Cenozoic denudation in the Marrakech High Atlas, Morocco: insight from apatite fission-track thermochronology

Cenozoic denudation in the Marrakech High Atlas, Morocco:insight from apatite fission-track thermochronology

Yves Missenard,1 Omar Saddiqi,2 Jocelyn Barbarand,3 Pascale Leturmy,1 Geoffrey Ruiz,4

Fatima-Zahra El Haimer2 and Dominique Frizon de Lamotte1

1Univ. Cergy-Pontoise, Departement des Sciences de la Terre (CNRS, UMR 7072), site de Neuville ⁄Oise, F-95 000 Cergy, France;2Departement de Geologie, Faculte des Sciences, Universite Hassan II Aın Chok, Route d�El Jadida, BP5366 Maarif – Casablanca, Maroc;3Departement des Sciences de la Terre (UMR 8148 IDES), Universite de Paris-Sud, Bat. 504, F-91405 Orsay Cedex, France; 4Institut de

Geologie et Hydrogeologie, Rue Emile Argand 11, CP 158, CH-2009, Neuchatel, Switzerland

Introduction

As it results from complex inter-actions between multiple processessuch as tectonic, climate or asteno-spheric flows, the evolution of thetopography in mountain belts andtheir forelands remains a matter ofdebate. In North Africa, the Atlasintraplate belt exhibits peaks at morethan 4000 m in Morocco for a meantopography of about 2000–2500 m. Itis demonstrated that these importantelevations result from the combina-tion of crustal shortening linked to theAfrica-Eurasia convergence with ther-mal doming related to a lithosphericthinning (Seber et al., 1996; Frizon deLamotte et al., 2004; Zeyen et al.,2005; Teixell et al., 2005; Missenardet al., 2006). If there is now a consen-sus on the uplift mechanisms, thechronology still remains a matter ofdebate (e.g. Beauchamp et al., 1999;Frizon de Lamotte et al., 2000; Tesonand Teixell, 2008).The aim of this study was to bring

new constrains on the timing of vertical

movements in the High Atlas belt andits borders, by combining the regionalgeological data with apatite fission-track (AFT) thermochronologyresults. At the moment, very few pub-lished fission-track data exist on theHigh Atlas belt except Barbero et al.(2007).Wewill focus on theMarrakechHigh Atlas (MHA – Fig. 1), becausethis part of the belt presents severaladvantages for fission-track analysiswith important altitude variations andfavourable lithologies. Furthermore,the lithospheric and crustal structuresare well constrained (Missenard et al.,2006, 2007).

Geological setting

The High Atlas belt as a whole is anintracontinental orogen (Mattaueret al., 1977) developed on previousmesozoic rifted basins preceding thespreading of the Central Atlantic.During the Upper Triassic-Lower

Jurassic rifting, the MHA belonged tothe so called West Morrocan Arch(WMA), a persistent topographic highseparating an Atlantic domain to thewest from a Tethyan domain to thenorth-east (Choubert and Faure Mur-et, 1962; Du Dresnay, 1972; Michard,1976; Medina, 1995). To the South,the rifted area is limited by a majordeformation zone probably inheritedfrom the Panafrican orogenesis, the

South Atlas Front (Fig. 1). The riftingended during the Middle Jurassic.After the Upper Jurassic and LowerCretaceous, dominated by the deposi-tion of continental red beds, theCenomano-Turonian transgressionled to the development of a carbonateshelf which covered the whole Atlassystem and large areas in WesternAfrica (see a review in Frizon deLamotte et al., in press).During the Cenozoic, the conver-

gence between Eurasia and Africatriggered the inversion of the riftedzones. Most of the previous publica-tions aiming to describe the geometryof the belt focus on the Central andWestern High Atlas (Fig. 1), wherethe Mesozoic cover is well preserved(Beauchamp et al., 1999; Frizon deLamotte et al., 2000; Benammi et al.,2001; Teixell et al., 2003; Frizon deLamotte et al., 2004). The shorteningvalues obtained from these studiesrange between 10 and 25 km. Acrossthe MHA, characterized by the abun-dance of basement rocks, the totalshortening including the displace-ments along the North Jebilet andAnti-Atlas Major Faults (AAMF)(Fig. 1) is of about 10 km (Missenard,2006). These relatively low values areconsistent with the poorly developedcrustal root under the High Atlas,imaged by seismic surveys and geo-physical modelling (Makris et al.,

ABSTRACT

In North Africa, the High Atlas belt culminates at more than4000 m. In Morocco, recent work shows that a lithosphericthinning explains about 1000 m of the mean topography, theremaining topography being related to crustal shortening. Wecombine regional geology with new apatite fission-track (AFT)ages to constrain the timing of these events in the MarrakechHigh Atlas (MHA). In the inner belt, 10 AFT ages are comprisedbetween 9 ± 1 and 27 ± 3 Ma. These Neogene ages indicatethat the MHA underwent significant denudation during that

time. In the southern foreland domain of the belt, threesamples give scattered AFT ages between 27 ± 2 and 87 ±5 Ma. Geological evidences allow us to constrain the age of amajor denudation event during Middle Miocene age. Wepropose that it is linked to the thermal doming highlighted inthe whole Moroccan Atlas domain.

