Geomorphology 104 (2009) 59–72
Contents lists available at ScienceDirect
Geomorphology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / g e o m o r p h
Tsunamis as geomorphic crises: Lessons from the December 26, 2004 tsunami in Lhok Nga, West Banda Aceh (Sumatra, Indonesia) Raphaël Paris a,⁎, Patrick Wassmer b, Junun Sartohadi c, Franck Lavigne b, Benjamin Barthomeuf a, Emilie Desgages a, Delphine Grancher b, Philippe Baumert b, Franck Vautier a, Daniel Brunstein b, Christopher Gomez b a b c
Géolab UMR 6042 CNRS, Clermont-Ferrand, France Laboratoire de Géographie Physique UMR 8591 CNRS, Meudon, France University Gadjah Mada, Yogyakarta, Java, Indonesia
A R T I C L E
I N F O
Article history: Accepted 26 March 2008 Available online 3 June 2008 Keywords: Tsunami Indonesia Sumatra Geomorphic crisis Coastal erosion Boulder deposition
A B S T R A C T Large tsunamis are major geomorphic crises, since they imply extensive erosion, sediment transport and deposition in a few minutes and over hundreds of kilometres of coast. Nevertheless, little is known about their geomorphologic imprints. The December 26, 2004 tsunami in Sumatra (Indonesia) was one of the largest and deadliest tsunamis in recorded human history. We present a description of the coastal erosion and boulder deposition induced by the 2004 tsunami in the Lhok Nga Bay, located to the West of Banda Aceh (northwest Sumatra). The geomorphological impact of the tsunami is evidenced by: beach erosion (some beaches have almost disappeared); destruction of sand barriers protecting the lagoons or at river mouths; numerous erosion escarpments typically in the order of 0.5–1.5 m when capped by soil and more than 2 m in dunes; bank erosion in the river beds (the retreat along the main river is in the order of 5–15 m, with local retreats exceeding 30 m); large scars typically 20–50 cm deep on slopes; dislodgement of blocks along fractures and structural ramps on cliffs. The upper limit of erosion appears as a continuous trimline at 20– 30 m a.s.l., locally reaching 50 m. The erosional imprints of the tsunami extend to 500 m from the shoreline and exceed 2 km along riverbeds. The overall coastal retreat from Lampuuk to Leupung was 60 m (550,000 m2) and locally exceeded 150 m. Over 276,000 m3 of coastal sediments were eroded by the tsunami along the 9.2 km of sandy coast. The mean erosion rate of the beaches was ~ 30 m3/m of coast and locally exceeded 80 m3/m. The most eroded coasts were tangent to the tsunami wave train, which was coming from the southwest. The fringing reefs were not efficient in reducing the erosional impact of the tsunami. The 220 boulders measured range from 0.3 to 7.2 m large (typically 0.7–1.5 m), with weights from over 50 kg up to 85 t. We found one boulder, less than 1 m large, at 1 km from the coastline, but all the others were transported less than 450 m (b 7 m a.s.l.). No fining landward boulder size distribution could be detected. The coincidence of different size modes, from boulders to fine sands, with independent spatial distributions, suggests that all the material was not transported in suspension, but rather by a combination of suspension and bed load transport. Finally, the spatial and size distributions of tsunami boulder deposits mostly depend on the location and characteristics of their source (coral reef, beach rock, platform, dams), together with clast and surface interference during transport. One year after, the coastal environment in northwest Sumatra is still in a post-tsunami dynamic. Thus, the difference between the largest tsunamis (height N 30 m) and the moderate tsunamis (height b 10 m) could be their long-term impact on coastal environments. © 2008 Elsevier B.V. All rights reserved.
1. Introduction The role of catastrophic events, such as tsunamis, hurricanes and storms, in the morphologic and sedimentologic evolution of coastal environments appears as a rising topic in the literature (Dawson, 1994; Bryant et al., 1996; Felton and Crook, 2003; Scheffers and Kelletat, 2003; Nott, 2004). Large tsunamis are major geomorphic ⁎ Corresponding author. E-mail address:
[email protected] (R. Paris). 0169-555X/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.geomorph.2008.05.040
crises, since they imply extensive erosion, sediment transport and deposition in a few minutes. Their imprint can stretch over hundreds of kilometres of shoreline, from shallow waters to several kilometres inland. The tsunami deposits are the geologic records of past tsunamis. Most publications about tsunamis focus on their depositional traces, such as on the U.S. Pacific coast (Atwater, 1987; Shennan et al., 1996; Williams et al., 2005), Hawaii (Moore and Moore, 1984; Moore et al., 1994; Goff et al., 2006); South America (Cantalamessa and Di Celma, 2005; Cisternas et al., 2005), New Zealand (Goff et al., 2001, 2004; Regnauld et al., 2004), Japan (Minoura and Nakaya, 1991; Shiki and
60
R. Paris et al. / Geomorphology 104 (2009) 59–72
Yamazaki, 1996; Fujiwara et al., 2000; Nanayama et al., 2000; Takashimizu and Masuda, 2000), Caribbean (Kelletat et al., 2004; Scheffers, 2004), North Atlantic coasts (Bondevik et al., 1997; Dawson and Smith, 2000; Bondevik et al., 2005), Canary Islands (Pérez Torrado et al., 2006), Mediterranean and Portugal (Andrade, 1992; Dawson et al., 1995; Hindson et al., 1996; Minoura et al., 2000; Luque et al., 2001; Tanner and Calvari, 2004). Their recognition and analysis provide clues to better understand the sedimentary signature of tsunamis, reconstruct their spatial distribution and determine earthquake recurrence intervals in highly seismic areas. Analysis of recent tsunami deposits and events that produced them were carried out by Shi et al. (1995), Minoura et al. (1997) in Florès (December 1992 tsunami), Bourgeois and Reinhart (1993) in Nicaragua (September 1992 tsunami), Nishimura and Miyaji (1995), Sato et al. (1995), in Hokkaido (July 1993), Dawson et al. (1996) in Java (June 1994), McSaveney et al. (2000), Gelfenbaum and Jaffe (2003) in Papua New Guinea (July 1998), Jaffe et al. (2003) in Peru (June 2001) and Szczucínski et al. (2005) in Thailand (December 2004). Nevertheless, we still know little about the geomorphologic imprints of tsunamis, and only a few authors intended to correlate tsunami deposits with erosional features and processes (Andrade, 1992; Shi et al., 1995; Bondevik et al., 1997; Gelfenbaum and Jaffe, 2003). Systematic and detailed surveys immediately following a tsunami are required in order to link geomorphological and sedimentological traces of past tsunamis with specific processes and features observed during and just after a tsunami event. The December 26, 2004 tsunami in Sumatra (Indonesia) was one of the largest and deadliest tsunamis in recorded human history. In this paper, we present a description of the coastal erosion and boulder deposition induced by the 2004 tsunami in the Lhok Nga Bay, located on the northwestern coast of Sumatra, near the city of Banda Aceh (Fig. 1). The Aceh province was poorly studied prior to the tsunami, because of a thirty years long civil war. The Lhok Nga Bay is 300 km from the Sunda subduction trench between the Indo-Australian plate and the Burma microplate. The December 26, 2004 main earthquake (magnitude 9.3) occurred 100 km east of the Sunda trench and 250 km south–southeast of the study area. The northwestern coast of Aceh has a contrasted relief of bays and creeks dominated by steep slopes typical of tropical environments. The cliffs display Palaeozoic limestones and dolomites, fractured and sometimes folded. Karst morphologies can be observed all along the coast. The embayments display a landscape of beaches, small estuaries with sand barriers, mangroves and lagoons. The Northern part of the Lhok Nga Bay is a 5 km large coastal embayment (25 km2) open to the west and delimited by steep slopes
Fig. 1. Location map of the Lhok Nga Bay (Sumatra, Indonesia).
