Holocene sea levels along the Andaman Sea coast of Thailand

Abstract

For the Malay-Thai Peninsula several sea-level curves for the younger Holocene, based on field evidence as well as on hydro-isostatic modelling of a far-field site, have been published. The general assumption is a rapid rise to a mid-Holocene maximum up to +5 m above present sea level, followed by a constant or oscillating regression. However, from the Andaman Sea coast of Thailand, which was affected by the 2004 tsunami, only isolated observations are available regarding Holocene sea levels. Thus, the timing and magnitude of the Holocene highstand as well as the course of the regression remain to be defined. As several palaeotsunamis could be detected in the meantime it is important to know the related sea levels as exactly as possible to judge the energy, inundation width and potential wave height of these events. Therefore, fixed biological indicators from the rocky coasts of the Phang-nga Bay and Phuket, as well as morphological indicators from beach-ridge and swale sequences along the exposed west coast (Ko Phra Thong) were studied, to gain information about the Holocene sea-level development in this region. While oyster and coral data from the Phang-nga Bay and Phuket document a Holocene maximum of +2.6 m at 5700 cal. BP, the ridge crests and swale bases in the northwest of the study area point to maximum heights of +1.5–2.0 m above the present level around 5300 years ago. During the last 3000 years, to when the largest part of the Holocene palaeotsunami deposits from Thailand was dated, relative sea levels (RSL) in both areas did not exceed +1.5 m.

Keywords

beach ridges biological indicators Holocene mid-Holocene highstand relative sea level Thailand

Introduction

In the aftermath of the Indian Ocean tsunami (IOT) of December 2004, numerous studies on the Holocene tsunami history of the affected areas have been undertaken (e.g. Brill et al., 2011; Dahanayake and Kulasena, 2008; Jankaew et al., 2008; Monecke et al., 2008; Rajendran et al., 2006). Along the west coast of southern Thailand sedimentary evidence of potential predecessors of the IOT in the form of sandy deposits and boulders has been reported from beach-ridge sequences on Ko Phra Thong (Fujino et al., 2009; Jankaew et al., 2008), coastal plains at Ban Bang Sak (Brill et al., 2011), Pakarang (Neubauer et al., 2011) and Thap Lamu (Rhodes et al., 2011; Yawsangratt et al., 2009) and caves in the Phang-nga Bay (Harper, 2005). While the oldest of these events were dated to more than 5000 years at Pakarang Cape (Neubauer et al., 2011), most evidence in Thailand documents palaeotsunamis (at least four different events) that occurred during the last 3000 years.

To distinguish deposits (sand sheets and/or boulders) of tsunamis from those of other extreme wave events, such as storms, that leave similar sedimentary evidence (e.g. Morton et al., 2007; Switzer and Jones, 2008; Tuttle et al., 2004) and to estimate the magnitudes of the associated events, it is important to know the water level at the time of impact. While in micro-tidal areas tsunami inundation depends nearly exclusively on the energy of the event, the topography and the relative sea level (RSL), in meso- to macrotidal areas the tide level during a tsunami is important too. Since RSL and the tidal range may have differed significantly from modern values during the last millennia, both RSL and tidal situation may make a difference of many metres in the level of wave or flow impact. While there is no possibility to calculate the tide level at the time of impact and potential variations of the tidal range, the positions of former sea levels can be reconstructed from morphological, biological and archaeological indicators, in the best case with a precision of ±5 cm and ±20 years (Milne et al., 2009; Yu et al., 2009). The resulting sea-level curves should give both the error margins regarding the altitudinal variability of an index point, as well as the error of the respective dating method. For the microtidal coast of southwest Thailand, a sea-level curve may help to identify the potential energy of palaeotsunamis by providing information about the maximum altitude above RSL reached by their deposits.

Sea-level curves constructed on the basis of field indicators reflect the local development of RSL, which is controlled by a variety of interacting processes (e.g. Gehrels, 1999; Pirazzoli, 1991). Generally, for continental coastlines in the lower latitudes (far-field sites of Clark et al., 1978), Holocene sea-level changes are mainly determined by ocean refill due to melting ice sheets, hydro-isostatic lowering of the shelf margins and geoidal eustasy (ocean siphoning) (Mitrovica and Milne, 2002; Mitrovica and Peltier, 1991; Peltier, 1999, 2002; Woodroffe and Horton, 2005). However, locally the RSL trend may significantly be influenced or even dominated by local phenomena such as tectonic movements, subsidence or auto-compaction of sediments (e.g. in Bangladesh, Islam and Tooley, 1999). Since SW Thailand is inferred to be more or less tectonically stable and other local factors are negligible at our study sites, the interplay between glacial eustasy, hydro isostasy and geoidal eustasy controlled the RSL development during the Holocene (Horton et al., 2005). In the course of postglacial ice melting a rapid rise of RSL took place (Hanebuth et al., 2000), resulting in a high stand several metres above present RSL (glacial eustasy). This maximum was followed by a more or less continuous fall (isostatic rebound and ocean siphoning), whereas spatial variability of these factors was capable of causing regional differences in magnitude and timing of the highstand. However, important for the interpretation of palaeotsunami deposits is solely the height of RSL at the time of impact. Thus, the main aim is to reconstruct the time and magnitude of the Holocene maximum, as well as the course of sea-level fall afterwards, while the determination of controlling processes is not an essential part of this study.

Holocene sea levels along the Malay-Thai Peninsula

Precise sea-level measurements for Thailand in the form of GPS-corrected tide gauge data and satellite altimetry are restricted to the last decades (Trisirisatayawong et al., 2011). For the time prior to measurements, sea levels have to be reconstructed either by field evidence or model approaches (glacial isostatic adjustment models (GIA), e.g. Mitrovica and Vermeersen, 2002). While no evidence of higher RSL than today has been found in the western Indian Ocean (Camoin et al., 1997, 2004), in most parts of the Indo-Pacific region (including the Malay-Thai Peninsula), in agreement with the typical sea-level pattern of far-field locations (Bradley et al., 2008; Mitrovica and Milne, 2002), sea level rose quickly to a mid-Holocene maximum, which was followed by a steady regression to the modern level (Woodroffe, 2009; Woodroffe and Horton, 2005). However, as a function of meltwater input, hydro isostasy and geoidal eustasy, the timing and magnitude of this high stand is spatially variable (Horton et al., 2005; Woodroffe, 2005; Zong, 2004).

For the Malay-Thai Peninsula the existence of a mid-Holocene sea-level maximum has already been postulated on the basis of geological field evidence from the Strait of Malacca (Geyh et al., 1979; Hesp et al., 1998), Malaysia (Tjia, 1996) and Thailand (Sinsakul et al., 1985, 1992). However, timing, magnitude and number of Holocene highstands varied significantly. For the coast of southern Thailand (Gulf of Thailand and Andaman Sea coast), a sea-level maximum of +4 m at 6000 BP, and two relative maxima at 4000 BP (+2.5 m) and 2600 BP (+2 m) were postulated by Tjia (1996) and Sinsakul (1992). However, considering the great altitudinal range of sea-level indicators used in those studies, the data are considered to be inadequate for constructing precise sea-level curves (Horton et al., 2005; Woodroffe and Horton, 2005). Recent sea-level reconstructions based on both reliable field evidence (Choowong et al., 2004; Horton et al., 2005; Scoffin and Le Tissier, 1998) and model results (Horton et al., 2005) predict a single Holocene high stand of +1–5 m, followed by a gradual fall to present RSL for the Malay-Thai Peninsula. Since hydro-isostatic effects resulted in significant spatial differences, each RSL curve is valid only for a small geographical area. So far, a detailed sea-level reconstruction for the study area (SW Thailand) is mainly based on GIA models which point to a high stand of +2–3 m at around 6000 years ago (Horton et al., 2005). The only reliable field evidence available for SW Thailand is intertidal reef-flat corals from southeast Phuket that indicate a gradual falling sea level from at least +1 m to the present level during the last 6000 years (Scoffin and Le Tissier, 1998). However, since their 11 index points provide only minimum indictors they are not capable of validating the modelled height and timing of the Holocene maximum.

