Several laterally extensive candidate tsunami deposits are preserved along coastlines facing the eastern Java Trench, indicating it has experienced mega-thrust earthquakes in the past. We investigated 37 coastal sites in Bali, Lombok, Sumba and Timor islands, many of which preserve course sand and pebble layers that overlie sharp basal contacts with scour marks into the mud, fine upward in grain size, and have bimodal grain size distributions. Other unique features are the common occurrence of marine fossils and concentrations of heavy minerals. The occurrence of these high-energy deposits interlayered with clay-rich units indicates the coarse clastics are anomalous because they were deposited in what is normally a very low-energy depositional environment. The lateral extent and paucity of thin, coarse clastic layers with marine organisms are inconsistent with local stream flood event, and the proximity to the equator of the sites diminishes the possibility of marine flood events from cyclones. The sparse, but consistent, the occurrence of at least two candidate tsunami deposits at depths of 1 and 2 meters over 950 km along the strike of the Java Trench may reveal that mega-thrust earthquakes have occurred there and generated giant tsunamis in the recent past.Five widely scattered imbricated boulder deposits are also found on Bali, Lombok, and Sumba. The boulders consist of slabs of hardpan up to 2.5 m in length and 80 cm thick that was torn from a near-shore seabed and stacked on top of one another. Some of the boulders were carried over the erosional coastal bank and deposited up to 100 meters inland. Comparisons with imbricated boulder ridges formed during the 1994 tsunami in east Java indicate that these deposits are from one or multiple tsunamis sourced by the Java Trench.Experiments in effective ways to communicate and implement tsunami disaster mitigation strategies have led us to train local communities about the 20-20-20 rule. If coastal communities experience more than 20 seconds of shaking from an earthquake, even if it is not intense, they should evacuate the coast. The time delay between the earthquake and arrival of tsunami waves is around 20 minutes, which is the time window for evacuation. Some tsunami waves may be as high as 20 meters, which is the target elevation for evacuation. Adopting the 20-20-20 rule could save thousands of lives throughout the region, especially in Bali where nearly 1 million people inhabit likely tsunami inundation zones.

