Éditeur(s) : Université de Brest
Résumé : The channel-lobe transition zone has been studied in two turbiditic systems: the Amazon channel-mouth lobe complex at the termination of the last eight channel-levee systems and the Neochannel termination, the last active channel of the Petit-Rhone turbidite system. - In the two cases, the internal geometry and stacking pattern of the deposits show a significant control of the configuration (size and shape) of the receiving basins, their morphology (gradients and subtle gradient changes), and their degree of confinement. - The Amazon and Petit-Rhone systems, with lobes built after the Last Glacial Maximum and the ensuing sea-level rise, evidence the major impact of external forcings (climatic events and eustatic variations) that induce variations in sediment fluxes delivered to the deep basin and therefore exert a control on channel-mouth lobes construction. The channel-mouth lobes of the Amazon and Petit-Rhone systems, even if they are located at great distances from the river mouths, at the termination of the turbidite systems and in areas characterised by subtle morphological variations, are both constrained by combined internal (accommodation and compensation) and external factors (post-glacial deconnection between the fluvial system and the deep basin).
La zone de transition chenal-levées/lobes a été étudiée dans 2 systèmes turbiditiques : le complexe de lobes à la terminaison des 8 derniers chenaux de l'éventail de l'Amazone et la terminaison du Néochenal, dernier chenal de l'éventail du Petit-Rhône. - Dans les deux cas, l'agencement des dépôts et leur évolution dans l'espace et le temps révèlent un fort contrôle de la configuration des réceptacles (taille et forme), de la morphologie des fonds (pentes et taux de changement des pentes) et du degré de confinement. - Les systèmes de l'Amazone et du Petit-Rhône, dont les lobes se sont construits à partir du Dernier Maximum Glaciaire et lors de la remontée du niveau marin qui a suivi, ont en outre permis de souligner l'impact majeur, sur la mise en place des lobes terminaux, de certains forçages externes (événements climatiques et variations eustatiques), qui déterminent les variations des flux sédimentaires arrivant au bassin. Ainsi, bien que situés à des distances considérables des sources fluviatiles et dans des zones distales où les variations de relief sont très ténues, les lobes terminaux de ces deux grands systèmes turbiditiques se sont révélés contraints à la fois par des processus internes (processus de compensation) et par des processus externes (déconnexion post-glaciaire du fleuve et du bassin).

Droits : info:eu-repo/semantics/openAccess
http://archimer.ifremer.fr/doc/2008/these-6796.pdf
http://archimer.ifremer.fr/doc/00000/6796/

Éditeur(s) : HAL CCSD Blackwell Publishing
Résumé : International audience
In the Causses platform (south-east France), Late Hettangian to Sinemurian deposits were interpreted previously as shallow-water carbonate ramp deposits. A new look at these deposits has shown a fault-controlled mosaic carbonate platform that is different from the carbonate ramp models. Within the platform mosaic, 15 lithofacies have been recognized, which are organized in four facies associations, including peritidal, restricted shallow sub-tidal, sand dunes and sub-tidal shelf facies associations. The rapid lateral and vertical facies changes, and the lack of consistent landward or seaward direction indicated by the pattern of facies changes, question the existence of a shoreline suggested by the traditional models for this region. Instead, the facies organization and cycle stacking pattern suggest deposition in a mosaic of intertidal islands between which sub-tidal restricted or open conditions could coexist in very close proximity. Such a platform mosaic would have been defined by tectonic activities along normal faults which segmented the shallow-water Causses platform. The facies and facies associations are arranged into metre-scale, peritidal and sub-tidal cycles that are also variable. Certain cycles show the same stacking pattern in all the sections and seem to be traceable over tens of kilometres. On the contrary, other cycles cannot be correlated; they are present only in specific sections and have a maximum lateral extension of 1 or 2 km. These metre-scale cycles stack to form four medium-scale cycles bounded by surfaces that display sub-aerial exposure features. Medium-scale cycles stack into two larger-scale cycles (tens of metres thick) and are bounded by well-defined karstic surfaces. Based on their lateral continuity and their stacking pattern, the metre-scale cycles are controlled probably by high frequency eustatic variations overprinting the topographic irregularities formed by differential subsidence of fault-bounded blocks. Episodic fault activities may reorganize the topography so that, even if eustatic changes may still be the major control of cycles, the expression and number of cycles could be different. Cycles of medium and large-scale are interpreted as being allogenic, controlled by changes in eustasy and/or subsidence rates as evidenced by their lateral continuity and the correlations of the large-scale cycles with third-order depositional sequences.
ISSN: 1469-3496