typical of tropical environments (Fig. 2). The coast prior to the tsunami event appeared as a continuous beach, breached at the mouth of the rivers during the wet season. The sea floor topography is almost flat, with a gentle slope break about 1 km from the shoreline. The lowest areas correspond to coastal lagoons. Hummocky terrains correspond to dunes and palaeo-dunes reaching 15 m a.s.l. (e.g. Lampuuk, Lam Lho). In the southern part of the bay, the coastal morphology becomes more contrasted, especially between the harbour and Labuhan Point, where the coast shows alternating cliffs and flat crescent-shaped bays and creeks. Fringing reefs are located between Lhok Nga Point and the main river mouth. 2. The December 26, 2004 tsunami in Lhok Nga The magnitude 9.3 earthquake was felt in Banda Aceh and Lhok Nga at 8:00 AM (7:58 local time, 00:58 UTC). The tsunami consisted in three main waves. A retreat of the sea was observed 10 min after the earthquake. The first wave came from the southwest three minutes later and was less than 5 m high, but very fast (Lavigne et al., 2006). The second and largest wave came from the west–southwest within 5 min after the first one and was at least 15–30 m high at the coast. Very few observations were available for the third wave. The return flow (backwash) was achieved 5 min after the third wave and its spatial distribution was controlled by the slope breaks and topography lows, such as rivers and lagoons (Lavigne et al., 2006; Paris et al., 2007). The eastern part of the bay was inundated for three days, because logjams of debris occurred along the river of Lhok Nga up to 4 km inland. Almost all the buildings and trees were swept away, all the roads and bridges were severely damaged or completely destroyed. Only the mosque and a few coconut and casuarina trees remained after the tsunami hit the area of Lampuuk. The harbour and the cement factory in the southern part of the bay suffered partial damage. The confirmed death toll in Lhok Nga was over 7000 people. The geomorphologic impact of the tsunami in Lhok Nga was evidenced by severe coastal erosion. The upper limit of destruction appears as a continuous trimline at 20–30 m a.s.l. The runup reached 36 m a.s.l. in Lhok Nga and 51 m a.s.l. in a small bay near Labuhan (Lavigne et al., 2006). It is the highest runup ever recorded in human history for a tsunami without coseismic landslides (NGDC, 2006). The erosional imprints of the tsunami extend to 500 m from the shoreline and exceed 2 km along the river beds. 3. Methodology The field investigations were carried out in Lhok Nga in August and December 2005 in the framework of the Tsunarisk project. A preliminary survey was undertaken in January 2005, three weeks after the tsunami (1st ITST — International Tsunami Survey Team). We also collected additional data and observations along the road from Lhok Nga to Calang (Labuhan, Leupung, Lhoong and Gleebruk). The characteristics of the tsunami in Lhok Nga (direction, velocity, height and number of waves) were documented by dozens of eyewitnesses. We also measured flow directions attested by tilted trunks and pillars, and debris wrapped around trees. The wave heights and runups on the slopes were estimated by measuring the elevation of broken limbs, uprooted bark, debris in trees or water level marks on buildings (data collected with handheld reflectorless laser measurement systems, see Lavigne et al., 2006). We did not estimate the wave heights from coconut trees, because their trunks were simply tilted from the top without breaking before returning back to their initial position after the tsunami. We surveyed and mapped 12.4 km of shoreline using a bifrequential DGPS (December 2005) and also by remote sensing (Ikonos images as of January 10, 2003 and December 28, 2004, Spot 5 image as of December 30, 2004 and an orthorectified aerial photo as of June 2005), in order to estimate the coastal retreat directly induced by the tsunami and coastline evolution during the following months.
R. Paris et al. / Geomorphology 104 (2009) 59–72
61
Fig. 2. Sketch map of the geomorphological impact of the December 26, 2004 tsunami in Lhok Nga (inset photo: northern part of the bay). As suggested by the inset graphics, there is not any relationship between the coastal retreat values (histograms) and the tsunami wave height values (curve).
Data were then corrected for tidal deviations. We calculated 125 values of instant coastal retreat from satellite imagery and laser measurements in the field, considering the direction of the tsunami waves (not always perpendicular to the coastline). We measured dozens of vertical erosion values and dug pits and trenches in the tsunami sand deposits with the aim of reconstruct the pre-tsunami topography. The vertical erosion was estimated by measuring the height between the base of the tsunami sand deposits covering eroded beaches, beach rocks, dunes or soils (a detailed study of the sand deposits has been carried out by Paris et al., 2007), and
preserved markers such as roads, exposed wells, foundations of buildings, non-eroded soils and roots. The main difficulties consisted in (1) determining the base of the tsunami sand deposits, which are enriched in bioclasts and rip-up clasts of soil, poorly sorted and coarser than the beach or dune sands, and (2) estimating the pretsunami topography in hummocky terrains (e.g. dunes near Lampuuk, northern part of the Lhok Nga Bay). Thus, we also used eyewitnesses accounts, photographs and videos of the area before the tsunami. We systematically measured the thickness of tsunami sand deposits and computed the data in a 100 m grid. Normally graded
62
R. Paris et al. / Geomorphology 104 (2009) 59–72
Fig. 3. Gleebruk (road from Banda Aceh to Calang): the coastal retreat is here so important (N 200 m) that the road is now directed toward the ocean.