Tsunamis are low amplitude, large wavelength waves that can significantly impact coastal regions. Although their destructive impacts are clear from recent events, the frequency with which tsunamis occur is less well constrained. To better understand the tsunami history and coastal impacts in Sri Lanka, this study compares sediments deposited by the December 26, 2004, tsunami to older lagoon sediments in search of evidence for paleotsunami deposits. Results from this study illustrate that the coastal lagoons in Sri Lanka preserve tsunami deposits and can provide the first steps towards constraining the paleotsunami history of the Indian Ocean. Because Sri Lanka is a far field location relative to the Sumatra-Andaman subduction zone, the preserved tsunami deposits are likely mega-tsunami events similar in size and destruction to the December 26, 2004, tsunami. The December 26, 2004, M 9.1?9.3 Sumatra-Andaman earthquake generated a massive tsunami that propagated throughout the Indian Ocean, causing extreme coastal inundation and destruction. The southeastern coastline of Sri Lanka was impacted by the 2004 tsunami where between one and three waves inundated coastal villages, lagoons, and lowlands, killing more than 35,000 people. Karagan Lagoon, located on the southeastern coast of Sri Lanka, was impacted by two waves from this tsunami. Although the lagoon commonly is dominated by organic-rich, siliciclastic clays, silts, and fine sands, the 2004 tsunami deposited a distinct layer of coarse quartz-dominated sand between 1 and 22 cm thick. The base of the 2004 deposit is sharp and erosional and some layers feature faint subparallel laminations. The 2004 tsunami deposit is generally continuous, fines landward, and is confined to the eastern portion of Karagan Lagoon, in the direction from which the tsunami arrived. Sri Lankan lore, in conjunction with reconstructed historical earthquake data, suggests that other tsunamis likely affected Sri Lanka in the past. To test this, twenty-two 1?4 m sediment cores were collected from Karagan Lagoon, providing key information for unraveling the pre-2004 tsunami history of southeastern Sri Lanka. At depth, sixteen cores from Karagan Lagoon contain as many as ten distinct sand layers, including the deposit from the 2004 tsunami. These cores feature siliciclastic clays, silts, and fine sands that dominate the background lagoonal sedimentation that are punctuated by coarse sand layers. These sand-rich layers feature sharp, erosional bases, coarsen and fine upwards, vary in thickness from 1 to 22 cm, and include varying percentages of fine to very coarse sand, with a low-abundance of silt and clay. In the best constrained interval, three coarse sand layers include composition, grain size, grading, and sedimentary structures similar to the sediments deposited by the December 26, 2004, tsunami. The layers are identified in five of the twenty-two cores, although the thicknesses vary. Six additional less well constrained sand layers are present in four of the twenty-two cores. Cores located closer to the lagoon mouth and the eastern coastline (the direction from which the 2004 tsunami arrived) contain more sand layers than cores farther away from the tsunami wave entry point. On the basis of their sedimentary structures, geometry, and extent, these sandy layers are interpreted to represent paleotsunami deposits. AMS radiocarbon dating was used to date the bulk organic sediment from above, between, and below the ten paleotsunami layers in sediment cores from Karagan Lagoon to constrain the timing of events in southeastern Sri Lanka. Material from within the deposit was not dated because it was likely transported from various sources during the event and thus does not represent the age of the tsunami. AMS radiocarbon dates from above and below the paleotsunami layers were calibrated from radiocarbon years before present to calendar years before present (Cal YBP) using OxCal v. 4.0 (Bronk Ramsey, 1995; Bronk Ramsey, 2001) with calibration curve IntCal04 (Reimer et al., 2004). The constraining time intervals of tsunami deposits II?VI were averaged to yield deposits of ages 226, 1641, 4198, 4457, 4924 Cal YBP. Tsunamis VII?X only had sediment dated immediately below the deposit and therefore were deposited prior to 6249, 6455, 6665, and 6840 Cal YBP. In total, ten tsunami deposits, including the 2004 event, are preserved in Karagan Lagoon on the southeastern coast of Sri Lanka. The Karagan Lagoon paleotsunami deposits provide constraints on the recurrence interval of tsunamis similar in magnitude to the 2004 event. The uppermost paleotsunami units were deposited 226, 1641, 4198, 4457, and 4924 Cal YBP, based on AMS radiocarbon dating. Thus, including the 2004 event, six tsunamis affected Karagan Lagoon in the past 5500 years, yielding a recurrence interval of approximately 916 years. Three of the six events, however, occur between ~4000 and 5500 years yielding a recurrence interval of approximately 500 years for this 1500 year period. Four additional older paleotsunami deposits occur in the deeper sections of the cores and were deposited prior to 6249, 6455, 6665, and 6840 Cal YBP, yielding a recurrence interval of approximately 200 years for this time period. Assuming that Karagan Lagoon contains a complete record of tsunami events, the recurrence of tsunamis similar in magnitude to the December 26, 2004, event can occur as often as 200 years. This ?recurrence interval? is illustrated by our data for the time period with increased tsunami activity from ~4000 to 7000 Cal YBP. Tsunamis may potentially affect Sri Lanka at relatively high frequency during certain time intervals though the overall recurrence pattern of these events displays a highly irregular distribution. This extreme variability needs to be taken into consideration when such events are related to earthquake recurrence intervals. Prior to the December 26, 2004, tsunami, paleotsunami deposits in the Indian Ocean were largely unstudied and consequently, Holocene tsunami chronology was incompletely understood for the Indian Ocean. The results from this study represents the first geologic evidence of paleotsunami deposits in Sri Lanka generated by tsunamis during the past 7000 years. The identification of these paleotsunami deposits illustrates that the 2004 tsunami was not a ?one-time event,? but in fact has ancient counterparts.

Several laterally extensive candidate tsunami deposits are preserved along coastlines facing the eastern Java Trench, indicating it has experienced mega-thrust earthquakes in the past. We investigated 37 coastal sites in Bali, Lombok, Sumba and Timor islands, many of which preserve course sand and pebble layers that overlie sharp basal contacts with scour marks into the mud, fine upward in grain size, and have bimodal grain size distributions. Other unique features are the common occurrence of marine fossils and concentrations of heavy minerals. The occurrence of these high-energy deposits interlayered with clay-rich units indicates the coarse clastics are anomalous because they were deposited in what is normally a very low-energy depositional environment. The lateral extent and paucity of thin, coarse clastic layers with marine organisms are inconsistent with local stream flood event, and the proximity to the equator of the sites diminishes the possibility of marine flood events from cyclones. The sparse, but consistent, the occurrence of at least two candidate tsunami deposits at depths of 1 and 2 meters over 950 km along the strike of the Java Trench may reveal that mega-thrust earthquakes have occurred there and generated giant tsunamis in the recent past. Five widely scattered imbricated boulder deposits are also found on Bali, Lombok, and Sumba. The boulders consist of slabs of hardpan up to 2.5 m in length and 80 cm thick that was torn from a near-shore seabed and stacked on top of one another. Some of the boulders were carried over the erosional coastal bank and deposited up to 100 meters inland. Comparisons with imbricated boulder ridges formed during the 1994 tsunami in east Java indicate that these deposits are from one or multiple tsunamis sourced by the Java Trench. Experiments in effective ways to communicate and implement tsunami disaster mitigation strategies have led us to train local communities about the 20-20-20 rule. If coastal communities experience more than 20 seconds of shaking from an earthquake, even if it is not intense, they should evacuate the coast. The time delay between the earthquake and arrival of tsunami waves is around 20 minutes, which is the time window for evacuation. Some tsunami waves may be as high as 20 meters, which is the target elevation for evacuation. Adopting the 20-20-20 rule could save thousands of lives throughout the region, especially in Bali where nearly 1 million people inhabit likely tsunami inundation zones.

