A simple model for calculating tsunami flow speed from tsunami deposits

This paper presents a simple model for tsunami sedimentation that can be applied to calculate tsunami flow speed from the thickness and grain size of a tsunami deposit (the inverse problem). For sandy tsunami deposits where grain size and thickness vary gradually in the direction of transport, tsunami sediment transport is modeled as a steady, spatially uniform process. The amount of sediment in suspension is assumed to be in equilibrium with the steady portion of the long period, slowing varying uprush portion of the tsunami. Spatial flow deceleration is assumed to be small and not to contribute significantly to the tsunami deposit. Tsunami deposits are formed from sediment settling from the water column when flow speeds on land go to zero everywhere at the time of maximum tsunami inundation. There is little erosion of the deposit by return flow because it is a slow flow and is concentrated in topographic lows. Variations in grain size of the deposit are found to have more effect on calculated tsunami flow speed than deposit thickness. The model is tested using field data collected at Arop, Papua New Guinea soon after the 1998 tsunami. Speed estimates of 14 m/s at 200 m inland from the shoreline compare favorably with those from a 1-D inundation model and from application of Bernoulli's principle to water levels on buildings left standing after the tsunami. As evidence that the model is applicable to some sandy tsunami deposits, the model reproduces the observed normal grading and vertical variation in sorting and skewness of a deposit formed by the 1998 tsunami.     

Distinguishing tsunami and storm deposits: An example from Martinhal, SW Portugal

Geological identification of past tsunamis is important for risk assessment studies, especially in areas where the historical record is limited or absent. The main problem when using the geological evidence is to distinguish between tsunami and storm deposits. Both are high-energy events that may leave marine traces in coastal stratigraphic sequences. At Martinhal, SW Portugal both storm surge and tsunami deposits are present at the same site within a single stratigraphic sequence, which makes it suitable to study the differences between them, excluding variations caused by local factors.

The tsunami associated with the Lisbon earthquake of November 1st 1755 AD, had a major impact on the geomorphology and sedimentology of Martinhal. It breached the barrier and laid down an extensive sheet of sand, as described in eyewitness reports. Besides the tsunami deposit the stratigraphy of Martinhal also displays evidence for storm surges that have breached and overtopped the barrier, flooding the lowland and leaving sand layers. Both marine-derived flood deposits show similar grain size characteristics and distinctive marine foraminifera. The most important differences are the rip-up clasts and boulders exclusively found in the tsunami deposit and the landward extent of the tsunami deposit that everywhere exceeds that of the storm deposits. Identification of both depositional units was only possible using a collection of different data and extensive stratigraphical information from cores as well as trenches.     

Diatom biostratigraphy of tsunami deposits: Examples from the 1998 Papua New Guinea tsunami

Distinctive diatom assemblages may be associated with tsunami sediments and may often contrast with the assemblages found within sediments underlying the tsunami deposit as well as those associated with the modern coastal environment. Sediments associated with the 1998 tsunami that destroyed much of the Sissano lagoon area in northern Papua New Guinea have been investigated. Surface sediments from three transects across the sediment spit near Warapu have been examined for diatom content and preservation. The preservation is variable, and the data show an, often chaotic, assemblage that can be attributed to the tsunami waves incorporating and depositing diatoms from distinctive habitat zones during their runup and subsequent backwash. The diatoms identified within the Warapu sediments indicate an origin from within the inter-tidal and offshore area rather than from the beach–sand spit complex. The sand deposits disclose a high percentage, in excess of 75%, of broken diatom valves, and a predominance of centric (circular) species due to preferential preservation. The study demonstrates that the application of diatom biostratigraphy to modern tsunami deposits can be used in conjunction with other stratigraphical lines of evidence to interpret the source and provenance of historical and palaeo-tsunami deposits.     

Physical criteria for distinguishing sandy tsunami and storm deposits using modern examples

Modern subaerial sand beds deposited by major tsunamis and hurricanes were compared at trench, transect, and sub-regional spatial scales to evaluate which attributes are most useful for distinguishing the two types of deposits. Physical criteria that may be diagnostic include: sediment composition, textures and grading, types and organization of stratification, thickness, geometry, and landscape conformity.