Regional setting

The Andaman Sea coast of southern Thailand (between 9° and 7°30′N) is part of the Malay-Thai Peninsula (Figure 1a), a drowned and eroded, N–S orientated mountain belt. Tectonically, this area is located on the relatively stable Sunda Plate (Tjia, 1996) with little tectonic activity during the Holocene compared with the seismically active zones to the west (Dheeradilok, 1995), where the Sunda Arc subduction zone is a major source of earthquakes and tsunamis. Since southwest Thailand is located in the sub-humid tropics with heavy rainfall (2500 to 3000 mm/yr) mainly during the summer monsoon and a dry period from December to February, dense vegetation and intensively developed soils (laterites on bedrock, podsols on beach ridges) are characteristic elements of the landscape.

Figure 1.

Geographical position of the study area. (a) Map of the Malay-Thai-Peninsula. (b) Overview of southwest Thailand with the locations of the study sites: (c) Ko Phra Thong, (d) Kamala Beach (western Phuket), (e) Ko Pha Nak and Ko Na Khae (Phang-nga Bay) and (f) Phi Phi Islands.

Normal coastal dynamics are controlled by micro- to mesotidal conditions with dominant semi-diurnal character (Admiralty Tide Tables, 2003). For the tide gauge station at the Phuket Royal Marina a maximum spring tide range of 2.82 m, a minimum neap tide range of 0.40 m and a mean tide range of 1.55 m were calculated from the data of 2010; mean sea level (MSL) is defined at +2.02 m above the lowest astronomical tide (LLW). Annual and seasonal differences in the order of <0.3 m are typical for the region and do not influence significantly the accuracy of sea-level reconstructions. The waves are generally moderate with heights <2 m (Choowong et al., 2008). Especially in the calm areas of the Phang-nga Bay this allows the formation of nearly horizontal bio-constructive benches. More energetic conditions, which have significant influence on coastal forming, are common during the summer monsoon. However, tropical cyclones are not reported from the west coast of Southern Thailand within the last 150 years (Murty and Flather, 2004).

The northern part of the study area, from Ko Phra Thong to the northern spit of Phuket, is a sedimentary coast, characterised by sandy beaches, tidal flats and barrier islands (Sinsakul, 1992). Although exposed to the leeward side of the trade winds and distant from the cyclone tracks (Murty and Flather, 2004; Singh et al., 2000), beach-ridge systems are a characteristic element of this area. Especially Phra Thong Island, located 125 km north of Phuket and separated from the mainland by tidal channels, is dominated by coast-parallel beach ridges and swales (Figure 1b).

In contrast, the southern part of the study area is dominated by rocky coastlines. While the coast of Phuket is formed by sandy beaches alternating with granite headlands, the islands of the Phang-nga Bay in the east of Phuket consist of Permian limestone (Watkinson et al., 2008).The spatial pattern of the islands as well as the form of single islands follow the structurally predetermined N–S direction (Tjia, 1996; Watkinson et al., 2008). Long-lasting tropical karstification has transformed the limestone into karst towers with steep slopes to the surrounding sea. These islands exhibit well developed karst features such as karren, sinkholes, speleothems and caves. In some cases caves connect deep, nearly vertical sinkholes in the interior of the islands with the sea, which are locally called ‘hong’. There, in contrast to the exposed cliffs outside the hongs, wave impact is reduced to a minimum. Because of the input of river sediments from the north, the inner parts of the Phang-nga Bay are characterised by forms of coastal progradation such as deltas and tidal flats, mostly covered by Rhizophora-dominated mangrove forests. Nautical charts show relatively shallow water (5–10 m water depth) in this part of the Phang-nga Bay with greater water depths only along former river channels, whereas at its southern end, around the Phi Phi Islands, deep water conditions (20–30 m water depth) occur immediately at the shoreline. In consequence, the wave impact from the Andaman Sea is stronger, which results in more extended beaches (e.g. the tombolo of Phi Phi Don). Since contemporaneously the input of fresh water and suspension diminishes, coral reefs are a frequent feature in the south of the Phang-nga Bay.

Sea-level indicators

In this study, two different groups of indicators were investigated for evidence of former sea and tide levels: (1) fixed biological indicators, attached to rocky shorelines in exposed and sheltered positions, and (2) morphological indicators of a beach-ridge plain. Other kinds of evidence often used for sea-level reconstruction, such as microfossils and pollen (Engelhart et al., 2007; Gehrels, 1999; Horton et al., 2003, 2005, 2007; Scott and Medioli, 2005; Van de Plassche, 1986), paralic peat (Brückner et al., 2010; Gehrels, 1999) or archaeological indicators (e.g. Devillers et al., 2007; Morhange et al., 2001), are missing in the coastal archives of SW Thailand.

Fixed biological indicators

As Tomanek and Helmuth (2002) stated, the rocky intertidal zone is among the most physically harsh environments on earth. Marine invertebrates and algae living in this habitat are alternatively pounded by waves and exposed to thermal extremes during low tide periods; additionally, they must deal with strong selective pressures due to predation and competition for space. As a result, the steep physical gradient and spatially condensed community has made the rocky intertidal zone an ideal ‘natural laboratory’ to study the distribution of organisms in relation to tidal levels (Baker and Haworth, 2000; Baker et al., 2001; Coates, 2008; Kelletat, 1997; Laborel and Laborel-Deguen, 1994, 1996, 2005; Pirazzoli, 2005; Stephenson and Stephenson, 1949; Tomanek and Helmuth, 2002). Generally, the greater the slope of the shore and the lesser the variation of the rock surface the more clearly delineated the distribution pattern will be. Along the limestone and granite coasts of the Phang-nga Bay and Phuket bio-erosive notches, benches of rock oysters, belts of boring bivalves and boring sea-urchins, as well as coral colonies exist. Since they occur in form of recent as well as dead or inactive formations, they were used to precisely determine past sea levels for southwest Thailand (Figure 2).

Figure 2.

Sea-level indicators used in this study. (a) Biological indicators of rocky coasts. (b) Morphological indicators of the beach-ridge plain on Ko Phra Thong.

There seems to be no general agreement about the relationship between indicators and tide levels. However, reliable altitudes can be obtained by comparison with recent counterparts at the same location (Pirazzoli, 2005). In most cases bio-erosive notches, formed by organisms living in the wave zone, give a good idea of the maximum tidal range (Kelletat, 2005). The recent notches in the study area are not totally exposed during mean low water (MLW), but drowned for almost 1 m; their base is treated as an indicator for MLW (±50 cm). While Laborel and Laborel-Deguen (2005) put boring bivalves (Lithophaga sp.) and sea urchins in the sublittoral below LLW, in the Phang-nga Bay living sea urchins and boring bivalves were found directly below MLW (±20 cm). Less disputable is the indicative range of oysters and corals. The upper limit of the closed belt of rock oysters (Saccostrea cucullata) marks mean high water (MHW) with an indicative range of ±10 cm; coral colonies grow at most to MLW and thus represent minimum indicators of past sea levels (Figure 2a).