hal-00411940
https://hal.archives-ouvertes.fr/hal-00411940
DOI : 10.1111/j.1365-3091.2007.00896.x

Éditeur(s) : HAL CCSD Geological Society of America
Résumé : International audience
Integrated Ocean Drilling Program Expedition 313 continuously cored uppermost Eocene to Miocene sequences on the New Jersey shallow shelf (Sites M27, M28, and M29). Previously, 15 Miocene (ca. 23-13 Ma) seismic sequence boundaries were recognized on several generations of multichannel seismic profi les using criteria of onlap, downlap, erosional truncation, and toplap. We independently recognize sequence boundaries in the cores and logs based on an integrated study of core surfaces, lithostratigraphy and process sedimentology (grain size, mineralogy, facies, and paleoenvironments), facies successions, stacking patterns, benthic foraminiferal water depths, downhole logs, core gamma logs, and chronostratigraphic ages. We use a velocitydepth function to predict the depths of seismic sequence boundaries that were tested by comparison with major core surfaces, downhole and core logs, and synthetic seismograms. Using sonic velocity (core and downhole), core density, and synthetic seismograms, we show that sequence boundaries correspond with acoustic impedance contrasts, although other stratal surfaces (e.g., maximum fl ooding and transgressive surfaces) also produce refl ections. Core data are suffi cient to link seismic sequence boundaries to impedance contrasts in 9 of 12 instances at Site M27, 6 of 11 instances at Site M28, and 8 of 14 instances at Site M29. Oligocene sequences have minimal lithologic and seismic expression due to deep-water locations on clinoform bottomsets. Miocene sequences (ca. 23-13 Ma) were sampled across several unconformity clinothems (prograding units) on topset, foreset, and bottomset locations. Excellent recovery allows core-seismic integration that confi rms the hypothesis that unconformities are a primary source of impedance contrasts. Our core-seismic-log correlations predict that key seismic surfaces observed in other sub surface investigations without core and/or well logs are stratal surfaces with sequence stratigraphic signifi cance.
EISSN: 1553-040X

insu-00946235
https://hal-insu.archives-ouvertes.fr/insu-00946235
DOI : 10.1130/GES00858.1