Terra Nova, 20, 221–228, 2008

Correspondence: Y. Missenard, Departe-

ment des Sciences de la Terre et de l�Envi-ronnement (CNRS,UMR7072), Universite

de Cergy-Pontoise, 95 031 Cergy Pontoise

Cedex, France. Tel.: +33 (0)1 34 25 73 67;

fax: +33 (0)1 34 25 73 50; e-mail: yves.

[email protected]

� 2008 Blackwell Publishing Ltd 221

doi: 10.1111/j.1365-3121.2008.00810.x

1985; Tadili et al., 1986; Wigger et al.,1992; Ramdani, 1998; Fullea Urchu-lutegui et al., 2006). According tothese authors, the Moho depth doesnot exceed 35–39 km. As such valuescannot maintain isostatically a meantopography of about 2500 m, at leastanother process is required to explainthe high topography of the belt (Sch-warz and Wigger, 1988; FulleaUrchulutegui et al., 2006).In 1996, Seber et al. imaged for the

first time a P-wave low velocity zoneunder the High Atlas. Recent workbased on geophysical modelling ofgravity data, geoid, topography andheat flow (following the proceduredeveloped by Zeyen and Fernandez,1994) has confirmed the existence of athinned lithosphere under the Anti-Atlas, the Marrakech and CentralHigh Atlas, and the Middle Atlas(Frizon de Lamotte et al., 2004; Zeyenet al., 2005; Teixell et al., 2005; Fullea

Urchulutegui et al., 2006; Missenardet al., 2006). Even if the origin of thisanomaly is still under debate, it is nowaccepted that an astenospheric processhas to be considered to explain thevery high relief of the Western Mag-hreb. Missenard et al. (2006) evaluatethe effect of the lithospheric thinningon the relief of the affected area andshow that it accounts for 1000 m ofadditional topography in the Anti-Atlas, the Central High Atlas and theMiddle Atlas.In the typical foreland basins, the

sedimentary record mainly reflects thetectonic load and is therefore fre-quently used to infer the timing ofdeformation. In Morocco, the Atlasforeland basins are poorly developedand the Cenozoic sedimentary pilesupposed to be associated with moun-tains building is locally up to 1700 mthick (Toundout area, Gorler et al.,1988) but generally does not exceed

1300 m (Mustaphi, 1997). The Neo-gene series always lie unconformablyon the Palaeogene. Furthermore, thedetermination of the orogenesischronology is complicated by the dif-ficulty to date detrital series. Untilnow, various interpretations havebeen proposed: following Beauchampet al. (1996) and Beauchamp et al.(1999), a first phase took place duringearly Cretaceous, but the major upliftphase occurred between 30 and 20 Ma(Oligo-Miocene) and corresponds tothe Alpine orogenic event. On thecontrary, Frizon de Lamotte et al.(2000), propose a chronology imply-ing two major tectonic events, the firstone being late Eocene, the second onebeing Plio-Quaternary. For Lavilleet al. (1995), the shortening began asearlier as Senonian. Teixell et al.(2005) and Missenard et al. (2006)estimate that the lithospheric thinningemplaced during Miocene on the basis

Fig. 1 Geological map of the Marrakech High Atlas and surrounding areas. Fission-track samples of the Marrakech High Atlasand Siroua plateau are localized. Lower left corner: main geological domain of Morocco and location of the studied area.

Cenozoic denudation in the Marrakech High Atlas • Y. Missenard et al. Terra Nova, Vol 20, No. 3, 221–228

.............................................................................................................................................................

222 � 2008 Blackwell Publishing Ltd

of the age of Moroccan volcanism.The lack of a unified scenario preventsto determine the exact role of thisorogen in the context of the Africa-Eurasia convergence and has to beconstrained with new arguments.

Sampling

The sampling follows a roughlyNorth–South cross-section throughthe MHA (see detailed sample posi-tions Fig. 1). The altitude variationsallow sampling at a wide range ofaltitudes, from 1000 m on the north-ern front to 2600 m at Ifni Lake. Themostly granitic and metamorphic Pre-cambrian to Cambrian basementcrops out widely, whereas in the Cen-tral and Western High Atlas theTriasico-Jurassic carbonates andsandstones dominate.Eight samples were taken in the

Ourika valley, in the northern part ofthe MHA (Fig. 1). This deeply incisedvalley erodes the crystalline basementon which the Upper Triassic �Ouk-aimeden sandstones� lies unconform-ably. Samples are gneisses (1OU,5OU, 13OU, 14OU), volcanodetriticmaterial (8OU), granodiorite (15OU,17OU) and amphibolite (20OU).On the southern flank of the belt,

eight samples were taken along a ver-tical cross-section, but only one gavesignificant apatite content (sample2SA, gneiss). A sample from the IfniLake (sample 1SA, granodiorite) hasbeen analysed and provided anotherage for this part of the belt (Fig. 1).

