T24A-01
Hunting for Ancient Tsunamis in the Tropics
Paleotsunami deposits may prove harder to find in tidal wetlands and beach-ridge plains around the tropical Indian Ocean than in temperate but otherwise comparable settings on the Pacific Rim. The reasons for this challenge are probably unrelated to tsunami size or recurrence. Estuarine marshes and grassy beach-ridge plains provide widespread opportunities for tsunamis to lay down preservable sand sheets in northeastern Japan, Kamchatka, the northwestern United States, and south-central Chile. The small plants of these lowlands offered little resistance to tsunami flow. Coseismic subsidence and net late Holocene submergence provided caps of tidal mud that help preserve the sand. By contrast in tidal wetlands overrun by the 2004 Indian Ocean tsunami, mangrove swamps and their inhibit the formation and preservation of tsunami sand sheets. For example, in mangrove swamps along tidal inlets near Ban Nam Kem and Tab Lamu, Thailand, sandy deposits of the 2004 tsunami were probably limited to feather- shaped channel-margin areas where the tsunami lost much of its momentum to the toppling of leafy trees. When these deposits were examined in July 2006, deposit-feeding crabs were busily mixing the sand into muddy, peaty mangrove soils. Such limitations of mangrove swamps as paleotsunami recorders may help explain why a reconnaissance in May 2006 turned up no sand sheets in the soils of mangroves near Cilacap, on the south coast of Java. Thus far it is unclear whether this coast, which faces the Sunda Trench, lacks potential for tsunamis as enormous as Aceh's in 2004, or whether it has a history of enormous tsunamis that simply failed to leave a long-lasting record in the Cilacap mangroves. Disturbance by humans limits the paleotsunami targets on beach-ridge plains facing the Indian Ocean. Thai coastal plains, though apparently grassy where undisturbed, have been extensively modified by placer mining for tin. In Java and southeastern India, most coastal plains have been under the plow for centuries. The outlook for paleotsunami hunting in South and Southeast Asia probably depends on new targets that include coral boulders and scarcely disturbed beach-ridge plains in Thailand; archaeological sites that provide cultural timelines in India; lagoons of Sri Lanka; coastal rivers that offer cutbanks and oxbows on beach-ridge plains of Java; delicately laminated deposits of salt flats routinely overrun by storm surges on the arid northern shores of the Arabian Sea; and records of prehistoric land-level change close to fault-rupture areas along the Sunda Trench.
T24A-02
A Pre-2004 Tsunami Deposit in Thailand
A sand layer from a predecessor to the 2004 Indian Ocean tsunami underlies a freshwater marsh on Phra Thong Island, Phang Nga Province, Thailand. At this marsh, about 300 meters from the sea, the 2004 tsunami left a sand layer about 13 cm thick on top of a peaty soil. The previous tsunami deposit, of similar thickness, rests on another peaty soil about 40 cm below present ground surface. It probably correlates with a pre-2004 tsunami deposit previously reported from Phra Thong Island in a web posting by Shigehiro Fujino. Observed in trenches and cores, the pre-2004 deposit begins with a razor-sharp contact with the underlying soil. The sand itself contains no fewer than 6 layers, all horizontal, defined by differences in particle size (fine sand and very fine sand) and by detrital plant fragments. We also noticed pebble-size clasts of peaty soil. These observations come from the vicinity of UTM coordinates 418975E, 1009496N, zone 47. Many of us participated in previous, unsuccessful attempts to find pre-2004 tsunami deposits in Thailand. We hunted for sand sheets between Phuket to the south and Kho Khao Island to the north, on beach-ridge plains disturbed by tin mining and in mangrove swamps. We also checked pristine, grassy beach-ridge plains of Phra Thong Island, the next island north of Kho Khao. None of these environments offer both a host deposit that contrasts with tsunami-laid sand and a lack of disturbance by tin miners, mud lobsters, and real-estate developers.
