Tectono-Sedimentary Cenozoic Evolution of the El Habt and Ouezzane Tectonic Units (External Rif, Morocco)

: An interdisciplinary study based on lithostratigraphic, biostratigraphic, petrographic and mineralogical analyses has been performed in order to establish the Cenozoic tectono-sedimentary evolution of the El Habt and Ouezzane Tectonic Units (External Intrarif Subzone, External Rif, Morocco). The reconstructed record allowed identiﬁcation of the depositional architecture and related sedimentary processes of the considered units. The Cenozoic successions were biochronologically deﬁned allowing, at the same time, identiﬁcation of unconformities and associated stratigraphic gaps. The presence of ﬁve unconformities allowed for the deﬁnition of the main stratigraphic units arranged in a regressive trend: (1) lower Paleocene interval (Danian p.p.) assigned to a deep basin; (2) Eocene interval (lower Ypresian-lower Bartonian p.p.) from a deep basin to an external carbonate-siliceous platform; (3) lower Rupelian-upper Chattian p.p. interval deposited on unstable slope with turbidite channels passing upward to an external siliciclastic platform; (4) Burdigalian p.p. interval from a slope; (5) Langhian-Serravallian p.p. interval from slope to external platform realms. The petrography of the arenites and calcarenites allowed for the identiﬁcation of the supplies derived from erosion of a recycled orogen (transitional and quartzose subtypes). The clay-mineralogy analysis indicates an unrooﬁng (ﬁrst erosion of Cretaceous terrains followed by upper Jurassic rocks) always accomplished by erosion of Cenozoic terrains. Several tectofacies checked in some stratigraphic intervals seem to indicate the beginning of deformation of the basement generating gentle folds and ﬁrst activation of blind thrusts, mainly during the Paleogene. A preorogenic tectonic framework is considered as responseto the generalized tectonic inversion (from extension to compression) as frequently registered in the central-western peri-Mediterranean areas. The large volumes of reworked terrigeneous supply during the latest Oligocene-Miocene p.p. indicates the beginningsof the synorogenic sedimentation (foredeep stage of the basins) controlled by active tectonics. peak areas of quartz (Qtz(001):Qtz(101) ratio hereafter) in the whole-rock XRD di ﬀ ractograms; the intensities ratio of the Sme(003):Sme(002) peak areas of smectite (Sme(003):Sme(002) ratio hereafter) from ethylene-glycol solvated clay-fraction XRD di ﬀ ractograms to di ﬀ erentiate dioctahedral and trioctahedral; and the intensities ratio of the Ill(002):Ill(001) peak areas of illite (Ill(002):Ill(001) ratio hereafter) from decomposed air-dried clay-fraction XRD di ﬀ ractograms to discern authigenic from mature illite. calcite (5–78%), dolomite (6–12%) and minor amounts of K-feldspar, plagioclase, opal CT and clinoptilolite. The clay fraction includes smectite (5–36%), illite (23–61%), mixed layer I-S (5–37%), kaolinite (5–60%), chlorite (5–7%) and palygorskite (5–14%). The clay-mineral associations consist of Ill + (I-S) ± Sme + Kln for Ypresian and Ill + Kln ± (I-S) + Sme + Chl for the Lutetian. The Oligocene portion (S42 to S44, samples) include quartz (5–33%), phyllosilicates (15–48%), calcite (21–71%) and minor amounts of K-feldspar and plagioclase. The clay fraction shows smectite (5–50%), illite (40–60%), mixed layer I-S (21–28%), kaolinite (21–25%), chlorite (5–8%)and palygorskite ( < 5–7%). The above data indicates Ill + Kln ± (I-S) + Sme + Chl association. The lower-middle Miocene (S45 to 49, samples) are made of quartz (9–23%), phyllosilicates (23–57%), calcite (19–61%) and minor amounts of K-feldspar, plagioclase, opal CT and clinoptilolite. The clay fraction includes smectite (5–7%), illite (45–61%), mixed layer I-S (10–14%), kaolinite (15–31%), palygorskite (5–8%) and minor amounts of chlorite. The Ill + (I-S) ± Sme + Kln + Chl clay-mineral association is represented.

