Levant jet system—effect of post LGM seafloor currents on Nile sediment transport in the eastern Mediterranean

Sedimentary Geology 329 (2015) 28–39

Contents lists available at ScienceDirect

Sedimentary Geology journal homepage: www.elsevier.com/locate/sedgeo

Levant jet system—effect of post LGM seafloor currents on Nile sediment transport in the eastern Mediterranean U. Schattner a,⁎, M. Gurevich a, M. Kanari b, M. Lazar a a b

Dr. Mosses Straus Department of Marine Geosciences, Leon H. Charney School of Marine Sciences, University of Haifa, Mount Carmel 31905, Israel Israel Oceanographic and Limnological Research, Haifa 31080, Israel

a r t i c l e

i n f o

Article history: Received 4 July 2015 Received in revised form 16 September 2015 Accepted 18 September 2015 Available online 28 September 2015 Editor: Dr. B. Jones Keywords: Continental shelf Seafloor currents Shelf break jet Contourite currents Nile Eastern Mediterranean

a b s t r a c t Sedimentary development of a continental margin is directly related to seafloor current dynamics. Yet the linkage between the processes remains vague due to the different time scales they represent. To narrow this gap we focus on the thoroughly studied distribution system of Nile derived sediments across the Levant continental margin (eastern Mediterranean). These sediments dominate the Late Quaternary stratigraphy of the entire margin. Their mobilization has been explained exclusively by longshore transport, while oceanographic evidence from the basin and margin are not incorporated in the known mechanism. New data indicates that longshore mechanism is part of a much larger system. Based on integrated interpretation of multibeam bathymetry, high-resolution single-channel seismic reflection and oceanographic (temperature, salinity and chlorophyll) data we suggest a jet current system mobilizes the Levant Surface Water (LSW), Levant Intermediate Water (LIW) and Atlantic Water (AW) northwards along the margin, between 0 and 400 m water depths. On the seafloor, contourite currents form elongated along-strike morphologies. Below 400 m along-dip gravity flows dominate sediment transport to down the slope, below the Eastern Mediterranean Deep Water (EMDW). Initiation of this mechanism during the Pleistocene–Holocene transition and not at the end of the Last Glacial Maximum (LGM) indicates a gradual recovery of the thermo-haline circulation. Current intensification in the early Holocene may have also increased water stratification. This comprehensive mechanism explains sediment transport along the entire depth range of the continental margin, while integrating seafloor currents, morphology, as well as their relation to sea level rise and stratigraphy of water masses in the Levant basin since the LGM. Given the consistency of seafloor currents throughout the Holocene we propose to address them as the Levant Jet System. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Sedimentary development of a continental margin is directly related to seafloor current dynamics. Formation of sedimentary patterns and the degree of their preservation are driven by physical (e.g., direction, velocity) and temporal (e.g., seasonal, annual) variations of the currents (e.g., Trincardi et al., 2007). However, while sediments record the geological past oceanographic measurements represent present-day conditions. Given the oceanographic variability, deduction on its influence on past accumulation/erosion budget is not straightforward. It requires knowledge of key variables such as sediment source, mode of supply over geologic times, its consistency, stratigraphic patterns, their age constraints and correlation to sea level stages. Deduction also requires key oceanographic variables—present-day water masses, their mobilizing currents, and their velocities. The aim of this study is to provide a comprehensive understanding of the interaction between sediment transport and deposition and bottom oceanographic currents. Due to large amount of variables in such a correlation the eastern ⁎ Corresponding author. E-mail address: [email protected] (U. Schattner).

http://dx.doi.org/10.1016/j.sedgeo.2015.09.007 0037-0738/© 2015 Elsevier B.V. All rights reserved.

Mediterranean was chosen as a case study because many of these are well constrained. 1.1. Geological settings Nile derived sediments dominate the Late Quaternary stratigraphy of the Levant continental margin, eastern Mediterranean (Figs. 1, 2). Their ubiquitous distribution extends from the shallowest continental shelf to the base of the continental slope down to a water depth of ~ 1 km. Their continuing supply is distributed northwards across the margins by anti-cyclonic circulation (Buchbinder et al., 1993; Ben-Gai et al., 2005). These homogeneous allochthonous terrigenous siliciclastic sediments settle along the northern Sinai and southern Levant margins (Coleman et al., 1981; Inman and Jenkins, 1984; Stanley, 1989; Frihy et al., 1991; Frihy and Lotfy, 1997; Almagor, 2000; Zviely et al., 2007; Hyams-Kaphzan et al., 2008), and coasts (Goldsmith and Golik, 1980; Rohrlich and Goldsmith, 1984; Carmel et al., 1985; Perlin and Kit, 1999). The Carmel Structure extends across the central part of the margin, over 10 km (Fig. 2; Garfunkel and Almagor, 1985; Ben-Gai and Ben-Avraham, 1995; Schattner et al., 2006). This elevated bedrock structure (~25 m water depth) obstructs northward sediment transport

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

29

Fig. 1. (a) Origin and migration path of the Levantine Intermediate Water (LIW) across the Mediterranean and into the Atlantic. (b) Close-up of the eastern Mediterranean, showing the LIW. In orange, currents transferring the Atlantic Water mass (AW): Libyo-Egiptian Current—LEC, Mid-Mediterranean Jet—MMJ, Asian Minor Current—AMC (after Malanotte-Rizzoli et al., 2014): (1) marks the location of the current study, (2) extent of Hecht et al. (1988), H—Haifa Bay, B—Beirut.

(e.g., Goldsmith and Golik, 1980). Sediments that bypass this structure accumulate in Haifa Bay, the northernmost limit of the Nile Littoral Cell (Pomerancblum, 1966; Nir, 1980; Inman and Jenkins, 1984; Almagor, 2000; Zviely et al., 2006, 2007; Avnaim-Katav et al., 2012).

Slope canyons that appear from the Carmel Structure northwards, channel surplus supplies to the deeper basin (Almagor, 2000). North of Haifa Bay, the shelf narrows from 3 km until disappearing off Beirut, while its slope steepens (Ben-Avraham et al., 2006; Carton et al., 2009).

30

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

Fig. 2. (a) Multibeam bathymetry map of the Levant continental margin and a section of the adjacent basin. Black lines—location of single channel seismic data (this study), black dots—current measurements of Rosentraub and Brener (2007). (b) Gradient azimuth shading on the bathymetry map. The southern continental slope (SL1) shows slumps and scars, while the northern slope (SL2) is incised by canyons. Two mega slides disrupt the continental slope—Palmahim in SL1 and Dor in SL2. Four morphologic belts follow isobath contours along the continental shelf and upper slope marked by 1–4 and correlate to B1–B4 in the text. The convex (Cv) and concave (Cc) parts of B1 are also marked. Cs—Carmel Structure, AchC—Achziv Canyon. White triangles on the continental rise and basin indicate incised paths of turbidites. (c) Simplified bathymetry map showing the depth ranges of B1—light green (0–80 m); B2—light blue (80–130 m); B3—orange (130–200 m); and B4—dark blue (200–350 m). Red lines—location of single channel seismic reflection profiles. Green lines—location of following figures. Blue line—location of profiles in Fig. 8. HB—Haifa Bay; Cs—Carmel Structure.