T24A-03
Estimating the Threat of Tsunamigenic Earthquakes and Earthquake Induced-Landslide Tsunami in the Caribbean
Deformation along the margin of the Caribbean Plate is the principal cause of the tsunami threat in the Caribbean. That margin parallels the northern coast of South America, the Eastern side of the Lesser Antilles, and extends along the Greater Antilles from Puerto Rico through Jamaica. The eastern boundary of the Caribbean plate near the Lesser Antilles is the locus of subduction of Atlantic seafloor. At least three distinct, shallow tectonic regimes parallel that margin. They are: an outer tectonic belt where the North/South America Plate bends to enter the subduction zone, the zone of contact between the plates, and an inner zone of intraplate activity in the overriding Caribbean Plate. The level of seismic activity and tsunami potential in each of these zones is influenced by the presence of rough seafloor on the downgoing plate. Aseismic Ridges may increase the probability slow earthquakes by changing the physical characteristics of the base of the accretionary prism. The northeastern corner of the Caribbean Plate margin undergoes a smooth transition from the relatively simple head-on subduction zone in the N. Lesser Antilles into a region of oblique convergence to the west. It is a complex margin dominated by microplate tectonics from near Puerto Rico through Hispaniola. Here too the same three tectonic zones can be defined, but the third zone, "intraplate activity in the Caribbean Plate", is more clearly delineated as microplate deformation in a wide plate boundary zone. Strike-slip tectonics dominates from Haiti westward to northern Honduras. Local bends in the transcurrent fault systems lead to vertical tectonics in the form of push-up and pull-aparts. Other potential sources lie along the northern coast of S. America and off the east coast of C. America. Manipulation of gridded bathymetry to produce maps of seafloor slope, slope direction, and residual bathymetry, aid the identification of potential slides/slumps, or reveal portions of subduction zones more likely to produce slow earthquakes. Subduction of rough seafloor may activate thrust faults within the accretionary prism above the main decollement, causing indentation of the prism toe. Later reactivation of a dormant decollement would enhance the possibility of slow earthquakes. Subduction of significant seafloor relief and corresponding indentation of the accretionary prism toe would then be another parameter to estimate the likelihood of slow earthquakes. Using these criteria, several regions of the Northeastern Caribbean stand out as more likely sources for slow earthquakes.
T24A-04
A multi-proxy approach to tsunami deposit characterization: searching wetlands and lagoons at the Mexican Pacific coast for a record of tsunami inundations
It is extremely important to integrate historic accounts with geologic evidence of past tsunamis to fully understand hazards to the highly populated Mexican coast. We aim to complete a record of tsunami events using a multi- proxy approach to identify tsunami deposits recovered from marsh lagoonal sediments on the Guerrero coast of Mexico. Core samples from nine study sites on tropical wetlands show distinct stratigraphic changes with depth, indicating clear rapid change in depositional environments over time. Core sequences recovered include evidence of land level changes and three probable tsunamis based on stratigraphic, granulometric, geochemical, and microfossil analysis (diatoms and pollen) of marsh sediments. Radiocarbon data indicate that the sediment units cored are up to ca. 6161+/- 53 yr BP. A probable tsunami event and coseismic subsidence dates back at 3800 yr BP. Discrete sand units fining up show sharp erosional contacts at the base, medium to poor sorting, fine to coarse sand, with clay/mud rip-up clasts. However, traces of these apparent tsunami deposits are not extensive over a large area and we could not correlate them from one site to another. We hypothesize that the distribution and preservation of tsunami deposits in these tropical lagoonal marshes and wetlands might be controlled by mangrove vegetation fringing coastal lagoons and estuaries. Recent observations after the 2004 Indian Ocean tsunami showed that mangroves may dissipate tsunami energy and result in flumed-shape sediment deposition. If this is the case, finding tsunami deposits over extensive areas in mangrove marshes and wetlands would prove very challenging.
T24A-05
Mangroves, Coral, and the Search for a Paleotsunami Deposit Along the Andaman Coast of Thailand.
Finding a preserved paleotsunami deposit along Thailand's Andaman coast has proven to be a major challenge. The coastline is tectonically stable precluding the preservation of tsunami deposits due to sudden sea level changes. Much of the coast is fringed by wide beach plains that have been thoroughly disturbed during a long history of placer tin mining. In the rare undisturbed areas, distinguishing potential paleotsunami sand layers from interlayered beach deposits is frustratingly difficult. In the central part of Phangnga Province, several large areas of mangrove-fringed coastal lagoons formed during Holocene sea level rise. The fine sand, mud and peat that accumulated in these quiet lagoons provide a suitably contrasting host for coarser tsunami deposits. The December 2004 tsunami entered these tidal channels destroying swaths of mangroves up to 200 m wide. Satellite images showing the patterns of destroyed mangroves suggest that the energy of the tsunami varied greatly along the lagoons' shores. Sampling in shallow trenches along transects across these swaths suggests that the sand layer marking this tsunami is thickest where the mangroves are destroyed, and thins rapidly landward into the standing mangroves. Intense bioturbation within the mangroves is already mixing the sand with the underlying peat and mud. Thus, deposition of any paleotsunami sand layer would be as discontinuous lenses along the paleoshoreline, and mixing with the enclosing sediment may mask it. Perhaps the best opportunity for discovering a paleotsunami deposit is to locate an unusually coarse deposit that could endure bioturbation. We found one such deposit in the Thap Lamu lagoon where the 2004 tsunami deposited a litter of angular coral-gravel blanketing an unusually large zone of destroyed mangroves. Cores collected along two transects across this zone penetrated a similar layer of coral gravel at 1-2 m depth. This coral-gravel layer thins landward, and cross cuts a facies boundary between peat and lagoonal mud formed as mangroves encroached across the lagoon, suggesting that the coral-gravel layer represents an event horizon and not an unusual facies. Preliminary AMS 14C dates are equivocal but suggest an age of around 3000 cy BP for the coral gravel layer. We hypothesize that both the modern and buried coral gravel layer represent tsunami deposits derived from a similar offshore point source. Discovery of other unusually coarse, paired deposits in this region would support our interpretation.
