Navigation – Plan du site

Nannoliths abundance and distribution in tsunami deposits: example from the December 26, 2004 tsunami in Lhok Nga (northwest Sumatra, Indonesia)

Abondance et distribution des nannolithes dans les dépôts de tsunami : l’exemple du tsunami du 26 décembre 2004 à Lhok Nga (nord-ouest de Sumatra, Indonésie)
Raphaël Paris, Mário Cachão, Jérôme Fournier et Olivier Voldoire
p. 109-118

Résumés

Les nannolithes sont un ensemble hétérogène de squelettes calcifiés de la taille des limons et argiles. Ils sont principalement composés de coccolithes, structures calcitiques complexes produites par les coccolithophores marins appartenant à la division des algues microscopiques Haptophyta. Les coccolithes étant présents en milieu océanique mais aussi côtier, ils peuvent être utilisés comme marqueurs paléoenvironnementaux et comme traceurs marins dans les sédiments côtiers. Parmi les structures microscopiques biogéniques généralement observées dans les dépôts de tsunami, les diatomées sont couramment utilisées pour déterminer l'origine marine et les sources sédimentaires mais très peu de travaux ont été réalisés sur les nannolithes. Cet article présente des données concernant l'abondance, l'origine et la répartition des nannolithes dans les dépôts du tsunami de 2004 à Lhok Nga (nord-ouest Sumatra, Indonésie), et les implications en termes de sources sédimentaires et d'hydrodynamique. Les dépôts du tsunami de 2004 présentent des abondances en nannolithes significativement supérieures (1,0 x 104 à 7,2 x 104 nn/g) aux autres sédiments côtiers du même secteur (4,5 x 103 nn/g) mais inférieures à celles des sédiments de plateforme péri-continentale (1,3 à 6,9 x 105 nn/g). Cet enrichissement en nannolithes pourrait donc représenter un outil de reconnaissance des paléo-tsunamis. La présence de nannolithes dans les dépôts du tsunami à Lhok Nga est singulière quand on considère leurs caractéristiques granulométriques grossières et leurs conditions de mise en place, peu favorables au dépôt des nannolithes. Cette relative abondance des nannolithes dans les dépôts du tsunami tend à décroître vers les terres et de bas en haut des coupes analysées, malgré les variations liées aux phases d'érosion et de sédimentation par les vagues successives et aux effets topographiques. L'étude de la distribution verticale des nannolithes, couplée aux analyses granulométriques, représente une source d'information complémentaire pour l'interprétation des dépôts de tsunami.

Haut de page

Notes de la rédaction

Article soumis le 24 avril 2009, accepté le 4 septembre 2009

Texte intégral

Introduction

1Estimate the magnitude (or “size”) of past tsunamis after their deposits is one of the major issues to be developed in studies on tsunami hazard assessment. Main limitations are (i) the great variability of tsunami sandy sheets deposited inland (the sediment source and the topography controlling many aspects of the deposition), (ii) the preservation of these soft sediments, and (iii) the distinction between tsunamis and other coastal depositional events (e.g., storms, hurricanes, seiches). A suite of diagnostic characteristics allows palaeo-tsunami deposits to be identified: they are locally extensive and generally finer landward and upwards (from base to top of the sections); distinct upper and lower sub-units can be identified; the lower contact is unconformable or erosional; the clast size varies from boulders to fine silt; there are fossil-rich sub-units (coral and shell fragments); rip-up clasts of reworked material can be found (e.g., soil, roots, wood, debris); clasts fabrics and laminasets may indicate landward, seaward and oscillatory palaeocurrents. These criteria have been confirmed and discussed through recent works on the December 26, 2004 tsunami deposits in Indonesia, Thailand, Sri Lanka and India (e.g., Bahlburg and Weiss, 2006; Goff et al., 2006a; Moore et al., 2006; Hori et al., 2007; Paris et al., 2007; Srinivasalu et al., 2007; Wassmer et al., 2007; Choowong et al., 2008 a and b; Morton et al., 2008).Nevertheless, tsunami sandy deposits analysed for one single event but in distinct areas may display the same grain-size trends and thickness, even if flow depths and velocities were completely different. Indeed, the sediment source and the topography control many aspects of the deposition inland: thickest deposits in the topographic lows, great spatial variations in thickness and upper laminated texture when the sedimentation was limited by steep slopes, very poor sorting and landward coarsening at the wave breaking line, bimodal or plurimodal grain-size distributions reflecting different sources of sediment (e.g., Nanayama and Shigeno, 2006; Paris et al., 2007; Choowong et al., 2008 a and b).Among microscopic biogenic structures commonly observed in tsunami deposits, diatoms are often used to determine the marine provenance and sediment source (e.g., Dawson, 2007), but few investigations about calcareous nannoplankton and other nannoliths were carried out so far (Andrade et al., 2003). Nannoliths are heterogeneous morphological suite of biogenic carbonate particles with silt-clay size dimensions. They are mainly composed of coccoliths, a suite of calcite complex structures produced by marine coccolithophores belonging to the Haptophyta microalgae division (Winter and Siesser, 1994; Billard and Inouye, 2004). In coastal facies, smaller-sized aragonite spicules of particular colonial ascidians can also be present, where most stellar-shape forms belong to the Didemnidae family (Monniot, 1970; Turon, 1986). Coccoliths are one of the main components of deep ocean carbonate oozes (Baumann et al., 2004; Ziveri et al., 2004). However, this group can also be found in coastal facies and thus is used as a palaeoenvironmental proxy and a marine tracer in coastal sediments (Ferreira and Cachão, 2005; Guerreiro et al., 2005; Alday et al., 2006, Drago et al., 2006). In this paper, we investigate the abundance, distribution and provenance of nannoliths observed in the 2004 tsunami deposits near Lhok Nga (northwest Sumatra, Indonesia) and their implications in terms of sediment sources and flow dynamics.

