Microatolls document the 1762 and prior earthquakes along ...

Contents lists available at ScienceDirect

Tectonophysics

journal homepage: www.elsevier.com/locate/tecto

Microatolls document the 1762 and prior earthquakes along the southeastcoast of Bangladesh

Dhiman R. Mondala,b,⁎,1, Cecilia M. McHugha,b,c, Richard A. Mortlockd, Michael S. Stecklerc,Sharif Mustaquea, Syed Humayun Akhtere

a School of Earth and Environmental Sciences, Queens College, City University of New York, Flushing, NY 11367, USAb Earth and Environmental Sciences, Graduate Center, City University of New York, New York, NY 10016, USAc Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY 10964, USAd Department of Earth and Planetary Sciences, Rutgers University, Piscataway, NJ 08854, USAe Department of Geology, University of Dhaka, Dhaka, Bangladesh

A R T I C L E I N F O

Keywords:1762 Arakan EarthquakeTectonics of Indo-Burma Fold BeltPaleoseismology using Porites coralsGPS survey of coralsU/Th chronology of Porites

A B S T R A C T

In order to understand the seismic hazard associated with the Sunda megathrust along SE Bangladesh, it isnecessary to document geologic evidence for the 1762 Arakan earthquake (M8.5–8.8) and prior events.Historical records indicated that the 1762 earthquake caused extensive damage along the Arakan segment of theSunda subduction system, but geologic evidence for the earthquake farther north is necessary to better under-stand its associated seismic hazard to the heavily populated nation of Bangladesh. To document the paleoseismichistory and understand ongoing tectonic deformation, we conducted detailed analyses of corals and microatollsin Saint Martin's Island, SE Bangladesh. The coral growth bands were studied, dated using U/Th systematics andtheir elevation was measured using high-precision GPS with RTX™ capability. The U/Th dating and> 2m ofelevation differences between dead and living corals provide strong evidence of the coseismic uplift of 1762Arakan earthquake. Based on annual banding patterns and reconstruction of the highest level of survival, coralmicroatolls indicate a tectonic subsidence rate of 2.2 ± 0.8mm/yr prior to their uplift, consistent with elasticstrain accumulation before the earthquake. In contrast, annual growth bands of a living coral microatoll indicatethat Saint Martin's Island is presently experiencing uplift, at a rate of 1.8 ± 0.1mm/yr. This suggests that theanticline underlying Saint Martin's Island is actively deforming following the earthquake. Our study also pro-vides evidence for two additional earthquakes taking place in ~700 and ~1140 CE. These findings suggest apreliminary earthquake recurrence interval of ~500 years. Based on our results, the 1762 rupture can be ex-tended from Faul Island (18°N) to Saint Martin's Island in SE Bangladesh (20.6°N). The motions documentedfrom the corals suggest that this segment of the Arakan collision zone is storing sufficient energy to cause afuture earthquake of M > 8.

1. Introduction

On April 2nd, 1762 the Arakan earthquake jolted Bangladesh,Myanmar, and parts of India (Akhter, 2010; Alam and Dominey-Howes,2014). The 1762 Arakan earthquake, estimated at M8.5–8.8 (Cummins,2007; Wang et al., 2013) had devastating effects including severeground shaking, liquefaction, landslides, and flooding of the coast alongthe Bay of Bengal. Surface deformation included subsidence and uplift,the formation of mud volcanoes, and ground compaction. The effects ofthe 1762 Arakan earthquake were documented along the west coast ofMyanmar and the southeast coast of Bangladesh (Akhter, 2010; Alam

and Dominey-Howes, 2014) (Fig. 1). The associated tsunami wavescaused 200 deaths in the Chittagong area in Bangladesh and 500 deathsnear Dhaka, mostly from overturned boats (Gulston, 1763; Verlest,1763). The severe seismic shaking initiated mud volcano eruptions atSitakund in the Chittagong district (Mallet, 1878; Akhter, 1979) and inCheduba, Ramree and Faul Islands along the west coast of Myanmar(Halsted, 1843). Most of the historical evidence and description of theeffects and damage were collected immediately after the earthquakewith additional evidence collected later in 1841 by the British surveyship Childers. However, some of the descriptions collected at that timewere somewhat colloquial and exaggerated (Cummins, 2007; Wang

https://doi.org/10.1016/j.tecto.2018.07.020Received 17 November 2017; Received in revised form 13 July 2018; Accepted 24 July 2018

⁎ Corresponding author.

1 Present address: Massachusetts Institute of Technology, Haystack Observatory, 99 Millstone Rd, Westford, MA 01886, USA.E-mail address: [email protected] (D.R. Mondal).

Tectonophysics 745 (2018) 196–213

Available online 29 July 20180040-1951/ © 2018 Elsevier B.V. All rights reserved.

T

http://www.sciencedirect.com/science/journal/00401951

https://www.elsevier.com/locate/tecto

https://doi.org/10.1016/j.tecto.2018.07.020


mailto:[email protected]


http://crossmark.crossref.org/dialog/?doi=10.1016/j.tecto.2018.07.020&domain=pdf

et al., 2013; Alam and Dominey-Howes, 2014). Consequently, historicalinformation did not provide sufficient detail towards understanding therupture mechanism (Guidoboni and Stucchi, 1993). It is, therefore,critical to document geological evidence of the 1762 rupture since thisearthquake remains the only reported pre-instrumental large magni-tude earthquake along this segment of the Sumatra-Andaman-Arakansubduction zone.

Geological evidence for coseismic uplift of the 1762 Arakan andprior earthquakes is well documented along the west coast of Myanmarfrom Cheduda and Ramree island to Sittwe island (Aung et al., 2008;Wang et al., 2013) (Fig. 1). The uplift measured from dead coral mi-croatolls in Ramree and Cheduda islands ranges from 3m to>6m(Fig. 1 Box 3; Wang et al., 2013). Three marine terraces have also beendocumented as having been uplifted by three other earthquakes thatoccurred circa 1395–740 BCE, 806–1220 CE and 1585–1810 CE alongthe Rakhine coast of Myanmar (Fig. 1; Box 2; Aung et al., 2008). Theyoungest terrace was uplifted by 3–7m and correlated to the 1762earthquake. But geologic evidence of the 1762 earthquake rupturealong the southeast coast of Bangladesh has not been previously es-tablished.

The main objectives of this paper are to document the effects of the1762 rupture along the SE coast of Bangladesh by providing geologicevidence of the timing and amount of uplift from corals in SaintMartin's Island. The results are largely based on 58 U-series (U/Th) ages

acquired on 15 coral heads combined with high-precision GPS elevationmeasurements.

2. Regional geology and tectonic setting

Cratonic India has been colliding with Eurasia since the Eocene(Powell and Conaghan, 1973; Ni and Barazangi, 1984; Dewey et al.,1989). India has acted as an indenter, pushing into the underbelly ofEurasia raising the Himalaya Mountains and Tibetan Plateau (Molnarand Tapponnier, 1975). On the eastern flank of the collision, there is5000 km long Sunda Megathrust, where the Eurasian plate overridesthe Indian plate to the north and Australian plate to the south. TheSunda Megathrust extends from south to north from Bali and SouthJava, Indonesia to the Naga-Disang thrust in northern India, wherecontinental collision is occurring. From south to north, the Sundamegathrust contains the Java, Sumatra, Andaman and Indo-Burmasubduction zones. The Arakan segment is the southern part of the Indo-Burma section of Sunda megathrust (Sieh, 2007). To the south, alongthe Java subduction zone, the Indo-Australian and Eurasia plates areconverging at the rate of ~70mm/yr (Michel et al., 2001; Calais et al.,2006; Tregoning et al., 1994). As it continues northward along thesubduction zone, the obliquity increases. Along the Andaman-Sumatrasegment, subduction is partitioned with the Andaman rift, Great Su-matra Fault and other structures absorbing most of the oblique motion

Fig. 1. Historical evidence for the 1762 CE earthquake and intensity in MMI (Modified Mercalli Intensity) scale estimated for different locations (modified after Alamand Dominey-Howes (2014)). Different symbols show the different types of evidence recorded, and the color of the symbols shows estimated intensity (MMI). Modernstudies of the 1762 earthquake (boxes). From S to N: Box-4 Aung et al. (2008), Box-3 Wang et al. (2013), Box-2: Aung et al. (2007, 2008), Box-1: this study (inset;Saint Martin's Island). The inset on the upper-right corner shows the locations of evidence documenting the 1762 earthquake around the Chittagong region.

D.R. Mondal et al. Tectonophysics 745 (2018) 196–213

197

and forming a forearc sliver (Sieh, 2007; Subarya et al., 2006). To thenorth, at the Indo-Burma segment of the subduction plate boundary theobliquity approaches 70° with an overall rate of ~46mm/yr (Ni et al.,1989; Satyabala, 2003; Nielsen et al., 2004; Steckler et al., 2016) andthe motion is partitioned with the strike-slip Sagaing fault accom-modating 21mm/yr right lateral fault parallel motion (Vigny, 2003;Maurin et al., 2010). However, this is only half of the oblique motion.The remaining 25mm/yr of the oblique motion is accommodated byadditional strike-slip faults within the non-rigid Myanmar platelet andby the subduction interface between India and Myanmar platelet (Niet al., 1989; Rao and Kumar, 1999; Socquet et al., 2006; Steckler et al.,2016). However, there remains 13–17mm/y of convergent motion onthe subduction zone (Steckler et al., 2016; Sloan et al., 2017). North ofthe Indo-Burma fold belt, Eurasia is colliding with cratonic India alongthe Naga thrust belt.

Active uplift and high orogenic precipitation cause the Himalayanmountain belt to rapidly erode and exhume (Menard, 1961; Lavé andAvouac, 2001; Burbank et al., 2003; Gabet et al., 2008; Burbank andAnderson, 2011) at a rate of 2.1–2.9 mm/yr (Galy and France-Lanord,2001). As a result, two river systems, the Ganges and Brahmaputra,carry> 1 Gigaton (1012 kg) of sediments per year (Kuehl et al., 1989)from the 8000m high Himalaya mountains onto the Bengal Basin(Alam et al., 2003). The extremely high sediment supply and modestdeltaic subsidence of ≤5mm/yr (Stanley and Hait, 2000; Grall et al.,2018) contribute to sediment accretion and progradation over a100,000 km2 region forming the world's largest delta, the Ganges-Brahmaputra Delta (Uddin and Lundberg, 2004). As a result, sedimentsare ~20 km thick (Singh et al., 2016; Howe et al., in prep., personalcommunication), in the northern part of the Sunda subduction system.Much thicker than that of the Sumatra segment where the sedimentthickness ranges from 2 to 4 km (Métivier et al., 1999).

As a result of the ≥20 km sediment thickness of the Ganges-Brahmaputra Delta, the accretionary prism developed into the 250 kmwide subaerial Indo-Burma fold-thrust belt (Curray, 2014). The easternboundary of this fold-thrust belt shows typical oceanic subductionfeatures such as a volcanic arc (Fig. 2; Johnson and Alam, 1991;Steckler et al., 2008). The central portion of the fold-thrust belt is≥2000m high and contains an oceanic subduction complex that pre-dates the Himalayan collision (Rangin et al., 2013; Sloan et al., 2017)(Fig. 2C). The broad active accretionary prism consists of Miocene andyounger Himalayan derived sediments (Uddin and Lundberg, 2004;Najman et al., 2012). The amplitude of the anticlines in this outer beltdecreases and their wavelength increases westward (Steckler et al.,2008, Betka et al., 2017). Along the westernmost margin of the fold-thrust belt, some of the anticlines are subaerially exposed while othersare buried under the sediments of the Ganges-Brahmaputra Delta bothon land and offshore (Rangin et al., 2013). Saint Martin's Island,south of the delta is underlain by one of these accretionary prism an-ticlines.

Structural measurements from Saint Martin's Island show that theanticline's eastern limb is gently dipping at 12–24° (Bastas-Hernandezet al., 2014). The axis of the anticline is located offshore at the west sideof the island (Bastas-Hernandez et al., 2014). The western limb is notexposed above sea level, which suggests that it may be dipping steeplyand the anticline is west-verging. The N-S strike of the anticline followsthe trend of the fold-thrust belt. The region around Saint Martin's an-ticline shows typical thin-skinned deformation features, such as nu-merous thrust faults (Sikder and Alam, 2003) above a low angle (0.1° to0.3°) shallow detachment fault (Rangin et al., 2013). Therefore, theanticline could well be a fault-propagation fold or detachment fold si-milar to other anticlines of the outer fold belt (Mandal andWoobaidullah, 2006; Betka et al., 2017). It is not known if the anticlineshave grown by coseismic rupture of underlying thrust faults, but in1999 the Mw 5.2 earthquake on Maheshkhali Island, another anticlineabout 100 km to the north, had a surface rupture (Steckler et al.,2008).

