1 Fault systems of the eastern Indonesian triple junction: evaluation

2
3 of Quaternary activity and implications for seismic hazards
4
5 IAN M. WATKINSON* & ROBERT HALL
6
7 SE Asia Research Group, Department of Earth Sciences, Royal Holloway
8 University of London, Egham, Surrey TW20 0EX, UK
9
10 *Correspondence: [email protected]
11
12
13 Abstract: Eastern Indonesia is the site of intense deformation related to convergence between
14 Australia, Eurasia, the Pacific and the Philippine Sea Plate. Our analysis of the tectonic geomor
phology, drainage patterns, exhumed faults and historical seismicity in this region has highlighted
15
faults that have been active during the Quaternary (Pleistocene to present day), even if instrumental
16 records suggest that some are presently inactive. Of the 27 largely onshore fault systems studied, 11
17 showed evidence of a maximal tectonic rate and a further five showed evidence of rapid tectonic
18 activity. Three faults indicating a slow to minimal tectonic rate nonetheless showed indications of
19 Quaternary activity and may simply have long interseismic periods. Although most studied fault
20 systems are highly segmented, many are linked by narrow (,3 km) step overs to form one or
21 more long, quasi continuous segment capable of producing M . 7.5 earthquakes. Sinistral shear
22 across the soft linked Yapen and Tarera Aiduna faults and their continuation into the transpres
23 sive Seram fold thrust belt represents perhaps the most active belt of deformation and hence
the greatest seismic hazard in the region. However, the Palu Koro Fault, which is long, straight
24
and capable of generating super shear ruptures, is considered to represent the greatest seismic
25 risk of all the faults evaluated in this region in view of important strike slip strands that appear
26 to traverse the thick Quaternary basin fill below Palu city.
27
28
29
30
31 Several of the devastating earthquakes that occurred Eastern Indonesia is a region of complex and
32 on faults around the world during the last decade rapid neotectonics. Convergence between Australia,
33 were either poorly understood or not recognized Eurasia, the Pacific and the Philippine Sea plates
34 at all. The Mw 6.6 Bam earthquake (Iran) of 26 (e.g. Hamilton 1979; DeMets et al. 1994; Hall
35 December 2003 ruptured a section of the Bam 1996, 2012; Bock et al. 2003; Charlton 2010) results
36 Fault that had a poor surface expression and had in both contraction and extension from subduction
37 not caused a destructive earthquake for 2000 years hinge rollback, lithospheric delamination and slab
38 (Eshghi & Zare 2003; Fu et al. 2004). The Mw 8.0 break off (e.g. Harris 1992; Spakman & Hall 2010;
39 Wenchuan earthquake (China) of 12 May 2008 Hall 2012).
40 resulted from complex rupture of part of the Lon Great uncertainty surrounds the position, tec
41 men Shan tectonic belt (Burchfiel et al. 2008), an tonic role and modern activity of eastern Indonesia’s
42 area that was previously considered not to be at many Quaternary faults (e.g. Hamilton 1979; Okal
43 risk from large earthquakes (Chen & Hsu 2013). 1999; Bailly et al. 2009; Charlton 2010). New
44 The Mw 7.1 Haiti earthquake of 12 January 2010 fault systems continue to be identified using both
45 occurred on the well known Enriquillo Fault, part modern geophysical/remote sensing and conven
46 of the fault system marking the northern boundary tional field techniques (e.g. Stevens et al. 2002;
47 of the Caribbean plate, but which had previously Spencer 2010, 2011; Watkinson et al. 2011; Pow
48 been mapped as having low seismic hazard based nall et al. 2013) and it is likely that many others
49 on recent seismicity (Stein et al. 2012). The Can remain unknown, with important implications for
50 terbury earthquake sequence (New Zealand) rup seismic hazard analysis.
51 tured the Greendale Fault, which was previously Despite intense seismicity in eastern Indonesia,
52 unrecognized because it was buried beneath alluvial there have been few catastrophic earthquake disas
53 sediments (Quigley et al. 2012). The Canterbury ters in the last 100 years compared with other rap
54 sequence culminated in the Mw 6.3 Christchurch idly deforming areas such as China, Iran, Japan
55 earthquake of 22 February 2011. These events and Pakistan (e.g. Holzer & Savage 2013; National
56 emphasize the need for the accurate identification Geophysical Data Center/World Data Service).
57 of faults that have been active during the Quaternary Significant events include: the 25 June 1976 Mw
58 and have the potential for modern tectonic activity. 7.1 Papua earthquake, which killed 3000 6000

From: Cummins, P. R. & Meilano, I. (eds) Geohazards in Indonesia: Earth Science for Disaster Risk Reduction.
Geological Society, London, Special Publications, 441, https://doi.org/10.1144/SP441.8
# 2016 The Author(s). Published by The Geological Society of London. All rights reserved.
For permissions: http://www.geolsoc.org.uk/permissions. Publishing disclaimer: www.geolsoc.org.uk/pub ethics
 I. M. WATKINSON & R. HALL

59 people; the 12 December 1992 Mw 7.8 Flores earth et al. 2005). Thus faults that show evidence of Qua
60 quake, which killed 2500 people and destroyed ternary activity may or may not also show evidence
61 31 800 houses; the 17 February 1996 Mw 8.2 Biak of palaeoseismicity, depending on whether they
62 earthquake, which caused a 7 m tsunami and killed have recently ruptured the surface, the rates of sed
63 at least 100 people (Okal 1999); the 16 November imentation and erosion, and whether they are truly
64 2008 Mw 7.4 Minahasa earthquake, which killed six ‘active’ in the sense that they have failed during
65 people and displaced 10 000; and the 16 June 2010 the Holocene. Equally, Quaternary faults may or
66 Mw 7.0 Yapen earthquake, which killed 17 people may not be present in records of instrumental or his
67 and destroyed 2556 houses (National Geophysical torical seismicity, depending on whether they have
68 Data Center/World Data Service; USGS Earth recently become inactive, have a long interseismic
69 quake Hazards Program). With increasing urban period, or have yielded historical earthquakes in
70 development and the replacement of traditional locations where there was no written documenta
71 wooden dwellings with concrete constructions, it tion. Quaternary fault activity is therefore distinct
72 is likely that damaging earthquakes will become from, but influential in, the field of active tectonics,
73 more frequent in the future (e.g. Wyss 2005). which includes future fault activity that may affect
74 This paper catalogues Quaternary fault systems human society (Wallace 1986).
75 onshore eastern Indonesia from Sulawesi to Papua, Quaternary fault activity was evaluated in
76 providing evidence for Quaternary tectonic activity this study by the following criteria: (1) instrumen
77 and a reconnaissance evaluation of the seismic haz tal/historical seismicity and geodetic observations;
78 ard of the faults (Fig. 1). (2) deformation of Quaternary sediments, often
79 indicated by topographic lineaments that could be
80 linked to an underlying fault; (3) the systematic
81 Methods offset of modern streams across a topographic line
82 Definitions and extent of study ament; (4) evidence of structurally controlled drain
83 age network modification where signs of an earlier
84 This study was concerned with evaluating Quater arrangement were preserved; (5) geomorphic indi
85 nary (Pleistocene and Holocene, 2.59 0 Ma) fault ces recording the relative youthfulness of fault
86 activity. Quaternary activity lies within the realm controlled mountain fronts; and (6) evidence of
87 of neotectonics, the study of broadly post Miocene, landslips localized to faults.
88 ‘young’ and still active tectonic events, the effects The study extent was a 2200 × 800 km swath
89 of which are compatible with modern seismotec of the Indonesian archipelago centred on the triple
90 tonics (Pavlides 1989). Neotectonics is distinct junction between Australia, Eurasia, the Pacific
91 from palaeoseismology the study of deformation and the Philippine Sea plates. It includes much of
92 related to specific past earthquakes (e.g. Michetti eastern Indonesia from Sulawesi eastwards, except
93
94
95
96
97
98
99
100
101
102
Colour
online/
colour
hardcopy
106
107
108
109
110
111
Fig. 1. Map of eastern Indonesia showing upper crustal structures with geomorphic evidence of Quaternary tectonic
112 activity and seismicity (1973 2014, focal depths ,35 km). CB, Cenderawasih Bay; GF, Gorontalo Fault; KaF,
113 Kolaka Fault; KF, Koor Fault; LF, Lawanopo Fault; MF, Matano Fault; MFTB, Mamberamo fold thrust belt;
114 MMC, Molino Metamorphic Complex; MSCC, Molucca Sea Collision Complex; NSS, North Sulawesi Subduction;
115 PKF, Palu Koro Fault; RS, Ransiki Fault; SF, Sorong Fault; SFTB, Seram fold thrust belt; TAF, Tarera Aiduna
116 Fault; YF, Yapen Fault. Locations of other figures as indicated.
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

117 the islands of the southern Banda Arc. Because of on a reconnaissance scale (Keller 1986). We utilized
118 the focus on geomorphic expression, the study mountain front sinuosity (Smf) and the valley floor
119 mainly dealt with onshore faults, except where mul width to valley height index (Vf) following the
120 tibeam bathymetry was available. method of Bull & McFadden (1977) and Bull
121 (1978). The key parameters are summarized in
122 Datasets Table 1. An excellent description of the method and
123 its uncertainties is given in Bull (2007). Although
124 Our interpretations of Quaternary fault activity are conventionally applied to normal faults, geomorphic
125 based on a variety of remote sensing data, field indices can been used in any setting where there is
126 observations by both authors and their students vertical motion, including regions of transpression
127 over several years (e.g. Roques 1999; Watkinson and transtension. However, they are of little value
128 2011; Pownall et al. 2013; Hennig et al. 2014) and in regions of pure strike slip and were not applied
129 published geodetic/geophysical data. Digital ele to pure strike slip segments in this study.
130 vation models (DEMs) based on Shuttle Radar Mountain front sinuosity is the ratio Smf ¼ Lmf/
131 Topography Mission (SRTM) 3 arc second/90 m Ls, where Lmf is the straight line length of the moun
132 resolution and ASTER 30 m resolution data were tain front and Ls is the true, or sinuous, length along
133 processed using ERMapper software. These data the mountain front following topographic contours
134 were also used to extract topographic contours and at the contact between alluvial fans and the solid
135 drainage networks using ArcGIS. Landsat TM and geology of the range front (Table 1). This method
136 ETM+ scenes composed of the 30 m resolution assumes that a fault bounded range front will
137 bands 432, 451, 531 and 742 (red green blue com become more sinuous over time in the absence of
138 binations) were used and, where appropriate, sharp tectonic activity (e.g. Bull & McFadden 1977;
139 ened with ETM+ band 8 panchromatic 15 m Rockwell et al. 1984). The method is well estab
140 resolution data. Where available, high resolution lished for Quaternary fault evaluation in regions of
141 visible spectrum imagery from Google Earth and extension (e.g. Ramı́rez Herrera 1998), contraction
142 Bing Maps (compiled from a variety of sources) and strike slip (e.g. Dehbozorgi et al. 2010), trans
143 and the ESRI World Imagery compilation, which tension (e.g. Silva et al. 2003; Yıldırım 2014), com
144 includes 2.5 m SPOT and ,1 m DigitalGlobe bined extension and contraction (Wells et al. 1988)
145 imagery, was also interpreted. ESRI World Imagery and differential uplift (e.g. Sohoni et al. 1999). Crit
146 is compiled from Esri, DigitalGlobe, GeoEye, ical uncertainties include the interpreter’s definition
147 Earthstar Geographics, CNES/Airbus DS, USDA, of the sinuous mountain front, which is partly
148 USGS, AEX, Getmapping, Aerogrid, IGN, IGP, dependent on the quality of the input satellite data,
149 swisstopo, and the GIS User Community. and the recognition of discrete mountain front seg
150 Multibeam bathymetric data (kindly provided by ments. Climate also has an impact on Smf indepen
151 TGS) from parts of the offshore Sorong Fault Zone dent of the tectonic rate. In a humid environment
152 and Cenderawasih Bay were interpreted in the like eastern Indonesia it is expected that erosion
153 same way as the DEMs. The multibeam data were and hence Smf will be higher than in an arid region
154 acquired using a Kongsberg Simrad EM120 Multi for a given tectonic rate.
155 beam Echo Sounder using 191 beams at equidistant The valley floor width to valley height index, Vf,
156 spacing. Positioning control used a C Nav Starfire measures the ratio between the valley floor width
157 DGPS. During processing, positioning, tidal and and the valley depth: Vf ¼ 2Vfw/(Eld Esc)
158 calibration corrections were applied, random noise (Erd Esc), where Vfw is the valley floor width,
159 and artefacts were removed, and a terrain model Eld and Erd are the topographic elevations of the
160 using a 25 m bin size was gridded and exported to left and right valley watersheds and Esc is the eleva
161 ESRI format. The multibeam data were further pro tion of the valley floor (Table 1). The method
162 cessed in ERMapper to remove voids. assumes that recently excavated river channels
163 All data were integrated in ArcGIS together (i.e. those into which a river has incised as a result
164 with previously published georeferenced maps. The of recent uplift) are V shaped and become more U
165 CMT focal mechanisms were from the International shaped over time (e.g. Bull & McFadden 1977;
166 Seismological Centre catalogue, plotted using Mir Rockwell et al. 1984). Like Smf, Vf has been applied
167 one software. We considered only earthquakes with in a wide range of tectonic settings (e.g. Wells et al.
168 a focal depth ≤35 km to avoid contamination from 1988; Ramı́rez Herrera 1998; Yıldırım 2014). Vf is
169 deeper structures that have little surface expression. sensitive to a number of variables apart from tec
170 tonic rate, so we standardized as much as possible
171 Geomorphic indices by: measuring Vf in all cases 1 km upstream from
172 the mountain front; measuring the valley width as
173 Geomorphic indices are a valuable tool to rapidly the width of the river channel visible on the highest
174 evaluate the relative tectonic rate of surface faults resolution satellite imagery available or the width
232
231
230
229
228
227
226
225
224
223
222
221
220
219
218
217
216
215
214
213
212
211
210
209
208
207
206
205
204
203
202
201
200
199
198
197
196
195
194
193
192
191
190
189
188
187
186
185
184
183
182
181
180
179
178
177
176
175

Table 1. Summary of geomorphic indices used in mountain front analysis, modified after Wells et al. (1988)

Parameter Definition Derivation* Measurement† Purpose Potential difficulties

Smf Sinuosity of topographic Lmf/Ls Defines the degree of topographic Defining actual topographic junction
mountain fronts modification of mountain front Defining discrete mountain
from the position of possible front segments
controlling structures
Vf Valley floor width to valley 2Vfw/[(Eld – Esc) – Defines the ratio of the valley Resolution of satellite imagery in defining
height index (Erd – Esc)] floor width to the mean height Vfw and divide elevations
of two adjacent divides, Effects of changes in lithology
measured at given locations Need to minimize variations in stream
along a stream channel within size (length and area)
the range block
I. M. WATKINSON & R. HALL

*Lmf, straight line length of mountain front; Ls, sinuous length along mountain front; Vfw, valley floor width; Eld and Erd, topographic elevations of left and right valley watersheds; Esc, elevation of valley floor.
†
Schematic map view for Smf, schematic cross-section view for Vf.
Both indices after Bull & McFadden (1977) and Bull (1978).
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

233 of the valley to the point where the floor rose Hamilton 1979; Silver et al. 1983a, b; Hall 1996).
234 10 m above the minimum elevation in individual North of Sulawesi, the Celebes Sea is being
235 transects; measuring only streams that reached the subducted beneath Sulawesi (e.g. Hamilton 1979;
236 mountain front without joining a higher order Silver et al. 1983a). Convergence across the sub
237 stream; and measuring only streams oriented ≥708 duction margin increases from 20 + 4 mm a 1 in
238 from the mountain front. Noise in the Vf signal the east to 54 + 10 mm a 1 in the west, associated
239 was reduced by averaging between three and ten with a clockwise rotation of about 48 Ma 1 about a
240 separate Vf measurements along each fault segment. pole close to Manado (Walpersdorf et al. 1998; Ran
241 High quality topographic maps are not available gin et al. 1999; Stevens et al. 1999; Beaudouin et al.
242 for eastern Indonesia, so both Smf and Vf were mea 2003). Immediately east of Sulawesi’s north ‘arm’,
243 sured in ArcGIS software using a combination of convergence between the Philippine Sea plate and
244 30 m ASTER GDEM satellite data and the ESRI Sundaland is partly accommodated by the Molucca
245 World Imagery compilation. This allowed the finest Sea double subduction and the overlying Sangihe
246 possible resolution of Ls and Vfw, which are crucial, and Halmahera thrusts (e.g. Rangin et al. 1999;
247 but potentially subjective, parameters. High quality Hall 2002; Beaudouin et al. 2003).
248 satellite imagery may be better for such measure Despite its setting within a collisional orogen,
249 ments than conventional maps (Bull 2007). Sulawesi is subject to widespread and young exten
250 Schemes for the classification of relative tec sion. Tomini Bay encloses a deep, enigmatic basin
251 tonic activity based on a combination of geomor containing up to 10 km of late Cenozoic sediments
252 phic indices have been proposed (e.g. Bull & (Jablonski et al. 2007; Pholbud et al. 2012).
253 McFadden 1977; Bull 1978, 2007). Here we applied Medium to high K Pliocene to modern volcanism
254 a modified scheme from McCalpin (2009). This in the Togian Islands within the bay results from Pli
255 uses Smf and Vf to classify relative tectonic activity ocene to Recent extension (Cottam et al. 2011) and
256 as follows: Smf , 1.1, mean Vf , 0.15, maximal onshore metamorphic core complexes are in the pro
257 activity; Smf 1.1 1.3, mean Vf 0.15, rapid activity; cess of being exhumed by processes related to
258 Smf 1.6 2.3, mean Vf 1.5, slow activity; crustal thinning (Kavalieris et al. 1992; van Leeu
259 Smf ≥ 2.5, Vf 1.7 2.5, minimal activity; and Smf wen et al. 2007; Spencer 2011).
260 2.6 4.0, mean Vf 7.4, inactive. This classification Active strike slip faults (e.g. Bellier et al. 2001),
261 allows a comparison between faults with different with left lateral slip rates of up to 39 mm a 1 (Soc
262 relative tectonic rates and corresponding geomor quet et al. 2006), characterize much of Sulawesi’s
263 phic expression. Because the indices record undated onshore Quaternary deformation. Often considered
264 Quaternary fault activity expressed by geomorphol to result from NW directed collision between the
265 ogy, the classes also correspond to a Quaternary tec Sula platform and Sulawesi (e.g. Silver et al.
266 tonic rate and not necessarily to a modern tectonic 1983b; Simandjuntak 1986), modern reconstruc
267 rate comparable with geodetic measurements. It tions emphasize the process of subduction hinge
268 should also be remembered that the schemes were rollback related to the substantial amounts of oce
269 developed using faults in arid areas of the western anic crust that have been, and continue to be, sub
270 USA where tectonic landforms are preserved for ducted around Sulawesi (e.g. Spakman & Hall
271 longer than in humid areas (e.g. Bull 1978), mean 2010; Hall 2012). The occurrence of Late Miocene
272 ing faults in the tropics will generally be classified to apparently modern north south directed conti
273 as tectonically ‘slower’ than equivalent faults at nental extension (e.g. Spencer 2011) in a broad
274 higher latitudes. region adjacent to the south directed Celebes Sea
275 We analysed both Smf and Vf for a total of 111 seg subduction means that a rollback mechanism must
276 ments from 24 fault systems across the study area be considered.
277 (Fig. 2a r, Table 2) and found a good correlation
278 between Smf and Vf (Fig. 3), supporting the reliability Palu Koro Fault
279 of each method. A previous study of geomorphic
280 indices along a segment of the Palu Koro Fault The Palu Koro Fault (Fig. 4) is the most prominent
281 (Vecchiotti 2008) obtained similar results to those active fault of Sulawesi and is of particular impor
282 presented here. However, we used these indices tance because it is straddled by Palu city (population
283 only as a simple quantitative means to support 340 000). The Palu Koro Fault appears to pass from
284 other evidence for Quaternary fault activity and did the SW corner of the Celebes Sea to a diffuse termi
285 not classify faults on the basis of these data alone. nation onshore at the northern end of Bone Bay, a
286 distance of 500 km, of which 220 km is onshore.
287 Sulawesi The fault’s tectonic role is disputed: sinistral
288 shear along a joint Palu Koro Matano Fault sys
289 Sulawesi lies at the triple junction between the Aus tem has been thought to accommodate clockwise
290 tralian, Eurasian and Philippine Sea plates (e.g. rotation and the northwards movement of a rigid
 I. M. WATKINSON & R. HALL