Published reports of Pacific Ocean tsunami impacts and our field observations suggest that sandy tsunami deposits are generally < 25 cm thick, extend hundreds of meters inland from the beach, and fill microtopography but generally conform to the antecedent landscape. They commonly are a single homogeneous bed that is normally graded overall, or that consists of only a few thin layers. Mud intraclasts and mud laminae within the deposit are strong evidence of tsunami deposition. Twig orientation or other indicators of return flow during bed aggradation are also diagnostic of tsunami deposits. Sandy storm deposits tend to be > 30 cm thick, generally extend < 300 m from the beach, and will not advance beyond the antecedent macrotopography they are able to fill. They typically are composed of numerous subhorizontal planar laminae organized into multiple laminasets that are normally or inversely graded, they do not contain internal mud laminae and rarely contain mud intraclasts. Application of these distinguishing characteristics depends on their preservation potential and any deposit modifications that accompany burial.

The distinctions between tsunami and storm deposits are related to differences in the hydrodynamics and sediment-sorting processes during transport. Tsunami deposition results from a few high-velocity, long-period waves that entrain sediment from the shoreface, beach, and landward erosion zone. Tsunamis can have flow depths greater than 10 m, transport sediment primarily in suspension, and distribute the load over a broad region where sediment falls out of suspension when flow decelerates. In contrast, storm inundation generally is gradual and prolonged, consisting of many waves that erode beaches and dunes with no significant overland return flow until after the main flooding. Storm flow depths are commonly < 3 m, sediment is transported primarily as bed load by traction, and the load is deposited within a zone relatively close to the beach.     

Identification of tsunami deposits considering the tsunami waveform: An example of subaqueous tsunami deposits in Holocene shallow bay on southern Boso Peninsula, Central Japan

This study proposes a tsunami depositional model based on observations of emerged Holocene tsunami deposits in outcrops located in eastern Japan. The model is also applicable to the identification of other deposits, such as those laid down by storms. The tsunami deposits described were formed in a small bay of 10–20-m water depth, and are mainly composed of sand and gravel. They show various sedimentary structures, including hummocky cross-stratification (HCS) and inverse and normal grading. Although, individually, the sedimentary structures are similar to those commonly found in storm deposits, the combination of vertical stacking in the tsunami deposits makes a unique pattern. This vertical stacking of internal structures is due to the waveform of the source tsunamis, reflecting: 1) extremely long wavelengths and wave period, and 2) temporal changes of wave sizes from the beginning to end of the tsunamis.

The tsunami deposits display many sub-layers with scoured and graded structures. Each sub-layer, especially in sandy facies, is characterized by HCS and inverse and normal grading that are the result of deposition from prolonged high-energy sediment flows. The vertical stack of sub-layers shows incremental deposition from the repeated sediment flows. Mud drapes cover the sub-layers and indicate the existence of flow-velocity stagnant stages between each sediment flow. Current reversals within the sub-layers indicate the repeated occurrence of the up- and return-flows.

The tsunami deposits are vertically divided into four depositional units, Tna to Tnd in ascending order, reflecting the temporal change of wave sizes in the tsunami wave trains. Unit Tna is relatively fine-grained and indicative of small tsunami waves during the early stage of the tsunami. Unit Tnb is a protruding coarse-grained and thickest-stratified division and is the result of a relatively large wave group during the middle stage of the tsunami. Unit Tnc is a fine alternation of thin sand sheets and mud drapes, deposited from waning waves during the later stage of the tsunami. Unit Tnd is deposited during the final stage of the tsunami and is composed mainly of suspension fallout. Cyclic build up of these sub-layers and depositional units cannot be explained by storm waves with short wave periods of several to ten seconds common in small bays.     

Sheltered coastal environments as archives of paleo-tsunami deposits: Observations from the 2004 Indian Ocean tsunami

The 2004 earthquake left several traces of coseismic land deformation and tsunami deposits, both on the islands along the plate boundary and distant shores of the Indian Ocean rim countries. Researchers are now exploring these sites to develop a chronology of past events. Where the coastal regions are also inundated by storm surges, there is an additional challenge to discriminate between the deposits formed by these two processes. Paleo-tsunami research relies largely on finding deposits where preservation potential is high and storm surge origin can be excluded. During the past decade of our work along the Andaman and Nicobar Islands and the east coast of India, we have observed that the 2004 tsunami deposits are best preserved in lagoons, inland streams and also on elevated terraces. Chronological evidence for older events obtained from such sites is better correlated with those from Thailand, Sri Lanka and Indonesia, reiterating their usefulness in tsunami geology studies.     