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
The contourite depositional system (CDS) along the southwestern Iberian Margin (SIM), within the Gulf of Cadiz and offshore areas of western Portugal bear the unmistakable signal of Mediterranean Outflow Water (MOW) exiting the Strait of Gibraltar. This locality records key information concerning the effects of tectonic activity on margin sedimentation, the effects of MOW dynamics on Atlantic circulation, and how these factors may have influenced global climate. Over the last four decades, numerous studies have been conducted on the late Miocene, Pliocene and Quaternary sedimentary stacking pattern of Neogene basins along the SIM for both academic and resources exploration purposes. However, understanding of the region rests primarily on basic seismic stratigraphy calibrated with limited data from only a few exploration wells. The Integrated Ocean Drilling Program (IODP) Expedition 339 recently drilled five sites in the Gulf of Cadiz and two sites on the western Iberian margin. The integration of core and borehole data with other geophysical databases leads us to propose a new stratigraphic framework. Interpretation of IODP Exp. 339 data along with that from industry sources and onshore outcrop analysis helps refine our understanding of the SIM's sedimentary evolution.We identify significant changes in sedimentation style and dominant sedimentary processes, coupled with widespread depositional hiatuses along the SIM within the Cadiz, Sanlucar, Doñana, Algarve and Alentejo basins. Following the 4.5 Ma cessation of a previous phase of tectonic activity related to the Miocene–Pliocene boundary, tectonics continued to influence margin development, downslope sediment transport and CDS evolution. Sedimentary features indicate tectonic pulses of about 0.8–0.9 Ma duration with a pronounced overprint of ~ 2–2.5 Ma cycles. These more protracted cycles relate to the westward rollback of subducted lithosphere at the convergent Africa-Eurasia plate boundary as its previous NW–SE compressional regime shifted to a WNW–ESE direction. Two major compressional events affecting to the Neogene basins at 3.2–3 Ma and 2–2.3 Ma help constrain the three main stages of CDS evolution. The stages include: 1) the initial-drift stage (5.33–3.2 Ma) with a weak MOW, 2) a transitional-drift stage (3.2–2 Ma) and 3) a growth-drift stage (2 Ma-present time) with enhanced MOW circulation into the Atlantic and associated contourite development due to greater bottom-current velocity. Two minor Pleistocene discontinuities at 0.7–0.9 Ma and 0.3–0.6 Ma record the effects of renewed tectonic activity on basin evolution, appearing most prominently in the Doñana basin. Several discontinuities bounding major and minor units appear on seismic profiles. Quaternary records offer the clearest example of this, with major units of about 0.8–0.9 Ma and sub-units of 0.4–0.5 Ma. Sedimentation is controlled by a combination of tectonics, sediment supply, sea-level and climate. This research identifies time scales of tectonic controls on deep-marine sedimentation, specifically over periods of 2.5–> 0.4 Ma. Shorter-term climatic (orbital) mechanisms control sedimentation at time scales of ≤ 0.4 Ma. The role of bottom water circulation and associated processes in shaping the seafloor and controlling the sedimentary stacking pattern on continental margins has to be seriously reconsidered in future multidisciplinary studies. This is not only because of the common occurrence of sandy contourite deposits in deep water setting and their economic interest for hydrocarbon exploration, but principally because they archive the heartbeat of the interior Earth and therefore have important sedimentary and paleoceanographic implications.
ISSN: 0025-3227

hal-01356166
https://hal.archives-ouvertes.fr/hal-01356166
DOI : 10.1016/j.margeo.2015.09.013

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Alpine Corsica consists of a stack of variably metamorphosed units of continental and Tethys-derived rocks. It represents an excellent example of high-pressure (HP) orogenic belt, such as the Western Alps, exposed over a small and accessible area. Compared to the Western Alps, the geology of Alpine Corsica is poorly unraveled. During the 1970s-80s, based on either lithostratigraphic or metamorphic field observations, various classifications of the belt have been proposed, but these classifications have been rarely matched together. Furthermore, through time, the internal complexity of large domains has been progressively left aside in the frame of large-scale geodynamic reconstructions. As a consequence, major open questions on the internal structure of the belt have remained unsolved. Apart from a few local studies, Alpine Corsica has not benefited of modern developments in petrology and basin research. This feature results in several uncertainties when combining lithostratigraphic and metamorphic patterns and, consequently, in the definition of an exhaustive architecture of the belt. In this paper we provide a review on the geology of Alpine Corsica, paying particular attention to the available lithostratigraphic and metamorphic classifications of the metamorphic terranes. These data are completed by a new and exhaustive metamorphic dataset obtained by means of thermometry based on Raman Spectroscopy of Carbonaceous Material (RSCM). This technique provides reliable insights on the peak temperature of the metamorphic history for CM-bearing metasediments. A detailed metamorphic characterization of metasediments, which have been previously largely ignored due to retrogression or to the lack of diagnostic mineralogy, is thus obtained and fruitfully coupled with the available lithostratigraphic data. Nine main tectono-metamorphic units are defined, from subgreenschist (ca. 280-300 °C) to the lawsonite-eclogite-facies (ca. 500-550 °C) condition. These units are homogeneous in metamorphism, laterally continuous and have characteristic lithostratigraphic features. This study also suggests a direct link between the pre-orogenic extensional setting and the present-day compressional structure of Alpine Corsica, indicating that large sections of subducted lithosphere were subducted and exhumed as coherent domains. These features provide important insight on the mechanism of stacking and exhumation of HP rocks, and make Alpine Corsica a unique reference for mountain-building processes in Tethyan-type orogens.
ISSN: 0012-8252

hal-00814433
https://hal.archives-ouvertes.fr/hal-00814433
DOI : 10.1016/j.earscirev.2012.10.003