291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330 Fig. 2. Maps showing fault segments analysed for geomorphic indices. Index map at top. Bold lines are the sinuous
331 mountain front trace (Ls) used in mountain front sinuosity calculations. Base map is a 90 m SRTM digital elevation
332 model. All maps (a r) drawn to the same scale. Fault segment codes correspond to the codes used in Table 2.
333
334
335 eastern Sulawesi block driven by collision of the can be associated with the overriding plate above
336 Banggai Sula block in the east (e.g. Hamilton a retreating subduction hinge, particularly in the
337 1979; Silver et al. 1983b; Beaudouin et al. 2003). early stages of continent continent collision (Roy
338 However, it is significant that the Palu Koro Fault den 1993). The orientation and kinematics of the
339 and the North Sulawesi Trench form the western Palu Koro Fault are compatible with an interpreta
340 and northern limits, respectively, of a region of tion that it is passively bounding a region of litho
341 late Cenozoic extreme continental extension that spheric extension driven by northwards rollback in
342 includes deep sedimentary basins (e.g. Jablonski the Celebes Sea, although it is unclear whether
343 et al. 2007; Pholbud et al. 2012), exhumation of there is a hard linkage between the fault and the
344 the mid to lower crust in settings similar to meta trench.
345 morphic core complexes (e.g. van Leeuwen et al. It is not disputed that the fault is an active zone
346 2007; Watkinson 2011), exhumed low angle normal of high strain. Geodetic measurements suggest a
347 faults (Spencer 2011) and decompression related 39 mm a 1 sinistral slip rate together with 11
348 mantle melts (Cottam et al. 2011). These features 14 mm a 1 of extension (Socquet et al. 2006),
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373 Fig. 2. Continued.
374
375
376 consistent with a 35 + 8 mm a 1 strike slip rate Fault and none had a focal mechanism indicating
377 determined from displaced alluvial fans 11 000 + left lateral slip along the Palu Koro trend.
378 2300 years old (Bellier et al. 2001). The Palu Koro Fault has the clearest geomor
379 There is palaeoseismic evidence for three Mw phological expression of any eastern Indonesian
380 6.8 8.0 earthquakes during the last 2000 years, sug fault. It occupies a steep sided, narrow valley along
381 gesting a recurrence interval of about 700 years much of its path through central Sulawesi, before
382 (Beaudouin 1998; Bellier et al. 1998). However, branching into the Palu valley, which is up to
383 even allowing for 10 m slip for each Mw 6.8 8.0 15 km wide (Fig. 5a). Two prominent scarps bound
384 event, the resultant 30 m total displacement in the valley and form the base of mountains that rise to
385 2000 years is less than the 54 86 m predicted from .2.3 km elevation. The western scarp is highly lin
386 Holocene slip rates (Bellier et al. 2001). Although it ear, particularly the remarkable central segments c.
387 has been proposed by these earlier researchers that 15 km south of Palu city. Mountain front sinuosity
388 the deficit is accommodated by aseismic creep, it values are consistently low at 1.08 1.09, indicating
389 is equally possible that large, undetected earth maximal tectonic activity, increasing to 1.28 1.56
390 quakes occurred on unobserved fault strands and at the northern and southern ends of the valley, indi
391 that the total recurrence interval for all the Palu cating rapid to moderate tectonic activity (Fig. 5a).
392 Koro Fault strands is much less than 700 years. The valley floor curvature is generally correspond
393 Socquet et al. (2006) proposed four parallel strands ingly tight, with an average Vf of 0.24 along the
394 across a zone c. 50 km wide, locked at depths western scarp.
395 between 0 and 5 km. Features such as prominent triangular facets,
396 Records of historical seismicity in Sulawesi are hanging valleys and steep sided, deeply incised
397 poor. Damaging earthquakes occurred along the streams are also focused along the central western
398 Palu Koro Fault in 1905, 1907, 1909, 1927, 1934, basin bounding segment (Fig. 5b). These landforms
399 1968 (c. Ms 6.7), 1985 and 1993 (c. Ms 5.7) (Katili support dominantly rapid normal faulting along the
400 1970; Hamilton 1979; Beaudouin 1998), but little basin margin faults. Wine glass canyons, in particu
401 detail is known. Large earthquakes close to the lar, indicate that the tectonic subsidence/uplift rate
402 fault zone occurred in 1996 (Mw 7.7) and 1998 is faster than erosion. Lateral offset of the alluvial
403 (Mw 6.6 and 6.0); the former caused a 2 4 m high fans and rivers across the mountain front have
404 tsunami in the Toli Toli region (Pelinovsky et al. been observed, notably in the northern and southern
405 1997). However, these earthquakes originated off segments of the fault system (e.g. Hamilton 1979;
406 shore, did not clearly lie on the active Palu Koro Bellier et al. 2006).
464
463
462
461
460
459
458
457
456
455
454
453
452
451
450
449
448
447
446
445
444
443
442
441
440
439
438
437
436
435
434
433
432
431
430
429
428
427
426
425
424
423
422
421
420
419
418
417
416
415
414
413
412
411
410
409
408
407
Table 2. Summary of measurements of mountain front sinuosity and average valley width to height ratio for analysed fault segments

Fault Segment Lmf* L†s S‡mf Ave. V§f Figure 2 Fault Segment Lmf* L†s S‡mf Ave. V§f Figure 2

Malino boundary ML1 45.16 27.17 1.66 1.01 a Kolaka KO1 8.79 8.38 1.05 1.25 h
Malino boundary ML2 52.00 49.60 1.05 0.33 a Kolaka KO2 52.12 33.80 1.54 1.12 h
Malino boundary ML3 28.79 27.53 1.05 0.64 a Kolaka KO3 10.07 8.24 1.22 1.68 h
Malino boundary ML4 22.99 16.08 1.43 0.29 a Kolaka KO4 10.25 7.91 1.30 0.23 h
Malino boundary ML5 38.22 33.86 1.13 0.22 a Kolaka KO5 49.44 30.21 1.64 1.19 h
Malino boundary ML6 27.41 17.27 1.59 0.29 a Average 1.35 1.09
Average 1.32 0.46 Mangole MN1 7.07 6.33 1.12 0.49 i
Gorontalo GO1 33.24 16.37 2.03 0.88 b Mangole MN2 9.64 6.60 1.46 0.49 i
Gorontalo GO2 39.29 16.64 2.36 1.27 b Mangole MN3 18.96 12.08 1.57 0.55 i
Gorontalo GO3 12.03 6.58 1.83 1.69 b Mangole MN4 4.56 3.86 1.18 N/A i
Average 2.07 1.28 Sanana SN1 2.45 1.93 1.27 0.44 i
Palu– Koro PK0 16.94 10.99 1.54 0.31 c Average 1.32 0.49
Palu– Koro PK1 8.94 6.60 1.35 0.22 c Rana RA1 2.86 2.83 1.01 0.35 j
Palu– Koro PK2 10.48 9.64 1.09 0.29 c Rana RA2 3.68 3.12 1.18 0.23 j
Palu– Koro PK3 7.24 6.69 1.08 0.21 c Rana RA3 8.19 7.73 1.06 0.18 j
Palu– Koro PK4 4.33 3.99 1.09 0.19 c Rana RA4 20.19 10.31 1.96 1.53 j
Palu– Koro PK5 9.43 7.90 1.19 0.99 c Rana RA5 24.35 18.31 1.33 0.28 j
Palu– Koro PK6 11.02 6.91 1.59 0.35 c Rana RA6 15.49 10.40 1.49 0.68 j
Palu– Koro PK7 7.15 6.44 1.11 0.56 c East Buru EB1 18.09 12.56 1.44 0.47 k
Palu– Koro PK8 10.78 9.61 1.12 0.89 e East Buru EB2 18.61 15.08 1.23 0.50 k
Palu– Koro PK9 9.72 6.22 1.56 0.20 c East Buru EB3 26.93 12.53 2.15 1.13 k
Palu– Koro PK10 16.34 12.80 1.28 1.10 c East Buru EB4 25.16 12.62 1.99 0.75 k
I. M. WATKINSON & R. HALL

Palu– Koro PK11 27.15 19.02 1.43 0.32 c Average 1.48 0.61
Palu– Koro PK12 64.23 27.88 2.30 0.80 c Southern Seram TE1 42.32 23.05 1.84 1.88 l
Average 1.36 0.47 Southern Seram TE2 49.68 23.89 2.08 1.36 l
Parigi boundary PA1 69.67 21.44 3.25 0.90 d Average 1.96 1.62
Parigi boundary PA2 72.43 26.17 2.77 0.78 d Kawa KA1 27.55 20.75 1.33 0.28 l
Parigi boundary PA3 62.33 19.92 3.13 1.45 d Kawa KA2 15.16 13.76 1.10 0.26 l
Parigi boundary PA4 17.86 13.48 1.32 0.50 d Average 1.21 0.27
Average 2.62 0.91 Bobol BB1 14.59 11.61 1.26 1.57 m
Sapu valley SV1 6.17 5.67 1.09 0.40 d Bobol BB2 51.83 26.37 1.97 2.60 m
Sapu valley SV2 3.64 3.38 1.08 N/A d Bobol BB3 35.56 17.84 1.99 1.44 m
Sapu valley SV3 4.99 3.90 1.28 N/A d Bobol BB4 23.05 12.73 1.81 1.04 m
Sapu valley SV4 6.30 5.60 1.13 N/A d Average 1.76 1.66
Sapu valley SV5 6.89 4.74 1.45 N/A d Sorong SG1 33.89 28.77 1.18 1.15 n
Sapu valley SV6 4.67 3.78 1.24 N/A d Sorong SG2 13.90 11.94 1.16 1.54 n
Sapu valley SV7 5.27 3.84 1.37 N/A d Sorong SG3 15.11 13.16 1.15 0.48 n
Average 1.23 0.40 Sorong SG4 8.54 5.35 1.60 0.59 n
522
521
520
519
518
517
516
515
514
513
512
511
510
509
508
507
506
505
504
503
502
501
500
499
498
497
496
495
494
493
492
491
490
489
488
487
486
485
484
483
482
481
480
479
478
477
476
475
474
473
472
471
470
469
468
467
466
465
Bada valley
Bada valley
BV1
BV2
5.97
10.09
5.73
9.47
1.04
1.07
0.57
0.36
e
e
Sorong
Sorong
SG5
SG6
29.39
10.19
10.53
8.94
2.79
1.14
4.90
0.27
n
n
Bada valley BV3 13.27 9.30 1.43 0.73 e Sorong SG7 36.50 27.47 1.33 0.45 n
Average 1.18 0.55 Sorong SG8 17.73 14.83 1.20 0.44 n
Poso area PO1 18.15 13.74 1.32 0.29 e Sorong SG9 31.90 18.34 1.74 1.33 n
Average 1.32 0.29 Sorong SG10 15.44 9.57 1.61 0.40 n
Balantak BK1 13.02 11.80 1.10 0.25 f Sorong SG11 65.53 40.33 1.62 0.42 o
Balantak BK2 6.85 6.59 1.04 0.47 f Sorong SG12 62.92 22.04 2.85 5.10 o
Balantak BK3 5.16 4.24 1.22 N/A f Sorong SG13 54.48 30.12 1.81 8.68 o
Average 1.12 0.36 Sorong SG14 22.17 17.39 1.27 0.32 o
Matano MA1 15.06 11.78 1.28 1.12 g Sorong SG15 11.44 7.61 1.50 0.49 p
Matano MA2 9.21 8.66 1.06 0.25 g Average 1.60 1.77
Matano MA3 12.54 10.57 1.19 0.45 g Ransiki RS1 18.83 17.71 1.06 N/A p
Matano MA4 12.56 11.63 1.08 0.23 g Ransiki RS2 31.31 11.87 2.64 N/A p
Matano MA5 23.51 12.35 1.90 0.79 g Average 1.85 N/A
Matano MA6 12.34 10.52 1.17 0.72 g Wandaman boundary WM1 17.22 13.76 1.25 N/A q
Matano MA7 8.46 8.31 1.02 0.84 g Wandaman boundary WM2 15.55 12.08 1.29 N/A q
Average 1.24 0.63 Wandaman boundary WM3 23.73 18.57 1.28 N/A q
Kendari KD1 24.61 14.03 1.75 0.52 h Wandaman boundary WM4 11.58 6.74 1.72 N/A q
Kendari KD2 24.62 20.42 1.21 0.58 h Wandaman boundary WM5 19.75 13.83 1.43 N/A q
Average 1.48 0.55 Wandaman boundary WM6 9.84 9.33 1.05 N/A q
Lawanopo LW1 43.13 28.29 1.52 0.83 g Wandaman boundary WM7 9.99 7.80 1.28 N/A q
Average 1.52 0.83 Wandaman boundary WM8 45.67 19.61 2.33 N/A q
Towuti bounding TO1 27.98 24.39 1.15 0.41 g Average 1.45 N/A
Towuti bounding TO2 9.58 9.30 1.03 0.56 g Tarera – Aiduna TA1 6.11 5.05 1.21 N/A r
Towuti bounding TO3 51.16 25.10 2.04 1.22 g Tarera – Aiduna TA2 20.74 19.19 1.08 N/A r
Average 1.41 0.73 Tarera – Aiduna TA3 32.26 19.75 1.63 N/A r
Tarera – Aiduna TA4 38.83 8.48 4.58 N/A r
Tarera – Aiduna TA5 35.31 18.46 1.91 N/A r
Average 2.08 N/A

*Straight line length of mountain front.

†
Sinuous length of mountain front.
‡
QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

Mountain front sinuosity (Smf ¼ Lmf/Ls).

§
Average valley floor width to valley depth ratio (Vf ¼ 2Vfw/(Eld 2 Esc) – (Erd 2 Esc), where Vfw is the valley floor width, Eld and Erd are the topographic elevations of the left and right valley watersheds and
Esc is the elevation of the valley floor).
Location of sinuosity segment on Figure 2.
 I. M. WATKINSON & R. HALL

523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558 Fig. 3. Graph of mountain front sinuosity (Smf) v. valley floor width to valley height index (Vf) for studied faults.
559 Grey boxes indicate tectonic activity rates, after McCalpin (2009), with average Vf marked by the darker grey bar.
560 BB, Bobol Fault; BK, Balantak Fault; BV, Bada valley faults; EB, East Buru faults; GO, Gorontalo Fault; KA,
561 Kawa Fault; KD, Kendari faults; KO, Kolaka Fault; LW, Lawanopo Fault; MA, Matano Fault; ML, Malino
562 boundary faults; MN, Mangole faults; PA, Parigi faults; PK, Palu Koro Fault; PO, Poso faults; RA, Rana Fault; SG,
563 Sorong Fault; SN, Sanana faults; SV, Sapu valley faults; TE, Southern Seram faults; TO, Towuti faults.
564
565
566 A 58 releasing bend/step over is required to link path directly along strike from the strike slip fault
567 the southern segments of the Palu Koro Fault, where it enters the Palu valley in the south (Fig.
568 where it emerges from its narrow valley at Pakuli, 6a, b). Many meanders have a square aspect with lin
569 with the northern segments NW of Palu city. In ana ear longitudinal segments parallel to the projected
570 logue models and other non linear strike slip faults, fault (Fig. 6c). In the south of the valley a linear
571 such releasing geometries are often associated with braided reach is similarly parallel to the projected
572 well defined oblique normal sidewall faults and a fault; individual braid channels are anomalously
573 cross basin fault system with a more subtle surface linear (Fig. 6d). Strands of the Palu Koro Fault cut
574 expression that accommodates most of the strike ting an alluvial fan and offsetting its incised drainage
575 slip strain (e.g. Mann et al. 1995; Mann 2007; Wu directly along strike to the south confirm that the
576 et al. 2009) (Fig. 6a inset). river is structurally controlled. It is more reasonable
577 Analysis of Palu River channels since 2003 from to project this southern fault strand directly north
578 satellite imagery and the pattern of older filled across the basin than it is to consider strike slip strain
579 oxbow lakes on the valley floor indicates that long transferring immediately to the western sidewall
580 reaches of the river rarely deviate from a linear fault between Pakuli and Bolongga, particularly as
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627 Fig. 4. Central Sulawesi overview digital elevation model (SRTM), CMT catalogue earthquakes ,35 km depth and
628 structures that show geomorphic evidence of Quaternary tectonic activity. Rivers marked in white. Illumination
from NE. Location shown in Figure 1.
629
630
631
632 geomorphic indices in that region indicate a rela now) (Figs 5a & 6a). The sidewall faults are largely
633 tively low tectonic rate (Fig. 5a). an extensional partition, explaining the lateral slip
634 Thus we propose that much of the Palu Koro deficit across them, noted by Bellier et al. (2001).
635 Fault strike slip strain through the Palu valley is Confinement of the Palu River meander belts within
636 not accommodated on the prominent sidewall faults, the strike slip cross basin fault system may be due
637 but on a cross basin fault system that is obscured by to the development of a subtle graben, or to changes
638 fluvial deposits during interseismic periods (as it is in permeability, cementation or compaction in the
 I. M. WATKINSON & R. HALL

639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690 Fig. 5. (a) The Palu and Sapu valleys showing structures that with geomorphic evidence of Quaternary tectonic
691 activity, plus topography and drainage. Mountain front sinuosity values in bold italic text. For location, see Figure 4.
692 Major drainage basins for Salo Sapu and Salo Wuno are marked, separated by uplift at the western end of the Sapu
693 valley fault system. (b) View of the Palu Koro Fault scarp from the Palu valley, showing geomorphic evidence of
694 Quaternary tectonic activity.
695
696
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
Fig. 6. Evidence of a cross basin fault system within the Palu valley Quaternary fill. (a) Overview ASTER digital
749
elevation model draped with ESRI imagery layer. Illumination from NW. Palu River channels traced from six separate
750 images from 2003 to 2015. Inset shows fault pattern developed in an analogue model of a releasing bend, modified after
751 Wu et al. (2009), reflected and rotated to mimic the Palu valley. Sidewall faults and cross basin fault system are
752 highlighted in the model and on the satellite imagery. (b, c) Laterally confined meander belts, interpreted as representing
753 minor subsidence within the cross basin fault system. (d) Laterally confined river channels directly along strike from a
754 Palu Koro Fault strand seen to offset alluvial fans in the south of the valley. (c, d, e) show ESRI imagery.
 I. M. WATKINSON & R. HALL

755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

813 Quaternary valley fill resulting from penetration by the projected trace of the fault system through the
814 strike slip strands. valley, implying fault penetration through the Qua
815 The valley’s eastern sidewall fault is generally ternary basin fill (Fig. 7c, d). In the same way as for
816 much more segmented and strongly eroded than in the Palu valley, this evidence supports a cross basin
817 the west, with gentle slopes and irregular mountain fault system that accommodates most of the strike
818 fronts (Fig. 6a). South of the intersection with the slip strain, whereas the prominent sidewall faults
819 Sapu valley fault system, Smf values are 1.19 1.59 are dominantly extensional structures. The cross
820 and Vf averages of 0.55 indicate rapid to moderate basin fault system is buried by fluvial sediments,
821 tectonic activity. North of the Sapu valley intersec but co seismic subsidence, or changes in permeabil
822 tion, Smf is 2.30 and the average Vf is 0.80, indicat ity, cementation or compaction caused by periodic
823 ing slow tectonic activity. surface rupture through the Quaternary basin fill
824 Further south along the Palu Koro Fault, the continue to influence the meander patterns.
825 Gimpu basin exists at a small releasing step over At the head of the valley the entire fault system
826 and the Leboni basin occupies a releasing bend curves to a more NNW SSE trend a restraining
827 near the southern termination of the fault (Fig. 4). geometry under sinistral shear. A broad, oversized
828 The Palu Koro Fault bounding these flat topped valley in the west is presently at 700 m elevation
829 Quaternary basins has Smf values of 1.11 and 1.12, (Fig. 7a), i.e. 100 m above the modern Sapu valley
830 respectively, and similarly low Vf values of 0.56 floor. Exhumed brittle SW dipping reverse sinistral
831 and 0.89, indicating rapid to moderate tectonic faults in mica schists along the uplifted valley
832 activity. support long lived uplift at this restraining bend
833 (Fig. 7e, f ). At the foot of the westernmost obli
834 Sapu valley fault system que reverse fault, Smf is 1.08, suggesting maximal
835 tectonic activity (Fig. 5a).
836 A complex NW SE trending, 75 km long fault sys Drainage networks extracted from SRTM data
837 tem cuts across crystalline basement between Palu show that there is presently a drainage divide sepa
838 valley and the Tokorondo Mountains in the east rating the Salo Wuno and Salo Sapu catchment
839 (Fig. 4). The fault system is dominated by a double basins at the position of the thrust related uplift and
840 bend: a releasing bend forming the intermontane oversized valley (Fig. 5a). Water presently exits
841 Sapu valley (c. 600 m elevation) and a restraining Sapu valley via a narrow, steep sided gorge (Fig.
842 bend associated with uplift at the head of the valley 7a). The extreme steepness and geomorphic imma
843 (Fig. 5a). Both bends are consistent with an overall turity of that gorge suggests that it has recently cap
844 left lateral shear sense for the fault system. Anec tured the Sapu valley drainage, perhaps in response
845 dotal reports from residents of the valley (various, to tectonic uplift of its former well established route
846 pers. comm. 2009) suggest that earthquakes are fre to the NW via Salo Wuno. It is likely that the Sapu
847 quent and well known, although there is little instru valley was internally drained for some time after
848 mental seismicity and no record of historical uplift in the NW and may have contained an inter
849 earthquakes. montane lake similar to Lake Lindu to the south
850 Sapu valley is an irregular rhomboidal basin (Fig. 4), explaining the flat base of the Sapu valley.
851 bounded by normal faults trending NNE SSE and Four lines of evidence suggest the Sapu valley
852 east west (Fig. 7a). Many of the faults are arcuate, fault system has been active during the Quaternary:
853 convex into the basin. Their range front slopes are (1) control of the modern river meander belts by a
854 generally gentle, but Smf values of 1.09 1.45 and cross basin fault system that traverses the Quater
855 an average Vf of 0.40 suggest rapid to moderate tec nary basin fill; (2) youthful geomorphic expression
856 tonic activity (Fig. 7b). A conspicuous feature of the of the Salo Sapu gorge where it has recently cap
857 basin floor is the strong confinement of river chan tured the Salo Sapu drainage in response to tectonic
858 nels to narrow linear meander belts (Fig. 7b), as dis uplift in the NW; (3) rapid to moderate tectonic
859 cussed earlier for the Palu River. Both modern and activity along the transtensional segment sidewall
860 abandoned channels have linear meander belt mar faults, indicated by geomorphic indices; and (4)
861 gins and square longitudinal sections parallel to maximal tectonic activity along the transpressional
862
863
864 Fig. 7. Details of the Sapu valley fault system. (a) ESRI imagery of the Sapu and central Palu valleys showing
major structural and geomorphic features, particularly the releasing restraining double bend and re routing of axial
865
drainage from the NW valley to the Salo Sapu gorge. (b) Detail of the Sapu valley showing drainage and
866 highlighting fault control of the axial river. Location shown on Figure 5a. (c, d) Laterally confined meander belts
867 and lineaments, interpreted as representing minor subsidence within the cross basin fault system. (e) Lineated
868 slickenside surface from an exhumed fault core within the Sapu restraining bend. Location shown in Figure 7a.
869 (f) Lower hemisphere stereographic projection of fault planes (great circles) and slickenside lineations (points) from
870 the fault shown in Figure 7e. ESRI imagery.
 I. M. WATKINSON & R. HALL

