GeoMod2010 Lisbon, 27-29 September 2010

1

G. Lymer (1), B. C. Vendeville (1), V. Gaullier (2) and J. Déverchère (3)

(*) Université Lille-Nord de France, Laboratoire Géosystèmes, FRE CNRS 3298, 59 655 Villeneuve d’Ascq cédex (France) (*)Laboratoire IMAGES, Université de Perpignan Via Domitia, 52 avenue Paul Alduy, 66860 Perpignan Cedex, (France) (*)Domaines Océaniques, UMR 6538, University of Brest (France) Introduction The Algerian margin is located north of the African continent and south of the Algero-Provencal basin (Fig. 1), which formed as a back-arc basin during subduction of the Tethyan ocean under the African plate during Oligo-Miocene times (Auzende et al., 1974). Back-arc extension started around 30–35 My ago, and ended when the internal zones of the arc, known as the AlKaPeCa domain (Al: Alboran, Ka: Kabylies, Pe: Peloritan, Ca: Calabria) drifted southward until it finally collided with the African plate 18-15 My ago (Gueguen et al., 1998; Frizon de Lamotte et al., 2000).

Figure 1: Bathymetric map showing the location of the Algerian margin (Obone-zué-obame., 2009)

In addition, the Algero-Provençal basin was affected by the Messinian Salinity Crisis (about 6 My ago), which lead to deposition of large volumes of mobile evaporites in the entire basin. Evaporite deposition was caused by the closure of the strait of Gibraltar following the rollback of the subduction slab toward the south, and the uplift of the Gibraltar area. Such closure disconnected the Mediterranean Sea from the Atlantic Ocean, allowing for massive evaporation of the Mediterranean sea water (Krijgsman et al., 1999). Déverchère et al. (2005) have demonstrated the presence of active crustal reverse faults off Algiers and Boumerdes, one of which would have cause the Boumerdes earthquake of 2003 (Déverchère et al., 2005 et Domzig et al., 2006). The line drawing in Figure 2 and the seismic profile in Figure 3, both oriented N-S, perpendicular to the coastline, show that movements along these thrusts faults have generated an uplifted plateau and a piggy-back basin, thereby deforming the overlying series made of Messinian evaporites and Plio-Quaternary overburden. It is important to point out that these thrusts do not extend laterally to the West and East of this particular margin segment, which suggests that regional compression is less intense along the adjacent segments.

Geometry and evolution of salt structures along the Algiers sector of the margin In the area uplifted by the basement thrusts, the geometry of the salt structures varies greatly whether observed along dip-oriented profiles (Figs 2 and 3) or strike-oriented profiles (Fig. 4). Dip profiles show only a few salt-cored anticlines that grew during Pliocene times, but are no longer active. The seismic resolution does not allow determining whether these structures started early, as the salt-overburden contact is not well imaged.

INFLUENCE OF REGIONAL TECTONIC STRESSES ON THE EVOLUTION OF GRAVITY SPREADING STRUCTURES ABOVE SALT: THE EXAMPLE OF THE ALGERIAN CONTINENTAL MARGIN TESTED BY PHYSICAL EXPERIMENTS

GeoMod2010 Lisbon, 27-29 September 2010

2

Figure 2: Line drawing (Déverchère et al., 2005) from interpreted seismic profile illustrating basement thrusts uplifting the pre-Messinian “basement” and the Messinian and post-Messinian series. Figure 3: Dip-oriented seismic profile (NS)

across the margin showing salt-cored anticlines active during Pliocene times (Obone-zué-obame., 2009). The geometry of the salt structures is, however, radically different on strike-oriented profiles (Figs. 4). There, the overburden, including the Upper evaporites, has deformed early, forming distinct overburden blocks separated by salt ridges having triangular geometry in cross section. This geometry is typical of that of reactive diapirs that form during thin-skinned extension above salt, as occurs during gravitational spreading of a sediment lobe.

Figure 4: Strike line (E-W) across the margin showing numerous salt structure that started to form early and continued

to grow as late as present (Obone-zué-obame., 2009)

Figure 5: Detail of the above profile illustrating the triangular geometry of the salt bodies, typical of en extensional origin (Obone-zué-obame., 2009)

Hypotheses on the evolution of salt structures along the Algiers sector of the margin The presence of numerous overburden blocks separated by salt ridges indicates that, as early as the time when the first post-Messinian clastic sediments were deposited, the sedimentary edifice spread gravitationally under its own weight, as has been described by Vendeville (2005) and Gaullier and Vendeville (2005). Typically, where the sediments are deposited as lobes, as opposed to linear wedges trending parallel to the margin, the edifice spread radially, forming both concentric and radial faults and associated salt ridges. Such an evolution would be compatible with the presence of several fans along this segment of the margin. However, one has to explain the disparities between the margin-parallel and margin-perpendicular profiles, which indicates that N-S (margin-perpendicular) structures formed early and lasted until now, whereas E-W-trending extensional salt structures either did not form (or they grew only during the earliest stages), stopped early then were reactivated as salt-cored anticlines during the Plio-Quaternary times.