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
The Miradouro da Lua cliffs, which are 60 km south of Luanda, record the building and uplift of the late Neogene Palaeo-Cuanza delta. The detailed study of the sedimentary architecture and stacking pattern permitted separation of the pile into five depositional units bounded by erosional surfaces and characterised by separate facies associations (genetic sequences = units in this paper). At the base of the series, aeolian deposits (Unit 1) mark the development of a possible coastal desert during the late Miocene aridification. The major Pliocene sea-level rise (Transgressive Systems Tract) led to the drowning of the continental platform into a discrete shoreface-foreshore sequence (Unit 2), followed by an expanded deltaic sequence (Unit 3) that represents the main outcrop of the area. The sedimentary fabric of this prograding wedge during the Highstand Systems Tract reveals laterally stacked pluri-hectometic mouth bars built by the abrupt switching of a bird-foot delta during the Pliocene highstand. The clinoforms are deeply incised by submarine gullies filled both by periodic river-driven turbidite and tidal currents (Unit 4) during the coeval growing of the delta. The topset of the prograding wedge and associated gullies infill is truncated by an overall erosional unconformity that marks the widespread development of an extensive braid-delta system (Unit 5) during the lower Pleistocene sea-level drop (Lowstand Systems Tract). The last 6 m of the Braid-delta unit is overprinted by a ferallitic profile, forming the surface of the plateau and indicating long-term subaerial exposure and weathering processes consistent with the maximum warming of the middle to late Pleistocene interglacial periods. The successive abrupt shifts of the depositional systems through the sedimentary pile indicate a high-amplitude sea level amplified by major coastal uplifts and the reorganisation of the fluvial network. In this context, the palaeo-Cuanza prograding wedge signals the westward tilting and 60 m total uplift of the coastal plain during the Pliocene. A new uplift pulse by differential tilt and slope increase is deduced from the widespread development of the braid-delta complex during the early Pleistocene. The last tectonic event corresponds to the 80- to 100-m uplift and ferallitic paedogenesis during the middle to late Pleistocene interglacial stages.
ISSN: 0899-5362

hal-01053716
https://hal.archives-ouvertes.fr/hal-01053716
DOI : 10.1016/j.jafrearsci.2014.01.013

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Volumetrically minor microsyenites, alkali microgranite and related trachytic dykes intrude early Pliocene OIB-like alkali basaltic and basanitic flows of the Meseta del Lago Buenos Aires in Central Patagonia (47 degrees S-71 degrees 30'W), and occur together with scarce trachytic lava flows. Whole-rock K-Ar ages between 3.98 and 3.08 Ma indicate that the emplacement of these felsic rocks occurred more or less synchronously with that of the post-plateau basaltic sequence that they intrude, during a bimodal mafic-felsic magmatic episode devoid of intermediate compositions. Chemically, these rocks have A(1)-type granitoid affinities and are characterized by high silica and alkali contents (60-68 wt.% SiO2; 8.7-10.8 wt.% Na2O+K2O), major and trace elements patterns evidencing evolution by low-pressure fractional crystallization, and Sr and Nd isotopic signatures similar to those of coeval basalts ((Sr-87/Sr-86)o=0.70488-0.70571; (Nd-143/Nd-144)(o)=0.512603-0.512645). Nevertheless, some of them have the most radiogenic Sr values ever reported for a magmatic rock in the Meseta and even in the whole Neogene Patagonian Plateau Lavas province ((Sr-17/Sr-86)(o)= 0.70556-0.70571; (Nd-143/Nd-144)(o)=0.512603-0.512608). In addition, very high contents of strongly incompatible elements in the most evolved rocks, together with Sr isotopic ratios higher than those of coeval basalts, suggest the occurrence of open-system magmatic processes. Continuous fractional crystallization from a primitive basaltic source, similar to post-plateau coeval basalts, towards alkali granites combined with small rates of assimilation of host Jurassic tuffs (AFC) in a shallow magmatic reservoir, best explains the geochemical and petrographic features of the felsic rocks. Therefore, A(1)-type magmatic rocks can be generated by open-system crystallization of deep asthenospheric melts in back-are tectonic settings. In Central Patagonia, these similar to 3-4 Ma old alkaline intrusions occur aligned along a similar to N160-170 trending lineament, the Zeballos Fault Zone, stacking the morphotectonic front of one segment of the Patagonian Cordillera. Intrusion along this fault zone
ISSN: 0024-4937