871 segment’s reverse faults implicated in uplifting the overs between strands. Each basin is 4 6 km wide
872 oversized palaeovalley in the east, indicated by and 20 30 km long. The Matano basin hosts Lake
873 geomorphic indices. Matano, which, at 590 m deep (Haffner et al.
874 2001), is the deepest lake in Indonesia and the
875 Matano Fault tenth deepest lake in the world. A fault passing
876 from the northern margin of the Pansu Basin is
877 The Matano Fault passes from southern central very prominent as it cuts through ultramafic rocks
878 Sulawesi through the island’s SE arm to Tolo Bay in the SW corner of Lake Matano, just south of
879 (Fig. 4). It is typically shown to mark the southern Desa Matano (Fig. 8a). The fault then steps to the
880 edge of the Sula Block, linking to the Palu Koro left to another very prominent fault in the NW of
881 Fault to the west and the North Sulawesi Trench to the lake, from where it passes across the northern
882 the north (e.g. Hamilton 1979; Rangin et al. 1999). margin of the Mahalona Basin. Rapid subsidence
883 A hard linkage between either the Lawanopo or in the lake and earthquake focal mechanisms record
884 Matano and Palu Koro faults is a requirement of ing east west extension close to the lake probably
885 many rigid block models for Sulawesi (e.g. Bellier result from this releasing geometry (McCaffrey &
886 et al. 2006; Socquet et al. 2006). However, Silver Sutardjo 1982). Two major pop ups associated
887 et al. (1983b) noted that the nature of the connection with the uplift, thrusting and exhumation of meta
888 was not known. Modern satellite imagery shows a morphic rocks and serpentinite at restraining bends
889 highly segmented and discontinuous westernmost occur east of the Mahalona Basin and west of the
890 Matano Fault curving towards the Palu Koro Pansu Basin (Fig. 8a).
891 Fault, but the two structures remain largely isolated A number of consistent left lateral stream off
892 either side of the Gunung Balease massif (Fig. 4). sets, evidence of stream capture across two fault
893 In the east, the Matano Fault passes into the strands west of Pansu Basin (Fig. 8b) and steep
894 northern Banda Sea. Some workers link it to the sided, narrow fault valleys (Fig. 8c) suggest youth
895 Tolo Thrust (sometimes referred to as the Hamilton ful fault activity. Geomorphic indices of oblique
896 Thrust or the East Sulawesi Trench) (Fig. 1), an basin bounding faults range from Smf 1.06 1.28,
897 ESE verging thrust zone NE of Buton. Silver et al. average Vf 0.69 (Pansu Basin), Smf 1.02 1.17, aver
898 (1983b) suggest that the Matano and Palu Koro age Vf 0.78 (Matano Basin), Smf 1.19, Vf 0.45
899 faults act as a trench trench transform between (Mahalona Basin) to Smf 1.08 1.9, average Vf
900 the north Sulawesi subduction and the Tolo Thrust. 0.51 (eastern termination splay) and indicate mostly
901 This thrust has been considered to accommodate rapid to moderate tectonic activity.
902 convergence between the Makassar block and the On 15 February 2011, a shallow focus Mw 6.1
903 Banda Sea block (e.g. Socquet et al. 2006). How earthquake near the western end of Lake Matano
904 ever, recent work suggests that the Tolo Thrust is (NEIC) had a focal mechanism consistent with left
905 a gravity driven feature at the foot of a series of lateral slip along the Matano Fault. The earthquake
906 slumps (Rudyawan 2011), rather than a structure caused damage to concrete walls and buildings,
907 bounding a tectonic block (e.g. Silver et al. 1983b; including a newly built hospital in the Mahalona
908 Rangin et al. 1999). valley (Fig. 8d). The earthquake’s location sug
909 Geological offsets (e.g. Ahmad 1978) and stream gested that the prominent fault segment that links
910 offsets (e.g. Hamilton 1979) across the Matano the NE corner of Lake Matano with the Mahalona
911 Fault confirm that it is a left lateral structure and Basin failed (Fig. 8e). ‘Surface cracks’ were
912 that it has been active during the Quaternary (Bellier reported by local people at the eastern end of the
913 et al. 2006). Laterally offset streams are routinely basin but, although we visited the area in October
914 used to assess the shear sense and Quaternary activ 2011, a surface rupture could not be located. Close
915 ity of strike slip faults, usually in arid environments to the lake, very high resolution satellite imagery
916 (e.g. Sieh & Jahns 1984), but also in humid, forested recently made available (Bing Maps) shows three
917 environments (e.g. Lacassin et al. 1998; Wang et al. clear lineaments cutting across boggy ground and
918 2014). Nonetheless, such observations must be low lying forest (Fig. 8f) along strike from a Mat
919 interpreted cautiously, as stream offset may result ano Fault strand that offsets drainage to the left.
920 from stream diversion along a fault and capture by Although it is not possible to confirm that they rep
921 another downstream reach, as well as by the genuine resent the 2011 surface rupture, these lineaments
922 tectonic displacement of a single stream (Wallace appear to be tectonic in origin and are clearly very
923 1990). No study has used such offsets to evaluate young. Linking these lineaments with the reported
924 Quaternary slip rates along the Matano Fault. surface cracks in the east, along a topographically
925 The Matano Fault is highly segmented and lacks clearly defined fault strand, yields a postulated sur
926 a single through going strand (Fig. 8a). Several lin face rupture length of .39 km, which is longer than
927 ear basins (e.g. Pansu, Matano and Mahalona) lie expected for a Mw 6.1 earthquake from empirical
928 within or adjacent to the fault zone, often at step relationships (Wells & Coppersmith 1994).
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979 Fig. 8. Details of the Matano Fault. (a) Map of the Matano Fault, Lake Towuti and the northern part of the
980 Lawanopo Fault. Base map is ASTER digital elevation model draped with ESRI imagery. Location shown in
Figure 4. (b) Systematic stream offsets along strands of the Matano Fault west of the Pansu Basin. (c) Deep,
981
steep sided valley marking the westernmost Matano Fault NE of Gunung Balease. (d) Hospital in the Mahalona
982 valley damaged during the 15 February 2011 Mw 6.1 earthquake. Location shown in Figure 8e. (e) Detail of the
983 eastern Matano and Mahalona valleys, showing features related to the 2011 earthquake and inferred surface rupture
984 extent. ASTER base map. (f) Imagery from Bing Maps showing strong topographic lineaments in low ground in the
985 NE corner of Lake Matano, inferred to represent recent (2011?) surface ruptures. Location shown in Figure 8e.
986 (b) and (f) images # 2016 DigitalGlobe.
 I. M. WATKINSON & R. HALL

987 Lawanopo Fault and Lake Towuti to the Lawanopo Fault to its north. It is equivalent
988 to the Mendoke Fault of Bellier et al. (2006). Ham
989 The Lawanopo Fault (Fig. 4) consists of several ilton (1979) interpreted the fault as a SW dipping
990 straight NW trending fault segments that cross thrust and Bellier et al. (2006) considered the fault
991 Sulawesi’s SE arm south of the Matano Fault. The as a pre Early Pleistocene strike slip continuation
992 Lawanopo Fault is used in preference to the Matano of the Palu Koro Fault, but there is little evidence
993 Fault by Socquet et al. (2006) as the southern mar to support either hypothesis. One strand of the
994 gin of the ‘East Sula Block’. However, discontinu Kolaka Fault is sealed by 4.4 + 0.2 Ma dacites,
995 ous and eroded fault traces along strands of the potentially placing a limit on the timing of faulting
996 Lawanopo Fault system suggest that it has been (White et al. 2014).
997 mostly inactive during the Quaternary (Bellier The fault is composed of several NE SW trend
998 et al. 2006; Natawidjaja & Daryono 2014). None ing, gently arcuate segments up to 45 km long in
999 theless, recent earthquakes close to Kendari may map view. Along the Bone Bay coast and at Kolaka
1000 indicate that at least some strands of the Lawanopo town the downthrown side is to the south and the
1001 Fault system remain active. An Mw 7.5 earthquake easternmost segment is downthrown to the north
1002 in the Banda Sea 170 km SE of Kendari on 19 Octo (Fig. 9a). The polarity shift occurs across a 10 km
1003 ber 2001 had a strike slip focal mechanism and may wide relay straddling the Anggowala mountains.
1004 have originated on the projected offshore trace of The orientation of these two apparently normal
1005 the Lawanopo Fault (Yeats 2010). fault systems is kinematically consistent with sinis
1006 Like the Matano Fault, the Lawanopo Fault is tral slip along the overall Kolaka trend.
1007 highly segmented and there is no through going Geomorphic indices are highest closest to
1008 strand at the surface (Fig. 4). Mountain front sinuos Kolaka town, where Smf values of 1.22 1.30 and
1009 ity values on the few segments associated with adja Vf values of 0.23 1.68 suggest that there is rapid
1010 cent basins range from 1.21 to 1.75 and valley depth to slow active dip slip across the fault, which has
1011 to width ratios average 0.55 0.83, indicating mod a clear surface expression and is marked by triangu
1012 erate to slow tectonic activity. lar facets (Fig. 9c). Along strike to the NW a series
1013 Lake Towuti, the largest of the Malili lakes, of linear valleys and low ridges near Lasusua may be
1014 occupies an intermontane basin at 318 m elevation a continuation of the Kolaka Fault (Fig. 9b). An
1015 and has a maximum water depth of 203 m (Haffner absence of fault scarps or clearly displaced features
1016 et al. 2001). The basin lies in the wedge between makes fault activity hard to evaluate, but meander
1017 the Matano and Lawanopo faults and is itself cut confinement within a linear graben across the Lasu
1018 by linear fault strands that internally deform the sua alluvial plain and asymmetrical subsidence
1019 wedge (Fig. 4). Two prominent curvilinear faults highlighted by the river’s proximity to the bounding
1020 lie along the south and east of the lake (Fig. 8a). fault suggests recent fault activity (Fig. 9d).
1021 The closest, trending NE SW and downthrown to Faults downthrown to the WSW at the western
1022 the NW, forms the linear eastern lake boundary end of the Kolaka Fault have very low Smf and Vf
1023 and is marked by a number of fans prograding into values (1.05 and 1.25, respectively), deeply incised
1024 the lake. Its high mountain front sinuosity (2.04) streams and well developed triangular facets, sug
1025 and valley depth to width ratio (1.22) suggest slow gesting Quaternary dip slip. These faults face into
1026 tectonic activity. However, a large earthquake Bone Bay and may be related to basin bounding
1027 along this .25 km long structure could cause a sub extensional structures accommodating subsidence
1028 stantial tsunami or seiche in the lake. The second in the bay (Camplin & Hall 2014).
1029 fault, to the east, is longer still (.55 km) and highly
1030 continuous. It intersects the Lawanopo Fault at a
small angle and may directly transfer slip away
Balantak Fault
1031
1032 from that structure. Mountain front sinuosity ranges A prominent ENE trending linear structure, the
1033 from 1.03 to 1.15, suggesting maximal to rapid tec Balantak Fault, lies at the eastern end of Sulawesi’s
1034 tonic activity, although the valley floors are rather east arm and separates the Batui thrust system in
1035 rounded (average Vf 0.49). Lake Towuti would rap the south from mountainous highlands in the north
1036 idly fill with sediment if it were not actively subsid (Fig. 10a). It has been considered to be part of
1037 ing, therefore the bounding normal faults must be the Batui thrust system (Silver et al. 1983b), but
1038 considered to be active during the Quaternary. its remarkably straight outcrop, field observations
1039 (Simandjuntak 1986) and along strike alternation
1040 Kolaka Fault between local uplift and subsidence suggest that it
1041 is a steep, possibly strike slip, fault.
1042 The Kolaka Fault (Simandjuntak et al. 1984, 1994; Onshore, where the fault bends gently to the
EDQ1 Surono 1994) (Fig. 9a) lies along the southern mar right, small, apparently Quaternary basins are devel
1044 gin of the Mengkoka mountains and is sub parallel oped (Fig. 10b). There is uplift where the fault bends
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074 Fig. 9. Details of the Kolaka Fault. (a) Overview map of the main Kolaka Fault segments. Base map is ASTER
1075 digital elevation model draped with ESRI imagery. Location shown in Figure 1. (b) Straight segment of the Kolaka
1076
EDQ2 Fault associated with linear ridges and valleys. ESRI imagery. Google Earth imagery. Image # 2016 DigitalGlobe.
1077 Image Landsat. (c) Linear fault bounded mountain front and triangular facets indicating Quaternary fault activity at
1078 Kolaka town. (d) Asymmetrical axial drainage at a splaying fault segment near the western fault termination,
1079 indicating Quaternary subsidence along the bounding fault system.
1080
1081
1082 gently to the left. Both observations are kinemati segments, an apparently antithetic sinistral fault
1083 cally compatible with a dextral shear sense. One and the orientation of folds and thrusts are all kine
1084 of the zones of Quaternary subsidence is shown in matically compatible with dextral shear along the
1085 Figure 10c. A basin bounding fault at a small clock Balantak Fault (Watkinson et al. 2011). Earth
1086 wise angle from the regional Balantak Fault trend is quakes located onshore and west of Poh Head also
1087 crossed by streams that show no systematic offset, suggest right lateral and reverse slip parallel to the
1088 suggesting dominant dip slip. To the north, a prom Balantak Fault (Fig. 10a). However, a swarm of off
1089 inent lineament crosses the basin, expressed by lines shore earthquakes between Peleng and Taliabu to
1090 of vegetation and slightly darker (moister?) soil. the east have focal mechanisms that support sinistral
1091 This lineament’s parallelism with the Balantak slip along the Balantak trend. This apparent contra
1092 Fault to the east and its negligible topographic relief diction is discussed in Watkinson et al. (2011). Here
1093 suggests it is the through going strike slip fault we conclude that the geological and geomorphic
1094 strand. Although stream avulsion across the flat evidence supports long term Quaternary dextral
1095 topped basin is too dynamic to preserve meaningful slip. Further work is required to understand the sig
1096 offsets, the clear expression of the fault in the young nificance of a small number of contradictory seis
1097 sediments suggests the Balantak Fault has been mological signals in the area.
1098 active during the Quaternary. The Balantak Fault is almost continuous for
1099 The Balantak Fault’s termination system off 54 km from Balantak town in the east to Poh Bay
1100 shore to the east of Poh Head is composed of left in the west, where it probably continues just off
1101 stepping segments separated by folds and thrusts shore for another .30 km. Extending to include the
1102 (Fig. 10d). Contraction between left stepping main dextral fault system offshore to the SE makes the
 I. M. WATKINSON & R. HALL

1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154 Fig. 10. (a) East arm of Sulawesi and Banggai Sula Islands digital elevation model (SRTM), multibeam
1155 bathymetry, CMT catalogue earthquakes ,35 km depth and structures showing geomorphic evidence of Quaternary
tectonic activity. After Watkinson et al. (2011). Location shown in Figure 1. (b) Subsidence and uplift associated
1156 with releasing and restraining segments of the onshore Balantak Fault. ASTER digital elevation model base map.
1157 (c) Detail of bounding fault system of a Balantak Fault releasing segment, showing a north dipping normal fault and
1158 sub parallel lineament in agricultural land, inferred to represent a through going strike slip strand. ESRI imagery
1159 base map. (d) Detail of the offshore Balantak Fault expressed in multibeam imagery (illumination from NW)
1160 showing evidence of dextral shear.
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

1161 fault up to 250 km long. The onshore fault scarp has and gneisses that may be an exhumed metamorphic
1162 exceptionally low Smf values, from 1.04 to 1.22 core complex (van Leeuwen & Muhardjo 2005).
1163 (Fig. 10b), with correspondingly low average Vf val The faults dip north and south on the north and
1164 ues of 0.36, suggesting maximal to moderate south sides of the complex, respectively, and have
1165 tectonic activity. crystalline basement in their footwalls. The southern
1166 segment has a curvilinear trace .75 km long with
1167 Gorontalo Fault extremely low Smf values (1.05) and well developed
1168 triangular facets at the end of V shaped valleys with
1169 The Gorontalo Fault (Katili 1973) (Fig. 11a) Vf values of 0.33 0.64 within an uplifted footwall
1170 has been considered to be one of the major block block. On this basis, combined with no evidence
1171 bounding structures of Sulawesi (e.g. Socquet of strike slip, it is interpreted as a normal fault.
1172 et al. 2006; Molnar & Dayem 2010). Geodetic mod Further SW, the Tomini Bay bounding faults are
1173 elling suggests a 11 mm a 1 dextral slip rate and crossed by a number of fan deltas prograding into
1174 10 km locking depth; however, because the obser the bay. These are surprisingly short (,3 km),
1175 vation points are widely spaced, it remains possible given the potential upstream sediment source, sug
1176 that global positioning system (GPS) data record gesting rapid and recent hanging wall subsidence
1177 rotation of the entire north arm of the island rather (Fig. 11d, e). Segments further south along the
1178 than discrete slip across a fault (Socquet et al. ‘neck’ have higher Smf values (1.66) and the fan
1179 2006). There is little modern shallow seismicity in delta lobe length increases to .10 km, suggesting
1180 the Gorontalo area, suggesting that the fault is inac less significant recent subsidence (Fig. 11f ).
1181 tive or remains locked (Fig. 11a). At the southern end of the neck, a NE dipping
1182 The fault is composed of several branching fault system, including the Tambarama Fault (Phol
1183 segments, including major c. 30 km long segments bud et al. 2012), forms an apparently continuous
1184 south and north of Gorontalo city (Fig. 11b). Lim arcuate trace at Parigi (Fig. 4), marking the boundary
1185 boto Lake lies in the 7 km wide step over between between the Palu Metamorphic Complex onshore
1186 these two segments, indicating local transtension. (van Leeuwen & Muhardjo 2005) and Tomini Bay
1187 The fault is expressed by highly eroded scarps subsidence offshore. Smf values are generally high
1188 passing along the Tomini Bay coast and bounding (2.77 3.25), although a short northern segment is
1189 the Gorontalo/Limboto depression. Geomorphic less sinuous at 1.32. A well developed apron of fan
1190 indices suggest that the segments experience slow deltas extends 6 km from the mountain front.
1191 to minimal tectonic activity, with Smf values rang
1192 ing from 1.83 to 2.36 and an average Vf of 1.28.
1193 Although there is considerable human development Maluku and North Maluku
1194 within the Gorontalo/Limboto depression, which
1195 may obscure neotectonic activity, there appears to Maluku and North Maluku are composed of numer
1196 be little evidence of deformation within the Quater ous islands affected by disparate neotectonic pro
1197 nary sediment fill, except for the presence of Lim cesses. In the north, Halmahera (Fig. 1) and the
1198 boto Lake subsidence at the releasing step over. Sangihe Arc are involved in the active collision of
1199 two accretionary complexes above the subducted
1200 Western Tomini Bay bounding faults Molucca Sea slab, where the Sangihe forearc is
1201 being thrust eastwards over the Halmahera forearc
1202 A series of faults along the margin of Tomini Bay (e.g. Silver & Moore 1978; Hamilton 1979; Hall
1203 shows evidence of recent activity. The faults are 1987; Hall et al. 1995). The entire system accommo
1204 arcuate and generally mark the boundary between dates 80 mm a 1 of the 105 mm a 1 Philippine Sea
1205 mountainous ground along Sulawesi’s narrow plate Sundaland convergence (Rangin et al. 1999).
1206 ‘neck’ and Tomini Bay, which is up to 2 km deep Splays of the left lateral Sorong Fault pass through
1207 and contains a sedimentary succession up to 10 km and to the south of Halmahera and Bacan, where
1208 thick (Jablonski et al. 2007; Pholbud et al. 2012). there is abundant modern seismicity (e.g. Ali &
1209 Extension and mantle decompression across the Hall 1995; Hall et al. 1995) (Fig. 1).
1210 bay are associated with Plio Pleistocene volcanism South of Bacan, islands with Australian conti
1211 in the Togian Islands and possibly with modern vol nental basement, such as the Banggai Sula Islands
1212 canism at Una Una volcano (Cottam et al. 2011), and Obi, are bounded by strands of the Sorong Fault
1213 supporting recent extensional faulting and litho and were for a long time considered to have been
1214 spheric thinning both onshore and offshore (Phol translated from New Guinea along a 1900 km long
1215 bud et al. 2012). Sorong Fault passing from northern Papua New
1216 The northernmost bounding fault bounds the Guinea towards Sulawesi (e.g. Visser & Hermes
1217 2.5 km high Molino Metamorphic Complex (Fig. 1962; Audley Charles et al. 1972; Hamilton 1979;
1218 11c), a suite of quartzo feldspathic mica schists Pigram et al. 1985; Garrard et al. 1988; Hutchison
 I. M. WATKINSON & R. HALL

