 1'2$4&3 8"i1" )r'vi-e 2ar*er 3i55&e33 i& 2'&*i$-i&a2 -ire1i'&
A1$a2 Sei34i1 Re3)'&3e ' !..1 Ear"#$a%e
Schools built under !"" Seismic Standard
All these schools e$perienced the &1 2une &''1 earth)uake. A22 those designed with the 1977 standards
suffered damage. 7n most cases related to short column behavior, in spite of a seismic separation specified
in blue prints and actually built. 8vidence of e$cessive lateral displacement could be observed in most
cases, although designs comply with the current !tandard at the time. As seen all in other seismic .areas
damage is related to e$cessive deformation. hotographs shown in 6igure 1' shown the type of damage
observed in these schools.
Schools built under !!" Seismic Standard
9one of the schools in the region designed and built under the 1997 !tandard suffered damage at all.
!tiffness was larger enough as to make seismic separation unnecessary. 6igure 11 shows post earth)uake
condition of these schools.
/i*$re 1.0 Da4a*e ' ?$i2-i&* -e3i*&e- $3i&* 1<>> Sa&-ar-0
8ven when peak ground acceleration must have been higher than design acceleration .record showed above
with almost o./g was registered 1''km south from Are)uipa region where schools were located, even
further from epicenter0 , there was no damage and the schools continue to operate unharmed.
/i*$re 110 P'3 ear"#$a%e 1'&-ii'& '5 31"''2 ?$i2-i&* -e3i*&e- $3i&* 1<<> Sa&-ar-0
Di3)2a1e4e&3 A&a263e3
!chool buildings are considered essential structures in the 1997 !tandard, the importance factor was
increased in relation to 1977 !tandard. 1esides this fact no other changes were introduced in the
computation for base shear. 3artel A,B.
A comparison of computed relative story displacements from a spectral dynamic analysis in the longitudinal
direction is shown in "able *. 7t can be seen than displacements of school under 1997 !tandard have been
reduced to almost one third of the older 1977 structure.
Ta?2e >0 C'4)ari3'& '5 )ara4eer3 5'r 31"''2 ?$i2-i&*30 N'e a 1=C 1"a&*e i& ?a3e 3"ear0
1977 !tandard 1997 !tandard
@ 1 '.5
: 1./ 1.+
! 1.5 1.5
# .mS$ for short '.5 &.+
periods0
%eduction 6actor %d 4 5 % 4 1'
1ase !hear N 41*.&M , or P 4 &1M
Ta?2e :0 Rai' (1<<>D1<>>+ '5 1'4)$e- i&er3'r6 -i3)2a1e4e&3 i& 2'&*i$-i&a2 (E+ -ire1i'&
!tory (iaphragm ;oad R 1977 R 1997 R 1997LR 1977
& (& !R '.'&*9* '.'1/&* '.5+*'7+
1 (1 !R '.'51*5 '.'1+/, './,711/
CONCLUSIONS
Sei34i1 Sa&-ar-3
#hange in eruvian !eismic !tandards resulted in higher computed lateral displacements. Around three
times that specified in the older 1977 !tandard. !tructures designed using 1997 new !tandard have to be
much more rigid than before.
A&a263i3
!pectral dynamic analysis performed on all school building structures showed a reduction to one third in
computed displacements.
Re3)'&3e
All buildings designed with 1977 !tandard e$perienced structural and nonstructural damage. 9one of the
schools in the region designed and built under the 1997 !tandard suffered damage. !tiffness was larger
enough as to make seismic separation unnecessary. 8ven when peak ground acceleration must have been
higher than design acceleration .record showed above with almost o./g was registered 1''km south from
Are)uipa region where schools were located, even further from epicenter0 , there was no damage and the
schools continue to operate unharmed. "herefore seismic response improved greatly with 1997 !tandard.
C'33
#hanges in structural element dimensions to achieve additional stiffness increase costs by 1+M. 9o cost
was involved after the earth)uake because of absence of damage. Nowever structures designed with 1977
!tandards had to be repaired, they could not be used for several months and cost of retrofitting an stiffening
reach up to 5'M of initial cost.
#learly designing and building stiffer structures has resulted in a much better performance, no damage and
continuity of service. eruvians have achieved what was always e$pected of seismic standards.
RE/ERENCES
1. @amudio J, Paldivia 7. >7ntensities 8valuation in #araveli and #amana .Are)uipa0 rovinces due to
the 2une &/rd. &''1 8arth)uake?. 9ational #enter for Ceophysical (ata.. Ceophysical 7nstitute of eru.
;ima. eru. &''1 .in !panish0
&. eruK2apan #enter for 8arth)uake 8ngineering %esearch and (isaster 3itigation, #7!37(, 6aculty
of #ivil 8ngineering, 9ational :niversity of 8ngineering, >reliminary %eport of the 1st. Pirtual #ongress
Are)uipa 8arth)uake of 2une &/rd 8arth)uake &''1T A year later. ;earned ;essons? . ;ima T eru .in
!panish0
/. 3inistry of "ransportation, #ommunications and Nousing. 9ational "echnical !tandard 8K'/'
>8arth)uake %esistant (esign?. ;ima, eru. 1997 .in !panish0
5. 3inistry of Nousing, >1asic !tandard for 8arth)uake %esistant (esign?. ;ima, eru, 1977 .in
!panish0
+. 3uIoz, A. >(esign 6orces and (isplacement #ontrol in the eruvian !tandard for 8arth)uake
%esistant (esign?. 8l 7ngeniero #ivil. 199*. .in !panish0
,. 3artel, . >#omparison of !eismic %esponse of !chool 1uildings designed with 1997 and 1977
!tandards subjected to the &''1 Are)uipa %egion earth)uake? "hesis to fulfill re)uirements to the of #ivil
8ngineer degree. 6aculty of #ivil 8ngineering. 9ational university of 8ngineering. ;ima. eru.&''5 .7n
!panish0