Updating the marine biostratigraphy of the Granada Basin (central Betic Cordillera). Insight for the Late Miocene palaeogeographic evolution of the Atlantic – Mediterranean seaway

Geobios 45 (2012) 249–263

Original article

Updating the marine biostratigraphy of the Granada Basin (central BeticCordillera). Insight for the Late Miocene palaeogeographic evolutionof the Atlantic – Mediterranean seaway§

Hugo Corbı a, Carlos Lancis a, Fernando Garcıa-Garcıa b, Jose-Antonio Pina a, Jesus M. Soria a,*,Jose E. Tent-Manclus a, Cesar Viseras c

a Departamento de Ciencias de la Tierra y del Medio Ambiente, Universidad de Alicante, Apdo Correos 99, 03080 Alicante, Spainb Departamento de Geologıa, Universidad de Jaen, Campus de Las Lagunillas s/n, 3071 Jaen, Spainc Departamento de Estratigrafıa y Paleontologıa, Universidad de Granada, Avda. Fuentenueva s/n, 18002 Granada, Spain

A R T I C L E I N F O

Article history:

Received 19 November 2010

Accepted 7 October 2011

Available online 19 May 2012

Keywords:

Biostratigraphy

Planktonic foraminifera

Calcareous nannoplankton

Evaporites

Granada Basin

Late Miocene

A B S T R A C T

The marine stratigraphic record of the Granada Basin (central Betic Cordillera, Spain) is composed of

three Late Miocene genetic units deposited in different sea-level contexts (from base to top): Unit I (sea-

level rise), Unit II (high sea-level), and Unit III (low sea-level). The latter mainly consists of evaporites

precipitated in a shallow-basin setting. Biostratigraphic analyses based on planktonic foraminifera and

calcareous nannoplankton indicate four late Tortonian bioevents (PF1-CN1, PF2, PF3, and PF4), which can

be correlated with astronomically-dated events in other sections of the Mediterranean. PF1-CN1 (7.89

Ma) is characterized by the influx of the Globorotalia conomiozea group (including typical forms of

Globorotalia mediterranea) and by the first common occurrence of Discoaster surculus; PF2 (7.84 Ma) is

marked by the first common occurrence of Globorotalia suterae; PF3 (7.69 Ma) is typified by the influx of

dextral Neogloboquadrina acostaensis; and PF4 (7.37 Ma) is defined by the influx of the Globorotalia

menardii group II (dextral forms). The PF1 event occurred in the upper part of Unit I, whereas PF2 to PF4

events occurred successively within Unit II. The age of Unit III (evaporites) can only be estimated in its

lower part based on the presence of dextral Globorotalia scitula, which, together with the absence of the

first common occurrence of the G. conomiozea group (7.24 Ma), points to the latest Tortonian.

Comparisons with data from the other Betic basins indicate that the evaporitic phase of the Granada

Basin (7.37–7.24 Ma) is not synchronous with those from the Lorca Basin (7.80 Ma) and the Fortuna Basin

(7.6 Ma). In the Bajo Segura Basin (easternmost Betic Cordillera), no evaporite deposition occurred

during the late Tortonian. The evaporitic unit of the Granada Basin (central Betics) records the late

Tortonian restriction of the Betic seaway (the marine connection between the Atlantic and

Mediterranean). The diachrony in the restriction of the Betic seaway is related to differing tectonic

movements in the central and eastern sectors of the Betic Cordillera.

� 2012 Published by Elsevier Masson SAS.

Available online at

www.sciencedirect.com

1. Introduction

Recognition of Milankovitch cycles in sedimentary sequenceshas notably refined the time scale for the Neogene period. Studiescombining orbitally-controlled cyclicity and magnetobiostratigra-phy led to the development of the modern Astronomically TunedNeogene Time Scale (ATNTS; Lourens et al., 2004). The mostsignificant calcareous plankton bioevents are accurately calibratedin astronomical age based on this scale. The bioevents related to

§ Corresponding editor: Frederic Quillevere.

* Corresponding author.

E-mail address: [email protected] (J.M. Soria).