1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

1277 1989). New interpretations based on evidence of north of the Banggai Sula margin (Engdahl et al.
1278 extreme crustal extension and mantle exhumation, 1998; Rangin et al. 1999; Beaudouin et al. 2003),
1279 mantle tomography and geodynamic models (e.g. indicating that there are few active structures, that
1280 Spakman & Hall 2010; Hall 2011; Spencer 2010, deformation is largely aseismic or that the main
1281 2011; Pownall et al. 2014) suggest that those faults have interseismic periods that exceed instru
1282 islands, together with others along the northern mental records. This is a marked contrast with the
1283 Banda Arc such as Buru and Seram, were part of a abundant shallow seismicity associated with the
1284 continental spur that was fragmented during Mio Molucca Sea collisional zone further north. How
1285 cene Pliocene times by lower crustal delamination ever, a number of focal mechanisms north and
1286 driven by Banda Sea rollback. south of the islands indicate that there is some resid
1287 Quaternary extension in Maluku appears to be ual left lateral slip on east west to NW SE trend
1288 as important as it is in Sulawesi, despite an overall ing faults (Fig. 10a).
1289 collisional tectonic setting. Young metamorphic Mangole Island appears to be bordered along its
1290 core complexes exhumed in Seram (Pownall et al. north and south sides by several linear east west
1291 2013) and possibly Buru (Roques 1999) are associ trending normal faults, indicated by straight traces
1292 ated with low angle and steep normal faults. A sig and well developed triangular facets (Fig. 12a, b).
1293 nificant component of the seismic moment release Mountain front sinuosity values range from 1.11
1294 in Maluku is by normal and strike slip earthquakes, to 1.57 and Vf is from 0.44 to 0.55, suggesting that
1295 alongside important thrusting in the Molucca Sea some of the structures have been active during the
1296 and north Seram (e.g. Rangin et al. 1999). Sinistral Quaternary. Sanana Island, topographically orthog
1297 transpression through Seram accommodates Austra onal to Mangole, is bounded by NNW SSE trend
1298 lia Pacific convergence and links into the Tarera ing faults that can be traced offshore in multibeam
1299 Aiduna Fault of West Papua (e.g. Rangin et al. bathymetry. The most prominent fault, on the east
1300 1999; Stevens et al. 2002; Teas et al. 2009). coast, forms a well defined scarp .20 km long, dip
1301 ping and downthrown to the east, making it likely to
1302 Banggai Sula Islands be a normal fault (Fig. 12c). Triangular facets, hang
1303 ing valleys, deeply incised streams (Fig. 12d) and an
1304 The Banggai Sula Islands (Fig. 10a) occupy a frag absence of subaerial prograding fan delta tops wider
1305 ment of continental crust of Australian affinity that than c. 400 m suggest rapid recent eastwards subsi
1306 has collided with the east arm of Sulawesi (e.g. dence along the fault, supported by Smf values of
1307 Audley Charles et al. 1972; Hamilton 1979; Pigram 1.27 1.34.
1308 et al. 1985; Garrard et al. 1988). The South Sula Taliabu Island (Fig. 10a) is cut by a number of
1309 Sorong Fault was interpreted by Hamilton (1979) east west and north south trending Quaternary
1310 to follow the break in slope south of Taliabu and faults. The north south trending faults in the
1311 pass between Mangole and Sanana. North of the west have a particularly fresh geomorphic expres
1312 Banggai Sula Islands the North Sula Sorong sion. A north coast bedding parallel dip slope
1313 Fault (e.g. Hamilton 1979; Norvick 1979; Silver dips 68 into the Molucca Sea (Fig. 13a). Offshore
1314 et al. 1983b; Sukamto & Simandjuntak 1983), pre to the north a planar detachment surface 34 km
1315 viously considered to pass from the Bird’s Head, wide exactly corresponds to the Taliabu dip slope
1316 past Obi and along the north margin of the Bang onshore and represents a submarine slope failure
1317 gai Sula Islands towards Sulawesi’s east arm, can (Watkinson et al. 2011). Both onshore and offshore
1318 not be detected in new geophysical data and must lie slopes appear to be part of a single large glide
1319 below the Molucca Sea collision complex to the surface of a mega debris slide that translated
1320 north (Ferdian et al. 2010; Watkinson et al. 2011). much of north Taliabu at least 37 km north into
1321 Despite the density of deformation in the area, the Molucca Sea, probably causing a significant
1322 there is very little shallow seismicity immediately tsunami.
1323
1324
1325
1326 Fig. 11. (a) North arm of Sulawesi digital elevation model (SRTM), CMT catalogue earthquakes ,35 km depth
1327 and structures showing geomorphic evidence of Quaternary tectonic activity. Rivers marked in white. Location
1328 shown in Figure 1. (b) Detail of the onshore Gorontalo Fault and associated basins. ASTER digital elevation model
draped with ESRI imagery layer. (c) Fault system bounding the Malino metamorphic complex showing remarkably
1329 straight and steep mountain front and well developed triangular facets. (d) Overview of fan deltas prograding into
1330 western Tomini Bay across the bounding fault system. (e) Narrow fan delta clearly cut by the basin bounding fault,
1331 indicating rapid subsidence. Google Earth imagery. Image # 2016 DigitalGlobe. Data SIO, NOAA, U.S. Navy,
1332 NGA, GEBCO. Image (2016) Terrametrics. Image Landsat. (f) Wide fan delta further south indicating a slower rate
1333 of hanging wall subsidence. Google Earth imagery. Image # 2016 DigitalGlobe. Data SIO, NOAA, U.S. Navy,
1334 NGA, GEBCO. Image (2016) Terrametrics. Image Landsat.
 I. M. WATKINSON & R. HALL

1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356 Fig. 12. Details of Quaternary faults in the Banggai Sula Islands. For locations, see Figure 10a. (a) Part of the
EDQ3 linear normal fault system bounding the southern margin of Mangole Island. Google Earth image. Data SIO, NOAA,
1358 U.S. Navy, NGA, GEBCO, Image Landsat. Image # 2016 TerraMetrics. (b) Triangular facets along the north coast
1359 of Mangole Island. Oblique view in Google Earth. Data SIO, NOAA, U.S. Navy, NGA, GEBCO. Image #
DigitalGlobe. Image # 2016 TerraMetrics. Image Landsat. (c) Fault control along the eastern coast of Sanana
1360 Island. Image from Google Earth (greyscale inverted for clarity). Image Landsat. Image # DigitalGlobe. Image #
1361 2016 TerraMetrics. Data SIO, NOAA, U.S. Navy, NGA, GEBCO. (d) Detail of the Sanana fault, showing the
1362 extremely linear mountain front, narrow V shaped valleys and triangular facets. Image # DigitalGlobe.
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
Colour
online/
colour
hardcopy
1383
1384
1385
1386
1387
1388 Fig. 13. (a) Mega debris slide and post collapse normal faults, north coast of Taliabu. Base map is SRTM
1389 topography onshore and multibeam bathymetry offshore. Location shown in Figure 10a. (b) Oblique perspective
1390 view from Google Earth of one of the normal faults on the north slope of Taliabu. White arrows mark fault tips.
1391 View to the west. Field of view is c. 1 km in foreground. View location and direction indicated by arrow in
1392 Figure 13a. Image Landsat. Image # DigitalGlobe. Data SIO, NOAA, U.S. Navy, NGA, GEBCO.
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

1393 The north Taliabu dip slope is truncated by sev along the northern edge of Seram (Fig. 14a) and
1394 eral prominent east west trending faults that dip entirely in the western part of the fold belt (Teas
1395 steeply north. The geomorphic expression is very et al. 2009).
1396 fresh (Fig. 13b). The footwall crests are only slightly Seram is centred on a belt of high mountains
1397 eroded and, in most cases, the drainage runs parallel (.3 km elevation), which include tracts of conti
1398 to fault scarps and has not cut across them, except nental metamorphic rocks, ultramafic rocks and
1399 for a few prominent high order streams. The faults the Earth’s youngest exposed ultra high tempera
1400 displace the dip slope and must therefore post date ture granulites, exhumed since 16 Ma (Pownall
1401 the mega debris slide. Although we have no abso et al. 2014). The Plio Pleistocene Wahai and Fufa
1402 lute constraint on the timing of the slide, reef build formations onlap the elevated pre Pliocene succes
1403 ups are conspicuously poorly developed along sion, forming low plains along the northern coast
1404 the section of coast at the foot of the dip slope, but (Pairault et al. 2003), and are themselves overlain
1405 are extensive along the coast and small islands on by modern alluvial and reef deposits. There is evi
1406 either side. The slide must have happened recently dence of active contraction within these plains.
1407 enough that corals have been unable to fully recolo On the north coast of Seram, onshore fold growth
1408 nize the new coastline, suggesting that the post slide affects the modern drainage, suggesting that the
1409 normal faults are late Quaternary and probably folds have been active during the Quaternary
1410 still active. (Fig. 15a). Three large rivers draining the northern
1411 slopes of the Kobipoto Mountains are deflected
1412 Sorong Fault from Obi to Waigeo from a linear route to the coast by two sets of seg
1413 mented east west to NW SE trending hills. Pro
1414 Westward splaying segments of the Sorong gressive migration of the rivers away from the
1415 Fault emanate from the western Bird’s Head and hilltops is recorded by a trail of abandoned and filled
1416 pass close to the islands of Salawati, Misool, Obi, river channels left behind by the deflected river,
1417 Bacan, south Halmahera and Waigeo (e.g. Katili expressed by oxbow shaped fields and areas of veg
1418 1975; Hamilton 1979; Ali & Hall 1995) (Fig. 1). etation (Fig. 15b, c). Larger hills, like that in the cen
1419 Although there is debate about whether the Sorong tre of Figure 15a, cause more deflection than smaller
1420 Fault onshore West Papua is tectonically active folds, like that in the east which only deflects Wai
1421 (discussed later in this paper), at the latitude of (stream) Kobi slightly. In all cases the abandoned
1422 Obi there is 19 + 8 mm a 1 left lateral displace channels are located upslope of the modern river,
1423 ment between Ternate and the Bird’s Head that suggesting that progressive uplift is forcing river
1424 may be accommodated by one or more strands of avulsion. This tendency for the hills to grow
1425 the Sorong Fault (Bock et al. 2003). Seismicity is symmetrically from a central axis, their elongate
1426 limited in the islands immediately west of the morphology and their asymmetry (steep northern
1427 Bird’s Head, but intense seismicity occurs around slopes, shallow southern slopes) support the inter
1428 Obi, Bacan and south Halmahera (Rangin et al. pretation that they are the surface expression of shal
1429 1999), which may be where sinistral strain is trans low, north vergent fault propagation folds above
1430 ferred from Seram into the Molucca Sea. south dipping thrusts (Fig. 15d).
1431 Abandoned meander channels and point bars on
1432 Seram fold thrust belt the coastal plain in the central part of Figure 15d are
1433 not associated with any obvious modern river, but
1434 Between northern Seram and the Bird’s Head is a seem to originate at the foot of the central frontal
1435 broad zone of transpression linked to convergence thrust. Abandoned remnants of a comparably large
1436 between Australia and the Pacific plate (Fig. 1). A river can also be observed in an uplifted valley
1437 deep bathymetric trough, the Seram Trough, lies immediately to the south, and directly north of a
1438 150 km north of Seram Island and curves around fourth major north flowing river, which presently
1439 the Banda Sea, linking to the Timor Trough and ulti abruptly curves around the eastern tip of the fault
1440 mately the Java Trench. The Seram Trough has been before joining Wai Musi. It is interpreted that the
1441 interpreted as a subduction trench (e.g. Hamilton abandoned channels here represent a river that
1442 1979), a foredeep ahead of a fold thrust belt (e.g. flowed directly north before the fold developed.
1443 Audley Charles 1986) and a hinge zone marking An uplifted valley across the mid point of the fold
1444 the northern limit of delaminated and subducted shows that the river attempted to downcut as the
1445 lower continental crust (Spakman & Hall 2010). fold grew, but was ultimately thwarted by a high
1446 Convergence across the Seram Trough is pres uplift rate and swung east to be captured by Wai
1447 ently 20 mm a 1 (Rangin et al. 1999; Stevens Musi. Deep lateral incision by the captured river
1448 et al. 2002) and is associated with intense seismicity into the back limb of the fold (Fig. 15b) suggests
1449 generated by shallow thrust faulting (McCaffrey that the fold growth, and presumably underlying
1450 1989; Engdahl et al. 1998) mainly concentrated thrust activity, is ongoing.
 I. M. WATKINSON & R. HALL

1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492 Fig. 14. (a) Seram digital elevation model (SRTM), CMT catalogue earthquakes ,35 km depth and structures
1493 showing geomorphic evidence of Quaternary tectonic activity. Rivers marked in white. Offshore structures from
1494 Teas et al. (2009). Location shown in Figure 1. (b) Normal faults along the south coast of Seram, marked by a linear
1495 mountain front and a prominent lineament crossing a narrow fan delta. Google Earth imagery. Data SIO, NOAA,
1496 U.S. Navy, NGA, GEBCO. Image # 2016 DigitalGlobe. Image # 2016 TerraMetrics. (c) Possible Quaternary fault
1497 SW of Ambon, marked by a lineament that crosses volcanic hills and Quaternary drift. Google Earth imagery.
1498 Image # 2016 DigitalGlobe. (d) Example of foliated gouge from a thick fault zone located where the lineament
1499 illustrated in Figure 14c reaches the coast. Pen is 14 cm long.
1500
1501
1502 A series of abandoned channels east of Wai Such evidence of recent hanging wall uplift
1503 Musi, the easternmost of which link to Wai Kobi, and tectonic folding, together with the low relief
1504 indicates that river itself may previously have of the range front, leads to the conclusion that
1505 been a tributary to Wai Kobi, before being deflected the faults are youthful, low angle, south to SW
1506 to the west and ultimately cut off from the trunk dipping thrusts supported by focal mechanisms
1507 stream, presumably by uplift above the eastern along the north coast (Fig. 14a). Uplifted coastal ter
1508 frontal thrust. races in the foreland of the onshore thrusts and a
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

1509 conspicuously wide coastal plain (Fig. 15d) suggest SE trending planes, whereas one close to the west
1510 additional young uplift north of the onshore thrusts, ernmost splay indicates sinistral slip (Fig. 14a).
1511 perhaps in response to a third set of active faults just The fault zone splays as it enters Taluti Bay in
1512 offshore. This is consistent with modern thrust the east (Fig. 16a). The splay strands are associated
1513 activity within the broad fold thrust belt offshore with well developed triangular facets that record
1514 (e.g. Engdahl et al. 1998; Teas et al. 2009) and a Quaternary normal faulting (Fig. 16d). The two
1515 1629 mega thrust earthquake probably originating major splays have low Smf values of 1.10 in the
1516 in the Seram Trough (Liu & Harris 2013). north and 1.33 in the south and an average Vf of
1517 0.27, indicating rapid to moderate tectonic activity.
1518 Kawa Fault The Kawa River flows hard against the southern
1519 splay, suggesting active subsidence along that
1520 The Kawa Fault (Pownall et al. 2013) lies in the segment, despite its higher Smf indicating slower
1521 prominent ESE WNW trending deep linear valley tectonic activity than in the north. However, the riv
1522 that passes through central Seram (Fig. 14a) and is er’s position may also be influenced by landslips,
1523 occupied by the Kawa River. The fault broadly sep debris flows and anticline growth in the northern
1524 arates upper greenschist to mid amphibolite facies part of its valley. In the west, the fault splays north
1525 Tehoru Formation rocks in the south from generally of Elaputih Bay, attaining a total onshore length of
1526 higher grade metamorphic rocks of the Saku and 90 km, or 120 km including a possible splay fault
1527 Taunusa complexes in the north (Germeraad 1946; along the north coast of Taluti Bay (Fig. 14a).
1528 Tjokrosapoetro et al. 1993; Pownall et al. 2013). Although the fault zone is thickly forested,
1529 The Kawa Fault coincides with the position of numerous landslip scars can be recognized along
1530 strongly mylonitic garnet bearing Tehoru Forma the fault, indicating recent seismicity (Fig. 16a). In
1531 tion schists with a steeply dipping foliation consid addition, the eastern termination is characterized
1532 ered by Linthout et al. (1991) to record dextral by a series of discontinuous tilted blocks suggestive
1533 shear, but now recognized to have been intensely of slope failure along the southern margin of the
1534 folded and possibly originating in a low angle nor Manusela Mountains (Fig. 16a). An M 7.8 earthquake
1535 mal fault, resulting in complexly re oriented kine in 1899 triggered landslides that caused a 12 m
1536 matic indicators (Pownall et al. 2013). high local tsunami at Tehoru (http://www.ngdc.
1537 A brittle fault zone up to 2 km wide (Pownall noaa.gov; Brune et al. 2010), although it is unclear
1538 et al. 2013) overprints the mylonitic rocks and con whether the source was the Kawa Fault or a more
1539 trols the modern topography (Fig. 16a, b). Fault distant earthquake. However, all evidence points to
1540 strands are generally parallel to the mylonitic folia the Kawa Fault being active during the Quaternary
1541 tion and contain abundant serpentinite slivers and and capable of generating large earthquakes.
1542 smears. Mid way along the fault is a prominent
1543 right step associated with uplift and a major drain Other active faults of Seram
1544 age divide, pointing to local transpression due to
1545 left lateral slip. Stream offsets measured from Land Along strike from the Kawa Fault on the east side
1546 sat and Google Earth imagery along the fault of Taluti Bay, a fault zone occupies the valleys of
1547 (Fig. 16a) range from 66 605 m of left lateral offset Wai Masumang and Wai Bobol and is here termed
1548 (22 measurements) to 62 334 m of right lateral off the Bobol Fault (Fig. 14a). It is highly segmented,
1549 set (five measurements). Most measurements have a although with a total onshore length of 100 km
1550 high uncertainty, increased by Seram’s extremely and possible along strike continuity with the Kawa
1551 humid climate and thick forest cover. Nonetheless, Fault, it is a significant structure. Four large basins
1552 some measurements for example, the 268 and are developed along its length, each bounded by
1553 253 m left lateral offsets (e.g. Fig. 16c) are con ESE WNW to SE NW trending normal faults.
1554 sidered to be robust because: (1) they lie on fault The mountain front sinuosity along these structures
1555 segments that are well defined (narrow linear val ranges from 1.26 in the central section to 1.99 in the
1556 leys with other independent evidence of a fault ori west and the average Vf is 1.66, indicating moderate
1557 gin such as triangular facets and steps/bends with to slow tectonic activity. There are a number of
1558 the corresponding uplift/subsidence appropriate to stream offsets both across the basin bounding faults
1559 the sense of river offset); (2) there is no evidence and across parallel faults in adjacent mountains
1560 of stream capture; and (3) upstream and downstream (Fig. 16e). Convincing displacements are all left
1561 valleys have a similar geomorphic character. A lateral and range from 310 m to 2.06 km (Fig. 16f ).
1562 left lateral shutter ridge displacement and a NW Most strike slip fault segments within the fault
1563 SE trending fold within the Kawa River delta (Fig. zone are parallel to the Kawa Fault and the two
1564 16a) support recent sinistral shear. A few earth fault systems appear to be tectonically related and
1565 quakes close to the western end of the fault yield part of a broader zone of active left lateral shear
1566 CMT solutions suggesting dextral slip along NW linking to the Tarera Aiduna Fault in West Papua.
 I. M. WATKINSON & R. HALL

1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

1625 The southern margin of Seram is locally formed Manipa Strait have CMT solutions indicating either
1626 by linear mountain fronts flanked by narrow fan NNE SSW dextral events or WNW ESE sinistral
1627 deltas not more than 1 km wide. The steep, linear events, including a 14 March 2006 Mw 6.7 earth
1628 aspect and high topographic relief of the mountain quake 25 km offshore. Most earthquakes have a
1629 fronts and the topographic lineaments that cross component of reverse slip; others are pure thrust
1630 the fans parallel to the mountain front (Fig. 14b) earthquakes with a NW SE trend.
1631 suggest that the mountain front is defined by Quater Most of Buru’s sparse population lives in the NE
1632 nary normal faults. However, the coastal range is of the island, including the major town, Namlea.
1633 deeply eroded, with Smf values of 1.84 2.08 and A 5 10 km wide system of NW SE trending
1634 an average Vf of 1.62, indicating slow tectonic activ faults cuts through the town, across Kayeli Bay,
1635 ity. Earthquake focal mechanisms towards the west and defines the coastline (Fig. 17b). The faults are
1636 of the coastal fault system in the region of Elaputih expressed in remote sensing data by linear hills
1637 Bay support shallow focus, broadly south directed, and sag ponds at releasing right step overs, notably
1638 steep normal faulting (Fig. 14a). at Jikumerasa (Fig. 17c). Fault strands that cut
1639 A number of other small suspected normal fault through basement metamorphic rocks and alluvial
1640 systems occur around the SW coast of Seram, fans show consistent stream offsets and pass directly
1641 including those bounding the Ambon Islands. One into Quaternary alluvium and control modern river
1642 fault along the northern coast of Hitu (Fig. 14a) is channels (Fig. 17d). Stream offsets of up to 85 m
1643 particularly steep and straight, with an Smf of 1.16 across individual strands are mostly right lateral;
1644 and well developed triangular facets along its where they are left lateral there is clear evidence
1645 16 km long trace. A NE SW trending lineament for stream capture. Variations in offset sense and
1646 that passes through Ambon city marks the southern amount are to be expected streams are dynamic
1647 coast of Ambon Bay (Fig. 14c) and is associated and are not passively offset like pre kinematic geo
1648 with a zone of fault breccia and foliated gouge sev logical markers. The process of offset, beheading
1649 eral metres thick (Fig. 14d). An M 7.6 earthquake and capture, leading to stream offsets of zero or
1650 occurred on 8 October 1950 close to the south opposite to the fault’s shear sense, is well docu
1651 coast of Ambon (Bath & Duda 1979), although mented and widely observable in active faults
1652 it is unlikely that such an event could have been worldwide (e.g. Wallace 1968; Sieh & Jahns 1984;
1653 caused by the relatively short, dominantly normal Huang 1993; Walker & Allen 2012). All these fea
1654 faults visible onshore. tures imply Quaternary NW SE trending dextral
1655 fault activity in NE Buru, despite the apparent dis
1656 Buru cordance with the few earthquake focal mechanisms
1657 recorded.
1658 Buru consists of a presumed Palaeozoic continental A broad fault zone 65 km long almost bisects
1659 metamorphic basement flanked by a Mesozoic sedi Buru from the NE to SW (Figs 17a & 18a). Identi
1660 mentary succession (Tjokrosapoetro et al. 1993), fied as left lateral on early geological maps (e.g.
1661 both of which are probably continuous with similar Tjokrosapoetro et al. 1981), little else is known
1662 units in Seram (e.g. Pigram & Panggabean 1984; about the fault zone, here termed the Rana Fault.
1663 Linthout et al. 1989). Young K Ar ages of 4 Danau (lake) Rana, in the centre of Buru, occupies
1664 5 Ma (Linthout et al. 1989) and an apatite fis an intermontane basin within a right step over
1665 sion track central age of 2.5 + 0.5 Ma suggest late between two segments of the Rana Fault, suggesting
1666 Neogene exhumation, possibly accommodated by that the fault is dextral. West of Wadule, Wa (river)
1667 low angle normal faults (Roques 1999) as similarly Geren is abruptly diverted 908 from a broad over
1668 postulated for western Seram (Pownall et al. 2013). sized valley, which would have taken it to the
1669 Intense shallow seismicity associated with coast in the NE of the island, into a narrow and
1670 Seram terminates abruptly in Manipa Strait, east of steep sided canyon (Fig. 18a) that links with Wa
1671 Buru (Fig. 17a). A broad belt of earthquakes in Apu and empties into Kayeli Bay further south
1672
1673
1674
16 EDQ3 Fig. 15. Evidence of Quaternary thrusting along the north coast of Seram. (a) ESRI image showing a number of
1676 NE flowing rivers flowing around linear elevated and forested regions. Location shown in Figure 14a. (b, c)
Migrating rivers marked by filled channels and oxbow lakes and incision into uplifting regions. Google Earth
1677
imagery. Image # 2016 DigitalGlobe. (d) Interpretation of Figure 15a. Thick arrows indicate progressive migration
1678 of river channels; short arrows show coastline regression. Abandoned channels at points A, B and C are interpreted
1679 to represent the previous route of a river that entered the sea at D north of a meander plain at C, but was cut off by
1680 thrust hanging wall (HW) uplift at B and was forced to divert east from point A to join Wai Musi, leaving previous
1681 channels abandoned. Other rivers show lateral migration away from the growing tips of thrusts in response to
1682 hanging wall fold growth. See text for further details.
 I. M. WATKINSON & R. HALL