Effects of rainy season on mobilization of contaminants from tsunami deposits left in a coastal zone of Thailand by the 26 December 2004 tsunami

Study on contamination of tsunami sediments deposited on 26 December 2004 conducted shortly after the tsunami in coastal zone of Thailand revealed elevated contents of salts in water-soluble and some heavy metals and arsenic in bioavailable fractions (Szczuciński et al. in Env Geol 49:321–331, 2005). Few months later rainy season started and effected in total rainfall of over 3,300 mm. This paper presents results of survey repeated 1 year after the tsunami. To assess the effects of rainy season on mobilization of previously determined potential contaminants, the same locations were sampled again and analysed with the same methods. The tsunami deposit layer was well preserved but in many locations the sediments were coarser than just after tsunami due to washing out of finer fractions. The water-soluble salts contents were strongly reduced after the rainy season. However, the concentrations of acid leachable heavy metals and metalloids were still elevated in comparison to reference sample from an area not impacted by tsunami. It is possible that the metals and metalloids are successively moved to more bioavailable fraction from forms which were more resistant to mobilization.     

Distribution and sedimentary characteristics of tsunami deposits along the Cascadia margin of western North America

Tsunami deposits have been found at more than 60 sites along the Cascadia margin of Western North America, and here we review and synthesize their distribution and sedimentary characteristics based on the published record. Cascadia tsunami deposits are best preserved, and most easily identified, in low-energy coastal environments such as tidal marshes, back-barrier marshes and coastal lakes where they occur as anomalous layers of sand within peat and mud. They extend up to a kilometer inland in open coastal settings and several kilometers up river valleys. They are distinguished from other sediments by a combination of sedimentary character and stratigraphic context. Recurrence intervals range from 300–1000 years with an average of 500–600 years. The tsunami deposits have been used to help evaluate and mitigate tsunami hazards in Cascadia. They show that the Cascadia subduction zone is prone to great earthquakes that generate large tsunamis. The inclusion of tsunami deposits on inundation maps, used in conjunction with results from inundation models, allows a more accurate assessment of areas subject to tsunami inundation. The application of sediment transport models can help estimate tsunami flow velocity and wave height, parameters which are necessary to help establish evacuation routes and plan development in tsunami prone areas.     

Hydraulic and numerical study on the generation of a subaqueous landslide-induced tsunami along the coast

By carrying out the hydraulic experiments in a one-dimensional open channel and two-dimensional basin, we clarified the process of how a landslide on a uniform slope causes the generation of a tsunami. The effect of the interactive force that occurs between the debris flow layer and the tsunami is significant in the generation of a tsunami. The continuous flow of the debris into the water makes the wave period of the tsunami short. The present experiments apply numerical simulation using the two-layer model with shear stress models on the bottom and interface, and the results are compared. The simulated debris flow shows good agreement with the measured results and ensures the rushing process into the water. We propose that the model use a Manning coefficient of 0.01 for the smooth slope and 0.015 for the rough slope, and a horizontal viscosity of 0.01 m²/s for the landslide; an interactive force of 0.2 for each layer is recommended. The dispersion effect should be included in the numerical model for the propagation from the shore.     

Gravelly shoreface deposits: a comparison of modern and ancient facies sequences

The morphology and dynamics of modern gravel shorefaces are poorly documented. This hinders the interpretation of possible ancient counterparts. A comparative study of a modern (Chesil Beach, England) and an ancient (Baytree Member of the Cardium Formation, Alberta) gravel shoreface shows that the two systems are very similar close to and above sea-level, with a high (about 1 m) gravel plunge step lying below plane-bedded sands and gravels of the beachface. The shoreface at Chesil Beach is dominated by asymmetrical gravel wave ripples. These are oriented offshore near the toe of the shoreface, and onshore in shallower depths. This may reflect offshore movement during storms and landward reworking during fair weather. The Baytree Member is over 12 m thick and comprises over 80% conglomerate. Conglomerate is decimetre-bedded, massive or cross-bedded, with sets over 60 cm thick produced by gravel bedforms migrating alongshore. It is interbedded with discontinuous cm- to dm-bedded sandstones which may be cross-bedded. Pebble fabric and cross-bed orientation both indicate strong alongshore sediment transport. Near the base of the section, pebble orientations suggest that gravel wave-ripples developed below the zone of strong longshore flows. Differences between these two examples may be attributed to different directions of wave approach.     