During the Last Glacial Maximum (LGM; ~ 18 kyr BP) global sea levels dropped by ~120 m (Fairbanks, 1989). Forced regression exposed the Levant shelf to erosion under aerial conditions. This formed a rugged regional unconformity of aeolianite calcareous sandstone ridges. It was previously used as a regional seismic marker due to the sharp lithologic contrast between clay and sand units from below and above (Almagor, 2000; Schattner et al., 2010). Post-LGM transgression re-established Nile sediment supply. Sediment retrograded inland as sea level rose, until sea level reached its present height ~ 6000 year BP. Younger sediments accumulate over the inner bay (Zviely et al., 2007) and prograded over its outer shelf (Schattner et al., 2010). The majority of previous studies widely agree that longshore transport is the only mechanism that mobilizes Nile derived sediments along the Levant coastline. This mechanism has been used to explain the stratigraphic development of the entire Nile littoral cell. These include theoretical studies (Neev, 1960; Goldsmith and Golik, 1980), later joined by bathymetric evidence (Golik, 1993, 1997, 2002; Shoshany et al., 1996; Golik et al., 1999), geological considerations (Buchbinder et al., 1993) and measurements of waves as well as geomorphology of coastal and nearshore bathymetry (Carmel et al., 1985; Perlin and Kit, 1999; Zviely et al., 2006). Longshore transport, however, is limited to shallow waters (~30 m) that extend ~350–400 m from the coast (Fig. 2). Therefore it cannot explain the arrival of Nile derived sediments to greater water depths (such as seen in the core of Box et al., 2011). The only suggestion for outer shelf and slope sedimentation was proposed by Ben-Gai et al. (2005). Their simulation of longshore transport showed that after accumulation, sediments move down slope where they are deposited. 1.2. Oceanographic settings Water stratigraphy in the Levant basin (area 2 in Fig. 1b) consists of four masses that circulate in the Mediterranean (Hecht et al., 1988). Their vertical extent varies with season. The high temperature and saline Levant Surface Water (LSW) occupies the top 15 m, 25 m and

70 m during spring, summer and fall (respectively). It develops during the summer. Below, the Atlantic Water (AW) extends between 25–100 m, 25–105 m and 80–110 m during spring, summer and fall (respectively). A jet stream carries AW across the Mediterranean, until it courses northwards along the Levant margin (Fig. 1; Millot, 1985; Baldacci et al., 2001; Ruiz et al., 2009; Malanotte-Rizzoli et al., 2014). Below, the Levantine Intermediate Water (LIW) appears between 170–310 m, 135–315 m, 220–270 m and 75–325 m during spring, summer, fall and winter (respectively). LIW originates southeast of Crete, circulates throughout the Levant and exits towards the Atlantic (e.g., Millot, 2009). The lowermost mass is the Eastern Mediterranean Deep Water (EMDW or in some papers DW). It is confined to depths greater than 680 m, 690 m, 675 m and 695 m during spring, summer, fall and winter, respectively (Hecht et al., 1988). EMDW originates from the Adriatic Sea and sporadically from the Aegean Sea northwest of the Levant basin. Its oceanographic characteristics remain constant throughout the year (Kontoyiannis et al., 1999; Kress and Herut, 2001). A high-velocity baroclinic jet current develops along Levant shelf and slope (area 1 in Fig. 1b) during summer and winter (Rosentraub and Brenner, 2007). Measurements over a 10-year period at depths of 30, 120 and 500 m (Fig. 2) indicate that maximum velocities of this anticyclonic current are confined to the upper water layer (~ 30 cm/s). They intensify seawards. Friction with the seafloor results in northward flow along the strike of the continental slope. Oceanographic circulation and its seasonal variability may play a key role in transport and distribution of Nile-derived sediments. However, comprehensive understanding correlating oceanographic and sedimentary processes is lacking. The present study aims to bridge this gap through analysis of newly acquired high resolution bathymetric, seismic reflection and oceanographic measurements from across the Levant margin. Based on seafloor and subsurface morphologies, we reconstruct patterns of seafloor currents from the LGM to the present responsible for sediment transport, accumulation and depletion along the Levant margin and Nile littoral cell.

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

2. Data and methods Multibeam bathymetry data analyzed here was collected between 2001-2007 aboard the R/V Eziona, using a Kongsberg Simrad EM 1002 and the R/V Mediterranean Explorer using an ELAC SeaBeam 3050. Data covers 15,462 km2 across the continental shelf, slope and proximal floor of the deep basin. The resulting grid resolution used here is a 50 m cell size. Acquisition and processing were conducted jointly by the Geological Survey of Israel and Israel Oceanographic and Limnological Research, as part of a national bathymetric survey (Fig. 2). Seismic datasets comprise 10 high-resolution single-channel seismic reflection surveys obtained between 2008 and 2015 aboard R/V Shikmona, R/V Eziona and R/V Mediterranean Explorer, with a total length of 3026 km covering an area of 3600 km2 (Fig. 2). Data were collected across most of the southern Levant continental shelf and upper slope. Average grid spacing between profiles is ~3 km. Denser grids of 100 m spacing were collected during two of the surveys, conducted across the northern flank of the Carmel Structure. These decimeter scale profiles were acquired using a Geo-Resources Geo-Spark-200 source with frequencies of 1–2.5 kHz. An eight-element streamer collected the signal. Processing included geometry, bandpass filtering, swell filtering and automatic gain. All data were interpreted using Schlumberger-Petrel and IHS-Kingdom Suite software packages, in a WGS-84 datum, UTM projection. Seismic attributes were calculated during interpretation, to emphasize structural and stratigraphic aspects of the data (e.g., amplitude, envelope, sweetness, chaos and variance attributes). Oceanographic measurements were collected in seven campaigns during November 2008, May 2010, June 2011, March 2012, October 2013 May 2014 and May 2015 aboard the R/V Shikmona. The constant transect was 27 km long and followed a N55W azimuth along Haifa