1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738 Fig. 16. Strike slip faults of southern Seram. (a) Overview map of the Kawa fault zone showing Quaternary fault
1739 strands, rivers, river offsets (in metres) and landslips. Left lateral offsets in black, right lateral offsets in grey. See
1740 Figure 14a for location. (b) ESRI image of the Kawa fault zone highlighting its clear geomorphic expression and
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
Fig. 17. Quaternary fault features in Buru. (a) Digital elevation model (SRTM), CMT catalogue earthquakes
1761 ,35 km depth and structures that show geomorphic evidence of Quaternary tectonic activity. Rivers marked in
1762 white. Location shown in Figure 1. (b) Overview of topographic lineaments passing through Namlea and NE Buru.
1763 ESRI imagery. (c) Detail showing possible sag pond developed at the releasing step over between right stepping
1764 fault segments. (d) Evidence of strike slip faulting along the Namlea lineament trend. Lineaments pass from
1765 basement rock through Quaternary drift and are associated with systematic right lateral stream offsets. Image #
1766 2016 DigitalGlobe.
1767
1768
1769 (Fig. 17a). This pronounced capture of a northern landslip scars in the surrounding hills close to
1770 drainage basin by a relatively minor tributary of the fault.
1771 Wa Apu appears to have been triggered by uplift A c. 10 m high scarp along the base of alluvial
1772 at a left bend in the Rana Fault immediately east fans in the valley, visible in high resolution Digital
1773 of the capture point (Fig. 18a), again suggesting Globe satellite imagery from Google Earth, has the
1774 Quaternary dextral shear. appearance of a normal fault surface rupture (Fig.
1775 Upstream of the Wa Geren stream capture, the 18d, e). The valley is relatively thinly vegetated
1776 Rana Fault has exceptionally fresh geomorphic and the scarp, discontinuous over c. 7.5 km, is well
1777 expression (Fig. 18b, c), with pronounced triangular preserved. Although in places it is parallel to the
1778 facets and very low Smf values from 1.01 to 1.18 modern river valley, the linear scarp also crosses
1779 along the southern valley slope and a correspond higher ground, proving that it is not simply an ero
1780 ingly low average Vf of 0.25, all suggesting a max sional feature. By analogy with proved historical
1781 imal to rapid tectonic rate. There are a number of earthquake surface ruptures with a similar topo
1782 beheaded and offset streams along the southern val graphic expression for example, the 1857 Lone
1783 ley slope, although there is no consistent tectonic Pine earthquake (Beanland & Clark 1994) and the
1784 lateral offset. The axial river has migrated system 1609 Hongyazi earthquake (Xu et al. 2010) the
1785 atically eastwards in two places, leaving behind Buru scarp may have formed during the last few
1786 abandoned channels uplifted up to 10 m above the hundred years. The entire 10 m throw could have
1787 modern river channel (Fig. 18c). The uplift defines developed during a single M 7.5 earthquake, accord
1788 a pair of low amplitude right stepping en echelon ing to empirical relationships (Wells & Copper
1789 periclines, consistent with Quaternary right lateral smith 1994), or during a number of smaller events,
1790 shear. There is abundant evidence of revegetated similar to the Star Valley Fault at Afton, Wyoming,
1791
1792
1793 Fig. 16. (Continued) thick forest cover. (c) Representative stream offset across the main Kawa Fault strand, image from
1794 Google Earth. (d) View into the Kawa Fault from the Wai Kawa delta showing the linear mountain front and triangular
1795 facets developed along the northern strand of the Taluti Bay splay. Image # 2016 DigitalGlobe. (e) Overview map of
1796 the Bobol fault zone showing Quaternary fault strands, rivers and left lateral river offsets (in metres). Bold italic
1797 numbers are Smf values. See Figure 14a for location. (f) Representative stream offset across the main Kawa Fault strand,
1798 also showing fault control of river channels. ESRI imagery.
 I. M. WATKINSON & R. HALL

1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
Fig. 18. Rana Fault, central Buru. (a) Overview of the Rana Fault, ASTER GDEM base map. Rivers marked in
1852 white, with white arrows showing the flow direction of major rivers discussed in the text. Location shown in
1853 Figure 17a. (b) ESRI image of the central part of the Rana Fault. (c) Interpretation of the image in (b), showing
1854 evidence of Quaternary fault activity. (d) Possible fault scarp along the foot of triangular facets marking the Rana
1855 Fault. (e) Detail of the possible fault scarp, showing steep dip, fresh geomorphic expression and straight trace.
1856 (d) and (e) are oblique views from Google Earth. Image # 2016 DigitalGlobe.
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

1857 where an 11 m high scarp formed during three late since 1938 was the tsunamigenic 17 February
1858 Quaternary earthquakes (Piety et al. 1992). This is 1996 Mw 8.2 Biak earthquake, which was also the
1859 perhaps a more likely scenario given the relatively largest thrust event worldwide since 1977 (Henry
1860 short length of the Rana Fault. & Das 2002) and may have been associated with
1861 Elsewhere in Buru the geomorphic expression of the 1979 M 7.9 Yapen earthquake (Okal 1999).
1862 other steep normal faults suggests rapid to moderate Left lateral strain of up to 80 mm a 1 resulting
1863 tectonic activity. Faults associated with the Rana from oblique Australia Pacific convergence is
1864 Lake basin have Smf values of 1.33 1.49 (Fig. 2j). accommodated across a 300 km wide zone of sinis
1865 Short fault segments in the SE of the island have tral shear (Stevens et al. 2002) focused on the Yapen
1866 Smf values of 1.23 and 1.44, whereas those on the Fault system in the north and stepping across Cen
1867 extreme east coast are more eroded, with Smf values derawasih Bay to the Tarera Aiduna Fault system
1868 of 1.99 and 2.14 (Fig. 2k), indicating that they have in the south, largely bypassing the antecedent
1869 been less active during the Quaternary. Sorong Fault in West Papua (e.g. Puntodewo et al.
1870 1994; McCaffrey 1996; Stevens et al. 2002; Bock
1871 et al. 2003). Left lateral shear is passed from the
1872 Papua and West Papua Tarera Aiduna Fault westwards into Maluku via
1873 the highly transpressive Seram fold thrust belt
1874 Oblique convergence at an angle of c. 608 between (Teas et al. 2009).
1875 Australia and the Pacific is accommodated across As in Sulawesi, Maluku and North Maluku,
1876 Papua and West Papua in a complex zone of strain extension is important within the overall convergent
1877 partitioning between shortening and left lateral orogen. Cenderawasih Bay and the adjacent Wai
1878 shear (e.g. Abers & McCaffrey 1988; McCaffrey poga Basin contain thick sediment piles (e.g. Dow
1879 1996). West of about 1388 E shortening is largely & Sukamto 1984; Pubellier et al. 1999; Charlton
1880 accommodated on a variety of structures in the 2010) and metamorphic core complex exhumation
1881 New Guinea Trench and Manokwari Trough, in at the Wandamen Peninsula (e.g. Bailly et al.
1882 the Mamberamo fold thrust belt and in the central 2009) indicates extreme lithospheric stretching.
1883 Highlands to the south (e.g. Milsom et al. 1992; Although extension may be related to processes
1884 Puntodewo et al. 1994; Stevens et al. 2002). The within the wide left lateral shear zone (Stevens
1885 largest earthquake to occur in eastern Indonesia et al. 2002), lessons from Sulawesi suggest that
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910 Fig. 19. (a) Bird’s Head (West Papua) digital elevation model (SRTM), multibeam bathymetry, CMT catalogue
1911 earthquakes ,35 km depth and structures showing evidence of Quaternary tectonic activity. Rivers marked in white.
1912 Location shown in Figure 1. Offshore structures north of the Koor Fault from Milsom et al. (1992). (b) Detail of the
1913 intersection between the Ransiki and Yapen faults, south of Manokwari. Eastern limit of image is to the east of the
1914 main map. SRTM onshore, multibeam bathymetry offshore.
 I. M. WATKINSON & R. HALL

1915 far field subduction related mechanisms may also present day left lateral strain accommodated across
1916 be significant. this zone. The April 1937 M 6.9 and April 1944 M
1917 7.2 and 7.4 earthquakes relocated by Okal (1999)
1918 Sorong Fault in West Papua were located on the onshore Sorong Fault 50
1919 100 km west of Manokwari and had focal mecha
1920 The Sorong Fault in West Papua is marked by a nisms indicating left lateral shear. Apparent right
1921 15 km wide zone of pronounced linear ridges and lateral motion between the Sorong and Biak GPS
1922 valleys trending ENE from northern Salawati stations, taken to lie on opposite sides of the Sorong
1923 through Sorong city and into the deep valley cutting Fault (Puntodewo et al. 1994), is complicated by
1924 across the northernmost mainland towards Manok other structures such as the Ransiki and Yapen
1925 wari in the east (Fig. 19a). Hamilton (1979) ques faults, which also lie between the stations.
1926 tioned whether this structure was significant in Numerous convincing left lateral stream offsets
1927 post Miocene tectonics, pointing out that parts of it of up to 300 m are documented in the central part
1928 were covered by post Miocene strata, and it is now of the fault valley (Dow & Sukamto 1984) (Fig.
1929 generally considered to be inactive (e.g. Puntodewo 20a). Similar sized displacements of Wallace
1930 et al. 1994; Decker et al. 2009; Charlton 2010). Creek crossing the San Andreas Fault have been
1931 There is little significant seismicity along dated to 13 259 years (Sieh & Jahns 1984). It is
1932 much of the fault and geodetic measurements sug unclear how long such offsets can be preserved in
1933 gest that both sides of the fault are broadly moving the landscape of an environment like West Papua,
1934 together and with the Pacific (e.g. Puntodewo et al. but it is unlikely they are pre Quaternary. Given
1935 1994; Stevens et al. 2002), with slight residual left that few such offsets are preserved in the more obvi
1936 lateral motion between the Sorong and Fakfak GPS ously active faults of eastern Indonesia, such as the
1937 stations possibly accommodated on the Sorong Palu Koro and Matano faults, the Sorong Fault
1938 Fault or the Koor Fault to the north (Bock et al. examples must reflect relatively recent and signifi
1939 2003). However, the Sorong GPS station is south cant strike slip. Mountain front sinuosity along
1940 of important strands of the Sorong Fault, which lie those segments of the fault associated with vertical
1941 offshore to the north and come onshore at Mega, motions is also conspicuously low, ranging from
1942 and the station is certainly south of the Koor Fault, 1.16 to 1.17 along segments NNE of Sorong city
1943 leaving substantial uncertainty in the amount of to 1.14 along the central section, where Dow &
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
Fig. 20. Evidence of Quaternary fault activity in the Bird’s Head. (a) Section of the onshore Sorong Fault showing
1968 a basin bounding normal fault in the north and two strike slip fault strands offsetting streams to the left in the south.
1969 Location shown in Figure 19a. ESRI imagery. (b) Ransiki delta at the southern end of the Ransiki Fault showing
1970 prominent western normal fault. Location shown in Figure 19b. (c) Region of co seismic subsidence showing
1971 flooded forest and buildings adjacent to the western normal fault. Google Earth Imagery. Image # 2016
1972 DigitalGlobe.
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

1973 Sukamto (1984) measured displaced streams and character to the Sorong Fault in West Papua (e.g.
1974 where triangular facets and shutter ridges are well Hamilton 1979; Dow & Sukamto 1984). In the
1975 developed. In the east, Smf values of 1.20 and 1.33 east, the Yapen Fault vanishes into the Mamberamo
1976 also suggest active tectonics. Faults adjacent to flat delta (Fig. 21a), where it forms a subtle linear valley
1977 topped Quaternary basins associated with Sorong delineated by active mud volcanoes (Dow &
1978 Fault releasing geometries are interpreted to be Sukamto 1984) and may dissipate into the Mambe
1979 dominantly normal faults (Fig. 20a) and these struc ramo fold thrust belt (Puntodewo et al. 1994). In
1980 tures have generally higher Smf values, including the west, the Yapen Fault has an unclear termina
1981 1.60, 1.61, 1.74 and 2.79. The average Vf value for tion, variously interpreted as being dextrally offset
1982 all these fault segments is 1.15, consistent with mod from the Sorong Fault along the Ransiki Fault
1983 erate to slow tectonic activity. (Puntodewo et al. 1994; Charlton 2010), linking/
1984 terminating against the Ransiki Fault (Milsom
1985 Koor Fault and Ransiki Fault et al. 1992) and unconnected to inactive Ransiki/
1986 onshore Sorong faults, but transferring strain
1987 The Koor Fault is an east west trending structure south to the Wandamen fault system (Bailly et al.
1988 20 30 km north of the Sorong Fault (Fig. 19a), 2009).
1989 which lies within a boundary zone between the oce Geodetic measurements indicate a fast left
1990 anic Pacific plate and continental crust in the south lateral slip rate of 46 + 12 mm a 1 across the Yapen
1991 (Dow & Sukamto 1984). The NNW trending Ran Fault (Bock et al. 2003), expressed by intense seis
1992 siki Fault (Fig. 19a) has been viewed as a dextral micity and focal mechanisms indicating left lateral
1993 shear zone linking the easternmost Sorong Fault slip along east west trending subvertical planes
1994 and the Yapen Fault (e.g. Robinson & Ratman (e.g. Okal 1999; Stevens et al. 2002). The 12 Sep
1995 1978; Milsom et al. 1992; Charlton 2010). tember 1979 M 7.9 tsunamigenic earthquake on
1996 Like the Sorong Fault in West Papua, both the the south coast of Yapen island (Fig. 21a) was asso
1997 Koor and Ransiki faults have been considered to ciated with sinistral slip along a ESE WNW trend
1998 be inactive (e.g. Hamilton 1979; Puntodewo et al. ing plane focused at a depth of 5 km and probably
1999 1994). However, a shallow M 7.6 earthquake on caused 2 m of displacement (Okal 1999).
2000 10 October 2002 at the southern end of the Ransiki The Randaway Fault Zone (Dow & Hartono
2001 Fault (Fig. 19b) had a focal mechanism and after 1982) is a set of NW SE trending faults onshore
2002 shock distribution consistent with dextral slip Yapen that link to strands of the Yapen Fault in
2003 along the Ransiki Fault (NEIC), although the possi the north (Fig. 21b). Interpreted as post Plio Pleisto
2004 bility of sinistral slip along a NE SW trending cene normal faults, they have previously been used
2005 splay of the Yapen Fault cannot be excluded. Topo to support a period of right lateral shear along the
2006 graphic and bathymetric data from the intersection Yapen Fault zone (Charlton 2010). However, we
2007 (Fig. 19b) could be interpreted to show the two saw no geomorphic evidence of significant normal
2008 structures curving gently into each other, leading faulting along the Randaway trend instead we
2009 to the possibility of contraction in the Ransiki area. saw a small linear basin and lake near the northern
2010 Mountain front sinuosity measured along two tip of the Randaway Fault at a left step over and a
2011 splays of the southern Ransiki Fault yields values deeply incised stream offset to the left by almost
2012 of 2.64 for a clearly inactive, c. north south 1 km both evidence of Quaternary sinistral shear
2013 trending southwestern strand, and 1.06 for the linear (Fig. 21b).
2014 fault bounding the southern margin of Ransiki delta Although the north coast of Yapen is remarkably
2015 (Figs 2p & 19b). The very low Smf and the asymmet straight and clearly fault controlled, the main fault
2016 rical position of the Ransiki River close to the fault mostly lies just offshore to the north, meaning that
2017 scarp (Fig. 20b) support recent extensional activity geomorphic indices could not be usefully measured
2018 along the fault. A 2 m high co seismic surface rup along the Yapen Fault. Multibeam bathymetry east
2019 ture formed close to the fault scarp during the of the island shows the Yapen Fault expressed by
2020 2002 earthquake and was associated with subsi a straight, narrow lineament marked by pressure
2021 dence of the delta that flooded a low lying church ridges and parallel to a prominent set of curvilinear
2022 (D. Gold, pers. comm. 2013), visible in satellite normal faults (Fig. 21c). Splays of the fault curving
2023 imagery to be coincident with a large region of to the WSW delimit at least two rhomboidal pull
2024 flooded forest (Fig. 20c). apart basins. At the western limit of the multibeam
2025 data a splay appears to enter a third pull apart basin,
2026 Yapen Fault which is associated with a prominent north south
2027 trending sidewall fault. It is significant that this
2028 The Yapen Fault (Fig. 21a) is a highly linear east structure is parallel to, and 60 km north, of the Wan
2029 west trending structure that crosses the 320 km damen Peninsula perhaps support for the south
2030 wide northern Cenderawasih Bay and is similar in wards transfer of sinistral shear from the Yapen
 I. M. WATKINSON & R. HALL

2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082 Fig. 21. Northern Papua and Cenderawasih Bay digital elevation model (SRTM), multibeam bathymetry, CMT
2083 catalogue earthquakes ,35 km depth and structures showing geomorphic evidence of Quaternary tectonic activity.
2084 Rivers marked in white. Location shown in Figure 1. (b) Expression of the Yapen and Randaway faults along the
northern coast of Pulau Yapen, showing evidence for Quaternary sinistral slip along the Randaway Fault. Inset
2085 shows the topography and major structures of Yapen. ESRI imagery. (c) Multibeam bathymetry detail showing the
2086 Yapen Fault to the west of Pulau Yapen; the southern strands appear to transfer to north south extension via a
2087 series of pull apart basins.
2088
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
Fig. 22. Evidence of Wandamen Peninsula Quaternary fault activity. (a) Overview digital elevation model (SRTM)
2123 showing bounding normal faults. Location shown in Figure 21a. (b) Pronounced triangular facets and hanging
2124 valleys along the eastern bounding fault system. ESRI imagery. (c) Inferred Quaternary fault trace across the top of
2125 alluvial fans crossing the western fault system. Google Earth imagery. Image # 2016 DigitalGlobe.
2126
2127
2128 Fault via a region of east west extension, as pro south (Puntodewo et al. 1994; McCaffrey 1996;
2129 posed by Bailly et al. (2009). Bock et al. 2003) (Fig. 1).
2130
2131 Mamberamo fold thrust belt Wandamen Peninsula faults
2132
2133 The Mamberamo fold thrust belt (Fig. 21a) proba The Wandamen Peninsula projects into Cenderawa
2134 bly accommodates some Australia Pacific shorten sih Bay from the eastern edge of the Lengguru fold
2135 ing in eastern Papua and lies north of the Highlands belt, and is bounded on the east and west sides by
2136 thrust belt of central New Guinea (e.g. Dow & north south trending faults (Fig. 22). We refer
2137 Sukamto 1984). Unlike the complex oblique con here specifically to these faults, not to the Wanda
2138 vergence and strain partitioning further west, the men Fault Zone of Dow & Sukamto (1984) that con
2139 belt contains relatively simple NW trending active nects the Sorong Fault with the Tarera Aiduna fault
2140 structures oriented normal to convergence (McCaf system via the Ransiki Fault.
2141 frey 1996). Despite intense and widespread seismic The peninsula is considered to represent the
2142 ity, less than 15 mm a 1 of shortening occurs across exhumed internal zone of the Lengguru fold belt
2143 the Mamberamo belt, leaving much of the remain and is composed of an amphibolite eclogite grade
2144 ing 45 mm a 1 Australia Pacific convergence and metamorphic dome rising to .2 km elevation
2145 100 mm a 1 of left lateral motion to offshore struc (Robinson et al. 1990; Bailly et al. 2009; Charlton
2146 tures to the north and the Highlands thrust belt to the 2010), which may be a metamorphic core complex
 I. M. WATKINSON & R. HALL