Last, four samples of 12 analysedfrom the Siroua Plateau revealed validages. This area is situated south of themain limit of the High Atlas belt, theSouth Atlas Front, and correspondsto a topographic high with Precam-brian basement cropping out at morethan 2500 m. On the plateau, theTriassic and Jurassic units are missing.Small remnants of continental LowerCretaceous and Cenomano-Turonianmarine limestones testified that thesurface of the plateau was at or belowsea level during this period. Sample02Si (granite), 01Si (volcanodetriticmaterial) and 03Si (granite) have beentaken respectively at 1, 5 and 15 km ofthe Cretaceous deposits (Fig. 1). Nomajor faults exist between the samplesand these remnants indicating that novertical displacement occurred be-tween them. A voluminous Mio-Plio-cene volcano (11–3 Ma, Berrahmaand Delaloye, 1989), the Siroua, liesunconformably on the basement.

Methodology

Fission-track analysis was performedby O. Saddiqi and Y. Missenard witha Zeiss microscope under dry ·1000magnification. The external detectormethod (Hurford, 1990) was used forthe age determination. Apatite wasmounted in epoxy, polished andetched with 0.8 weight % HNO3

during 45 s at ambient temperaturefor the spontaneous track revelation.The plastic external detectors areetched with boiling blench during

40 min at ambient temperature forrevealing the tracks induced duringthe irradiation under thermal neu-trons flux at the ORPHEE facility(CEA, Saclay, France). Zeta (f) factor(Hurford and Green, 1983) used forthe age calculation is 396 (measure-ment by Yves Missenard) for the ageson the Siroua Plateau and south partof the belt, and 321 (measurement byOmar Saddiqi) for the Ourika valleyages.

Results

The analytical results are presented onTable 1. Three groups of ages are iden-tified (Table 1 and Fig. 2). In the nor-thern part of the belt, all the eight agesare comprised between 17 ± 1 and9 ±1 Ma (Middle to Upper Miocene).On the southern flank, the two samplesgive close values (24 ± 3 and 27 ±3 Ma).South of the South Atlas Front,

ages are scattered between 27 ± 2and 87 ± 5 Ma. Sample 03Si has anAFT age of 63 ± 4 Ma, whereas 04Siis 27 ± 2 Ma in AFT age. Sample04Si has been sampled at <1 km tovolcanic vents and lava flows. Wecannot exclude that this age has beenpartially rejuvenated by volcanic heat-ing, even if the thermal effect ofvolcanic extrusion on the underlyingrocks remains weak (Gallagher et al.,1994), depending of the lava thicknessand temperature (assumed here to bebelow 900 �C considering the acidmaterial). This result will therefore

Table 1 Apatite fission-track analytical results.

Samples

Altitudes

(m) Lithologies n

qs

(·106 t cm)2) Ns

qi

(·106 t cm)2) Ni

qd

(·105 t cm)2) Nd

P(v2)

%

Ages ± 1r(Ma)

1OU 1360 Biotite and

hornblende-bearing gneiss

17 0.354 519 1.57 2298 3.95 15672 66 14.3 ± 0.7

5OU 1230 Gneiss 19 0.188 408 1.27 2747 3.95 15672 62 9.2 ± 0.5

8OU 1100 Volcanodetritic 14 0.138 121 0.7 613 3.95 15672 100 12.5 ± 1.2

130U 1265 Gneiss 17 0.331 486 1.49 2180 3.95 15672 65 14.1 ± 0.7

140U 1560 Eyed-Gneiss 18 0.896 715 3.35 2675 3.95 15672 80 16.9 ± 0.7

15OU 1580 Granodiorite 15 0.253 261 1.21 1253 3.95 15672 95 13.2 ± 0.9

17OU 1480 Granodiorite 14 0.353 436 1.91 2356 3.95 15672 75 11.5 ± 0.6

20OU 1520 Amphibolite 10 0.628 450 2.83 2027 4.08 15672 82 14.5 ± 0.8

1SA 1900 Volcanodetritic 17 0.140 79 0.289 163 4.25 2127 100 24.6 ± 3.4

2SA 2150 Volcanodetritic 18 0.194 174 0.367 329 4.25 2127 75 26.9 ± 2.6

01Si 2100 Volcanodetritic 22 0.845 1005 0.473 563 2.22 16656 38 47.5 ± 2.9

02Si 1340 Granite 19 0.973 973 0.563 563 4.25 2127 60 87.5 ± 5.4

03Si 2130 Granite 20 1.395 848 1.128 686 4.25 2127 46 62.7 ± 3.8

04Si 2150 Granite 55 0.204 242 0.391 464 4.25 2127 79 26.5 ± 2.3

n is the number of grains counted, qs, qd and qi are respectively the track density for the standard, the fossil track density and the induced track density. Ns, Ni and Nd

are the corresponding number of traces. P(v2) is the chi-squared probability. AFT ages provided here are central ages.