. (A) Simplified map of the north-western African area; (B) Simplified tectonic sketch map of the central-western peri-Mediterranean chains (after [11], modified); (C) Geological sketch map of the central-western Rif Chain (after [21], modified).

Geological Setting and Previous Work
The External Rif Domain ( Figure 1C) comprises the sedimentary cover of the North-African passive paleo-margin in contact with the southern Maghrebian Tethyan Ocean branch ( [22][23][24]; among others). This passive margin was affected during the Mesozoic by several phases of sedimentation and subsidence.

Geological Setting and Previous Work
The External Rif Domain ( Figure 1C) comprises the sedimentary cover of the North-African passive paleo-margin in contact with the southern Maghrebian Tethyan Ocean branch ( [22][23][24]; among others). This passive margin was affected during the Mesozoic by several phases of sedimentation and subsidence.
The subdivision of the ERZ in Intrarif, Mesorif and Prerif proposed by Suter [14] has been generally accepted. This author further subdivides the Intrarifian units into: (i) Intrarifian origin nappes and (ii) an Intrarifian Zone comprising from internalto external positions, the Ketama, Internal Tanger, External Tanger andLoukkos Units. These units also show differences on the stratigraphic record Suter [14]. So, the Ketema unit is considered the basement of the Tanger unit showing only Jurassic and Lower Cretaceous rocks with a slight metamorphism sometimes. The Internal Tanger unit is only made of Upper Cretaceous rocks. The External Tanger unit shows Upper Cretaceous rocks (similar to those of the Tanger-Interne) and also Cenocoic rocks. Loukkos is also made of Upper Cretaceous to Cenozoic rocks but representing a shallow marine area while the other units show deep marine environments. However, this approach has created confusion concerning the correlation of the stratigraphic records in different areas of the same subzone because of the use of differentstratigraphic and tectonic criteria by different authors. This problem is further worsened by the fact that units attributed to the Intrarif are: deeply rooted, more or less detached from their original basement (parautochthonous/allochthonous) and affected by superficial gravity-driven tectonics. This ensemble is also affected by two major ENE to NE striking strike-slip faults: the Jebha and Nekor faults related to the westward movement of the Alboran Domain relative to the African and Iberian plates.
So, the Intrarif represents the most internal subzone of the ERZ and it is constituted by various tectono-sedimentary units which are differently defined and interpreted based on different criteria and methodological approaches ( [21,[27][28][29][30][31][32][33][34][35][36][37][38]; among others and references therein). Indeed, many authors have mentioned the tectonic complexity of the Intrarif subzone, underlining the existence of multiple nappe units displaced from their original paleogeographic locations and overriding more external subzones. These tectonic translations have obscured original spatial and temporal relationships between stratigraphic successions.
In this way, the Intrarif subzone is subdivided into the Internal and the External Intrarif. The Internal Intrarif comprised the Tanger-Ketama and Loukkos units (para-autochtonous). In turn, the Tanger Unit is further divided into a Cretaceous internal subunit and a Cenozoic external one. The External Intrarif is made of several other unrooted units or tectonic klippen (Aknoul, Tsoul, El Habt, Ouezzane Units, etc.) which are considered to be of Intrarifian origin (e.g., [14]).
Recently, we have previously worked on other sectors of the western External Riffian Zones. In concrete, we have published two papers on the External Tanger Unit [21,38]. Our studies have furnished new stratigraphic, paleogeographicand paleotectonic data on the Cenozoic successions of that sector.
Despite these studies and due to the vastness of the Rif, many problems are still unresolved concerning the ERZ which might also be due to a lack of homogeneous data and poor integration. Main problems arise from the nomenclature mixing stratigraphic and tectonic terms. Other problems are related to the lack of agreement in the original paleogeographic position of the External Intrarif units. Another problem is that recently an oceanic or thinned crust has been proposed for the upper Jurassic-Miocene Mesorif units [52,53]. Moreover, the relationships of the ERZ with the innermost domain, represented by the Maghrebian Flysch Basin (deposited on oceanic crust), are still unknown. In addition, the nature of the basement (oceanic, transitional or continental) of the most Internal Intrarif Zone (Tanger unit) is under discussion. Nevertheless, a general framework of the ERZ is presented by Frizon de Lamotte et al. [23] which can constitute a useful reference. Other stratigraphic reconstructions of the ERZ have been recently addressed to the location of the hydrocarbon reservoir [17,19,20,54].