Morphological indicators

Sedimentary coastlines offer an independent archive for former RSL. The interpretation of the beach ridges and swales on Ko Phra Thong, regarding their indicative meaning in respect to RSL (Figure 2b), was based on the following arguments:

Beach ridges, as can be seen by their internal structure, are built by waves which reached at least the height of their crest, normally have been even higher (Otvos, 2000; Taylor and Stone, 1996). Single ridges are formed by the swash of waves higher than normal (the berm is the initial stage) and ridge crests may be formed decimetres to metres above spring tide high water (SpTHW) (Otvos, 2000; Scheffers et al., 2011). The base of the ridge generally starts above the intertidal zone around MHW but, depending on the spacing between neighbouring ridges, can be slightly above or below.

If sediment supply is sufficient, ridges will more or less continuously be added at the seaward site, forming a succession of ridges and swales (depression between two successive ridges), whereas the cyclicity of ridge formation may be explained by small-scale sea-level variations, occasional storm impact or sediment supply. For densely spaced beach ridges the intervening swales are rather narrow. In this case, the altitude of the swale bases should be at MHW or slightly above. The base of wider ridges may form as deep as at MLW (Taylor and Stone, 1996).

It is likely that the ridge crests in the last centuries and millennia have lost some of their initial height (e.g. by strong rain, storm overwash or tsunami inundation), whereas the swales have been filled up with the deposits moved by these processes plus the organic material of peaty soils. On Phra Thong Island the erosion of the ridge crests is documented by the lack of an Ah-horizon in the site specific podsols, while the swale bases are mainly covered by several decimetres of peaty soil and tsunamigenic sand sheets.

The indicative height of the ridge crests was considered to be too imprecise to mark a specific sea level, therefore, they were only interpreted in their position relative to the recent berm. Instead, the bases of the rather narrow swales were used as an indicator for MHW ±0.5 m, the accuracy of ±0.5 m being a rather conservative empirical value (Figure 2b).

Methods

To establish a sea-level curve for southwest Thailand, all index points were related to the astronomical tides. For this, principally two different approaches can be used: one possibility is to relate the indicators to the exact tide level during measurements, by taking the astronomical tide level from tide gauge data and correcting them for the exact position (if not identical with the tide gauge station) and the actual air pressure. In the absence of disturbances by waves, accuracy within the variation of tidal markers (e.g. MHW), which may differ from year to year and season to season in the amount of 1–2 decimetres, can be achieved. The second approach is based on relating the indicators to significant markers in the field which precisely correlate with specific tide levels. Since along the rocky coasts of the study area such a marker was always present in the form of the uppermost closed belt of rock oysters (Saccostrea cucullata), which in sheltered positions represents MHW ±10 cm, it was used to determine the height of the biological sea-level indicators. The levelling was carried out with a levelling staff over short, nearly vertical distances, which allowed measurements with an insignificant error margin. At the sedimentary coast of Phra Thong, where a constant sea-level marker was missing, the levelling followed the first approach (the tide level during measurements was calculated from the gauge data of Kuraburi) using a differential GPS (Leica SR 530). To get the height of the swale bases, the thickness of the terrestrial infill (peaty soil and tsunamigenic sand sheets) was measured for most swales (in some cases it could only be estimated) and subtracted from the surface height.

To establish a chronology for the levelled index points, three different dating methods were applied: (1) a speleothem from a bio-erosive notch was dated by means of the U/Th technique at the University of Heidelberg (Germany). (2) The ages of corals, molluscs and organic macro remains were determined by radiocarbon dating at the laboratories of Beta Analytic (Florida) and the University of Georgia (Athens), using radiometric dating or AMS. A marine reservoir effect of ΔR = −2 was adopted from Southon et al. (2002). All data were calibrated to 2σ-sidereal years using the IntCal09 and Marine09 data sets of Reimer et al. (2009). (3) To date littoral sand from beach ridges and swales, the optically stimulated luminescence (OSL) technique was applied, which generally proved to be applicable for the dating of beach ridge plains (e.g. Nielsen et al., 2006; Roberts and Plater, 2006); the preparation and measurement of OSL samples followed the procedure described in Brill et al. (2012) for littoral deposits at the same location (Ko Phra Thong).

The sample processing was carried out in subdued red light. Dried samples were sieved and treated with HCl, H₂O₂ and sodium oxalate to remove carbonate, organic material and clay. To obtain pure quartz, the sediment was separated with heavy liquid and finally etched with concentrated HF. Measurements were performed on a Risø TL/OSL DA 20 with a ⁹⁰Sr/⁹⁰Y beta source. Luminescence signals were detected through a Hoya U340 filter (7.5 mm) after blue LED stimulation, carried out at a temperature of 125°C for 50 s. The determination of equivalent doses followed the SAR protocol of Murray and Wintle (2000, 2003). The effects of thermal pretreatment and the luminescence characteristics were analysed by performing preheat-plateau, thermal-transfer and dose recovery tests. Mean equivalent doses were calculated by means of the central age model (Galbraith et al., 1999). Dose rates were calculated from K, Th and U concentrations (determined by means of NAA at Becquerel Laboratories, Canada) and in situ water contents of the samples, using the ADELE software (Kulig, 2005).

Results

Figure 1 shows the areas where the sea-level index points of this study were taken from. These are the beach-ridge sequences of Phra Thong Island as well as the granite headland Laem Son at the southern end of Kamala Beach (Phuket), the island group around Ko Pha Nak and Ko Hong in the inner part of the Phang-nga Bay and the Phi Phi Islands, southeast of Phuket.

Holocene sea-level evidence at rocky coasts

Laem Son (Kamala Beach)

The west coast of Phuket is characterised by granitic headlands and bays with narrow, sandy beaches. Laem Son is the headland south of Kamala Beach (Figure 1d). At its northern spit (the southern end of Kamala Beach) the solid granite rocks are covered by large corestones, resting on a rough, seaward dipping slope. Between the granite corestones, coral boulders as well as small, dead corals in living position (mostly Porites lutea) are hidden. Today these in situ corals are located well above the highest level of living corals (i.e. MLW). The thin crusts are attached to the surrounding boulders and mostly covered by green algae and bored by worms and tiny crabs (Figure 3c, d). A total of nine samples of dead in situ corals, the lowest at −1.36 m below MHW and the highest at 0.6 m above MHW, were taken (Figure 3a, b) and dated to ages between 530–373 and 6877–6618 cal. BP (Thai 11-14-22, Table 1). Considering a tidal range of 1.60 m, the index points are located between 0.24 and 2.20 m above their maximum living level (MLW) and thus indicate higher sea levels during the last 6800 years with a maximum of at least +2.20 m at 4700–4400 cal. BP.

Figure 3.

Sea-level indicators at Laem Son (Kamala Beach). (a) Schematic overview of the sampling site; dead in situ corals are located between granitic core stones well above the zone of modern coral growth. (b) Photograph of the corestone-covered beach; the zone with dead in situ corals reaches from MLW to 60 cm above MHW. (c) Dead in situ coral in the gap between the granitic corestones; (d) the corals form thin, c. 3 cm thick crusts, which are attached to the corestones.

Table 1.

Radiocarbon (a) and U/Th (b) dating results for the index points from the Phang-nga Bay and Phuket. Radiocarbon ages were measured at Beta Analytic (Florida), U/Th ages were measured at the Institut für Umweltphysik, Heidelberg (Prof. Dr A Mangini); calibrated ages are based on the Marine09 data set of Reimer et al. (2009); a marine reservoir effect of ΔR = −2 was adopted from Southon et al. (2002).

