indonesia sumatra northern report

8/13/2019 Indonesia Sumatra Northern Report

1/8

1

pec a ar qua e epor ugus

Learning from Earthquakes

The Northern Sumatra Earthquake of March 28, 2005

tor s ote: eg nn ng w e

March issue and in each one sub-sequently, we presented six reports

y e many eams a o serve

the effects of the December 26,

2004, earthquake and tsunami in

coun r es aroun e n an cean.

This report adds information on

structural damage, largely from

shaking, on two islands west of

Sumatra in a subsequent large

earthquake in roughly the same

epicentral area as the December

qua e. u ca on o s repor s

supported by funds from the Na-tional Science Foundation through

s earn ng rom ar qua es

Program under grant # CMS-

0131895.

Within a few days of this earth-

quake, a team set out from Jakarta

o s u y e ear qua e amage on

the islands of Nias and Simeulue.

The team leader and author of this

report was Teddy Boen, consulting

eng neer n a ar aan sen or a v -sor, WSSI. Survey team members

on Nias were Teddy Boen, G. Hen-

ra, . rus, . o n, ames ong,

M. Agus, and B. Bernard. Survey

team members on Simeulue were

e y oen, . en ra, an .

Karsono.

n ro uc on

On Monday night March 28, 2005

at 11:09 pm local time, the Nias and

meu ue s an s were s a en y a

M8.7 earthquake with a depth of

30 km. The epicenter of the earth-

qua e was on t e reat umatran

fault, between the two islands at

2.074N, 97.013E (see Figure 1).

s s m rom unung to ,

the district capital of Nias, and 84 km

from Sinabang on Simeulue. In

unung to an e u a am,

many ouse s ops co apse an

e aroun , peop e w e t ey

were either sleeping or trying to runfrom the second and third oors. In

meu ue, some ouse s ops co -

apse , an a out peop e were

killed. There was a rush to exit

ouses an run to g er p aces e-

cause the people feared a tsunami

similar to that of December 26, 2004.

There was a local tsunami, but not

of the size or ferocity of the previous

one.

as s an s part o ort umatra

Province and Simeulue Island is partof Aceh Province. The population of

as s , an t at o meu ue

is 80,000. Most of the people on Nias

and Simeulue were already asleep,

or a out to e, w en t e eart qua e

occurre . g t a ter t e eart -

quake, electric power was cut offand telecommunication disrupted.

n unung to an e u a am,

as, t e amage was concentrate

in certain areas close to the beach.

n meu ue t e amage was a so

concentrate n areas c ose to t e

beach. Strong aftershocks (32 km

from Gunung Sitoli [M 6.0] and 41

km from Sinabang (M 5.5]) did not

cause any further damage to either

damaged or undamaged buildings.

Tectonic Effects

ur ng t e survey n meu ue as

well as in Nias Island, we observed

some subsidence in certain very

m te areas. ere were a so pop-

Figure 1.

Main shock and

a ers oc s

of the Nias/

Simeulue

ear qua e,

March 28, 2005

(USGS).


2/8

2

peca ar qua e epor ugus

up beaches in Nias and Simeulue,

caused by uplift (see Figure 2). Staffmembers of NGOs who worked in

several places across Simeulue re-

porte t at t ey a so o serve s m -

lar pop-up beaches all over the is-

land.

After the earthquake, a local tsunami

hit limited areas. In Nias, along the

roa enter ng ora e an n ora e

itself, the tsunami height was ap-

proximately 1.80 m, as can be seen

on t e wa o ouse n amaetan -

ha, approximately 1 km from Sorake

(see Figure 3). In Sorake, several

building foundations were scoured

by the tsunami, and some buildings

collapsed. The local tsunami wentn an up to m n amaetan -

a. rom t e amage, t appeare

that the ow force was not too

strong. n mue ue, traces o oca

tsunam cou e seen n a uan

Bajau primary school # 2 (see Fig-

ure 4). The tsunami height was ap-

prox mate y . m an no ev ence

of ow force could be seen.