Aim and Methods
This study consists of a better definition of the Cenozoic stratigraphic record and lateral correlations supported by petrographic and mineralogical characterization of the El Habt and Ouezzane Tectonic Units belonging to the Intrarif subzone. This new contribution, and two previous ones [21,38] constitutes a progressive study of the ERZ of our research group.
This research is based on interdisciplinary methodological approach (stratigraphic reconstructions, biostratigraphy dating, mineralogical and petrographic analysis, etc.) of the upper Cretaceous-Cenozoic succession. More in detail, the utilized methodology consists of: (i) field analyses (lithostratigraphic reconstructions and sampling of specific stratigraphic sections and tectonic observations) and (ii) laboratory analysis: biostratigraphy, petrography of arenites, clay mineralogy, also used to improve the stratigraphic interpretation.
Data integration allowed space-time reconstructions of the events including specific objectives such as (i) characterization of the Intrarifian Origin Units; (ii) identification of sedimentary gaps and angular unconformities, sequential stratigraphy, recognition of the main stratigraphic and tectonic events, and, finally, a paleogeographic and paleotectonic reconstruction.
The biostratigraphic study is based on the planktonic foraminifera found in 55 beds (9 samples in N Tahar Village section, 7 in Mezgalef, 18 in Douar Ahel Chane and 21 in Oulad Ktir) of marly and clays subjected to the traditional washing and fractionation by sieves of 150 µm (main studied fraction) and 125 µm. The subtropical biozonations established by Berggren et al. [55], Olsson et al. [56], Berggren and Pearson [57,58] and revised by Wade et al. [59] were considered in the Paleogene sediments. The standard zonation of Blow [60] was used for the Neogene. The most significant bioevents of the Mediterranean area [61][62][63] were used together for the Global Standard Chronostratigraphic Scale [64].
Preliminary mineralogical analyses on the whole-rock and <2 µm grain-size fraction (clay fraction hereafter) mineralogy of selected 19 samples were performed. The samples were ground and the obtained powders were analyzed by X-ray diffraction (XRD) using a PANalytical X'Pert Pro diffractometer (Cu-Kα radiation, 45 kV, 40 mA), equipped with an X'Celerator solid-state lineal detector. The diffraction patterns were obtained by a continuous scan from 3 • 2θ to 60 • 2θ, with a 0.01 • 2θ resolution. The XPOWDER ® program [72] was used for semiquantitative mineral composition interpretation of the whole-rock and clay-fraction samples.

Stratigraphy of the El Habt and Ouezzane Nappes
For each unit, two stratigraphic sections were lithostratigraphically reconstructed ( Figure 1C). In each of them, field observations (lateral relationships, internal and external structures of the beds, sampling, etc.) ( Figure 2A) together with biochronostratigraphy analyses ( Figure 2B) were performed.

El Habt Tectonic Unit
The El Habt Unit (External Intrarif Subdomain) consists of a duplex structure in which subunits show a westward displacement. Two logs (Log1, N Tahar Village; and Log 2, Mezgalef) were reconstructed, measured and sampled (Table 1, Figure 2). In this area, the succession of the N Tahar Village is overthrusted by the Mellouse Flysch Nappe and the External Tanger Unit northward. In addition, this last unit thrusts the most Internal Prerif in the NW sector where the Mezgalef log has been studied. Table 1. Main field data (units, logs, localities, UTM coordinates, thickness, samples, boundary types, lithofacies and lithological associations) and laboratory results (ages, petrofacies and mineralogical associations). * Classification (NCE/CI/CE) according to Zuffa [68,69]; ** Classification (QFL) according to Dickinson [66]; Zuffa [68] and Ingersoll et al. [67].