Previous investigations in Lhok Nga

2The Lhok Nga Bay is a 5 km large coastal embayment (25 km²) opened to the west and delimited by calcareous steep slopes (fig. 1). The coast prior to the 2004 tsunami appeared as a continuous beach, breached by small estuaries during the wet season. Main fringing reefs are located between Lampuuk and the Lhok Nga River. The lowest areas correspond to lagoons, swamps and rice crops. Small hills and hummocky terrains refer to dunes, beach ridges and palaeo-dunes reaching 15 m a.s.l. (e.g., Lampuuk, Lam Lho). In the southern part of the bay, the coastal morphology becomes more contrasted, especially south of the harbour, where the coast shows alternating cliffs and flat crescent-shaped bays and creeks.The December 26, 2004 tsunami in Sumatra was one of the largest tsunamis in recorded human history. In Lhok Nga, the tsunami waves were almost 30 m high and runups reached 51 m a.s.l. (Lavigne et al., 2009). Eyewitnesses reported 10 to 12 waves, the second and third ones being the highest. The sea was observed to recede 10 mn after the earthquake and the first wave came from the southwest a few minutes later. The first wave moved rapidly landward as a turbulent flow with depths ranging from 0.5 to 2.5 m from ground. The second and largest wave (i.e., the tsunami bore) came from the west-southwest within 5 mn after the first one and was 15-30 m high at the coast. Few eyewitness’ accounts are available for the outflow (backwash). Both inflow and outflow produced extensive erosion, sediment transport and deposition until 5 km inland (Umitsu et al., 2006; Wassmer et al., 2007; Paris et al., 2009). The erosional imprints of the tsunami extend to 500 m from the shoreline and exceed 2 km along the river beds. The most eroded coasts were tangent to the tsunami wave train, which came from the southwest. The fringing reefs were not efficient in reducing erosion and destruction inland (Baird et al., 2005). R. Paris et al. (2009) estimated that the volume of beach eroded by the tsunami (ca. 1.5 x 105 m³) represents less than 10% of the sediments deposited inland (ca. 1.5 x 106 m³).The sandy deposits laid down by the tsunami inland record the successive inflows and a final outflow (the backwash), as shown by normally graded couplets or triplets of layers (Moore et al., 2006; Paris et al., 2007). The topmost layers, interpreted as the backwash deposition, describe a seaward sequence of increasing mean grain size and decreasing degree of sorting. Composition of the deposits and plurimodal grain size distribution reflect different sources of sediments. The local effects of the topography on depositional conditions could be identified: thickest deposits in the topographic lows (50-80 cm), large-scale spatial variations in thickness and upper laminated texture when the sediment accretion was limited by steep slopes (oscillations), flame structures in the riverbeds, landward coarsening and very poor sorting at the wave breaking point. The tsunami was also able to detach and transport coral boulders in excess of 10 t over 500-700 m and megaclasts from the tidal flat in excess of 85 t over a few metres (Paris et al., 2009). The coincidence of different size modes, from boulders to fine sands suggests that all the material was not transported in suspension, but rather through a combination of rolling, saltation and suspension.The 2004 tsunami deposits in Lhok Nga are particularly rich in bioclasts: coral fragments, coral and sponge triaxones spicules, rhodophyta calcareous algae, nannoliths, diatoms, benthic foraminifers, gastropods, echinoderms and rare bryozoans (Paris et al., 2007). These faunistic and floristic species are typical of a shallow marine environment.

Fig. 1 – Location of the Lhok Nga Bay and samples collected for analysis of nannoliths.
Fig. 1 – Carte de localisation de la Baie de Lhok Nga et des échantillons récoltés pour l'analyse des nannolithes.

Fig. 1 – Location of the Lhok Nga Bay and samples collected for analysis of nannoliths.Fig. 1 – Carte de localisation de la Baie de Lhok Nga et des échantillons récoltés pour l'analyse des nannolithes.

Method for the construction of bathymetry (5-m dashed contour lines) and topography (10-m contour lines) is presented in F. Lavigne et al. (2006, 2009).
La méthode de réalisation de la bathymétrie (équidistance 5 m, isobathes  en pointillés) et de la topographie (équidistance 10 m) est présentée in F. Lavigne et al. (2006, 2009).