Damage to Buildings

wo types o u ngs were a ect-ed: engineered and nonengineered.

In Nias and Simeulue, there are few

engineered buildings except for

district head ofces, hospitals, andsome other government buildings.

Engineered buildings consist mostly

o re n orce concrete com ne

with masonry walls. Almost 95% of

the buildings are nonengineered,

ma e o con ne masonry a -

burnt brick, concrete block, and

timber.

ng neere u ngs: Some en-

gineered buildings in Nias and Sim-

eu ue were amage ur ng t e

eart qua e, an some co apse .

There are two main reasons for this.

The reinforced concrete buildings

Figure 2.Pop up beach at Matanurung, East Simeulue.gure . oca sunam e g approx ma e y . m.

gure . amage y oca sunam n a uan a au

school, Simeulue. gure .District head ofce in Simeulue.


3/8

3


were not designed for the seismicity

of the islands. We also found thatpoor qua ty o concrete an eta -

ng contr ute to t e co apses.

Even though Indonesia has a seis-

m c co e, t was not en orce n

e t er s an . t ona y, t e u -

ing permit system was not yet in

place.

For example, both district heads

ofces were damaged (see Figure

. e surmse t at n eac case,

the structure was analyzed as 2-D,

with the torsional effect overlooked.rom t e amage, t was ev ent

that the concrete quality was low

and the detailing was not in accor-

dance with requirements for earth-

quake-resistant construction (seegures an .

Nonengineered Buildings: Most of

t e u ngs were noneng neere

that collapsed in Gunung Sitoli in

Nias, Sinabang City in Simeulue,

and villages along the road from

na ang to a uan a au an rom

Gunung Sitoli to Teluk Dalam in Nias.

There are three main categories of

noneng neere u ngs: one or

two-story houses of burnt brick or

concrete block masonry with sand/cement mortar; two to our-story

shop houses of burnt brick or con-

crete block masonry with sand/

cement mortar; and 3) timber build-

ngs. n some areas, sea san sused for concrete mix and, of

course, results in poor quality

concrete.

The one or two-story buildings were

most y ouses, sc oo s, an com-

munity health centers. The roofs of

the houses mostly consist of gal-

van ze ron s eets or t atc ma e

of coconut tree leaves; other build-

ings use galvanized iron sheets.

ese masonry u ngs surv ve

t e s a ng w t out co apse,

though some had cracks in the

walls (see Figure 8).

Figure 7.Damage to district head ofce in Simeulue.Figure 6.Damage to district head ofce in Nias.

gures a an . oneng neere masonry ouses amage an co apse n as.


4/8

4


e two to our-story s op ouse

buildings also use burnt bricks or

concrete block masonry with sand

an ort an cement mortar. e

roofs of these shop houses are

usually RC slabs. All the buildings

use pract ca co umns anbeams as connement. The term

practical is used because the

specs are not ase on any ca cu-

at ons an are a opte as a stan-

dard practice. The shop houses are

built according to prevailing practice

by contractors and foremen, without

structural analysis or drawings. In

neither Nias or Simeulue was the

u ng perm t system en orce ;

gure . m er ouses co apse n as ue o ac o

ma n enance.

gure . ra ona n genous ouse n as, no

amage .

Figures 11a and b.Traditional indigenous houses damaged due to lack of maintenance in Bawogosali Village, Teluk

Dalam, Nias.

Figure 12.

n genous ousebuilt during Dutch

occupation, slight-

y amage ue o

lack of mainten-

ance in Nias.


5/8

5


t ere ore no structura c ec wasperformed to catch poor-quality

materials or poor workmanship.