Terra Nova, Vol 20, No. 3, 221–228 Y. Missenard et al. • Cenozoic denudation in the Marrakech High Atlas

.............................................................................................................................................................


not be considered in the followingdiscussion. Samples 01Si, 02Si and03Si have been taken, respectively, at13, 10 and 5 km of the volcano.Sample 02Si has been taken in adeeply incised valley that is unlikelyto have been filled by the Pliocene lavaflows, thus excluding any volcanicheating. It is slightly older than 03Si,at 87 ± 5 Ma. Sample 01Si is theyoungest of the Siroua Plateau sam-ples at 47 ± 3 Ma.Low track density for all samples,

combined with the poor content inapatite of the rocks, did not allow usto perform length track analysis.Therefore, ages data have been inter-preted in the light of the geologicalconstraints.

AFT ages discussion

High Atlas domain

In the northern part of the MHA, thesample ages are comprised between 9and 17 Ma, indicating denudationbetween early and middle Miocene.The two ages obtained on samples inthe southern flank of the MHA areslightly older, about 25 Ma. They arelocated at higher altitudes than theother samples but the existence offaults delimitating structural blocks inthe belt prevents from calculating aglobal exhumation rate from all sam-ples. Samples 1OU, 13OU, 14OU and20OU are taken in the same block andallow calculating an exhumation rateof about 80 m Ma)1. The same calcu-lation for samples from the southflank also provides a similar result

about 100 m Ma)1. These rates indi-cate that between 27 and 14 Ma, boththe north and south flanks of theMHA have been slowly eroded atthe same rate. The current altitudediscrepancy is thus more probablyrelated to Plio-Quaternary tectonicactivity or to enhanced climaticerosion on the northern flank.The AFT ages we produced in the

inner MHA are all testifying of asignificant Neogene denudation.These results contrast with the onesof Barbero et al. (2007) in the CentralHigh Atlas where AFT ages all rangesbetween 76 and 270 Ma except fortwo syenites whose Eocene AFT agesdate their emplacement. The MHAtherefore appears to be the solelyregion of the whole Moroccan HighAtlas and surroundings (including Je-bilet massif, El Haimer, 2005 andAnti-Atlas, Malusa et al., 2007) whereCenozoic AFT ages are revealed. Thisparticular region therefore underwentmore denudation in the last 30 Mathan the other parts of the belt. As aconsequence, we have to consider thatthe almost complete absence of Meso-zoic rocks in the MHA is not onlyinherited from the rifting phase (the�West Moroccan Arch� concept, seeabove) but also a consequence of amore important exhumation duringthe Neogene, in agreement with thehigh topography of the region. Thisdifference in denudation between theMHA and the Central High Atlas ismost probably related to significantchanges in the geometry of the belt.Indeed, the Central High Atlas iswider than the MHA (115 and

30 km respectively). As a conse-quence, the shortening is distributedon multiple faults in the Central HighAtlas (see for instance Teixell et al.,2003) whereas deformation is concen-trated on few major faults in theMHA (Missenard et al., 2007). Thus,considering similar shortening valuesfor both domains, expected verticalmovements are more important in theMHA than in the in the Central HighAtlas.As previously discussed, two mech-

anisms can account for the Lower-Middle Miocene erosion phase putforward by our AFT results in theMHA: thermal doming related to thethinning of the lithosphere, or crustalshortening. To determine which ofthese processes account for this ero-sion phase, we analysed the AFT agesof the South foreland in the light ofthe well-constrained geological frame.As the lithospheric thinning affect thewhole Atlas system, its imprint shouldhave been recorded in the foreland.

Siroua Plateau

The outcrops of lower Cretaceous andCenomano-Turonian limestones onthe Siroua Plateau, just above thesampled Precambrian rocks, bringimportant constraints on their thermalhistory: they imply that the sampleswere at the surface at that time. Thus,one must expect fission-track agesolder than 110 Ma. Our fission-trackages from the Siroua Plateau are allcomprised between 87 and 47 Ma,indicating that they have been partlyrejuvenated. These ages are thus not

Cenozoic Triassic Paleozoic Precambrian Main faults

Assarag valley

03Si04Si

8OU 5OU 13OU1OU

14OU16 ±1

20OU14 ±1

15OU13 ±1

17OU

1SA

01Si 02Si

2SA11 ±1

87 ± 5

27 ± 3

24 ± 3

47 ± 3 63 ± 4

12 ± 1 9 ± 1 14 ± 114 ± 1

27 ± 2

0

1000

2000

3000

4000

Ele

vatio

n (m

)

0 10 20 km

Haouz Basin

ToubkalMassif

Siroua PlateauSAF

AAMF

North South

NAF

Fig. 2 Section across the Marrakech High Atlas and the Siroua Plateau. The fission-track ages have been projected. Three groupsof ages can be identified. In the northern part of the belt, ages are comprised between 9 and 17 Ma. On the southern border, 1SAand 2SA samples are 25 Ma old. The ages obtained on the Siroua Plateau are scattered between 47 and 87 Ma.