couplets or triplets of sand layers were previously used by Paris et al. (2007) to identify the inflows of successive waves and the final outflow (backwash). In addition, we measured the 3 axis of 220 boulders, their imbrications if any, and their distance to the shoreline (considering the direction of the tsunami waves). We also noted their lithology, took samples and calculated their volume, density and weight. Finally, all the data were integrated in a GIS database for statistical and spatial analysis. 4. Field observations and results 4.1. Coastal retreat In the field, the coastal erosion caused by the tsunami mainly consisted of severe beach erosion (some beaches have almost disappeared, e.g. around Lhok Nga Point), attested by the exposure of the beach rock (e.g. Lampuuk and near the swimming pool of Lhok Nga), destruction of the sand barriers protecting the lagoons or at the mouths of the rivers (e.g. Lhok Nga main river), and numerous erosion escarpments. The most spectacular evidence are the sections of missing roads or bridges. In Gleebruk (on the road from Lhok Nga to Calang), the coastal retreat was so important (N 200 m) that the road is now directed toward the ocean (Fig. 3). The overall coastal retreat, from Lampuuk to Leupung was 60 m when considering the area of removal divided by the coastline length, and 56 m when considering the mean of the 125 values of retreat (Fig. 2). Over 550,000 m2 of sediments were eroded by the tsunami along the 9.2 km of sandy coast. In Lhok Nga, the largest retreats correspond to coastal lagoons and topographic insets invaded and enlarged by the tsunami (e.g. 197 m between Lhok Nga and Lampuuk, 111 m in south Lhok Nga). From the north to the south, we divided the coast in 10 morphological terrain units (Table 1). With less than 35 m of maximum retreat and less than 20 m of mean retreat, the small bays between Labuhan and
the harbour were less affected, perhaps due to the low approaching angle of the tsunami wave train. The coast of the golf course was also surprisingly preserved (mean coastal retreat: 10 m). The most eroded sectors were Lampuuk (mean retreat: 64 m), the Lhok Nga Point (86 m) and the Leupung Bay (123 m). These three coastlines were tangent to the tsunami wave train coming from the Southwest, and are located at the northern extremities of the two large bays. The small fringing reef of Lhok Nga was not efficient in reducing the erosional impact of the tsunami. In north Leupung, the sand barrier — coastal lagoon systems were considerably vulnerable and retreat values are typically in the order of 100–150 m (Fig. 4). An issue in question is the relationship between the coastal erosion induced by the tsunami and the effects of the earthquake and subsequent aftershocks. The numerous submerged roots and trunks in Leupung and Lhok Nga, especially in the southern part of the bay, suggest a land subsidence in the order of 1–2 m (Fig. 5). Comparatively, the land has subsided 30 to 60 cm in the city of Banda Aceh (Yuichiro Tanioka and Yudhicara, pers. comm.). Thus, our estimations
Table 1 Coastal retreat induced by the December 26, 2004 tsunami in Lhok Nga Coast length
Lampuuk Lhok Nga Point Golf course Main river South Lhok Nga Harbour Labuhan bay 1 Labuhan bay 2 Labuhan bay 3 Leupung Total
Mean retreat
Area affected 2
m
m
m
900 1600 600 900 1850 110 900 200 480 1680 9220
64 86 10 48 38 46 16 20 17 123 60
58,000 137,400 5300 42,900 70,000 5100 14,250 3980 8360 206,400 551,690
Volume eroded m3/m 25–40 30–80 b 40 20–35 20–80
20–50 30
R. Paris et al. / Geomorphology 104 (2009) 59–72
63
Fig. 4. Satellite images showing the coast in the north part of the Leupung Bay, before (Ikonos image of January 10, 2003) and after the tsunami (CNES-Spot 5 image of December 30, 2004). The coastal retreat exceeds 150 m and is attested by the complete destruction of the sand barrier, thus leaving the lagoons opened to the sea and modifying the coastal drainage.
of coastal retreat are maximum values and more investigations should be carried out in order to detail the spatial distribution and rates of land subsidence. 4.2. Vertical erosion The instant erosion induced by the tsunami was attested by multiple sets of fresh escarpments typically in the order of 0.5–1.5 m when the surface was capped by the soil and the roots (Fig. 6). The
mean vertical erosion was 1.3 m, although the surface micromorphology affected the tsunami flows and locally increased flow turbulence, both during runup and backwash. As a consequence, the instant vertical erosion was higher near walls, sea-walls and roads (e.g. 2.3 m between the dam and the swimming pool in south Lhok Nga, 2.2 m behind a wall in Lampuuk). The erosion escarpments occasionally described coalescent topographic insets and outsets, determined by the previous topography, the geometry of the infrastructures or deeprooted trees. In South Lhok Nga, the main erosional insets were
64
R. Paris et al. / Geomorphology 104 (2009) 59–72
Fig. 5. The submerged roots and trunks in south Lhok Nga suggest a land subsidence in the order of 1–2 m, due to the Mw 9.3 earthquake.
Fig. 6. The erosional impact of the tsunami is evidenced by beach erosion (A: vertical erosion of respectively 2 m near the swimming pool; B: 1.5 m at the contact with the road which is also partly destroyed), fresh escarpments of soil erosion (C: typically 0.5–1.5 m), and bank erosion along the river beds (D: Lampuuk).
R. Paris et al. / Geomorphology 104 (2009) 59–72
located behind the trenches cut by the tsunami in a coastal dam, following the direction of the wave train (Fig. 7). There was a notable difference between the vertical erosion in the outsets (1.1 m) and in the insets (1.4 m). The highest vertical erosion values (N2 m) were observed in the small dune field near Lampuuk (Fig. 8). The previous hummocky topography was heavily modified by the tsunami and some dunes are completely destroyed (N4 m of abrasion). A brown palaeo-soil was exposed by the destruction of the dunes, and even partly eroded by the tsunami. The density of the vegetation covering the dunes may have played a role in their degree of destruction. Secondary scarps on the flanks of the eroded dunes suggest that the backwash also produced significant erosion. We estimated eroded volumes per metre of coast when the values of coastal retreat were completed by vertical erosion rates along beach profiles. The mean eroded volume of the beaches is ~30 m3/m of coast
65
(Table 1), locally exceeding 80 m3/m of coast in South Lhok Nga (near the cement factory and the harbour) and at Lhok Nga Point (in the axis of the fringing reef). Assuming a mean coastal retreat of 60 m, a mean vertical erosion of 1 m at the top of the beach profiles (which is a conservative value) and an affected area of 551,690 m2, the volume of subaerial sediments transported by the tsunami may be around 276,000 m3 along 9.2 km of coast. 4.3. Erosion of river beds and banks Bank erosion along rivers, gullies and coastal lagoons was more difficult to quantify, since no markers were available. After eyewitness accounts and using the Ikonos images of January 10, 2003 and December 28, 2004, we can affirm that the bank retreat along the main river (Lhok Nga bridge) was typically in the order of 5–15 m, with local retreats exceeding 30 m. No difference was found between
Fig. 7. Aerial orthorectified photo (June 2005) of south Lhok Nga, showing a boulder field (N1000 boulders) and erosion escarpments behind a coastal dam. Note that the erosional insets and boulder concentrations are in the axis of the trenches cut by the tsunami in the dam.