Timber houses consist of a timber

frame and timber plank walls, and

ga van ze ron s eet or t atc

roofs. In spite of the closeness to

the epicenter of the earthquake, the

amage was not s gn cant n rea-

sona y u t t m er structures. ery

few one-story timber houses col-

lapsed during the earthquakes. The

damage that occurred resulted fromdeterioration of the poorly main-

tained columns (see Figure 9). In

a t on, some t m er co umns

were not anchored to the founda-

tion. There were no tie beams to

t e t e co umns toget er, so t e

columns moved in different direc-

tions during the earthquake. Some

t m er ouses were ean ng a ter

t e eart qua e ecause t e con-

nections were not appropriately

one.

Two types of indigenous nonengi-

neere u ngs st ex st on as.

rst, t e tra t ona ouses u t

some 200 years ago are earth-

quake-resistant and withstood the

s a ng see gure . en

these traditional buildings were

damaged, the main cause was

eterorat on ue to ac o ma n-

tenance (Figure 11). These types ofu ngs are not u t anymore e-

cause of the lack of expertise in this

generation, and because they are too

cost y to u . e ot er u ngs,

u t y t e utc some years

ago (Figure 12), are also earthquake-

resistant when properly maintained.

Causes of collapse: he main

causes of collapse in the two to four-

story s op ouses were poor mate-

rials and poor workmanship, exacer-

bated by the prevailing procedureso u ng n stages, epen ng on

the availability of funds. The long

intervals between the construction

o var ous e ements resu t n wea

o nts.

rst t e oun at on s u t w t t e

column-reinforcing bars already in

place. If additional funds are avail-

able, the columns are poured. The

size of the columns average 30 x 30

(40) cm with eight reinforcing bars

of diameter 12 mm and stirrups

usua y not proper y ent o mm

at an average distance of 20 cm.

Subsequently, the walls are built.

onstruct on o some u ngs stops

after the walls are completed. The

next stage is constructing the beams

an oor s a s or t e next story w t

splices for the upper-story columns.

Most buildings are in use as soon as

the rst story is completed. Only af-ter severa mont s or years, w en

more funds are available, do the

second-story columns and walls get

u t; t e process o construct ng t e

rst story is essentially repeated.

Some people continue the process

for the third and fourth stories. Col-

umn re n orcements an st rrups are

not sufcient if the structure works

as an open frame, particularly in the

ong tu na rect on.

In the damaged structures weoo e at, t e engt o t e co umn

splices for the upper oors con-

struction was generally insufcient,

an t e concret ng at t e sp ces

was very poor. s resu te n t e

shear failure of the columns, partic-

u ar y at t e o nts; t s was t e

main reason for the pancake or

soft-story collapse of shop houses,

particularly in Nias (Gunung Sitoli

and Teluk Dalam), that resulted in

most of the deaths (see Figures 13,

14 and 15).

For one-story conned masonry

houses, the construction is also

one n stages. rst t e oun at on

is built with column reinforcement

embedded. The column size is 12 x

cm w t our re n orc ng ars o

diameter 8 mm and stirrups of dia-

meter 6 mm at an average distance

Figure 13.Pancake type of collapse of shop houses

that killed many people in Teluk Dalam, Nias.gure . no er panca e co apse n unung

Sitoli, Nias.


6/8

6


of 20 cm. If enough funds

are ava a e, t e co umns

are poured. At a later date,

he walls and the ring

eams are u t an , na y,

he roof is constructed.

e o serve severa s opouse u ngs t at were

built and supervised by the

owners, and did not suffer

amage. e earne t at

hey had been built contin-

uously, not in stages with

ong nterva s cause y

shortages of funds. The

workmanship and mate-

r a s use were aso o

better quality.

rospects for Change

n all areas in Indonesia, a good

eart qua e-res stant es gn eature

is available, namely reinforced con-

crete framing consisting of practical

columns and beams. However, in

many places in both Nias and Sim-

eulue, the columns and beams

were not constructed in accordance

w t preva ng pract ce on ava. n

spite of that, most did not collapse

in the M8.7 earthquake shaking.

uc masonry structures ave e-

come a new culture in Indonesia,

and from past earthquakes it has

ecome ev ent t at t ey are ap-

propriately resistant to the earth-

quake forces identied in the Indo-

nesian seismic hazard map. This new

cu ture s assoc ate w t soc a sta-

tus: as t e r econom c con t on m-

proves, people tend to build masonry

houses at least partly because they

are a sign of status.