.............................................................................................................................................................


true cooling ages but only the result ofpartial annealing of the AFT system.So, the Siroua Precambrian rocksforming the top of the plateau musthave been buried significantly after thelate Cretaceous and exhumed duringthe Tertiary up to their present alti-tude. Consequently, they testify of amajor erosion event during late Cre-taceous or Cenozoic. The erodedrocks are therefore necessarily Seno-nian or younger.To the south, the Siroua volcano

lies unconformably on the Precam-brian rocks of the Plateau. Its firstactivity began at 11 Ma (Berrahmaand Delaloye, 1989) and constrainsthe upper boundary for the age of theeroded rocks that must be older thanthis age. Undated detritic rocks aresandwiched between the volcano andthe Precambrian basement (Imzi for-mation, De Beer et al., 2000). De Beeret al. (2000) propose a Miocene agefor this formation. A minimum thick-ness of 2000 m of sediments musthave been reached to explain therejuvenation of the fission-track agesconsidering a 30� per km geothermalgradient and a 60 �C value for thepartial annealing zone upper bound-ary (Gallagher et al., 1998). Thisthickness is reduced to 1000 m ifconsidering a high geothermal gradi-ent of 60 �C km)1 that could havebeen reached during the Mio-Pliocenevolcanic activity. The infill of theadjacent sedimentary basins givessome constraints on the nature andage of the eroded series.In the Ouarzazate basin (Fig. 1),

seismic profiles show thicknesses up to1.5 km for the Upper Cretaceous andPalaeogene units (Beauchamp et al.,1999). The Palaeocene–Eocene is upto 600 m thick (El Harfi et al., 2001).As the top of the Neogene series israrely preserved, this value of 1.5 kmcorresponds to a minimum thickness.In the Souss basin (Fig. 1), the

Upper Cretaceous is 700 m thick(Mustaphi et al., 1997). The Palaeo-cene and Eocene crop out in theeastern part of the basin. Its thicknessis <500 m, but here again, its top israrely preserved and this value is aminimum. The Miocene and Oligo-cene series are poorly developed. Theyonly crop out along the northernborder of the basin, where they arealways <300 m thick (Frizon deLamotte et al., 2000).

The Ouarzazate and Souss basinsand the Siroua Plateau are covered bythe same Cenomano-Turonian marinelimestones testifying of similar lowbathymetry marine palaeoenvironne-ments and resulting from a generaltransgression 200 m above the presentsea level (Haq et al., 1987). Thisindicates that at that time, the Ou-arzazate, Souss and Siroua domainformed a single domain and were atsimilar altitudes.The thicknesses of the Upper Cre-

taceous strata do not exceed 1500 mand is too low to explain the rejuve-nation of the fission-track ages. Itmust therefore be consider that theUpper Cretaceous to Eocene, andmaybe Oligo-Miocene series were alsodeposited on the Siroua Plateau toexplain the rejuvenation. As a conse-quence, the erosional event must beyounger than Eocene.Three main causes, namely tectonic,

thermal and volcanic, contribute tothe uplift of the Siroua Plateau:

1 The Anti-Atlas Major Fault(Fig. 1), situated at the southernborder of the plateau (Fig. 1), is acrustal scale fault zone affecting thewhole Siroua Plateau. The shift ofthe Mio-Pliocene Siroua lava flowstestify of a post-Miocene verticalthrow of about 500 m. The Ceno-mano-Turonian strata, north of theplateau lies at about 2100 m, that is1000 m higher than the same unit inthe Ouarzazate basin. As a conse-quence, 500 m of vertical throwmust have occurred after the depo-sition of the Cenomano-Turonian,but before the lava emplacementduring the late Miocene;

2 The lithospheric thinning identifiedacross the whole Morocco includingAnti-Atlas, Central High Atlas andMiddle Atlas (Fullea Urchuluteguiet al., 2006; Missenard et al., 2006)affects the Siroua plateau and theMHA. The surface uplift is of about1000 m (see Fig. 6B in Missenardet al., 2006);

3 The magmatic activity has also aneffect on the topography of thePlateau. The Siroua strato-volcano,covering more than 500 km2 em-placed from 11 to 3 Ma (Berrahma,1982; Berrahma and Delaloye,1989). To quantify the effect ofcrustal magmatic injections belowthis volcano, we reconstructed the

geometry of the ante volcano (i.e.ante 11 Ma) surface. We digitizedthe lava-basement contact on thebasis of the recent 1 ⁄50 000 geo-logical maps of the Siroua andTachoukart areas (De Beer et al.,2000). We then extrapolated thesedata using a minimum curvaturedigital method (Smith and Wessel,1990) to obtain an envelope map ofthe basement top below the volcano(Fig. 3). The presence of theAAMF in the southern part ofthe plateau has been considered inthe gridding method. North of thefault the basement is affected by anelongated dome (15 · 35 km) fol-lowing a N160� direction. The con-tinuation of the dome to the southof the AAMF cannot be evidenceddue to the small area covered by thelavas, but this doming may explainthe altitude increase of the base-ment from West to East. Along thisdome, the basement reaches2600 m, that is 500 m above itsaltitude along the borders of theplateau. The orientation of thisbulge is inconsistent with the maincompressive trends lying betweenN50�E and N90�E. As the mainvolcanic centres are aligned alongthis dome (Fig. 3), we propose thatits geometry is linked to themagmatic activity and may berelated to magma injections underthe volcano.