(a) Radiocarbon data
Sample	Location	Altitude (m a MHW)	Ind. Height (m a. RSL)	Lab code	Material	δ¹³C (‰)	Conv. Age (yr BP)	Cal. age (2σ) (yr cal. BP)	Cal. Age (ad/bc, 2σ)
Thai 11-14	Laem Son	−1.36	≥0.24	Beta-298446	coral	−2.3	6280±50	6877–6618	4928–4669 bc
Thai 11-15	Laem Son	−1.18	≥0.42	Beta-298447	coral	−1.5	1280±40	914–726	ad 1036–1224
Thai 11-16	Laem Son	−1.18	≥0.42	Beta-298448	coral	−1.8	1220±40	873–675	ad 1078–1276
Thai 11-17	Laem Son	−1.03	≥0.57	Beta-298449	coral	−2.4	1120±40	765–610	ad 1186–134
Thai 11-18	Laem Son	−1.00	≥0.6	Beta-298450	coral	−2.2	5760±50	6282–6026	4333–4077 bc
Thai 11-19	Laem Son	−0.97	≥0.63	Beta-298451	coral	−1.6	830±50	534–333	ad 1417–1618
Thai 11-20	Laem Son	−0.89	≥0.71	Beta-298452	coral	−0.9	5960±60	6518–6259	4569–4310 bc
Thai 11-8	KPN (H)	0.70	0.7±0.1	Beta-298444	oyster	−0.1	1610±30	1256–1083	ad 695–868
Thai 11-9	KPN (H)	0.85	0.85±0.1	Beta-298426	oyster	−0.6	2730±30	2575–2328	626–379 bc
Thai 11-3	KPN (E)	1.00	1±0.1	Beta-298441	oyster	−0.1	2490±30	2281–2052	332–103 bc
Thai 11-7	KNK	1.50	1.5±0.1	Beta-298443	oyster	−0.1	3790±30	3830–3631	1881–1682 bc
Thai 11-21	Laem Son	0.08	≥1.68	Beta-298453	coral	−2.9	830±40	530–373	ad 1421–1578
Thai 11-11	KPN (H)	1.70	1.7±0.1	Beta-298445	oyster	−0.8	2740±30	2599–2336	650–387 bc
Thai 11-12	KPN (H)	1.90	1.9±0.1	Beta-298427	oyster	−1	3560±30	3544–3364	1595–1415 bc
Thai 11-22	Laem Son	0.60	≥2.2	Beta-298454	coral	−1	4380±50	4711–4390	2762–2441 bc
Thai 11-13	KPN (H)	2.60	2.6±0.1	Beta-298428	oyster	0.4	5350±40	5845–5605	3896–3656 bc
Thai 11-23	KPP MB	1.50	≥3.1	Beta-298429	coral	−8.1	23,190±130	28,000–26,910	26,051–24,961 bc
Thai 11-2	KPN (E)	1.50	dislocated	Beta-298440	oyster	−3.3	> 43,500	49,026–44,848	47,077–42,899 bc
Thai 11-4	KPN (E)	2.40	dislocated	Beta-298442	Marine gastropod	−0.7	30,1240±640	36,759–34,500	34,810–32,551 bc

b) U/Th data
Sample	Location	Altitude (m a. MHW)		Lab code	Material		Uncorrected age (yr)	Corrected age (yr BP, 2σ)	Corrected age (ad/ bc, 2σ)

Thai 11-5	KPN (E)	3.60		HD-5353	stalagmite		21400	19000±1300	17,000±1300 bc

	δU (‰)	²³⁸U (µg/g)		²³²Th (ng/g)	²³⁰Th (pg/g)

	209.6±7.3	0.0557±0.0001		5.43±0.024	0.1963±0.0053

KPN (E): Ko Pha Nak East; KPN (H): Ko Pha Nak Hong; KNK: Ko Na Khae; KPP MB: Ko Phi Phi Maya Beach.

Phang-nga Bay

The area of the Phang-nga Bay is characterised by limestone islands. Their steep coasts show features of bio-construction (oysters) as well as bio-erosion (borings and notches).The most impressive forms are horizontal notches that are the result of bio-erosion (bio-corrosion and bio-abrasion) by littorinids, limpets, chitons and other gastropods (plus micro-organisms) grazing on endolithic cyanophyceae and chlorophyceae. Today only the lowermost notch is actively formed. The highest notch, generally much more incised than the active one, often shows a remarkably flat roof, cave openings and a thick cover of flowstone on its floor, as well as curtains of speleothems with diameters of up to 3 m. As the speleothems are formed in an open notch situation and always above the reach of waves (not dissolved if drowned in sea water, but bored by micro-organisms and bivalves), the sharp lower limits of hanging stalactites may also be used as a sea-level indicator.

On the east coast of Ko Pha Nak (KPN (E) in Figure 1e) up to three different notch levels are developed. The upper one is incised more than 10 m deep at a height of 2.50–4.50 m above MHW. Its floor is covered with a thick flowstone and well developed speleothems partly hide the notch from outside. Borings of bivalves cover its walls and the roof up to 4.50 m above MHW. The middle notch (at 2 m above MHW) is much less incised than both the upper notch and the active one (Figure 4). A stalagmite from the inner part of the highest notch was U/Th dated to 19,000±1300 years (Thai 11-5, Table 1). Dislocated molluscs, incorporated in lithified sediments that cover the basal flowstone in the upper notch (at 2.50 m above MHW) and a vertical crack in the limestone cliff (at 1.50 m above MHW, Figure 4d, f), yielded radiocarbon ages of 36,759–34,500 and 49,026–44,848 cal. BP (Thai 11-2 and -4, Table 1). An in situ oyster found in another crack at +1 m above MHW was dated to 2281–2052 cal. BP (Thai 11-3, Table 1). With a tidal range of 1.60 m, higher sea levels of up to +4.10±0.5 m (upper notch, height of the base above MLW), +6.10±0.2 m (bivalve borings) and +1.0±0.10 m (oysters) are indicated.

Inside a hong of Ko Pha Nak (KPN (H) in Figure 1e) notch levels above the active one are missing. However, in a sheltered position of the hong, dead in situ oysters (Saccostrea cucullata) reached altitudes of up to +3 m above the belt of living organisms (Figure 4g). A total of five samples between 0.85 and 2.6 m above MHW yielded ages between 1256–1083 and 5845–5605 cal. BP (Thai 11-8, -9 and -11, -12, -13, Table 1), indicating a maximum sea-level height of +2.60±0.10 m at 5850–5600 cal. BP.

Along the northern coast of Ko Na Khae (KNK in Figure 1e) a narrow beach is developed in front of the limestone cliff. Here the upper notch reaches from 3.50 to 4.90 m above MHW. While sea-urchin borings cover the cliff up to 1.30 m above MHW, the highest bivalve borings were found at the roof of the upper notch (4.90 m above MHW). An in situ oyster (Saccostrea cucullata) at 1.5 m above MHW, which was hidden in a vertical crack, gave an age of 3830–3631 cal. BP (Thai 11-7, Table 1). Considering a tidal range of 1.60 m, notches indicate higher sea levels of up to +5.10±0.5 m (height of the base above MLW), bivalve borings of +6.50±0.2 m, sea-urchin borings of +2.90±0.20 m and oysters of +1.50±0.1 m.

While elevated notch levels on the Phi Phi Islands were found along several sections of the coast, bio-constructive forms others than of recent organisms were missing. Only at Maya Beach (MB in Figure 1f) a single coral at 1.50 m above MHW (3.10 m above MLW), recovered from a crack in the limestone cliff, could be dated to 28,000–26,910 cal. BP (Thai 11-5, Table 1). Another kind of young sea-level indicator is given by a nearly horizontal rock platform (about 6–7 m wide) developed in sandstone at the west coast of Phi Phi Don. The platform is situated close to MHW; less than 1 m above it, honeycomb weathering and small tafoni were found.

Figure 4.