hal-00412030
https://hal.archives-ouvertes.fr/hal-00412030
DOI : 10.1016/j.lithos.2007.07.002

Éditeur(s) : HAL CCSD Springer Verlag
Résumé : International audience
Unravelling the detailed pressure–temperature–time-deformation (P–T–t-D) evolution of magmatic and metamorphic rocks provides essential insights into the timing and duration of partial melting and related plutonism during crustal flow and migmatitic dome formation. The Montagne Noire Axial Zone (MNAZ) is a migmatitic dome located within the Variscan orogen in the southern French Massif Central. The timing of the main thermal event that was responsible for intense partial melting is still highly debated. In this study we present new laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) age data on micaschists, migmatites and granites that clarify the P–T–t-D evolution of the MNAZ. Structurally controlled samples were collected in order to constrain the timing of metamorphism, migmatization and plutonism regarding the main structural pattern D1, D2 and D3. D1 and D2 correspond to nappe stacking and dextral transpression, respectively. D3 is related to vertical shortening and coaxial thinning with a preferential NE–SW- to E–W-directed stretching. LA-ICP-MS analyses on the syntectonic Anglès, Soulié and Martys granites yielded U–Th/Pb monazite ages of 305 ± 1.5, 306 ± 1.9 and 314 ± 2 Ma, respectively. Five migmatitic rocks sampled in the eastern and central Espinouse area yielded in situ ages ranging between 312 ± 2 and 301 ± 2 Ma. Along the dome envelope, two garnet–staurolite-bearing micaschists near Saint-Pons-de-Thomières village gave in situ U–Th–Pb ages of 312.1 ± 2.1 and 309.0 ± 3.1 Ma. A fine-grained gneiss with a D3 fabrics in the eastern dome envelope yield a 208Pb/232Th mean age at 305.7 ± 3.9 Ma. All ages obtained in this study for the micaschists, migmatites and granites range between 315 and 301 Ma. We interpret this time span as the record of the high thermal event responsible for intense crustal partial melting within the lower and middle crust. The onset of partial melting occurred at ca. 315 Ma that marked the beginning of transpressional deformation D2. Based on structural and petrological studies, our new U–Th–Pb results suggest that (1) partial melting may have started at ca. 315 Ma and lasted 15–10 Myr and (2) D2 et D3 developed between 315 and 300 Ma and were synchronous. D1 deformation ended at 315 Ma. The onset and duration of D1 related to nappe stacking and crustal thickening is still uncertain.
ISSN: 1437-3254

hal-01526942
https://hal.archives-ouvertes.fr/hal-01526942
DOI : 10.1007/s00531-016-1417-x

