Submarine slides during relative sea level rise: three late-Quaternary examples from the Central Mediterranean

SUBMARINE SLIDES DURING RELATIVE SEA LEVEL RISE: TWO EXAMPLES FROM THE EASTERN TYRRHENIAN MARGIN F. TRINCARDI, A. CATTANEO, A. CORREGGIARI, S. MONGARDI Istituto di Geologia Marina (CNR), v. Gobetti 101, 40129 Bologna, Italy A. BREDA, A. ASIOLI Dipartimento di Geologia, Paleontologia e Geofisica, v. Giotto 1, 35137, Padova, and Istituto di Geoscienze e Georisorse (CNR), c. Garibaldi 37, 35137, Padova, Italy

Abstract Two extensive mass-failure deposits originated during the late-Quaternary sea level rise on the eastern Tyrrhenian margin. The deposits that failed had markedly different architectures: offshore Cape Licosa, a shelf-margin low-stand wedge failed along its basal downlap surface; in Paola slope basin, extensive failure on the upper slope involved a few-m-thick mud drape and older consolidated units. Regardless of their geometric differences, both failures occurred close to an interval of accelerated lateQuaternary sea-level rise (ca.13.8 cal. kyr BP). This evidence suggests that rapid drowning of unconsolidated sediment resulted in increased water load; enhanced pore pressure played a role in favoring failure. Keywords: Sea-level rise, melt-water pulse, sediment failure, weak layer, geochronology 1. Introduction Sea level lowering is commonly invoked as an important predisposing factor or potential trigger for sediment failure of unconsolidated sediment deposited during previous highstand conditions on continental shelves and slopes (Vail et al., 1977; Coleman et al., 1983; Mutti, 1985). On Mediterranean margins, a significant portion of the PlioQuaternary stratigraphic record consists of mass-failure deposits including deep-sea fans (Droz and Bellaiche 1985) and base-of-slope deposits (Farrán and Maldonado 1990; Nelson et al. 1991). These are clear examples of extensive deep-water mass wasting deposits originated during the low sea level stand of the Last Glacial Maximum (LGM). Despite this compelling evidence, studies from Quaternary continental margins increasingly document sediment failure during times of relative sea-level rise (Torres et al., 1995; Lastras et al., 2002); these results hint to a more complex relation between changing relative sea level and the generation of mass wasting (Ross et al., 1994). Additional examples of sediment failure during intervals of sea level rise come from two deposits (the Licosa slide and the Paola Basin mud flow deposit) on the eastern Tyrrhenian margin (Gallignani, 1982; Trincardi and Field, 1992; Trincardi et al., 1995). Herein, we review the architecture and age of these deposits to determine if a relation can be established between sediment failure and precise intervals of increased rates of sea level rise during the late Quaternary (Fig. 1). This relationship, however, does not exclude the role of seismicity (Soloviev, 1990) in triggering failure on the tectonicallyactive eastern Tyrrhenian margin. 469

470

Trincardi et al.

2. Methods This paper is based on the interpretation of 3.5-kHz sub-bottom profiles collected using a 4-transducer hull-mounted system fired at 0.5-s intervals with a pulse length of 2 ms. Navigation relied on Loran C and limited D-GPS control to correct the data and shift them to a WGS84 datum. Fourteen sediment cores were analysed through sedimentologic description, magnetic-susceptibility scans, grain-size analyses and micro-paleontological content. Geo-chronological control comes from biostratigraphic correlations based on planktic foraminifera ecostratigraphy (Capotondi et al., 1999; Asioli et al., 1999) supported by AMS 14C dates and tephrochronology (Calanchi et al., 1996). Whole-core magnetic susceptibility logs allowed precise physical correlations among coring sites.

Figure 1. Location of Licosa (LS) and Paola Basin (PB) slides on the eastern Tyrrheninan margin. Both slides occur on the upper slope and resulted in