66
R. Paris et al. / Geomorphology 104 (2009) 59–72
Fig. 8. Abrasion of dunes near Lampuuk. The vertical erosion is more than 4 m, as attested by this exposed well. The secondary scarp on the flank of the dune, below the maximum inundation height and pre-tsunami surface, probably reflects the erosion by the backwash.
the two banks of the river. The flow orientation data that we collected (Lavigne et al., 2006) shows that the backwash was mostly concentrated in the topographic lows, thus playing a secondary role in bank erosion. Along the west coast on the road to Calang, the sand barriers at the mouths of the rivers were systematically removed, leaving the lagoons and river beds opened to the sea. Bank erosion was also severe in the areas of Leupung and Lhoong (small bays located south of Leupung). Satellite imagery confirms that the path of some rivers shifted and palaeo-channels were reactivated (e.g. west Banda Aceh). This is the result of the intense erosion of the topographic lows in a few minutes, followed by inundations because of the voluminous logjams of debris. The geometry of the hydrographic network and the spatial distribution of the debris during the backwash may determine the posttsunami distribution of the drainage.
Nga and Gleebruk. The vegetation and the soil were almost washed over behind the tsunami's trimline on the cliffs and slopes. The erosional features consisted in erosion escarpments at the foot of the slopes, large scars typically 20–50 cm deep, and dislodgment of blocks along the fractures and structural ramps of the cliffs (the Palaeozoic limestone is particularly fractured and karstified on this coast). When the rock was previously weathered and fractured, the tsunami was able to mobilise entire slope sections. On the road from Lhoong to Gleebruk, the main road was cut and a new one has been built in a trench made by the tsunami through the slope. Limestone megaclasts resulting from this erosion were deposited on the rice fields behind. Nevertheless, this kind of spectacular erosional feature remains rare, despite the considerable intensity of the tsunami.
4.4. Cliff and slope erosion
Although the deposition by the tsunami is mainly represented by extensive sheets of sand up to 5 km inland (Paris et al., 2007), megaclasts of soil, road, cement and boulders of coral (mainly Porites), beach rock and limestone were also deposited onshore. The 220 boulders measured range from 0.3 to 7.2 m on the main axis. Most of them are 0.7–1.5 m large (0.1–0.9 m3). The beach rock clasts are slightly smaller (mean volume: 0.33 m3) than the coral clasts (mean
Rocky coasts and slopes undoubtedly underwent erosion, but its estimation is almost impossible (Fig. 9). The pre-tsunami surface is not easily identifiable in the field and the resolution of the satellite imagery is not high enough to estimate retreats. Nevertheless, we could notice erosion evidence on all the rocky coasts between Lhok
4.5. Boulder deposits
R. Paris et al. / Geomorphology 104 (2009) 59–72
67
Fig. 9. The erosion of the cliffs and slopes is attested by dislodgment of blocks along the fractures and structural ramps of the cliffs (A), soil erosion and escarpments at the foot of the slopes (B), plurimetric fresh scars typically 20–50 cm deep (C), and rarely by trenches enlarged by the tsunami (D).
volume: 0.84 m3). Considering densities of 1.8 t/m3 for the beach rock, 2.4 t/m3 for the limestone and 1.1–1.4 t/m3 for the coral clasts, we estimated weights from over 50 kg up to 85 t. We found one boulder, less than 1 m large, at 1 km from the coastline (7 m a.s.l.), but all the others occurred at less than 450 m (b7 m a.s.l.). We could not detect any clear relationship between the coral and limestone boulder sizeweight and their distance from the shoreline (Fig. 10). The beach rock clasts display a landward fining. The trenches we dug around the boulders indicate that they are partly set in sandy tsunami deposits, without any ejecta or impact features (Fig. 11). The beach rock clasts were transported almost the same distance as denser limestone clasts (b350 m), because their transportation needs prior erosion of the beach by the tsunami. The spatial distribution of the boulders helps to distinguish four main sectors of deposition. We found 15 isolated coral boulders on the Lhok Nga Point. Their weight ranges from 160 kg to 11.4 t, five of them exceeding 3 t. We also found a small isolated beach rock boulder (90 kg) in the southeastern part of the point. The structure and shape of the Porites boulders show that they were in overturned position. Their long axis was tangent to the direction of the tsunami runup (NW to SSW). Thus, the boulders were not removed by the backwash. Their surface morphology displayed occasional evidence of impact during transport and the coral branches are completely destroyed. The most probable source for these boulders is the fringing reef about 250–500 m west from the shoreline. Thus, the transport figure of the biggest boulders ranges from 8000 to 45,000 (transport figure = weight × height × distance to coastline, see Scheffers and Kelletat, 2003; Kelletat et al., 2004). The spatial distribution of the coral boulders was concentrated behind the
reef by tsunami wave diffraction, as shown by the oriented markers on the field (Lavigne et al., 2006). Another field of ~100 boulders (~ 1 km2) is located between the Lhok Nga Point and the main river mouth, including the golf course. It is composed of 65 coral boulders and 14 beach rock clasts, less than 2 m large and not exceeding respectively 1.1 t and 2.5 t. Some of them were broken into several parts, but the clusters were not dispersed. The direction of the tsunami waves suggests that the coral boulders came directly from the southern prolongation of the fringing reef, thus covering a distance of 300–700 m. Hence, the maximum transport figure of the biggest boulder is 4600. The source of the beach rock clasts is more difficult to constrain, since erosion and exposure of the beach rock can be observed all along the coast. A minimum 500 m distance and a transport figure around 5000 can be expected for the biggest clast (2.5 t). The third sector is a 1.5 km2 field of limestone boulders, resulting from the destruction of a sea-wall by the tsunami (Fig. 4). We mapped and measured 114 boulders, ranging from 0.5 m to 3.3 m (3.2 m3), but the total amount of boulders may be over 1000. They are all located along a 250 m strip between the sea-wall and the main road. The lateral extension of the boulder field corresponds to the length of the sea-wall. The dispersion of boulders by the tsunami was limited. Although their source is clearly identified, there is not a landward fining of the boulders. Numerous small clasts around the boulders, together with impact features on their surface, suggest interference during their transport. Considering the weight (b7.7 t) and elevation (b8 m) of the limestone boulders, the transport figures are lower than 10,000.
68
R. Paris et al. / Geomorphology 104 (2009) 59–72
superposition of lithological facies). Orientations of these features show that some of the megaclasts are overturned. The submerged scars can be observed when the sea is calm. Thus, the distance covered by these megaclasts is only a few metres (b100 m) and the transport figure must be lower than 13,000 (85 t × 1.5 m × 100 m). 5. Discussion 5.1. Boulder deposits and the reconstruction of palaeo-tsunamis
Fig. 10. Bivariate plots of the characteristics of tsunami boulder deposits in Lhok Nga, compared to other coastal boulders in the world (data after Bourrouilh-Le Jan and Talandier, 1985; Harmelin-Vivien and Laboute, 1986; Hearty, 1997; Nott, 1997; Mastronuzzi and Sansò, 2000; Bryant and Nott, 2001; Kelletat and Schellmann, 2002; Kelletat et al., 2004; Nott, 2004; Scheffers, 2004; Whelan and Kelletat, 2005; Goff et al., 2006).