31

Bay across the slope to the basin (blue line in Fig. 2). Conductivity, temperature and depth (CTD) were collected at depths of 15 m and 40 m over the inner and outer shelf of southern Haifa Bay; 200 m and 500 m over the upper and mid continental slope; and down to 1000 m over the deep basin. Sampling interval was every 1 m at each station. Results were converted into water salinity and density parameters at each measured location. 3. Results 3.1. Bathymetry Bathymetric data show four distinct belts that differ in morphology, deposition and erosional patterns (Figs. 2–6). Boundaries between the belts are sharp and consistently follow isobath contours (Figs. 2, 4, inset). The coastline bounds the belts to the east. Towards the basin, they extend to the upper slope, down to 350 m. This width narrows northwards due to slumps and canyon that cut the continental shelf and steepen the slope. Along the southern margin, the Palmahim slide diverts the orientation of the belts (Figs. 2, 3, inset). Farther northwards, only the easternmost belt (B1, light green, Fig. 2) appears across Haifa Bay (Fig.2). Drastic narrowing of the shelf by the Achziv canyon truncates this belt. The upper belt (B1) extends between water depths of 0–80 m (Figs. 2, 4). Its profile in the east is concave and slopes from the coastline to reach a local minimum at a depth of ~40 m. Farther west, the profile becomes convex. Occasional aeolianite sandstone ridges pierce the otherwise smooth relief. In places, local depressions appear north of these ridges. Morphology of B1 becomes elevated, rough and rugged as it approaches the Carmel Structure from the south (Fig. 5). It climbs to a water depth of 26 m before dropping back towards Haifa Bay, where

Fig. 3. E–W single channel seismic reflection profile crossing the southernmost portion of the study area, across the Levant margin, south of the Palmahim slide (P). Belts B1 to B4 are presented. Lower right—location of the profile (blue) on gradient shading of the bathymetry. Note the difference in slope morphology between slump (S) dominance in the south and canyons (C) in the north. D—Dor slide. 1–4 mark the location of belts B1–B4 on the continental shelf and upper slope. B3 and B4 extend laterally northwards (to the left) until truncated by a steeper slope carved by canyons. White line marks this break. Note the lack of belts below B4, i.e., below 350 m. Green profile—Fig. 4; Red—Fig. 5; Yellow—Fig. 6. Upper section—flattened profile of the upper 70 msec of the seismic data. Note the typical bathymetric appearance of B1 to B4. Thickening of post Surface A stratigraphy below B1 and B2 represents a clinoform. Yellow arrows—truncation, green arrows—onlap and downlap, G—acoustic blanking possibly by gas.

32

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

Fig. 4. (a,c) E–W single channel seismic reflection profile crossing the Levant margin north of the Palmahim slide. B1 shows inner shelf concave (Cc) and outer shelf convex (Cv) morphologies. Yellow arrows mark truncation of reflectors, green–onlap. Sediment waves extend across the post Surface A succession. 1—Sediment wave pattern below Surface A, i.e., older than LGM. 2—Pre-LGM lowstand unconformity. (c) Enlargement of B2 is divided to units S1–S4 according to Schattner et al. (2010). See text for details. Most of the sediment waves on the seafloor and subsurface of B2 are asymmetric (3) while some are more symmetric (4). In general, the stoss side is downslope and the lee side faces upslope. In places, the accumulation pattern between waves reflects contourite current activity (5). (b) Location of the profile on gradient shading of the bathymetry. Note the close correlation between B1–B4 and the isobath contours. S—slump scars.

Fig. 5. (a, c) E–W single channel seismic reflection profile crossing the Levant margin south of the Carmel Structure. (a, b) The erosive concave (Cc) part of B1 exposes aeolian calcareous ridges across the inner shelf (ACR). Here the concave (Cv) part of B1 also shows truncation (yellow arrows). B2 narrows relative to Fig. 4. In addition to low amplitude asymmetric sediment waves (1), it shows drift (2) and moat (3) patterns, typical products of contourite currents. In places, the accumulation pattern between waves reflects contourite current activity (4). (b) Location of the profile on gradient azimuth shading of the bathymetry map. White triangles mark location of bathymetric notches, similar to the moat (3) marked over B2. The numbers 1, 2, 3 mark the location of B1 to B3. B4 is absent. White circles mark isobath contours 80, 130, 200 and 350 m from right to left. CS—Carmel Structure.

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

33

Fig. 6. NNE–SSW single channel seismic reflection profile crossing two elevated NW extensions of the Carmel Structure (CS). South of CS, sediments accumulated in morphologic pockets above Surface A. They are truncated close to the surface in several locations (1) due to seafloor currents along B1. Hardly any post LGM sediment accumulates on the southern flank of CS (2). The northern CS flank is covered by a thin northward progradation unit (3). This unit is divided into three bathymetric steps, by currents exiting southern Haifa Bay towards the west (4, and Fig. 9). Occasional gas escape features appear across the profile (5), similar to observations north of Carmel Structure (Schattner et al., 2012). CS—Carmel Structure, HB—Haifa Bay.

the typical profile pattern of B1 is retained—concave over the inner shelf and convex over the outer shelf (Fig. 6). However, its axis is shifted eastwards in a way that the concave part of B1, south of the Carmel Structure, faces the convex part of B1 north of the Carmel structure (Fig. 2). B2 follows the shelf break, between isobaths 80 and 130 m (Figs. 3, 4). Isobath orientation changes from N31E south of the Palmahim slide to N16E south of the Carmel Structure, and N15E north of it. B2 narrows northwards starting at 4.2 km, while the slope of B2 steepens northwards, from 0.63° to 0.84°. Over this entire depth range B2 is characterized by elongated linear sediment waves. Both their maximum amplitude and frequency decrease northwards: from 3.7 m to 2.3 m, and from 3.84 to 0.9 waves per kilometer (respectively). The waves are asymmetric. Their stoss side faces the basin while the lee faces upslope. Throughout B2 their orientation maintains an angle of 10°–11° with the isobath contours bounding the belt. The sediment waves disappear northwards where the slope of the shelf break exceeds 0.8°o (transition between SL1 and SL2, Fig 2). To the west of this transition the continental slope also steepens relative to its gradient in the south and canyons begin to appear across the slope. Relief of B2 becomes smoother except for local comet structures and notches oriented sub-parallel to the isobaths. B2 disappears south of the Carmel Structure, where the slope steepens to over 10°o. Morphology of B3 is smooth (Figs. 2, 3, 4). It is bounded between 130 and 200 m isobaths and runs parallel to B2, maintaining a slope of ~1.3°. B3 terminates northwards close to the transition of B2 from sediment waves to a smooth topography. At this point, B3 steepens, narrows and is cut by local slope canyon heads. The lowermost belt B4 is somewhat wider and shorter. It is 4.8 km wide in the south and narrows northwards until truncated by the canyons. Along this route its slope steepens from 1.7° to 1.9°. B4 is bounded between isobaths 200 and 350 m. Its southern portion is characterized by elongated, linear and asymmetric sediment waves that gradually dissipate north of the Palmahim slide (Fig. 2). Similar to B2, their stoss side faces downslope. Their amplitude increases from a maximum of 3.16 m in the south to 7.72 m in the north. Their frequency decreases northwards from 2.75 to 1.85 waves per kilometer. The angle between the axis of B4 sediment waves and belt orientation changes from 16° south of the Palmahim slide, to 11° north of it. No further distinct belts can be observed downslope. 3.2. Seismic reflection The division of belts (B1-4) extends to the subsurface through a succession of finely layered (~1 m) reflections. These overlay a corrugated