Sea-level indicators on Ko Pha Nak (Phang-nga Bay). (a–f) Ko Pha Nak East: along the steep limestone coast to the left of the narrow beach a deep bio-erosive notch is developed several metres above the recent notch (a–c); the walls and the floor are covered with flowstones and stalagmites (b, c); Lithophaga borings can be found at its roof (e). To the right of the beach a flowstone-covered crack in the limestone cliff contains a lithified shell deposit with dislocated oysters (d, f). (g) Ko Pha Nak Hong: dead in situ oysters were sampled up to 2.60 m above the living zone of modern oysters.

Holocene sea-level evidence at the sedimentary coast of Ko Phra Thong

Phra Thong Island is dominated by sequences of more or less shore parallel beach ridges and swales, occasionally cut by discordances that document erosion events (e.g. Scheffers et al., 2011). While in the eastern part these structures are predominantly covered by mangroves and dense vegetation, grassy ridges and wet, tree lined swales build up a 2.5–3 km wide open beach-ridge plain in the west of the island (Figures 1c and 5). In the northern part of this beach-ridge plain four shore-perpendicular transects were levelled (transects A–D in Figure 5). The chronology was based on a total of 13 OSL samples, taken from ridges (KPT 12, 24–26, 28, 32, 40 and 43–46; Table 2 and Figure 5) and intertidal sand at the swale bases (KPT 2 and 36; Table 2). Additionally, three organic samples from peaty soils at the swale base were dated by means of radiocarbon (Table 1b). Further chronological information was obtained from OSL results of Prendergast et al. (2012). Figure 5 documents the results of transect A: from the east to the west the ages decline from 5300±500 to 1600±200 years, revealing no age inversions. In the same direction the heights of the ridge crests (from 3.30 to 1.50 m above MHW) and the swale bases (from 2.30 to 1.15 m above MHW) show a decreasing trend. Transects B and C reveal similar tendencies (oldest ages in the east and youngest ages in the west as well as decreasing topography towards the present shoreline), with only a single age inversion in transect C (KPT 32). Because of the high dating uncertainties of the ages for KPT 24, 25 and 26 (all ages overlap within their error margins), no clear trend could be observed in transect D.

Figure 5.

The beach-ridge plain on Ko Phra Thong. (a) In the northern part of the island four transects were levelled and sampled for dating. (b) Topographic cross-section along transect A: the heights of the ridge crests and swale bases are plotted against the distance from the present shoreline. Since the island developed from the east to the west, the ages become relatively older towards the east. In the same direction the heights of ridge crests and swale bases show a slightly increasing trend.

Table 2.

Luminescence (a) and radiocarbon (b) dating results for the index points from Ko Phra Thong. OSL data were carried out by DB and NK in the Marburg Luminescence Lab (*ages for R1–3 and S1 are taken from Prendergast et al., 2012). Radiocarbon samples were measured at the University of Georgia (Athens); calibrated ages are based on the IntCal09 data set of Reimer et al. (2009).

(a) OSL data
Sample	Location	Distance (m)	Altitude (m a. MHW)	Lab code	U (ppm)	Th (ppm)	K (%)	Water (%)	De (Gy)	Age (years)
R1	ridge	180	1.92							1600±200*
R2	ridge	300	2.25							2100±270*
S1	swale	370	−0.45							2500±250*
R3	ridge	450	2.40							2560±350*
KPT 2j	swale	740	0.68	MR-0787	4.1±0.13	17.7±0.40	0.15±0.01	23.10	4.43±0.060	2480±240
KPT 12	ridge	770	3.12	MR-0791	0.8±0.05	2.7±0.18	0.03±0.01	11.90	2.73±0.08	2324±197
KPT 40	ridge	954	2.54	MR-1522	5.9±0.61	26.7±1.34	0.2±0.01	1.17	7.66±0.43	4416±421
KPT 43	ridge	1050	2.75	MR-1524	2.7±0.07	6.1±0.27	0.05±0.01	1.20	2.94±0.07	4252±378
KPT 32	ridge	1170	3.32	MR-1514	3.7±0.16	17.8±0.91	0.4±0.01	6.05	5.75±0.17	4516±454
KPT 36	swale	1300	0.70	MR-1519	2.3±0.106.5±0.56	9.1±0.4816.8±0.84	0.5±0.010.6±0.02	16.70	5.00±0,06	3028±351
KPT 44	ridge	1310	2.99	MR-1526	1.9±0.06	7.6±0.27	0.2±0.01	2.85	3.66±0.07	5025±400
KPT 45	ridge	1500	3.03	MR-1528	1.2±0.064.7±0.14	4.2±0.2310.8±0.73	0.2±0.010.05±0.01	1.65	4.10±0.1	4645±347
KPT 46	ridge	1730	3.36	MR-1531	4.5±0.191.2±0.08	12.4±0.643.22±0.15	0.2±0.010.07±0.01	2.16	4.81±0.19	5310±520
KPT 28	ridge	1800	3.53	MR-1512	2.8±0.13	12.9±0.67	0.3±0.01	1.63	5.55±0.06	5394±535
KPT 25	ridge	4500	3.25	MR-1509	3.4±0.14	18.7±0.96	0.17±0.01	0.23	151.5±7.70	126,301±14,398
KPT 24	ridge	4600	3.47	MR-1506	1.2±0.064.2±0.12	3.8±0.2120.7±0.75	0.14±0.010.01±0.01	2.42	123.6±8.45	104,587±10,565
KPT 26	ridge	4900	3.90	MR-1511	1.5±0.07	6.0±0.32	0.26±0.01	2.43	121.0±4.76	106,950±8318

(b) Radiocarbon data
Sample	Location	Distance (m)	Altitude (m a. MHW)	Lab code	Material	δ¹³C (‰)	Conv. Age (yr BP)	Cal. Age (yr cal. BP, 2σ)		Cal. Age (cal. ad/bc, 2σ)

KPT 35B/1	swale	1300.00	0.70	UGAMS-8050	wood	−28.90	3040±25	3345–3165		1396–1216 bc
KPT 37/2	swale	1300.00	−0.69	UGAMS-8051	plant fragment	−29.80	2760±20	2923–2785		974–836 bc
KPT 23/7	swale	1550.00	−0.20	UGAMS-8045	wood	−29.30	3080±30	3370–3217		1421–1268 bc

Discussion

The chronology of sea-level index points

Galleries of sea-urchin and Lithophaga borings, often found around 4.5–5.5 m above MSL as well as directly above the living oyster belt (from MHW to nearly +0.90 m above MHW), certainly belong to higher sea levels than today. Similarly, bio-erosive notches indicate sea levels up to 3.50–5.50 m above the present one. However, as pure destructive forms they could not be dated directly. Mainly, even relative ages can only be estimated: from the thick flowstones and the very well developed speleothem inside, some of them about 3 m in diameter, pre-Holocene ages are suspected for the upper notch; a long-lasting sea-level highstand of pre-Holocene age is also supported by the deep incision of the upper notch (up to 10 m (Figure 4c, d), deeper caves leading into the islands).The intermediate notch, which is only preserved as remnants at isolated places, seems to be eroded by both the recent and the upper notch and thus is certainly the oldest of the three. Therefore, it must result from a sea-level highstand of at least last-interglacial age (probably even older). Borings of bivalves and sea-urchins located at the roof of the upper notch must be younger than this, but still are estimated to be of pre-Holocene age. In some cases minimum ages could be achieved by dating constructive forms (not necessarily connected to sea level), inside the notches: according to a speleothem age of 20 ka, the upper notch at the eastern side of KPN definitely pre-dates the Holocene.