T24A-06
Characteristics of nearshore marine tsunami sediments from SE coast of India
The geologic records of tsunami deposits on coastal lands have been studied along the southeast coast of India. The December, 2004 tsunami sediments deposited in marine environment of Tamil Nadu State on the southeast coast of India were sampled within 5 days after the tsunami event and the samples were compared with the samples collected from the same locations 30 days before tsunami. Two core samples and 10 surface sediment samples were collected along two transects. The change in the bathymetry along transects, between pre and post tsunami conditions indicate both erosion and deposition in the continental shelf. Overall, there were erosion at 20 m water depth and deposition at 5 to 10 m water depth. The core samples collected from 20 m water depth show an abrupt change in sediment characteristics at 25 cm depth. A sand sheet is observed up to 25 cm depth and it has been recognized as tsunami deposit. The deposits are also enriched in heavy minerals, low in organic carbon, broken foraminiferal species, abraded test species indicating large amount of backwash from the land area.
T24A-07
Sedimentary deposits from the July 17th 2006 Java tsunami on the West Australian coastline
On July 17th 2006, an Mw = 7.7 earthquake south of Java generated a tsunami that devastated parts of the Javanese coast, killing more than 500 people. The tsunami also affected parts of the Western Australian coast. Within a week of the event, a post tsunami survey was carried out near Steep Point, Western Australia. Tsunami inundation and run-up were mapped on the basis of eyewitness accounts, debris lines, vegetation damage and the occurrence of recently deposited fish, starfish, corals and sea urchins well above high-tide mark. Eyewitnesses reported three waves in the tsunami wave train, the second being the largest. A topographic survey using kinematic GPS with accuracies of 0.02 metres in the horizontal and 0.04 metres in the vertical recorded inundation depths of between 1-2 m, inundation of up to 200 m inland, and a maximum recorded run-up of 7.9 m AHD (Australian Height Datum). The tsunami caused widespread erosion in the littoral zone, extensive vegetation damage, destroyed several campsites (including transporting a large vehicle ten metres inland) and deposited extensive sediment sheets over coastal dunes. At their seaward edge, these sediments are up to 14 cm thick, tapering landwards over approximately 200 m. The deposits are composed of moderately well sorted, medium grained silicic sand with some gravel and organic debris. A basal unconformity defines the boundary between tsunami sediments and underlying aeolian dune sand. Buried green vegetation stems with roots traceable to the lower dune unit occur within the tsunami deposit. No roots are present within the tsunami-deposited sediments. This is taken as evidence for the recent mobilisation of the sediments within the tsunami sediment sheet. Evidence for up to three individual waves is preserved as normally graded sequences mantled by layers of dark grey, organic-rich fine silty sand. Grainsize and microfossil analyses are in progress. Given the strong wind regimes in the area, and the similarity of the underlying dune deposits to the tsunami sediments, it is likely that seasonal erosion will remove all traces of these sediment sheets within years to decades.
T24A-08
Geological recognition of tsunami and storm deposits in two coastal areas of Portugal
Abrupt marine invasions such as tsunamis and storms are particularly devastating for coastal areas. They may also leave a permanent record in sedimentary deposits. In historical times the most destructive tsunami that affected Europe was the Lisbon AD 1755. This presentation aims to contribute to a better understanding of the signature left by abrupt marine invasions in coastal stratigraphy by investigating the nature of the sedimentary record associated with tsunamis in a region of their known impact. The techniques used include stratigraphic description, grain size analysis, digital and x-ray photography, magnetic susceptibility, macrofossils analysis, geochemical analysis and 210Pb and Optically Stimulated Luminescence dating. The investigated areas (e.g. Lagoa de Óbidos (Central Portugal) and Martinhal (South Portugal)) were affected by the AD 1755 tsunami. The locations have similar geomorphological features and are both susceptible to major abrupt marine invasions. Results show that an abrupt event deposited unique sedimentary units in both locations. A similar age for the event was established. A considerable number of tsunami sedimentary characteristics, (stratigraphical, granulometric, palaeontological and geochemical) were detected in both sites. Moreover, other abrupt marine invasions were detected in the lithostratigraphy of the two areas. However, a key outcome of this research is the demonstration of the difficulty of distinguishing between sedimentary deposits laid down by tsunamis, and those deposits resulting from storm action.