Transcript
Journal of Earthquake and Tsunami, Vol. 1, No. 1 (2007) 21–47c World Scientific Publishing Company THE SUMATRAN FAULT ZONE — FROM SOURCETO HAZARD DANNY HILMAN NATAWIDJAJA Research Center for Geotechnology, Indonesian Institute of Sciences (LIPI)LIPI – Bld.70, Sangkuriang Street, Bandung, 40135, Indonesia WAHYU TRIYOSO Department of Geophysics & Meteorology, Bandung Institute of Technology (ITB)Ganesha Street, Bandung, Indonesia The substantial portion of the dextral component of the Sumatran oblique convergenceis accommodated by the Sumatran fault. This 1900km-long active strike-slip fault zoneruns along the backbone of Sumatra pose seismic and fault hazards to dense populationon and around the fault zones. The Sumatran fault is highly segmented, and consistsof 20 major geometrically defined segments, which range in length from about 60 to200km. These segment lengths influenced seismic source dimensions and have limitedthe magnitudes of large historical fault ruptures to between M w 6.5 and about 7.7. Sliprates along the fault increase northwestward, from about 5mm/yr around the SundaStrait to 27mm/yr around Toba Lake. These sliprate values provide a quantitativebasis for calculation of average expected recurrence periods for large earthquakes oneach segment. Deterministic and probabilistic hazard assessments are constructed basedon these active fault data. 1. Introduction — Plate-Tectonic Environment The island of Sumatra sits atop the Southeast Asian plate, which overrides the sub-ducting Indian and Australian oceanic plates that converges obliquely at about 50to 70mm/yr [Prawirodirdjo et al. , 2000].The oblique convergenceis partitioned intotwo components: the dip slip is accommodated on the subduction interface, and thestrike-slip component is accommodated largely by the Sumatran fault [McCaffrey,1992; Sieh and Natawidjaja, 2000]. Other strike-slip faults that occur in similarsettings include the left-lateral Philippine fault (which is parallel to the Luzon andPhilippine trenches), the right-lateral Median Tectonic Line (which is parallel tothe Nankai trough, Japan), and the Atacama fault (which lies parallel to the SouthAmerican trench, offshore Chile).The 1900km long Sumatran fault zone (SFZ) traverses the back-bone of Sumatra, within or near the active volcanic arc. [Sieh and Natawidjaja, 2000].At its northern terminus, the Great Sumatran fault zone transforms into thespreading centers of the Andaman Sea [Curray and others, 1979]. At its south-ern end, around the Sunda Strait, the fault curves southward toward and possibly 21 22 D. H. Natawidjaja & W. Triyoso Fig. 1. The Great Sumatran fault zone is a trench-parallel, right-lateral strike-slip fault. It tra-verses the hanging-wall block of the Sunda trench from the Sunda Strait to the spreading centersof the Andaman Sea. Solid arrows relative plate motions from NUVEL-1 [Larson et al. , 1997].Ellipsoids along the fore-arc regions are the seismic sources of the recent-megathrust earthquakes. intersects the Sunda trench [Le Pichon et al. , 1981; Diament et al. , 1992; Sieh andNatawidjaja, 2000] (Fig. 1).The Sumatran fault zone poses major hazards, particularly to the highlypopulated areas on and around the active fault trace. More than a dozens largeearthquakes have occurred historically in the past 200 yeas. They caused greatloss of life and property. However, until now this natural disaster threats have notbeen considered seriously for hazard-mitigation acts and for spatial planning andbuilding codes. 