3. Global average sea level and relative sea level changes

Global sea level affects all coasts on the Earth surface. The globalaverage, or eustatic, sea level variations are caused by effects such asmelting of glaciers or ice sheet loading, thermal expansion of watermasses and volume changes of the ocean basins (Peltier and Fairbanks,2006; Milne and Mitrovica, 2008; Rovere et al., 2016). Locally, absolutesea level changes differ from the global average, however, estimates ofrates in the NE Bay of Bengal are close to the mean (Meyssignac et al.,2012; Meyssignac and Cazenave, 2012). Relative sea level, the observedchange of the sea from land, also includes local vertical motions such assediment compaction, tectonics, isostasy and stored elastic strain fromregional earthquakes (Rovere et al., 2016). The coast along the Bay ofBengal has been affected by global average sea level (eustatic) andregional effects manifested as relative sea level changes (Warrick et al.,1996).

Global climate has been characterized by distinctively warmer andcolder periods during the past 2000 years and these climatic changeshave impacted global average sea level (Kemp et al., 2011). For ex-ample, during the Medieval Warm Period (800–1400 CE) globalaverage sea level rose at an average rate of 0.6mm/year (Fig. 3).During the Little Ice Age (1650–1850 CE) sea level may have fallen at arate of 0.1mm/year before accelerating to 2.1mm/year of sea level risefrom 1900–2000 CE (Fig. 3). These changes in the global average ratesof sea level rise/fall are much more modest than the rates of sea levelrise recorded during the late stages of deglaciation when global averagerates of sea level rise averaged 10mm/yr and approached ~40mm/yrduring Meltwater Pulse 1B, 11,400 years ago (Abdul et al., 2016).

As the Sunda and Indo-Burma subduction systems are tectonicallyactive, ongoing tectonic deformation is one of the leading causes con-tributing to relative sea level changes. Tectonic deformation includesboth sudden coseismic displacement due to earthquakes and the slowinterseismic deformation that occurs between earthquakes. The mag-nitude of coseismic uplift and subsidence can reach meters and the ratesof interseismic deformation, 0–8mm/yr (Meltzner et al., 2015), can becomparable or larger than eustatic rates. Therefore, records of relativesea level changes as those provided by corals (Meltzner and Woodroffe,2015) have the potential of providing critical data on earthquake de-formation cycles, especially along the southeast coast of the Bay ofBengal.

4. Corals as paleo-tide gauges and paleo-geodetic markers

Corals from the Porites genera have been used as paleo tide gauges(Taylor et al., 1987; Zachariasen et al., 1999a). These types of coralsbegin to grow on a nucleus, which could be a rock fragment or anothercoral head. The coral grows by adding new skeleton onto the old one ata rate of 1–2 cm/yr (Taylor et al., 1987; Zachariasen et al., 1999a;Briggs et al., 2006). The rate of coral growth is influenced by manyfactors such as species, water temperature, water depth, turbidity,salinity, available nutrients, and pollutants (Buddemeier et al., 1974;Scoffin et al., 1978). Corals grow both upward and laterally until theyreach a level in which the polyps cannot survive, such as where there isprolonged subaerial exposure during spring low tide levels. At thispoint, the coral head only grows laterally and becomes flat; it is thentermed a microatoll (Smithers and Woodroffe, 2000). The elevationabove which a coral microatoll stops growing is known as the highestlevel of survival (HLS) (Taylor et al., 1987). Extreme low water (ELW)refers to the lowest water level achieved over a period of time. Therelation between ELW and HLS for any species can be different fordifferent regions (Meltzner et al., 2015). At a given time, the highestelevation at which the coral can grow at its maximum growth rate isknown as highest level of growth (HLG) (Meltzner et al., 2015, 2010;Weil-Accardo et al., 2016). The term HLG is used for those years whenHLS cannot be determined.

A coral microatoll HLS history can be used to deduce relative sea


198

level changes if the difference between the ELW and HLS is known forthat coral species in that region. Since only the relative sea level isknown, it is very difficult to separate the signal of global average sealevel change, long-term slow tectonic signal and global isostatic ad-justment from the coral microatoll. During past 2000 years eustatic sealevel averaged a 0.5 m rise (taking into consideration a −1mm/yearlowering of sea level during the Little Ice Age). If eustatic sea levelremains unchanged a coral could provide evidence for relative sea levelchanges related to either tectonic uplift or subsidence including global

isostatic adjustment (GIA). However, according to VM2 GIA model(viscosity model based on the Bayesian inversion of the GIA data), thecoast of Bay of Bengal is experiencing sea level change at a rate of0–0.2mm/yr relative to the eustatic curve (Peltier, 2002; Spada andGalassi, 2012). In the absence of a local sea level curve, we use theglobal eustatic curve to represent local sea level change and neglect alocal GIA component. We interpret the microatolls submergence andemergence in response to the relative sea level changes as the differencebetween eustatic sea level and local land motion. Table 1 summarizes

Fig. 2. Tectonic setting of the study area. A)Tectonic elements and earthquake rupturesegmentation along the Sunda Megathrust.The color background shows elevation andbathymetry of this region. B) Rupture dis-tributions along the megathrust. From −5°N to 15°N, almost the entire segment rup-tured in recent times (from 2004 to 2007)(Natawidjaja et al., 2006; Konca et al.,2008; Banerjee et al., 2007; Ambikapathyet al., 2010). The northern segment, northof 15°, hasn't ruptured since 1762. C)Structural cross-section of the study areaadapted from Rangin et al. (2013). The lo-cation of the profile is shown on map A by ayellow line. The overlaid earthquake seis-micity and focal mechanisms were down-loaded from the Global Earthquake Model(https://www.globalquakemodel.org). (Forinterpretation of the references to color inthis figure legend, the reader is referred tothe web version of this article.)


199

https://www.globalquakemodel.org

the growth of microatolls under different tectonic conditions.The growth pattern of a coral microatoll in response to varying

tectonic conditions assuming a steady slow and constant rate of sealevel may be reconstructed by the annual growth band patterns pre-served in the coral microatoll (Fig. 4). For example, the coral microatollgrows upward at a constant rate when relative sea level rises at a slowand steady rate. A slow steady rate of relative sea level rise allows forthe coral to grow both sideways and upwards (Fig. 4). In the case of aminor relative sea level drop, the coral microatoll ceases to grow up-ward but continues to grow from a new HLS (Zachariasen et al., 2000).Inter-annual climatic variability such as El-Nino Southern Oscillation(ENSO) and associated sea level variations can cause small disruptionsin the growth of coral microatolls (Philibosian et al., 2014). Othernatural phenomena such as cyclones and tsunamis can disrupt the coralgrowth by modifying the environment in which the coral microatollwas growing (Scoffin et al., 1978; Smithers and Woodroffe, 2001).Sudden uplift by an earthquake can dramatically lower relative sealevel which can kill the entire microatoll (Meltzner and Woodroffe,2015).

The morphology of coral microatolls can also be used to extractinformation about longer and slower rates of relative sea level change.For instance, hat-shaped or cone-shaped coral microatolls are producedwith a growth during a steady rate of relative sea level lowering(Fig. 4C), such as associated with slow rates of tectonic uplift. On theother hand, a cup-shaped microatoll is suggestive to an emergent event(Fig. 4D).

5. Methods

To document evidence of the 1762 earthquake rupture along theArakan segment of the Sunda Subduction zone, we conducted paleo-seismological investigations in Saint Martin's Island and the Teknafpeninsula along the southeastern coast of Bangladesh in January 2013,

2014 and 2015. In all field seasons coral samples were collected duringlow spring tides (tidal range in Saint Martin's Island is ~2.8 m), a timewhen both living and dead coral heads were exposed and accessible.The elevation of each sampled coral head was measured by high pre-cession, fast static GPS with RTX™ (trademarked to Trimble NavigationLimited) capability to an accuracy of± 10 cm.

5.1. Sampling corals in Saint Martin's Island

In this study, we only sampled Porites Lutea/Lobata corals that werein-situ and showed no or minimal visible signs of disturbance and di-agenesis. We employed several sampling techniques to obtain coralsamples. We collected large pieces of corals using a chisel and hammerand used a powered hand drill equipped with a 2-inch drill bit to obtainoriented cores ~10 to 20 cm in length. We also used a gasoline operatedchainsaw with a diamond chain to cut 2-inch thick vertical slabs or-iented from the center to the edge of the coral head. The slabs werelater X-rayed in a local hospital. The X-ray images were examined on alight table to identify growth bands that were then digitized and ana-lyzed in order to map both continuous and interrupted growth patterns.

The corals from the Saint Martin's Island are in danger by manyanthropogenic activities such as, collecting and selling as showpieces,uncontrolled waste disposal from the resorts and hotels, physical de-struction by fishing and anchoring boat, poorly controlled sanitationand agricultural practices that introduced nutrients and pesticides asidentified by Hossain and Islam (2006). We do not encourage andsupport any activity that destroys the habitat of coral. Keeping that inmind, we restricted our sampling to dead coral heads and collected onlya minimum number of representative samples from each location. Werestricted our sampling to a single living coral, which was needed to“ground truth” U/Th dating accuracy and to obtain the growth historyfor geophysical interpretation, without destroying the coral head.

Fig. 3. GIA (Glacial isostatic adjustment) adjusted sealevel variations for the past 1400 years (modified fromKemp et al., 2011). Three green bell curves show thetiming (2σ uncertainties) of the change in the rate of sealevel rise. The average rates of sea level in between thechange points are shown in the boxes above the x-axis.(For interpretation of the references to color in this figurelegend, the reader is referred to the web version of thisarticle.)

Table 1The growth of a coral microatoll in response to different tectonic conditions during sea level rise.

Global sea level change Tectonic uplift/subsidence Resulting relative sea level change Microatoll response

Rise Uplift Uplift rate > global sea level rise Drop EmergenceUplift rate= global sea level rise No change No change observedUplift rate < global sea level rise Slight rise Slow submergence

Rise Remains unchanged Moderate rise Moderate submergenceSea level rise Subsidence Rise Fast submergence


200

5.2. U-Th analysis of the coral samples and age correction

All coral samples were subsampled using concentric 10mm and6mm drill plugs. The following steps were performed: each subsamplewas mechanically cleaned with high-frequency sonication, first withMICRO detergent, then by 30% H2O2, and finally with double deionizedH2O (DDW). Cleaned coral sub-samples were then crushed to powderand analyzed for evidence of secondary calcite by X-ray diffractionanalyses (XRD). Three samples (SM-D13, D-03-01, D-03-02) failed topass a “no detectable” calcite (< 0.2 wt%) criteria (Chiu et al., 2005)(Table A1).

Our approach in this study was to conduct U-Th dating of coralsamples rather than radiocarbon (14C) dating, which is commonly ap-plied to marine carbonate materials. While 14C dating of fossil coralscan provide precise ages all dates must be calibrated to calendar age.However, it is necessary to correct 14C dates for the age differencebetween the contemporaneous atmosphere and the seawater in whichthe coral grew (the marine reservoir effect). These age corrections canbe hundreds of years, can be variable, and are not well known for ourstudy site. U-Th dating of relatively young corals has previously beendemonstrated in the dating of earthquake events (Edwards et al., 1988)and for absolute dating of climate records (Cobb et al., 2003). For the U-Th dating, approximately 0.5 g of coral was spiked with a calibrated229Th-233U solution and dissolved in 1ml of concentrated nitric acid.Organics were digested with the addition of HClO3 acid. Uranium andThorium isotopes were separated through anion exchange columnchemistry following co-precipitation as iron hydroxide. Due to the largesample masses required for accurate dating of young corals, the Thfraction occasionally required a second column (100 ul) cleaning step inorder to remove excess U.

U-series (U-Th) dating was determined by Inductively Coupled MassSpectrometry ICP-MS (NEPTUNE PLUS) in the Department of Earth andPlanetary Sciences at Rutgers University, applying the methods and

protocols described in Mortlock et al. (2005) as modified by Abdul et al.(2016). Specifically, the peak-jumping mode was used to measure theisotopes of U and Th with a secondary electron multiplier (SEM) posi-tioned behind a retarding potential quadrupole (RPD). U isotopes weremeasured with sample-standard bracketing (SSB) to correct for SEMgain efficiency and drift. Instrument mass fractionation was correctedby the monitoring the 235U/238U ratio. Measurement accuracy wasconfirmed with measurements of certified reference materials (CRM112a, CRM U010). The average precision of the age determination (at2σ=2 Standard Deviations) is between± 0.5% and ± 0.2‰ de-pending on the age of the sample (Table: A1). Replicate samples wereobtained parallel to visible growth bands and were used to determineisochron ages. Procedural blanks were<0.1 to 0.5 fg and 2 to 20 pg for230Th and 232Th, respectively. Sample ages have not been corrected forprocedural blanks since they contribute to corrections of 5 years or lessand are an order of magnitude less than corrections due to 230Th initial,as discussed below.

The accuracy of U-Th dating in carbonates may at times be limitedby corrections for non-radiogenic or initial 230Th. This component,termed initial 230Th, does not result from in situ decay of 238U but in-stead is added to a precipitating carbonate from the surrounding wa-ters. Generally, corrections for initial 230Th in surface corals are notnecessary since its concentration in seawater is extremely low as it is aninsoluble element that easily adsorbs to the sediment grains and effi-ciently removed by scavenging onto particles (Cheng et al., 2000).However, in young samples or samples containing high concentrationsof 232Th age corrections for initial 230Th may be important (Cobb et al.,2003).

Saint Martin's Island is at the boundary between clean marine wa-ters and sediment-rich waters derived from the Naf and Ganges-Brahmaputra river systems. Indeed, surface seawater we collected atSaint Martin's pier was found to have 232Th concentrations in excess of3 ppb (Table 2), orders of magnitude higher than typical seawater.

