1During the ten last years, several marine-coastal sequences of the Last Interglacial times (Marine Isotope Stages 5.5, 5.3 and 5.1), formerly called “Ouljian” through the whole Morocco, were performed along the Atlantic Moroccan and Tangiers Peninsula coasts with remarkable differences. New records from Rabat-Temara, Agadir-Tamraght and Tangier-Strait of Gibraltar, Dhâda (tab. 1) are sampled here and their chronological, paleo-environmental and tectonic implications briefly discussed.
Tab. 1: U/Th, OSL and IRSL datings for MIS 5 fossil sea-levels at Tangier-Dhâda, Rabat-Temara and Agadir-Tamraght. In blue: new datas, in this volume.
Fig. 1: Morocco: main coastal landforms and geological structure; location of the three studied examples.
(G. Mahieux infography)
2In this quite tectonic-stable area, the Meseta coast, beneath a +150 m high “Moghrebian” (Early Pleistocene) wide platform, is a typical “oulja” coast, where the Paleozoic basement and its covering strata rarely appear at the shore, producing small monoclinal capes called “skhour”. On this basement, the “oulja” geomorphological system of the Rabat-Temara area develops mainly in the Pleistocene marine and aeolian sandstones, with three classical landforms occurring seawards (fig. 2A): i) an inner dead-cliff cut in the eolianites, with basal caves at ca +8 m; the caves infillings constitute the six main prehistorical sites of the area (El Hajraoui et al., 2012); ii) a more or less wide and low depression (the “oulja” sensu stricto), sometimes enclosing a lagoon; the oulja s.s. corresponds to the former shore platform and its more or less thick successive marine and continental covers; iii) an external near-shore ridge of cemented coastal dunes where present low cliffs occur. This shoreline is underlined by coastal features of karstic dissolution developed in the eolianites (Guilcher et Joly, 1954; Gomez-Pujol et Fornos, 2010). Their present zonation shows mainly coastal karren and pools in the upper-tidal spray zone, with sometimes blow-holes related to the basal caves; such detail landforms provide good indicators for past fossil shorelines (fig. 3);
Fig. 2: Coastal geomorphologic patterns. A – « Oulja »-type coast at El Harhoura-Temara, after Chahid & al., in this volume; B – Stepped-like terraces coast at Arhoud, Atlantic Atlas, after Weisrock, 1980, modified (today, the Holocene terrace has totally disappeared).
Fig. 3: Fossil coastal karren and pot-hole at the top of Dar Es Soltane I Cave (Rabat, 2014 June).
3Geomorphological differences with the Meseta are due to the uplift and folding of the Mesozoic substrate, inducing the formation of stepped-like marine terraces with thicker detrital cover. The lowest fossil terrace likes an oulja, but appears rarely complete, due to the progressive marine erosion towards inland. In most cases, it becomes a more or less wide lower platform, where preserved marine, fluvio-marine, eolian and continental-colluvial deposits lie in discontinuous superposed layers (fig. 2B). Due to the mountainous inland, the thickness of this platform is highly variable; in most cases, the colluvial debris cover masks the inner dead cliff. At Tamraght however, this composite platform stays beneath +8/10 m, basis of the inner dead cliff, and is cut in lower active external cliffs at the present shoreline.
4This point, at the middle part of the Strait’s southern coast, consists of high cliffs cut in the flyschs of the Cenozoic “Beni-Ider nappe”, which presents vertically dipping layers of sandstones, limestones and clay-marls. The cliffs cut obliquely the Rifian arch thrusted structures but are active today only at their foot, showing fossil marine deposits on a marine abrasion surface at ca. +15 m high. It is a great similarity between the fossil deposits and the present ones; in both cases, early transgressive deposits with blocks are thicker and better developed on clay-marl bedrock.
