Recognition of diagnostic criteria for recent- and paleo-tsunami sediments from Sri Lanka
Introduction
The undersea tsunamigenic Sumatra–Andaman earthquake occurred on the morning (00:58:53 universal time (U.T)) of 26 December 2004 off the west coast of northern Sumatra registering a magnitude of 9.3 (Kruger and Ohrnberger, 2005, Ishii et al., 2005, Subarya et al., 2006). The 2004 tsunami caused destruction and human casualties rising to about 300,000 in many coastal countries of the Indian Ocean with over 35,000 deaths reported in Sri Lanka.
Tsunami waters started arriving in Sri Lankan coastal regions at different times about 2 h after the earthquake event. The initial water movement was characterized by a rapid drawdown or lowering of the sea surface at the coast as the water moved into the area of seabed displacement. In different parts of Sri Lanka, the drawdown was equivalent to hundreds of meters horizontally. According to eye witness accounts at the village of Peraliya, there were three inundations. The first wave to arrive was 1 m high. This was followed 20 min later by a wave that was 3–8 m high traveling inland at speeds of 30–40 km per hour (Tanioka et al., 2004, Liu et al., 2005). The second wave was characterized as being dark coloured with suspended fine grained sediment. The villagers attributed an open sea origin to this sediment. The third wave was smaller with heights varying from 5 to 6 m. The tsunami waves entered the land via embayments such as harbours, bays, estuaries, lagoons and along rivers (Fig. 1). Normal tidal range in these areas is < 0.8 m, but storm surges can reach elevations of 3 to 5 m.
The 2004 Indian Ocean tsunami inundated southern and eastern coasts of Sri Lanka and encroached inland to locations extending to several kilometers (Fig. 1). Tsunami waters receded from inland areas several hours after inundation at different times depending on the local topography. The tsunami event has left significant geological signatures with changes in coastal geomorphology and deposition of sediments correlative to landward and seaward flows along the coast. Continuous or discontinuous layers of sediments were deposited in the coastal plain. They consisted of sequences of basal coarse layers with uppermost fine fractions. These sediments formed relatively thicker sequences in shallow coastal depressions which can be circular or elongated with diameters ranging from 1 to 10 m and depths generally less than 50 cm. Generally, the lowermost part of a sequence may consist of disorganized coarse grained beds including rock fragments, boulders and even vehicle debris, various artifacts and missiles (see also Smoot et al., 2000, Takashimizu and Masuda, 2000). The uppermost part can be considered as the deposits of calm sedimentation conditions with fine grained sediments characterized by fining up character. The entire thickness of a tsunami-deposited sequence, including the fining upward cap, is comprised of a heterogeneous mixture of marine sediments (such as mollusks, gastropods, foraminifera, radiolarians etc.) and materials locally derived on shore (such as pieces of buildings and vehicle debris etc.).
The mixture of sediments is due to landward and seaward flow that is characteristic of tsunami waves (Fujiwara et al., 2000). As the tsunami waves recede, backflow of the flood waters moves sediment to produce features indicating a reversal of flow direction from the initial flood deposits. Microfossil assemblages (ostracods, diatoms, foraminifera, radiolarians and pollen) provide evidence of sediments that were transported and deposited by tsunamis (see also Hickman et al., 2001, Prendergast, 2006). The foraminifera assemblages can provide information on the pathway of a tsunami by indicating the nature of sediment eroded offshore and along the coast (Uchida et al., 2005). These assemblages can differentiate pre-tsunami sediment from tsunami sediment (Hawkes et al., 2007).
Following the 2004 tsunami event, the sediments have steadily been altered by normal depositional and erosional processes in both the marine and terrestrial environments. Distinctive sedimentary structures have been obliterated by weathering and biological reworking. However, there are a few local conditions that allow the nearly pristine preservation of the tsunami deposits. Small water filled depressions allow preservation of the suspended tsunami sediment as thin distinctive layers that are protected from destructive overprints and buried beneath normal sediments.
Detailed sedimentological research studies on Sri Lankan tsunami deposits are rare. This paper describes granulometric and biological characteristics of 2004 tsunami sediments from Sri Lanka. Based on these findings, an effort was made to correlate probable past tsunami (paleo-tsunami) deposits located in shallow depressions close to the present coastal region of southern Sri Lanka. Historical texts of Sri Lanka refer to at least two past tsunami events that had occurred between 2000 to 3000 B.C (Geiger, 1934, Suraweera, 2000, Stoddart, 2005, Dahanayake, 2006).
Section snippets
Sampling sites
Tsunami deposits are thin or absent in the area near the shore line, whereas they tend to progressively thicken landward, particularly in depressions and coastal stream channels.
Our initial field observations carried out about a few days after the tsunami event showed the extremely heterogeneous and complex character of the tsunami sediments. In the context of our observations, questions arose as to the nature of the diagnostic tsunami sediments and where they could be located.