Fig. 4. Growth pattern of coral microatolls in response to relative sea level (RSL) changes (Modified from Zachariasen et al., 2000). Graphs show relative sea levelchange; the corresponding cartoons show the microatoll growth under that sea level change scenario. The interannual variability is not considered. The dotted line onthe figure shows the growth bands however it does not represent the number of bands. A) Microatoll growth below RSL. The growth bands do not reach the HLS. Inthis case, the elevation of the growth bands is described as HLG. B) The microatoll grows upward and reaches the HLS and then grows laterally. C). When sea leveldrops, the coral microatoll grows laterally, maintaining a new HLS. D) In the case of a sudden sea level rise (scenario following C), the microatoll starts to growupwards creating new growth bands. E) Microatoll shows a complex HLS history, the color line shows HLG, HLS and a sea level drop. Following an emergence event,the microatoll grows upwards until it reaches the HLS.


201

Therefore, we determined it necessary to corrected for 230Th due to therelatively young age of the corals that are the focus of our study.

Two approaches are usually considered for estimating and cor-recting U-Th ages for initial 230Th. One approach is to correct for initial230Th by combining the coral's measured 232Th content with a230Th/232Th atom ratio of 4.4× 10−6, which is the value of averagecrustal materials at secular equilibrium assuming a bulk earth232Th/238U ratio of 3.8. A second, and perhaps preferred technique is toconstruct isochrons, where sub-samples of same age coral are analyzedfor their [230Th/232Th] and [232Th/238U] activity ratios in order touniquely constrain [230Th/232Th] initial values (Edwards et al., 1988;Richards and Dorale, 2003). To reduce age uncertainty, our strategywas to construct isochrons on a representative number of samples tobest constrain [230Th/232Th] initial, its range and variability.

Isochrons obtained from coral microatolls SM-C5, SM-C10 and SM-C16 generated [230Th/232Th] initial activity ratios of 4.03 ± 0.90,0.32 ± 0.23 and 3.36 ± 5.25 respectively. Two corals have[230Th/232Th] initial activity ratios much higher than the averagecrustal ratio (~0.8) suggesting a source of excess 230Th besides detrital,possibly via the remobilization of 230Th associated with reversible ex-change reactions on particles carried in river waters and seawater. A[230Th/232Th] initial of 0.32 is less than the crustal value and may re-flect the more typical environment in which 230Th has been effectivelystripped out from surface seawater, thereby resulting in minimal or noage corrections. The range of initials (0.32 to 4.03 from the isochrons)displayed in the Saint Martin's Island corals are very similar to the in-itial (3.8) used to correct U-Th ages in relatively young corals of knownage from Central Pacific Ocean (Cobb et al., 2003).

Where we have isochron analyses, coral ages were corrected for230Th initial using the measured intercept obtained from isochron plots.In addition, we applied age corrections to all corals using a[230Th/232Th] initial range of 0.8 to 4, thereby spanning correctionsbased on the crustal values and including the highest values determinedby our isochron analyses. We recognize all age corrections provide“minimum” U-Th ages and ranges. However, as a support of our ap-proach, a living Porites (Q07-02-1) yielded a U-Th corrected date of1981 ± 7 CE. This date range is in excellent agreement with the as-signed date of ~1982 CE obtained from counting the annual growthbands (Table A1). This demonstrates that using a range of[230Th/232Th] initial of 0.8 to 4 to correct the ages can have the re-quired accuracy for dating of corals used in this study. Throughout thiswork, we have reported concentration data, activity ratios, and

uncorrected ages are reported at 2σ. Corrected dates are expressed ei-ther as the sample-specific isochron age and uncertainty (based on themeasured 230Th/232Th initial and uncertainty: 1σ; only for SMC05,SMC10 and SMC16) or the average of the two dates and range based oncorrections using [230Th/232Th] initials of 0.8 and 4 (Table A1). Finally,all coral samples yielded δ234Uinitial (144 to 150‰) that is within therange of seawater and measured in modern corals known to havemaintained and grown in a closed system (Cheng et al., 2013).

5.3. GPS survey of the coral heads

The elevation of the coral heads was measured by dual frequencyfast static GPS during our surveys. In 2013, a Trimble NetR9 GPS wasset up as a base station for the duration of the fieldwork. The position ofthe base station was determined using GAMIT/GLOBK including ourlocal network in Bangladesh (Steckler et al., 2016). A second TrimbleNetR9 GPS was used as the rover GPS and the data were processedusing Trimble Business Center to a vertical accuracy of± 11 cm. In2015 and 2016, coral head elevations were measured using a singleTrimble NetR9 GPS with RTX™ capabilities. The OMNISTAR high pre-cision service provided elevations within±10 cm. Each data point wassurveyed for 30min using a geodetic tripod with the antenna mountedon it. The EGM96 geoid height was subtracted from the GPS ellipsoidheight determined for each station to estimate the orthometric height(which approximates the height above mean sea level). The spring tidalrange was measured using the rover GPS, to determine the elevation ofthe local mean sea level. Our observations reveal that there is an offsetof 1.4 m between the observed and theoretical local mean sea level.Therefore, the GPS measurements have been transferred into the localtidal datum (Table 3).

6. Results

6.1. Evidence of uplift related to the 1762 earthquake

6.1.1. Distribution of coral colonies in Saint Martin's IslandAlong the surveyed regions, the coral heads are not distributed

uniformly. There are three main dead coral colonies located along theeastern coast and one colony along the western coast of the island(Fig. 5). The identified colonies contain many different varieties ofcorals including Porites Lutea/Lobata.

The southern part of Saint Martin's Island, called Chheradip, haseast dipping indurated strata composed of Miocene sandstone(Chowdhury et al., 1997). This indurated sandstone is more resistant toerosion and has formed parallel ridges that bound areas of low relief.These areas of low relief were likely composed of strata that were re-moved by erosion and have formed bays that provide protected areasfor the corals to grow. These bays remain connected to the sea duringlow tide that suggests the living corals are not ponded. At least six deadcoral colonies were mapped within these bays along the southern,

Table 3Elevation of coral heads measured by fast static GPS.

Coral ID Longitude in decimal degrees E Latitude in decimal degrees N Elevation (m) In Local Tidal Datum (m)

SM-C05 92.3373 20.5761 2.23 ± 0.11 0.83 ± 0.15SM-C06 92.3375 20.576 2.03 ± 0.03 0.63 ± 0.11SM-C10 92.3369 20.5827 1.99 ± 0.13 0.59 ± 0.17SM-C11 92.3372 20.5828 2.02 ± 0.08 0.62 ± 0.13SM-C16 92.3318 20.6041 2.02 ± 0.09 0.62 ± 0.14SM-D04 92.3372 20.5761 1.94 ± 0.14 0.54 ± 0.18SM-D07 92.3245 20.6018 2.72 ± 0.17 1.32 ± 0.20SM-D09 92.3369 20.5829 2.57 ± 0.25 1.17 ± 0.27SM-D13 92.3319 20.6047 2.02 ± 0.05 0.62 ± 0.12SM-A05 92.3369 20.5825 2.31 ± 0.02 0.91 ± 0.11SM-Q07 92.3393 20.5823 0.06 ± 0.09 −1.34 ± 0.14

Table 2Concentration of 232Th in Saint Martin's seawater.

Sample name 232Th(ppb)

St Martin SW-1 (2015) 3.42 ± 0.01St Martin SW-2 (2015) 3.20 ± 0.01


202

central and northern parts of the island (Fig. 5). Fig. 5C shows a coralmicroatoll with a completely flat upper surface. This indicates thatmicroatoll experienced constant low water during its lifespan, thus was

most likely ponded during its growth. We specifically did not samplemicroatolls from potentially ponded regions. Numerous live coral co-lonies were identified offshore during low tide particularly on thesoutheast segment of the island (Chheradip). HLS for live coral micro-atoll was estimated to be 9.52 ± 1.9 cm that we obtained by surveying18 microatolls (Fig. 6). The dead coral colonies are located within theintertidal zone with a measured spring high tide of 1.43 ± 0.06mabove mean sea level and spring low tide of 1.49 ± 0.09m belowmean sea level (Fig. A1). The dead coral colonies are barely submergedduring the spring high tide and remain exposed otherwise (Fig. 7). Thelive coral colonies, located below the subtidal region near the springlow tide level, generally remain submerged and occasionally becomeexposed during spring low tide.

6.1.2. Morphology, ages and locations of the coral headsWe sampled and obtained U-Th ages of corals in three different

regions of Saint Martin's Island: south Chheradip, north Chheradip andcentral Saint Martin's Island (Figs. 5 and 8).

6.1.2.1. South Chheradip. Three coral heads, SM-C05, SM-D04 and SM-C06, were sampled in a bay on the eastern part of the island. Numerousdead coral heads were found in this bay. However, we found manycoral heads that were broken apart and considered not to be in-situ. SM-C05 is a microatoll and has a diameter of 1.5 m (Fig. A2). The top of thiscoral head has flat to semi-flat top with the presence of severalconcentric ridges although they are highly eroded. The elevation ofthis coral head, obtained by GPS from its center is 0.91 ± 0.11mabove local mean sea level. A drill core was collected from the center ofthe coral head and chunk samples were collected from the edge. The

Fig. 5. A) Distribution of live and dead coral colonies in Chheradip (southern part of St. Martin's Island) and central and northern parts of the island. Dashed lineshows low tide level, red lines indicate the location of dead coral colonies and blue lines indicate the locations of live coral colonies. B, C and D show the location ofthe photos of dead and live coral colonies. B) Dead coral colony. C) Dead coral microatolls with flat upper surface morphology. D) Living tissue of Porites. (Forinterpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 6. Highest Level of Survival (HLS) measured from different live coralscolonies. Measurements were taken during spring low tide.


203

isochron age of the edge sample is 1780 ± 24 CE (Fig. 9). The center ofthis microatoll (core A02-01) is dated back to 1670 ± 95 CE or 50 to100 years older than the coral edge.

Coral head SM-D04, located 4m from SM-C05, has a radius of 2.2m.The top of this microatoll is flat to semi flat with irregular sharp ridgesand crests, and exhibits concentric annuli and overgrowths. Surfacemorphology, such as irregular sharp ridges suggests that the coral topmay have eroded. Some pieces of this coral head had fallen away fromthe center and were found around the coral head (Fig. A6). The ele-vation obtained from the outer most layer is 0.54 ± 0.18m above localmean sea level. The core collected 30 cm from the edge of this micro-atoll was dated at 1485 ± 21 CE (sample SM-D04-01; Table A2).Another core collected from the center (2.2 m from the edge) was datedat 1455 ± 28 CE. However, the lower elevation of this coral head and

older age relative to adjacent SM-C05 suggest that the coral head mayhave been significantly eroded. Another possibility could be that thiscoral head died before SM-C05 started to grow.

Coral head SM-C06 is located 6m away from the SM-C05. The top ofcoral head SM-C06 is semi-flat to rounded. Concentric ridges are notvisible on the top of the head. Thus, we think that this coral head maynot be a microatoll (Fig. A3). The elevation at the center of this coralhead is 0.63 ± 0.11m above local mean sea level. Samples dated at thecoral edge yielded a corrected date of 1624 ± 7 CE.

6.1.2.2. North Chheradip. The dead coral colony in the northernChheradip is in a bay that becomes submerged during spring hightide. We collected small pieces, cores, and slab samples from four coralheads.

Fig. 7. Profile from land to offshore showing the location and elevation of dead and live corals. The location of this profile is shown by a dark line on Fig. 5A(Northern Chheradip region). The tide level was measured with GPS during spring high tide and spring low tide.

Fig. 8. Panel A, B and C show the location and elevation ofcoral microatolls relative to sea level. The horizontal axis foreach graph represents time. The ages of the coral heads andmicroatolls are the average ages from multiple samplesobtained from the edge of each microatolls. The horizontalerror bars represent the range of the ages. The ages of SM-C05, SM-C10, SM-C16 were obtained from isochrons andthe rest of the ages were calculated using the [232Th/230Th]of 0.8 to 4. The vertical error bars (red) represent the ele-vation uncertainty. SHW=Spring High Water,LMSL=Local Mean Sea Level, SLW=Spring Low Water.(For interpretation of the references to color in this figurelegend, the reader is referred to the web version of this ar-ticle.)


204

SM-A05 is the best-preserved coral microatoll from this region. Thecoral head is perfectly rounded and the top of the head shows threeconcentric ridges on the growth bands. A slab from the center to edgewas taken for HLS analysis (Fig. 10). The elevation of this microatoll is0.91 ± 0.11 above mean sea level, 2.34 m above present-day livingcoral heads, and has a radius of 1.02m. The outermost growth bandyielded a corrected date of 1784 ± 7 CE. The center of the coral headwas dated at 1691 ± 9 CE. The detailed sea level history obtained fromthe coral growth patterns of SM-A05 will be discussed later.