京都大学 / 新制・課程博士 / 博士(理学) / 甲第23455号 / 理博第4749号 / 新制||理||1681(附属図書館) / 京都大学大学院理学研究科地球惑星科学専攻 / (主査)准教授 成瀬 元 助教 松岡 廣繁, 教授 生形 貴男 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

Samples from 13 different sites along the south coast of West Java yield 7 candidate paleo-tsunami sands, which may represent 4 different paleo-tsunami events. Ages obtained from one deposit may document a tsunami and coastal subsidence from an earthquake in 1,053 AD. The tsunami deposit from this event is preserved in an uplifted marine terrace exposed at Panto Cape, Banten Province. We speculated that the terrace has been uplifted about 4.6 m to the present height of 2 m above sea level, since the 1053 AD event at a rate of 4.8 mm/a. This uplift is strong evidence that strain is accumulating at the Java Trench and enough has already accumulated to generate a megathrust earthquake event.Numerical models using ComMIT of possible megathrust earthquake scenarios were constructed using the 2004 Sumatra earthquake, 30-m fault slip, and the 2011 Japan earthquake as proxies. These three scenarios yield earthquakes of Mw 9.3, 9.5 and 8.9, respectively. The worst case scenario is used to estimate the extent of tsunami inundation of the SW coast of Java, which totals 643 km2. The total number of people who inhabit the inundation area is around 451,000. Some coastal configurations cause a no escape situation where the modeled tsunami arrives in less than 20 minutes, which is not enough time for those near the coast to escape far enough inland or to a sufficient elevation to avoid the tsunami. These areas include the coastlines of Sukabumi, Cianjur and west Garut Regencies and the Pameungpeuk area.

Samples from 13 different sites along the south coast of West Java yield 7 candidate paleo-tsunami sands, which may represent 4 different paleo-tsunami events. Ages obtained from one deposit may document a tsunami and coastal subsidence from an earthquake in 1,053 AD. The tsunami deposit from this event is preserved in an uplifted marine terrace exposed at Panto Cape, Banten Province. We speculated that the terrace has been uplifted about 4.6 m to the present height of 2 m above sea level, since the 1053 AD event at a rate of 4.8 mm/a. This uplift is strong evidence that strain is accumulating at the Java Trench and enough has already accumulated to generate a megathrust earthquake event.Numerical models using ComMIT of possible megathrust earthquake scenarios were constructed using the 2004 Sumatra earthquake, 30-m fault slip, and the 2011 Japan earthquake as proxies. These three scenarios yield earthquakes of Mw 9.3, 9.5 and 8.9, respectively. The worst case scenario is used to estimate the extent of tsunami inundation of the SW coast of Java, which totals 643 km2. The total number of people who inhabit the inundation area is around 451,000. Some coastal configurations cause a no escape situation where the modeled tsunami arrives in less than 20 minutes, which is not enough time for those near the coast to escape far enough inland or to a sufficient elevation to avoid the tsunami. These areas include the coastlines of Sukabumi, Cianjur and west Garut Regencies and the Pameungpeuk area.