Éditeur(s) : Université de Bucarest - Université de Bretagne occidentale
Résumé : This study is focusing on the architecture and recent sedimentary evolution of the Danube channel, the youngest channel-levee system in the Danube deep-sea fan. The study was conducted as part of the BlaSON French-Romanian Project, and combined high-resolution seismic-reflection profiles and chirp profiles with multibeam bathymetry and piston cores. This data set was acquired in 1998 during a joint survey IFREMER-GeoEcoMar of the north-western Black Sea. Previous seismic and acoustic data were also used. The Danube deep-sea fan is a large passive-margin mud-rich fan. Like the other systems of this type (Amazon fan, Mississippi fan or Indus fan) the Danube fan consists of stacked channel-levee systems intercalated with mass-transport deposits. Seismic and sedimentary facies in the Danube fan are similar to those identified in most of the mud-rich systems. Nevertheless, the Danube fan is distinguished by a specific feature: its development in a freshwater environment. This is due to the peculiar water-level history of the Black Sea controlled by the link to the Mediterranean through the Strait of Bosphorus and the Sea of Marmara. This connexion was successively interrupted (during sea level lowstands) and re-established (when the sea level was rising above the Bosphorus). Temporary absence of marine water influx during lowstands together with large freshwater inputs from the Danube and other major rivers changed the Black Sea into a freshwater lake during times of fan activity. This peculiarity possibly favourised the development of hyperpycnal flow at the Danube mouth and the initiation of turbidity currents in the deep-sea fan. The Danube channel is directly connected to the large shelf-indenting Danube canyon (also known as Viteaz canyon). The Danube canyon is deeply incised into the shelf margin for 26 km landward of the shelfbreak. During lowstands this canyon acted as the most important path for sediment supply to the deep sea in this part of the continental margin. It consists of a main trough with steep flanks, and a meandering thalweg cut into the flat canyon floor, attesting for the development of the canyon by erosion in the entrenched axial thalweg. Sections with specific morphology, orientation and gradient identified along the canyon, are interpreted as phases of landward expansion of the canyon. Internal structure of the canyon shows several erosional surfaces, which indicate that the present morphology of the canyon is the result of its polyphasic evolution. Instability in the zone of the canyon is related to the important sediment supply at the Danube mouth, to the presence of the gas in the surficial sediment, and possibly under a structural control. The upper part of the Danube channel (between the Danube canyon and ~1400 m depth) consists in a single leveed-channel that has undergone significant overbank deposition, as attested by the well-developed levees. The levees are strongly asymmetrical, being higher and wider on the right-hand side looking downstream. This type of asymmetry is rather common in deep-sea fans, and is generally attributed to the Coriolis effect (Menard, 1955). The channel is slightly sinuous, partially filled and incised by an entrenched thalweg, connected to the axial thalweg of the Danube canyon. Detailed seismic investigation inside the channel trough documented several depositional phases within the channel fill, separated by erosional surfaces. These surfaces are associated with distinct terraces identified on the multibeam bathymetry, that can be followed downward along the main trough axis. The valley fill deposits (where not removed by the subsequent erosional event) show an axial HAR (High Amplitude Reflections) seismic facies with lateral lower amplitude continuous reflections consisting in a levee facies, as proved by sampling. This indicates that filling up was associated with flow within the channel, and not with interruption of fan activity. On the middle slope below 1400 m, this single channel bifurcates through repeated avulsions. As a result, several highly meandering channels developed. The onlap relationships between these channels indicate that only one channel was active at a time. Each phase of avulsion resulted in a depositional unit consisting in a basal unchannelized lobe defined as High Amplitude Reflection Packets (HARP, Flood et al., 1991) that underlies a channel-levee system. The deposition of HARPs was associated with the readjustment of the longitudinal profile of the channel after the breaching of a levee, which