Coral head SM-D09 is the largest coral microatoll from this area.The radius of this coral head is ~2.5m and it is located ~10m fromSM-A05. This massive coral microatoll has several concentric ridgeswhich make the surface of this coral microatoll very irregular. A slabwas cut from the outer section of this microatoll. This slab was not usedfor HLS study because the coral head was tilted. The elevation of thiscoral is 1.17 ± 0.27m above local mean sea level and ~2.6m abovepresent-day living corals. The edge of this coral head yielded a cor-rected age of 1774 ± 16 CE (Fig. 8).

Coral heads SM-C10 and C11 are ~3m apart from each other andare located ~12m from SM-A05. The radius of the SM-C10 is 1.2m andSM-C11 is 1.45m. Both coral heads have flat to semi flat tops withconcentric ridges, which are not prominent and have been highlyeroded. The elevation of the two microatolls is 0.59 ± 0.17m and0.62 ± 0.12m above local mean sea level, respectively. The outermostlayers of SM-C10 and SM-C11 yielded corrected dates of 1125 ± 47 CEand 1201 ± 55 CE, respectively (Fig. 8). Although SM-C10, SM-C11,SM-A05 and SM-D09 were all found in the same colony, the elevationsof the coral heads SM-C10 and SM-C11 are 0.5 m lower and their ages~550–600 years older than SM-A05 and SM-D09. The older corals maybe associated with an earlier seismic event.

6.1.2.3. Central Saint Martin's Island. Most of the coral heads in thisarea were found to be reworked and are not considered not to be in-situ.We sampled two coral heads from the eastern coast (SM-D13 and SM-C16) and one from the western coast (SM-D07) that appeared to be wellpreserved. The coral heads from the eastern part of the island are theoldest coral heads that were dated in our study. Both corals from theeastern coast show concentric ridges and annuli but they are highlybroken. Thus, we think both SM-D13 and SM-C16 are microatolls. Wecollected one sample from the edge ridge of each of these coral heads.The corrected ages for the samples collected from the edge of SM-D13 is871 ± 2 CE. SM C16 yielded a wide range of possible dates, extending

from 506 to 1161 CE. due to a wide range in measured 232Th and basedon sample-specific initial corrections. The elevation of the two coralheads is 0.62 ± 0.13m above local mean sea level.

The SM-D07 coral sample is the only sample that we collected fromthe western coast of Saint Martin's Island. This coral head did not ex-hibit the concentric ridges and annuli. Thus, we think that this coralhead may not be a microatoll. The sample from the edge of the SM-D07coral head yielded a corrected age extending from 1529 to 1967 CE.The large range in the age of coral SM-D07 reflects the range in 230Thinitial age corrections due to its extremely high 232Th content(12.2 ppb). The elevation of this coral is 1.32 ± 0.20m from presentlocal mean sea level. Due to the large uncertainty in the age, the coralhead SM-D07 is not considered in the interpretation of results.

We also obtained samples from the edges of two other coral headsSM16-D1 and SM16-D2 from eastern central Saint Martin's Island thatwere dated at 1760 ± 8 CE and 1264 ± 6 CE, respectively (Fig. 5).

6.1.2.4. Northern Saint Martin Island. From the northern part of theisland, we sampled two corals (at the edge of the head): SM16-C1 andSM16-D3 that were dated at 1767 ± 16 CE and 995 ± 9 CE,respectively. Unfortunately, we were not able to document theelevation of these coral heads.

6.1.3. Microatoll development before 1762 earthquakeWe sub-sampled a slab cut from coral head SM-A05 from northern

Chheradip and obtained multiple U-Th ages in order to reconstruct itsHLS (highest level of survival) history (Fig. 10). Our results fromcounting the growth bands suggest that the coral head SM-A05 firstbegan to grow around 1645 CE. The coral grew, nearly vertically, untilit reached HLS level in ~1686 CE. After that time the coral growthbands are oriented laterally (outward) for 5 to 6 years. We interpret thisto reflect a near constant relative sea level after which time the coralappears to have experienced a die down at ~1691 CE (Fig. 10), perhapsfrom prolonged subaerial exposure. The lower part of the coral becamerecolonized and started to grow again. The orientation of the following~15 growth bands suggest vertical growth upwards. After the 15 years,the growth bands are laterally oriented for a duration of 16 years. Thissuggests that during these 16 years the coral growth occurred during atime in which sea level did not change or there was a “constant rate” ofrelative sea level. At this point, relative sea level was ~5 cm higher thanat 1691 CE (Fig. 10). The coral banding indicates that an event occurredand resulted in 5 cm of sea level drop. This was followed by 15 cm ofrelative sea level rise. From ~1719 CE the coral grew vertically for16 years until the growth bands reached the HLS. The new HLS is thesame as the HLS prior to 1719 CE. The coral head then recorded a smalldie-down of ~5 cm at ~1741 CE. After the die-down the coral brieflygrew upward and then laterally to its highest recorded elevation, whichimplies a permanent change of HLS which could be have caused by asmaller earthquake. Finally, in ~1762 CE the entire coral head died andstopped growing.

6.1.4. Post 1762 microatoll developmentResults from the live coral colonies reveal that the coral microatoll

SM-Q07 started growing ~1.3 m below local mean sea level, and 2.5mbelow the dead coral microatolls. This microatoll began growingaround 1800 CE (as suggested by U-Th dating) and grew vertically untilit reached relative HLS level (Fig. 11). Based on the annual bandcounting here are three apparent die-downs documented in this mi-croatoll growth bands: ~1866, ~1906 and ~1970 CE. There is noevidence that these die-downs are related to local tectonics since thepre and post-earthquake relative sea level appear to be the same. Thedie-down in ~1866 CE resulted in HLS lowering of 15 cm after whichtime the coral grew upward and laterally, reaching same HLS level in15 years. Two events, in 1906 and 1970 CE resulted in nearly identical8 cm HLS lowering. In both cases, the coral grew upwards and laterallyreaching its previous HLS level within 6 to 7 years.

Fig. 9. Isochron analyses of coral samples SM-C5, SM-C10, SM-C16 as obtainedfrom replicate samples taken from a single growth band. The samples wereobtained from the edge of the microatoll.


205

7. Discussion

7.1. Evidence for the 1762 and prior earthquakes

The ages obtained from U/Th dating and the GPS elevation mea-surements of the coral microatolls above sea level document the 1762and two prior paleoearthquakes.

7.1.1. Timing of the coral die-downThe U-Th corrected ages obtained from the last or terminal growth

band of coral heads, show three temporal distributions (Fig. 12b). Theage of the last growth band is important because it reveals the timewhen a coral head presumably died. Five coral heads cluster in datingfrom 1751 to 1805 CE (Table 4). Three coral heads (SM-C10, SM-C11

and SM16-D3) have dates ranging from 986 to 1193 CE and one coralhead (SM-C16) was dated at 606–712 CE (Table 4). As discussed earlier,SM-C16 is located in an area where numerous similar coral heads arepresent. These corals, however, showed signs of physical and chemicalweathering. The majority of these coral heads were broken and theirshape was not preserved. Therefore, additional corals were not selectedfor U-Th dating.

7.1.2. Non-tectonic sea level variationOur studies of two coral slabs (SM-A05, SM-Q07) reveal that the

corals display complex and dissimilar growth patterns as both showperiods of growth interruption that temporarily halted growth but didnot kill-off the microatolls altogether. This phenomenon is verycommon for microatolls and has been observed in other regions, such as

Fig. 10. Coral microatoll SM-A05. A) X-ray. B) Drawing of the growth bands. C) Plot of relative present day mean sea level versus age showing the growth history.


206

Sumatra and Haiti (Meltzner et al., 2010; Weil-Accardo et al., 2016).Possible explanations for these growth interruptions could be related tosea-level variations caused by temporary oceanographic changes such

as those documented by Philibosian et al. (2014) and Meltzner andWoodroffe (2015). In Sumatra, these oscillations are caused by the In-dian Ocean Dipole Moment and are regional in extent. The Indian

Fig. 11. History of live coral SM-Q07. A) Drawing of the growth bands. B) Plot of local mean sea level versus age revealing the coral's growth history. The growthbands shown by open circles represent broken segment.

Fig. 12. A) Elevation of dead and living coral headsmeasured by GPS and adjusted to local mean sea level.The red curve is the global sea level estimate for the last~1400 years (Kemp et al., 2011). The age vs elevationsuggests that the coral data encompassed a long-termrelative sea level rise at the rate of 0.2 mm/yr. SamplesSM-D07 and SM-D09 are excluded from the fit becauseSM-D07 is on the western coast of the island, and SM-D09is believed to be a tilted coral head. B) The elevation ofthe coral heads after subtracting the global sea level riserates from obtained from Fig. 3. The corrected data sug-gests subsidence at the rate of 2.2 ± 0.8mm/yr prior tothe 1762 earthquake, as recorded by SM-A05 and uplift ata rate of 1.8 ± 0.1mm/yr as recorded by SM-Q07. Thedashed line shows projected subsidence rate. The greencurve above the x-axis shows the age density estimatedistribution for the 15 dated corals ages. (For inter-pretation of the references to color in this figure legend,the reader is referred to the web version of this article.)


207

Ocean Dipole Moment (IODM) is an irregular oscillation of sea surfacetemperature in the Indian Ocean, in which the western ocean becomeswarmer and colder alternatively than its eastern part (Saji et al., 1999).During these IODM events the sea surface level changes on the order of5–10 cm, depending on the strength of the events (Clarke and Liu,1994). These relative sea level variations are not permanent and sealevel returns to its original position within one or two years. However,if the topmost corals are killed, it can take several years of growth forthe coral to reach HLS once again. Microatoll SM-A05 showed evidencefor relative sea level variations in ~1691, ~1719 and ~1741 CE, andthe coral microatoll SM-Q07 showed evidence of relative sea levelchange in ~1866, ~1906 and in ~1970 CE (Fig. 11). Previous studiesindicate that the geographic extent of non-tectonic sea level variationsin the Indian Ocean equatorial regions, Sumatra for instance, is sub-stantial and can cause large-scale changes in the growth patterns ofcoral microatolls (Taylor et al., 1987; Zachariasen et al., 1999b). Exceptfor one die down event near 1970 that could possibly be caused byIODM events mentioned by Abram et al. (2003, 2007), there is no directlink between IODM events in the Bay of Bengal as those documented inthe equatorial regions of the Indian Ocean.

Other possible causes for interruption of coral growth could be adecrease in salinity or increased turbidity related to local fluvial dis-charge. The geographic setting of Saint Martin's Island is influenced bysediment discharge that is> 1 GT/yr from 2 main rivers (Barua, 1990;Kuehl et al., 2005). Fluvial discharge in the SE Bay of Bengal is directlyrelated to the strength of the Indian Monsoon. The Ganges-Brahma-putra River system, in response to the monsoon rains, has a mean dis-charge of 38,000m3/s with peak rates that can exceed 100,000m3/sduring the summer monsoon (Barua, 1990). Most of the discharge flowswestward from the river mouth away from St. Martin's Island. However,oceanographic events such as IODM could direct more of the river flowtowards and along the SE coast of Bangladesh. In addition, flash floodson the Naf River can transport turbid waters and reduce salinity, whichcould affect the coral growth. However, historical records of pre-cipitation, flood and cyclones for the region are not known for timescorresponding to the coral die down events.

Another possible cause considered for ceasing of coral growth arealgal blooms (Charles et al., 1997; Abram et al., 2003, 2007;Natawidjaja, 2004; Natawidjaja et al., 2007). A number of factors cancause algal blooms. For example, upwelling of nutrient-rich deep water,heavy rainfall on the land washing off phosphates and Vitamin B12from the soil and salt-marsh areas into the sea, and blue-green algaehave been suggested as possible causative agents for the algal blooms(Lapointe, 1997, 1999; Hughes et al., 1999). Other environmentalvariables such as sea surface temperature extremes and annual solarradiation changes could also affect the calcification rate of the corals.The calcification rate eventually affects the thickness of the growthbands A slow calcification rate could lead to the growth of a widebandor no-growth (Lough and Barnes, 2000; Lough and Cooper, 2011).

7.1.3. Evidence for coral upliftResults from the elevation survey at Saint Martin's Island revealed

that dead microatolls are positioned 1.8 to 2.5m higher than present-day living corals. Living colonies can grow no higher than the justabove the spring low tide level since they cannot survive prolongedexposure to air and sunlight (Lough and Barnes, 2000; Lough andCooper, 2011). Today, all dead coral microatolls are only submergedonly during spring high tide. As there is little evidence for changes~2m amplitude changes in global sea level during the past millennium,we ascribe this elevation difference to local tectonics. To assess whetherthe coral microatolls were elevated due to a sudden event or by slowgradual tectonic uplift, we carefully studied the shape of the micro-atolls. Slow and gradual uplift would result in an upside down “hat”shaped coral (Zachariasen et al., 2000). In contrast, Saint Martin's mi-croatolls are either cup-shaped or flat-topped. This observation in-dicates that the interruption in coral growth was likely due to a suddenuplift or emergence of the land, such as that caused by an earthquake.

Based on the corrected U/Th ages and GPS elevation measurementsof dead coral microatolls we conclude that the 1762 Great Arakanearthquake ruptured the Saint Martin anticline leading to the ~2muplift of the southern part of the Saint Martin's Island and the death ofthe microatolls. The U-Th ages at the center of the living coral SM-Q07,and its present location and elevation support our conclusion. Thecenter of this coral dates to< 1813 ± 22 CE and indicates that thecoral SM-Q07 began to colonize after the 1762 earthquake. The suddenuplift during the 1762 earthquake lowered the preferred water depthfor existing corals and required that new colonies grow farther offshorefrom the dead corals.