GeoMod2010 Lisbon, 27-29 September 2010

3

One possible parameter could be the regional NS compressional stress field that applies to this particular segment of the margin. Initially, immediately after deposition of the Messinian mobile evaporites, the first sediment lobes could freely spread radially because in the absence of distal overburden, the compressional stresses could not be transmitted to the sediment lobes. This set up gradually changed as the distal overburden thickened by deposition of turbiditic and pelagic series, thereby transmitting more and more of the regional compressional stress onto the spreading lobe. This, in turn opposed further spreading to the North, effectively blocking any E-W-trending normal faults. By contrast, N-S-trending faults, could continue to grow because some accommodation existed both to the East and the West of the area. The resulting pattern was asymmetric, with E-W normal faults and associated salt ridges that aborted early, whereas N-S structures continued to grow. Experimental modelling The experiment was designed following the observations and hypotheses described above. A tabular 2.2-cm-thick layer of viscous silicone polymer (SGM 36, from Dow Corning; density:1), simulating the mobile Messinian evaporites rested on a rigid, horizontal substrate. As analogues of the brittle Plio-Quaternary overburden, we used alternating layers of fine dry quartz sand (density 1.5) and PVC powder (density : 0.8) in order to obtain an average density similar to that of the mobile layer, as is the case in the Western Mediterranean.

Figure 6. Initial set up of the experiment (see from above). The red body is the initial sediment lobe.

First, we deposited an initial lobe made of PVC and fine sand (black, white, and green, each 3 mm thick) then let it spread radially (Fig 7 A) so that both radial and concentric faults and salt ridges formed.

Figure 7. Overhead photographs of the experiment. A: early stage; B: after the onset of compression; C: Final stage; D: Top of the mobile layer after removal of the overburden

After 30 hours, we deposited a thin (2mm) distal series of PVC across the entire model, simulating the distal overburden, and we started applying some N-S shortening in the middle segment of the model (Fig. 7B). We continued adding episodically additional layers of sand and PVC on top of the initial lobe, then stopped the experiment after a total running time of 84 hours. As soon as regional shortening started, differences in evolution appeared between the NS-trending salt structures and the EW-trending ones. NS structures continued to grow, whereas the EW-stuctures,

GeoMod2010 Lisbon, 27-29 September 2010

4

whose growth was impeded by the increasing NS compressional stress, started to slow down and eventually stopped altogether. At the end of the experiment, we vacuumed the sand/PVC overburden and shot photographs of the top of the mobile layer (Figs. 7 D and 8). The photographs clearly show that EW-trending are the highest and have pierces the youngest overburden strata, whereas EW-trending ridges have stopped rising much earlier. Conclusion The experiment results described in this article are in good agreement with the hypotheses raised by the analysis of data from the Algiers segment of the Algerian margin. In the absence of thick distal overburden and significant regional compressional stress field, sediment lobes can freely spread radially, leading to the formation and growth of radial and concentric grabens and underlying salt ridges. The presence of a thick distal overburden, capable of transmitting a regional stress field tends to block further growth of the EW-trending fault and ridge family, thus promoting the growth of only the NS-trending structures.

Figure 8. Oblique photographs of the experiment after removal of the overburden

REFERENCES • Auzende J.M., Bonnin J. et Olivet J.L., 1974. Le détroit sardano-tunisien et la zone de fracture nord-tunisienne.

Tectonophysics, 21, 357–374. • Déverchère, J., Yelles, K., Domzig, A., Mercier de Lépinay, B., Bouillin, J.-P., Gaullier, V., Bracène, R., • Calais, E., Savoye, B., Kherroubi, A., Le Roy, P., Pauc, H., and Dan G. (2005), Active thrust faulting offshore

Boumerdes, Algeria, and its relations to the 2003 Mw 6.9 earthquake, Geophys. Res. Lett., 32, L04311, doi:10.1029/2004GL021646.

• Domzig A., 2006. Déformation active et récente, et structuration tectono-sédimentaire de la marge sous-marine algérienne. Thèse de l’université de Bretagne occidentale, Brest, 343p.

• Domzig A., Yelles K., Le Roy C., Déverchère J., Bouillin J.-P., Bracène R., de Lépinay B. M., Le Roy P., Calais E., Kherroubi A., Gaullier V., Savoye B., Pauc H., 2006. Searching for the Africa-Eurasia Miocene boundary offshore western Algeria (MARADJA’03 cruise). C. R. Geoscience, 338: 80–91.

• Frizon de Lamotte, D., Saint Bezar, B., Bracène, R., et Mercier, E., 2000. The two main steps of the Atlas building and geodynamics of the western Mediterranean, Tectonics, 19, 740-761.

• Gaullier, V., Le Roy, P., Bettich, M., Déverchère, J., Mercier de Lépinay, B., Domzig, A., Sage, F., and the MARADJA Team (2005), The Sedimentary and Tectonic Consequences of the Messinian Salinity Crisis on the Algerian Margin, Southwestern Mediterranean: Insights from the MARADJA Cruise, EGU meeting, Vienna, EGU05-A-09984.

• Gaullier, V. and Vendeville, B. C., 2005, Salt tectonics driven by sediment progradation: Part II – Radial spreading of sedimentary lobes prograding above salt. American Association of Petroleum Geologists Bulletin , v. 89, pp. 1081-1089.

• Gueguen E., Doglioni C., Fernandez M., 1998. On the post-25 Ma geodynamic evolution of the western Mediterranean. Tectonophysics, 298: 259–269.

• Krijgsman W., Hilgen F.J., Raffi I., Sierro F. J. et Wilson D.S., 1999a. Chronology, causes and progression of the Messinian salinity crisis. Nature 400: 652–655.

• Obone-zué-obame E.M., 2009. Conséquences sédimentaires et tectoniques de la crise de salinité Messinienne en Méditerranée occidentale. Thèse de l’Université de Perpignan Via Domitia. 290p.

• Vendeville, B. C., 2005, Salt tectonics driven by sediment progradation: Part I – Mechanics and kinematics. American Association of Petroleum Geologists Bulletin , v. 89, pp. 1071-1079.