The fourth sector is a small bay south of the harbour, on the road to Labuhan and Leupung. There, we measured 5 coral boulders close to the shoreline (b115 m), the largest one being 2.8 m with a weight of 6.5 t (transport figure = 3000). The most impressive feature in this sector is a rocky platform eroded by the tsunami and associated with imbricate clusters and solitary megaclasts (Fig. 12). The megaclasts are tabular, 4.9–7.2 m long and 2.1–3.5 m thick. Their surface morphology and lithological structure can be matched with similar features of the platform itself (flat or rough surface, broken branches of young coral,
Large boulders in coastal areas have been reported from various places in the world, such as in French Polynesia (Bourrouilh-Le Jan and Talandier, 1985; Harmelin-Vivien and Laboute, 1986), Australia (Nott, 1997, 2004; Bryant and Nott, 2001), Hawaii (Goff et al., 2006), Bahamas (Hearty, 1997; Kelletat et al., 2004), Netherlands Antilles (Scheffers, 2004), Italy (Mastronuzzi and Sansò, 2000, 2004), Cyprus (Kelletat and Schellmann, 2002) and Spain (Whelan and Kelletat, 2005). They range from over 10 up to 1000 m3 and, depending on the rock density, their mass can exceed 2000 t (Scheffers and Kelletat, 2003). They have been found at various elevations from the intertidal zone to a few tens of meters above the present sea level, and their emplacement is usually attributed to tsunamis, hurricanes or powerful storms. Shi et al. (1995) reported that hundreds of boulders were deposited as far as 200 m inland by the December 12, 1992 tsunami in Flores (Indonesia), especially in the area of Riangkroko where the runup reached 26 m. Eyewitness accounts confirm that the boulders were deposited by the second and highest wave. The first wave, although lower than 5 m, may have prepared the detachment of the boulders by the second one. Thus, our study confirms that a large tsunami with wave heights in the order of 25–30 m, is able to detach and transport coral boulders with weights more than 10 t over 500–700 m landward, and megaclasts of the platform with weights in excess of 85 t over a few metres. The transport figures for the December 26, 2004 tsunami in Lhok Nga range from 8000 to 45,000 for the boulders of the coral reef, less than 13,000 for the megaclasts of the platform and less than 10,000 for the limestone boulders of the dam. These values are in the range of the transport figures of large past-events on the coasts of Italy (1456 Ionian tsunami: Mastronuzzi and Sansò, 2000), Spain (1755 Lisbon tsunami: Whelan and Kelletat, 2005) and Hawaii (1946 Aleutian tsunami: Noormets et al., 2002). Transport figures exceeding 70,000 and 100,000 were calculated for coastal boulders in the Netherlands Antilles and in Australia (Scheffers and Kelletat, 2003). Nevertheless, these estimations are not only based upon the weight of the boulder, but also on the elevation and distance from the coastline, which could have been different at the time of deposition. Furthermore, Nott (2003) points out that the pre-transport environment of the boulders along with their shape, size and density determines the energy required for them to be transported. Nott (2003) distinguishes the subaerial from the submerged boulders and the megaclasts derived from joint bounded blocks on shore platforms. Indeed, the transportation of boulders requires large drag-and-lift forces particularly if the megaclasts are broken off from rocky platforms, terraces or reefs. The sea withdrawal prior to the tsunami may also play a significant role on the preparation, as it occurred in Lhok Nga (~1 km after eyewitnesses). The entrapment of air in the joints and fractures by the first onrushing wave could have provided sufficient hydraulic lift to detach some boulders. Following the equations of Nott (1997, 2003) for subaerial boulders, a 1.5 m high wave is required to move the limestone boulders of the dam in south Lhok Nga (b7.7 t). For submerged boulders like the coral boulders in Lhok Nga Point (max. 11 t), a 2.5 m high wave is required. The mushroom-like shape of the Porites, with an increasing weight from the base to the top, makes them particularly vulnerable. The tsunami wave height required to detach the megaclasts from the platform south of the harbour (85 t) is up to 15 m. These values are lower than
R. Paris et al. / Geomorphology 104 (2009) 59–72
69
Fig. 11. Coral boulder deposits in Lhok Nga. The boulder on the photo above is 2.5 m large and weighs more than 4 t. The coral boulders are partly set in the sandy tsunami deposits, without any ejecta or impact features. The coincidence of two size modes deposited at the same time suggests both suspension and bed load transport mechanisms.
the tsunami heights collected on the field (15–30 m at the coast). Greater transport figures could be expected for the December 26, 2004 tsunami, but it is not the case. Furthermore, our bivariate analysis of the weight or main axis of the boulders compared to their distance to the sea do not show any landward fining, except for the beach rock located at the mouth of the main river. The spatial distribution and position of the boulders provide useful information about the direction and transport mechanism of palaeotsunamis. Our observations on the boulder field in Lhok Nga Point confirm that the elongated boulders tend to dispose their imbrication or long axis tangent to the direction of the tsunami wave train, as suggested by many authors (Mastronuzzi and Sansò, 2000; Scheffers, 2004; Whelan and Kelletat, 2005). The long axis and imbrication axis distribution can thus help to reconstitute the direction of a tsunami wave train and to distinguish different flows (successive waves, backwash), even though powerful storms are able to modify the position of large boulders (Noormets et al., 2002; Felton and Crook, 2003). The surface morphology and structure of the largest boulders in Lhok Nga show that they are sometimes in overturned positions. This is not surprising for the Porites boulders, which could not rely on their narrow base. Overturning of large boulders suggests a mechanism for detachment and inversion that does not involve suspension in a high-velocity flow or in a breaking wave (Felton and Crook, 2003).
The fact that the sandy and boulder deposits appear to be contemporaneous is not in favour of a suspension mechanism for boulder transportation. The coincidence of different size modes with independent spatial distributions is a common feature of tsunami deposits (Scheffers, 2004; Goff et al., 2006) and suggests that all the material is not transported in suspension, but rather that there is a combination of suspended and bed load. Finally, the spatial and size distributions of tsunami boulder deposits mostly depend on the location and characteristics of their source, together with clast and surface interference during transport. 5.2. Instant and long-term geomorphic impact of a large tsunami Most of the erosional features induced by the December 26, 2004 tsunami in Lhok Nga were previously described for recent tsunamis, such as erosion escarpments in soils, beaches, banks and riverbeds, and slopes during the 1994 tsunami in Southeast Java (Maramai and Tinti, 1997), 1992 tsunami in Flores (Shi et al., 1995) and 1998 tsunami in Papua New Guinea (Gelfenbaum and Jaffe, 2003). Sand barriers were also removed by the 1755 tsunami in Portugal (Andrade, 1992) and dunes destroyed by Holocene tsunamis in New Zealand (Regnauld et al., 2004). The vertical erosion rates estimated by Maramai and Tinti (1997), Shi et al. (1995) and Gelfenbaum and Jaffe (2003) for tsunamis
70
R. Paris et al. / Geomorphology 104 (2009) 59–72
Fig. 12. Imbricate clusters and megaclasts coming from the coastal platform (small bay south of the Lhok Nga harbour, on the road to Calang). The largest megaclasts are 6–7 m large and weigh 85 t. Note also the trimline left by the tsunami on the cliff at approximately 20 m a.s.l. and the coral rubble deposits (left corner).
