David R. Tappin

Seafloor morphology of the Sumatran subduction zone: Surface rupture during megathrust earthquakes?

High-resolution multibeam bathymetry data from the Sumatran subduction zone reveal the regional and local morphology, including small-scale fault-related features and landslides that may be linked to earthquakes in the recent geological past. The accretionary prism is steeply sloped and pervasively eroded, with evidence of unusual landward vergence (seaward fault dip) of the frontal thrusts. Small-scale (5–100 m height) fault scarps, folds, and troughs are common along the seaward edge of the frontal thrust at the deformation front. A model of back-thrust fault slip or bending moment folding during plate-boundary slip, such as during the 2004 M9.2 megathrust earthquake, can explain the position of these features on the seaward fold limb of a seaward-dipping thrust. We infer that in major Sumatran (and other similar settings) plate-boundary earthquakes, coseismic surface rupture may occur at the prism toe.

Megatsunami deposits on Kohala volcano, Hawaii, from flank collapse of Mauna Loa

The origin of coastal and high-elevation marine gravels on the Hawaiian islands of Lanai and Molokai is controversial, because the vertical tectonics of these islands is poorly constrained. The gravels are either from eustatic highstands or were left by massive tsunamis from offshore giant landslides. In contrast, at Kohala on the island of Hawaii, where continuous subsidence is well established, lithofacies analysis and dating of a fossiliferous marine conglomerate 1.5–61 m above present sea level support a tsunami origin and indicate a runup of >400 m >6 km inland. The conglomerate age, 110 ± 10 ka, suggests a tsunami caused by the ca. 120 ka giant Alika 2 landslide from nearby Mauna Loa volcano.

Submarine mass failures as tsunami sources: their climate control

Recent research on submarine mass failures (SMFs) shows that they are a source of hazardous tsunamis, with the tsunami magnitude mainly dependent on water depth of failure, SMF volume and failure mechanism, cohesive slump or fragmental landslide. A major control on the mechanism of SMFs is the sediment type, together with its post-depositional alteration. The type of sediment, fine- or coarse-grained, its rate of deposition together with post-depositional processes may all be influenced by climate. Post-depositional processes, termed sediment ‘preconditioning’, are known to promote instability and failure. Climate may also control the triggering of SMFs, for example through earthquake loading or cyclic loading from storm waves or tides. Instantaneous triggering by other mechanisms such as fluid overpressuring and hydrate instability is controversial, but is here considered unlikely. However, these mechanisms are known to promote sediment instability. SMFs occur in numerous environments, including the open continental shelf, submarine canyon/fan systems, fjords, active river deltas and convergent margins. In all these environments there is a latitudinal variation in the scale of SMFs. The database is limited, but the greatest climate influence appears to be in high latitudes where glacial/interglacial cyclicity has considerable control on sedimentation, preconditioning and triggering. Consideration of the different types of SMFs in the context of their climate controls provides additional insight into their potential hazard in sourcing tsunamis. For example, in the Atlantic, where SMFs are common, the tsunami hazard under the present-day climate may not be as great as their common occurrence suggests.

Mass Wasting Processes - Offshore Sumatra

Earthquakes are a commonly cited mechanism for triggering submarine landslides that have the potential to generate locally damaging tsunamis. With measured runups of over 35 metres in northern Sumatra from the December 26th 2004 tsunami source, these runups might be expected to be due, in part, to local submarine landslides. Mapping of the convergent margin offshore of Sumatra using swath bathymetry, single channel seismic and seabed photography reveals that seabed failures are common, but mainly small-scale, and composed of blocky debris avalanches and sediment flows. These failures would have contributed little to local tsunami runups. Large landslides are usually formed where there is significant sediment input. In the instance of Sumatra, most sediment is derived from the oceanic plate, and there is little sediment entering the system from the adjacent land areas. Input from the oceanic source is limited because of the diversion of sediment entering the subduction system off of Sumatra, that is attributed to collision between the Ninetyeast ridge and the Sunda Trench at approximately 1.5 million years ago.

Evidence for kilometre-scale Neogene exhumation driven by compressional deformation in the Irish Sea basin system

Abstract Large tracts of the NW European continental shelf and Atlantic margin have experienced kilometre-scale exhumation during the Cenozoic, the timing and causes of which are debated. There is particular uncertainty about the exhumation history of the Irish Sea basin system, Western UK, which has been suggested to be a focal point of Cenozoic exhumation across the NW European continental shelf. Many studies have attributed the exhumation of this region to processes associated with the early Palaeogene initiation of the Iceland Plume, whilst the magnitude and causes of Neogene exhumation have attracted little attention. However, the sedimentary basins of the southern Irish Sea contain a mid–late Cenozoic sedimentary succession up to 1.5 km in thickness, the analysis of which should permit the contributions of Palaeogene and Neogene events to the Cenozoic exhumation of this region to be separated. In this paper, an analysis of the palaeothermal, mechanical and structural properties of the Cenozoic succession is presented with the aim of quantifying the timing and magnitude of Neogene exhumation, and identifying its ultimate causes. Synthesis of an extensive apatite fission-track analysis (AFTA), vitrinite reflectance (VR) and compaction (sonic velocity and density log-derived porosities) database shows that the preserved Cenozoic sediments in the southern Irish Sea were more deeply buried by up to 1.5 km of additional section prior to exhumation which began between 20 and 15 Ma. Maximum burial depths of the preserved sedimentary succession in the St Georges Channel Basin were reached during mid–late Cenozoic times meaning that no evidence for early Palaeogene exhumation is preserved whereas AFTA data from the Mochras borehole (onshore NW Wales) show that early Palaeogene cooling (i.e. exhumation) at this location was not significant. Seismic reflection data indicate that compressional shortening was the principal driving mechanism for the Neogene exhumation of the southern Irish Sea. Coeval Neogene shortening and exhumation is observed in several sedimentary basins around the British Isles, including those along the UK Atlantic margin. This suggests that the forces responsible for the deformation and exhumation of the margin may also be responsible for the generation of kilometre-scale exhumation in an intraplate sedimentary basin system located >1000 km from the most proximal plate boundary. The results presented here show that compressional deformation has made an important contribution to the Neogene exhumation of the NW European continental shelf.