The reconstruction period is a good

opportun ty to ntro uce t e u ng

permit system and to start enforcing

the seismic code. Ofcials must start

ma ng qua ty contro part o t e cu -ture in order to improve the perfor-

mance of buildings in earthquakes.

gure .

Cracks onroad in Lasara,

Gunung Sitoli,

as, cause

by lateral

spreading.

Infrastructure Damage

ost o t e amage to n rastruc-

ture occurred to bridges, and there

were ssures along the roads pro-

uce y su s ence an atera

sprea ng. any r ges n ot

Simeulue and Nias were displaced

rom t e r supports rom cm

to cm. any r ge a utments

settled during the earthquake,

thereby damaging bridge ap-

proaches. The differential settle-

ment was jointly caused by lurch-ing, liquefaction, and uneven com-

pact on, part cu ar y c ose to t e

abutments (see Figures 16 and 17).

e ma n e ectr c power generat ng

plant in Gunung Sitoli in Nias was

damaged by the shaking (Figure 18).

rans ormers were s g t y over-

turned (Figure 19) and some gen-

erator foundations settled. But the

ca e networ n t e c ty as we as

along the road from Gunung Sitoli

to Teluk Dalam was not affected,

an a ter t e power p ant was re-

pa re , e ectrc power resume . n

Teluk Dalam, the cable network

was practically intact. The power-

generat ng p ant n meu ue was

not damaged, but some cable

network to Lasikin was damaged

an too a ew ays to restore.

gures a an . oneng neere masonry

shop houses damaged and collapsed due to

ea co umns n agun r , unung o , as.


7/8

7


Figure 17a.Movement of end support, bridge in Kelapa

Street, Gunung Sitoli, Nias.

Figure 17b.Settlement of bridge approach in Teupah

Barat, Simeulue.

Figures 19a and b. Transformers overturned at power generating plant in Gunung Sitoli, Nias.

gures a an . Power generating plant in Gunung Sitoli, Nias.


8/8

8


Figure 21a.

e orme e y

ue o s ear

failure of columnjoints, Simeulue.

gure .

Shear failure of

the column joint,

meu ue.

e water pur cat on p ant n u-

nung Sitoli, Nias, was not dam-

aged, but the piping network was

srupte y t e tsunam see

Figure 20). Distribution was done

by water tank trucks. In Sinabang,

Figure 20.

Damage to drink-

ng wa er p pes

y oca sunam

in Lagundri,

Hiliamaetaniha,

as.

Simeulue, the water puri-

cat on p ant was not act -

vated prior to the earth-

quake because of a change

n t e p p ng system ue to

the widening of the main

road. After the earthquake, water

purication was provided by a

erman , an water was

e vere y tan truc s.

One part of the newly constructed

jetty at Gunung Sitoli, Nias, sank

into the sea, most probably be-cause the piles were not driven

own to t e ar ayer. part rom

that portion, the remaining jetty was

still intact, which was fortunate in it-

se ecause o ow p es a een

used. These should not, in fact, be

used for undersea structures be-

cause t e t c ness o t e concrete

cover is not sufcient to protect the

reinforcing bars from corrosion. In

meu ue, many o t e same type

of piles supporting the jetty were

sheared at the connection betweent e p e ea an t e p es, ecause

the upper parts of the columns were

already deteriorated and weakened

(see Figure 21).

Airport runways at Gunung Sitoli

in Nias and Lasikin in Sinabang

were not amage ; a rp anes were

able to land and take off right after

the earthquake. At Lasikin Airport,

some u ngs co apse , name y

the terminal building (Figure 22), a

storage building, and some airportsta res ent a ouses.

Figure 22. Collapsed Lasikin Airport Terminal

u ng, meu ue.

indonesia sumatra northern report

Documents