with wave heights lower than 5–10 m are in the same order of those we measured in Lhok Nga, except for the dunes in Lampuuk (N 4 m). Shi et al. (1995) calculated coastal retreats from 10 to 50 m and a local erosion rate of 43 m3/m. The erosion rates induced by the 2004 tsunami in Lhok Nga range between 20 and 80 m3/m, with a mean erosion rate of 30 m3/m. We could not observe sculptured bedrock forms in the rocky platforms and cliffs, as those described by Bryant et al. (1996) and Bryant and Nott (2001) in southeastern Australia. With wave heights more than 15 m and runups more than 80 m a.s.l., the tsunami simulated by Bryant and Nott (2001) is comparable with the 2004 tsunami in Lhok Nga, but the lithologic and morphologic setting is completely different. The genesis of these coastal bedrock erosional features is far from being resolved (Felton and Crook, 2003; Nott, 2004). The succession of outsets and insets, especially along the dam of South Lhok Nga, and the enlarged lagoons remind the chevron ridges described in the Bahamas by Hearty et al. (1998) and in Australia by Bryant et al. (1996), but they are 15 to 30 times smaller in size. Although the 2004 tsunami was one of the largest tsunami of human recorded history, its geomorphic impact is more impressive in spatial extent than in intensity. After our estimations, 16.5 millions of m3 of coastal sediments (550,000 m2) were moved over 9.2 km of coast. The tsunami also affected submarine sediments. The spatial distribution of the tsunami sand deposits inland mostly depends on the topography, despite a landward thinning tendency (Fig. 13). In the northern part of the Lhok Nga Bay, a break in the sedimentation is associated with a gentle slope break at 1–1.5 km from the coast. The thickest accumulations correspond to sediment traps (e.g. gullies, swamps, dams). The thickness of tsunami deposits, when computed in a 100 m grid, allows us to estimate the volume of sediments deposited inland by the tsunami. This volume ranges from 500,000 to 1,500,000 m3 between Lampuuk and the Lhok Nga River, where 120,000 m3 of coastal sediments were moved by the tsunami. Thus, we can conclude than more than 75% of the tsunami deposits came from offshore. Numerous eyewitness accounts collected in Lhok Nga confirmed that the tsunami waves were particularly dark because of their high sediment content (Lavigne et al., 2006). Gelfenbaum and
Jaffe (2003) also estimated that as much as 2/3 of the deposits of the 1998 tsunami in PNG came from offshore. Sato et al. (1995) found that deposits of the 1983 and 1993 tsunami in Japan came mainly from the beaches and berms. More than 300 km of coast were severely affected by the tsunami in Sumatra. The difference between the largest tsunamis (N30 m) and moderate tsunamis (height b 10 m) could be their spatial extent and their long-term impact on coastal environments. Bryant et al. (1996) suggest that large tsunamis have been the dominant factor for the Holocene coastal evolution in southeastern Australia. We have started to follow the post-tsunami coastal evolution in Lhok Nga and Banda Aceh, from January 2005 to December 2005. Our preliminary investigations in Lhok Nga show a beach growth at fast rates, especially in April and May 2005 (end of the wet season), on the south side of the Lhok Nga Point and in south Lhok Nga. All the lagoons and river mouths were enclosed by sand barriers during the 2005 dry season. In contrast, the sectors of Lampuuk and the golf course displayed more or less the same coastline in January and June 2005. On the north coast of Banda Aceh, the destruction of the polders, lagoons and aquacultures resulted in an incredible coastal retreat (locally 900–1600 m). The aerial photo of June 2005 and our surveys in August and December 2005 confirm that the northern districts of Banda Aceh are systematically affected by inundations during the highest tides. Further investigations will focus on the long-term impact of the tsunami on coastal sediment transfers and recommendations for management in tsunami-resilient coasts. 6. Conclusions The erosional imprints of the December 26, 2004 tsunami in Lhok Nga extend to 500 m from the shoreline and exceed 2 km along the river beds. The mean erosion rate of the beaches is ~30 m3/m of coast and locally exceeds 80 m3/m, but more than 75% of the sediments deposited inland came from offshore. The correlation between erosion rates and tsunami heights remains unclear. The most eroded coasts were tangent to the tsunami wave train, which came from the southwest. Furthermore, the morphology of the coast and the local
R. Paris et al. / Geomorphology 104 (2009) 59–72
71
Fig. 13. Thickness of tsunami sand deposits in the northern part of the Lhok Nga Bay, inferred from field observations computed in a 100 m grid (UTM coordinates). The spatial distribution of the deposits mostly depends on the topography, despite a landward thinning tendency. Note the breakage in sedimentation associated with a gentle slope break at 1– 1.5 km from the coast (thickness b 5 cm). The thickest accumulations correspond to sediment traps (e.g. gullies, swamps, sea-walls).
topography (dams, cliffs, buildings) greatly influence the erosional impact of a tsunami. The fringing reefs were not efficient in reducing the erosional impact of the tsunami. The large tsunami waves (25–30 m high) were able to detach and transport coral boulders in excess of 10 t over 500–700 m and megaclasts of the platform in excess of 85 t over a few metres. No landward fining trend in the boulder size distribution could be detected. The spatial and size distributions of tsunami boulder deposits mostly depend on the location and characteristics of their source (coral reef, beach rock, platform, dams), together with clast and surface interference during transport. More systematic investigations on historic and future tsunamis are needed to confirm that the spatial distribution and position of the boulders provide useful information about the direction and transport mechanism of a tsunami. At present, the boulder fields do not appear as powerful indicators of the magnitude and hydrodynamic of a palaeo-tsunami. Nevertheless, the coincidence of different size modes, from boulders to fine sands suggests that all the material was not transported in suspension, but rather that there was a combination of bed load and suspended transport modes.