Historic records of teletsunami in the Indian Ocean and insights from numerical modelling

Following the catastrophic “Great Sumatra–Andaman” earthquake- tsunami in the Indian Ocean on the 26th December 2004, questions have been asked about the frequency and magnitude of tsunami within the region. We present a summary of the previously published lists of Indian Ocean Tsunami (IOT) and the results of a preliminary search of archival materials held at the India Records Office, at the British Library in London. We demonstrate that in some cases, normal tidal movements and floods associated with tropical cyclones have been erroneously listed as tsunami. We summarise archival material for tsunami that occurred in 1945, 1941, 1881, 1819, 1762 and a little known tsunami in 1843. We present the results of modelling of the 2004, 1861 and 1833 tsunami generated by earthquakes off Sumatra and the 1945 Makran earthquake and tsunami, and examine how these results help to explain some of the historical observations. The highly directional component to tsunami propagation illustrated by the numerical models may explain why we are unable to locate archival records of the 1861 and 1833 tsunami at important locations like Rangoon, Kolkata (formally Calcutta) and Chennai (formally Madras), despite reports that these events created large tsunami that inundated western Sumatra. The numerical models identify other areas (particularly the central and southern Indian Ocean islands) where the 1833 tsunami may have had a large enough effect to produce a historic record. We recommend further archival research, coastal geological investigations of tsunami impacts and detailed modelling of tsunami propagation to better understand the record and effects of tsunami in the Indian Ocean and to estimate their likelihood of occurring in the future.     

The 2004 Sumatra earthquake and Indian Ocean tsunami: What happened and why?

The December 26, 2004 Sumatra earthquake and the tsunami that followed killed over 300,000 people. In this paper, we analyze and discuss the geologic causes for this earthquake, the mechanisms that generated it, and follow up with a discussion on ways to prevent this type of disaster in the future.     

Sedimentology of tsunami inflow and backflow deposits: key differences revealed in a modern example

Onshore tsunami deposits may consist of inflow and backflow deposits. Grain sizes can range from clay to boulders of several metres in diameter. Grain‐size distributions reflect the mode of deposition and may be used to explore the hydrodynamic conditions of transport. The absence of unique sedimentary features identifying tsunami deposits makes it difficult in some cases to distinguish inflow from backflow deposits. On Isla Mocha off central Chile, the 27 February 2010 tsunami left behind inflow and backflow deposits of highly variable character. Tsunami inflow entrained sands, gravels and boulders in the upper shoreface, beach, and along coastal terraces. Boulders of up to 12 t were transported up to 300 m inland and 13 m above sea‐level. Thin veneers of coarse sand were found up to the maximum runup at 600 m inland and 19 m above sea‐level. Backflow re‐mobilized most of the sands and gravels deposited during inflow. The orientation of erosional structures indicates that significant volumes of sediment were entrained also during backflow. A major feature of the backflow deposits are widespread prograding fans of coarse sediment developed downcurrent of terrace steps. Fan sediments are mostly structureless but include cross‐bedding, imbrication and ripples, indicating deposition from bedload traction currents. The sediments are poorly sorted, grain sizes range between medium to coarse sand to gravel and pebbles. An assessment of the backflow transport conditions of this mixed material suggests that bedload transport at Rouse numbers >2·5 was achieved by supercritical flows, whereas deposition occurred when currents had decelerated sufficiently on the low‐gradient lower coastal plain. The sedimentary record of the February 2010 tsunami at Isla Mocha consists of backflow deposits to more than 90%. Due to the lack of sedimentary structures, many previous studies of modern tsunami sediments found that most of the detritus was deposited during inflow. This study demonstrates that an uncritical use of this assumption may lead to erroneous interpretations of palaeotsunami magnitudes and sedimentary processes if unknowingly applied to backflow deposits.     

Effect of a submerged bay-mouth breakwater on tsunami behavior analyzed by 2D/3D hybrid model simulation

The hybrid numerical model had been developed to simulate a complicated 3D flow around structures generated by tsunami. In the model, the conventional 2D model is adopted for the wide region far from structures and the 3D non-hydrostatic pressure model is used in the limited region adjacent to structures. The applicability of the model is shown by comparisons of the numerical results with the experimental results of the laboratory model tests and the numerical analysis results of the conventional whole 2D simulation. In addition, the effect of a submerged structure at the opening of a breakwater is discussed from the numerical simulations by the hybrid model. The submerged structure improves the stability of the rubble mound and reduces the tsunami inflow into the bay, while it increases the water surface velocity around the opening of the breakwater. The increase of surface velocity causes the increases of impulsive forces by collision with drafts and so on.     