During our field
Analytical techniques
Sediment samples collected from sites described above as well those from storm surge and near shore deposits were air dried and grain size analysis was done using 1 mm, 0.5 mm, 0.25 mm, 0.212 mm, 0.125 mm, and 0.063 mm sieves. Cumulative curves (as in Tickell, 1965) were constructed for typical samples from each sampling site from the 2004 tsunami (Figs. 3, 2A) and for the upper and lower brownish layer from each augur hole sample site (Fig. 3, 2B, C). All of the 2004 tsunami samples were
Results
Both 2004 and probable buried tsunami sediment types studied were brownish yellow in colour and of fine sand size, generally showing a fining upward trend in the field. The 2004 tsunami sediments formed layers of thickness varying from 5 to 12 cm. Buried tsunami layers had thicknesses varying from about 5 cm (P1) and about 12 cm in P2. Both sediment types in their entirety showed the characteristic brownish yellow colour, which reflects the high microfossil contents of about 50% of the total
Discussion
Our findings based on granulometric, mineralogical and micro-paleontological studies as well as reflection and SEM microscopy observations on sediment samples derived from 2004 tsunami, but are strikingly different from recent storm-surge deposits and samples of near shore sediments. The characteristic dark colour, relatively high heavy mineral content and the scarce presence of open marine microfossils help to diagnose the storm-surge sediments. The near shore sediments are characterized by
Conclusions
Our study has identified diagnostic criteria for distinguishing tsunami sediments from storm-surge sediments in southern Sri Lanka. In the tsunami sediments, reworked marine microfauna are abundant, quartz sand is not well rounded, and heavy minerals are rare, compared with storm-surge sediments. Tsunami sediments are less well sorted than storm-surge and nearshore sediments. These criteria allowed us to successfully extend our findings from 2004 tsunami sediments to locate paleo-tsunami
Acknowledgements
We thank our colleagues Dr H.A. Dharmagunawardena, Dr A. Pitawala, Dr Sudharma Yatigammana and Ms Kumudu Kumarihamy for their assistance in various ways. Ms Sepa Nanayakkara of the Industrial Technology Institute (ITI), Colombo patiently assisted us during several SEM sessions. This work was funded by the National Science Foundation (NSF), Sri Lanka by way of a generous research grant (RG/2005/DMM/05) to one of us (K.D).
References (23)
- et al.
Tsunami deposits in Holocene bay mud in southern Kanto region, Pacific Coast of central Japan
Sediment. Geol.
(2000) - et al.
Sediments deposited by the 2004 Indian Ocean Tsunami along the Malaysia–Thailand Peninsula
Mar. Geol.
(2007) - et al.
Sedimentary differences between the 1993 Hokkaido-nansei-oki tsunami and the 1959 Miyakojima typhoon at Taisei, southwestern Hokkaido, northern Japan
Sediment. Geol.
(2000) - et al.
Sedimentary record of the 1872 earthquake and “tsunami” at Owens Lake, southeast California
Sediment. Geol.
(2000) - et al.
Depositional facies and sedimentary successions of earthquake-induced tsunami deposits in Upper Pleistocene incised valley fills, central Japan
Sediment. Geol.
(2000) Microfossils
(1980)Science at the Solstice: a day in the life of a scientific planet, Abstract on Tsunami sediments
Nature
(2006)The Mahawamsa (The Great Chronicle of Ceylon–Sri Lanka)
(1934)- et al.
Integrated Principles of Zoology
(2001) - et al.
Extent, duration and speed of the 2004 Sumatra–Andaman earthquake imaged by the Hi-Net array
Nature
(2005)
Tracking the rupture of the Mw = 9.3 Sumatra earthquake over 1150 km at teleseismic distance
Nature
Cited by (46)
Characterization of storm surge deposits along the shore of the Gulf of Gdańsk (Baltic Sea) applying heavy mineral analysis
2022, Quaternary InternationalCitation Excerpt :In our study average heavy mineral content was also site depended, however despite the sampling site, storm deposits showed higher heavy mineral concentrations than sediments of adjacent area (Table 2). Dahanayake and Kulasena (2008) successfully used heavy minerals to distinguish deposits of storm and tsunami origin and noted significant differences in their HM concentrations. In this study, deposits interpreted to be of storm origin were richer in HM fraction than sediments from adjacent sedimentary environments or deposits of ambiguous origin (Table 2).
Onshore archives of tsunami deposits
2020, Geological Records of Tsunamis and Other Extreme WavesQuartz grains reveal sedimentary palaeoenvironment and past storm events: A case study from eastern Baltic
2018, Estuarine, Coastal and Shelf ScienceCitation Excerpt :Limited active wave movement is also probably a cause of not fresh percussion marks. Storm quartz grains are, in contrast, characterized by more percussion marks and fresh surfaces compared to potential source material (Bruzzi and Prone, 2000; Costa et al., 2017), suggesting derivation from offshore areas not subjected to beach rounding (Dahanayake and Kulasena, 2008). Therefore, cracked grains with numerous fresh big-sized conchoidal features (Fig. 4B), for example as observed in the sediment deposited at 5900 cal BP, and fresh grains (NU; Fig. 4A) visible in a few sediment horizons, suggest a storm record.
Holocene hydrological events and human induced environmental changes reflected in a southeastern Mediterranean fluvial archive
2017, Palaeogeography, Palaeoclimatology, Palaeoecology