One year after the disaster, the coastal environment in northwest Sumatra was still in a post-tsunami dynamic. The large tsunamis (N30 m) may have more long-term impact on coastal environments than moderate tsunamis (height b 10 m). Thus, the recognition and study of past geomorphic crisis could help in assess the intensity of palaeo-tsunamis and present-day associated hazards. Acknowledgements Funding came from the Délégation Interministérielle pour l'aide Post-Tsunami (DIPT, project no. 161), the French Embassy in Indonesia and the Centre National de la Recherche Scientifique (CNRS) in France. We would like to thank the Tsunarisk team, François Flohic, Benjamin De Coster and Damien Le Floch (Planet Risk), Taufik Gunawan, Fachrizal, Iman and Syahnan (Indonesian Meteorological and Geophysical Agency), Adi Widagdo, Rino Cahyadi, Anggri Setiawan, Djati Mardiatno, Mujiono and Syamsul (University Gadjah Mada) for their collaboration on the field in August and December 2005. We are also grateful to the CNES which provided Spot 5 images at a reasonable rate. Mehdi Adjeroud, Michel Le Pichon and Bernard Salvat provided
72
R. Paris et al. / Geomorphology 104 (2009) 59–72
useful comments about the coral boulders. Finally, we would like to thank F. Javier Gracia and Giuseppe Mastronuzzi for their encouraging and meticulous reviews. References Andrade, C., 1992. Tsunami generated forms in the Algarve barrier island (South Portugal). In: Dawson, A.G. (Ed.), European Geophysical Union 1992 Tsunami Meeting. Science of Tsunami Hazard, vol. 10, pp. 21–34. Atwater, B.F., 1987. Evidence for great Holocene earthquakes along outer coast of Washington State. Science 236, 942–944. Bondevik, S., Svendsen, J.I., Mangerud, J., 1997. Tsunami sedimentary facies deposited by the Storegga tsunami in shallow marine basins and coastal lakes, western Norway. Sedimentology 44, 1115–1131. Bondevik, S., Mangerud, J., Dawson, S., Dawson, A., Lohne, O., 2005. Evidence for three North Sea tsunamis at the Shetland Islands between 8000 and 1500 years ago. Quaternary Science Reviews 24, 1757–1775. Bourgeois, J., Reinhart, M.A., 1993. Tsunami deposits from 1992 Nicaragua event: implications of paleo-tsunami deposits, Cascadia Subduction Zone. EOS, Fall Meeting of the American Geophysical Union, Washington DC, p. 350. Bourrouilh-Le Jan, F.G., Talandier, J., 1985. Major high-energy events in a reef environment: tsunamis, hurricanes and tropical cyclones and their effects on the sedimentology and geomorphology of an atoll: Rangiroa, Tuamotu (SE Pacific). Marine Geology 37, 263–272. Bryant, E.A., Nott, J., 2001. Geological indicators of large tsunami in Australia. Natural Hazards 24 (3), 231–249. Bryant, E.A., Young, R.W., Price, D.M., 1996. Tsunami as a major control on coastal evolution, Southeastern Australia. Journal of Coastal Research 12, 831–840. Cantalamessa, G., Di Celma, C., 2005. Sedimentary features of tsunami backwash deposits in a shallow marine Miocene setting, Mejillones Peninsula, northern Chile. Sedimentary Geology 178, 259–273. Cisternas, M., Atwater, B.F., Torrejon, F., Sawai, Y., Machuca, G., Lagos, M., Eipert, A., Youlton, C., Salgado, I., Kamataki, T., Shishikura, M., Rajendran, P.R., Malik, J.K., Rizal, Y., Husni, M., 2005. Predecessors of the giant 1960 Chile earthquake. Nature 437, 404–407. Dawson, A.G., 1994. Geomorphological effects of tsunami run-up and backwash. Geomorphology 10, 83–94. Dawson, S., Smith, D.E., 2000. The sedimentology of Middle Holocene tsunami facies in northern Sutherland, Scotland, UK. Marine Geology 170, 69–79. Dawson, A.G., Hindson, R., Andrade, C., Freitas, C., Parish, R., Baternan, M., 1995. Tsunami sedimentation associated with the Lisbon earthquake of 1 November 1755: Boca do Rio, Algarve, Portugal. The Holocene 5 (2), 209–215. Dawson, A.G., Shi, S., Dawson, S., Takahashi, T., Shuto, N., 1996. Coastal sedimentation associated with the June 2nd and 3rd, 1994 tsunami in Rajegwesi, Java. Quaternary Science Reviews 15, 901–912. Felton, E.A., Crook, K.A.W., 2003. Evaluating the impact of huge waves on rocky shorelines: an essay review of the book ‘Tsunami — The Underrated Hazard’. Marine Geology 197, 1–12. Fujiwara, O., Masuda, F., Sakai, T., Irizuki, T., Fuse, K., 2000. Tsunami deposits in Holocene bay mud in southern Kanto region, Pacific coast of central Japan. Sedimentary Geology 135, 219–230. Gelfenbaum, G., Jaffe, B., 2003. Erosion and sedimentation from the 17 July, 1998 Papua New Guinea tsunami. Pure and Applied Geophysics 160, 1969–1999. Goff, J.R., Chagué-Goff, C., Nichol, S.L., 2001. Palaeotsunami deposits: a New Zealand perspective. Sedimentary Geology 143, 1–6. Goff, J.R., McFadgen, B.G., Chagué-Goff, C., 2004. Sedimentary differences between the 2002 Easter storm and the 15th-century Okoropunga tsunami, southeastern North Island, New Zealand. Marine Geology 204, 235–250. Goff, J.R., Dudley, W.C., deMaintenon, M.J., Cain, G., Coney, J.P., 2006. The largest local tsunami in 20th century Hawaii. Marine Geology 226, 65–79. Harmelin-Vivien, M.L., Laboute, P., 1986. Catastrophic impact of atoll outer reef slopes in the Tuamotu (French Polynesia). Coral Reefs 5, 55–62. Hearty, P., 1997. Boulder deposits from large waves during the last interglaciation on north Eleuthera Island, Bahamas. Quaternary Research 48, 326–338. Hearty, P., Neuman, A.C., Kaufman, D.S., 1998. Chevron ridges and runup deposits in the Bahamas from storms late in oxygen-isotope substage 5e. Quaternary Research 50, 309–322. Hindson, R.A., Andrade, C., Dawson, A.G., 1996. Sedimentary processes associated with the tsunami generated by the 1755 Lisbon earthquake on the Algarve coast, Portugal. Physics and Chemistry of the Earth 21 (12), 57–63. Jaffe, B., Gelfenbaum, G., Rubin, D., Peters, R., Anima, R., Swensson, M., Olcese, D., Bernales, L., Gomez, J., Riega, P., 2003. Identification and interpretation of tsunami deposits from the June 23, 2001 Peru tsunami. Proceedings of the International Conference on Coastal Sediments 2003, World Scientific Publishing Corp and East Meets West Productions, Corpus Christi, TX, USA. 13 pp. Kelletat, D., Schellmann, G., 2002. Tsunamis on Cyprus: field evidences and 14C-dating results. Zeitschrift für Geomorphologie NF 46, 19–34. Kelletat, D., Scheffers, A., Scheffers, S., 2004. Holocene tsunami deposits on the Bahaman Islands of Long Island and Eleuthera. Zeitschrift für Geomorphologie 48, 519–540. Lavigne, F., Paris, R., Wassmer, P., Gomez, C., Brunstein, D., Grancher, D., Vautier, F., Sartohadi, J., Setiawan, A., Syahnan, Gunawan, T., Fachrizal, Waluyo, B., Mardiatno, D., Widagdo, A., Cahyadi, R., Lespinasse, N., Mahieu, L., 2006. Learning from a major disaster (Banda Aceh, December 26th, 2004): a methodology to calibrate simulation codes for tsunami inundation models. Zeitschrift für Geomorphologie. Supplementband 146, 253–265.