Bio-constructive indicators (oysters and corals) as well as sedimentary sea-level markers (ridges and swales) could be dated with isotopic methods (radiocarbon) and luminescence dating. In the case of in situ organisms and morphological structures, the dating results reflect the age of the sea level indicated by this index point. If organisms have been dislocated (gastropod in the upper notch and oyster in the crack fill at KPN), they only provide maximum ages for the corresponding sea levels. Considering this, bio-erosive indicators cannot be used to reconstruct a sea-level curve. However, they prove the existence of up to 6 m higher sea levels during the younger Pleistocene. On the basis of ridge-crest data from the eastern part of Phra Thong (KPT 24, 25, 26), the last pre-Holocene highstand can be related to the last interglacial around 125 ka BP, a period of widespread 4–6 m higher eustatic sea levels (e.g. Church et al., 2008; Rohling et al., 2008). Although the corresponding OSL data are too imprecise to define an exact time (covering an age range of 94–140 ka within their error ranges) and the recent altitude of the ridges is a sea-level indicator too poor to give an exact altitude (compare section ‘Sea level indicators’), they must be related to the Eem, since sea level was many metres below its present position during the intervening glacial periods. In contrast, it was possible to combine all data of in situ oysters and corals, as well as ridges and swales to a Holocene sea-level curve (section ‘Holocene sea-level curves’).

Holocene sea-level curves

Our tentative sea-level curves cover only the ages we have found in our samples. Nevertheless, those from the Laem Son coral profile, as well as those from oysters in the Phang-nga Bay show the same trend and maximum altitude lower than +3 m compared with modern RSL (Figure 6a); the maximum of +2.5–3.0 m was timed to c. 5700 cal. BP. Both time and magnitude of the highstand match the GIA model predictions (Horton et al., 2005) quite precisely. From the reconstruction of former sea levels back to about 5300 BP out of the sedimentary archive of the ridge and swale sequences of Phra Thong Island we tentatively deduce a highstand around 5300 BP in the order of +1.5–2.5 m above RSL (Figure 6b). The limited height of the uppermost beach ridges just below +5 m above MSL and their difference in altitude of about 1 m to the original swale bottoms point to a slightly lower highstand in the northwest of the study area. At least three possible explanations can be mentioned for this difference:

there is the possibility of local effects in the form of neo-tectonic movements in the order of 1 m during the younger Holocene. Although the assumption of complete neotectonic quiescence during the Holocene (Tjia, 1996) might be too simplistic (Fenton and Sutiwanich, 2005), the vague data of historical earthquakes and seismic activity in prehistoric times (Department of Mineral Resources, http://www.dmr.go.th/main.php?filename=ActiveFault_Eng) do not provide any quantitative information about amount, timing and spatial distribution of Holocene tectonic movements. Recent GPS measurements support the assumption of tectonic activity during the Holocene, pointing to cycles of long-term slow uplift and shorter periods (several decades) of accelerated subsidence induced by strong earthquakes at the Sunda Fault subduction zone (Meltzner et al., 2006; Trisirisatayawong et al., 2011). However, because of the repeated switchover between uplift and subsidence, which predominantly compensate each other, it is very difficult to quantify the absolute effect on the Holocene sea-level curve. Moreover, there is no hint for significant spatial variations which could explain the differences in the sea-level curves of Phuket and Phra Thong.

The second possible reason is compaction of the ridge-and-swale-sediment sequence. But as it consists of fine to medium well sorted sand, this is a vague argument, not supported by similar observations elsewhere.

The third, probably most reasonable explanation relates the difference to spatially variable hydro-isostatic effects (Horton et al., 2005).

Figure 6.

Sea-level curves for the study area. (a) Sea-level curve based on bio-constructive index points from Kamala Beach and the Phang-nga Bay; the reconstructed sea-level trend is nearly identical with that predicted by GIA models. (b) Sea-level curve for Ko Phra Thong based on ridge crest and swale base data; compared with the area around Phuket and the Phang-nga Bay (a), the Holocene maximum is slightly smaller.

Based on our index points (limited in number, but taken from exposed and sheltered sites and from very different indicators) we conclude that slightly higher sea and tide levels in the order of present spring tide range (about 2.5 m) existed for the Holocene.

Implications for the interpretation of palaeotsunami deposits

In case of palaeotsunamis that are only preserved in the form of sedimentary evidence, maximum inland extent, indicating inundation, and maximum height above RSL, indicating run-up or wave height, of the deposits can be used to estimate event magnitudes. Considering the sea-level fluctuations described in the section ‘Holocene sea-level curves’, the reference level for these indicators, i.e. RSL, was at the most 2.5 m above present level for palaeotsunami evidence from western Thailand, which covers only the last 5000 years (Brill et al., 2011; Jankaew et al., 2008; Rhodes et al., 2011). This may mean that sea levels during low water had been at the levels of modern high water (considering constant spring tide ranges of ~2.5 m), all in all a nearly comparable situation for palaeotsunami landfalls during the last millennia. For example, the well constrained palaeotsunami deposits from Ko Phra Thong (Jankaew et al., 2008) and Ban Bang Sak (Brill et al., 2011) comprise one event at 500–700 cal. BP and 2–3 additional tsunamis younger than 2500 cal. BP. Taking today’s sea level as a basis, the waves of these events had to override ridges up to 5 m (above low water) and inundation reached up to 1300 m inland. Considering the corresponding RSL of +1–1.5 m at 500–2500 years ago, the minimum waves that had the potential to cross the barrier (reduced to 3.5–4 m at low water) could have been slightly lower than at present RSL. However, this difference of at the most 1.5 m is completely compensated by the potential differences in tide levels (diurnal variation between high water and low water) of 1.5–2 m.

Much lower RSL compared with the present level (and therefore with more influence on the spread of tsunami deposits) are characteristic for the early Holocene prior to 7000 cal. BP, a period that lacks contemporaneous tsunami evidence in Thailand. During the last 7000 years, instead, coastline progradation (e.g. of the Phra Thong beach-ridge plain) had much more influence on the distribution of tsunamites than sea-level movements. However, only the combined data of palaeotsunami deposits, palaeogeography and RSL at a given time allow a well-grounded interpretation of event magnitudes.

Conclusions

For the west coast of Southern Thailand, glacial isostatic adjustment models predict a mid-Holocene sea-level highstand with regionally variable altitude. Field evidence in the form of fixed biological indicators from rocky coasts as well as beach ridges and swales from a sedimentary coast, was used to quantify the exact timing and altitude of this sea level maximum: (1) Sea level had reached its maximum of +2.5–3 m 5700 years ago. Following this highstand, sea level gradually dropped to its present level; oscillations have not been observed. (2) Regional differences between Phra Thong Island in the north and Phang-nga Bay in the south, probably as a result of hydro-isostasy, are in the order of at the most 1 m at the time of the Holocene maximum.

These sea level variations effect the interpretation of palaeotsunami deposits regarding the magnitude of associated events. However, RSL during the ages so far elaborated for Thailand’s palaeotsunamis, i.e. the last 5000 years, was in the order of +2.5 m and thereby comparable with the difference between high water and low water. Thus, its influence on deposition by tsunamis is too low to be distinguished from diurnal variations of tide levels.

Footnotes

Acknowledgements

Augusto Mangini, Heidelberg, kindly provided the U/Th age of a stalagmite. We thank Kruawun Jankaew and Thanakorn Jiwarungruangkul for assistance during the 2010 field work on Phra Thong Island and Anne Hager for graphical assistance. Raphaël Paris, Clemand-Ferrand, provided a thoughtful and constructive review. We appreciate the administrative and logistic support of Penjai Sompongchaiyakul, Suratta Bunsomboonsakul (both from Chulalongkorn University, Bangkok) and Klaus Schwarzer (University of Kiel).

Funding

This study was integrated in the Thai-German cooperation TRIAS (Tracing Tsunami Impacts On- and Offshore in the Andaman Sea Region), financially supported by NRCT (National Research Council of Thailand) and DFG (German Research Foundation) (DFG ref. nos.: PAK 228; BR 877/27-2), which is gratefully acknowledged.