Survey presents broad approach to tsunami studies

The Sumatra Earthquake and Tsunami Offshore Survey (SEATOS) investigated the seafloor area near the 26 December 2004 earthquake. An early look at the data suggests that only small ground motions occurred. This raises major questions about how the enormous amount of energy from that earthquake could dissipate over relatively short distances, and whether the ground motion is characteristic of other active margins. SEATOS, which included an international, interdisciplinary team of scientists, conducted a detailed seafloor expedition in the vicinity of the epicenter of the Great Sumatra earthquake, in May 2005. The success of this interdisciplinary effort has led to a follow-on workshop and a Union session at the upcoming AGU Fall Meeting on 5–6 December in San Francisco.

Architecture and Failure Mechanism of the Offshore Slump Responsible For the 1998 Papua New Guinea Tsunami

After considerable controversy over the origin of the July 1998 PNG tsunami, there is now a large body of evidence that supports a sediment slump offshore of the devastated area. Between 1999 and 2000, four surveys were carried out offshore of the affected area, acquiring bathymetry, sediment cores, 3.5kHz seismic, multi-channel seismic and seabed imagery. In 2001, the same area was surveyed using single channel seismic that has been used to interpret the northern margin of PNG and the internal architecture of the slump. The susceptibility to slumping of the area offshore of northern PNG can be more definitively assessed.

Volcaniclastic gravity flow sedimentation on a frontal arc platform: The Miocene of Tonga

Abstract Marine volcaniclastic gravity flow deposits of Miocene age are described from island exposures on the Tongan frontal arc platform (southwest Pacific Ocean). Background sedimentary rocks between gravity flow beds include non‐calcareous brown mudstone, calcareous pebbly sandstone, and chalk. Depositional environments inferred from microfaunas, macrofaunas, trace fossils, and sedimentary structures range from shallow (shelf) to deep water (c. 1500 m). The depth range of the deposits is considered deeper than continental shelf, and shallower than typical non‐volcanic large‐scale depositional gravity flow environments such as submarine fans. Six lithofacies are distinguished. They embrace a wide range of gravity flow deposits, but within each lithofacies/environment there is one dominant association. The lithofacies contain varying proportions of mafic and silicic volcanic clasts. Some are solely mafic, some contain interleaved mafic and silicic intervals, and some contain mixed mafic and silicic clasts in the same beds. Clast size ranges from silt (<1/16 mm) to boulders (>64 cm). Accretionary lapilli are present in three lithofacies. The dominant gravity flow mechanisms were turbidity currents and debris flows. Derivation from underwater eruptions is likely in some lithofacies, while others are likely to be from subaerial eruptions. It is rarely possible to make the distinction from the clasts themselves. On Mango Island, bouldery debris flow material was transferred directly from a probable subaerial volcano to the basin. In all other cases a marked upper limit of clast size suggests that eruption process(es), or processes in the transfer of sediment before generation of gravity flows, effectively removed the largest clasts (>5 cm). The overall control on deposition is considered to be eruption‐controlled sediment supply.

Mass Transport Events and Their Tsunami Hazard

Mass transport events, such as those from submarine landslides, volcanic flank collapse at convergent margins and on oceanic islands, and subaerial failure are reviewed and found to be all potential tsunami sources. The intensity and frequency of the tsunamis resulting is dependent upon the source. Most historical records are of devastating tsunamis from volcanic collapse at convergent margins. Although the database is limited, tsunamis sourced from submarine landslides and collapse on oceanic volcanoes have a climate influence and may not be as hazardous as their frequency suggests. Conversely, tsunamis sourced from submarine landslides at convergent margins may be more frequent historically than previously recognized and, therefore, more hazardous.

Digital elevation models in the marine domain: investigating the offshore tsunami hazard from submarine landslides

Abstract Digital elevation models (DEMs) of seabed relief are now commonly available at a number of scales. On a global scale three-dimensional (3D) relief maps of the ocean floor are derived from satellite gravity measurements validated by single-beam echo soundings. On a smaller, more local, scale, the development of multibeam bathymetric mapping technology provides detailed seabed data from which DEMs are derived. Over the past 30 years multibeam bathymetry has replaced single-beam echo soundings as the main tool used to map the sea floor. Multibeam bathymetry has revolutionized our ability to interpret seabed morphology. It has the capability to provide complete seabed coverage and gives a 3D visualization of the seabed not previously available. DEMs derived from multibeam are comparable to those on land. One aspect of the improved seabed visualization is in mapping marine geohazards. Here three DEMs, from Papua New Guinea, Hawaii and the Indian Ocean, are presented. These DEMs have been used to investigate submarine seabed failure and volcanic flank collapse in the context of their tsunami hazard. For these three areas the DEMs contribute to an improved interpretational capability in marine geohazards. In addition, the DEMs underpin newly developed modelling methodologies of landslide-generated tsunami.