Two other populations of microatolls cluster around at 986 to1193 CE and 606 to 712 CE, thereby raising the possibility that two pre-historical earthquakes that uplifted Saint Martin's Island, causing thedeath of the microatolls. However, our evidence for these two earth-quakes is not as robust as for the 1762 earthquake because of limitedsampling and U-Th dating due to poor preservation of coral heads. Inaddition, the elevation of these coral microatolls is similar to that of thecoral microatolls that died and were elevated during the 1762 earth-quake. This suggests that coseismic, postseismic and interseismic mo-tions may balance each other such that the corals remain at similarelevations following each earthquake cycle (Fig. 12).

We conclude from the evidence (U-Th dating, changes in elevation,growth patterns) that the primary cause of death for several coral headswas coseismic uplift during the 1762 earthquake. The effects due tonon-tectonic or changing oceanographic conditions (e.g., temperature,salinity) would be different as revealed from the growth band analysisof the coral slabs obtained from SM-A05 and SM-Q07 coral heads. Theseeffects could not account for the large relative sea level shift in eleva-tion between the modern and dead coral colonies.

7.2. Interseismic and postseismic deformation from the microatolls

During the past decade, new knowledge has been gained about largemegathrust earthquakes such as the 2004 Sumatra-Andaman, the 2010Maule and the 2011 Tohoku-oki (e.g., Sun et al., 2014; Sun and Wang,2015; Wang and Tréhu, 2016; Bedford et al., 2016). During large sub-duction earthquakes, the frontal part of the upper plate moves tren-chward and upwards releasing the stored strain while the downdip endof the earthquake subsides (Wang and Tréhu, 2016) (Fig. 13). After theearthquake, the postseismic motion is initially dominated by afterslipand viscous relaxation. While this motion is often in the same directionas the earthquake (e.g., Hsu et al., 2006), it can also have the oppositesense, as for example, the offshore area of the 2011 Tohoku-okiearthquake (e.g., Sun et al., 2014). This motion reflects deformationdriven by the stresses from the earthquake. After a period of time that isdependent on the magnitude of the earthquake, the postseismic motionfades and the interseismic motion dominates. This is primarily elasticloading due to the dragging of the upper plate, which is locked to the

Table 4U/Th ages reveal three major coral killing events (separated by horizontallines).

Uncorrected CE Corrected for 230Th/232Th initial

SM-C05 1734 ± 1 1780 ± 24 CESM-A05 1773 ± 1 1784 ± 7 CESM-D09 1750 ± 2 1774 ± 16 CESM16-C1 1743 ± 3 1767 ± 16 CESM16-D1 1747 ± 1 1760 ± 8 CE

SM-C10 1147 ± 4 1125 ± 47 CESM-C11 1115 ± 10 1201 ± 55 CESM16-D3 979 ± 4 995 ± 9CE

SM-C16 630 ± 5 659 ± 52 CE


208

lower plate. As a result of interseismic viscoelastic stress relaxation(Wang et al., 2012), the strain rate is not completely linear with timeand the motion does not precisely match elastic backslip models(Savage, 1983; Okada, 1985). Interseismic motion has the oppositesense of the earthquake, with the updip part of the locked fault movinglandward and downwards.

In order to examine the subduction earthquake cycle related to the1762 earthquake, we compared the annual growth bands in SM-A05that died as a result of the 1762 earthquake with those of SM-Q07 thatgrew after the 1762 earthquake. The geometry of the growth bandsprovides a tool for tracking relative sea level change, which can aid inreconstructing the interseismic vertical deformation. The two coral re-cords combined provide a nearly continuous record from ~1650 CE topresent with 67 years missing right after the 1762 earthquake. In orderto obtain an accurate estimate of the tectonic deformation, relative sealevel must be first corrected for eustatic sea level change (Meltzneret al., 2015). The rate of sea level change has varied over the lastmillennium. Since the Little Ice Age, there has been an increase in therate of sea level rise (Kemp et al., 2011). We have taken the rates es-timated by Kemp et al. (2011) (Fig. 3) and subtracted those from theHLS change rate estimated from the coral microatolls (SM-A05 and SM-Q07) for corresponding time (Fig. 12). Rates were then calculated byapplying a linear fit to the growth bands that represent the highest levelof survival (HLS) with the correction for eustatic rise of sea level. Thegrowth bands suggest 2.2mm/year subsidence and 1.8 mm/year ofuplift before and after the earthquake, respectively (Fig. 12B).

As the change in eustatic sea level rise was minimal during the LittleIce Age, microatoll SM-A05 tracks a relative sea level rise of about15 cm over 71 years (Fig. 10C). This slow rate of change of sea levelsuggests a slow interseismic subsidence at a rate of ~2.2mm/yr (1648to 1762 CE) prior to the 1762 earthquake, similar to the uncorrectedrate. On the other hand, the slow relative sea level rise of only 0.3 mm/yr recorded in microatoll SM-Q07 (Fig. 11) is slower than the rate of theeustatic rise over the last 200 years. The very slow relative sea level risein this coral thus suggests an uplift rate of 1.8 mm/yr (1850 to2016 CE).

The slab sampled from coral head SM-Q07 provides the HLS recordfrom 1805 to 2016 CE and implies that the immediate post-earthquakedeformation following the 1762 event was not recorded. Postseismicdeformation generally lasts from years to several decades. However, thedeformation was not recorded by SM-Q07 until 70 years after theearthquake and appears to have been relatively constant over the life-span of this coral. Thus, the postseismic deformation is not recorded.

The interseismic deformation derived from HLS data in microatollSM-Q07 might be expected to be a continuation of subsidence, similarto the interseismic deformation recorded by microatoll SM-A05.However, the slow lateral growth of the microatoll during an interval ofincreasing eustatic sea level indicates that it may have undergone tec-tonic uplift that is nearly equal to the rate of sea level rise. We suggestthat a continuation of uplift, post-1762, may be due to aseismic de-formation of the Saint Martin's anticline. It is unclear whether theshallow detachment that underlies the frontal part of the Indo-Burmasubduction zone is seismogenic (Steckler et al., 2016, 2008). Even if it isseismogenic, it is possible that the megathrust may not rupture theupdip region (see scenarios in Wang et al., 2013). One possible inter-pretation is that the rupture stopped at or near the Saint Martin's an-ticline, contributing to the coseismic uplift that occurred there. Thedeformation and stress at the front of the earthquake rupture may thendrive aseismic folding and uplift of the anticline. This deformation ofthe anticline overlying the megathrust would propagate the deforma-tion aseismically into the unruptured outer part of the foldbelt fol-lowing the 1762 earthquake.

The Indo-Burma subduction zone (Fig. 2) differs other subductionzones in the extreme thickness and volume of sediments on the in-coming plate (Singh et al., 2016) and the 250-km wide accretionaryprism with numerous anticlinal folds (Steckler et al., 2008). The

Fig. 13. Schematic of the viscoelastic deformation observed from the coralmicroatolls A). Interseismic deformation prior to the 1762 earthquake, whichmatches the subsidence recorded by microatoll SM-A05. B) Coseismic andpostseismic deformation during and after the earthquake. The net uplift of SaintMartin's Island is recorded from the ages and elevations of coral heads. C)Present day postseismic uplift recorded in coral head SM-Q07, which we in-terpret as due to aseismic folding. D) Future interseismic deformation caused byelastic loading.


209

extremely low surface slope (0.1°) implies that the prism and themegathrust detachment are extremely weak. Mapping and structuralanalysis of the anticlines in the Indo-Burma foldbelt reveal a mixture offault propagation folds and detachment folds over the relativelyshallow décollement (Betka et al., 2017). The detachment folds maywell develop relatively aseismically. This uplift at Saint Martin's Islandmay be similar to the afterslip observed following the Nias-Simeulue2005 earthquake when significant postseismic slip updip of the M8.7rupture was observed (Hsu et al., 2006b). However, the continuation ofthe uplift at Saint Martin's Island for an extreme amount of time,≥250 years, in this extreme accretionary prism suggest that there maybe a considerable aseismic deformation of the nascent folds near thedeformation front. The outer part of the accretionary prism may beadvancing westward ≥25 km/Myr (Maurin and Rangin, 2009; Betkaet al., 2017). Whether the buried nascent folds at the front of the ra-pidly growing accretionary prism are capable of seismic rupture is notknown. If the frontal part of the accretionary prism is able to deformaseismically, even in part, it could greatly decrease the seismic hazardfor Dhaka. This megacity, the capital of Bangladesh, lies near the de-formation front in the next segment of the megathrust to the north.

7.3. Reconstruction of coseismic and postseismic uplift

The reconstruction of coseismic uplift associated with an ancientearthquake is problematic due to a lack of constraints. The oldestdated coral at Saint Martin's Island is ~700 CE and its present-dayelevation suggests it may record an ancient earthquake event.However, we have limited information on the interseismic deforma-tion after that earthquake. Considering fast postseismic relaxationfollowed by interseismic subsidence at the rate of 1–5mm/yr up to thefollowing earthquake, the net sum of the coseismic and postseismicuplift could range from 0.5 m to 2.0 m. (Fig. 12). The distribution ofcoral ages also suggests an earthquake at around 1140 CE. The netuplift from the coseismic and postseismic motion from that event at1140 CE can be estimated using the deformation recorded in the coralhead SM-A05 where we estimated interseismic subsidence at the rateof 2 mm/yr. Extrapolating the subsidence rate back to 1140 CE gen-erates a net uplift of ~1 m. The present elevation difference betweenthe dead coral SM-A05 from before the 1762 earthquake and live coralSM-Q07 measured from the GPS is 2.4 m. Considering postseismicdeformation followed by slow uplift at the rate of 1.8 mm/yr, thecoseismic uplift of the 1762 earthquake could range from 1.8 to 2.0 m(Fig. 12). It is important to note that the pattern of immediate post-seismic deformation for these earthquakes is not recorded in any coralheads and thus it is not known. We can only, at best, estimate the sumof the coseismic and postseismic motion.

7.4. Recurrence interval

U-Th dating of Saint Martin Island corals indicates that the mostlikely cause of coral mortality has been earthquakes. The probabilitydistribution of ages suggests three coral-killing events. The most recentof these corresponds to the 1762 Arakan earthquake. The other eventscorrespond to 1126–1161, and 620–750 CE. (Fig. 12). The distributionof ages suggests a recurrence interval of 500–700 years, similar to thatestimated by Wang et al. (2013) farther south. One caveat is that thedistribution of earlier events is not well constrained with our currentsampling and dating. Nor can we determine if these earlier earthquakesmay have only broken part of the zone that ruptured in 1762. However,two additional ages of 984 to 1270 CE, obtained from coral headsSM16-D2 and SM16-D3 in the northern part of the island, correspond to986–1193 CE earthquake event. Therefore, a 500–700 year recurrenceinterval is not an unreasonable preliminary estimate, given the limita-tions of the data set.

7.5. Implication of future hazards

Estimated uplift at 1 to 2mm/yr documented in this study for SaintMartin's Island and previously for Ramree and Cheduba Islands (Wanget al., 2013) show that strain is accumulating in this region, whichsuggests that the subduction boundary could fail again in the future andgenerate another large earthquake. However, whether the whole seg-ment will break or whether it would break in smaller patches remainsundetermined, at least with available data. If we assume that a totalaccumulated slip/strain will be released during one earthquake, con-sidering the time and the accumulated slip, this 500 km long and150 km wide fault is capable of causing an earthquake as large as Mw8.0–8.7M earthquake (Hanks and Kanamori, 1979). Using the em-pirical relation between fault length and magnitude (Wells andCoppersmith, 1994), this fault has the potential for generating a Mw 7.9earthquake. An earthquake of magnitude 8.0–8.7 would obviouslyprove devastating to neighboring cities and populations.

The current uplift of Saint Martin's Island suggests that at least someof the deformation of the frontal part of the accretionary prism may beoccurring aseismically. We cannot tell if this frontal zone does notrupture in large earthquakes, or only in some earthquakes as was thecase for Tohoku. A better understanding of the deformation modes nearthe deformation front is critical for the megacity of Dhaka, which liesastride the deformation front in the next segment to the north.

8. Conclusions

Porites coral microatolls in Saint Martin's Island at southernmostBangladesh provide evidence of past earthquakes along the Chittagong-Arakan coast. U-Th ages reveal the timing of a massive coral microatollkilling, corresponding to the 1762 Arakan earthquake. Our GPS surveyindicates all dead corals are found at elevations ~2m above livingcolonies. The flat-topped microatolls suggest that the uplift associatedwith the Arakan earthquake was rapid. Previously, the evidence ofrupture associated with the 1762 event has only been documented fromthe west coast of Myanmar (Aung et al., 2007, 2008; Wang et al., 2013),located over 100 km to the south of our study area. Thus, our studyprovides strong evidence that deformation was related to the Arakanearthquake.