2147 (e.g. Hill et al. 2002). Seismicity and GPS vectors On the west of the peninsula, Smf values range
2148 either side of Cenderawasih Bay (Stevens et al. from 1.05 to 1.43, indicating maximal to rapid
2149 2002) suggest active extension accommodated on tectonic activity. A 21 km long section of the west
2150 north south trending structures close to the Wan ern fault system passing through Wasior shows
2151 damen Peninsula, which may connect to the western evidence of recent normal faulting (Fig. 22c).
2152 releasing termination array of the Yapen Fault in the Upper modern fan deltas are abruptly terminated
2153 north (Fig. 21c). by a linear scarp, above which are narrow truncated
2154 Normal faults bounding the peninsula are palaeofans. Rivers vertically incised into footwall
2155 expressed by curvilinear en echelon segments up palaeofans show little evidence of lateral erosion
2156 to 20 km long trending north south to NNW and small landslides are localized along the over
2157 SSE. These make up the east and west detachment steepened scarp. The scarp is marked by a linear
2158 systems of Bailly et al. (2009). Triangular facets, change in topography, lines of vegetation and
2159 hanging valleys and V shaped valleys are common often an abrupt change from meandering rivers up
2160 and indicate rapid tectonic activity (Fig. 22b). stream to anastomosing rivers downstream of the
2161 Two tiers of hanging valleys on the eroded scarp scarp. A southern continuation of the Wandamen
2162 of the eastern fault system are defined by changes fault system bounds the eastern margin of the
2163 in valley width or orientation at common elevations Wasimi delta and has an Smf value of 2.33, indicat
2164 along the scarp. They probably record variations ing slow to minimal tectonic activity.
2165 in the tectonic rate or climate during exhumation
2166 of the fault surface. Mountain front sinuosity values Other circum-Cenderawasih Bay structures
2167 of four segments on the east side are uniform
2168 at 1.25, 1.28 and 1.29, with one more eroded seg The locus of active Australia Pacific left lateral
2169 ment of 1.72. Fan deltas are well developed at strain partitioning shifts from the Yapen Fault sys
2170 relays between the fault segments, notably at Goni tem to the Tarera Aiduna Fault system across Cen
2171 and another smaller delta 21 km further north derawasih Bay, defining a 300 km wide shear zone
2172 (Fig. 22a). As well as localizing sediment transport, that involves a complex array of Quaternary faults
2173 the relays are likely to be sites of active displace within the two bounding strike slip zones (e.g. Ste
2174 ment minima, allowing subaerial delta progradation vens et al. 2002; Bock et al. 2003). Along the east
2175 on the hanging wall. ern margin of Cenderawasih Bay, the NE trending
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202 Fig. 23. Southern West Papua and Cenderawasih Bay digital elevation model (SRTM), multibeam bathymetry,
2203 CMT catalogue earthquakes ,35 km depth and structures showing geomorphic evidence of Quaternary tectonic
2204 activity. Rivers marked in white. Location shown in Figure 1. Offshore structures from Teas et al. (2009).
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

2205 Lowlands Fault Zone (bounding the Waipoga zone (Fig. 24a). In the west it passes across a low
2206 Trough of Visser & Hermes 1962) and the Paniai lying mangrove plain with minimal topographic
2207 Fault Zone are associated with thrust and left lateral relief. It is possible to trace several fault strands
2208 strike slip earthquakes (Fig. 23), offset drainage and from linear features revealed by abandoned river
2209 high fault scarps, indicating modern tectonic activ channels and coastline segments (Fig. 24b). Its cen
2210 ity (Pubellier et al. 1999; Stevens et al. 2002). The tral section is expressed by a series of linear ridges
2211 faults have a soft linkage with the Tarera Aiduna of moderate relief bounding a wide rhomboidal
2212 Fault system in the south and splays curve into par basin (Fig. 24c), across which the captured Aru
2213 allelism with the Yapen Fault and Mamberamo River passes into the Uruma River in the south.
2214 fold thrust belt in the north. The river is abruptly deflected as it crosses two
2215 The Lengguru fold belt (Visser & Hermes 1962) prominent fault strands, with 65 75 m left lateral
2216 lies SW of Cenderawasih Bay and the Wandamen displacement, which may reflect recent Tarera
2217 Peninsula, east of Bintuni Bay and the Bomberai Aiduna Fault slip (Fig. 24d, e), although this offset
2218 Peninsula, and is bounded by the Tarera Aiduna is rather speculative.
2219 fault system in the south (Fig. 23). Compressional An asymmetrical graben developed at the
2220 deformation terminated during the Pleistocene eastern termination of the Tarera Aiduna Fault is
2221 (Decker et al. 2009) and the belt is presently largely bounded by NE SW trending normal faults (Fig.
2222 inactive, except for a few earthquakes related to 24f ). Rivers pressed hard against the NW dipping
2223 gravitational collapse (Bailly et al. 2009), often bounding faults and a SE dipping set of antithetic
2224 with a left lateral component related to residual Tar faults indicate active subsidence. The easternmost
2225 era Aiduna strain. Tarera Aiduna Fault itself has a significant dip slip
2226 component, forming the northern margin of an
2227 Tarera Aiduna Fault 800 m high ridge. The Tarera Aiduna Fault and
2228 the eastern bounding normal fault have Smf values
2229 The Tarera Aiduna Fault (Visser & Hermes 1962) of 1.08 and 1.21, respectively, indicating that they
2230 is an east west trending left lateral shear zone that are both active. Bounding faults along the northern
2231 forms the southern boundary of the Lengguru fold margin of the rhomboidal basin, including segments
2232 belt and passes offshore to the west, north of the corresponding to the Aria River Fault of Hamilton
2233 Aru Trough (Fig. 23). The Tarera Aiduna Fault (1979), have Smf values of 1.63, 1.91 and .4.00,
2234 sensu stricto is part of a wide system of faults that pointing to slow to inactive tectonics.
2235 pass, via a diffuse zone of sinistral transpression,
2236 into the Seram fold thrust belt in the west (Teas
2237 et al. 2009). The fault system is at least 130 km Discussion
2238 long onshore (Fig. 24) and is expressed by straight Challenges
2239 lineaments clearly visible on satellite imagery
2240 (Hamilton 1979) and a set of en echelon folds (Katili The identification of Quaternary/modern fault
2241 1986). Including possible soft linkage to Seram via activity in eastern Indonesia has historically proved
2242 sinistral transpression within the Seram fold thrust difficult (e.g. Hamilton 1979; Dow & Sukamto
2243 belt, imaged in multibeam bathymetric data (Teas 1984; Puntodewo et al. 1994; Socquet et al. 2006;
2244 et al. 2009), the whole fault system may be Bailly et al. 2009; Teas et al. 2009). In part, this is
2245 .700 km long. Geodetic measurements show high because eastern Indonesia cannot be well described
2246 relative motion between the Bird’s Head north of in terms of rigid plate tectonics, involving instead
2247 the Tarera Aiduna Fault and GPS stations south diffuse boundaries and boundary linkages, litho
2248 of the fault, such as Aru and Timika (Bock et al. spheric strength heterogeneity and lower crustal
2249 2003). Earthquake focal mechanisms showing sinis flow (Hall 2011). All the fault zones in the region
2250 tral slip along east west trending vertical planes that are relatively well constrained by geodetic
2251 (e.g. Seno & Kaplan 1988) suggest that the motion data display strain gradients that can be explained
2252 onshore is seismic and occurs along a broad zone in terms of multiple fault strands, distributed defor
2253 (Fig. 23). Seismicity is largely absent west of the mation or elastic strain surrounding a locked fault
2254 Bomberai peninsula, suggesting either a wide zone (e.g. Walpersdorf et al. 1998; Rangin et al. 1999;
2255 of aseismic deformation linking the Tarera Aiduna Stevens et al. 2002; Bock et al. 2003; Socquet
2256 Fault with the Seram sinistral transpression (Teas et al. 2006). Poor historical earthquake records
2257 et al. 2009), a region of seismic deformation and few palaeoseismic data mean that it is difficult
2258 with recurrence times longer than the instrumental to distinguish between these options and so attention
2259 record, or no structural connection between the is naturally focused on geomorphologically promi
2260 two regions. nent faults and lineaments or structures with instru
2261 The onshore Tarera Aiduna Fault has a geomor mentally recorded seismicity. Faults or segments of
2262 phic expression typical of a major strike slip fault fault systems with recurrence intervals greater than
 I. M. WATKINSON & R. HALL

2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315 Fig. 24. (a) Map of the onshore Tarera Aiduna Fault showing structures with geomorphic evidence of Quaternary
2316 tectonic activity. Bold italic numbers are Smf values. Location shown in Figure 23. (b) Detail from greyscale
2317 Landsat TM 432 image showing linear confinement of abandoned River Aru channels, indicating strike slip strands
2318 across the plain. (c) Major strand of the Tarera Aiduna Fault bounding a steep sided ridge and rhomboidal basin.
2319 (d, e) Possible river offset across the Tarera Aiduna Fault. (f) Termination extensional fault array developed at the
2320 eastern end of the main Tarera Aiduna Fault strand. (d, e, f) from ESRI imagery.
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

2321 the short period of instrumental or even historical straight strike slip faults, is the presence of discon
2322 seismic records inevitably remain undocumented. tinuities or step overs (e.g. Segal & Pollard 1980;
2323 Additional challenges to the identification of Sibson 1985; Barka & Kadinsky Cade 1988). The
2324 Quaternary faults include thick forest over most of majority of historical strike slip earthquake ruptures
2325 the islands (e.g. Pubellier et al. 1999), the abun were arrested by step overs wider than 3 5 km
2326 dance of important structures located entirely off (Lettis et al. 2002; Wesnousky 2006). For example,
2327 shore and not readily available for study (e.g. the 1999 Mw 7.1 Düzce earthquake ruptured a 40 km
2328 Silver et al. 1983b; Henry & Das 2002; Teas et al. segment of the North Anatolian Fault (Aydın &
2329 2009; Liu & Harris 2013), the rapid erosion of tec Kalafat 2002) and terminated in the .4 km wide
2330 tonic landforms in the humid environment, the Eften releasing bend in the west and the 4 5 km
2331 rapid burial of co seismic features by a high sedi wide Bakacak releasing step over in the east
2332 ment flux (e.g. Suggate & Hall 2003) and the high (Duman et al. 2005). Straight, continuous faults are
2333 density of active and inactive structures within a therefore capable of generating larger earthquakes
2334 large region (e.g. Puntodewo et al. 1994; Stevens than curved or segmented faults, of generating rup
2335 et al. 2002). tures that penetrate below the seismogenic layer
2336 Of the 27 fault systems described here, none (King & Wesnousky 2007) and of sustained super
2337 can be confidently described as inactive during the shear rupture propagation, causing enhanced ground
2338 Quaternary. Eleven show evidence of ‘maximal’ motion (Robinson et al. 2010). Eastern Indonesia’s
2339 tectonic activity according to the classification sum major strike slip faults show a variety of levels of
2340 marized in McCalpin (2009) and a further five show segmentation, which may be viewed as an indi
2341 evidence of ‘rapid’ tectonic activity (Table 3). It is cation of their structural maturity, with high cumu
2342 important to note that the Quaternary faults dis lative displacements empirically known to remove
2343 cussed here are not exhaustive there are numerous fault zone complexities (e.g. Wesnousky 1988;
2344 other active faults in the region, in addition to major Stirling et al. 1996; King & Wesnousky 2007).
2345 offshore seismic sources, such as the Molucca Sea Other properties such as block rotation and pre
2346 collision complex, the Banda Sea and Molucca existing weaknesses may complicate this simple
2347 Sea subducted slabs, and ongoing subduction of relationship.
2348 the Celebes Sea (e.g. Cardwell & Isacks 1978; Sil The Matano Fault is an example of a structurally
2349 ver & Moore 1978; Cardwell et al. 1980; Silver immature fault zone. Its onshore length of 195 km is
2350 et al. 1983a; Engdahl et al. 1998), which also punctuated by three major basins, each one 4 6 km
2351 need to be taken into account in any hazard analysis. wide, and two major restraining bends. The resul
2352 tant maximum potential rupture length is 90 km
2353 Quaternary fault geometry and earthquakes (Table 3). Empirical rupture length magnitude
2354 relationships (Wells & Coppersmith 1994) suggest
2355 The largest earthquakes in eastern Indonesia have a potential M 7.4 earthquake for such a rupture
2356 been thrust and mega thrust events, including those length. Uncertainties in this estimate include the
2357 of the Seram Trough (1629, M . 8.5), the Banda unknown ability of a rupture to bypass the relatively
2358 Sea (1938, M . 8.0) and Biak (1996, Mw 8.2) gentle restraining bend east of the Mahalona Basin,
2359 (e.g. Wichmann 1918; Henry & Das 2002; Okal & the possibility of a through going strike slip fault at
2360 Reymond 2003; Liu & Harris 2013). However, seismogenic depths below Lake Matano, the effect
2361 many major historical earthquakes in the studied on fault strength of widespread serpentinite smears
2362 region have occurred on strike slip faults, in along the fault zone and the unknown length to
2363 cluding the Sorong Fault (1944, M 7.5), the Yapen which the fault continues offshore to the east.
2364 Fault (1979, M 7.9) the Ransiki Fault (2002, M The Sorong Fault in West Papua, part of the fault
2365 7.6) and perhaps the Kawa Fault (1899, M 7.8) system at the southern end of the Philippine Sea
2366 (e.g. Okal 1999; Brune et al. 2010; NEIC). Sixteen plate, is a much more established fault zone with a
2367 of the studied faults are dominantly strike slip and long history of slip (e.g. Ali & Hall 1995), reflected
2368 an additional five may have a substantial strike slip in an apparent absence of step overs .1 km and a
2369 component (Table 3). As they are often long, straight, continuous, straight onshore length of 420 km
2370 geometrically simple and subvertical, strike slip equating to a potential M . 8.0 earthquake if the
2371 faults are capable of generating large, shallow and entire linked system failed. The Mw 7.9 Yapen
2372 damaging earthquakes for example, the 1906 M earthquake of 1979 ruptured an unknown length of
2373 7.7 San Francisco earthquake (e.g. Wald et al. the potentially 420 km long quasi continuous
2374 1993), the 2001 Mw 7.8 Kunlun Shan earthquake Yapen Fault (Okal 1999), showing that such a sce
2375 (e.g. Lin et al. 2003) and the 2002 Mw 7.9 Denali nario is possible. Despite evidence that most left
2376 earthquake (e.g. Haeussler et al. 2004). lateral strain is focused south of the Sorong Fault
2377 A crucial barrier to the propagation of lateral in West Papua, a conservative slip rate estimate of
2378 ruptures and hence earthquake magnitude, even on 2 mm a 1 could accumulate 2 m of elastic
2436
2435
2434
2433
2432
2431
2430
2429
2428
2427
2426
2425
2424
2423
2422
2421
2420
2419
2418
2417
2416
2415
2414
2413
2412
2411
2410
2409
2408
2407
2406
2405
2404
2403
2402
2401
2400
2399
2398
2397
2396
2395
2394
2393
2392
2391
2390
2389
2388
2387
2386
2385
2384
2383
2382
2381
2380
2379

Table 3. Summary of observations made from Quaternary faults in eastern Indonesia, with hypothetical earthquake magnitudes, styles and tsunami risk

Fault Typical Maximum Step-over/ Potential Attributable Notable Smf range Vf range Tectonic Potential Potential Associated
segment observed total relay width rupture length seismicity historical activity earthquake earthquake tsunami?
length (km) length (km) (km)* events class† magnitude‡ style
(km)

Malino 25 –75 130 1.2 130 Y 1.05 –1.66 0.22– 1.01 Maximal to 7.6 Normal Y
boundary slow
Gorontalo 30 95 7 35 N 1.83 –2.36 0.88– 1.69 Slow to 6.9 Strike-slip Y
minimal
Palu–Koro 10 –35 220 ,1 135 Y Mw7.7, 1.08 –2.30 0.24– 0.89 Maximal to 7.6 Strike-slip Y
1996 slow
Parigi boundary 10 –45 95 3 80 Y 1.32 –3.25 0.50– 1.45 Minimal 7.3 Normal Y
Sapu valley 5–20 75 ? ,2 75 N 1.08 –1.45 0.40 Maximal 7.3 Strike-slip N
to
moderate
Balantak 54 250 10 (offshore) 54 Y 1.04 –1.22 0.25– 0.47 Maximal to 7.1 Strike-slip Y
moderate
I. M. WATKINSON & R. HALL

Matano 10 –60 195 6 90 Y Mw 6.1, 1.02 –1.9 0.23– 0.78 Maximal to 7.4 Strike-slip Y (lake)
2011 slow
Lawanopo and 10 –45 200 7 70 ? Mw 7.5, 1.21 –1.75 0.55– 0.83 Moderate to 7.2 Strike-slip N
Kendari 2001? slow
Towuti 25 –55 55 ,1 55 N 1.03 –2.04 0.41– 1.22 Maximal to 7.1 Normal Y (lake)
bounding slow
Kolaka 5–45 175 10 50 Y 1.05 –1.64 0.23– 1.68 Maximal to 7.0 Strike-slip Y
slow
Mangole 20 135 2 135 Y 1.12 –1.57 0.49– 0.55 Rapid to 7.6 Normal/ Y
slow strike-slip
Sanana 5–20 60 ? 60 Y 1.27 0.44 Rapid 7.2 Normal/strike-slip Y
Rana 10 65 3 –4 .40 N Sfc rupture? 1.01 –1.96 0.23– 1.53 Maximal to .6.9 ?Strike-slip N
slow
East Buru 10 48 ,1 48 Y 7.0 Strike-slip Y
Southern Seram 5–15 60 2 60 Y ?M 7.6, 1.84 –2.08 1.36– 1.88 Slow 7.2 Normal Y
1950
2494
2493
2492
2491
2490
2489
2488
2487
2486
2485
2484
2483
2482
2481
2480
2479
2478
2477
2476
2475
2474
2473
2472
2471
2470
2469
2468
2467
2466
2465
2464
2463
2462
2461
2460
2459
2458
2457
2456
2455
2454
2453
2452
2451
2450
2449
2448
2447
2446
2445
2444
2443
2442
2441
2440
2439
2438
2437
Kawa 15 –40 120 2 120 Y 1.10 –1.33 0.26– 0.28 Rapid to 7.5 Strike-slip Y
moderate
Bobol 10 –15 100 2.5 100 N 1.26 –1.99 1.04– 2.60 Moderate to 7.4 Strike-slip Y
minimal
Combined 10 –40 240 2.5 240 Y ?M 7.8, 1.10 –1.99 0.26– 2.66 Rapid to 7.8 Strike-slip Y
Kawa –Bobol 1899 minimal
Seram FTB 5–15 135 .2 ?135 Y 7.6 Thrust Y
Sorong 45 –75 420 ,1 420 Y M 7.4, 1944 1.14 –2.85 0.27– 8.68 Rapid to .8.0 Strike-slip Y
(West Papua) minimal
Koor 15 –35 100 6 75 Y 7.3 Strike-slip Y
Ransiki 20 –50 100 ,1 100 Y M 7.6, 2002 1.06 –2.64 Maximal to 7.4 Strike-slip Y
minimal
Yapen 30 –50 420 2 –3 420 Y M 7.9, 1979 .8.0 Strike-slip Y
Mamberamo 10 –20 180 ,2 ?180 Y Numerous 7.7 Thrust Y
Wandamen 6–20 55 2 55 Y Sfc rupture? 1.05 –2.33 Maximal to 7.1 Normal Y
boundary slow
Lowlands 30 –70 220 ? ? Y ? Normal/strike-slip N
Paniai ? 150 ? ? Y ? Normal/strike-slip N
Tarera –Aiduna 30 –60 130 7 90 Y 1.08 –4.58 Maximal to 7.4 Strike-slip Y
minimal

*Maximum length of segment(s) separated by step-overs ,3 km wide.