undulating unconformity of high-amplitude lower-frequency reflectivity (Surface A; Figs. 3, 4, 5; as defined by Schattner et al., 2010, for Haifa Bay). Seismic data show that the convex section of B1 consists of basinward dipping prograding reflectors at the outer shelf (equivalent to S4b of Schattner et al., 2010). These are truncated across the inner shelf forming the concave section of B1's bathymetric expression (Figs. 3, 4, 5). Close to the southern flank of the Carmel Structure, post-LGM reflectors fill morphological pockets between local highs of Surface A (Fig. 6). These reflectors are truncated close to the seafloor. Hardly any post-LGM reflectors are visible on the structure itself. The northern flank is covered by a thin set of northward prograding reflectors. They are truncated at the top and appear on the bathymetry as three steps (Fig. 6). A succession of sediment waves appear in B2 (Fig. 4). They first appear above Surface A as minor undulations. Based on the seismic facies of reflectors we correlate this initial undulating reflector to the top of S1 of Schattner et al. (2010). The low reflectivity parasequence above corresponds to S2 while the high reflectivities above this unit are correlated to S3 and S4 (Fig. 4). The amplitude of sediment waves increases through S2 while their crest axis tends to migrate upslope. In S3 and S4 these axes become steeper. Along B2, the wavelength of sediment waves varies between 60 and 585 m with an average of 221 m. Their amplitude ranges between 1.5 and 5.1 m. Contourite drifts and moat structures occasionally appear in between the slanted successions of sediment waves. Farther northwards, additional local drift and moat structures appear across B2, this time not nested between sediment waves (Fig. 5). These structures correspond to the bathymetric notches described above. North of the Carmel Structure the entire post-LGM sedimentary section is evident along B1. However, unlike its southern counterpart, the differences between its concave and convex sections represent variations in the amount of deposition rather than erosion. Further downslope B3 appears as a set of conformable reflectors. Flattening of seismic data (Fig. 3) emphasizes that B3 marks the transition from slope to a shelf depositional environment. It is located immediately west of the post-LGM sigmoidal clinoform. Across B3, S1 reflectors are semi-parallel to Surface A. S2 reflectors onlap and downlap S1 from west and east (respectively). S3–4 appear thinner across B3 relative to their thickness over B4 and B2. Stratigraphy across B4 is correlative to the reflectors of B2 (Fig. 3). Amplitudes of S1 are low compared to S2–4. Its reflectors are semi parallel to Surface A. Undulations begin to appear in S2 and develop into asymmetric sediment waves farther up towards the seafloor. The lateral frequency of these waves is denser than the sediment waves of B2. The

34

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

succession of reflectors building the sediment waves is thinner than in B2. 3.3. Oceanography Integrated temperature-salinity diagrams suggest that the water column is comprised of four masses (Fig. 7). The range of salinity and temperature values is wider across the margin (our measurements) than in the basin (Hecht et al., 1988). Across the margin, salinity ranges between 38.74 and 39.56 ppm, and temperature between 13.69 °C and 25.42 °C, compared with 38.68–39.08 ppm and 13.56 °C and 21.46 °C in the basin. Based on the strong resemblance to Hecht et al. (1988) we hereon use the same terminology. At the bottom of the diagram, the EMDW shows low salinity and temperature values that remain relatively stable throughout the year and between the basin and margin. Overall, temperature increases steadily throughout the diagram while salinity increases from the basin towards the margin in all three layers (except for AW in spring 2015). Seasonal variations between the measured curves appear mainly in salinity values of LIW, AW and LSW. Salinity reaches a local peak that defines LIW and drops to a local minimum defining the AW. Further up, the curve climbs steeply towards the top to define the LSW. Stratification of the water column is evident in individual profiles of temperature, salinity and chlorophyll (Fig. 8). Data indicate peak values for these three indices across the shelf in both spring and summer. Deepening of the mixed layer equalizes peak salinity and temperature values across the entire profile, while chlorophyll values remain higher across the shelf than in the upper waters of the basin. Stratification of the water column is enhanced during the spring and summer. High salinity values of LSW, decline downwards to a local minimum a few tens of meters below the sea surface (AW). Maximum chlorophyll values appear across LIW. This N150 m thick layer appears slightly higher during the spring than the summer and fall. Salinity reaches maximum values around depths of 200 m during summer, fall and spring. At grater depths below 400 m all three indices remain low and constant, across the EMDW. 4. Discussion The present study challenges the prevailing paradigm that the only mechanism for sediment transport along the Levant continental margin

is by longshore transport and is limited to water depths shallower than 30 m. Based on integrated interpretation of bathymetric, seismic and oceanographic data we show for the first time that transport extends beyond the inner continental shelf. Our results indicate a dynamic connection between seafloor currents and morphology since the LGM, as well as their relation to sea level rise and stratification of water masses in the Levant basin during this time frame. The following paragraphs describe this connection in detail. Interpretation integrates the new data with previous measurements of water masses and currents (e.g., Hecht et al., 1988; Rosentraub and Brener, 2007), seismic stratigraphy of the post LGM sedimentary succession and its correlation to sea level fluctuations (Schattner et al., 2010). At the base of the post-LGM sedimentary infill Surface A extends across the entire Levant margin, from the present-day shoreline basinwards (Figs. 3, 4, 6). This irregular erosional unconformity serves as a key marker for stratigraphic correlation across the Carmel Structure, from north (Almagor, 2000; Schattner et al., 2010) to the southern Levant margin. Most of Surface A is covered by post LGM infill. However, its rugged relief appears sporadically along B1, across the inner shelf (Figs. 2, 6). During post-LGM transgression B1 was the last to receive sediments, and soon after was subjected to erosional seafloor currents (see description of S4 below). Four major sedimentary parasequences progressively accumulated above Surface A (S1-S4; Figs. 3–6). While the elevated Carmel Structure divides between Haifa Bay and the southern Levant, the unique seismic facies of these parasequences is similar between the regions. Their morphology, however, differs. At the base of the infill, parasequence S1 directly overlies Surface A. It is sub-parallel to the rough relief of the erosional unconformity. At the base of S1, a succession of reflectors onlap Surface A, representing progressive transgression that covered the outer shelf with sediments, while the inner shelf remained exposed (Fig. 5). Schattner et al. (2010) correlated this stage to ~ 18–11 kyr BP. Close examination of this parasequence under B2 and B3 shows sub-parallel concordant reflectors with minor undulations. This pattern indicates that low velocity seafloor currents led to the deposition of sediments in this parasequence. This seismic facies stands in contrast to the sediment waves that begin to appear only in the parasequences above S1. This contrast suggests that during early transgressional stages, thermohaline circulation in the Levant basin, as part of the Mediterranean

Fig. 7. Temperature–salinity diagram of six measurement campaigns conducted between 2008 and 2015 across the northern Levant margin (blue transect in Fig. 2). Red curve represents the results of Hecht et al. (1988) measured across area 2 in Fig. 1b. Water masses are marked based on correlation between the two datasets: LSW—Levantine Surface Water, AW—Atlantic Water, LIW—Levantine Intermediate Water, EMDW—Eastern Mediterranean Deep Water.