This last process only affects thePrecambrian basement directly underthe Siroua volcano and cannot ex-plain alone the uplift and denudationof the whole plateau. The verticalpost-Mio-Pliocene throw is about500 m, as testified by the shift of theSiroua lava flows. The Cenomano-Turonian strata, north of the plateau,actually lies at about 2100 m, that is1000 m higher than the same unit inthe Ouarzazate basin. As a con-sequence, 500 m of additional verticalthrow must have occurred after thedeposition of the Cenomano-Turo-nian, but before the lava emplacementduring Miocene. Therefore, this aloneprocess cannot explain the erosion ofmore than 1500 m of sediments. Con-sequently, the main process at theorigin of the uplift and erosion of theSiroua basin is likely the lithosphericthinning-related doming. As thisevent is a large scale process, it might


.............................................................................................................................................................


also affect the adjacent areas and maybe at the origin of the erosion eventpreceding the so-called Pontianmolasses deposition in the High Atlasforeland domain and in the Anti-Atlas (Choubert and Faure Muret,1962).To summarize, fission-track ages in

the MHA foreland domain indicatethat post-Cenomano-Turonian serieswere deposited at the emplacement ofthe Siroua Plateau. Geological datashow that they were eroded betweenthe Upper Eocene and the Middle-Upper Miocene. Field analysis andgeophysical modelling show that oneof the main processes at the origin ofthis erosion is an uplift associated with

a major lithospheric thinning. The ageof the relief building related to thisthinning is comprised between theUpper Eocene and the Upper Mio-cene. It is worth noting that thisUpper Eocene–Upper Miocene periodis also characterized by the onset ofintense magmatic activity in the Atlassystem, with a highest alkaline volca-nic production during the MiddleMiocene (see review in El Azzouziet al., 1999).

Conclusion

The AFT ages discussed in this studyare the first from the inner MHA.They are also the first fission-track

ages in the Moroccan High Atlastestifying of significant Cenozoicdenudation.The analysis of these new AFT ages

in the light of the geological con-straints shows that the Siroua Plateaumust have been subsiding duringUpper Cretaceous to Eocene times astestify by the partial reset of the AFTages. As this plateau is currentlymostly formed of Precambrian base-ment, an erosion phase occurred dur-ing Lower or Middle Miocene in thisforeland domain. This event is alsorecorded in the High Atlas belt, withAFT ages ranging between 9 ± 1 and27 ± 3 Ma.The review of mechanisms inducing

uplift in the High Atlas forelandshows that this large scale erosionevent is probably related to the thin-ning of the lithosphere under the Atlasdomain, synchronously with the in-tense magmatic activity in this area(El Azzouzi et al., 1999).

Acknowledgements

This research was funded by the CNRS-INSU Relief programme and by the �Ac-tion Integree� Volubilis, ProjectMA ⁄05 ⁄125. We are grateful to C. Dogli-oni (Ed.) for his comments on this manu-script. We thank J. Van den Driessche andA. Teixell for their careful reviews thatstrongly enhanced the text. We are alsograteful to A. Michard for his help andfruitful discussions.

References

Barbero, L., Teixell, A., Arboleya, M.-L.et al., 2007. Jurassic-to-present thermalhistory of the central High Atlas (Mor-occo) assessed by low-temperature ther-mochronology. Terra Nova, 19, 58–64.Doi: 10.1111/j.1365-3121.2006.00715.x.

Beauchamp, W., Barazangi, M., Demnati,A. et al., 1996. Rifting and Inversion:Missour basin and Atlas Mountains,Morocco. AAPG Bulletin, v. 80, p. 1459–1482.

Beauchamp, W., Allmendinger, R.W.,Barazangi, M. et al., 1999. Inversiontectonics and the evolution of the HighAtlas Mountains, Morocco, based on ageological-geophysical transect. Tecton-ics, 18, 163–184.

Benammi, M., Arbi Toto, E. and Chakiri,S., 2001. Les chevauchements frontauxdu Haut Atlas central marocain: stylesstructuraux et taux de raccourcissementdifferentiel entre les versants nord etsud: on the differential structuralstyles and rates of shortening on the

Fig. 3 Reconstruction of the basement topography under the Siroua volcanic massif;see localization Fig. 1. Black crosses correspond to basement ⁄ lavas contact digitizedfrom the 1 ⁄50 000 geological maps (De Beer et al., 2000) to reconstruct the ante-volcano surface. Their altitudes have then been extrapolated using minimumcurvature. The Anti-Atlas Major Fault is included and acts as a barrier betweenthe northern and southern blocks, which appear shifted of about 500 m, confirmingthe throw observed on the field. In the northern block, the basement forms a 500-m-high dome over the plateau borders, lying at about 2100 m. The 2300 m and 2400 misolines are drawn in bold to underline the elongated geometry of this dome, which islocalized exactly under the main eruptive centers (black circles).


.............................................................................................................................................................


northern and southern bordering thrustsof the Moroccan Central Atlas. C. R.Acad. Sci. IIA Earth Planet. Sci., 333,241–247.