resulted in remobilization of upslope channel deposits and eroded levees. When this adjustment was complete, erosion ceased and levees began to develop above the HARPs (Pirmez et al., 1997). All the identified phases of avulsion followed the same pattern: (1) breaching of the lower and narrower left levee; (2) building of a unit of High Amplitude Reflector Packets (HARP) basinward of the bifurcation point by the unchannelized flow, while the former channel was abandoned; and (3) initiation of a new meandering leveed channel. The northward migration of the resulting units through repeated bifurcations is influenced by the asymmetry between levees (hence by the Coriolis effect), and confined between the high levees of the initial phase of the Danube channel (to the south) and the steep relief of the Dniepr fan (to the north). Structure of the fan valley fill indicate that the erosional surfaces inside the upper channel could be formed in response of successive avulsions, by the adjustment of the longitudinal profile of the channel following the breaching of a levee wall. Sediments removed by erosion formed the HARP lobes basinward of the avulsion point. When this adjustment was complete, a channel-levee system developed downward of the bifurcation, overlying the HARPs, but also upward of this point, as a confined channel-levee system inside the erosional trough of the fan valley. Fluvial incisions identified on the continental shelf, together with the coastline location during the last active period of the Danube channel, indicated that the paleo-Danube was directed towards the head of the Danube canyon. Paleo-Danube mouth was fairly close (ca. 10 km) to the Danube canyon, supplying sediment to the Danube channel. Furthermore, hyperpycnal flow probably prevailed in the freshwater environment that characterized the Black Sea during times of fan activity. These conditions would have enabled the development of a quasi-continuous river-canyon-deep-sea fan system, ensuring the effective transfer of the sediment between the coastal zone and the deep sea.
Ce travail est consacré à l'étude de l'architecture et de l'évolution sédimentaire récente de l'éventail profond du Danube, en particulier de son dernier système chenal-levée: le chenal du Danube. L'étude a été réalisée dans le cadre du projet de coopération franco-roumaine BlaSON, à partir des données sismiques, bathymétriques-acoustiques et sédimentologiques acquises en 1998 lors d'une campagne en mer Noire réalisée par IFREMER et GeoEcoMar. Des données sismiques et acoustiques antérieures ont été également utilisées. L'éventail profond du Danube s'enserre dans la catégorie des grands éventails vaseux. Comme les autres systèmes de ce genre (tel que les éventails de l'Amazone, du Mississippi ou de l'Indus) il est constitué d'une succession de systèmes chenaux-levées intercalés avec des dépôts de transport en masse. Le fonctionnement de l'éventail était conditionné par la baisse du niveau marin lors des périodes glaciaires. Ses faciès sismiques et sédimentaires s'apparentent aux faciès qui caractérisent la plupart des éventails de ce type. L'éventail du Danube constitue néanmoins un cas particulier parmi les autres systèmes étudiés, du fait de son fonctionnement dans un bassin lacustre. Ceci est dû à la situation spécifique de la mer Noire dont la connexion avec la Méditerranée, par le détroit de Bosphore et la mer de Marmara, a été successivement interrompue (au cours de périodes de bas niveau) et reprise (quand le niveau marin remontait en dépassant le seuil du Bosphore). L'absence de l'apport d'eau salée pendant les périodes glaciaires, associée avec l'augmentation de l'apport fluvial, ont déterminé l'installation d'un milieu d'eau douce dans la mer Noire à chaque fois que la baisse du niveau permettait la reprise du fonctionnement de l'éventail profond. Cette situation particulière aurait favorisé la formation des courants hyperpycnaux à l'embouchure d'un fleuve du débit du Danube qui déversait ses eaux turbides dans un bassin lacustre, et aurait donc influencé l'apparition de courants de turbidité dans l'éventail profond. Le chenal du Danube s'est développé sur la pente continentale en prolongation du canyon du Danube (ou Viteaz) auquel il est directement connecté. Le canyon est incisé de manière significative (26 km) dans la plate-forme continentale. Au cours de bas-niveaux marins il constituait la principale voie de transfert des sédiments terrigènes vers le bassin profond dans cette partie de la marge. Le canyon est constitué par une entaille avec des flancs abrupts et un talweg axial incisé, qui montre