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

35

Fig. 8. Temperature, salinity and fluorescence profiles measured during 2011 (summer), 2013 (fall) and 2015 (spring), across the northern Levant margin (blue transect in Fig. 2). Shaded gray areas represent the continental shelf and slope. B1 to B4 are marked according to their extent on the bathymetry, between isobaths of 0, 80, 130, 200 and 350 m. Further details appear in text.

system, still had not recovered from the extreme conditions of the LGM. As a result, seafloor currents were too weak to either erode or deform the seafloor of the Levant margin. The transition between deposition of S1 and S2 is marked by an increase in reflector amplitude, frequency and alteration of spatial arrangement. The sedimentary succession under B2 began to form as a series of sediment waves. At the base of S2 most waves are symmetrical with small amplitudes. Further up in the succession their amplitude increases. Most become asymmetric, while the succession of their crests migrate upslope. This increase suggests that either the capacity of the current to transport sediment increased or that current velocity increased, or a combination of the two factors occurred. However flow did not become erosive. Data also exhibit occasional drifts and moats

between the sediment waves. These indicate that the activity of alongslope contourite seafloor currents advanced along the small-scale topography created by local sediment wave crests. Along with progressive development of sediment waves under B2 and B4, reflectors of B3 and B1 remain parallel and concordant. These sharp dissimilarities suggest differences in rates of suspension fallout and in velocity of seafloor currents. Field evidence and numerical analysis indicate that formation of sediment waves is associated with an intrinsic instability that is sensitive to seafloor roughness (Lee et al., 2002). The degree of seafloor roughness (perturbations) dictates the drag coefficient of overriding seafloor currents. An increase in the drag coefficient will result in a shift in accumulation patterns from sediment drape, to the formation of a field of

36

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

upslope migrating sediment waves. This is possible in cases where sediment supply is available and seafloor currents are able to transport these sediments (Lee et al., 2002). Parasequences S3 and S4 accumulated across the shelf and slope, along the entire Levant margin. The shift from low to high reflectivity between S2 and S3–4 reflects a change in sediment composition or temporal hiatus in sedimentation. This may have resulted from a pause in current flow or alternatively, from a surge. Either way, a change in sediment supply can be ruled out since it was not interrupted during the Holocene (Schattner et al., 2010). Typical morphologies of B4, B3 and B2 appear throughout S3 and S4 parasequences. The sediment waves in B2 and B4 climb upslope with time. In this diagonal out-of-phase climbing pattern, the crest of a young sediment wave lies at a slight offset upslope relative to older ones. For the majority of waves, the climbing angle between the crest and the average slope is consistently larger than the angle between the stoss side and the slope. These waves are equivalent to supercritical climbing sediment waves. The entire pattern is comparable to Facies 1 of Jobe et al. (2012). This suggests that depositing currents were relatively long-lived. Over time, sedimentary patterns exhibit a balance between a high suspension fallout rate that exceeds transport and overall rapid currents. The dynamic equilibrium can be described by small changes over the central part of the Hjulström curve, between transport, erosion and deposition (Hjulström, 1939). However, this remains an estimation due to lack of actual grain-size data. Formation of sediment waves is dictated by seafloor currents. On a relatively flat seafloor their orientation tends to be arranged perpendicular to the current. Currents flow over the stoss to the lee side of the waves. However, orientation changes when the angle of the seafloor increases (slopes). In these areas, local isobaths dictate the orientation of sediment waves. The angles between isobath contours and sediment wave axes range between 10°–50° (Flood et al., 1993). Under the influence of a seafloor current, axes of sediment waves tend to migrate upslope with time, as well as downstream, i.e. with the flow direction (Wynn and Stow, 2002). The angle of upslope migration is subjected to the Coriolis Effect. In the northern hemisphere crests of sediment waves tend to migrate to the right of flow direction (Flood, 1994). Our measurements show that in B2 this angle ranges between 10° and 11°, and in B4 it is between 11° and 16° (Fig. 2). Looking northwards, their axes migrate to the right in both vertical (upslope) and horizontal planes. Integration of surface and subsurface evidence with previous oceanographic studies (Hecht et al., 1988; Rosentraub and Brenner, 2007; Malanotte-Rizzoli et al., 2014) suggests that sediment waves of B2 and B4 are mainly a product of along-slope contourite currents. The contour currents transport and deposit sediments while eroding and reshaping the seabed. The net result is the formation of elongated sediment waves. Their orientation is consistent across B2 and B4. Sediments are channeled between local crests of the elongated sediment waves. Local helical flow deposits sediments to the left of these features where they accumulate upon the lee side of the nearby sediment wave. This flow occasionally erodes the stoss side of the nearby wave to the right, or prevents deposition (upslope) (Figs. 3–6). In places where crests are vertically stacked, these currents carve a contourite moat along both sides of the crest. Velocity of seafloor currents is such that they allow quick burial and preservation of the original structure, along with minor erosion. Halfway through B2, sediment wave morphology transforms northwards to smooth reflectors and seafloor, which are cut by local moats. This transition occurs where the average seafloor slope steepens from 0.8° to 1.3°. It is possible that sediment wave structure has reached the limit of stability. Stratigraphic correlation with Haifa Bay (Schattner et al., 2010) suggests that S3 and S4 accumulated since the early Holocene. During this period, sea levels climbed rapidly from -40 m over the first ~ 5 kyr, while for the rest of the period, they remained at about what they are today (Sivan et al., 2001). Data show that during this entire rise B2, B3