References

Admiralty Tide Tables (2003) Indian Ocean and South China Sea, Volume 3. Hydrographer to the Navy, Admiralty Hydrography Department, 357 pp.

Baker

RGV

Haworth

(2000) Smooth or oscillating late Holocene sea-level curve? Evidence from cross-regional statistical regressions of fixed biological indicators. Marine Geology 163: 353–365.

Baker

RGV

Haworth

Flood

(2001) Inter-tidal fixed indicators of former Holocene sea-levels in Australia: A summary of sites and a review of methods and models. Quaternary International 83–85: 257–273.

Bradley

Milne

Zong

. (2008) Modelling sea-level data from China and Malay-Thai Peninsula to infer Holocene eustatic sea-level change. American Geophysical Union, Fall Meeting 2008, abstract #GC33A-0763.

Brill

Brückner

Jankaew

. (2011) Potential predecessors of the 2004 Indian Ocean Tsunami – Sedimentary evidence of extreme wave events at Ban Bang Sak, SW Thailand. Sedimentary Geology 239: 146–161.

Brill

Klasen

Brückner

. (2012) OSL dating of tsunami deposits from Phra Thong Island, Thailand. Quaternary Geochronology in press.

Brückner

Kelterbaum

Marunchak

. (2010) The Holocene sea level story since 7500 BP – Lessons from the Eastern Mediterranean, the Black and the Azov Seas. Quaternary International 225: 160–179.

Camoin

Colonna

Montaggioni

. (1997) Holocene sea level changes and reef development in the southwestern Indian Ocean. Coral Reefs 16: 247–259.

Camoin

Montaggioni

Braithwaite

CJR

(2004) Late glacial to post glacial sea levels in the Western Indian Ocean. Marine Geology 206(1–4): 119–146.

10.

Choowong

Phantuwongraj

Charoentitirat

. (2008) Beach recovery after 2004 Indian Ocean tsunami from Phang-nga, Thailand. Geomorphology 104: 134–142.

11.

Choowong

Ugai

Charoentitirat

. (2004) Holocene biostratigraphical records in coastal deposits from Sam Roi Yod National Park, Prachuap Khiri Khan, Western Thailand. The Natural History Journal of Chulalongkorn University 42: 1–18.

12.

Church

White

Aarup

. (2008) Understanding global sea levels: Past, present and future. Sustainable Science 3: 9–22.

13.

Clark

Farrell

Peltier

(1978) Global changes in post glacial sea level: A numerical calculation. Quaternary Research 9: 265–287.

14.

Coates

(2008) A comparison of tidal assemblages on exposed and sheltered tropical and temperate rocky shores. Global Ecology and Biogeography Letters 7(2): 115–124.

15.

Dahanayake

Kulasena

(2008) Recognition of diagnostic criteria for recent- and paleo-tsunami sediments from Sri Lanka. Marine Geology 254: 180–186.

16.

Devillers

Excoffon

Morhange

. (2007) Relative sea-level changes and coastal evolution at Forum Julii (Fréjus, Provence). Comptes Rendus Geosciences 339(5): 329–336.

17.

Dheeradilok

(1995) Quaternary coastal morphology and deposition in Thailand. Quaternary International 26: 49–54.

18.

Engelhart

Horton

David

. (2007) Mangrove pollen of Indonesia and its suitability as a sea-level indicator. Marine Geology 242: 65–81.

19.

Fenton

Sutiwanich

(2005) Paleoseismic Investigation of the Ranong and Khlong Marui faults, Chumphon Province, Southern Thailand. American Geophysical Union, Fall Meeting 2005.

20.

Fujino

Naruse

Matsumoto

. (2009) Stratigraphic evidence for pre-2004 tsunamis in southwestern Thailand. Marine Geology 262: 25–28.

21.

Galbraith

Roberts

Laslett

. (1999) Optical dating of single grains of quartz from Jinmium rock shelter, northern Australia. Part I: Experimental design and statistical models. Archaeometry 41: 339–364.

22.

Gehrels

(1999) Middle and late Holocene sea-level changes in eastern Maine reconstructed from foraminiferal saltmarsh stratigraphy and AMS ¹⁴C dates on basal peat. Quaternary Research 52: 350–359.

23.

Geyh

Kudrass

H-R

Streif

(1979) Sea-level changes during the late Pleistocene and Holocene in the Straits of Malacca. Nature 278: 441–443.

24.

Hanebuth

Stattegger

Grootes

(2000) Rapid flooding of the Sunda Shelf: A Late-Glacial sea-level record. Science 288: 1033–1035.

25.

Harper

(2005) Bedded shell deposit at Ao Nang, Krabi Province, southern Thailand; a record of a prehistoric tsunami event or extreme storm event or neither [abstract]. Abstracts with Programs – Geological Society of America 37(7): 75.

26.

Hesp

Hung

Hilton

. (1998) A first tentative Holocene Sea level curve for Singapore. Journal of Coastal Research 14: 308–314.

27.

Horton

Gibbard

Milne

. (2005) Holocene sea levels and palaeoenvironments, Malay-Thai Peninsula, southeast Asia. The Holocene 15(8): 1199–1213.

28.

Horton

Larcombe

Woodroffe

. (2003) Contemporary foraminiferal distributions of a mangrove environment, Great Barrier Reef coastline, Australia: Implications for sea-level reconstructions. Marine Geology 198: 225–243.

29.

Horton

Zong

Hillier

. (2007) Diatoms from Indonesian mangroves and their suitability as sea-level indicators for tropical environments. Marine Micropaleontology 63: 155–168.

30.

Islam

Tooley

(1999) Coastal and sea-level changes during the Holocene in Bangladesh. Quaternary International 55: 61–75.

31.

Jankaew

Atwater

Sawai

. (2008) Medieval forewarning of the 2004 Indian Ocean tsunami in Thailand. Nature 455: 1228–1231.

32.

Kelletat

(1997) Mediterranean coastal biogeomorphology: Processes, forms, and sea-level indicators. In: Briand

Maldonado

(eds) Transformations and Evolution of the Mediterranean Coastline. CIESM Science Series no.3 (Commission Internationale pour l’Exploration Scientifique de la mer Méditerranée). Bulletin de L’Institut océanographique, Monaco, numéro spécial 18: 209–226.

33.

Kelletat

(2005) Notches. In: Schwartz

(ed.) Encyclopedia of Coastal Science. Springer, pp. 728–730.

34.

Kulig

(2005) Erstellung einer Auswertesoftware zur Altersbestimmung mittels Lumineszenzverfahren unter spezieller Berücksichtigung des Einflusses radioaktiver Ungleichgewichte in der 238-U-Zerfallsreihe. Bakkalaureusarbeit Network Computing, TU Freiberg.

35.

Laborel

Laborel-Deguen

(1994) Biological indicators of relative sea-level variations and of co-seismic displacements in the Mediterranean region. Journal of Coastal Research 10(2): 395–415.

36.

Laborel

Laborel-Deguen

(1996) Biological indicators of Holocene sea-level and climatic variations on rocky coasts of tropical and subtropical regions. Quaternary International 31: 53–60.

37.

Laborel

Laborel-Deguen

(2005) Sea-level indicators, biologic. In: Schwartz

(ed.) Encyclopedia of Coastal Science. Springer, pp. 833–834.

38.

Meltzner

Sieh

Abrams

. (2006) Uplift and subsidence associated with the great Aceh-Andaman earthquake of 2004. Journal of Geophysical Research 111: doi:10.1029/2005JB003891.

39.

Milne

Gehrels

Hughes

. (2009) Identifying the causes of sea-level change. Nature Geoscience 2: 471–478.