3. Results We reconstruct the key architectural elements of two complex mass-failure deposits on the eastern Tyrrhenian margin (Fig. 1). By reconstructing geometries and stratigraphic contexts we expect to improve the definition of the timing and mechanisms of these failure deposits. The timing of both failures is constrained by the age of the underlying deposits and overlying sediment drapes. Herein we adopt the event stratigraphy proposed for the late-Glacial-Interglacial Transition by Bjorck et al. (1998) and extended to the Mediterranean as proposed by Asioli et al. (1999). 3.1 LICOSA SLIDE Setting. Offshore Cape Licosa, the growth of a tectonic slope ridge generated a narrow E-W-trending slope basin in water depths between 250 and 300 m. Along its axis, this elongated basin shows a slightly shallower sill separating an eastern confined basin from a western basin connected to a deeply-incised canyon system. On the upper slope, Licosa slide generated rafted blocks and flow deposits that accumulated in both subbasins (Figs. 2 and 3). The unit that failed consists exclusively of shelf-margin progradational deposit that accumulated during the Last Glacial Maximum (Trincardi and Field, 1991). The original internal geometry of these deposits is characterized by low-angle downlapping reflectors that are truncated at their top by a shelf-wide erosion surface and converge seaward into a draped unit of plane parallel reflectors (Trincardi

Submarine slides during relative sea-level rise, eastern Tyrrhenian margin

471

and Field, 1992). Sediment cores show that the prograding deposit is composed of finegrained sand in proximal areas and becomes muddy on the upper slope; volcanogenic material, including large pumice pebbles, is present at several stratigraphic intervals. The depocentre of the shelf-margin deposit that partially failed is elongated parallel to the shelf margin indicating the importance of near-shore sediment redistribution at sea level low stand. On seismic profiles, the downlap surface at the base of the shelf-margin deposit is affected by short-distance amplitude variations and irregularities that are consistent with the presence of gas charged sediment accompanied by incipient deformation (Fig. 3, upper). Licosa slide is about 30 km2 in extent and occupies a restricted portion of this broader mobilisation area (Fig. 2). Gentle compression above the basal surface occurs upslope of the slide scar (Trincardi and Field, 1992 and Fig. 3). Architecture. Licosa slide is composed of distinctive depositional elements above the basal mobilisation surface (Figs. 2 and 3): 1) Remnant blocks, seaward of the slide scar, are characterized by a very irregular bathymetry, locally accompanied by diffraction hyperbolae on 3.5-kHz profiles (Fig. 3). The blocks are in situ on the basal slip plane; no evidence of rotation is observed because: a) where detectable, plane-parallel reflectors within the blocks parallel the basal surface and the section below it; b) the block height never exceeds the thickness of the failed progradational wedge at the slide scar; c) no blocks appear to have been thrusted thereby generating duplications of section.

Figure 2. Location and internal subdivisions of Licosa slide, affecting a progradational shelf-margin deposit along its basal downlap surface. The thickness distribution of the shelf-margin deposit exceeds 45 ms in the depocenter but less than 15 ms in the area affected by failure. Note that evidence of sediment mobilisation on the basal downlap surface is broader than the extent of the failure itself.

472

Trincardi et al.

Figure 3. Seismic profiles illustrating the main elements of Licosa slide. The evidence of mobilization on the basal downlap surface extends landward, beyond the slide scar. See Fig.2 for patterns on profile location map.

2) Exposed basal surface is at the seafloor in steeper slope areas (comprised between 1Û and 3Û ZKHUHWKHHQWLUHPRELOLVHGVHFWLRQZDVUHPRYHG)LJVDQG IXUWKHUVHDZDUG the basal surface plunges beneath the mass-transport deposits that fill the slope basin. 3) Slope basin fill presents two styles. In the eastern basin, extensive slide slabs have preserved internal stratigraphy and appear only gently folded indicating that compression occurred against the walls of the slope basin; in the western basin, translated deposits underwent more intense remolding as indicated by the presence of acoustically transparent seismic facies. Part of these deposits spilled into a canyon and moved further downslope to the west. Age of slide. The downlap surface at the base of the sediment section that failed represents the base of the Last Glacial Maximum (LGM). Failure occurred after lowstand progradation had ceased, because the slide scar is not filled by similar progradational deposits. A mud unit of roughly uniform thickness drapes the slide area (Fig. 4); micro-paleontological determinations indicate that this drape encompasses the interval since the upper portion of GI1 interstadial (Bijork et al., 1998; Asioli et al., 1999). Interestingly, in all cores the lower portion of GI1 is reduced or missing at the base of the post-slide drape; this stratigraphic incompleteness extends to the C2 tephra

Submarine slides during relative sea-level rise, eastern Tyrrhenian margin

473

(Neapolitan Yellow Tuff, ca. 14.3 cal. kyr BP), even though the study area is not far from the source of this widely reported Mediterranean tephra (Calanchi et al., 1996).