and B4 maintained their position and did not migrate laterally towards the coast or basin. This was made possible by a stable dynamic equilibrium between three components dictating the morphological development of the seafloor: 1) the velocity of seafloor currents, their direction of flow and capacity to transport sediments; 2) physical characteristics of water masses and their stratigraphy in the basin; and 3) sediment supply and its grain size range, which are carried by seafloor currents. Sea level rise caused the submergence of the Carmel Structure at ~ 8 ky BP, which was exposed until then (Zviely et al., 2006). Hence, S1–S4 sediments reached Haifa Bay only by bypassing the structure. When sea level rose above − 26 m, currents also began to flow over the Carmel Structure into the bay (Fig. 6; e.g., Zviely et al., 2007). Schattner et al. (2010) show that the upper part of S4 (S4b) accumulated by progradation only along the outer shelf and upper slope. The bathymetric analysis presented here shows that B1 has an inner shelf concave profile that corresponds to a N–S strip of erosion (Figs. 2, 4, 5). The convex part of B1, at the outer shelf, was created by basinward progradation. We suggest that a strong northward seafloor current has been eroding B1 since the stabilization of sea levels. This current is responsible for carving the concave section of B1 south of the Carmel Structure. Local bathymetric highs along its route, such as aeolianite calcareous sandstone ridges, resulted in the formation of comet structures (e.g., Werner and Newton, 1975). Comets also appear in the northern section of B2. The direction of their tail point to the direction of the dominant seafloor current (Wynn and Masson, 2008 Stow et al., 2008). The erosion products are carried further northwards and cross the corrugated Carmel Structure into Haifa Bay. A fraction of these sediments fallout at the northern flank of the Carmel Structure and progradate northwards (Figs. 2, 6, 9). The rest accumulate as S4b along the outer shelf of Haifa Bay, progradating westwards toward the basin. We propose that sediment transport along B1 is equivalent to the previously suggested longshore transport. 4.1. Stratigraphy of the water Integration of morphological and seismic data with oceanographic evidence indicates that sediment transport is not limited to a narrow strip along the Levant shores. Salinity-temperature diagrams indicate a strong correlation between our measurements and those of Hecht et al. (1988) (Figs. 7). Stratigraphy of the water column seems to be related to the morphological belts interpreted here. Several studies have shown that northward currents dominate flow along the Levant margin in the water column (e.g., Hecht et al., 1988; Rosentraub and Brener, 2007; Malanotte-Rizzoli et al., 2014). Our results show that at present and since the LGM, seafloor currents also flow northwards along the margin, as contour currents. It appears that the LSW is responsible for the formation of longshore transport over B1 (Fig. 9). Salinity and chlorophyll measurements presented here show lateral variability from Haifa Bay to the open sea. This may represent branching of the northwards current after it crosses the Carmel Structure into Haifa Bay, in agreement with the trends marked by Zviely et al. (2007). However, seafloor and subsurface appearance of the S4 parasequence indicates that transport continues northwards, beyond the bay. The Carmel Structure also causes basinwards branching of the upper flow (Fig. 9), as shown by Efrati et al. (2013). Below the LSW, the AW intersects with the continental slope at roughly 80–130 m, along B2. This unit coincides with the jet current reported along the margin by Rosentraub and Brener (2007). It also corresponds to the upper peak of jet velocity (30–40 cm/s) reported by Hecht et al. (1988) at 80 m water depth. The intersection of the LIW with the slope is responsible for the morphology of B4 at 200–350 m. Rosentraub and Brener (2007) show values close to 30 cm/s along the slope at these depths. Hecht et al. (1988) indicate a second peak of velocity (20 cm/s) at 230 m. The smooth seafloor and subsurface reflectors of B3 represent a transition between velocities and oceanographic parameters of AW

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

37

Fig. 9. Schematic illustration of seafloor currents flowing northwards along the Levant continental margin, presented over a slope shading of the bathymetry. Levantine Surface Water (LSW, light green) flows over B1. Part of it branches WNW south of the Carmel Structure (CS). The main flow crosses CS, deposits sediments over its northern flank (Fig. 6) and continues northwards to deposit unit S4 of Schattner et al. (2010). In Haifa Bay (HB) LSW branches eastwards, turns clockwise to the inner bay, where sediments are deposited (e.g., Zviely et al., 2007). LSW is then injected westwards and erodes the northern flank of CS into three bathymetric steps (Fig. 6). Below, Atlantic Water mass (AW) flows along B2, between isobath contours of 80–130 m. Levantine Intermediate Water mass (LIW) follows the slope below AW. Both LIW and AW deflect locally over Palmahim slide (P). Close to the Carmel Structure, the continental slope steepens. AW and LIW project NNW. Below 350 m morphology of the continental slope shows down-slope features rather than along-slope. Numerous slumps (S) appear across the southern slope (SL1), while canyons (C) incise into the northern slope (SL2). White arrows mark down-slope sediment migration along the canyons and turbidite channels in the basin. D—Dor slide.

and LIW. Both seafloor currents transporting AW and LIW do not form a typical morphological signature north of the Carmel Structure. It is assumed that the Carmel Structure diverts both currents from alongslope trajectory into the basin, in a N to NNW direction. This agrees with the near surface anti-cyclonic circulation pattern (e.g., Rosentraub and Brener, 2007; Malanotte-Rizzoli et al., 2014 and references therein). 4.2. Implications on lower slope morphology Below ~400 m the continental slope shows two main morphologies. Numerous small slumps (few km in length) scar the southern slope. This segment is also interrupted by the Palmahim slide (Figs. 2, 3, 9). The northern termination of this segment is marked by narrowing and steepening of the continental slope. Slope canyons begin to appear and continue further north along the entire northern Levant margin, along northern Israel, Lebanon and Syria (e.g., Ben-Avraham et al., 2006; Carton et al., 2009). The transition between the southern and northern segments of the slope (slumps vs. canyons, respectively) truncates the northern edge of B4 (Fig. 2). It also coincides with the shift in B2—from sediment waves to a smooth morphology (Fig. 4). We suggest that seafloor currents following B2 and B4 northward along the slope arrive at a sudden bathymetric drop in the transition between the southern to northern morphological segments (SL1 to SL2 in Figs. 2b, 3). As a result, their capacity for transporting sediments in suspension and/or in traction along the seafloor decreases. This fallout is channeled basinwards through the slope canyons. WNW trending scars crossing the deep basin in the north suggest that on occasions, downslope fallout might have been catastrophic, generating turbidity currents (Figs. 2, 9).