Reconstructing the pattern and depth of flow onshore in a palaeotsunami from associated deposits

The widespread sheets of fine particulate sediment frequently deposited by tsunami constitute valuable evidence from which to reconstruct tsunami inundation. This is illustrated with evidence from three sites near Montrose, in eastern Scotland, U.K., where a horizon of mainly sand, laid down during the Holocene Storegga Slide palaeotsunami of circa 8000 BP is examined. The horizon is remarkably consistent in its distribution, morphology, stratigraphy, and particle size characteristics. These properties allow inferences to be made on the nature of tsunami flow onshore and run-up. It is suggested that estimates can be made of the possible depth of water involved from the characteristics of the sediment, and thus of the extent of inundation involved in the tsunami at these sites.     

Evaluation of tsunami impacts on shallow marine sediments: An example from the tsunami caused by the 2003 Tokachi-oki earthquake, northern Japan

A combined approach of field geology and numerical simulation was conducted for evaluating the tsunami impacts on the shelf sediments. The 2003 Tokachi-oki earthquake, M 8.0, that occurred on 25 September 2003 off southeastern Hokkaido, northern Japan, generated a locally destructive tsunami. Maximum run-up height of the tsunami waves reached 4 m above sea level. In order to estimate the tsunami impacts on shallow marine sediments, we compared pre- and post-tsunami marine sediments in water depths of 38–112 m in terms of grain size, sedimentary structure, and microfossil content. Decreases of fine fractions, especially finer than very fine sand, which led to coarsen the mean grain size, were detected in the inner shelf of the northern part of the study area. Foraminiferal assemblages also changed in the coarsened sediments. On the other hand, the other shelf sediments largely unchanged or slightly fined. We also simulated the tsunami wave velocity and direction, and grain size entrained by the modeled tsunami. The numerical simulation resulted in that the 2003 tsunami could transport very fine sand in water depths shallower than 45–95 m at the northern part of the study area. This is comparable with the actual grain-size changes after the tsunami had passed. However, some storms and tidal currents might also be possible to stir the surface sediments after the pre-tsunami survey, so we could not conclude that the grain-size changes had been caused only by the tsunami. Nevertheless, a combined approach of sampling and modeling was powerful for estimating the tsunami impacts under the sea.     

Sedimentology of the December 26, 2004, Sumatra tsunami deposits in eastern India (Tamil Nadu) and Kenya

The December 26, 2004 Sumatra tsunami caused severe damage at the coasts of the Indian ocean. We report results of a sedimentological study of tsunami run-up parameters and the sediments laid down by the tsunami at the coast of Tamil Nadu, India, and between Malindi and Lamu, Kenya. In India, evidence of three tsunami waves is preserved on the beaches in the form of characteristic debris accumulations. We measured the maximum run-up distance at 580 m and the maximum run-up height at 4.85 m. Flow depth over land was at least 3.5 m. The tsunami deposited an up to 30 cm thick blanket of moderately well to well-sorted coarse and medium sand that overlies older beach deposits or soil with an erosional unconformity. The sand sheet thins inland without a decrease of grain-size. The deposits consist frequently of three layers. The lower one may be cross-bedded with foresets dipping landward and indicating deposition during run-up. The overlying two sand layers are graded or parallel-laminated without indicators of current directions. Thus, it remains undecided whether they formed during run-up or return flow. Thin dark laminae rich in heavy minerals frequently mark the contacts between successive layers. Benthic foraminifera indicate an entrainment of sediment by the tsunami from water depths less than ca. 30 m water depth. On the Indian shelf these depths are present at distances of up to 5 km from the coast. In Kenya only one wave is recorded, which attained a run-up height of 3 m at a run-up distance of ca. 35 m from the tidal water line at the time of the tsunami impact. Only one layer of fine sand was deposited by the tsunami. It consists predominantly of heavy minerals supplied to the sea by a nearby river. The sand layer thins landward with a minor decrease in grain-size. Benthic foraminifera indicate an entrainment of sediment by the tsunami from water depths less than ca. 30 m water depth, reaching down potentially to ca. 80 m. The presence of only one tsunami-related sediment layer in Kenya, but three in India, reflects the impact of only one wave at the coast of Kenya, as opposed to several in India. Grain-size distributions in the Indian and Kenyan deposits are mostly normal to slightly positively skewed and indicate that the detritus was entrained by the tsunami from well sorted pre-tsunami deposits in nearshore, swash zone and beach environments.     