†
Using scheme modified after McCalpin (2009).
‡
Based on potential rupture length estimate and empirical length –magnitude relationships from Wells & Coppersmith (1994).
QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA
 I. M. WATKINSON & R. HALL

2495 displacement across the northern Bird’s Head (sim models (e.g. Wu et al. 2009) and natural strike slip
2496 ilar to the 1979 Yapen earthquake release; Okal basins (e.g. the Clonard Basin, Haiti; Mann et al.
2497 1999) in 1000 years. Even if all of this occurred 1995), means that the Palu Koro Fault may be
2498 west of the 1937 and 1944 earthquakes, and assum straighter and more continuous than previously sug
2499 ing complete stress release during those events, the gested and palaeoseismic trenches across the border
2500 remaining c. 200 km western portion of the fault faults may not record major historical strike slip
2501 could still generate a M . 7.7 earthquake. earthquakes. The postulated buried and locked
2502 The apparently very young and highly seg section alone is 50 km long and is thus capable of
2503 mented Tarera Aiduna fault zone and structures generating an M 7.0 earthquake. The total onshore
2504 in the near offshore Seram fold thrust belt (Teas length of the Palu Koro Fault between Leboni
2505 et al. 2009) and onshore Seram (Kawa and Bobol valley and Palu city, lacking step overs wider than
2506 faults) could be part of a single soft linked fault 1 km and bends greater than 58, is 135 km. As
2507 system and seem to partition much of the present such, the Palu Koro Fault must qualify as a ‘fault
2508 day left lateral motion between Australia and the superhighway’, potentially capable of sustained
2509 Bird’s Head. This fault system may thus be taking super shear rupture speeds (Robinson et al. 2010)
2510 over the Pre Pleistocene role of the Sorong Fault. and earthquakes up to M 7.6.
2511 Although there is not yet a through going fault on Other smaller structures that are geologically
2512 the scale of the Sorong Fault, the individual compo less significant because they are either not associ
2513 nents of the Tarera Aiduna Fault and left lateral ated with instrumental seismicity (e.g. the Sapu
2514 faults in Seram are each capable of generating valley fault system), have very low geomorphic
2515 M . 7 earthquakes. Geomorphic observations sug tectonic activity indices (e.g. the Gorontalo Fault)
2516 gest that they have all been active during the Quater or are composed of short and discontinuous fault
2517 nary, even if some segments (e.g. the Bobol Fault) segments (e.g. the Namlea fault system) are of par
2518 lack instrumental seismicity records. A major uncer ticular importance from a hazard analysis perspec
2519 tainty in assessing the Tarera Aiduna fault system tive because of their proximity to large population
2520 is the type and degree of linkage along its segments. centres with little to no earthquake resistance. Sim
2521 The longest segment onshore with geomorphic evi ilarly, structures such as the Kolaka Fault, which has
2522 dence for rapid activity is 60 km long and may its most geomorphologically youthful segment
2523 be traced, via an abrupt releasing bend 3 km wide, bounding steep uplifted topography immediately
2524 another 30 km to the east. Assuming rupture is not adjacent to Kolaka town, may also be associated
2525 terminated by the bend, an M 7.4 earthquake is pos with secondary seismic hazards such as landslides.
2526 sible on this 90 km long segment. It is reasonable to Large earthquakes along many of the faults, partic
2527 assume that the fault passes some distance offshore ularly the Palu Koro, Matano and Balantak faults
2528 before the next terminating step over, so the maxi and the Molino, Towuti and Wandamen Peninsula
2529 mum magnitude is likely to be larger. boundary faults, may also trigger local tsunami, as
2530 In a similar manner, the maximum potential has been already demonstrated in Palu and Taluti
2531 magnitudes for observed quasi continuous seg bays (e.g. Prasetya et al. 2001; Brune et al. 2010).
2532 ments of the Kawa and Bobol faults of southern
2533 Seram are 7.5 and 7.4, respectively, but a continuous
2534 rupture linking across Taluti Bay could achieve a Conclusions
2535 length of 240 km and an earthquake magnitude of
2536 7.8. The 1899 M 7.8 event, which caused slope fail Neotectonic deformation in eastern Indonesia is
2537 ure north of Tehoru and a tsunami around Taluti rarely focused on discrete shear zones bounding
2538 Bay, could have originated from such a rupture. rigid blocks, although this is often how it is inter
2539 The Palu valley has previously been considered preted. The pattern of seismicity and the broad
2540 to represent a pull apart basin between two strands distribution of Quaternary faults suggests that the
2541 of the Palu Koro Fault (Bellier et al. 2001, 2006; region is more closely approximated by continuum
2542 Beaudouin et al. 2003; Socquet et al. 2006). The mechanics than by rigid microplates (e.g. Thatcher
2543 width between the two strands would be about 1995). All of the studied faults show geomorphic
2544 6 km, ample to terminate earthquake rupture, limit evidence of Quaternary tectonic activity, even in
2545 ing the maximum length and magnitude of Palu areas where high strain rates are not inferred from
2546 Koro Fault earthquakes to the segments north and geodetic measurements (e.g. Buru, south Seram
2547 south of Palu valley. However, the possibility of a and northern West Papua).
2548 continuous, buried cross basin fault system within The zone of left lateral deformation that includes
2549 the Palu valley as proposed here has significant the Yapen Fault, the Tarera Aiduna Fault and
2550 implications for seismic hazard assessment in the strike slip associated with the Seram fold thrust
2551 densely populated valley. A continuous cross basin belt is perhaps the most active onshore/nearshore
2552 fault within the Palu valley, as seen in analogue fault system of eastern Indonesia as recorded by
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

2553 instrumental seismicity and geodetics. However, in Asia Research Group, Royal Holloway University of
2554 terms of seismic risk, the Palu Koro Fault is con London.
2555 sidered to be the most significant structure due to
2556 its proximity to Palu city, the possibility of a cross
2557 basin fault system close to the city, the fault’s unpre References
2558 dictability as a result of its poorly known seismic
2559 history, and the fault’s potential to cause large, shal Abers, G. & McCaffrey, R. 1988. Active deformation in
2560 the New Guinea Fold and Thrust Belt: seismological
low focus, super shear earthquakes. Additional fac
evidence for strike slip faulting and basement
2561 tors increasing the risk of the Palu Koro Fault involved thrusting. Journal of Geophysical Research,
2562 include the possibility of liquefaction in the deep 93, 13332 13354.
2563 Quaternary sedimentary basin on which Palu city Ahmad, W. 1978. Geology along the Matano Fault Zone,
2564 is built and the low lying city’s vulnerability to tsu East Sulawesi, Indonesia. In: Wiryosujono, S. &
2565 nami travelling down the narrow Palu Bay. Sudradjat, A. (eds) Proceedings: Regional Confer
2566 The Sorong Fault in West Papua should be ence on the Geology and Mineral Resources of South
2567 viewed as the wildcard of eastern Indonesian active East Asia, Jakarta, Indonesia, 4 7 August 1975. Indo
2568 tectonics. Although GPS measurements appear to nesian Association of Geologists, 143 150.
2569 Ali, J.R. & Hall, R. 1995. Evolution of the boundary
show little sinistral strike slip motion, or even a
between the Philippine Sea Plate and Australia: palae
2570 degree of dextral slip, station locations in Sorong omagnetic evidence from eastern Indonesia. Tectono
2571 and Biak cannot resolve the complexity of the physics, 251, 251 275.
2572 Sorong Yapen Ransiki Fault and may omit shear Audley Charles, M.G. 1986. Timor Tanimbar Trough:
2573 to the north. Convincing left lateral stream offsets the foreland basin to the evolving Banda orogen. In:
2574 and low mountain front sinuosity values show that Allen, P.A. & Homewood, P. (eds) Foreland Basins.
2575 the fault has been active during the Quaternary. A International Association of Sedimentologists, Special
2576 dearth of seismicity, rather than indicating that the Publications, 8, 91 102.
2577 fault is benign, may instead indicate that it is locked Audley Charles, M.G., Carter, D.J. & Milsom, J.
2578 1972. Tectonic development of Eastern Indonesia in
and accumulating elastic strain. Magnitude 6.9, 7.2
relation to Gondwanaland dispersal. Nature, 239,
2579 and 7.4 earthquakes in 1937 and 1944, located on 35 39.
2580 the fault west of Manokwari, prove that the fault is Aydın, A. & Kalafat, D. 2002. Surface ruptures of the
2581 capable of generating large earthquakes. The Sor 17 August and 12 November 1999 Izmit and Duzce
2582 ong Fault’s contribution to the seismic hazard of earthquakes in northwestern Anatolia, Turkey: their
2583 West Papua should not be underestimated, parti tectonic and kinematic significance and the associated
2584 cularly given its proximity to large towns such as damage. Bulletin of the Seismological Society of
2585 Sorong and Manokwari. America, 92, 95 106.
2586 There is great potential for the palaeoseismic Bailly, V., Pubellier, M., Ringenbach, J. C., de
2587 Sigoyer, J. & Sapin, F. 2009. Deformation zone
study of some of the faults discussed in this paper
‘jumps’ in a young convergent setting; the Lengguru
2588 to confirm Quaternary activity and to provide fold and thrust belt, New Guinea Island. Lithos, 113,
2589 more detailed answers to questions about seismic 306 317.
2590 hazards, particularly characteristic earthquake sizes Barka, A. & Kadinsky Cade, K. 1988. Strike slip fault
2591 and recurrence intervals. It is recommended that geometry in Turkey and its influence on earthquake
2592 trenching work is carried out across possible surface activity. Tectonics, 7, 663 684.
2593 ruptures identified along the Matano, Balantak, Bath, M. & Duda, S.J. 1979. Some Aspects of Global
2594 Rana, Ransiki and Wandamen Peninsula faults. Seismicity. Seismological Institute, Uppsala, 1 79.
2595 Geophysical studies to image shallow fault strands Beanland, S. & Clark, M. 1994. The Owens Valley
2596 Fault Zone, Eastern California, and Surface Faulting
in the Quaternary sedimentary fill of several strike
Associated with the 1872 Earthquake. US Geological
2597 slip basins, including the Palu, Sapu and Mahalona Survey Bulletin, 1982. US Government Printing
2598 valleys, would help to confirm the existence of Office, Washington DC.
2599 cross basin strike slip fault systems that may pose Beaudouin, T. 1998. Tectonique active et sismotectoni
2600 a previously unrecognized seismic hazard. que du système de failles décrochantes de Sulawesi
2601 Central. Thèse de Doctorat, Universitá Paris Sud.
2602 We are grateful to TGS (http://www.tgs.com), who pro Beaudouin, T., Bellier, O. & Sebrier, M. 2003.
2603 vided the multibeam data, to John Decker, Phil Teas and Present day stress and deformation fields within the
2604 the crew of the Shakti dive boat for support and valuable Sulawesi Island area (Indonesia): geodynamic implica
discussions during Banda Sea fieldwork. Alfend Rudya tions. Bulletin de la Société géologique de France, 174,
2605
wan, Dave Gold, Ferry Yulian, Ega Abang Surya Nugraha, 305 317.
2606 Jonathan Pownall and Benjamin Sapiie are thanked for Bellier, O., Beaudouin, T. et al. 1998. Active faulting
2607 their assistance and contributions to the ideas discussed in central Sulawesi (eastern Indonesia). In: Wilson, P.
2608 here. Two anonymous reviewers are thanked for detailed & Michel, G.W. (eds) The Geodynamics of S, SE Asia
2609 and constructive reviews, which considerably improved (GEODYSSEA Project). GeoForshingsZentrum, Pots
2610 the final paper. Recent fieldwork was funded by the SE dam, 276 312.
 I. M. WATKINSON & R. HALL

2611 Bellier, O., Sebrier, M. et al. 2001. High slip rate for a Asia Collision. Geological Society, London, Special
2612 low seismicity along the Palu Koro active fault in cen Publications, 355, 177 202, https://doi.org/10.
2613 tral Sulawesi (Indonesia). Terra Nova, 13, 463 470. 1144/SP355.9
2614 Bellier, O., Sébrier, M., Seward, D., Beaudouin, T., Decker, J., Bergman, S.C., Teas, P.A., Baillie, P. &
Villeneuve, M. & Putranto, E. 2006. Fission Orange, D.L. 2009. Constraints on the tectonic evolu
2615 track and fault kinematics analyses for new insight tion of the Bird’s Head, West Papua. In: Proceedings,
2616 into the Late Cenozoic tectonic regime changes in Indonesian Petroleum Association, 33rd Annual Con
2617 West Central Sulawesi (Indonesia). Tectonophysics, vention and Exhibition, IPA09 G 139.
2618 413, 201 220. Dehbozorgi, M., Pourkermani, M., Arian, M., Mat
2619 Bock, Y., Prawirodirdjo, L. et al. 2003. Crustal motion kan, A.A., Motamedi, H. & Hosseiniasl, A. 2010.
2620 in Indonesia from global positioning system measure Quantitative analysis of relative tectonic activity in
2621 ments. Journal of Geophysical Research, 108, 2367. the Sarvestan area, central Zagros, Iran. Geomorphol
2622 Brune, S., Babeyko, A.Y., Ladage, S. & Sobolev, S.V. ogy, 121, 329 341.
2623 2010. Landslide tsunami hazards in the Indonesian DeMets, C., Gordon, R., Argus, D. & Stein, S. 1994.
Sunda Arc. Natural Hazards Earth Systems Science, Effects of recent revisions to the geomagnetic reversal
2624 10, 589 604. time scale on estimates of current plate motions. Geo
2625 Bull, W.B. 1978. Geomorphic Tectonic Activity Classes physical Research Letters, 21, 2191 2194.
2626 of the South Front of the San Gabriel Mountains, Cal Dow, D.B. & Hartono, U. 1982. The nature of the crust
2627 ifornia. Final Report, US Geological Survey, Contract underlying Cenderawasih (Geelvink) Bay, Irian Jaya.
2628 No. 14 08 0001 G 394. Proceedings of the Indonesian Petroleum Association,
2629 Bull, W.B. 2007. Tectonic Geomorphology of Mountains: 11, 203 210.
2630 a New Approach to Paleoseismology. Blackwell Pub Dow, D.B. & Sukamto, R. 1984. Western Irian Jaya: the
2631 lishing, Malden, MA. end product of oblique plate convergence in the Late
2632 Bull, W.B. & McFadden, L.D. 1977. Tectonic geomor Tertiary. Tectonophysics, 106, 109 139.
phology north and south of the Garlock fault, Califor Duman, T.Y., Emre, O., Dogan, A. & Ozalp, S. 2005.
2633 nia. In: Doehering, D.O. (ed.) Geomorphology in Arid Step over and bend structures along the 1999 Duzce
2634 Regions. Proceedings at the Eighth Annual Geomor Earthquake surface rupture, North Anatolian Fault,
2635 phology Symposium. State University of New York, Turkey. Bulletin of the Seismological Society of
2636 Binghamton, NY, 115 138. America, 95, 1250 1262, https://doi.org/10.1785/
2637 Burchfiel, B.C., Royden, L.H. et al. 2008. A geologi 0120040082
2638 cal and geophysical context for the Wenchuan earth Engdahl, E.R., van der Hilst, R. & Buland, R.
2639 quake of 12 May 2008, Sichuan, People’s Republic 1998. Global teleseismic earthquake relocation with
2640 of China. Geological Society of America Today, 18, improved travel times and procedures for depth deter
2641 4 11. mination. Bulletin of the Seismological Society of
Camplin, D.J. & Hall, R. 2014. Neogene history of None America, 88, 722 743.
2642 Gulf, Sulawesi, Indonesia. Marine and Petroleum Eshghi, S. & Zare, M. 2003. Bam (SE Iran) Earthquake
2643 Geology, 57, 88 108. of 26 December 2003, Mw 6.5. A Preliminary Recon
2644 Cardwell, R.K. & Isacks, B.L. 1978. Geometry of the naissance Report, www.iiees.ac.ir/en/wp content/
2645 subducted lithosphere beneath the Banda Sea in eastern uploads/2010/12/Bam report zare.pdf [last accessed
2646 Indonesia from seismicity and fault plane solutions. 29 October 2016].
2647 Journal of Geophysical Research, 83, 2825 2838. Ferdian, F., Hall, R. & Watkinson, I. 2010. A structural
2648 Cardwell, R.K., Isacks, B.L. & Karig, D.E. 1980. The re evaluation of the north Banggai Sula area, eastern
2649 spatial distribution of earthquakes, focal mechanism Indonesia. In: Proceedings, Indonesian Petroleum
2650 solutions, and subducted lithosphere in the Philippine Association, 34th Annual Convention, IPA07 G 009,
and northeastern Indonesian islands. In: Hayes, D.E. 1 20.
2651 (ed.) The Tectonic, Geologic Evolution of Southeast Fu, B., Ninomiya, Y., Lei, X., Toda, S. & Awata, Y.
2652 Asian Seas, Islands. American Geophysical Union, 2004. Mapping active fault associated with the 2003
2653 Geophysical Monograph, 23, 1 35. Mw 6.6 Bam (SE Iran) earthquake with ASTER 3D
2654 Charlton, T.R. 2010. The Pliocene Recent anticlock images. Remote Sensing of Environment, 92, 153 157.
2655 wise rotation of the Bird’s Head, the opening of the Garrard, R.A., Supandjono, J.B. & Surono, 1988. The
2656 Aru Trough Cenderawasih Bay sphenochasm, and geology of the Banggai Sula microcontinent, eastern
2657 the closure of the Banda double arc. In: Proceedings, Indonesia. In: Proceedings, Indonesian Petroleum
2658 Indonesian Petroleum Association, 34th Annual Con Association, 17th Annual Convention, 23 52.
2659 vention and Exhibition. IPA10 G 008. Germeraad, J.H. 1946. Geology of central Seran. In:
Chen, W. P. & Hsu, L. 2013. Historic seismicity near the Rutten, L. & Hotz, W. (eds) Geological, Petro
2660 source zone of the great 2008 Wenchuan earthquake: graphical, and Palaeontological Results of Explora EDQ4
2661 implications for seismic hazards. Tectonophysics, tions Carried Out September 1917 June 1919,
2662 584, 114 118. Island of Ceram. de Bussy, Amsterdam.
2663 Cottam, M.A., Hall, R., Forster, M.A. & Boudagher Haeussler, P.J., Schwartz, D.P., Dawson, T.E.,
2664 Fadel, M.K. 2011. Basement character and basin for Stenner, H.D., Lienkaemper, J.J., Sherrod, B. &
2665 mation in Gorontalo Bay, Sulawesi, Indonesia: new Personius, S.F. 2004. Surface rupture and slip distri
2666 observations from the Togian Islands. In: Hall, R., bution of the Denali and Totschunda faults in the 3
2667 Cottam, M.A. & Wilson, M.E.J. (eds) The SE November 2002 M 7.9 earthquake, Alaska. Bulletin
2668 Asian Gateway: History, Tectonics of the Australia of the Seismological Society of America, 94, S23 S52.
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

2669 Haffner, G.D., Hehanussa, P.E. & Hartoto, D. 2001. Province, Eastern China: implications for repeated
2670 The biology and physical processes of large lakes of great earthquakes in the Holocene. Tectonophysics,
2671 Indonesia: Lakes Matano and Towuti. In: Munawar, 219, 283 304.
2672 M. & Hecky, R.E. (eds) The Great Lakes of the Hutchison, C.S. 1989. Geological Evolution of South
World (GLOW): Food Web, Health, Integrity, Ecovi East Asia. Clarendon Press, Oxford.
2673 sion. Ecovision World Monograph Series. Backhuys Jablonski, D., Priyono, P., Westlake, S. & Larsen,
2674 Publishers, Leiden, 183 192. O.A. 2007. Geology and exploration potential of the
2675 Hall, R. 1987. Plate boundary evolution in the Halmahera Gorontalo basin, central Indonesia eastern extension
2676 region, Indonesia. Tectonophysics, 144, 337 352. of the North Makassar Basin? In: Proceedings, Indone
2677 Hall, R. 1996. Reconstructing Cenozoic SE Asia. In: sian Petroleum Association, 31st Annual Convention
2678 Hall, R. & Blundell, D.J. (eds) Tectonic Evolution and Exhibition. IPA07 G 083.
2679 of SE Asia. Geological Society, London, Special Pub Katili, J.A. 1970. Additional evidence of transcurrent
2680 lications, 106, 153 184, https://doi.org/10.1144/ faulting in Sumatra and Sulawesi. Bandung National
2681 GSL.SP.1996.106.01.11 Institute of Geology and Mining Bulletin, 3, 15 28.
Hall, R. 2002. Cenozoic geological and plate tectonic Katili, J.A. 1973. On fitting certain geological and geo
2682 evolution of SE Asia and the SW Pacific: computer physical features of the Indonesian island arc to the
2683 based reconstructions, model and animations. Journal new global tectonics. In: Coleman, P.J. (ed.) The
2684 of Asian Earth Sciences, 20, 353 434. Western Pacific: Island Arcs, Marginal Seas, Geo
2685 Hall, R. 2011. Australia SE Asia collision: plate tecton chemistry. Western Australia University Press, Perth,
2686 ics and crustal flow. In: Hall, R., Cottam, M.A. & 287 305.
2687 Wilson, M.E.J. (eds) The SE Asian Gateway: History, Katili, J.A. 1975. Volcanism and plate tectonics in the
2688 Tectonics of the Australia Asia Collision. Geological Indonesian island arcs. Tectonophysics, 26, 165 188.
2689 Society, London, Special Publications, 355, 75 109, Katili, J.A. 1986. Geology and hydrocarbon potential of
2690 https://doi.org/10.1144/SP355.5 the Arafura Sea. In: Halbouty, M.T. (ed.) Future
Hall, R. 2012. Late Jurassic Cenozoic reconstructions of Petroleum Provinces of the World. AAPG Memoirs,
2691 the Indonesian region and the Indian Ocean. Tectono 40, 487 501.
2692 physics, 570, 1 41. Kavalieris, I., van Leeuwen, T.M. & Wilson, M. 1992.
2693 Hall, R., Fuller, M., Ali, J.R. & Anderson, C.D. 1995. Geological setting and styles of mineralization, north
2694 The Philippine Sea plate: magnetism and reconstruc arm of Sulawesi, Indonesia. Journal of Southeast
2695 tions. In: Taylor, B. & Natland, J.H. (eds) Active Asian Earth Sciences, 7, 113 130.
2696 Margins, Marginal Basins: a Synthesis of Western Keller, E.A. 1986. Investigation of active tectonics: use
2697 Pacific Drilling Results. American Geophysical of surficial Earth processes. In: Wallace, R.E. (ed.)
2698 Union Monograph, 88, 371 404. Active Tectonics, Studies in Geophysics. National
2699 Hamilton, W. 1979. Tectonics of the Indonesian Region. Academy Press, Washington, DC, 136 147.
US Geological Survey, Professional Papers, 1078. King, G.C.P. & Wesnousky, S.G. 2007. Scaling of fault
2700 Harris, R.A. 1992. Peri collisional extension and the for parameters for continental strike slip earthquakes. Bul
2701 mation of Oman type ophiolites in the Banda Arc and letin of the Seismological Society of America, 97,
2702 Brooks Range. In: Parson, L.M., Murton, B.J. & 1833 1840.
2703 Browning, P. (eds) Ophiolites, their Modern Oceanic Lacassin, R., Replumaz, A. & Leloup, P.H. 1998.
2704 Analogues. Geological Society, London, Special Pub Hairpin river loops and slip sense inversion on
2705 lications, 60, 301 325, https://doi.org/10.1144/ Southeast Asian strike slip faults. Geology, 26,
2706 GSL.SP.1992.060.01.19 703 706.
2707 Hennig, J., Advokaat, E., Rudyawan, A. & Hall, R. Lettis, W., Bachhuber, J. et al. 2002. Influence of
2708 2014. Large sediment accumulations and major sub releasing step overs on surface fault rupture and fault
sidence offshore; rapid uplift on land: consequences segmentation: examples from the 17 August 1999
2709 of extension of Gorontalo Bay and Northern Sulawesi. Izmit earthquake on the North Anatolian Fault, Turkey.
2710 In: Proceedings, Indonesian Petroleum Association, Bulletin of the Seismological Society of America, 92,
2711 38th Annual Convention and Exhibition, IPA14 19 42.
2712 G 304. Lin, A., Kikuchi, M. & Fu, B. 2003. Rupture segmenta
2713 Henry, C. & Das, S. 2002. The Mw 8.2, 17 February 1996 tion and process of the 2001 Mw 7.8 central Kunlun
2714 Biak, Indonesia, earthquake: rupture history, after earthquake, China. Bulletin of Seismological Society
2715 shocks, and fault plane properties. Journal of Geo of America, 93, 2477 2492.
2716 physical Research, 107, https://doi.org/10.1029/ Linthout, K., Helmers, H., Sopaheluwakan, J. &
2717 2001jb000796 Surya Nila, E. 1989. Metamorphic complexes in
Hill, K.C., Hoffman, N., Lunt, P. & Paul, R. 2002. Buru and Seram, northern Banda Arc. Netherlands
2718 Structure and hydrocarbons in the Sareba block, Journal of Sea Research, 24, 345 356.
2719 ‘Bird’s Neck’, West Papua. In: Proceedings, Indone Linthout, K., Helmers, H. & Andriessenm, P.A.M.
2720 sian Petroleum Association, 28th Annual Convention, 1991. Dextral strike slip in Central Seram and 3
2721 IPA01 G 115, 227 248. 4.5 Ma Rb/Sr ages in pre Triassic metamorphics
2722 Holzer, L.T. & Savage, J.C. 2013. Global earthquake related to Early Pliocene counterclockwise rotation
2723 fatalities and population. Earthquake Spectra, 29, of the Buru Seram microplate (E. Indonesia). Journal
2724 155 175. of Southeast Asian Earth Sciences, 6, 335 342.
2725 Huang, W. 1993. Morphologic patterns of stream chan Liu, Z.Y. C. & Harris, R.A. 2013. Discovery of possible
2726 nels on the active Yishi Fault, southern Shandong mega thrust earthquake along the Seram Trough from
 I. M. WATKINSON & R. HALL