El Habt Tectonic Unit
The El Habt Unit (External Intrarif Subdomain) consists of a duplex structure in which subunits show a westward displacement. Two logs (Log1, N Tahar Village; and Log 2, Mezgalef) were reconstructed, measured and sampled (Table 1, Figure 2). In this area, the succession of the N Tahar Village is overthrusted by the Mellouse Flysch Nappe and the External Tanger Unit northward. In addition, this last unit thrusts the most Internal Prerif in the NW sector where the Mezgalef log has been studied.
Log 1 (N Tahar Village). The considered succession is about 130 m thick. It belongs to a monoclinal structure where an unconformity separates different stratigraphic intervals: (i) a Cretaceous substrate made of greenish pelites (samples S1 and S2); (ii) a first interval 75 m thick of greyish pelites with variable content in marls or sand; (iii) follows a second interval of 55 m thick of varicoloured pelites, marls, calcareous marls and arenites to the top. The Cretaceous and lower Cenozoic intervals show signs of carbonate dissolution (S3 to S5 samples) as also the radiolarians and rare planktonic foraminifers attest in the lower Cenozoic levels. The presence of A. soldadoensis, M. subbotinae and M. aragonensis (cf. M. lensiformis (Subbotina) foraminifera indicate the E4-E5 zones referable to middle-late Ypresian. Upwardly, some levels of the log (samples S6 and S7) are azoics but in the uppermost one (S8) A. bullbrooki, A. cuneicamerata, I. broedermanni, P. micra and forms similar to T. possagnoensis indicatethe lower-middle Lutetian (E8-E9 zones). The features of this bioassembalge could indicate a deep basin realm. Table 1. Main field data (units, logs, localities, UTM coordinates, thickness, samples, boundary types, lithofacies and lithological associations) and laboratory results (ages, petrofacies and mineralogical associations). * Classification (NCE/CI/CE) according to Zuffa [68,69]; ** Classification (QFL) according to Dickinson [66]; Zuffa [68] and Ingersoll et al. [67]. Log 2 (Mezgalef). The succession, 330 m thick, is part of a monoclinal structure where three stratigraphic intervals can be distinguished. To the top, an interval of cross-laminated Quaternary sands of beach environment is overcome by means an unconformable surface. The lower interval (80 m thick) is made of brownish pelites and turbidite beds (arenites with a variable content in marls) with added sands and Fe-Mn and chert nodules. The microfauna is mainly made of agglutinated foraminifera and/or radiolarians. The set of features indicates a deep realm referable to a probable slope/channel. The upper interval consists of channelized and amalgamated lenticular arenites of up to 30 m thick with occasional calcarenites referrable to a submarine fan. This last interval is followed by 70 m of brownish pelites with arenites intercalations and of occasional microconglomerate beds (10-30 cm thick). The identification of G. ampliapertura group, G. eocaena group, and P. opima (samples S17 to S21) in this last interval level indicates middle Rupelian (O2 zone). The succession of this log is arranged into minor depositional cycles with a shallowing trend.