Figure 4. Physical correlation scheme for Licosa slide (based on magnetic susceptibility logs) with main biostratigraphic control from cores. The 14C date on a boreal mollusc (P. septemradiatum) close to the basal downlap surface is consistent with a LGM origin of the progradational wedge (dark gray, below). The post slide drape (light gray on core correlation scheme) encompasses the last ca. 13,000 cal years BP. Upper and lower black intervals within Holocene drape denote climatic optimum and sapropel S1, respectively.

3.2 PAOLA BASIN MUD-FLOW DEPOSIT Setting. A basin-wide (> 600 km2) failure deposit is found on the sea floor of Paola slope basin (Fig. 5). The deposit is elongated parallel to the regional slope in water depths between 400 and 700 m. The deposit is 5x109 m3 in volume and appears separated into two major elements by the depositional relief created by an older overbank wedge. The mass-failure deposit is composed of fine-grained sediment derived from the failure of a relatively thin section of late Quaternary deposits over a broad slope region. Although

474

Trincardi et al.

difficult to resolve on the steep upper slope, several minor coalescing slide scars are found along the shelf edge region; the failed section was quite thin but failure affected a broad area. An acoustically transparent drape covers the slide and adjacent sea floor. On the slide, where sea-floor topography is irregular, the drape is not resolved on 3.5-kHz profiles, but can be resolved in cores and dated.

Figure 5. Thickness distribution of Paola Basin slide. Maximum thickness is in the proximal subunit at the base of the upper slope. Note deposit thickening in the distal subunit in an area of sub-horizontal sea floor.

Submarine slides during relative sea-level rise, eastern Tyrrhenian margin

475

Figure 6. Sesismic profile across the distal portion of Paola Basin slide and adjacent basin floor where cores were collected to constrain the age of the deposits.

Figure 7. Core correlation scheme through Paola Basin slide and adjacent undeformed basin floor deposits. Biostratigraphy from Capotondi et al. (1999) and 14C dates from Mongardi (1994) in calibrated years BP. Timing of slide deposit is within GI-1 (Asioli et al., 1999). Black intervals within Holocene are as in Fig. 4.

Architecture. Three subunits characterize the deposit on 3.5-kHz profiles: 1) The proximal subunit (Figs. 5 and 6) shows hyperbolic returns that correspond to the abrupt edges of larger blocks. This subunit occupies the base of the slope down to 550600 m water depth, where gradients are between 2Û DQG Û DQG KDV D OLPLWed alongmargin extent. The thickness of this subunit is up to 25 m.

476

Trincardi et al.

2) The transitional subunit (Figs. 5 and 6) is widespread but rather thin (less than 4 m) and occurs where the gradient on the sea floor is typically 1.5Û WKH DFRXVWLF IDFLHV LV transparent with minor hyperbolae on the irregular top. 3) The distal subunit rests on the sub-horizontal sea floor of the slope basin and shows a greater thickness (up to about 8 m) compared to the transitional subunit (Figs. 5 and 6). The distal subunit is transparent on acoustic profiles and rests on turbidite-bearing plane-parallel beds through an erosional contact (Fig. 6). Age of slide. Sediment cores reached the mass-failure deposit and older units on the adjacent basin floor, beneath a uniform mud drape of about 2 m (Fig. 7). The age of the mass-failure deposit is between 13 and 14 cal. kyr BP. On the basin floor adjacent to the mass-failure deposit, gravity cores penetrated several turbidite beds deposited during the Last Glacial Maximum and in the lower part of GI-1. Seismic profiles and core correlation allow tracing this turbidite-bearing unit under the mass-failure deposit (Fig. 6), constraining its age to be younger than approximately 14 cal. kyr BP. This age assignment is consistent with the age of the basin wide drape mantling the deposit. This mud drape contains the tephra layer of the 79 A.D. Plinian eruption of Vesuvius and a thin turbidite bed of ca 12.9 cal. kyr close to its base (Fig. 7). The recognition of main planktonic foraminifera ecozones typical of Allerod and younger intervals (Capotondi et al., 1999) confirms that the drape deposited during the last ca. 13.5 cal. kyr (Tab. 1). Table 1. Samples calibration of AMS 14C ages in cores PB91-1, -2, and -8 according to Stuiver and Reimer (1993). Regional mean value of delta R for the Tyrrhenian Sea cames from “The marine reservoir correction database” compiled by Paula Reimer available online at: http://depts.washington.edu/qil/marine/. *Calibrated age is given as: 1sigma maximum cal. age (cal. age intercepts) minimum cal. age. Depth in core CAMS calibrated age 14 Core C Radiocarbon AMS age lab n. (cm) yrs BP* PB91-1 12134 186-188 10190±70 12894(12720)12419 PB91-2 12136 144-145 10180±60 11287(10946)10835 PB91-2 12132 173-175 11340±90 12971(12888)12659 PB91-2 12133 247-249 13240±70 15530(15338)14409 PB91-2 12135 403-405 17050±90 20004(19672)19353 PB91-8 12137 62-64 7270±60 7736(7669)7618 PB91-8 12127 93-94 9490±70 10295(10260)9869 PB91-8 12128 203-205 15510±80 18187(17900)17630 PB91-8 12129 403-405 27210±260 PB91-8 12130 540-542 34770±660 -