Berrahma, M., 1982. Volcanisme Mio-Pliocene de la partie Nord Ouest duMassif du Siroua (Anti-Atlas Central :Maroc); Etude structurale et petrologi-que. Unpublished PhD thesis, Paris XI,Orsay, France, 183 p.

Berrahma, M. and Delaloye, M., 1989.Donnees geochronologiques nouvellessur le massif volcanique du Siroua(Anti-Atlas, Maroc). J. Afr. Earth Sci.,9, 651–656.

Choubert, G. and Faure Muret, A., 1962.Evolution du domaine atlasique marocaindepuis les temps paleozoıques, Livre a lamemoire du Professeur Paul Fallot.Societe Geologique de France, Paris,pp. 447–527.

De Beer, C.H., Chevalier, L.P., De Kocks,G.S. et al., 2000. Memoire explicatif dela carte geologique du Maroc au 1 ⁄50000, Feuille Sirwa. Notes Mem. Serv.Geol. Maroc, 395 bis, 86 p.

Du Dresnay, R., 1972. Les phenomenes debordure des constructions carbonateesdu Lias moyen du Haut Atlas oriental(Maroc). C. R. Acad. Sci. Paris, 275,535–537.

El Azzouzi, M., Bernard-Griffiths, J.,Bellon, H. et al., 1999. Evolution of thesources of Moroccan volcanism duringthe Neogene. C. R. Acad. Sci. IIA, 329,95–102.

El Haimer, F.Z., 2005. Thermochronologiepar traces de fission sur apatite dans ledomaine mesetien (Jebilet) : implicationspour l�evolution du relief du Marococcidental. Unpublished, Memoire deDESA de l�Universite Hassan II –Mohammedia, Departement de Geolo-gie, Casablanca.

El Harfi, A., Lang, J., Salomon, J. et al.,2001. Cenozoic sedimentary dynamics ofthe Ouarzazate foreland basin (CentralHigh Atlas Mountains, Morocco). Int.J. Earth Sci., 90, 393–411. doi: 10.1007/s005310000115.

Frizon de Lamotte, D., Saint Bezar, B.,Bracene, R. and Mercier, E., 2000. Twostep Atlas building and geodynamics ofthe West Mediterranean. Tectonics, 19,740–761.

Frizon de Lamotte, D., Crespo-Blanc, A.,Saint Bezar, B. et al., 2004. TRASNS-MED-transect I [Betics, Alboran Sea,Rif, Moroccan Meseta, High Atlas, JbelSaghro, Tindouf basin]. In: TheTRANSMED Atlas – the Mediterraneanregion from Crust to Mantle (W. Cavaz-za, F. Roure, W. Spakman, G.M.Stampfli and P.A. Ziegler, eds), pp. 91–96. Springer, Berlin, Heildelberg.

Frizon de Lamotte, D., Zizi, M., Misse-nard, Y. et al., 2008 (in press). The Atlas

system. In: Continental Evolution: TheGeology of Morocco. Structure, Stratig-raphy, and Tectonics of the Africa-Atlantic-Mediterranean Triple Junction(A. Michard, O. Saddiqi, A. Chalouanand D. Frizon de Lamotte, eds),Springer, Berlin, Heildelberg.

Fullea Urchulutegui, J., Fernandez, M.and Zeyen, H., 2006. Lithosphericstructure in the Atlantic-Mediterraneantransition zone (southern Spain, north-ern Morocco): a simple approach fromregional elevation and geoid data.C. R. Geosci., 338, 140–151.

Gallagher, K., Hawkesworth, C.J. andMantovani, M.S.M., 1994. The denuda-tion history of the onshore continentalmargin of SE Brazil inferred fromapatite fission track data. J. Geophys.Res., 99, 18117–18145.

Gallagher, K., Brown, R. and Johnson, C.,1998. Fission track analysis and itsapplications to geological problems.Annu. Rev. Earth Planet. Sci., 26,519–572.

Gorler, K., Helmdach, F.F., Gaemers, P.et al., 1988. The uplift of the CentralHigh Atlas as deduced from Neogenecontinental sediments of the Ouarzazateprovince, Morocco. In: The Atlas Systemof Morocco (V. Jacobshagen, ed.),pp. 361–404. Springer, Berlin.

Haq, U.J., Hardenbol, J. and Vail, P.R.,1987. Chronology of fluctuating sealevels since the Triassic. Science, 235,1156–1167.

Hurford, A.J., 1990. Standardization offission track dating calibration: recom-mendation by the Fission Track Work-ing Group of the IUGS. Chem. Geol., 80,171–178.

Hurford, A.J. and Green, P.F., 1983. Thezeta calibration of fission track dating.Chem. Geol., 1, 285–317.

Laville, E., Charroud, A., Fedan, B. et al.,1995. Inversion negative et rifting atlas-ique: l�exemple du bassin triasique deKerrouchene, Maroc. Bull. Soc. Geol.Fr., 166, 364–374.