Subduction seismic cycle deformation is recorded from pre and post-earthquake corals. Annual banding in a coral growing prior to 1762suggests an interseismic subsidence at ~2mm/y, consistent with theelastic loading of the fault. The post-earthquake coral record begins~70 years after the earthquake when the new corals reached HLS andthus did not record immediate post-earthquake deformation. The sub-sequent motion, when corrected for sea level rise, shows uplift at a rateof 1.8 mm/yr. We suggest that this may be due to the aseismic foldingof the Saint Martin's anticline following loading by the 1762 earth-quake. This raises the question of how much of the deformation in thefrontal part of the accretionary prism may be absorbed aseismically.

Our study provides evidence for two other earthquakes in ~700 andin ~1140 CE. If so, the recurrence interval of large earthquakes is onthe order of 500–700 years. This interval is consistent with an estimateby from the coral and terrace study along the western coast ofMyanmar.

The evidence of interseismic subsidence, coseismic uplift and theongoing deformation suggest that subduction zone is storing strain.While time since the last earthquake is still significantly less than theestimated recurrence interval, when the next earthquake occurs, thestored energy could be sufficient to cause an earthquake of Mw ≥8.

Acknowledgments

This work was supported by the National Science Foundation grantOISE 09-68354; Office of Naval Research grant ONR N00014-11-1-0683; Queens College, City University of New York; the Gural


210

Foundation Scholarship; and a Graduate Center Dissertation Fellowshipof the City University of New York. We would like to thank our re-viewers for improving the manuscript. We also would like to thankPeter Knappett from Texas A&M University for the use of his geodeticequipment; Pritam Shaha, Masud Iqbal, and Alamgir Hossain fromDhaka University for assistance with field work, and Nicole Abdul fromRutgers University for the assistance with the U/Th analysis. Lamont-Doherty Earth Observatory publication number 8241.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.tecto.2018.07.020.

References

Abdul, N.A., Mortlock, R.A., Wright, J.D., Fairbanks, R.G., 2016. Younger Dryas sea leveland meltwater pulse 1B recorded in Barbados reef crest coral Acropora palmata.Paleoceanography 31https://doi.org/10.1002/2015PA002847. 2015PA002847.

Abram, N.J., Gagan, M.K., McCulloch, M.T., Chappell, J., Hantoro, W.S., 2003. Coral reefdeath during the 1997 Indian ocean dipole linked to Indonesian wildfires. Science301, 952–955. https://doi.org/10.1126/science.1083841.

Abram, N.J., Gagan, M.K., Liu, Z., Hantoro, W.S., McCulloch, M.T., Suwargadi, B.W.,2007. Seasonal characteristics of the Indian Ocean Dipole during the Holocene epoch.Nature 445, 299–302. https://doi.org/10.1038/nature05477.

Akhter, S.H., 1979. Structure, Stratigraphy and Sedimentology of the Upper TertiarySediments of the Central Part of the Sitakund Hill Range, Chittagong, Bangladesh.M.Sc thesis. Geology Department, University of Dhaka, Dhaka, Bangladesh (143p).

Akhter, S.H., 2010. Earthquakes of Dhaka. In: Islam, M.A. (Ed.), Environment of CapitalDhaka - Plants Wildlife Gardens Parks Air Water and Earthquake. Celebration SeriesAsiatic Society of Bangladesh, pp. 401–426.

Alam, E., Dominey-Howes, D., 2014. An analysis of the AD1762 earthquake and tsunamiin SE Bangladesh. Nat. Hazards 70, 903–933. https://doi.org/10.1007/s11069-013-0841-5.

Alam, M., Alam, M.M., Curray, J.R., Chowdhury, M.L.R., Gani, M.R., 2003. An overviewof the sedimentary geology of the Bengal Basin in relation to the regional tectonicframework and basin-fill history. Sediment. Geol. 155, 179–208. https://doi.org/10.1016/S0037-0738(02)00180-X.

Ambikapathy, A., Catherine, J.K., Gahalaut, V.K., Narsaiah, M., Bansal, A., Mahesh, P.,2010. The 2007 Bengkulu earthquake, its rupture model and implications for seismichazard. J. Earth Syst. Sci. 119, 553–560. https://doi.org/10.1007/s12040-010-0037-2.

Aung, T., Satake, K., Okamura, Y., Shishikura, M., Swe, W., Saw, H., Swe, T., Tun, S.,Aung, T., 2007. Paleoseismoloigcal evidence of the 1762 and earlier earthquakes offMyanmar. In: AGU Fall Meet. Abstr. 24. pp. 6.

Aung, T.T., Satake, K., Okamura, Y., Shishikura, M., Swe, W., Saw, H., Swe, T.L., Tun,S.T., Aung, T., 2008. Geologic evidence for three great earthquakes in the past 3400years off Myanmar. J. Earthq. Tsunami 2, 259–265. https://doi.org/10.1142/S1793431108000335.

Banerjee, P., Pollitz, F., Nagarajan, B., Bürgmann, R., 2007. Coseismic slip distributions ofthe 26 December 2004 Sumatra–Andaman and 28 March 2005 Nias Earthquakesfrom gps Static Offsets. Bull. Seismol. Soc. Am. 97, S86–S102. https://doi.org/10.1785/0120050609.

Barua, D.K., 1990. Suspended sediment movement in the estuary of the Ganges-Brahmaputra-Meghna river system. Mar. Geol. 91, 243–253. https://doi.org/10.1016/0025-3227(90)90039-M.

Bastas-Hernandez, A., McHugh, C.M., Mondal, D.R., Seeber, L., Steckler, M.S., Marsh, J.,Gurung, D., Mustaque, S., Akhter, S.H., 2014. Sedimentary record of paleodeforma-tion of the Saint Martin anticline reveals the interaction between tectonics, sedi-mentation processes and relative sea-level changes: Ganges-Brahmaputra DeltaBurma Arc collision, SE Bangladesh. In: AGU Fall Meet. Abstr.

Bedford, J., Moreno, M., Li, S., Oncken, O., Baez, J.C., Bevis, M., Heidbach, O., Lange, D.,2016. Separating rapid relocking, afterslip, and viscoelastic relaxation: an applicationof the postseismic straightening method to the Maule 2010 cGPS. J. Geophys. Res.Solid Earth 121https://doi.org/10.1002/2016JB013093. 2016JB013093.

Betka, P., Sebber, L., Thomson, S., Steckler, M., 2017. Shallow, Weak Megathrust BeneathIndoBurman Accretionary Prism. (submitted).

Briggs, R.W., Sieh, K., Meltzner, A.J., Natawidjaja, D., Galetzka, J., Suwargadi, B., Hsu, Y.,Simons, M., Hananto, N., Suprihanto, I., 2006. Deformation and slip along the Sundamegathrust in the great 2005 Nias-Simeulue earthquake. Science 311, 1897–1901.

Buddemeier, R.W., Maragos, J.E., Knutson, D.W., 1974. Radiographic studies of reef coralexoskeletons: rates and patterns of coral growth. J. Exp. Mar. Biol. Ecol. 14, 179–199.https://doi.org/10.1016/0022-0981(74)90024-0.

Burbank, D.W., Anderson, R.S., 2011. Rates of erosion and uplift. In: TectonicGeomorphology. John Wiley & Sons, Ltd, pp. 195–242.

Burbank, D.W., Blythe, A.E., Putkonen, J., Pratt-Sitaula, B., Gabet, E., Oskin, M., Barros,A., Ojha, T.P., 2003. Decoupling of erosion and precipitation in the Himalayas.Nature 426, 652–655. https://doi.org/10.1038/nature02187.

Calais, E., Dong, L., Wang, M., Shen, Z., Vergnolle, M., 2006. Continental deformation inAsia from a combined GPS solution. Geophys. Res. Lett. 33, L24319. https://doi.org/10.1029/2006GL028433.

Charles, C.D., Hunter, D.E., Fairbanks, R.G., 1997. Interaction between the ENSO and theAsian monsoon in a coral record of tropical climate. Science 277, 925–928. https://doi.org/10.1126/science.277.5328.925.

Cheng, H., Edwards, R.L., Hoff, J., Gallup, C.D., Richards, D.A., Asmerom, Y., 2000. Thehalf-lives of uranium-234 and thorium-230. Chem. Geol. 169, 17–33. https://doi.org/10.1016/S0009-2541(99)00157-6.

Cheng, H., Edwards, R.L., Shen, C.C., Polyak, V.J., Asmerom, Y., Woodhead, J.,Hellstrom, J., Wang, Y., Kong, X., Spötl, C., Wang, X., 2013. Improvements in 230Thdating, 230Th and 234U half-life values, and U–Th isotopic measurements by multi-collector inductively coupled plasma mass spectrometry. Earth Planet. Sci. Lett. 371,82–91.

Chiu, T., Fairbanks, R.G., Mortlock, R.A., Bloom, A.L., 2005. Extending the radiocarboncalibration beyond 26,000 years before present using fossil corals. Quat. Sci. Rev. 24,1797–1808. https://doi.org/10.1016/j.quascirev.2005.04.002.

Chowdhury, S.Q., Fazlul, A.T.M., Hasan, H.K., 1997. Beachrock in St. Martin's Island,Bangladesh: implications of sea level changes on beachrock cementation. Mar. Geod.20, 89–104. https://doi.org/10.1080/01490419709388097.

Clarke, A.J., Liu, X., 1994. Interannual Sea Level in the Northern and Eastern IndianOcean. J. Phys. Oceanogr. 24, 1224–1235. https://doi.org/10.1175/1520-0485(1994)024<1224:ISLITN>2.0.CO;2.

Cobb, K.M., Charles, C.D., Cheng, H., Kastner, M., Edwards, R.L., 2003. U/Th-datingliving and young fossil corals from the central tropical Pacific. Earth Planet. Sci. Lett.210, 91–103. https://doi.org/10.1016/S0012-821X(03)00138-9.

Cummins, P.R., 2007. The potential for giant tsunamigenic earthquakes in the northernBay of Bengal. Nature 449, 75–78. https://doi.org/10.1038/nature06088.

Curray, J.R., 2014. The Bengal Depositional System: from rift to orogeny. Mar. Geol.https://doi.org/10.1016/j.margeo.2014.02.001.

Dewey, J.F., Cande, S., Pitman, W.C., 1989. Tectonic evolution of the India/Eurasiacollision zone. Eclogae Geol. Helv. 82, 717–734.

Edwards, R.L., Taylor, F.W., Wasserburg, G.J., 1988. Dating earthquakes with high-pre-cision thorium-230 ages of very young corals. Earth Planet. Sci. Lett. 90 (4), 371–381.

Gabet, E.J., Burbank, D.W., Pratt-Sitaula, B., Putkonen, J., Bookhagen, B., 2008. Modernerosion rates in the High Himalayas of Nepal. Earth Planet. Sci. Lett. 267, 482–494.https://doi.org/10.1016/j.epsl.2007.11.059.

Galy, A., France-Lanord, C., 2001. Higher erosion rates in the Himalaya: geochemicalconstraints on riverine fluxes. Geology 29, 23–26. https://doi.org/10.1130/0091-7613(2001)029<0023:HERITH>2.0.CO;2.

Grall, C., Steckler, M.S., Pickering, J.L., Goodbred, S., Sincavage, R., Paola, C., Akhter,S.H., Spiess, V., 2018. A base-level stratigraphic approach to determining Holocenesubsidence of the Ganges–Meghna–Brahmaputra Delta plain. Earth Planet. Sci. Lett.499, 23–36.

Guidoboni, E., Stucchi, M., 1993. The contribution of historical records of earthquakes tothe evaluation of seismic hazard. Ann. Geophys. 36. https://doi.org/10.4401/ag-4264.

Gulston, E., 1763. An account of an earthquake at Chattigaon: translated from the Persianby Mr. Edward Gulston, in the service of the Honourable East India Company, andcommunicated by him to the Reverend Mr. Hirst. Philos. Trans. 53, 251–256.

Halsted, E., 1843. Report on the Island of Cheduba. J. Asiatic Soc. Bengal 43–46.Hanks, T.C., Kanamori, H., 1979. A moment magnitude scale. J. Geophys. Res. Solid Earth

84, 2348–2350. https://doi.org/10.1029/JB084iB05p02348.Hossain, M.M., Islam, M.H., 2006. Status of the biodiversity of St. Martin's Island, Bay of

Bengal, Bangladesh. Pakistan J. Mar. Sci. 15 (2), 201–210.Howe, M., Kim, W.Y., Akhter, S.H., Mondal, D.R., Seeber, L., Steckler, M.S., Crustal

Velocity Structure of the Bengal Basin from Regionally Recorded Earthquakes, (inpreparation).

Hsu, Y.-J., Simons, M., Avouac, J.-P., Galetzka, J., Sieh, K., Chlieh, M., Natawidjaja, D.,Prawirodirdjo, L., Bock, Y., 2006. Frictional afterslip following the 2005 Nias-Simeulue Earthquake, Sumatra. Science 312, 1921–1926. https://doi.org/10.1126/science.1126960.

Hughes, T., Szmant, A.M., Steneck, R., Carpenter, R., Miller, S., 1999. Algal blooms oncoral reefs: what are the causes? Limnol. Oceanogr. 44, 1583–1586. https://doi.org/10.4319/lo.1999.44.6.1583.