l'importance du processus d'érosion du fond pour le développement du canyon. Les segments qui ont été identifiés le long du canyon, avec des morphologies, des orientations et des pentes spécifiques, sont interprétés comme des phases d'avancement du canyon vers la côte. Plusieurs incisions sont visibles dans la structure interne du canyon et témoignent que la morphologie actuelle du canyon est le résultat de son évolution polyphasée. L'instabilité de la zone du canyon est en relation avec les apports sédimentaires importants à l'embouchure du Danube, avec la présence du gaz dans les sédiments superficiels, et possiblement sous un contrôle structural. Sur la pente supérieure (entre le canyon du Danube et environ 1400 m de profondeur) le chenal du Danube présente des levées bien développées et fortement asymétriques, avec la levée droite plus haute et plus large que la levée gauche. Ce type d'asymétrie, fréquemment décrit dans les éventails profonds est généralement attribué à l'effet Coriolis (Menard, 1955). Le chenal est légèrement sinueux, partiellement comblé et incisé par un talweg axial qui représente la continuation sur la pente du talweg incisé dans le canyon du Danube. L'analyse sismique détaillée du remplissage de la vallée montre plusieurs phases de dépôt, séparées par des discontinuités érosives. Ces surfaces d'érosion correspondent à des terrasses emboîtées, relativement parallèles le long de la vallée, visibles dans la bathymétrie. Les dépôts qui constituent le remplissage du chenal présentent un faciès sismique de type HAR (High Amplitude Reflections) dans l'axe du chenal, partiellement (ou parfois totalement) enlevé par les phases d'érosion subséquentes, qui continue latéralement avec des réflexions litées correspondant à un faciès sédimentaire de levée. Le remplissage de la vallée a été donc associé avec des écoulements dans le chenal, et non pas avec l'interruption de son fonctionnement. Sur la pente inférieure, le chenal unique bifurque plusieurs fois par avulsion et forme de nouveaux systèmes chenaux-levées méandriformes. Ces systèmes se succèdent verticalement en onlap, ce qui montre qu'un seul chenal a été actif à la fois. Chaque phase d'avulsion a eu comme résultat la mise en place d'une unité constituée par un lobe défini comme "High Amplitude Reflection Packets" (HARP, Flood et al., 1991) à la base, et un système chenal-levée au sommet. Le dépôt d'un lobe HARP est associé avec de l'érosion dans le chenal en amont du point d'avulsion pour l'ajustement de son profil après la rupture de la levée. Quand le chenal a retrouvé son profil d'équilibre, l'érosion a cessé et des levées ont commencé à se développer au-dessus des HARPs (Pirmez et al., 1997). Toutes les phases d'avulsion se sont développées d'après le même modèle: (1) la rupture de la levée gauche, plus étroite; (2) le dépôt d'un lobe HARP par les écoulements non-chenalisés en aval du point d'avulsion, et l'abandon de l'ancien chenal; (3) l'initiation d'un nouveau système chenal-levée. La migration systématique du chenal vers le nord est influencée par l'assymétrie des levées (donc par la force de Coriolis), et confinée entre les grandes levées de la phase initiale du chenal du Danube, au sud, et le relief abrupt de l'éventail du Dniepr au nord. La structure sédimentaire du chenal du Danube indique que les surfaces érosives à l'intérieur du remplissage du chenal se seraient formées en réponse aux avulsions, du fait de l'ajustement du profil du chenal après la rupture d'une levée. Les sédiments du chenal érodés au cours de ce processus ont formé les lobes HARP. Quand le chenal a retrouvé son profil d'équilibre, un système chenal-levée s'est développé en aval du point d'avulsion au dessus du lobe HARP, mais aussi en amont de ce point, où il se trouve confiné dans la vallée érosive. Les incisions fluviatiles identifiés sur la plate-forme continentale et la position de la ligne de côte pendant la dernière période d'activité du chenal du Danube montrent que le paléo-Danube se dirigeait directement vers la tête du canyon du Danube. Son embouchure était située à proximité du canyon, qui alimentait le chenal du Danube. Cependant, les courants hyperpycnaux devaient prévaloir dans le milieu de salinité réduite qui caractérisait la mer Noire lors des périodes actives de l'éventail. Ces conditions auraient favorisé la mise en place d'un système quasi-continu fleuve-canyon-éventail profond, qui contrôlait le transfert des sédiments entre la côte et le bassin profond.

Droits : info:eu-repo/semantics/openAccess
http://archimer.ifremer.fr/doc/2002/these-1206.pdf
http://archimer.ifremer.fr/doc/00000/1206/