Methodology

3Samples were selected according to longitudinal transects in the 2004 tsunami deposits (fig. 1; NGA and LHO samples). Additional samples from a pre-tsunami beach (NGA1), and a post-tsunami beach (LPK2), lagoons (LAG) and the inner shelf (m0) were also selected in order to compare compositions and abundances of modern nannolith thanatocoenoses (post-mortem assemblages recently deposited into surface sediments) with the tsunami deposit itself. All samples were prepared following the random settling procedure (Flores and Sierro, 1997) adapted to paralic facies (Ferreira and Cachão, 2003). Due to sample coarseness weights of dried raw sediment varied from 2 to 2.5 g (with 1 x 10-4 g precision) while pipetted volumes varied from 4 to 5 ml. Cover slips were permanently mounted on slides using the synthetic balsam Entellan. For each slide, a 3-cm continuous row was screened through optical cross-polarizing microscope at x 1250 magnification and all specimens tabulated into species whenever possible or into broader taxa. The taxonomic frameworks of K. Perch-Nielsen (1989) and P. Bown (1999) and references therein have been followed. All nannoliths with more than half of the structure preserved were included in the counts.Additionally, the CaCO3 content of the 2004 tsunami deposits was determined using a Bernard calcimeter. This method consists of quantifying the CO2 released when the sample is treated with hydrochloric acid. In a closed system, under a constant pressure and temperature, and without any other gases involved, the quantity of CO32− is directly proportional to the volumetric increase resulting from the release of the CO2. The stratigraphy, internal structure and grain-size analyses of these tsunami deposits are already presented in a previous contribution (Paris et al., 2007).In order to infer the transport capacity of the tsunami offshore, we applied the methodology used for instance by F. Maeno and F. Imamura (2007) and A. Noda et al. (2007). Tsunami waves raised the bed shear velocities to levels above critical values for the entrainment of submarine sediments. We estimated the critical particle size for entrainment based upon the simulated current velocities of the tsunami waves (data provided by A. Loevenbruck et al., 2007). We assumed that the flow type generated by the tsunami is a unidirectional flow, because the wave periods were very long, and that the predominantly sandy sediments behave in an essentially non-cohesive manner (sand ripples).Threshold shear velocity (u*) at the bed can be related to threshold flow velocity U100 (nominally taken as 100 cm above the bed) using Karman-Prandtle equation:

(1)

Critical threshold shear velocity (u*cr)of each grain-size class (D) is then derived through the A. Shields (1936) entrainment function:

(2)

Threshold Shields parameter θcr for cohesionless grains (e.g., Soulsby and Whitehouse, 1997)is given by:

(3)

with dimensionless grain size (Yalin, 1972):

(4)

where κ is the Prandtle constant (κ≈ 0.4), z is the height above the bed (in cm), zo is the roughness length (0.5 cm for irregular or rippled bed of sand or gravel), ρs is the density of the boulder (in kg/m³ or g/cm³), ρwis the water density (ρw = 1.025 g/ml for sea “clear” water), g is the gravitational acceleration (g = 9.8 m/s²) and v is the sea water viscosity at 25°C (0.00896 m²/s).We thus illustrate the relationship between the critical threshold shear velocity (u*cr) of a specific grain size (D) and the simulated tsunami shear velocity (u*) each 5 mn of simulation. Sediment transport occurs when u* > u*cr. We calculated and compared the critical threshold shear velocities of specific grain sizes and the simulated tsunami shear velocities using an 18 m bathymetric grid realised in December 2005 (Lavigne et al., 2006). Tsunami flow velocities were extracted from a numerical simulation by A. Loevenbruck et al. (2007).

Results

4The 2004 tsunami deposits in Lhok Nga are although almost silty-clay void coarse bioclastic sands. However, they revealed the presence of nannoliths (present day coccoliths and Ascidian spicules), with abundances between 1.0 and 7.2 x 104 per grams(data matrix available at http://mcprojectos.fc.ul.pt/​papers/​Lhok_Nga/​Lhok_Nga_nanno.pdf). These values are one order of magnitude lower than the innershelf present-day sediments off Lhok Nga (from 1.3 to 6.9 x 105 nannoliths per grams). Highest abundance was found in sample LAG6 of the Lampuuk lagoon (8.5 x 105 nn/g). Large amounts of sediment were trapped in this lagoon during the tsunami outflow (or backwash) because the fringing reef acted as a topographical barrier. Sediments collected in LAG6 are medium sand rich in bioclasts, and numerous fragments of centimetric Acropora coral branches. Beach samples have expectable low abundance of nannoliths (ca. 4.5 x 103 nn/g), while bryozoan fragments (fig. 2: images 26-29) and ascidian spicules (fig. 2: images 16-25) are, comparably, more abundant. There is no difference between pre-tsunami (4.4 x 103 nn/g) and post-tsunami beach samples (4.5 x 103 nn/g). Tsunami deposits sampled close to the beach have abundances of nannoliths ten times higher (2.19 x 104 nn/g for NGA4). Simple nannolith diversity (number of distinct non-reworked present day taxa) varies in relation with abundance (r = 0.65). Highest diversities were found for samples NGA-7A and LAG 6, with values similar to diversities obtained for the neritic samples (>10).

Fig. 2 - Nannoliths observed in the 2004 tsunami deposits of the Lhok Nga Bay(calcareous nannoplankton, ascidian spicules, bryozoa fragments and reworked calcareous nannofossils).
Fig. 2 – Nannolithes observés dans les dépôts du tsunami de 2004 dans la Baie de Lhok Nga (nannoplancton, spicules d’ascidies, fragments de bryozoaires et nannofossiles calcaires remaniés).

Fig. 2 - Nannoliths observed in the 2004 tsunami deposits of the Lhok Nga Bay(calcareous nannoplankton, ascidian spicules, bryozoa fragments and reworked calcareous nannofossils).Fig. 2 – Nannolithes observés dans les dépôts du tsunami de 2004 dans la Baie de Lhok Nga (nannoplancton, spicules d’ascidies, fragments de bryozoaires et nannofossiles calcaires remaniés).