2. Active Fault Mapping The grossest features of the Sumatran fault have long been known from analysisof small-scale topographic and geologic maps [Katili and Hehuwat, 1967]. Moredetailed small-scale maps of the fault, based upon analysis of satellite imagery,have been produced more recently [Bellier et al. , 1997; Detourbet et al. , 1993]. The Sumatran Fault Zone 23 The unavailability of stereographic imagery, however, limited the resolution andthe reliability of these small-scale maps. To be of use in seismic hazard assessmentsthe active fault map must be constructed on a scale that is large enough to clearlydiscriminate fault strands, changes in strike and structural discontinuities betweenstrands.Sieh and Natawidaja [2000] mapped the Sumatran fault zone based primarilyupon inspection of its geomorphic expression on 1:50,000-scale topographic mapsand 1:100,000-scale aerial photographs. These data were digitized and attributed,using the Geographic Information System (GIS) software. Geomorphic expressionis especially reliable for mapping high slip rate faults, where tectonic landformscommonly develop and are maintained at rates that exceed local rates of erosionor burial [Yeats et al. , 1997, Chapter 8]. Examples of geomorphologically basedregional maps of active faults include active fault maps of Japan, Turkey, China,Tibet, and Mongolia [Research Group for Active Faults, 1980; Saroglu et al. , 1992;Tapponnier and Molnar, 1977] as well as most maps of submarine active faults.However, the geomorphic expressions of active faults with slip rates that are lowerthan or nearly equal to local rates of erosion or burial is likely to be obscure. This isespecially likely if the faults are short, have small cumulative offset, or have no com-ponent of vertical motion. Because of our reliance on geomorphic expression, theirmap of the Sumatran fault undoubtedly excludes many short, low-rate active faultstrands. The 1:1,000,000 scale version of their map can be viewed and downloadedfrom http://www.tectonics.caltech.edu/sumatra/sumatranfault.htm. The SFZ ishighly segmented. These fault segments are separated by more than a dozen dis-continuities, ranging in width from ∼ 4 to 12km [Sieh and Natawidjaja, 2000].Theoretically, these discontinuities and bends in the fault are large enough to influ-ence the seismic behavior of the fault [Harris et al. , 1991; Harris and Day, 1993]. Infact, the ends of historical earthquake ruptures along the SFZ seem to have beenconstrained by these large fault stepovers [e.g. Natawidjaja et al. , 1995].Sieh and Natawidjaja [2000] divide the Sumatran fault into 20 major segmentsincluding Batee fault ranging in length from 35km to 200km (Fig. 2 and Table 1).For systematic and consistency, they named each segment by the name of a majorriver or bay along the segment. In older literatures, names of fault segments derivedvariously from nearby cities, districts, basins, and rivers. These include BandaAceh Anu, Lam Teuba Baro, Reuengeuet Blangkejeren, Kla-Alas, Ulu-Aer, Batang-Gadis, Kepahiang-Makakau, Ketahun, Muara Labuh, and Semangko [e.g., see Katiliand Hehuwat, 1967; Cameron et al. , 1983; Durham, 1940]. Sieh and Natawidjaja[2000] also mapped an active thrust fold belt called the Toru active thrust-fold beltwest of SFZ around the 1N latitude (Fig. 2).Historical earthquake records along the Sumatran fault suggest that faultstepovers constrained past fault ruptures and hence are likely to govern future rup-ture dimensions. An excellent example is mechanisms of the 1926 and 1943 eventsin west Sumatra (Fig. 3). The 1926 event is a double-mainshock event. The firstshock ruptured the segment between Dibawah and Singkarak lakes. Then, about 24 