L’étude des dépôts de tsunamis est une science relativement jeune, et si de nombreux progrès ont été faits, en particulier au cours des trois dernières décennies, les techniques disponibles à l’heure actuelle ne permettent pas l’étude exhaustive des dépôts, que ce soit à cause de problèmes de conservation, fréquemment rencontrés, ou de limites liées aux techniques analytiques elles-mêmes. En effet, leur résolution spatiale s’avère insuffisante pour les dépôts fins (fractions sableuses et inférieures), soit par leur principe même (absence de visualisation en trois dimensions sur des lames minces, par exemple), soit à cause des techniques d’échantillonnage dont elles dépendent. Ce travail a permis d’explorer les applications d’une technique analytique jusqu’à présent inutilisée dans ce domaine, la microtomographie par rayons X, appliquée à trois dépôts de tsunamis : les dépôts du tsunami de Lisbonne en 1755 sur les côtes Andalouses, ceux d’un tsunami causé en 1996 dans le Lac Karymskoye au Kamchatka, et un potentiel dépôt de tsunami datant de l’Holocène à Ninety Mile Beach, sur le littoral ouest de l’Australie. Les problématiques liées aux techniques d’échantillonnage et au traitement des données brutes et à leur exploitation sont passées en revue et les solutions retenues sont exposées. Les résultats obtenus sont comparés à ceux obtenus par l’utilisation de techniques analytiques dont la pertinence et l’utilité ont été démontrées lors de précédentes études. Cette technique se révèle extrêmement utile pour les analyses structurales (organisation interne des dépôts), texturales (distributions de tailles de grains, fabrique sédimentaire et dynamiques d’écoulements), ainsi que les analyses de composition (abondances en bioclastes et en minéraux lourds par exemple). Le croisement des études structurales et texturales du dépôt du tsunami de 1755 mettent en avant des dynamiques de mise en place allant parfois à l’encontre des scenarii généralement admis et offre un nouveau point de vue sur les dépôts de tsunami. / The study of tsunami deposits is a relatively young science, and even if substantial progress has been made (especially during the last three decades) the techniques available at present day do not allow exhaustive studies of the deposits, weither poor conservation of the deposits, which is a common problem, is to blame, or because of limitations of the techniques themselves. Their spatial resolution often appears to be insufficient for fine deposits (sand size fractions and finer) because of their working principle (lack of three dimensional visualisation of structures on thin sections for instance) or because of the sampling techniques they depend on. This work allowed exploring the applications of an analytic technique, X-ray computed microtomography, which had to this day never used before for the study of tsunami deposits. Three distinct case studies have been made: deposits from the 1755 Lisbon tsunami on the western coast of Spain (Andalusia), deposits from the 1996 tsunami triggered in Karymskoye Lake, and a potential mid-Holocene tsunami deposit at Ninety Mile Beach, western Australia. Problematic linked to sampling techniques and raw data processing and exploitation are addressed. Chosen solutions are exposed. The results are compared to those obtained using techniques whose relevance and usefulness have been proven in previous studies. This technique proves to be extremely useful for structural (deposits inner organisation), textural (grain size distribution, sedimentary fabrics and flow dynamics), and composition (bioclasts and heavy minerals abundances) analyses. Crossing structural and textural analysis of the 1755 Lisbon tsunami deposit unravels deposition modes which go against the usually admitted scenario and offers a new point of view on tsunami deposits.

Maps and models of seismic and tsunami risk are constructed from a variety of measurements taken in Indonesia, which have the potential to reduce loss of life and infrastructure. The first study uses the multichannel analysis of surface waves (MASW) method to calculate the time-averaged shear wave velocity to 30 m depth (Vs30). These measurements were taken at 58 sites in the city of Pacitan, Java and on the islands of Lombok, Ambon, and the Banda Islands. Vs30 calculations are compared with local geologic maps to extrapolate site class for locations not measured directly. Site class maps are then compared with Modified Mercalli Intensity (MMI) observations for three earthquake events that impacted Lombok and Ambon to identify regions where the MMI and Vs30 do and do not corroborate one another. Consistent with other Vs30 studies, the lowest values are observed on coastal alluvial plains and the highest values on steeper hillsides underlain by volcanic deposits. The second study focuses on a potential sector collapse of the volcano Banda Api within the Banda Islands. A field survey of its summit identified a steeply dipping normal fault striking NNE-SSW. This, along with the fissure geometry of the volcano's most recent eruption, reveals a failure plane along which a future sector collapse could occur. The numerical model Tsunami Squares (TS) predicts that the tsunami produced by this landslide would inundate an estimated 63% of buildings on the Banda Islands with waves as high as 82 m. These findings highlight the importance of installing a GPS receiver array on Banda Api to monitor the motion of its slopes. The third study analyzes sediment from trenches on the Banda Islands and Ambon to test if historical tsunamis that have impacted the area are preserved in the geological record. Potential tsunami deposits were identified by the presence of marine sand and larger clasts of marine carbonate in an environment which otherwise lacks large storms to bring such material onshore. Several dating methods constrain the ages of at least seven candidate tsunami deposits found in trenches onshore. One of these historical tsunamis (the event of November 26, 1852) is described in significant detail from several locations across the Banda Sea, which enables modeling of the event using a Bayesian statistical approach. The posterior of this model predicts the most likely epicenter was SW of Seram with a mean magnitude of Mw 8.8. It also makes other predictions about fault parameters. The region exhibits a marked slip deficit based on instrumental records of earthquakes in the area.