All images are at the same magnification (scale bar = 5 mm) and taken under cross nicols. 1-6: Gephyrocapsa oceanica (1: sample NGA 2A; 2: sample NGA 7A; 3: sample NGA 8; 4: sample NGA 18; 5: sample LAG 6; 6: sample LAG 8). 7: Gephyrocapsa ericsonii (sample NGA 7A). 8: Umbilicosphaera sibogae (sample NGA 7A). 9-11: Helicosphaera carteri (9: sample NGA 7A; 10: sample NGA 18; 11: sample LAG 6). 12-15: Helicosphaera wallichii (12-13: sample NGA 2D; 14-15: sample LAG 6). 16-25: Ascidian spicules (16-17: distinct focal planes, sample NGA 7A; 18: sample NGA 18; 19: sample LAG 6; 20-21: distinct focal planes, sample LAG 6; 22-23: distinct focal planes, sample LAG 6; 24-25: sample LAG 8). 26–29: isolated fragments of Bryozoan carbonate skeleton (sample NGA 9A). 30–31: reworked calcareous nannofossil (Dictyococcites) (30: sample NGA 7A; 31: sample NGA 18). 32: reworked calcareous nannofossil (Watznaueria; sample NGA 2B).
Toutes les images sont prises au même grossissement (échelle 5 mm). 1-6 : Gephyrocapsa oceanica (1 : échantillon NGA 2A ; 2 : échantillon NGA 7A ; 3 : échantillon NGA 8 ; 4 : échantillon NGA 18 ; 5 : échantillon LAG 6 ; 6 : échantillon LAG 8). 7 : Gephyrocapsa ericsonii (échantillon NGA 7A). 8 : Umbilicosphaera sibogae (échantillon NGA 7A). 9-11 : Helicosphaera carteri (9 : échantillon NGA 7A ; 10 : échantillon NGA 18 ; 11 : échantillon LAG 6). 12-15 : Helicosphaera wallichii (12-13 : échantillon NGA 2D ; 14-15 : échantillon LAG 6). 16-25 : spicules d'ascidie (16-17 : échantillon NGA 7A ; 18 : échantillon NGA 18 ; 19 : échantillon LAG 6 ; 20- 21 : échantillon LAG 6 ; 22-23 : échantillon LAG 6 ; 24-25 : échantillon LAG 8). 26-29 : fragments isolés de squelette carbonaté de bryozoaires (échantillon NGA 9A). 30-31 : nannofossile calcaire remanié (Dictyococcites ; 30 : échantillon NGA 7A ; 31 : échantillon NGA 18). 32 : nannofossile calcaire remanié (Watznaueria ; échantillon NGA 2B).

5Gephyrocapsids, namely Gephyrocapsa oceanica, a distinct feature of coastal nannolith thanatocoenoses, dominate in all samples (Okada, 1983; Ferreira and Cachão, 2005; Guerreiro et al., 2005). This is corroborated by much less diversity than the oceanic (Indian Ocean) realm (see Andruleit, 2007), lower abundances of small placoliths (Emiliania huxleyi + Gephyrocapsa ericsonii), and the occurrence of ascidian spicules (Alday et al., 2006; Turon, 1986). On the other hand, the common presence of Umbilicosphaera sibogae (fig. 2: image 8) is indicative of warm water assemblages (Winter and Siesser, 1994). Small quantities of reworked forms, mainly Neogene (Dictyococcites, Sphenolithus, Reticulofenestra) and in lesser degree Cretaceous (Watznaueria), were detected which indicatives active processes of erosion from Mesozoic and Cenozoic nearby outcrops (fig. 2: images 30-32). The apparent correlation between extant and reworked forms seems to indicate that the latter were present inside the surface sediments before the 2004 tsnami and not preferentially eroded during the tsunami itself. The coastal and shallow-water signature of the nannoliths observed indicates that there is not an increase of open-ocean over coastal taxa brought by the tsunami.

6Nannoliths, together with other marine biogenic elements (tests of small planktonic foraminifera, sponge spicules, bryozoa colony fragments), could be traced along the 3.5 km long NGA transect in the northern part of the Lhok Nga Bay and compared to samples LHO in the southern part (Paris et al., 2007). LHO3 deposits display abundance (1.4 x 104 nn/g) in the same order of magnitude than NGA samples, but LHO4 has low to very low abundances (from 0 to 4.9 x 103 nn/g). Along the NGA transect (fig. 3), the mean abundances from the shore to 1 km inland are quite high, but remain variable from one site to another (from 1 x 104 nn/g to 5 x 104 nn/g), and then decrease landward (< 2 x 104 nn/g). This longitudinal trend is also observed for the CaCO3 content. Samples NGA 2 to NGA 12 (< 1 km inland) contain more than 25% of carbonates (26-57%), whereas NGA 14 to 18 (from 1.5 km to 3.5 km inland) have less than 25% (3-25%).

7Nevertheless, abundances are highly variable when considered in vertical sections of tsunami deposits. The vertical distribution of nannoliths and the CaCO3 content are not clearly correlated with other sedimentological trends (e.g., thickness, mean size, sorting). Sections NGA2 and NGA 7 (fig. 4) are located respectively 175 m and 330 m from the shore, on the banks of brackish lagoons scoured by the tsunami. These sections thus represent a complete sediment record of the event. R. Paris et al. (2007) used successions of normally graded couplets or triplets of layers to identify the signature of successive waves. Deposits of the first wave were eroded by the subsequent waves, except in some topographic lows (e.g., NGA2-A at 52 cm deep). For both sections, the carbonate content decreases upward, with a slight increase at the arrival of the third wave (NGA2-E and NGA7-D). The second and main wave (i.e., the tsunami bore) is recorded by three layers fining upward (NGA2-B-C-D and NGA7-A-B-C). A first decrease of the number of nannoliths is rapidly followed by a significant increase (NGA2-D and NGA7-C). Same trend occur from NGA7-D and NGA 7-F (third wave?), but with lower abundances, whereas the second decrease upward is continuous along NGA2 section. Finally, the two sections both display a two-phase distribution of nannoliths, but peaks of abundance appear at the beginning of the second phase for NGA 2 (7.2 x 104 nn/g at NGA2-D) and at the beginning of the first phase for NGA 7 (4.5 x 104 nn/g at NGA7-A). A similar vertical trend is observed for section NGA9, located 475 m inland.