Ouezzane Tectonic Unit
The sedimentary record of the Ouezzane Unit has been reconstructed in Log 3 (Douar Ahel Chane) and Log 4 (Oulad Ktir logs) (Table 1, Figure 2). In this unit the age of the sedimentary record is wider than that of the El Habt Unit since the Paleocene-middle Miocene is well represented and the older stratigraphic beds seem to be pelagic and frequently below the CCD level. This nappe thrusts the Internal Prerif (in both studied logs) while in other sectors thrusts the External Prerif.  T a-b , T a-b-c , T a-b-c-d , T a-c ) terrigeneous arenites (S42: litharenites) and calcarenites. This interval (sample S43) shows Oligocene microfauna (Zone O2: middle Rupelian) characterized by the presence of G. ampliapertura, G. increbescens, G. eocaena group (including G. corpulenta) and C. dissimilis, while Eocene typical species as P. micra are absent. Upwards (sample S44) the late Rupelian is dated by the presence of P. opima. The above seem to indicate that the unconformity comprises a gap affecting the late Eocene (most part of the Bartonian and the Priabonian) and the lowermost Oligocene (Lower Rupelian). The Oligocene interval seems to belong to slope realms with the presence of diverse turbidite facies.
A new interval, 150 m thick, is deposited after a new unconformity with a gap comprising the late Oligocene up to the base of the Burdigalian. This interval is made of thick calcarenites, brownish quartzose arenites, limestones and calcareous-marls. The arenites show internal Bouma structures like : T a-b , T a-b-c ,  T a-b-c-d , and T a- After an angular unconformity, the second interval (320 m thick) is made of yellowish-whitish marls and local calcareous marls. This interval is dated (Sample S52) to Paleocene p.p. (P2 zone) by the presence of P. pseudobulloides, P. uncinata, P. inconstans, and G. compressa and the absence of Acarinina and Morozovella.
After a second unconformity the third interval, 60 m thick, shows yellowish-grey marls, thin arenites and occasional conglomerates containing polygenic and cherty clasts at the base. The lower portion of this interval is referable to the lower Ypresian (E3 zone; sample S53) for the presence of After a new unconformity with a gap comprising the upper Eocene and part of the lower Oligocene, a fifth interval of 247 m in thickness is made of brownish to cream marls dated at upper Rupelian (S62 and S63 samples) by the disappearance of G. ampliapertura (marker of the O2/O3 zonal boundary) and the occurrence of G. angulisuturalis (O3/O4 zonal boundary of the uppermost Rupelian). Later, the disappearance of the P. opima (O5/O6 boundary, early Chattian) is registered in the sample S64 where G. oecaena group, G. cf. primordius and morphotypes close to G. praebulloides (with supplementary dorsal apertures) appear indicating the late Chattian.
After a new unconformity, a sixth interval (15 m thick) is made of calcareous marls and marly limestones with intercalations of fine-medium amalgamated arenitic beds (with breccias levels of 5-10 cm thick at the base) appears dated as lower Burdigalian (S65 to S67) by the presence of G. altiaperturus, G. trilobus, G. peripheroronda, G. suteri and C. dissimilis and the absence of G. subquadratus. Upwards the same lithofacies follow to the end. This interval is dated (S68 to S72 samples) at Lower Langhian (Zone N8), in the lower part, by the presence of G. bisphaericus, P. sicanus and P. glomerosa, and, in the upper part, as late Langhian (S73 sample) by the presence of Orbulina.
In a similar way to the Log 3, the Miocene succession seems to indicate the transition from the slope to the external platform environments.
In short, the El Habt Unit is prevalently characterized by the widespread presence of thick quartzarenites, pelites and marly pelites, while the Ouezzane Unit is characterized by a wider diversity of lithofacies showing immature arenites, calcarenites and poligenic conglomerates. Both successions are affected by many unconformities and biostratigraphic gaps, the meaning of which will be discussed below.

Discussion and Interpretations
The data above exposed allowed some advancement on the Cenozoic tectono-sedimentary evolution, on the sedimentary paleoenvironments and on the source areas of the El Habt and Ouezane subdomains of the ERZ.