Johnson, S.Y., Alam, A.M.N., 1991. Sedimentation and tectonics of the Sylhet trough,Bangladesh. Geol. Soc. Am. Bull. 103, 1513–1527. https://doi.org/10.1130/0016-7606(1991)103<1513:SATOTS>2.3.CO;2.

Kemp, A.C., Horton, B.P., Donnelly, J.P., Mann, M.E., Vermeer, M., Rahmstorf, S., 2011.Climate related sea-level variations over the past two millennia. Proc. Natl. Acad. Sci.108, 11017–11022. https://doi.org/10.1073/pnas.1015619108.

Konca, A.O., Avouac, J.-P., Sladen, A., Meltzner, A.J., Sieh, K., Fang, P., Li, Z., Galetzka,J., Genrich, J., Chlieh, M., Natawidjaja, D.H., Bock, Y., Fielding, E.J., Ji, C.,Helmberger, D.V., 2008. Partial rupture of a locked patch of the Sumatra megathrustduring the 2007 earthquake sequence. Nature 456, 631–635. https://doi.org/10.1038/nature07572.

Kuehl, S.A., Hariu, T.M., Moore, W.S., 1989. Shelf sedimentation off the Ganges-Brahmaputra river system: evidence for sediment bypassing to the Bengal fan.Geology 17, 1132. https://doi.org/10.1130/0091-7613(1989)017<1132:SSOTGB>2.3.CO;2.

Kuehl, S.A., Allison, M.A., Goodbred, S.L., Kudrass, H., 2005. The Ganges-BrahmaputraDelta. 83. Spec. Publ.-SEPM, pp. 413.

Lapointe, B.E., 1997. Nutrient thresholds for bottom-up control of macroalgal blooms oncoral reefs in Jamaica and Southeast Florida. Limnol. Oceanogr. 42, 1119–1131.

Lapointe, B.E., 1999. Simultaneous top-down and bottom-up forces control macroalgalblooms on coral reefs (Reply to the comment by Hughes et al.). Limnol. Oceanogr. 44,1586–1592. https://doi.org/10.4319/lo.1999.44.6.1586.

Lavé, J., Avouac, J.P., 2001. Fluvial incision and tectonic uplift across the Himalayas ofCentral Nepal. J. Geophys. Res. Solid Earth 106, 26561–26591. https://doi.org/10.1029/2001JB000359.

Lough, J.M., Barnes, D.J., 2000. Environmental controls on growth of the massive coralPorites. J. Exp. Mar. Biol. Ecol. 245, 225–243. https://doi.org/10.1016/S0022-0981(99)00168-9.

Lough, J.M., Cooper, T.F., 2011. New insights from coral growth band studies in an era ofrapid environmental change. Earth-Sci. Rev. 108, 170–184. https://doi.org/10.1016/j.earscirev.2011.07.001.


211



https://doi.org/10.1002/2015PA002847

https://doi.org/10.1126/science.1083841

https://doi.org/10.1038/nature05477

http://refhub.elsevier.com/S0040-1951(18)30263-4/rf0020






https://doi.org/10.1007/s11069-013-0841-5

https://doi.org/10.1007/s11069-013-0841-5

https://doi.org/10.1016/S0037-0738(02)00180-X

https://doi.org/10.1016/S0037-0738(02)00180-X

https://doi.org/10.1007/s12040-010-0037-2

https://doi.org/10.1007/s12040-010-0037-2




https://doi.org/10.1142/S1793431108000335

https://doi.org/10.1142/S1793431108000335

https://doi.org/10.1785/0120050609

https://doi.org/10.1785/0120050609

https://doi.org/10.1016/0025-3227(90)90039-M

https://doi.org/10.1016/0025-3227(90)90039-M






https://doi.org/10.1002/2016JB013093






https://doi.org/10.1016/0022-0981(74)90024-0




https://doi.org/10.1029/2006GL028433

https://doi.org/10.1029/2006GL028433

https://doi.org/10.1126/science.277.5328.925

https://doi.org/10.1126/science.277.5328.925

https://doi.org/10.1016/S0009-2541(99)00157-6

https://doi.org/10.1016/S0009-2541(99)00157-6






https://doi.org/10.1016/j.quascirev.2005.04.002

https://doi.org/10.1080/01490419709388097

https://doi.org/10.1175/1520-0485(1994)024<1224:ISLITN>2.0.CO;2

https://doi.org/10.1175/1520-0485(1994)024<1224:ISLITN>2.0.CO;2

https://doi.org/10.1016/S0012-821X(03)00138-9


https://doi.org/10.1016/j.margeo.2014.02.001





https://doi.org/10.1016/j.epsl.2007.11.059

https://doi.org/10.1130/0091-7613(2001)029<0023:HERITH>2.0.CO;2

https://doi.org/10.1130/0091-7613(2001)029<0023:HERITH>2.0.CO;2





https://doi.org/10.4401/ag-4264

https://doi.org/10.4401/ag-4264





https://doi.org/10.1029/JB084iB05p02348





https://doi.org/10.4319/lo.1999.44.6.1583

https://doi.org/10.4319/lo.1999.44.6.1583

https://doi.org/10.1130/0016-7606(1991)103<1513:SATOTS>2.3.CO;2

https://doi.org/10.1130/0016-7606(1991)103<1513:SATOTS>2.3.CO;2

https://doi.org/10.1073/pnas.1015619108



https://doi.org/10.1130/0091-7613(1989)017<1132:SSOTGB>2.3.CO;2

https://doi.org/10.1130/0091-7613(1989)017<1132:SSOTGB>2.3.CO;2





https://doi.org/10.4319/lo.1999.44.6.1586

https://doi.org/10.1029/2001JB000359

https://doi.org/10.1029/2001JB000359

https://doi.org/10.1016/S0022-0981(99)00168-9

https://doi.org/10.1016/S0022-0981(99)00168-9

https://doi.org/10.1016/j.earscirev.2011.07.001

https://doi.org/10.1016/j.earscirev.2011.07.001

Mallet, F., 1878. Mud volcanoes of Ramri and Chedooba (No. 11). Geol. Surv. of India 11,188–207.

Mandal, B.C., Woobaidullah, A.S.M., 2006. Sedimentary Tectonics of the Eastern FoldBelt of the Bengal Basin, Bangladesh. Geol. Soc. India 68, 115–128.

Maurin, T., Rangin, C., 2009. Structure and kinematics of the Indo-Burmese Wedge: re-cent and fast growth of the outer wedge. Tectonics 28 (2).

Maurin, T., Masson, F., Rangin, C., Min, U.T., Collard, P., 2010. First global positioningsystem results in northern Myanmar: constant and localized slip rate along theSagaing fault. Geology 38, 591–594. https://doi.org/10.1130/G30872.1.

Meltzner, A.J., Woodroffe, C.D., 2015. Coral microatolls. In: Shennan, I., Long, A.J.,Horton, B.P. (Eds.), Handbook of Sea-Level Research. John Wiley & Sons, Ltd, pp.125–145.

Meltzner, A.J., Sieh, K., Chiang, H.-W., Shen, C.-C., Suwargadi, B.W., Natawidjaja, D.H.,Philibosian, B.E., Briggs, R.W., Galetzka, J., 2010. Coral evidence for earthquakerecurrence and an A.D. 1390–1455 cluster at the south end of the 2004Aceh–Andaman rupture. J. Geophys. Res. 115. https://doi.org/10.1029/2010JB007499.

Meltzner, A.J., Sieh, K., Chiang, H.-W., Wu, C.-C., Tsang, L.L.H., Shen, C.-C., Hill, E.M.,Suwargadi, B.W., Natawidjaja, D.H., Philibosian, B., Briggs, R.W., 2015. Time-varying interseismic strain rates and similar seismic ruptures on the Nias–Simeuluepatch of the Sunda megathrust. Quat. Sci. Rev. 122, 258–281. https://doi.org/10.1016/j.quascirev.2015.06.003.

Menard, H.W., 1961. Some rates of regional erosion. J. Geol. 69, 154–161.Métivier, F., Gaudemer, Y., Tapponnier, P., Klein, M., 1999. Mass accumulation rates in

Asia during the Cenozoic. Geophys. J. Int. 137, 280–318. https://doi.org/10.1046/j.1365-246X.1999.00802.x.

Meyssignac, B., Cazenave, A., 2012. Sea level: a review of present-day and recent-pastchanges and variability. J. Geodyn. 58, 96–109. https://doi.org/10.1016/j.jog.2012.03.005.

Meyssignac, B., Becker, M., Llovel, W., Cazenave, A., 2012. An assessment of two-di-mensional past sea level reconstructions over 1950–2009 based on tide-gauge dataand different input sea level grids. Surv. Geophys. 33, 945–972. https://doi.org/10.1007/s10712-011-9171-x.

Michel, G.W., Yu, Y.Q., Zhu, S.Y., Reigber, C., Becker, M., Reinhart, E., Simons, W.,Ambrosius, B., Vigny, C., Chamot-Rooke, N., Le Pichon, X., Morgan, P., Matheussen,S., 2001. Crustal motion and block behaviour in SE-Asia from GPS measurements.Earth Planet. Sci. Lett. 187, 239–244. https://doi.org/10.1016/S0012-821X(01)00298-9.

Milne, G.A., Mitrovica, J.X., 2008. Searching for eustasy in deglacial sea-level histories.Quat. Sci. Rev. 27, 2292–2302. https://doi.org/10.1016/j.quascirev.2008.08.018.

Molnar, P., Tapponnier, P., 1975. Cenozoic tectonics of Asia: effects of a continentalcollision. Science 189, 419–426.

Mortlock, R.A., Fairbanks, R.G., Chiu, T., Rubenstone, J., 2005. 230Th/234U/238U and231Pa/235U ages from a single fossil coral fragment by multi-collector magnetic-sectorinductively coupled plasma mass spectrometry. Geochim. Cosmochim. Acta 69,649–657. https://doi.org/10.1016/j.gca.2004.06.033.

Najman, Y., Allen, R., Willett, E.A.F., Carter, A., Barfod, D., Garzanti, E., Wijbrans, J.,Bickle, M.J., Vezzoli, G., Ando, S., Oliver, G., Uddin, M.J., 2012. The record ofHimalayan erosion preserved in the sedimentary rocks of the Hatia Trough of theBengal Basin and the Chittagong Hill Tracts, Bangladesh. Basin Res. 24, 499–519.https://doi.org/10.1111/j.1365-2117.2011.00540.x.

Natawidjaja, D.H., 2004. Paleogeodetic records of seismic and aseismic subduction fromcentral Sumatran microatolls, Indonesia. J. Geophys. Res. 109. https://doi.org/10.1029/2003JB002398.

Natawidjaja, D.H., Sieh, K., Chlieh, M., Galetzka, J., Suwargadi, B.W., Cheng, H.,Edwards, R.L., Avouac, J.-P., Ward, S.N., 2006. Source parameters of the greatSumatran megathrust earthquakes of 1797 and 1833 inferred from coral microatolls.J. Geophys. Res. Solid Earth 111, B06403. https://doi.org/10.1029/2005JB004025.

Natawidjaja, D.H., Sieh, K., Galetzka, J., Suwargadi, B.W., Cheng, H., Edwards, R.L.,Chlieh, M., 2007. Interseismic deformation above the Sunda Megathrust recorded incoral microatolls of the Mentawai islands, West Sumatra. J. Geophys. Res. Solid Earth112, B02404. https://doi.org/10.1029/2006JB004450.

Ni, J., Barazangi, M., 1984. Seismotectonics of the Himalayan Collision Zone: geometry ofthe underthrusting Indian Plate beneath the Himalaya. J. Geophys. Res. Solid Earth89, 1147–1163. https://doi.org/10.1029/JB089iB02p01147.

Ni, J.F., Guzman-Speziale, M., Bevis, M., Holt, W.E., Wallace, T.C., Seager, W.R., 1989.Accretionary tectonics of Burma and the three-dimensional geometry of the Burmasubduction zone. Geology 17, 68–71. https://doi.org/10.1130/0091-7613(1989)017<0068:ATOBAT>2.3.CO;2.

Nielsen, C., Chamot-Rooke, N., Rangin, C., 2004. From partial to full strain partitioningalong the Indo-Burmese hyper-oblique subduction. Mar. Geol. 209, 303–327. https://doi.org/10.1016/j.margeo.2004.05.001.

Okada, Y., 1985. Surface deformation due to shear and tensile faults in a half-space. Bull.Seismol. Soc. Am. 75, 1135–1154.

Peltier, W.R., 2002. Global glacial isostatic adjustment: palaeogeodetic and space-geo-detic tests of the ICE-4G (VM2) model. J. Quat. Sci. 17, 491–510. https://doi.org/10.1002/jqs.713.

Peltier, W.R., Fairbanks, R.G., 2006. Global glacial ice volume and Last Glacial Maximumduration from an extended Barbados sea level record. Quat. Sci. Rev. 25, 3322–3337.

Philibosian, B., Sieh, K., Avouac, J.-P., Natawidjaja, D.H., Chiang, H.-W., Wu, C.-C.,Perfettini, H., Shen, C.-C., Daryono, M.R., Suwargadi, B.W., 2014. Rupture andvariable coupling behavior of the Mentawai segment of the Sunda megathrust duringthe supercycle culmination of 1797 to 1833: 1797/1833 Mentawai RuptureSequence. J. Geophys. Res. Solid Earth 119, 7258–7287. https://doi.org/10.1002/2014JB011200.