Fig. 3 – Abundance of nannoliths and thickness of the 2004 tsunami deposits near Lampuuk (northwestern coast of Sumatra).
Fig. 3 – Quantité de nannolithes et épaiss>5 ) (mi de Toutes les images ude the 20and Siesser, 199Céchan4echinodepos of nannoliths and thiconpm>) (mi de Toan_;,) (nannoplankton, ascidian spicules, bryozoa fragments and reworked calcareous nannofossils).Fig.72 – Nannolithes obsevés dans les dépôts du tsunamns les dépôts du tsunamnction. Finally, the two sections both display a two-phase distribution of nannoliths, 4em>FiVted as tof the 2in sample LAG6 of the La-E ands and reas theng of the second phase for NGA 2 (7.2 x 104 nn/g at NGA2-D) and at the beginni4em>FiVted as tores et leursaniq. 3 – Quantité de nanryozoaires et nannofossiles calcaireslar vertical trend is observed for section NGA9, located 475 m inland.

8

Fig. 34em>FiVted as tof the 2in sample LAG6 of the La-E ands and reas theng of the second phase for NGA 2 (7.2 x 104 nn/g at NGA2-D) andginni4em>FiVted as tores et leursaniq. 3 – Quantité de nanryozoaires et nannofossiles calcaireslar vertical trend is observed for section NGA9es images ude the 20and Siesser, 199Céchan4echinodepos of nannoliths and thiconpm>) (mi de Toan_;,) 8nannoplankton, ascidian spicules, bryozoa fragments and reworked calcareous nannofossils).Fig.82 – Nannolithes obse and samples collected for analy construction of bathymetry (5-m dashed contour li5r. We calculate5 and compared er 2005 (Lfra). The Us calb>cont triplets of>er 2005 (Lion rppeared af these tsunamiiculesssocinami wavs of tilisées pcm long triplets ofions), soted w The low abundanod LHO es ions)crosconocribe a ser"iffe andgnt appresent, where most stellar-a suins can also be pliths (preset degranvb>*et ertical st that ther"paranu"texte" dis wavs ss="papthoreline and ex-12 me suobsermaranu (iiishredbe a >er 2005 (Lion re tsunamiiculerd (<itedex 10,ithesce 5 m,uous baves, excwons rll nble low abundan865/img-2.pitecol as tof tiv crch-Nwami in Suon was le t75 om 2 tree CreseawarcospOrig-of-viewoair-31 : nannl trends (of td upphe ts3< and NGch, br5 (stormsnd is also observegments and cidianrsese tsunamiihe west-dia-B-C-D and of n) in thre 4 sr varied. Nofig. depositony f innaueria), ewmber">3McospOrig-of-viewoates that there tsunaminland) copographic labovusatedts of el="r2006;awer t2008 a and b)diashru mn of sielsNwami ire uhes. Beairide the surtflow the tsunami. Th). Tsunami arbona texture when the sup>4erved in ns utng">Orid Ca., also be prn o-29) difFloresnnolik_Ngart irtinnd Inanceranple /ml fonot trerms ch i couplets or t a and b)diashru mn ,boulders in examples, (Andradet al., 2003).specimen the soted w The Nga (nannoplancton, sp5 m,uous bavt isup>ng">and of n resp tsunaminlssocinami wavhills anable from one site to another (from 1 x 104 nn/g to 5 x 104<9shear velrbona ifFlNGA mi wavbeified; the:es wereumann etm>), were detected which i of the second phase for NG nd thus is sed as a palaeoenvironmentalween teted as theble low abun decrease latsunami wavese07; Choowong caNevlackg theurm>Dictyorces of ss. The mam Lho). In the southern part ofTe critical theng of m²/s).We thus illustrat> 5)s can beives acDecemed 2 km along the river beds. The w (backwasher gase. Bt des échliths (presendia ( ths areated with hydrochloric acid. In a closedouthern part s invm>., 2007; Paris et al., 2009). The erosional imprints of the 0.1ts were tangent to the n vertical. Teline and exceed 2 km along the river beds. The most eroded coasts were tangent to the t cent works on tsser degree Cretaceous (Watznaueria 4.9 x6nal (>etm>), were dets (Okrric16-25)ndicatives active processes of erosion from Mesozoic and Cenozoic nearby outcrops (fig. 2: imaDaippare beinWe thus illustrate tttp://mcprojevssure alon NGA b 1.5on tsser ave iculesuleumen thuite h size a. The first wave mos thus ran class=Aceh b 1.5 (2700-23st de of b wavese0ictyoarc dir="ltpapth ce (1.st dndguobsermaranu zt wRSigna were detecleftds. The mucem 4 to 5 mlemte" ter vis riverbeds, and NGAused successions of, ndts, ed 2 km 7 (fig. 4) areeally sh size aamples weremblage> m³ calcaire n layc loccording the th/em>. 1.5 x

This longitudinals 2in sample LAG6 of the Lade l bio of openv claobhe fatsunami wavese07; Choowong et al., thic LHO3 deposcedriver bedher gase( and NGpt/​pah at the Choowong ods whenebottospanic3">Methodoclass="icer. The first wave moretion hodotra,zt whene. The first pnable from one site to another (from 1 x 104 nn/g to 5 x 104<10shear vWe (oscilldes ac ulseheble low abun decrease nlandthan 25% of cA .g., thic L and veryamparablst wave moretioeported m>, 2 thi. Weposplote sedima CM06;awramacid thg30 : 64emniletIconA, 20the Decembf open-B-C-D wav 4 to 5 mldriver bedtraroductio(png"load)d consisdee tsunami wam> ths Roductiosub>

eated 475 m inland.