5On the southern side of the Gibraltar Strait, two superposed transgressive marine units U1 and U3 highlight two transgressive pulsations at Dhâda and neighbouring sections at ca. +15 m. Despite the fact that only U3 is (by U/Th on coral) dated, from at the most 119,6 ± 2,3 ka, El Abdellaoui et al. (2016, in this volume) deduced an age of ca. 125-130 ka for U1 and concluded at the occurrence of two MIS 5.5 pulsations. The other results: 111,2 ± 2,1 and 106,7 ± 3,3, 96,5 ± 2,3 can however suggest a MIS 5.3 age for U3, and consequently even also for U1, due to the fact that the intermediate layer U2 is very thin. In fact, the basal point of the dead cliff related to the marine platform of U1 was not yet observed. The main result is the high tectonic uplift rate evidence in the Strait of Gibraltar, with an elevation of ca. 8-11 m for the Dhâda terrace since the MIS 5.5. If it was since the MIS 5.3, we do find a higher marine abrasion terrace for MIS 5.5, and that remain to search further. Even if it seems not be the case at Dhâda, many higher stepped marine terraces are observed eastwards from this site (Chalouan et al., 2008). The high uplift rate keep the marine erosion active, and in contrast with the Atlantic coast, all more recent and lower fossil coastal landform disappeared.
6The Rabat-Temara area is the best documented for MIS 5.5 (tab. 1), with numerous age measurements made on the marine layers of the basal infilling of five well-known caves, due to the archeological study of their occupation by the Aterian Prehistoric Man (El Hajraoui et al., 2012). In spite of numerous records, substages MISS 5.5.1 and MISS 5.5.2 cannot be statistically distinguished (Jacobs et al., 2012), but we can say that MIS 5.5 high sea-level is detected in all caves: i) at Dar Es Soltan I at +4 m, with two OSL measurements in “basal marine coarse sands and lumachelles”: 151,4 ± 9,1 ka and 131,4 ± 17,7 ka (Barton et al., 2009; Schwenninger et al., 2010); ii) at El Mnasra at +8/+10 m by two OSL dating in “marine sands of level 13”: 118,6 ± 20,1 ka (Schwenninger et al., 2010) and 134,5 ± 2,2 in level 12 (Jacobs et al., 2012); iii) at Contrebandiers Cave, by two OSL measurements in “beach deposits” at ca. +8 m: 129,0 ± 6,6 ka and 121,8 ± 13,8 ka (Schwenninger et al., 2010); Iv) at Dar Es Soltan II, at ca. +4 m, by one OSL date at 121,7 ± 8,2 ka in “basal sands” (Schwenninger et al., 2010); v) at El Harhoura II, at ca. +8 m, by one OSL date at 124 ± 7 ka in “level 11, basal sands” (Jacobs et al., 2012). Furthermore, the probably eolian sands of “level 8” in the same cave El Harhoura II, giving a 107 ± 7 ka OSL age (Ibid.), may indicate the nearness of the MIS 5.3 high sea-level. This fact is corroborate out of the caves, on the linked seawards marine abrasion surface, where the most scenic landform consists in a high continuous coastal aeolian sandy ridge of MIS 5.3 age, well defined at El Harhoura with four OSL dates between 103,8 ± 8,2 ka and 93,9 ± 7,1 ka (Chahid et al. (2016, in this volume). Even if some uncertainties remain about the number of the dead-cliffs and the altitudes of the correlative sea-levels, this bulk of chronological records make the concluding remarks of Barton et al. (2009), Schwenninger et al. (2010), Jacobs et al. (2012) very consistent for the MIS 5.5 genesis of the main “Ouljian” dead-cliff and its basal caves at ca. +8 m. MIS 5.3 seal-level don’t reached again this altitude, but left abundant coastal dunes. After this coastal dune edification, one or more new high sea-level invaded the coastal area and mainly the “oulja”, but their precise dating remains not known at this day.