Overall, any downslope sediment transport propagated at the base of the EMDW mass, where no significant lateral velocity has been reported. 4.3. Implications of the Levant jet system through time Seismic data show that B2–B4 began to form during the Pleistocene– Holocene transition. Their initiation suggests that the thermo-haline circulation of the eastern Mediterranean gradually recovered from the extreme conditions of glacial periods over thousands of years. In the early Holocene, currents intensified, transporting sediments and nutrients along the margin. Intensification of currents mobilized the water masses more efficiently. However, at the same time, they might have also increased water stratification. This process matured during the first half of the Holocene. It possibly facilitated preferable conditions for the formation of the Sapropel S1 unit (Rohling et al., 2015). During the second half of the Holocene, the sea reached its present level and erosion along B1 commenced. Given the consistency of seafloor currents throughout the Holocene we propose to address them as the Levant Jet System. While the currents in the upper-most water mass are driven by baroclinic conditions, the deeper circulation is driven by thermo-haline and geostrophic conditions. 5. Conclusions The study combines interpretation of multibeam bathymetry, highresolution single-channel seismic and oceanographic (temperature, salinity and chlorophyll) data collected along the Levant continental

38

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39

margin, eastern Mediterranean. A gap between oceanographic and sedimentologic knowledge led the vast majority of studies from the last several decades to explain sediment mobilization along the margin only by longshore transport. This transport is widely recognized as the system building the Levant margin during the Late Quaternary. However, longshore transport is limited to water depths of 30 m. It occurs along the innermost shelf, up to ~ 400 m from the coastline, while the margin extends down to water depths of 1500 m across ~20 km. Our results show that fast seafloor currents (jets) flow along the continental margin, from south to north. These currents follow isobath contours between 0 and 350 m water depths, i.e., across the shelf and upper slope. Four distinct seafloor morphologies indicate differences in velocities. Correlation with oceanographic studies indicates that these morphologies are formed when water masses move along the seafloor. In particular, we suggest that mobilization of the base of the LSW along the shelf forms B1 morphology, between 0 and 80 m water depths. Seafloor shelf-break jet currents at the base of the AW mass form the upper sediment wave unit of B2, between 80 and 130 m. A transition between velocities of the AW and the LIW mass from below forms the smooth morphology of B3, between 130 and 200 m. Below, the LIW mass forms sediment wave morphology along B4, between 200 and 350 m. Lack of along-slope morphologies in deeper water reflects slower vorticity of the lowermost water mass, the EMDW mass. Morphologies mapped here developed throughout the post LGM sedimentary succession. Their appearance is consistent despite seasonal/annual variations in oceanographic parameters. This consistency allows for the extrapolation of contemporary oceanographic conditions in order to explain the formation of similar morphologies in the past. Hence, we propose to address the sediment transport system as the Levant Jet System. The Levant Jet System developed along with the recovery of eastern Mediterranean thermo-haline circulation from glacial lowstand conditions. Seafloor current velocities gradually developed between the LGM and Holocene, along with the capacity of the currents to carry sediments. During this period, sedimentation was conformable. The jet system matured during the Holocene while sea levels rose. Intensification of the jet system possibly reflects enhanced water stratification. The youngest and shallowest morphological unit formed during the second half of the Holocene when the sea reached its present level, the mature jet system stabilized and erosion commenced along the shelf. Our results agree with previous studies regarding longshore transport of Nile sediments northwards. However, longshore transport is only one component of a larger jet system. Acknowledgements We thank Barak Herut, Gideon Amit, Gidon Tibor, Giora Boxer and the crews of the R/V Eziona and R/V Shikmona, including students of Department of Marine Geosciences, University of Haifa who collected data during educational cruises. We thank Paradigm Geophysical, IHS Kingdom Suite and Petrel-Schlumberger for providing academic licenses which enabled seismic processing and interpretation. We also thank the editor of Sedimentary Geology and two reviewers whose comments have improved the manuscript. References Almagor, G., 2000. The Mediterranean Coast of Israel. GSI/13/022. Jerusalem, Geological Survey of Israel. Avnaim-Katav, S., et al., 2012. The chronostratigraphy of a quaternary sequence at the distal part of the Nile littoral cell, Haifa Bay, Israel. Journal of Quaternary Science 27 (7), 675–686. Baldacci, A., et al., 2001. A study of the Alboran sea mesoscale system by means of empirical orthogonal function decomposition of satellite data. Journal of Marine Systems 29 (1), 293–311. Ben-Avraham, Z., et al., 2006. Segmentation of the Levant continental margin, eastern Mediterranean. Tectonics 25, TC5002. Ben-Gai, Y., Ben-Avraham, Z., 1995. Tectonic processes in offshore northern Israel and the evolution of the Carmel structure. Marine and Petroleum Geology 12 (5), 533–548.