40.

Mitrovica

Milne

(2002) On the origin of late Holocene sea-level highstands within equatorial ocean basins. Quaternary Science Reviews 21: 2179–2190.

41.

Mitrovica

Peltier

(1991) On postglacial geoid subsidence over the equatorial oceans. Journal of Geophysical Research 96: 20,053–20,071.

42.

Mitrovica

Vermeersen

LLA

(2002) Glacial Isostatic Adjustment and the Earth System: Sea Level, Crustal Deformation, Gravity and Rotation. Washington DC: American Geophysical Union.

43.

Monecke

Finger

Klarer

. (2008) A 1,000-year sediment record of tsunami recurrence in northern Sumatra. Nature 455: 1232–1234.

44.

Morhange

Laborel

Hesnard

(2001) Changes of relative sea level during the past 5000 years in the ancient harbor of Marseilles, Southern France. Palaeogeography, Palaeoclimatology, Palaeoecology 166: 319–329.

45.

Morton

Gelfenbaum

Jaffe

(2007) Physical criteria for distinguishing sandy tsunami and storm deposits using modern examples. Sedimentary Geology 200: 184–207.

46.

Murray

Wintle

(2000) Luminescence dating of quartz using an improved single aliquot regenerative-dose protocol. Radiation Measurements 32: 57–73.

47.

Murray

Wintle

(2003) The single aliquot regenerative dose protocol: Potential for improvements in reliability. Radiation Measurements 37: 377–381.

48.

Murty

Flather

(2004) Impact of storm surges in the Bay of Bengal. Journal of Coastal Research 12: 149–161.

49.

Neubauer

N-P

Brill

Brückner

. (2011) 5000 Jahre Tsunami-Geschichte am Kap Pakarang (Thailand). Coastline Reports 17: 75–92.

50.

Nielsen

Murray

Pejrup

. (2006) Optically stimulated luminescence dating of a Holocene beach ridge plain in Northern Jutland, Denmark. Quaternary Geochronology 1: 305–312.

51.

Otvos

(2000) Beach ridges – Definitions and significance. Geomorphology 32: 83–108.

52.

Peltier

(1999) Global sea level rise and glacial isostatic adjustment. Global and Planetary Change 20: 93–123.

53.

Peltier

(2002) On eustatic sea level history: Last Glacial Maximum to Holocene. Quaternary Science Reviews 21: 377–396.

54.

Pirazzoli

(1991) World Atlas of Holocene Sea-level Changes. Elsevier Oceanography Series 58, 300 pp.

55.

Pirazzoli

(2005) Sea-level indicators, geomorphic. In: Schwartz

(ed.) Encyclopedia of Coastal Science. Springer, pp. 836–838.

56.

Prendergast

Cupper

Jankaew

. (2012) Indian Ocean tsunami recurrence from optical dating of tsunami sand sheets in Thailand. Marine Geology 295–298: 20–27.

57.

Rajendran

Machado

. (2006) Evidence of ancient sea surges at the Mamallapuram coast of India and implications for previous Indian Ocean tsunami events. Current Science 91(9): 1242–1247.

58.

Reimer

Baillie

MGL

Bard

. (2009) IntCal09 and Marine09 radiocarbon age calibration curves, 0-50,000 years cal BP. Radiocarbon 51(4): 1111–1150.

59.

Rhodes

Kirby

Jankaew

. (2011) Evidence for a mid-Holocene tsunami deposit along the Andaman coast of Thailand preserved in a mangrove environment. Marine Geology 282(3–4): 255–267.

60.

Roberts

Plater

(2006) Reconstruction of Holocene foreland progradation using optically stimulated luminescence (OSL) dating: An example from Dungeness, UK. The Holocene 17(4): 495–505.

61.

Rohling

Grant

Hemleben

. (2008) High rates of sea-level rise during the last interglacial period. Nature Geocsience 1: 38–42.

62.

Scheffers

Engel

Scheffers

. (2011) Beach ridge systems – Archives for Holocene coastal events? Progress in Physical Geography 36: 5–37.

63.

Scoffin

le Tissier

MDA

(1998) Late Holocene sea level and reef flat progradation, Phuket, South Thailand. Coral Reefs 17: 273–276.

64.

Scott

Medioli

(2005) Sea-level indicators – Biological in depositional sequences. In: Schwartz

(ed.) Encyclopedia of Coastal Science. Springer, pp. 835–835.

65.

Singh

Ali Khan

Rahman

(2000) Changes in the frequency of tropical cyclones over the North Indian Ocean. Meteorology and Atmospheric Physics 75: 11–20.

66.

Sinsakul

(1992) Evidence of quaternary sea level changes in the coastal areas of Thailand: A review. Journal of Southeast Asian Earth Science 7: 23–37.

67.

Sinsakul

Sonsuk

Hasting

(1985) Holocene sea levels in Thailand: Evidence and basis for interpretation. Journal of Geological Society Thailand 8: 1–12.

68.

Southon

Kashgarian

Fontugne

. (2002) Marine reservoir corrections for the Indian Ocean and Southeast Asia. Radiocarbon 44: 167–180.

69.

Stephenson

(1949) The universal features of zonation between tide-marks on rocky coasts. Journal of Ecology 37: 289–305.

70.

Switzer

Jones

(2008) Large-scale overwash sedimentation in a freshwater lagoon from the southeast Australian coast: Sea-level change, tsunami or exceptionally large storm? The Holocene 18(5): 787–803.

71.

Taylor

Stone

(1996) Beach-ridges: A review. Journal of Coastal Research 12: 612–621.

72.

Tjia

(1996) Sea-level changes in the tectonically stable Malay-Thai Peninsula. Quaternary International 31: 95–101.

73.

Tomanek

Helmuth

(2002) Physiological ecology of rocky intertidal organisms: A synergy of concepts. Integrative and Comparative Biology 42: 771–775.

74.

Trisirisatayawong

Naeije

Simons

. (2011) Sea level change in the Gulf of Thailand from GPS-corrected tide gauge data and multi-satellite altimetry. Global and Planetary Change 76(3–4): 137–151.

75.

Tuttle

Ruffman

Anderson

. (2004) Distinguishing tsunami from storm deposits in eastern North America: The 1929 Grand Banks Tsunami versus the 1991 Halloween Storm. Seismological Research Letters 75(1): 117–131.

76.

Van de Plassche

(ed.) (1986) Sea-Level Research: A Manual for the Collection and Evaluation of Data. Norwich: GEO-Books.

77.

Watkinson

Elders

Hall

(2008) The kinematic history of the Khlong Marui and Ranong Faults, southern Thailand. Journal of Structural Geology 30: 1554–1571.

78.

Woodroffe

(2005) Late Quaternary sea-level highstands in the central and eastern Indian Ocean: A review. Global and Planetary Change 49: 121–138.

79.

Woodroffe

(2009) Testing models of mid to late Holocene sea-level change, North Queensland, Australia. Quaternary Science Reviews 28: 2474–2488.

80.

Woodroffe

Horton

(2005) Late and post-glacial sea level changes of the Indo-Pacific: A review. Journal of Asian Earth Science 5(8): 29–43.

81.

Yawsangratt

Szczucinski

Chaimanee

. (2009) Depositional effects of 2004 tsunami and hypothetical paleotsunami near Thap Lamu Navy Base in Phang Nga Province, Thailand. Polish Journal of Environmental Studies 18(1): 17–23.

82.

K-F

Zhao

J-X

Done

. (2009) Microatoll record for large century-scale sea-level fluctuations in the mid-Holocene. Quaternary Research 71: 354–360.

83.

Zong

(2004) Mid-Holocene sea-level highstand along the Southeast Coast of China. Quaternary International 117: 55–67.