9nannoplankjp="textIcon">

Fig. 35em>FiM pneng of m²/s).We thus illusyt-tsunamsu*g/e sea wateramples, o able to detliths (presedginni5collected Lapai vall s Lapa cisa dile LAGC (0figllu*g/e sea n de la Baeossiles calcairesau observ/em>

) (mi de Toan_;,) 9nannoplankjp=" ascidian spicules, bryozoa fragments and reworked calcareous nannofossils).Fig.9kjp=" Nannolithejpeg, 88és dans les dépôts du tsunami de 2004 dans la Baie de Lhok Nga (nannoplLity y-clay B-C-D pres-clok Ngabein y-clayn, sed.hing to 12 waves,ssions and flatch ridges and palafor thy B-w sper m>

)onpas oeses, sa the"nofo,3="d/sb. Au grai/em40-4cleon NGitical th,Baeossiles cambinahilltura 5 nannoldere atoes (leatioderreiroudualisaifLfrangeaTheses in Lhooochantillon LAG 6 ; 20- 21 : échantillon LAG 6 ; 22-2ction. Finally, the two sections both display a two-phase distribution of nannoliths, 6collecM0puantintmem>Splow abun decrease lang of the second phase for NGthymetric gd at the beginni6colleIae

floristic species are typical of a shal0low marine environment.

Fig. 36collecM0puantintmem>Splow abun decrease lang of the second phase for NGthymetric gdginni6colleIae

Fig. 1 – Locat02 – Nannolithes obsep>
                  <blockquote>
 e 2004 dans la Baie de Lhok Nga (nannople beMf the boul5t al</emeent toetrait" d1 er 1.3 aeopeumees, sa tocid thg30 : 64 emniletIconA, d n" dir="ltr">Method for the construction of bathymetry (5-m dashed contour li6r. We calculate6 and compared6r3 cont of thestilisées pcmtriplets ofw The ed w The low abundanod LHO e. 2: imn sample LAG6 of the Lade of the second phase for NGtedis wavits display abundaositiothus rpaippareeted as the uhes. >e.g., other / deposits anal NGA7> yGA .g.
, thic LHO3 do). These va, as shknWup>43. ofened is also observe Tsunami ch, binvm>.,g as the-and)eedmi wavancl"icyt yc ulsehe ua deposit 3.5 km inland) havmem>Splow abun ami. ThLadod) co B-C-D riticonAnannly. A12 wavev clasn sinsravo doI,G6 of the La-B-C-D ad A. seful34d) 8 ; 1could bFceanh Emb (fyicsoIndoeraiaG em>Asisk LoeveTIP mber">4secttanceé ( A w LoevJoh" als Steathrayt om"textce a ser"md tnts pf rolling, saltation an const

Annexe

Version française abrégée

L'un des principaux enjeux actuels dans l'étude des tsunamis concerne l'interprétation de leurs dépôts et leur exploitation en terme de reconstitution des anciens tsunamis.Les principales limites demeurent la grande variabilité des dépôts sableux laissés à terre par ces événements, leur préservation, et leur distinction avec les dépôts de tempête ou ceux consécutifs des cyclones et des seiches. Les caractéristiques (amplitude, vitesse, direction, turbulence) d'un tsunami, la topographie et le contexte morphosédimentaire déterminent la répartition spatiale et les caracté­ristiques de ces dépôts. Les éléments sui­vants permettent de les distinguer d’autres dépôts, comme par exemple les dépôts de tempête : ils sont étendus à toutes les côtes d’accumulation affectées par le tsunami ; leur taille moyenne décroît généralement en fonction de l’altitude et de la distance par rapport à la côte (landward et upward fining) ; on distingue souvent une stratification interne avec des unités inférieures plus grossières ; ils reposent en discontinuité sur le substratum, ce dernier présentant parfois des formes d’érosion ; la granularité varie des sables fins aux blocs plurimétriques ; ils sont riches en bioclastes (coquilles, fragments de coraux, etc.), fragments de sol (rip-up clasts), de plantes et de bois ; les figures d’imbrication indiquent des paléo-courants dirigés à la fois vers la terre et vers la mer ainsi que des phases d'oscillation. Parmi les structures microscopiques biogéniques généralement observées dans les dépôts de tsunami, les diatomées sont couramment utilisées pour déterminer l'origine marine et les sources sédimentaires, mais très peu de travaux ont été réalisés sur d'autres marqueurs tels que les nannolithes. Ces derniers sont principalement composés de coccolithes, structures calcitiques complexes produites par les coccolithophores marins appartenant à la division des algues microscopiques Haptophyta. Les coccolithes sont présents en milieu océanique mais aussi côtier et peuvent donc être utilisés comme marqueurs paléoenvironnementaux et comme traceurs marins dans les sédiments côtiers.