2727 records of 1629 tsunami in eastern Indonesian region. mechanism to a historical event. Earth and Planetary
2728 Natural Hazards, 72, 1311 1328. Science Letters, 216, 1 15.
2729 Mann, P. 2007. Global catalogue, classification and Pairault, A.A., Hall, R. & Elders, C.F. 2003. Tectonic
2730 tectonic origins of restraining and releasing bends Evolution of the Seram Trough, Indonesia. In:
on active and ancient strike slip fault systems. In: Proceedings, Indonesian Petroleum Association, 29th
2731 Cunningham, W.D. & Mann, P. (eds) Tectonics of Annual Convention, 355 370.
2732 Strike Slip Restraining and Releasing Bends. Geologi Pavlides, S.B. 1989. Looking for a definition of neotec
2733 cal Society, London, Special Publications, 290, tonics. Terra Nova, 1, 233 235, https://doi.org/10.
2734 13 142, https://doi.org/10.1144/SP290.2 1111/j.1365 3121.1989.tb00362.x
2735 Mann, P., Taylor, F.W., Lawrence Edwards, R. & Pelinovsky, E., Yuliadi, D., Prasetya, G. & Hidayat,
2736 Teh Lung, Ku. 1995. Actively evolving microplate R. 1997. The 1996 Sulawesi tsunami. Natural Hazards,
2737 formation by oblique collision and sideways motion 16, 29 38.
2738 along strike slip faults: an example from the northeast Pholbud, P., Hall, R., Advokaat, E., Burgess, P. &
2739 ern Caribbean plate margin. Tectonophysics, 246, Rudyawan, A. 2012. A new interpretation of
1 69. Gorontalo Bay, Sulawesi. In: Proceedings, Indonesian
2740 McCaffrey, R. 1989. Seismological constraints and spec Petroleum Association, 36th Annual Convention,
2741 ulations on Banda arc tectonics. Netherlands Journal IPA12 G 029, 197 224.
2742 of Sea Research, 24, 141 152. Piety, L., Sullivan, J.T. & Anders, M.H. 1992. Segmen
2743 McCaffrey, R. 1996. Slip partitioning at convergent plate tation and paleoseismicity of the Grand Valley Fault,
2744 boundaries of SE Asia. In: Hall, R. & Blundell, D.J. southeastern Idaho and western Wyoming. In: Link,
2745 (eds) Tectonic Evolution of Southeast Asia. Geological P.K., Kuntz, M.A. & Platt, L.B. (eds) Regional
2746 Society, London, Special Publications, 106, 3 18, Geology of Eastern Idaho and Western Wyoming.
2747 https://doi.org/10.1144/GSL.SP.1996.106.01.02 Geological Society of America, Memoirs, 179,
2748 McCaffrey, R. & Sutardjo, R. 1982. Reconnaissance 155 182.
microearthquake survey of Sulawesi, Indonesia. Geo Pigram, C.J. & Panggabean, H. 1984. Rifting of the east
2749 physical Research Letters, 9, 793 796. ern margin of the Australian continent and the origin of
2750 McCalpin, J.P. (ed.) 2009. Palaeoseismology. 2nd ed. some microcontinents in Indonesia. Tectonophysics,
2751 International Geophysics Series, 95. Academic Press, 107, 331 353.
2752 Burlington, USA. Pigram, C.J. Surono & Supandjono, J.B. 1985. Origin
2753 Michetti, A.M., Audemard, F.A. & Marco, S. 2005. of the Sula Platform, eastern Indonesia. Geology, 13,
2754 Future trends in palaeoseismology: integrated study 246 248.
2755 of the seismic landscape as a vital tool in seismic haz Pownall, J.M., Hall, R. & Watkinson, I.M. 2013.
2756 ard analyses. Tectonophysics, 408, 3 21. Extreme extension across Seram and Ambon, eastern
2757 Milsom, J., Masson, D., Nichols, G., Sikumbang, N., Indonesia: evidence for Banda slab rollback. Solid
Dwiyanto, B., Parson, L. & Kallagher, H. 1992. Earth, 4, 277 314.
2758 The Manokwari Trough and the western end of the Pownall, J.M., Hall, R., Armstrong, R.A. & Forster,
2759 New Guinea Trench. Tectonics, 11, 145 153. M.A. 2014. Earth’s youngest known ultrahigh temper
2760 Molnar, P. & Dayem, K.E. 2010. Major intercontinental ature granulites discovered on Seram, eastern Indone
2761 strike slip faults and contrasts in lithospheric strength. sia. Geology, 42, 279 282.
2762 Geosphere, 6, 444 467. Prasetya, G.S., de Lange, W.P. & Healy, T.R. 2001.
2763 Natawidjaja, D.H. & Daryono, M.R. 2014. The Lawa The Makassar Strait Tsunamigenic Region, Indonesia.
2764 nopo Fault, Central Sulawesi, East Indonesia. In: Pro Natural Hazards, 24, 295 307.
2765 ceedings, 4th International Symposium on Earthquake Pubellier, M., Desfontaines, B., Chorowicz, J.,
2766 and Disaster Mitigation 2014 (ISEDM 2014). 11 12 Rudant, J. P. & Permana, H. 1999. Active denuda
November 2014, Bandung, Indonesia, https://doi. tion morphostructures from SAR ERS 1 images (SW
2767 org/10.1063/1.4915009 Irian Jaya). International Journal of Remote Sensing,
2768 NATIONAL EARTHQUAKE INFORMATION CENTRE (NEIC). 20, 789 800.
2769 http://earthquake.usgs.gov/contactus/golden/neic.php Puntodewo, S.S.O., McCaffrey, R. et al. 1994. GPS
2770 [last accessed 29 October 2016]. measurements of crustal deformation within the
2771 NATIONAL GEOPHYSICAL DATA CENTER/WORLD DATA SER Pacific Australia plate boundary zone in Irian Jaya,
2772 VICE. Global Significant Earthquake Database. National Indonesia. Tectonophysics, 237, 141 153.
2773 Geophysical Data Center, NOAA, https://doi.org/10. Quigley, M., van Dissen, R. et al. 2012. Surface rupture
2774 7289/V5TD9V7K [last accessed 29 October 2016]. during the 2010 Mw 7.1 Darfield (Canterbury) earth
2775 Norvick, M.S. 1979. The tectonic history of the Banda quake: implications for fault rupture dynamics and
arcs, eastern Indonesia: a review. Journal of the Geo seismic hazard analysis. Geology, 40, 55 58.
2776 logical Society, London, 136, 519 527, https://doi. Ramı́rez Herrera, M.T. 1998. Geomorphic assessment
2777 org/10.1144/gsjgs.136.5.0519 of active tectonics in the Acambay Graben, Mexican
2778 Okal, E.A. 1999. Historical seismicity and seismotectonic Volcanic Belt. Earth Surface Processes and Land
2779 context of the great 1979 Yapen and Biak, Irian Jaya forms, 23, 317 332.
2780 earthquakes. Pure and Applied Geophysics, 154, Rangin, C., Le Pichon, X. et al. 1999. Plate conver
2781 633 675. gence measured by GPS across the Sundaland/Philip
2782 Okal, E.A. & Reymond, D. 2003. The mechanism of pine Sea plate deformed boundary: the Philippines and
2783 great Banda sea earthquake of 1 February 1938: apply eastern Indonesia. Geophysical Journal International,
2784 ing the method of preliminary determination of focal 139, 296 316.
 QUATERNARY FAULT ACTIVITY IN EASTERN INDONESIA

2785 Robinson, D.P., Das, S. & Searle, M.P. 2010. Earth Socquet, A., Vigny, C., Chamot Rooke, N., Simons,
2786 quake fault superhighways. Tectonophysics, 493, W., Rangin, C. & Ambrosius, B. 2006. India and
2787 236 243. Sunda plates motion and deformation along their boun
2788 Robinson, G.P. & Ratman, N. 1978. The stratigraphic dary in Myanmar determined by GPS. Journal of Geo
and tectonic development of the Manokwarai area, physical Research, 111, https://doi.org/10.1029/
2789 Irian Jaya. BMR Journal of Australian Geology & Geo 2005JB003877
2790 physics, 3, 19 24. Sohoni, P.S., Malik, J.N., Merh, S.S. & Karanth, R.V.
2791 Robinson, G.P., Ryburn, R.J., Harahap, B.H., Tobing, 1999. Active tectonics astride Katrol Hill Zone,
2792 S.L. & Achdan, A. 1990. Geology of the Steenkool Kachchh, Western India. Journal of the Geological
2793 Sheet Area, Irian Jaya. 1:250,000 Scale. Geological Society of India, 53, 579 586.
2794 Research and Development Centre, Bandung. Spakman, W. & Hall, R. 2010. Surface deformation and
2795 Rockwell, T.K., Keller, E.A. & Johnson, D.L. 1984. slab mantle interaction during Banda Arc subduction
2796 Tectonic geomorphology of alluvial fans and mountain rollback. Nature Geoscience, 3, 562 566.
2797 fronts near Ventura, California. In: Morisawa, M. Spencer, J.E. 2010. Structural analysis of three exten
(ed.) Tectonic Geomorphology. Proceedings of the sional detachment faults with data from the 2000
2798 15th Annual Geomorphology Symposium. Allen and Space Shuttle Radar Topography Mission. Geological
2799 Unwin, Boston, MA, 183 207. Society of America Today, 20, 4 10.
2800 Roques, D. 1999. The Metamorphic Core of Buru. South Spencer, J.E. 2011. Gently dipping normal faults identi
2801 east Asia Research Group, London, Report No. 204. fied with Space Shuttle radar topography data in central
2802 Royden, L.H. 1993. Evolution of retreating subduction Sulawesi, Indonesia, and some implications for fault
2803 boundaries formed during continental collision. Tec mechanics. Earth and Planetary Science Letters, 308,
2804 tonics, 12, 629 638. 267 276.
2805 Rudyawan, A. 2011. Tectonostratigraphy and structural Stein, S., Geller, R.J. & Liu, M. 2012. Why earthquake
2806 style of the South Banggai Sula Block and NW Banda hazard maps often fail and what to do about it. Tecto
Basin, Indonesia. MSc thesis, University of London. nophysics, 562– 563, 1 25.
2807 Segal, P. & Pollard, D.D. 1980. Mechanics of discontin Stevens, C.W., McCaffrey, R. et al. 1999. GPS
2808 uous faults. Journal of Geophysical Research, 85, evidence for rapid rotations about a vertical axis in a
2809 4337 4350. collisional setting: the Palu fault of Sulawesi, Indone
2810 Seno, T. & Kaplan, D.E. 1988. Seismotectonics of west sia. Geophysical Research Letters, 26, 2677 2680.
2811 ern New Guinea. Journal of Physics of the Earth, 36, Stevens, C.W., McCaffrey, R., Bock, Y., Genrich,
2812 107 124. J.F., Pubellier, M. & Subarya, C. 2002. Evidence
2813 Sibson, R.H. 1985. Stopping of earthquake ruptures at for block rotations and basal shear in the world’s fastest
2814 dilational fault jogs. Nature, 316, 248 251. slipping continental shear zone in NW New Guinea. In:
2815 Sieh, K.E. & Jahns, R.H. 1984. Holocene activity of the Stein, S. & Freymueller, J. (eds) Plate Boundary
San Andreas fault at Wallace Creek, California. Jour Zones. American Geophysical Union, Geodynamics
2816 nal of Geophysical Research, 95, 883 896. Series, 30, 87 99.
2817 Silva, P.G., Goy, J.L., Zazo, C. & Bardajı́, T. 2003. Stirling, M.W., Wesnousky, S.G. & Shimazaki, K.
2818 Fault generated mountain fronts in southeast Spain: 1996. Fault trace complexity, cumulative slip, and
2819 geomorphologic assessment of tectonic and seismic the shape of the magnitude frequency distribution for
2820 activity. Geomorphology, 50, 203 225. strike slip faults: a global survey. Geophysical Journal
2821 Silver, E.A. & Moore, J.C. 1978. The Molucca Sea International, 124, 833 868.
2822 collision zone, Indonesia. Journal of Geophysical Suggate, S. & Hall, R. 2003. Predicting sediment yields
2823 Research, 83, 1681 1691. from SE Asia: a GIS approach. In: Proceedings, Indo
2824 Silver, E.A., McCaffrey, R. & Smith, R.B. 1983a. Col nesian Petroleum Association, 29th Annual Conven
lision, rotation and the initiation of subduction in the tion, IPA03 G 015, 289 304.
2825 evolution of Sulawesi, Indonesia. Journal of Geophys Sukamto, R. & Simandjuntak, T.O. 1983. Tectonic
2826 ical Research, 88, 9407 9418. relationship between geologic provinces of western
2827 Silver, E.A., McCaffrey, R., Joyodiwiryo, Y. & Ste Sulawesi, eastern Sulawesi and Banggai Sula in the
2828 vens, S. 1983b. Ophiolite emplacement by collision light of sedimentological aspects. Bulletin Geological
2829 between the Sula Platform and the Sulawesi Island Research and Development Centre, Bandung, 7, 1 12.
2830 Arc, Indonesia. Journal of Geophysical Research, 88, Surono & Bachri, S. 1994. Stratigraphy of the southeast
2831 9419 9435. Sulawesi continental terrane, Eastern Indonesia. Jour
2832 Simandjuntak, T.O. 1986. Sedimentology and tectonics nal of Geology and Mineral Resources, 4, 4 11.
2833 of the collision complex in the East Arm of Sulawesi, Teas, P.A., Decker, J., Orange, D. & Baillie, P.
Indonesia. PhD thesis, University of London. 2009. New insight into structure and tectonics of the
2834 Simandjuntak, T.O., Surono & Sukido. 1984. Geolog Seram Trough from SEASEEPTM high resolution
2835 ical Report of the Kolaka Quadrangle, 394 Sulawesi, bathymetry. In: Proceedings, Indonesian Petroleum
2836 Scale 1:250,000. Open File Report. Indonesian Geo Association, 33rd Annual Convention and Exhibition,
2837 logical Research and Development 395 Centre, IPA09 G 091.
2838 Bandung. Thatcher, W. 1995. Microplate v. continuum descrip
2839 Simandjuntak, T.O., Surono & Sukido. 1994. Geo tions of active tectonic deformation. Journal of Geo
2840 logical Map of the Kolaka Quadrangle, Sulawesi, physical Research, 100, 3885 3894.
2841 1:250,000 scale. Geological Research and Develop Tjokrosapoetro, S., Budhitrisna, T. & Rusmana, D.
2842 ment Centre, Bandung. 1981. Geological Map of the Buru Island Quadrangle,
 I. M. WATKINSON & R. HALL

2843 Maluku, 1:250,000 Scale. Geological Research and (Sulawesi) by GPS. Geophysical Research Letters,
2844 Development Centre, Bandung. 25, 2313–2316.
2845 Tjokrosapoetro, S., Achdan, A., Suwitodirdjo, K., Wang, Y., Sieh, K., Soe Thura Tun, Lai, K.-Y. & Than
2846 Rusmana, E. & Abidin, H.Z. 1993. Geological Map Myint, 2014. Active tectonics and earthquake poten-
of the Masohi Quadrangle, 2200 Maluku, 1:250000. tial of the Myanmar region. Journal of Geophysical
2847 Geological Research and Development Centre, Research: Solid Earth, 119, 3767–3822, https://doi.
2848 Bandung. org/10.1002/2013JB010762
2849 USGS EARTHQUAKE HAZARDS PROGRAM. Significant Watkinson, I.M. 2011. Ductile flow in the metamorphic
2850 Earthquakes, http://earthquake.usgs.gov/earthquakes/ rocks of central Sulawesi. In: Hall, R., Cottam,
2851 browse/significant.php [last accessed 29 October 2016]. M.A. & Wilson, M.E.J. (eds) The SE Asian Gateway:
2852 van Leeuwen, T.M. & Muhardjo, 2005. Stratigraphy History and Tectonics of the Australia– Asia Collision.
2853 and tectonic setting of the Cretaceous and Paleogene Geological Society, London, Special Publications,
2854 volcanic-sedimentary successions in northwest Sula- 355, 157 –176, https://doi.org/10.1144/SP355.8
2855 wesi, Indonesia implications for the Cenozoic evolu- Watkinson, I.M., Hall, R. & Ferdian, F. 2011. Tectonic
tion of Western and Northern Sulawesi. Journal of re-interpretation of the Banggai-Sula –Molucca Sea
2856 Asian Earth Sciences, 25, 481–511. margin, Indonesia. In: Hall, R., Cottam, M.A. &
2857 van Leeuwen, T., Allen, C.M., Kadarusman, A., Wilson, M.E.J. (eds) The SE Asian Gateway: History
2858 Elburg, M., Palin, J.M., Muhardjo & Suwijanto and Tectonics of the Australia–Asia Collision. Geo-
2859 2007. Petrologic, isotopic, and radiometric age con- logical Society, London, Special Publications, 355,
2860 straints on the origin and tectonic history of the Malino 203–224, https://doi.org/10.1144/SP355.10
2861 Metamorphic Complex, NW Sulawesi, Indonesia. Wells, D.L. & Coppersmith, K.J. 1994. New empirical
2862 Journal of Asian Earth Sciences, 29, 751–777. relationships among magnitude, rupture length, rupture
2863 Vecchiotti, F. 2008. Calculating geomorphic indices width, rupture area, and surface displacement. Bulletin
2864 in SE Asia using a SRTM derived DEM: a worked of the Seismological Society of America, 84, 974–1002.
example from West Sulawesi, Indonesia. In: Michel, Wells, S.G., Bullard, T.F. et al. 1988. Regional varia-
2865 U., Civco, D.L., Ehlers, M. & Kaufman, H.J. (eds) tions in tectonic geomorphology along a segmented
2866 Remote Sensing for Environmental Monitoring. GIS convergent plate boundary, Pacific Coast of Costa
2867 Applications and Geology VIII, SPIE Proceedings, Rica. Geomorphology, 1, 239–265.
2868 7110, https://doi.org/10.1117/12.800240 Wesnousky, S.G. 1988. Seismological and structural evo-
2869 Visser, W.A. & Hermes, J.J. 1962. Geological Results of lution of strike-slip faults. Nature, 335, 340– 342.
2870 the Exploration for Oil in Netherlands New Guinea. Wesnousky, S.G. 2006. Predicting the endpoints of earth-
2871 Verhandelingen Koninklijk Nederlands Geologisch quake ruptures. Nature, 444, 358– 360.
2872 en Mijnbouwkundig Genootschap, Geologische Serie White, L.T., Hall, R. & Armstrong, R.A. 2014. The age
2873 20. Staatsdrukkerij-en Uitgeverijbedrijf, The Hague. of undeformed dacite intrusions within the Kolaka
Wald, D.J., Kanamori, H., Helmberger, D.V. & fault zone, SE Sulawesi, Indonesia. Journal of Asian
2874 Heaton, T.H. 1993. Source study of the 1906 Earth Sciences, 94, 105–112.
2875 San Francisco earthquake. Bulletin of the Seismologi- Wichmann, A. 1918. Die Erdbeben des Indischen Archi-
2876 cal Society of America, 83, 981– 1019. pels Bis Zum Jahre 1857. Verhandelingen der Konin-
2877 Walker, F. & Allen, M.B. 2012. Offset rivers, drainage klijke Akademie van Wetenschappen te Amsterdam,
2878 spacing and the record of strike-slip faulting: the Kuh 20, 193 [in Dutch].
2879 Banan Fault, Iran. Tectonophysics, 530–531, 251–263. Wu, J.E., McClay, K., Whitehouse, P. & Dooley, T.
2880 Wallace, R.E. 1968. Notes on stream channels offset by 2009. 4D analogue modelling of transtensional pull-
2881 the San Andreas fault, southern Coast Ranges, Cali- apart basins. Marine and Petroleum Geology, 26,
2882 fornia. In: Dickinson, W. & Grantz, A. (eds) Confer- 1608– 1623.
ence on Geologic Problems of San Andreas Fault Wyss, M. 2005. Human losses expected in Himalayan
2883 System. Proceedings. Stanford University Publications earthquakes. Natural Hazards, 34, 305–314.
2884 in the Geological Sciences, 11, 6 –21. Xu, X., Yeats, R.S. & Yu, G. 2010. Five short historical
2885 Wallace, R.E. 1986. Active Tectonics: Impacts on Soci- earthquake surface ruptures near the silk road, Gansu
2886 ety. Studies in Geophysics. National Academy Press, Province, China. Bulletin of the Seismological Society
2887 Washington, DC. of America, 100, 541– 561.
2888 Wallace, R.E. (ed.) 1990. The San Andreas Fault System, Yeats, R. 2010. Active Faults of the World. Cambridge
2889 California. US Geological Survey, Professional University Press, Cambridge.
2890 Papers, 1515. Yıldırım, C. 2014. Relative tectonic activity assessment
2891 Walpersdorf, A., Vigny, C., Subarya, C. & Manur- of the Tuz Gölü Fault Zone; Central Anatolia, Turkey.
ung, P. 1998. Monitoring of the Palu-Koro Fault Tectonophysics, 630, 183– 192.
2892
2893
2894
2895
2896
2897
2898
2899
2900