0016-6995/$ – see front matter � 2012 Published by Elsevier Masson SAS.

http://dx.doi.org/10.1016/j.geobios.2011.10.006

the planktonic foraminifer Globorotalia conomiozea group are ofparticular interest. Its first regular occurrence correlates with theTortonian/Messinian boundary at 7.24 Ma; however, this grouphas been recorded at several points of the Mediterranean 650 kabefore this major bioevent (Hilgen et al., 1995; Krijgsman et al.,2000; Husing et al., 2009), defining a marked influx at 7.89 Ma (lateTortonian). In this paper, and for the first time, the G. conomiozea

group influx is described for the Granada Basin, in addition to otherplanktonic foraminiferal and calcareous nannoplankton bioeventsthat have been astronomically tuned in the Mediterranean as well(first common occurrence of Discoaster surculus at 7.88 Ma, by Raffiet al., 2006; first occurrence of Globorotalia suterae at 7.84 Ma, bySprovieri et al., 1999; first influx of dextral Neogloboquadrina

acostaensis at 7.697 Ma, by Krijgsman et al., 2000; first influx of

H. Corbı et al. / Geobios 45 (2012) 249–263250

Globorotalia menardii group II at 7.355 Ma, by Hilgen et al., 1995).Based on these new data, the goal of our study is to:

� establish the timing of the main palaeogeographic events thatoccurred in the Granada Basin, especially the evaporitic phasemarking the marine restriction and the subsequent transforma-tion to a continental basin;� to assess the importance of these bioevents for the stratigraphic

correlation at a regional scale. The data presented here are ofimportance for understanding the palaeogeographic evolution ofthe marine connections between the Mediterranean and theAtlantic Ocean (Betic and Rif seaways) before the onset of theMessinian Salinity Crisis (5.96 Ma, according to Krijgsman et al.,1999). It appears that, during the latest Tortonian, the Beticseaway was partially closed, undergoing evaporitic precipitation(Tortonian Salinity Crisis; Krijgsman et al., 2000). The isochronyof this regional palaeogeographic event is discussed in light ofnew biostratigraphic data presented here.

2. Geological evolution and general stratigraphy of the GranadaBasin

The intramontane Granada Basin is located in the central sectorof the Betic Cordillera (Fig. 1). The basin’s sedimentary infillunconformably overlies the two main domains of the Cordillera:the allochthonous complexes of the Internal Zones (Palaeozoic toTriassic metamorphic rocks) and the parautochthonous units ofthe External Zones (Mesozoic carbonate sedimentary rocks). Bothdomains crops out to the south and north of the basin, respectively(Fig. 2). The Granada Basin formed during the late orogenic phaseof the Betic Cordillera (Rodrıguez Fernandez and Sanz de Galdeano,2006), after the westward drift of the Internal Zones and theircollision against the External Zones. From the late Tortonian to theQuaternary, the Granada Basin became a typical extensional basindelimited by high-angle faults (Rodrıguez Fernandez and Sanz deGaldeano, 2006).

The sedimentary record of the Granada Basin is traditionallysubdivided into two main groups: a lower marine one, late

Fig. 1. General geological map of the Betic Cordillera with the location of the Granada

references to colours, see the web version of this article.

Tortonian in age, and an upper continental one, deposited from theTurolian to the Quaternary (Fernandez et al., 1996). The lowergroup comprises three genetic units related to sea-level changes(from bottom to top: Units I, II, and III). The upper assemblage isformed of three continental units which, from bottom to top, are:Unit IV (Turolian), Unit V (latest Turolian-Pliocene), and Unit VI(Pliocene-Quaternary). These three units contain several small-mammal localities that have enabled the recognition of the MN12to MN15 biozones (Garcıa-Alix et al., 2008). In this study, we focuson the Tortonian marine units (Fig. 3).

Unit I overlies the metamorphic basement of the Internal Zones.Locally, beneath Unit I, there are gypsum-bearing silts and sandsassigned to the Serravallian-earliest Tortonian (Gonzalez Donoso,1977–1978; Rodrıguez Fernandez, 1982). This unit is formed oftwo gradually superimposed lithological intervals. The lower(Subunit Ia) is composed of nearshore bryozoan-rich siliciclasticsands, gradually passing upward to shelf skeletal sandstones. Theupper (Subunit Ib) is represented by marls rich in calcareousplankton. According to Fernandez and Rodrıguez Fernandez (1989)and Fernandez et al. (1996), Unit I records both the beginning ofmarine sedimentation in the Granada Basin during the Tortonianas well as a progressive deepening in a sea-level rise context.