delta R 45±21 45±21 45±21 45±21 45±21 45±21 -

4. Discussion and conclusion Licosa and Paola basin mass-failure deposits are rather different in style, location relative to the shelf edge, and extent of down-slope translation. However, regardless of their architectural differences, both deposits formed during the late-Quaternary relative sea level rise and reinforce the evidence for sediment failure during rising sea levels. The sediment draping both failures shows an incomplete basal portion of the Late Glacial-Interglacial Transition (lower part of GI-1). Based on the timing of the onset of the post-slide drapes, failure occurred close to melt-water pulse 1A (mwp1A), the

Submarine slides during relative sea-level rise, eastern Tyrrhenian margin

477

interval of maximum rate of relative sea level rise during the last Termination (Fairbanks, 1989; Bard et al., 1996; Fig. 8). The precise timing of this interval of peak melt water production has been constrained to be close to 13.8 cal. kyr BP (Bard et al., 1996). Recent estimates of the rate of sea level rise during mwp1A are in the order of 20 m in 500 years (Clark et al., 1996). This evidence suggests that rapid loading by drowning of unconsolidated sediment played a role in generating failure. This view is consistent with the evidence that, in both areas, failure coincides with a marked change in sedimentation style. Such change reflects a substantial landward shift of sediment entry points and decrease in sediment accumulation rates, both consistent with the draped stile of post-failure deposition. If refined dating becomes available, an increasing number of failure events in the Mediterranean may show a relation between the timing of sediment failure and the timing of increased rate in relative sea-level rise during melt water pulses: mwp1A (e.g., this study) or mwp1B (e.g., BIG’95 debris flow; Lastras et al., 2002). We speculate that both Licosa and Paola failures formed in response to the following combination of predisposing factors and triggers: 1) during the LGM, rapid deposition on the upper slope resulted in the formation of a potentially unstable sediment section resting on a well defined basal surface; 2) when sea level rise reached peak rates (mwp1A), the hydrostatic load increased the pore pressure within the recently-deposited sediments; 3) failure occurred in response to this increase of hydrostatic load and/or in response to cyclic loading imparted by earthquake shocks, which are common and large on the tectonically-active eastern Tyrrhenian margin.

Figure 8. Geochronology of Licosa and Paola Basin slides relative to the timing of increased melt-water production (mwp1A) and enhanced late-Quaternary relative sea-level rise. Similarly, BIG’95 (from Lastras et al., 2002), appears associated to mwp1B. Timing of mwp1A and 1B is from Bard et al. (1996), and Clark et al. (1996).

5. Acknowledgements Data for this paper were re-interpreted within the Italian GNDT project on geologic risks in offshore areas. We thank Franco Ricci Lucchi, Marco Roveri and Rob Wheatcroft for their thorough reviews and stimulating comments. This is IGM contribution n 1305.

478

Trincardi et al.