Tectono-Sedimentary Evolution
Five main unconformities to which different biostratigraphic gaps associate have been detected (Figure 4). The unconformity boundaries have been located at the base of the older stratigraphic level dated in each depositional sequence.
The oldest unconformity (Unconformity-1) recognized in the Ouezzane Unit (Log 4) shows a gap extending from the uppermost Cretaceous to the Lowermost Paleocene, observable only in the Ouezzane Tectonic Unit (Log 4). The Unconformity-2 is located at the Paleocene-Eocene boundary with a gap affecting the middle Paleocene-lowermost Ypresian and it is recognized in the El Habt Tectonic Unit (Log 1) and in the Ouezzane Unit (Log 4). The Unconformity-3 concerns the Eocene-Oligocene boundary checked only in the Ouezzane Unit (Logs 3 and 4) showing a gap affecting the upper Eocene-lowermost Ruppelian. The Unconformity-4 is also exposed only in the Ouezzane Unit (Logs 3 and 4) with a gap marking the Chattian p.p.-lowermost Burdigalian time span. The youngest unconformity (Unconformity-5) is only restricted at the Burdigalian-Langhian boundary in the Ouezzane Unit (Log 4). This unconformity apparently laterally passes to continuity in the succession of the Log 3.
The checked unconformities allow for the division of the stratigraphic record considered into five stratigraphic units with the degree of depositional sequences (  2, 3 and 4), whose lithofacies indicate a sedimentation in slope-channel paleoenvironments; (4) Burdigalian p.p. reconstructed in the succession of the Ouezzane Unit (Logs 3 and 4) indicating a slope to external platform paleoenvironments; (5) Langhian-Serravallian p.p. checked in the Ouezzane Unit (Logs 3 and 4) deposited in an external platform. The above exposed indicates a Cenozoic main regressive cycle (from deeper deposits to platform-slope realms).
In the gaps, two portions can be differentiated: the portion below the unconformity boundary and that located above the unconformity (Figure 4). The lowest portion is related to erosion, probably during lowstand intervals (erosional gap), while the uppermost is related to non-deposition over paleogeographic paleorelifs (depositional gap) by means of an onlap device, probably during transgressive intervals. The defined sequences are arranged into transgressive-regressive intervals in which the recorded intervals usually belong to transgressive intervals, while the regressive correspond to the erosive intervals of the gaps. evolution, on the sedimentary paleoenvironments and on the source areas of the El Habt and Ouezane subdomains of the ERZ.

Tectono-Sedimentary Evolution
Five main unconformities to which different biostratigraphic gaps associate have been detected The oldest unconformity (Unconformity-1) recognized in the Ouezzane Unit (Log 4) shows a gap extending from the uppermost Cretaceous to the Lowermost Paleocene, observable only in the Ouezzane Tectonic Unit (Log 4). The Unconformity-2 is located at the Paleocene-Eocene boundary The proposed stratigraphic architecture clearly indicates a non tabular sedimentation with marked lateral changes as suggested by lateral variation of the: lithofacies, thicknesses, and diachronous stratigraphic boundaries. According to regional data [8][9][10][11][12]80], after the Cretaceous a paleogeographic reorganization of the external basins of the African Margin took place as a consequence of the beginning of the tectonic inversion (from extensional to compressional) between Europe and Africa plates. In short, this structuring is related to the neo-Alpine phase (Pyrenean of [81]) as response to the beginning of the Africa/Europe convergence during the Eocene and Oligocene. This reorganization would have caused modifications recorded in the stratigraphic framework as unconformities and lateral variations.
In addition, across the Eocene-Oligocene boundary, the occurrence of the well-known sea level fall in the entire globe due to a climatic cooling caused a generalized regression everywhere. The beginning of convergence (prenappe tectonics) is characterized by differential basement uplift that generated mild folds (also reflected on surface) and initiated blind thrusts leading to the end of the preorogenic sedimentation in the central-western Mediterranean area [80,[82][83][84]. The beginning of the large and growing volumes of reworked terrigeneous supply (turbidites, slumps, olistostromes, etc.) from the latest Oligocene and during the Miocene p.p. indicates the starts of the synorogenic sedimentation (foredeep stage of the basins) controlled by active tectonics contemporary to the sedimentary processes.