Makris, J., Demnati, A. and Klussmann,J., 1985. Deep seismic soundings inMorocco and a crust and upper mantlemodel deduced from seismic and gravitydata. Ann. Geophys., 3, 369–380.

Malusa, M.G., Polino, R., Feroni, A.C.,Ellero, A., Ottria, G., Baidder, L. andMusumeci, G., 2007. Post-Variscantectonics in eastern Anti-Atlas (Mor-occo). Terra Nova 19, 481–489. doi:10.1111/j.1365-3121.2007.00775.x.

Mattauer, M., Tapponier, P. and Proust,F., 1977. Sur les mecanismes de forma-tion des chaınes intracontinentales.L�exemple des chaınes atlasiques duMaroc. Bull. Soc. Geol. Fr., 7, 521–526.

Medina, F., 1995. Syn- and postrift evolu-tion of the El Jadida-Agadir basin

(Morocco): constraints for the riftingmodel of the central Atlantic. Can. J.Earth Sci., 32, 1273–1291.

Michard, A., 1976. Elements de geologiemarocaine. Notes Mem. Serv. Geol.Maroc, 252, 408 p.

Missenard, Y., 2006. Le relief des AtlasMarocains : contribution des processusasthenospheriques et du raccourcissementcrustal, aspects chronologiques. Unpub-lished PhD thesis, Universite de CergyPontoise, France, 236 pp.

Missenard, Y., Zeyen, H., Frizon deLamotte, D. et al., 2006. Crustal versusAsthenospheric Origin of the Reliefof the Atlas Mountains of Morocco.J. Geophys. Res, v. 111; B03401.doi: 10.1029/2005JB003708.

Missenard, Y., Taki, Z., Frizon de Lamo-tte, D. et al., 2007. Tectonic styles inthe Marrakesh High Atlas (Morocco):the role of heritage and mechanicalstratigraphy. J. Afr. Earth Sci., 48,247–266.

Mustaphi, H., 1997. Apport des donnees desubsurface (forage et profils sismiques) etde la modelisation a l�etude de l�evolutiondu bassin du Souss du Trias a l�actuel.Unpublished PhD thesis, UniversiteMohammed V, Rabat, Morocco.

Mustaphi, H., Medina, F., Jabour, H.et al., 1997. Le bassin du Souss (Zone deFaille du Tizi n�Test, Haut Atlas occi-dental, Maroc): resultat d�une inversiontectonique contrlee par une faille dedetachement profonde. J. Afr. EarthSci., 24, 153–168.

Ramdani, F., 1998. Geodynamic implica-tions of intermediate-depth earthquakesand volcanism in the intraplate Atlasmountains (Morocco). Phys. EarthPlanet. In., 108, 245–260.

Schwarz, G. and Wigger, P.J., 1988. Geo-physical studies of the Earth�s crust andupper mantle in the Atlas System ofMorocco. In: The Atlas System of Mor-occo (V. Jacobshagen, ed.), pp. 339–357.Springer, Berlin.

Seber, D., Barazangi, M., Tadili, B. et al.,1996. Three-dimensional upper mantlestructure beneath the intraplate Atlas andinterplate Rif mountains of Morocco.J. Geophys. Res, 101, 3125–3138.

Smith, W.H.F. and Wessel, P., 1990.Gridding with continuous curvaturesplines in tension. Geophysics, 55,293–305.

Tadili, B., Ramdani, M., Ben Sari, D.et al., 1986. Structure de la croute dansle Nord du Maroc. Ann. Geophys., 4,99–104.

Teixell, A., Arboleya, M.-L., Julivert, M.et al., 2003. Tectonic shortening andtopography of the central High Atlas(Morocco). Tectonics, 22, 1051.

Teixell, A., Ayarza, P., Zeyen, H. et al.,2005. Effects of mantle upwelling in a


.............................................................................................................................................................


compressional setting: the Atlas Moun-tains of Morocco. Terra Nova, 17,456–461.

Teson, E. and Teixell, A., 2008. Sequenceof thrusting and syntectonic sedimenta-tion in the eastern Sub-Atlas thrust belt,(Dades and Mgoun valleys, Morocco).International Journal Earth Sciences(Geol. Rundschau), v. 97-1, Doi:10.1007/s00531-006-0151-1.

Wigger, P., Asch, G., Giese, P. et al.,1992. Crustal structure along a traverse

across the Middle and High Atlasmountains derived from seismicrefraction studies. Geol. Rundsch., 81,237–248.

Zeyen, H. and Fernandez, M., 1994. Inte-grated lithospheric modeling combiningthermal, gravity, and local isostasyanalysis: application to the NE SpanishGeotransect. J. Geophys. Res., 99,18089–18102.

Zeyen, H., Ayarza, P., Fernandez, M.et al., 2005. Lithospheric structure under

the western African-European plateboundary: A transect across the AtlasMountains and the Gulf of Cadiz.Tectonics, 24, doi: 10.1029/2004TC001639.

Received 20 November 2007; revised versionaccepted 1 April 2008


.............................................................................................................................................................


Cenozoic denudation in the Marrakech High Atlas, Morocco: insight from apatite fission-track thermochronology

Documents