6. References Asioli, A., Trincardi, F., Lowe, J.J., Oldfield, F., 1999. Short-term climate changes during the Last GlacialHolocene transition: comparison between Mediterranean records and the GRIP event stratigraphy. Journal of Quaternary Science, 14: 373-381. Bard, E., Hamelin, B., Arnold, M., Montaggioni, L., Cabioch, G., Faure, G., Rougerie, F., 1996. Deglacial sea-level record from Tahiti corals and the timing of global meltwater discharge. Nature, 382: 241244. Bjorck, S., Walker, M.J.C., Cwynar, L., Johnsen, S., Knudsen, K.L., Lowe, J.J., Wohlfarth, B., 1998. An event stratigraphy for the Last Termination in the North Atlantic region based on the Greenland icecore record: a proposal by the INTIMATE group. Journal of Quaternary Science, 13: 283-292. Calanchi, N., Dinelli, E., Lucchini, F., Mordenti, A., 1996. Chemostratigraphy of late Quaternary sediments from Lake Albano and central Adriatic Sea cores (PALICLAS Project). Mem. Ist. Ital. Idrobiol., 55: 247-263. Capotondi, L., Borsetti, A.M., Morigi, C., 1999. Foraminiferal ecozones, a high-resolution proxy for the late Quaternary biochronology in the central Mediterranean Sea. Marine Geology, 153: 253-274. Clark, P.U., Alley, R.B., Keigwin, L.D., Licciardi, J.M., Johnsen, S.J., Wang, H., 1996. Origin of the first global meltwater pulse following the last glacial maximum. Paleoceanography, 11: 563-577. Coleman, J.M., Prior, D.B., Lindsay, J.F., 1983. Deltaic influences on shelf edge instability processes. In Stanley, D.J., Moore G.T. (Eds.). The shelf break: Critical interface on continental margins. SEPM Special Publication, 33: 121-137. Droz, L., Bellaiche, G., 1985. Rhône deep-sea fan: morpho-structure and growth pattern. American Association of Petroleum Geologists Bulletin, 69: 460-479. Fairbanks, R.G., 1989. A 17,000 year glacio-eustatic sea-level record: Influence of glacial melting rates on the younger Dryas event and deep ocean circulation. Nature, 342: 637-642. Farrán, M., Maldonado, A., 1990. The Ebro continental shelf: Quaternary seismic stratigraphy and growth patterns. Marine Geology, 95: 298-312. Gallignani, P., 1982. Recent sedimentation processes on the Calabrian continental shelf and slope (Tyrrenian Sea, Italy). Oceanologica Acta, 5: 493-500. Lastras, G., Canals, M., Hughes-Clarke, J.E., Moreno, A., DeBatist, M., Masson, D.G., Cochonat, P., 2002. Seafloor imagery from the BIG’95 debris flow, western Mediterranean. Geology, 30: 871-874. Mongardi, S., 1994. Studio sismo-stratigrafico e sedimentologico dei depositi tardo-quaternari sul margine tirrenico orientale (bacino di Paola, Calabria). Unpublished thesis, University of Bologna, Italy, pp. 166. Mutti, E., 1985. Turbidite systems and their relation to depositional sequences. In Zuffa, G.G., ed., Provenance of Arenites: NATO-ASI Series. Dordrecht, Reidel Publishing, pp. 65-93. Nelson, C.H., Maldonado, A., Barber, J.H., Alonso, B., 1991. Modern sand-rich and mud-rich siliciclastic aprons: alternative base-of-slope turbidite systems to submarine fans. In Weimer, P., Link, M.H. (eds), Seismic Facies and Sedimentary Processes of Submarine Fans and Turbidite Systems: New York, Springer-Verlag, pp. 171-190. Ross, W.C., Halliwell, B.A., May, J.A., Watts, D.E., Syvitski, J.P.M., 1994. Slope readjustment: a new model for the development of submarine fans and aprons. Geology, 22: 511-514. Soloviev, S.L., Kuzin, I.P., Kovachev, S.A., Ferri, M., Guerra, I., Luongo, G., 1990. Microearthquakes in theTyrrhenian Sea as revealed by joint land and sea-bottom seismographs. Marine Geology, 94: 131-146. Stuiver, M., Reimer, P. J., 1993. Extended 14C database and revised CALIB radiocarbon calibration program. Radiocarbon, 35: 215-230. Torres, J., Savoye, B., Cochonat, P., 1995. The effects of late Quaternary sea-level changes on the Rhone slope sedimentation (northwestern Mediterranean), as indicated by seismic stratigraphy. Journal of Sedimentary Research, 65: 368-387. Trincardi, F., Field, M.E., 1991. Geometry, lateral variation, and preservation of downlapping regressive shelf deposits: eastern Tyrrhenian Sea margin, Italy. Journal of Sedimentary Petrology, 61: 775790. Trincardi, F., Field, M.E., 1992. Collapse and flow of lowstand shelf-margin deposits: an example from the Eastern Tyrrhenian, Italy. Marine Geology, 105: 77-94. Trincardi, F., Correggiari, A., Field, M.E., Normark, W.R., 1995. Turbidite deposition from multiple sources; Quaternary Paola Basin (eastern Tyrrhenian Sea). Journal of Sedimentary Research, 65: 469-483. Vail, P.R., Mitchum, R.M.Jr., Thompson, S., III, 1977. Seismic stratigraphy and global changes of sea level; Part 3, Relative changes of sea level from coastal onlap. In Payton, C. E. (Ed). Seismic stratigraphy; applications to hydrocarbon exploration. AAPG Memoir, 26: 63-81.