Supply Origin
Modal analyses of arenites indicate an origin from "recycled orogen" tectonic setting [85] with a trend from Paleogene to Miocene from the quartzose recycled to the transitional recycled ( Figure 5). Recycled quartz grains (second-cycle) together with frequent fragments of sedimentary rock are probably due to the erosion of older siliciclastic and carbonate formations. The abundance of recycled monocrystalline quartz, the shortage of feldspars, micas (unstable minerals) and the presence of zircon (ultra-stable minerals) indicate a multicyclic origin derived very probably from the African craton in tune with paleogeographic position of the basins.   [21]) and Ill + Sme ± (I-S) + Kln [88] for late Cretaceous and Paleogene marine formations, respectively. The Ill + (I-S) ± Sme + Kln claymineral association identified in the Eocene and Oligocene deposits might come from a mixing from Albian-Cenomanian to Paleocene terrains, while the Ill + (I-S) ± Sme + Kln + Chl clay-mineral association identified in the Burdigalian deposits might derive from a mixing from upper Jurassic to Paleocene terrains. The source-area history described suggests a complex erosional evolution since deposits were initially fed by Paleogene and Cretaceous terrains; upper Jurassic to Paleocene terrains, later.

Correlations with Regional Tectonic Phases and Deformation
Several papers have been published in the last two decades in the Betic, Rifian and Tellian External Zones proposing tectonic phases associated to unconformities recognized in the Cenozoic sedimentation [10,21,83,84,[89][90][91][92][93][94]. The papers from Khomsi et al. [83,88] presented structural interpretations on seismic reflection sections, as well as seismic sequential analysis in the Paleogene of the eastern Tunisian Tell. In concrete, these authors mentioned a D5 unconformity (Eocene-Oligocene boundary in age) correlatable with our Unconformity-3. According to these authors this tectonic phase should be associated to a flexuration in the front of the Atlas ("Atlas event"). This  [21]) and Ill + Sme ± (I-S) + Kln [88] for late Cretaceous and Paleogene marine formations, respectively. The Ill + (I-S) ± Sme + Kln clay-mineral association identified in the Eocene and Oligocene deposits might come from a mixing from Albian-Cenomanian to Paleocene terrains, while the Ill + (I-S) ± Sme + Kln + Chl clay-mineral association identified in the Burdigalian deposits might derive from a mixing from upper Jurassic to Paleocene terrains. The source-area history described suggests a complex erosional evolution since deposits were initially fed by Paleogene and Cretaceous terrains; upper Jurassic to Paleocene terrains, later.

Correlations with Regional Tectonic Phases and Deformation
Several papers have been published in the last two decades in the Betic, Rifian and Tellian External Zones proposing tectonic phases associated to unconformities recognized in the Cenozoic sedimentation [10,21,83,84,[89][90][91][92][93][94]. The papers from Khomsi et al. [83,88] presented structural interpretations on seismic reflection sections, as well as seismic sequential analysis in the Paleogene of the eastern Tunisian Tell. In concrete, these authors mentioned a D5 unconformity (Eocene-Oligocene boundary in age) correlatable with our Unconformity-3. According to these authors this tectonic phase should be associated to a flexuration in the front of the Atlas ("Atlas event"). This Eocene-Oligocene boundary was also mentioned by Belayouni et al. [91] also in the Tunissian Tell, by Guerrera et al. [83], Guerrera et al., [84] in the External Betic Cordillera, and Maaté et al.) [21] in the Intrarif of the External Riffian Zone. These last authors also recognized an unconformity Cretaceous-Paleocene boundary in age correlatable with our Unconformity-1.
In the case of the Miocene interval, several works pointed out Miocene unconformities associated to tectonic phases. In the Internal Rif Zone [93,94] by means of a structural analysis study recognized their D2 and D3 compressive phases correlables with our Unconformity-4 and Unconformity-5. Similar Miocene tectonic phases has also been detected in the Tunisian Tell [92], the External Betic Zone [10,84], and the Intrarif of the External Riffian Zone [21]. All of the above suggests a common regional tectonic cadre at western Mediterranean scale probably related to the Africa-Iberia-Europe plate motion.
The source area of the studied succession is correlated with a "recycled orogen" with a trend from Paleogene to Miocene from the quartzose recycled to the transitional recycled. A multicyclic origin derived very probably from the African craton and more in detail of the Atlas Zone is evoked. The few deformational and exhumation data of the Atlas Zone indicate rising from the Triassic onwards [24,95]. In detail, the emersion of the Atlas Zone is supposed at the end of the Cretaceous with a rapid rising in Miocene times due to the thrusting of the High and Middle Atlas over the more external areas [24].