Cet article présente des données concernant l'abondance, l'origine et la répartition des nannolithes dans les dépôts du tsunami de 2004 à Lhok Nga (nord-ouest de Sumatra, Indonésie) et les implications en termes de sources sédimentaires et d'hydrodynamique. La Baie de Lhok Nga se poursuit à terre sous la forme d'une plaine littorale large de 5 km, qui a été touchée très sévèrement par le tsunami du 26 décembre 2004 et intégralement inondée (vagues de 15 m à 30 m à la côte, runups de 20 m à 50 m sur les falaises et les versants). Des relevés ont été effectués le long de transects dans les dépôts sableux du tsunami orientés dans la direction des vagues. Des relevés complémentaires ont été effectués en mer, au large et derrière les récifs frangeants. Les abondances en nannolithes ont été complétées par des analyses au calcimètre Bernard et des données granulométriques (Paris et al., 2007). La capacité du tsunami à mobiliser des sédiments au large est appréhendée par une modélisation numérique à partir des équations de Karman-Prandtle.

Les résultats montrent que les dépôts du tsunami de 2004 présentent des abondances en nannolithes significativement supérieures (1,0 x 104 à 7,2 x 104 nn/g) aux autres sédiments côtiers du même secteur (4,5 x 103 nn/g) mais inférieures aux sédiments de plateforme péri-continentale (1,3 à 6,9 x 105 nn/g). Les nannolithes observés sont typiques des zones côtières peu profondes, suggérant que les sédiments plus profonds mobilisés en charge de fond par le tsunami ont été bloqués au niveau du talus continental. L'enrichissement relatif en nannolithes dans des sédiments côtiers pourrait donc représenter un outil de reconnaissance des paléo-tsunamis, même si les résultats restent préliminaires. La présence de nannolithes dans les dépôts du tsunami à Lhok Nga est surprenante du fait de la granularité des sédiments (sables moyens à grossiers, mal triés) et de leurs conditions de mise en place, peu favorables au dépôt des nannolithes. Cette relative abondance des nannolithes dans les dépôts du tsunami tend à décroître vers l'intérieur des terres et de bas en haut des coupes stratigraphiques analysées, malgré des variations liées aux phases d'érosion et de sédimentation par les vagues successives (pulsations) et aux effets topographiques, particulièrement influents lors du flux de retrait (backwash). Finalement, l'étude de la distribution verticale des nannolithes couplée à la granularité des sédiments représente une source d'information complémentaire pour l'interprétation des dépôts de tsunami. Cependant, les études à venir devront expliquer les décalages observés entre les variations de taille moyenne des dépôts, de tri granulométrique et d'abondances en nannolithes.