Luque, L., Lario, J., Zazo, C., Goy, J.L., Dabrio, C.J., Silva, P.G., 2001. Tsunami deposits as paleoseismic indicators: examples from the Spanish coast. Acta Geologica Hispanica 3–4, 197–211. Maramai, A., Tinti, S., 1997. The 3 June 1994 Java tsunami: a post-event survey of the coastal effects. Natural Hazards 15 (1), 31–49. Mastronuzzi, G., Sansò, P., 2000. Boulders transport by catastrophic waves along the Ionian coast of Apulia (southern Italy). Marine Geology 170, 93–103. Mastronuzzi, G., Sansò, P., 2004. Large boulder accumulation by extreme waves along the Adriatic coast of southern Apulia (Italy). Quaternary International 120, 173–184. McSaveney, M., Goff, J., Darby, D., Goldsmith, P., Barnett, A., Elliott, S., Nongkas, M., 2000. The 17th July 1998 tsunami, Sissano Lagoon, Papua New Guinea — evidence and initial interpretation. Marine Geology 170, 81–92. Minoura, K., Nakaya, S., 1991. Traces of tsunami deposits preserved in intertidal lacustrine and marsh deposits: some examples from Northeast Japan. Journal of Geology 99, 265–287. Minoura, K., Imamura, F., Takahashi, T., Shuto, N., 1997. Sequence of sedimentation processes caused by the 1992 Flores tsunami: evidence from Babi Island. Geology 25, 523–526. Minoura, K., Imamura, F., Kuran, U., Nakamura, T., Papadopoulos, G.A., Takahashi, T., Yalciner, A.C., 2000. Discovery of Minoan tsunami deposits. Geology 28, 59–62. Moore, J.G., Moore, G.W., 1984. Deposit from a giant wave on the island of Lanai, Hawaii. Science 226, 1312–1315. Moore, J.G., Bryan, W.B., Ludwig, K.R., 1994. Chaotic deposition by a giant wave, Molokai, Hawaii. Geological Society of America Bulletin 106, 962–967. Nanayama, F., Shigeno, K., Satake, K., Shimokawa, K., Koitabashi, S., Miyasaka, S., Ishii, M., 2000. Sedimentary differences between the 1993 Hokkaido-nansei-oki tsunami and the 1959 Miyakojima typhoon at Taisei, southwestern Hokkaido, northern Japan. Sedimentary Geology 135, 255–264. NGDC (National Geophysical Data Center), 2006. Tsunami Event Database. National Oceanic and Atmospheric Association (NOAA). http://www.ngdc.noaa.gov/seg/ hazard/tsu_db.shtml. Nishimura, Y., Miyaji, N., 1995. Tsunami deposits from the 1993 southwest Hokkaido earthquake and the 1640 Hokkaido Komagatake eruption, Northern Japan. Pure and Applied Geophysics 144, 719–733. Noormets, R., Felton, E.A., Crook, K.A.W., 2002. Sedimentology of rocky shorelines: 2 — Shoreline megaclasts on the north shore of Oahu, Hawaii — origins and history. Sedimentary Geology 150, 31–45. Nott, J., 1997. Extremely high wave deposits inside the Great Barrier Reef, Australia: determining the cause—tsunami or tropical cyclone. Marine Geology 141, 193–207. Nott, J., 2003. Waves, coastal boulders and the importance of the pre-transport setting. Earth and Planetary Science Letters 210, 269–276. Nott, J., 2004. The tsunami hypothesis — comparisons of the field evidences against the effects, on the Western Australian coast, of some of the most powerful storms on Earth. Marine Geology 208, 1–12. Paris, R., Lavigne, F., Wassmer, P., Sartohadi, J., 2007. Coastal sedimentation associated with the December 26, 2004 in Lhok Nga, west Banda Aceh (Sumatra, Indonesia). Marine Geology 238, 93–106. Pérez Torrado, F.J., Paris, R., Cabrera, M.C., Schneider, J.L., Wassmer, P., Carracedo, J.C., Rodriguez Santana, A., Santana, F., 2006. The Agaete tsunami deposits (Gran Canaria): evidence of tsunamis related to flank collapses in the Canary Islands. Marine Geology 227, 137–149. Regnauld, H., Nichol, S.L., Goff, J.R., Fontugne, M., 2004. Maoris, middens and dune front accretion rate on the northeast coast of New Zealand: resilience of sedimentary system after a tsunami. Géomorphologie 1, 45–54. Sato, H., Shimamoto, T., Tsutsumi, A., Kawamoto, E., 1995. Onshore tsunami deposits caused by the 1993 Southwest Hokkaido and 1983 Japan Sea earthquakes. Pure and Applied Geophysics 144, 693–717. Scheffers, A., 2004. Tsunami imprints on the Leeward Netherlands Antilles (Aruba, Curaçao, Bonaire) and their relation to other coastal problems. Quaternary International 120, 163–172. Scheffers, A., Kelletat, D., 2003. Sedimentologic and geomorphologic tsunami imprints worldwide — a review. Earth-Science Reviews 63, 83–92. Shennan, I., Long, A.J., Rutherford, M.M., Green, F.M., Innes, J.B., Lloyd, J.M., Zong, Y., Walker, K.J., 1996. Tidal marsh stratigraphy, sea-level change and large earthquakes, I: a 500 years record in Washington, U.S.A. Quaternary Science Reviews 15, 1023–1059. Shi, S., Dawson, A.G., Smith, D.E., 1995. Coastal sedimentation associated with the December 12th, 1992 tsunami in Flores, Indonesia. Pure and Applied Geophysics 144, 525–536. Shiki, T., Yamazaki, T., 1996. Tsunami-induced conglomerates in Miocene upper bathyal deposits, Chita Peninsula, central Japan. Sedimentary Geology 104, 175–188. Szczucinski, W., Niedzielski, P., Rachlewicz, G., Sobczynski, T., Ziolla, A., Kowalski, A., Lorenc, S., Siepak, J., 2005. Contamination of tsunami deposits in a coastal zone inundated by the 26 December 2004 tsunami in Thailand. Environmental Geology 49, 321–331. Takashimizu, Y., Masuda, F., 2000. Depositional facies and sedimentary successions of earthquake-induced tsunami deposits in Upper Pleistocene incised valley-fills, central Japan. Sedimentary Geology 135, 231–239. Tanner, L.H., Calvari, S., 2004. Unusual sedimentary deposits on the SE side of Stromboli volcano, Italy: products of a tsunami caused by the ca. 5000 BP Sciarra del Fuoco collapse? Journal of Volcanology and Geothermal Research 137, 329–340. Whelan, F., Kelletat, D., 2005. Boulder deposits on the southern Spanish Atlantic coast: possible evidence for the 1755 AD Lisbon tsunami? Science of Tsunami Hazards 23 (3), 25–38. Williams, H.F.L., Hutchinson, I., Nelson, A.R., 2005. Multiple sources for late-Holocene tsunamis at Discovery Bay, Washington State, USA. The Holocene 15, 60–73.