1.
Five unconformities with associate gaps were detected. The Unconformity-1, only recognized in the Ouezzane Unit, has a gap affecting the uppermost Cretaceous to the Lowermost Paleocene. The Unconformity-2, recognized in both tectonic units, is located at the Paleocene-Eocene boundary with a gap affecting the middle Paleocene-lowermost Ypresian. The Unconformity-3, only visible in the Ouezzane Unit, appears at the Eocene-Oligocene boundary with a gap affecting the upper Eocene-lowermost Ruppelian. The Unconformity-4 is also only exposed in the Ouezzane Unit and has a gap marking the time span Chattian p.p.-lowermost Burdigalian. The Unconformity-5 is very reduced and only restricted at the Burdigalian-Langhian boundary in the Ouezzane Unit passing laterally to continuity.

3.
On the basis of lithofacies features, fossils association and minerals distribution some generic reconstruction of the sedimentary realms can be proposed: (i) the upper Cretaceous-lower Paleocene succession was assigned to a deep basin; (ii) the lower Eocene p.p.-upper Eocene p.p. to a deep basin to external carbonate-siliceous platform; (iii) the Oligocene p.p. sequence was deposited on unstable slope with turbidite channels passing upward to an external siliciclastic platform; (iv) the two upper depositional sequences (Burdigalian to Serravallian period) indicate the upward transition from slope to external platform. 4.
The sequences are arranged into transgressive-regressive intervals where the deposits are related to the transgressive interval; meanwhile, the gaps are interpreted as erosive periods due to regressions. The whole Cenozoic sedimentation represents a major regressive cycle. Climate driven sea-level changes are only modest (≈100 m) and do not offer a satisfactory interpretation for alternating erosive and depositional stages spanning millions of years each. Therefore, these must reflect alternating tectonic uplift and subsidence due to a regional tectonics at western Mediterranean scale, probably related to the Africa-Iberia-Europe plate motion.

5.
The petrographic preliminary study allows proposing a source area belonging to rocks deriving from "recycled orogen" (transitional recycled and quartzose recycled subtypes). The clay-mineralogy preliminary study indicates a complex unroofing with erosion of Cenozoic terrains in the whole period accomplished by Cretaceous terrains erosion, first, followed by upper Jurassic erosion later. An Atlas origin is evoked for these terrigenous supplies. 6.
The study area was affected by a prenappe tectonics resulting in blind thrusts and growth folds during the Paleogene after the generalized tectonic inversion (from extension to compression) in the Perimediterranean area. After the Alpine extensional phase, the Paleogene tectonic evolution (after the tectonic inversion to compression) can be correlated with a prenappe compressional tectonics consisting of gentle deformation prevalently resulting in a Mesozoic substrate growing folds and blind thrusts. This evolution predated the Miocene nappe tectonics and it is recognized the whole Central-Western Mediterranean Chains. Both Paleogene and Miocene tectonic phases have been also recognized in other sectors of the Rif, in the Tunisian Tell and in the External Betic Cordillera. 7.
The beginning of the large and growing volumes of reworked terrigeneous supply (turbidites, slumps, olistostromes, etc.) during the latest Oligocene and the Miocene p.p. indicates the starts of the synorogenic sedimentation (foredeep stage of the basins) controlled by active tectonics during the sedimentary processes