D. H. Natawidjaja & W. Triyoso Fig. 2. This map of the principal active traces of the Sumatran fault zon (SFZ) is based uponinterpretation of 1:100,000 stereographic aerial photographs and 1:50,000 topographic maps. TheSFZ can be divided into 20 fault segments. Ends of segments are mostly major fault stepovers of 4km-width or more of separations. The Sumatran Fault Zone 25 T a b l e 1 . F a u l t S e g m e n t a t i o n o f t h e S u m a t r a n F a u l t Z o n e t h a t c o n t r o l s t e r m i n a t i o n s o f e a r t h q u a k e r u p t u r e s [ m o d i fi e d f r o m S i e h a n d N a t a w i d - j a j a , [ 2 0 0 0 ] . L o c a t i o n H i s t o r i c a l M a g n i t u d e S l i p R a t e S l i p R a t e L e n g t h E a r t h q u a k e s b y G e o l . b y G P S S e c t i o n I n d e x # Y 1 Y 2 ( k m ) Y e a r ( M ) G e o m o r p h i c F e a t u r e s M 1 M a x M 2 M a x m m / y r m m / y r 1 S u n d a 1 S − 6 . 7 5 − 5 . 9 1 5 0 N o n e - b u t m a n y s u b m a r i n e g r a b e n 7 . 6 7 . 7 n / a n / a r e c e n t M 4 - 6 2 S e m a n g k o 1 R − 5 . 9 − 5 . 2 5 6 5 1 9 0 8 e a s t f a c i n g s c a r p 7 . 2 7 . 2 n / a n / a 3 K u m e r i n g 1 Q − 5 . 3 − 4 . 3 5 1 5 0 1 9 3 3 ( M s = 7 . 5 ) ; S u o h g e o t h e r m a l v a l l e y 7 . 6 7 . 7 n / a n / a 1 9 9 4 ( M w = 7 . 0 ) 4 M a n n a 1 P − 4 . 3 5 − 3 . 8 8 5 1 8 9 3 m o u n t a i n o u s r a n g e o n 7 . 3 7 . 4 n / a n / a e a s t s i d e o f t h e f a u l t 5 M u s i 1 O − 3 . 6 5 − 3 . 2 5 7 0 1 9 7 9 ( M s = 6 . 6 ) v a l l e y , d e p r e s s i o n 7 . 2 7 . 3 1 1 n / a 6 K e t a u n 1 N − 3 . 3 5 − 2 . 7 5 8 5 1 9 4 3 ( M s = 7 . 3 ) ; d e p r e s s i o n v a l l e y a n d 7 . 3 7 . 4 1 1 n / a 1 9 5 2 ( M s = 6 . 8 ) K a b a v o l c a n o 7 D i k i t 1 M − 2 . 5 − 2 . 4 6 0 n o r e c o r d n / a 7 . 2 7 . 2 1 1 n / a 8 S i u l a k 1 L − 2 . 2 5 − 1 . 7 7 0 1 9 0 9 ( M s = 7 . 6 ) ; L a k e K e r i n c i a n d 7 . 2 7 . 3 1 1 2 3 1 9 9 5 ( M w = 7 . 0 ) K u n y i t v o l c a n o 9 S u l i t i 1 K − 1 . 7 5 − 1 9 5 1 9 4 3 ( M s = 7 . 4 ) s m a l l d e p r e s s i o n , c a l d e r a s 7 . 4 7 . 4 1 1 2 3 ± 5 a n d y o u n g v o l c a n i c c o n e 1 0 S u m a n i 1 J − 1 − 0 . 5 6 0 1 9 4 3 ( M s = 7 . 6 ) ; L a k e D i a t a s , c a l d e r a s 7 . 2 7 . 2 1 1 2 3 1 9 2 6 ( M s ∼ 7 ) a n d T a l a n g v o l c a n o 1 1 S i a n o k 1 I − 0 . 7 0 . 1 9 0 1 9 2 6 ( M s ∼ 7 ) L a k e S i n g k a r a k 7 . 3 7 . 4 1 1 2 3 ± 3 1 2 S u m p u r 1 H 0 0 . 3 3 5 n o r e c o r d w i d e d e p r e s s i o n a s s o c i a t e d 6 . 9 6 . 9 n / a n / a w i t h n o r m a l f a u l t s 1 3 B a r u m u n 1 G 0 . 3 1 . 2 1 2 5 n o r e c o r d l o n g ( S u m p u r ) v a l l e y 7 . 5 7 . 6 n / a 4 a l o n g t h e f a u l t 1 4 A n g k o l a 1 F 0 . 3 1 . 8 1 6 0 1 8 9 2 ( M s = 7 . 7 ) m o u n t a i n o u s r a n g e s o n 7 . 6 7 . 7 n / a 1 9 ± 4 b o t h s i d e s o f t h e f a u l t 1 5 T o r u 1 E 1 . 2 2 9 5 1 9 8 7 ( M s = 6 . 6 ) u p l i f t e d h i l l o n t h e e a s t 7 . 4 7 . 4 n / a 2 4 s i d e o f t h e b e n d 1 6 R e n u n 1 D 2 3 . 5 2 2 0 1 9 1 6 ; 1 9 2 1 ( m b = 6 . 8 ) ; T a r u t u n g V a l l e y 7 . 8 7 . 9 2 7 2 6 ± 2 1 9 3 6 ( M s = 7 . 2 ) 1 7 T r i p a 1 C 3 . 4 4 . 4 1 8 0 1 9 3 6 ( M s 7 . 2 ) ; A l a s V a l l e y 7 . 7 7 . 8 n / a n / a 1 9 9 0 ( M s = 6 ? ) 1 8 A c e h 1 A 4 . 4 5 . 4 2 0 0 n o r e c o r d m o u n t a i n o u s r a n g e , 7 . 7 7 . 9 n / a n / a a s s o c i a t e d w i t h t h r u s t s 1 9 S e u l i m e u m 1 B 5 5 . 9 1 2 0 1 9 6 4 ( M s = 6 . 5 ) S m a l l d e p r e s s i o n o n 7 . 5 7 . 6 n / a 1 3 d i l a t a t i o n a l s t e p o v e r