Recognition of ancient carbonate wind deposits: lessons from a modern analogue, Chrissi Island, Crete

Carbonate aeolian deposits are common along arid to semiarid, wind-exposed, present-day coastlines bordered by productive carbonate ramps. Lithified carbonate dunes (aeolianites) have been described around the world in marine terraces of Quaternary age, but these deposits have seldom been identified in the Pre-Quaternary record. Several authors have suggested that this scarcity reflects that these deposits form and are preserved only during icehouse periods characterized by high-amplitude sea-level changes. Others [e.g. McKee and Ward Carbonate Depositional Environments (1983) , AAPG Memoirs, Vol. 33, pp. 131–170] suggest that the scarcity of aeolianites in the Pre-Quaternary record could reflect the ‘great difficulty in recognising wind blown carbonate deposits and in differentiating between them [aeolianites] and other carbonate sands of nearshore environments’. It has been considered that carbonate shoreface/foreshore deposits are very difficult to discriminate petrographically from backshore deposits. This petrographic study of recent sediments from the shoreface to backshore along the northern coast of Chrissi Island, Crete, confirms that carbonate aeolian sands can be very easily misinterpreted as shoreface deposits. Textural examination of thin sections by image analysis techniques indicates, however, that grain orientation patterns differ between facies. Shoreface deposits exhibit a unimodal distribution of grain orientation (flat rose diagram), whereas backshore deposits show a tendency towards a bimodal distribution with a significant proportion of vertical grains. This observation has been confirmed in Pleistocene aeolianites from Tunisia and Western Australia. Grain verticality thus seems to be a reliable criterion for discriminating wind-lain carbonate bodies from shoreface deposits. Vertical grains in aeolian carbonate deposits could reflect gravity effects (e.g. reorientation of grains because of meteoric water percolation and air pull-up). Laboratory experiments conducted on carbonate sands under the action of percolating waters confirm this hypothesis. This reorganization process is preferentially developed in recently deposited and loosely packed sands resulting from grainfall and/or grainflow. In addition, this suggests that the presence of vertical grain orientation might be an indicator of the frequency and intensity of rainfalls during deposition.     

X‐ray tomography of tsunami deposits: Towards a new depositional model of tsunami deposits

X‐ray computed microtomography is used to obtain high resolution imagery of a historical tsunami deposit in Andalusia, Spain (1755 Lisbon tsunami). The technique allows characterization of grain‐size distribution, structures, component analysis and sedimentary fabric of fine‐grained unconsolidated tsunami deposits at resolutions down to particle scale. The results are validated by comparing to data obtained using other techniques such as laser diffraction, anisotropy of magnetic susceptibility and X‐ray microfluorescence on the same deposits. Specific technical details such as sampling, scanning and image processing methods, and further improvements are addressed. The use of X‐ray computed microtomography provides new insights into the stratigraphy of the deposits and gives access to significantly more detailed view of key sedimentary features such as mudlines, rip‐up clasts, crude laminations, convolutions, floating outsized clasts and contacts between successive units. This analysis of the 1755 tsunami deposits using X‐ray computed microtomography allows the proposal of new hypotheses for the sedimentary processes forming tsunami deposits. Deposition by settling is limited and the section analysed here is dominated by a high shear stress leading to the development of traction carpets, with laminated mudlines corresponding to the basal frictional region of these carpets. The onset of the tsunami backwash is marked by a micro‐vortex resembling Kelvin–Helmoltz instabilities.     

Misattributed tsunami: Chile, Sumatra and the subduction model

Tsunami intensity is poorly correlated with earthquake magnitude. The distribution of aftershocks that immediately followed the 2010 Maule (Chile), the 2004 Sumatra–Andaman and the 2005 Nias (Indonesia) events supports the view that faulting within an accretionary wedge or an outer rise can sometimes disrupt the seafloor more effectively than a megathrust even if the associated seismicity is minor. Monitoring offshore faults would thus seem an effective way to supplement modes of tsunami early warning which hinge on instrumental earthquake detection or wave height and period.