7The late MIS 5 coastal history gets some more precise knowledges at Agadir-Tamraght, at the mouth of Assif (wadi) Tamraght (Weisrock et al., 2016, in this volume): at both sides of this mouth, two sea-levels are identified. The first of them (MIS 5.3) reached certainly more than +3 m and is dated by U/Th at 107,2 ± 0,9 ka and by IRSL at 101 ± 12 ka. The second (MIS 5.1) occurs between +1,5 m, altitude of its basal marine abrasion surface, and culminate at +5/+6 m with fluvio-marine and near-shore aeolian deposits, where five datations by U/Th, OSL and IRSL are giving ages between 76,9 ± 1,4 ka and 93 ± 11 ka. Situated at sometimes more than 500 m from the main dead-cliff, the linked fossil landforms (marine abrasion surfaces covered by marine and fluvio-marine deposits, related dead-cliffs) were well preserved, due to the early carbonated cementation, the thickness of the sus-jacent continental deposits (fig. 2B) such as that of the wide coastal dune system of MIS 4 age and the later abundant colluvial debris furniture of the inner slopes, according to a continuous but relatively moderate uplift rate and successive wet climatic episodes (Weisrock, 1980).
8So we can conclude that the present Atlantic coastal landforms of Morocco are all greatly dependant of the MIS 5 fossil features related with the last Interglacial, but with geomorphological and chronological differences. Geomorphological differences are mainly due to regional structures such as different rates of uplift, local coastal dynamic processes such as upwelling or marine current influences, and climatic conditions of preservation such as early marine cementations or continental development of calcrusts. Chronological measurements show that the different MIS 5 fossil features may also belong to different MIS 5 substages. As demonstrated by Guérémy (2003) for the Mediterranean Calabrian coast, further detailed geomorphological studies of the Atlantic Moroccan coast may give more proofs of millennial-scale climatic oscillations of the sea-level.
9As suggested by Schwenninger et al. (2010), Jacobs et al. (2012), El Hajraoui et al. (2012), Stoetzel et al. (2014), Campmas et al. (2016), for the Atlantic coast, and Ramos-Munoz et al. (2016) for the Strait of Gibraltar, the MIS 5 coastal environments gave attractive features for the Aterian (MSA) Man and contemporaneous fauna settlements during wet periods of savannahs development. So, the several millennial-scale regressive marine episodes (MIS 5.4, MIS 5.2) are thought to be correlable with the growth of vegetal cover and the occurrence of Mammals and MSA hunters-gatherers. This conclusion is well documented at El Harhoura II and El Mnasra Caves: above the high sea-level basal deposit of MIS 5.5, a notable human occupation is found at ca. 116 ± 6 ka, as the shore was not yet very distant (less than a few kilometres), giving a food complement of fishes and molluscs. This favourable situation of shore nearness occurs again at ca. 106 ± 6 ka (just before the MIS 5.3 transgressive maximum) at both caves, and even also at ca. 75 ± 4 ka at El Harhoura II (beginning of the regressive phase after the MIS 5.1 maximum). However, about the numerous sterile levels of the caves fills, such correlations remain highly hypothetical: a high sea-level may correspond to a warmer and possibly locally drier episode, with wide invasion of coastal dunes linked to the shore strait position; more simply, high sea-level is the time of coastal cave genesis and not of coastal cave occupation by man! But abandonments can be also attributed at many other reasons that are not inferred at the geomorphological features.
10Althought the always present risk of tsunami, and joint to the global warming and sea-level rise, present off-shore and on-shore anthropic impact and above all the explosive coastal urban development of Rabat-Temara, Agadir and Tangier induces very destructive activities for the past fossil landforms. If they facilitated the archeological discoveries, the sand and stone quarries destroyed irreversibly the fossil coastal dunes. Planned managements such as dams on the neighbouring fluvial basins perturbed drastically the sedimentary supply to the coastal regeneration of beaches and dunes. Harbour installations modified greatly the coastal marine processes and induced pollutions. For any successful management of the coastal problems, the whole regional situation must be taken into account. A preliminary safeguard of the most valuable original archeological and geomorphological patrimonial sites, through coastal national parks, seems to be evident.