Powell, C.M., Conaghan, P.J., 1973. Plate tectonics and the Himalayas. Earth Planet. Sci.Lett. 20, 1–12. https://doi.org/10.1016/0012-821X(73)90134-9.

Rangin, C., Maurin, T., Masson, F., 2013. Combined effects of Eurasia/Sunda obliqueconvergence and East-Tibetan crustal flow on the active tectonics of Burma. J. AsianEarth Sci. 76, 185–194. https://doi.org/10.1016/j.jseaes.2013.05.018.

Rao, N.P., Kumar, M.R., 1999. Evidences for cessation of Indian Plate Subduction in theBurmese Arc Region. Geophys. Res. Lett. 26, 3149–3152. https://doi.org/10.1029/1999GL005396.

Richards, D.A., Dorale, J.A., 2003. Uranium-series chronology and environmental ap-plications of speleothems. Rev. Mineral. Geochem. 52, 407–460. https://doi.org/10.2113/0520407.

Rovere, A., Stocchi, P., Vacchi, M., 2016. Eustatic and relative sea level changes. Curr.Clim. Change Rep. 2, 221–231. https://doi.org/10.1007/s40641-016-0045-7.

Saji, N.H., Goswami, B.N., Vinayachandran, P.N., Yamagata, T., 1999. A dipole mode inthe tropical Indian Ocean. Nature 401, 360–363. https://doi.org/10.1038/43854.

Satyabala, S.P., 2003. Oblique plate convergence in the Indo-Burma (Myanmar) sub-duction region. Pure Appl. Geophys. 160, 1611–1650.

Savage, J.C., 1983. A dislocation model of strain accumulation and release at a subduc-tion zone. J. Geophys. Res. Solid Earth 88, 4984–4996. https://doi.org/10.1029/JB088iB06p04984.

Scoffin, T.P., Stoddart, D.R., Rosen, B.R., 1978. The nature and significance of micro-atolls. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 284, 99–122. https://doi.org/10.1098/rstb.1978.0055.

Sieh, K., 2007. The Sunda megathrust—past, present and future. J. Earthq. Tsunami 1,1–19.

Sikder, A.M., Alam, M.M., 2003. 2-D modelling of the anticlinal structures and structuraldevelopment of the eastern fold belt of the Bengal Basin, Bangladesh. Sediment. Geol.155, 209–226.

Singh, A., Bhushan, K., Singh, C., Steckler, M.S., Akhter, S.H., Seeber, L., Kim, W.-Y.,Tiwari, A.K., Biswas, R., 2016. Crustal structure and tectonics of Bangladesh: newconstraints from inversion of receiver functions. Tectonophysics 680, 99–112.https://doi.org/10.1016/j.tecto.2016.04.046.

Sloan, R.A., Elliott, J.R., Searle, M.P., Morley, C.K., 2017. Chapter 2 active tectonics ofMyanmar and the Andaman Sea. Geol. Soc. Lond. Mem. 48, 19–52. https://doi.org/10.1144/M48.2.

Smithers, S.G., Woodroffe, C.D., 2000. Microatolls as sea-level indicators on a mid-oceanatoll. Mar. Geol. 168, 61–78. https://doi.org/10.1016/S0025-3227(00)00043-8.

Smithers, S.G., Woodroffe, C.D., 2001. Coral microatolls and 20th century sea level in theeastern Indian Ocean. Earth Planet. Sci. Lett. 191, 173–184. https://doi.org/10.1016/S0012-821X(01)00417-4.

Socquet, A., Vigny, C., Chamot-Rooke, N., Simons, W., Rangin, C., Ambrosius, B., 2006.India and Sunda plates motion and deformation along their boundary in Myanmardetermined by GPS. J. Geophys. Res. 111. https://doi.org/10.1029/2005JB003877.

Spada, G., Galassi, G., 2012. New estimates of secular sea level rise from tide gauge dataand GIA modelling. Geophys. J. Int. 191, 1067–1094. https://doi.org/10.1111/j.1365-246X.2012.05663.x.

Stanley, D.J., Hait, A.K., 2000. Deltas, radiocarbon dating, and measurements of sedimentstorage and subsidence. Geology 28, 295–298. https://doi.org/10.1130/0091-7613(2000)28<295:DRDAMO>2.0.CO;2.

Steckler, M.S., Akhter, S.H., Seeber, L., 2008. Collision of the Ganges–Brahmaputra Deltawith the Burma Arc: Implications for earthquake hazard. Earth Planet. Sci. Lett. 273,367–378. https://doi.org/10.1016/j.epsl.2008.07.009.

Steckler, M.S., Mondal, D.R., Akhter, S.H., Seeber, L., Feng, L., Gale, J., Hill, E.M., Howe,M., 2016. Locked and loading megathrust linked to active subduction beneath theIndo-Burman Ranges. Nat. Geosci. 9, 615–618. https://doi.org/10.1038/ngeo2760.

Subarya, C., Chlieh, M., Prawirodirdjo, L., Avouac, J.-P., Bock, Y., Sieh, K., Meltzner, A.J.,Natawidjaja, D.H., McCaffrey, R., 2006. Plate-boundary deformation associated withthe great Sumatra–Andaman earthquake. Nature 440, 46–51. https://doi.org/10.1038/nature04522.

Sun, T., Wang, K., 2015. Viscoelastic relaxation following subduction earthquakes and itseffects on afterslip determination. J. Geophys. Res. Solid Earth 120https://doi.org/10.1002/2014JB011707. 2014JB011707.

Sun, T., Wang, K., Iinuma, T., Hino, R., He, J., Fujimoto, H., Kido, M., Osada, Y., Miura, S.,Ohta, Y., Hu, Y., 2014. Prevalence of viscoelastic relaxation after the 2011 Tohoku-oki earthquake. Nature 514, 84–87. https://doi.org/10.1038/nature13778.

Taylor, F.W., Frohlich, C., Lecolle, J., Strecker, M., 1987. Analysis of partially emergedcorals and reef terraces in the central Vanuatu Arc: comparison of contemporarycoseismic and nonseismic with quaternary vertical movements. J. Geophys. Res. SolidEarth 92, 4905–4933. https://doi.org/10.1029/JB092iB06p04905.

Tregoning, P., Brunner, F.K., Bock, Y., Puntodewo, S.S.O., McCaffrey, R., Genrich, J.F.,Calais, E., Rais, J., Subarya, C., 1994. First geodetic measurement of convergenceacross the Java Trench. Geophys. Res. Lett. 21, 2135–2138. https://doi.org/10.1029/94GL01856.

Uddin, A., Lundberg, N., 2004. Miocene sedimentation and subsidence during con-tinent–continent collision, Bengal basin, Bangladesh. Sediment. Geol. 164, 131–146.https://doi.org/10.1016/j.sedgeo.2003.09.004.

Verlest, M., 1763. An Account of the Earthquakes That Have Been Felt in the Province ofIslamabad, With the Damages Attending Them, From 2nd to the 19th of April, 1762:Translated From the Persian, and Communicated to Henry Vansittart, Esq; Presidentand Governor of Fort William in Bengal, by Mr. Verlest, Chief of the Hon (No. Philos.Trans. (1683–1775)). (East India Company's Affairs at Islamabad).

Vigny, C., 2003. Present-day crustal deformation around Sagaing fault, Myanmar. J.Geophys. Res. 108. https://doi.org/10.1029/2002JB001999.

Wang, K., Tréhu, A.M., 2016. Invited review paper: some outstanding issues in the studyof great megathrust earthquakes—the Cascadia example. J. Geodyn. 98, 1–18.https://doi.org/10.1016/j.jog.2016.03.010.

Wang, K., Hu, Y., He, J., 2012. Deformation cycles of subduction earthquakes in a vis-coelastic Earth. Nature 484, 327–332. https://doi.org/10.1038/nature11032.

Wang, Y., Shyu, J.B.H., Sieh, K., Chiang, H.-W., Wang, C.-C., Aung, T., Lin, Y.N., Shen, C.-C., Min, S., Than, O., Lin, K.K., Tun, S.T., 2013. Permanent upper plate deformationin western Myanmar during the great 1762 earthquake: implications for neotectonicbehavior of the northern Sunda megathrust. J. Geophys. Res. Solid Earth 118,1277–1303. https://doi.org/10.1002/jgrb.50121.

Warrick, R.A., Bhuiya, A.K.A.H., Mitchell, W.M., Murty, T.S., Rasheed, K.B.S., 1996. Sea-level changes in the Bay of Bengal. In: Warrick, R.A., Ahmad, Q.K. (Eds.), The


212







https://doi.org/10.1130/G30872.1




https://doi.org/10.1029/2010JB007499

https://doi.org/10.1029/2010JB007499




https://doi.org/10.1046/j.1365-246X.1999.00802.x

https://doi.org/10.1046/j.1365-246X.1999.00802.x

https://doi.org/10.1016/j.jog.2012.03.005


https://doi.org/10.1007/s10712-011-9171-x

https://doi.org/10.1007/s10712-011-9171-x

https://doi.org/10.1016/S0012-821X(01)00298-9

https://doi.org/10.1016/S0012-821X(01)00298-9




https://doi.org/10.1016/j.gca.2004.06.033

https://doi.org/10.1111/j.1365-2117.2011.00540.x

https://doi.org/10.1029/2003JB002398

https://doi.org/10.1029/2003JB002398

https://doi.org/10.1029/2005JB004025

https://doi.org/10.1029/2006JB004450


https://doi.org/10.1130/0091-7613(1989)017<0068:ATOBAT>2.3.CO;2

https://doi.org/10.1130/0091-7613(1989)017<0068:ATOBAT>2.3.CO;2





https://doi.org/10.1002/jqs.713

https://doi.org/10.1002/jqs.713



https://doi.org/10.1002/2014JB011200

https://doi.org/10.1002/2014JB011200

https://doi.org/10.1016/0012-821X(73)90134-9

https://doi.org/10.1016/j.jseaes.2013.05.018

https://doi.org/10.1029/1999GL005396

https://doi.org/10.1029/1999GL005396

https://doi.org/10.2113/0520407

https://doi.org/10.2113/0520407

https://doi.org/10.1007/s40641-016-0045-7

https://doi.org/10.1038/43854





https://doi.org/10.1098/rstb.1978.0055

https://doi.org/10.1098/rstb.1978.0055







https://doi.org/10.1144/M48.2

https://doi.org/10.1144/M48.2

https://doi.org/10.1016/S0025-3227(00)00043-8

https://doi.org/10.1016/S0012-821X(01)00417-4

https://doi.org/10.1016/S0012-821X(01)00417-4

https://doi.org/10.1029/2005JB003877

https://doi.org/10.1111/j.1365-246X.2012.05663.x

https://doi.org/10.1111/j.1365-246X.2012.05663.x

https://doi.org/10.1130/0091-7613(2000)28<295:DRDAMO>2.0.CO;2

https://doi.org/10.1130/0091-7613(2000)28<295:DRDAMO>2.0.CO;2

https://doi.org/10.1016/j.epsl.2008.07.009

https://doi.org/10.1038/ngeo2760



https://doi.org/10.1002/2014JB011707

https://doi.org/10.1002/2014JB011707



https://doi.org/10.1029/94GL01856

https://doi.org/10.1029/94GL01856

https://doi.org/10.1016/j.sedgeo.2003.09.004






https://doi.org/10.1029/2002JB001999



https://doi.org/10.1002/jgrb.50121

Implications of Climate and Sea–level Change for Bangladesh. Springer, Netherlands,pp. 97–142. https://doi.org/10.1007/978-94-009-0241-1_3.

Weil-Accardo, J., Feuillet, N., Jacques, E., Deschamps, P., Saurel, J.-M., Thirumalai, K.,Demeza, S., Anglade, D., 2016. Relative sea-level changes during the last centuryrecorded by coral microatolls in Belloc, Haiti. Glob. Planet. Chang. 139, 1–14.https://doi.org/10.1016/j.gloplacha.2015.12.019.

Wells, D.L., Coppersmith, K.J., 1994. New empirical relationships among magnitude,rupture length, rupture width, rupture area, and surface displacement. Bull. Seismol.

Soc. Am. 84, 974–1002.Zachariasen, J., Sieh, K., Taylor, F.W., Edwards, R.L., Hantoro, W.S., 1999. Submergence

and uplift associated with the giant 1833 Sumatran subduction earthquake: evidencefrom coral microatolls. J. Geophys. Res. Solid Earth 104, 895–919. https://doi.org/10.1029/1998JB900050.

Zachariasen, J., Sieh, K., Taylor, F.W., Hantoro, W.S., 2000. Modern vertical deformationabove the Sumatran subduction zone: Paleogeodetic insights from coral microatolls.Bull. Seismol. Soc. Am. 90, 897–913.


213

https://doi.org/10.1007/978-94-009-0241-1_3

https://doi.org/10.1016/j.gloplacha.2015.12.019




https://doi.org/10.1029/1998JB900050

https://doi.org/10.1029/1998JB900050




Microatolls document the 1762 and prior earthquakes along ...

Documents

Transcript of Microatolls document the 1762 and prior earthquakes along ...