Haut de page

Table des illustrations

Titre Fig. 1 – Location of the Lhok Nga Bay and samples collected for analysis of nannoliths.Fig. 1 – Carte de localisation de la Baie de Lhok Nga et des échantillons récoltés pour l'analyse des nannolithes.
Légende Method for the construction of bathymetry (5-m dashed contour lines) and topography (10-m contour lines) is presented in F. Lavigne et al. (2006, 2009).La méthode de réalisation de la bathymétrie (équidistance 5 m, isobathes  en pointillés) et de la topographie (équidistance 10 m) est présentée in F. Lavigne et al. (2006, 2009).
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-1.png
Fichier image/png, 309k
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-2.png
Fichier image/png, 1,5k
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-3.png
Fichier image/png, 2,1k
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-4.png
Fichier image/png, 3,6k
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-5.png
Fichier image/png, 2,1k
Titre Fig. 2 - Nannoliths observed in the 2004 tsunami deposits of the Lhok Nga Bay(calcareous nannoplankton, ascidian spicules, bryozoa fragments and reworked calcareous nannofossils).Fig. 2 – Nannolithes observés dans les dépôts du tsunami de 2004 dans la Baie de Lhok Nga (nannoplancton, spicules d’ascidies, fragments de bryozoaires et nannofossiles calcaires remaniés).
Légende All images are at the same magnification (scale bar = 5 mm) and taken under cross nicols. 1-6: Gephyrocapsa oceanica (1: sample NGA 2A; 2: sample NGA 7A; 3: sample NGA 8; 4: sample NGA 18; 5: sample LAG 6; 6: sample LAG 8). 7: Gephyrocapsa ericsonii (sample NGA 7A). 8: Umbilicosphaera sibogae (sample NGA 7A). 9-11: Helicosphaera carteri (9: sample NGA 7A; 10: sample NGA 18; 11: sample LAG 6). 12-15: Helicosphaera wallichii (12-13: sample NGA 2D; 14-15: sample LAG 6). 16-25: Ascidian spicules (16-17: distinct focal planes, sample NGA 7A; 18: sample NGA 18; 19: sample LAG 6; 20-21: distinct focal planes, sample LAG 6; 22-23: distinct focal planes, sample LAG 6; 24-25: sample LAG 8). 26–29: isolated fragments of Bryozoan carbonate skeleton (sample NGA 9A). 30–31: reworked calcareous nannofossil (Dictyococcites) (30: sample NGA 7A; 31: sample NGA 18). 32: reworked calcareous nannofossil (Watznaueria; sample NGA 2B).Toutes les images sont prises au même grossissement (échelle 5 mm). 1-6 : Gephyrocapsa oceanica (1 : échantillon NGA 2A ; 2 : échantillon NGA 7A ; 3 : échantillon NGA 8 ; 4 : échantillon NGA 18 ; 5 : échantillon LAG 6 ; 6 : échantillon LAG 8). 7 : Gephyrocapsa ericsonii (échantillon NGA 7A). 8 : Umbilicosphaera sibogae (échantillon NGA 7A). 9-11 : Helicosphaera carteri (9 : échantillon NGA 7A ; 10 : échantillon NGA 18 ; 11 : échantillon LAG 6). 12-15 : Helicosphaera wallichii (12-13 : échantillon NGA 2D ; 14-15 : échantillon LAG 6). 16-25 : spicules d'ascidie (16-17 : échantillon NGA 7A ; 18 : échantillon NGA 18 ; 19 : échantillon LAG 6 ; 20- 21 : échantillon LAG 6 ; 22-23 : échantillon LAG 6 ; 24-25 : échantillon LAG 8). 26-29 : fragments isolés de squelette carbonaté de bryozoaires (échantillon NGA 9A). 30-31 : nannofossile calcaire remanié (Dictyococcites ; 30 : échantillon NGA 7A ; 31 : échantillon NGA 18). 32 : nannofossile calcaire remanié (Watznaueria ; échantillon NGA 2B).
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-6.png
Fichier image/png, 264k
Titre Fig. 3 – Abundance of nannoliths and thickness of the 2004 tsunami deposits near Lampuuk (northwestern coast of Sumatra).Fig. 3 – Quantité de nannolithes et épaisseur des dépôts du tsunami de 2004 près de Lampuuk (côte nord-ouest de Sumatra).
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-7.png
Fichier image/png, 24k
Titre Fig. 4 – Variation of the abundance of nannoliths along cross-section of the 2004 tsunami deposits near Lampuuk (northwestern coast of Sumatra).Fig. 4 – Variation verticale de la quantité de nannolithes dans les dépôts du tsunami de 2004 près de Lampuuk (côte nord-ouest de Sumatra).
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-8.png
Fichier image/png, 29k
Titre Fig. 5 – Maps of threshold shear velocity values (u*) calculated for the 2004 tsunami off Lhok Nga.Fig. 5 – Cartes des vitesses de cisaillement critique (u*) calculés pour le tsunami de 2004 au large de Lhok Nga.
Légende Low values appear in blue and high values in red. 30 mn after the main earthquake, the tsunami wave train approaches Lhok Nga. u* values are already high enough (> 0.5 cm/s) to mobilize submarine sediments (u*cr of very coarse sands is ≈ 0.3 cm/s). At 40-45 mn, the tsunami has invaded the Lhok Nga Bay and high values of u* (10-20 cm/s) are observed at the coast, especially around the fringing reef in Lampuuk.Les faibles valeurs apparaissent en bleu et les fortes valeurs en rouge. 30 mn après la secousse sismique principale, le train d'onde approche de Lhok Nga. Les valeurs de u* sont déjà suffisamment élevées (> 0,5 cm/s) pour mobiliser des sédiments au fond (u*cr pour des sables grossiers ≈ 0,3 cm/s). Au bout de 40-45 mn de simulation, le tsunami a envahi les terres et de fortes valeurs de u* (10-20 cm/s) sont observées à la côte, surtout autour du récif frangeant de Lampuuk.
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-9.jpg
Fichier image/jpeg, 88k
Titre Fig. 6 – CM patterns and nannolith abundances of the 2004 tsunami deposits in Lhok Nga.Fig. 6 – Images CM et abondances en nannolithes des dépôts du tsunami de 2004 à Lhok Nga.
Légende M50 corresponds to the median and C95 to the fifth coarsest percentile (Passega, 1964; Allen, 1971).M50 représente la médiane et C95 le 5e percentile plus grossier (Passega, 1964 ; Allen, 1971).
URL http://journals.openedition.org/geomorphologie/docannexe/image/7865/img-10.png
Fichier image/png, 21k
Haut de page

Pour citer cet article

Référence papier

Raphaël Paris, Mário Cachão, Jérôme Fournier et Olivier Voldoire, « Nannoliths abundance and distribution in tsunami deposits: example from the December 26, 2004 tsunami in Lhok Nga (northwest Sumatra, Indonesia) », Géomorphologie : relief, processus, environnement, vol. 16 - n° 1 | 2010, 109-118.

Référence électronique

Raphaël Paris, Mário Cachão, Jérôme Fournier et Olivier Voldoire, « Nannoliths abundance and distribution in tsunami deposits: example from the December 26, 2004 tsunami in Lhok Nga (northwest Sumatra, Indonesia) », Géomorphologie : relief, processus, environnement [En ligne], vol. 16 - n° 1 | 2010, mis en ligne le 01 avril 2012, consulté le 17 décembre 2017. URL : http://journals.openedition.org/geomorphologie/7865 ; DOI : 10.4000/geomorphologie.7865

Haut de page

Auteurs

Raphaël Paris

GEOLAB UMR 6042 CNRS-UBP - Clermont-Ferrand - France (raparis@univ-bpclermont.fr)

Articles du même auteur

Mário Cachão

Centro de Geologia - Universidade de Lisboa - Campo Grande - 1749-016 Lisboa - Portugal

Jérôme Fournier

BOREA UMR 7208 CNRS-MNHN - CRESCO - 38, rue du Port Blanc - 35800 Dinard - France

Articles du même auteur

Olivier Voldoire

GEOLAB UMR 6042 CNRS-UBP - Clermont-Ferrand - France

Haut de page

Droits d’auteur

© Groupe français de géomorphologie

Haut de page
  • Logo CNRS - Institut des sciences humaines et sociales
  • Logo Groupe français de géomorphologie
  • OpenEdition Journals