Ben-Gai, Y., Ben-Avraham, Z., Buchbinder, B., Kendall, C.G.S.C., 2005. Post-Messinian evolution of the Southeastern Levant Basin based on two-dimensional stratigraphic simulation. Marine Geology 221 (1-4), 359–379. Box, M., Krom, M., Cliff, R., Bar-Matthews, M., Almogi-Labin, A., Ayalon, A., Paterne, M., 2011. Response of the Nile and its catchment to millennial-scale climatic change since the LGM from Sr isotopes and major elements of East Mediterranean sediments. Quaternary Science Reviews 30 (3), 431–442. Buchbinder, B., Martinotti, G.M., Siman-Tov, R., Zilberman, E., 1993. Temporal and spatial relationships in Miocene reef carbonates in Israel. Palaeogeography Palaeoclimatology Palaeoecology 101, 97–116. Carmel, Z., Imman, D., Golik, A., 1985. Directional wave measurements at Haifa, Israel, and sediment transport along the Nile littoral cell. Coastal Engineering 9, 21–36. Carton, H., et al., 2009. Seismic evidence for Neogene and active shortening offshore of Lebanon (Shalimar cruise). Journal of Geophysical Research 114, B07407. http://dx. doi.org/10.1029/2007JB005391. Coleman, J., Roberts, H., Murray, S., Salama, M., 1981. Morphology and dynamic sedimentology of the eastern Nile delta shelf. Developments in Sedimentology 32, 301–326. Fairbanks, R., 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. Flood, R.D., 1994. Abyssal bedforms as indicators of changing bottom current flow: examples from the US East Coast continental rise. Paleoceanography 9 (6), 1049–1060. Flood, R.D., Shor, A.N., Manley, P.L., 1993. Morphology of abyssal mud waves at Project MUDWAVES sites in the Argentine Basin. Deep Sea Research Part II: Topical Studies in Oceanography 40 (4), 859–888. Frihy, O., Lotfy, M., 1997. Shoreline changes and beach-sand sorting along the northern Sinai coast of Egypt. Geo-Marine Letters 17 (2), 140–146. Frihy, O.E., Fanos, A.M., Khafagy, A.A., Komar, P.D., 1991. Patterns of nearshore sediment transport along the Nile Delta, Egypt. Coastal Engineering 15 (5), 409–429. Garfunkel, Z., Almagor, G., 1985. Geology and structure of the continental margin off northern Israel and the adjacent part of the Levantine Basin. Marine Geology 62, 105–131. Goldsmith, V., Golik, A., 1980. Sediment transport model of the southeastern Mediterranean coast. Marine Geology 37, 147–175. Golik, A., 1993. Indirect evidence for sediment transport on the continental shelf off Israel. Geo-Marine Letters 13, 159–164. Golik, A., 1997. Dynamics and management of sand along the Israeli coastline. Bulletin. Institut Océanographique (Monaco) 97–110. Golik, A., 2002. Pattern of sand transport along the Israeli coastline. Israel Journal of Earth Sciences 51. Golik, A., Shoshany, M., Golan, A., Haimi, O., 1999. Sediment dynamics in Haifa Bay. IOLR Rep. Hp. 17. Hecht, A., Pinardi, N., Robinson, A.R., 1988. Currents, water masses, eddies and jets in the Mediterranean Levantine Basin. Journal of Physical Oceanography 18 (10), 1320–1353. Hjulström, Filip, 1939. Transportation of detritus by moving water. In: Trask, P.D. (Ed.), Recent Marine Sediments, a Symposium. American Association of Petroleum Geologists, Tulsa, Okla, pp. 5–31. Hyams-Kaphzan, O., Almogi-Labin, A., Sivan, D., Benjamini, C., 2008. Benthic foraminifera assemblage change along the southeastern Mediterranean inner shelf due to fall-off of Nile-derived siliciclastics. Neues Jahrbuch für Geologie und Paläontologie Abhandlungen 248 (3), 315–344. Inman, D.L., Jenkins, S.A., 1984. The Nile littoral cell and man’s impact on the coastal zone of the southeastern Mediterranean. Scripps Institution of OceanographyReference Series 84-31. University of California, La Jolla, p. 43. Jobe, Z.R., Lowe, D.R., Morris, W.R., 2012. Climbing‐ripple successions in turbidite systems: depositional environments, sedimentation rates and accumulation times. Sedimentology 59 (3), 867–898. Kontoyiannis, H., Theocharis, A., Nittis, K., 1999. Structures and Characteristics of Newly Formed Water Masses in the NW Levantine during 1986, 1992, 1995, The Eastern Mediterranean as a Laboratory Basin for the Assessment of Contrasting Ecosystems. Springer, pp. 465–473. Kress, N., Herut, B., 2001. Spatial and seasonal evolution of dissolved oxygen and nutrients in the Southern Levantine Basin (Eastern Mediterranean Sea): chemical characterization of the water masses and inferences on the N:P ratios. Deep Sea Research Part I: Oceanographic Research Papers 48 (11), 2347–2372. Lee, H.J., et al., 2002. Distinguishing sediment waves from slope failure deposits: field examples, including the ‘Humboldt slide’, and modelling results. Marine Geology 192 (1), 79–104. Malanotte-Rizzoli, P., et al., 2014. Physical forcing and physical/biochemical variability of the Mediterranean Sea: a review of unresolved issues and directions for future research. Ocean Science 10, 281–322. Millot, C., 1985. Some features of the Algerian Current. Journal of Geophysical Research 90 (C4), 7169–7176. Millot, C., 2009. Another description of the Mediterranean Sea outflow. Progress in Oceanography 82 (2), 101–124. Neev, D., 1960. A pre-Neogene erosion channel in the southern Coastal Plain of Israel. Bulletin. Geological Survey of Israel 25, 20. Nir, Y., 1980. Recent sediments of Haifa bay. MG/11/80, Geological Survey of Israel report, Jerusalem. Perlin, A., Kit, E., 1999. Longshore sediment transport on the Mediterranean coast of Israel. Journal of Waterway, Port, Coastal, and Ocean Engineering 125, 80–87. Pomerancblum, M., 1966. The distribution of heavy minerals and their hydraulic equivalents in sediments of the Mediterranean continental shelf of Israel. Journal of Sedimentary Research 36 (1). Rohling, E.J., Marino, G., Grant, K.M., 2015. Mediterranean climate and oceanography, and the periodic development of anoxic events (sapropels). Earth-Science Reviews 143, 62–97.

U. Schattner et al. / Sedimentary Geology 329 (2015) 28–39 Rosentraub, Z., Brenner, S., 2007. Circulation over the southeastern continental shelf and slope of the Mediterranean Sea: direct current measurements, winds, and numerical model simulations. Journal of Geophysical Research, Oceans 112 (C11), 1978–2012. Ruiz, S., Pascual, A., Garau, B., Faugère, Y., Alvarez, A., Tintoré, J., 2009. Mesoscale dynamics of the Balearic Front, integrating glider, ship and satellite data. Journal of Marine Systems 78, S3–S16. Schattner, U., Ben-Avraham, Z., Lazar, M., Hübscher, C., 2006. Tectonic isolation of the Levant basin offshore Galilee-Lebanon - Effects of the Dead Sea fault plate boundary on the Levant continental margin, eastern Mediterranean. Journal of Structural Geology 28, 2049–2066. Schattner, U., Lazar, M., Tibor, G., Ben-Avraham, Z., Makovsky, Y., 2010. Filling up the shelf—A sedimentary response to the last post-glacial sea rise. Marine Geology 278 (1), 165–176. Schattner, U., Lazar, M., Harari, D., Waldmann, N., 2012. Active gas migration systems offshore northern Israel, first evidence from seafloor and subsurface data. Continental Shelf Research 48, 167–172. Shoshany, M., Golik, A., Degani, A., Lavee, H., Gvirtzman, G., 1996. New evidence for sand transport direction along the coastline of Israel. Journal of Coastal Research 311–325.

39

Sivan, D., Wdowinski, S., Lambeck, K., Galili, E., Raban, A., 2001. Holocene sea-level changes along the Mediterranean coast of Israel, based on archaeological observations and numerical model. Palaeogeography Palaeoclimatology Palaeoecology 167, 101–117. Stanley, D.J., 1989. Sediment transport on the coast and shelf between the Nile Delta and Israeli margin as determined by heavy minerals. Journal of Coastal Research 5, 813–828. Trincardi, F., Verdicchio, G., Miserocchi, S., 2007. Seafloor evidence for the interaction between cascading and along‐slope bottom water masses. Journal of Geophysical Research - Earth Surface 112 (F3), 2003–2012. Werner, F., Newton, R.S., 1975. The pattern of large-scale bed forms in the Langeland Belt (Baltic Sea). Marine Geology 19 (1), 29–59. Wynn, R.B., Stow, D.A., 2002. Classification and characterisation of deep-water sediment waves. Marine Geology 192 (1), 7–22. Zviely, D., et al., 2006. Holocene evolution of the Haifa Bay area, Israel, and its influence on ancient tell settlements. The Holocene 16, 849–861. Zviely, D., Kit, E., Klein, M., 2007. Longshore sand transport estimates along the Mediterranean coast of Israel in the Holocene. Marine Geology 237, 61–73.