Unit II consists of two correlative lithological assemblages: amarginal, coarse-grained one, attached to the extensional faultslimiting the Granada Basin and the relief of the Sierra Nevada; andanother dominated by marls in the basin centre. The marginalassemblage, previously interpreted as a marine delta complex (e.g.,Garcıa-Garcıa et al., 1999), is represented by matrix-supportedbreccias and conglomerates that cyclically alternate with plank-ton-rich silts. The basinal marls are characterized by an abundantand diversified association of planktonic and benthic microfossils.Overall, Unit II has been interpreted in a high sea-level context(Fernandez et al., 1996), recording the late Tortonian maximummarine extension in the Granada Basin.

Evaporitic sedimentation occurs within Unit III in the GranadaBasin (Rouchy, 1976; Rouchy and Pierre, 1979; Dabrio et al., 1982;Rodrıguez Fernandez, 1982). Lithologically, this unit is representedby a well-stratified succession in which gypsum layers alternate

Basin and other Neogene-Quaternary basins cited in the text. For interpretation of

Fig. 2. Simplified map of the Granada Basin showing the location of the three stratigraphic sections studied (La Malaha, Agron, and Genil River). For interpretation of

references to colours, see the web version of this article.

H. Corbı et al. / Geobios 45 (2012) 249–263 251

with laminated marls. Locally, this succession is capped bystrontium-mineralized stromatolites (Martın et al., 1984). UnitIII overlies only the marls of Unit II in the basin centre. For thesedimentary environment, we agree with the marine originsuggested by Martın et al. (1984), who proposed a shallow-waterand shallow-basin evaporitic model (sensu Hardie and Eugster,1971; Kendall, 1979). The micropalaeontological analysis wecarried out in the laminated marl confirmed its marine origin.These marls contain a varied and well preserved, though dwarffaunal assemblage consisting of planktonic and benthic foraminif-era. Unit III is the record of a restriction stage in the evolution of theGranada Basin (Martın et al., 1984), characterized by a relative sea-level fall (Fernandez et al., 1996).

The only data available on the planktonic foraminiferalbiostratigraphy of the Granada Basin are those from Dabrioet al. (1972) and Gonzalez Donoso (1977–1978). Dabrio et al.(1972) focused their study on the marine sediments beneath theevaporites, for which a late Tortonian to early Messinian age wasproposed. The work by Gonzalez Donoso (1977–1978) presentedtwo particularly relevant biostratigraphical contributions. The firstwas the recognition of N. acostaensis and Globigerinoides extremus

within the skeletal sandstones making up the lower terms of themarine fill (Unit I), indicating an early Tortonian age (following the

zonal scheme of D’Onofrio et al., 1975). The second contributionwas the identification of G. suterae in the stratigraphically highestmaterials of Unit II, indicating a late (not end) Tortonian age asmost likely. This latter dating is truly important for the chronologyof events recorded in the Granada Basin since it first suggested thatthe beginning of the evaporitic sedimentation was older in theGranada Basin than elsewhere in the Mediterranean.

The biostratigraphic data of Gonzalez Donoso (1977–1978)have been used in numerous works to assume that the marinestratigraphic record of the Granada Basin extends chronologicallyfrom the early to the late Tortonian, whereas doubt remainsconcerning the existence of an early Messinian marine sedimen-tation. Re-examination of the temporal significance of these twobiostratigraphic markers following recent zonal schemes (Spro-vieri et al., 1999; Iaccarino et al., 2007) has enabled updating of thebasin marine infill. Occurrences of N. acostaensis and G. extremus inUnit I (skeletal sandstones) fits within biozone MMi12a (lateTortonian, from 8.35 to 7.84 Ma). On the other hand, thedetermination of G. suterae at the top of Unit II would indicatethe uppermost zone of the Tortonian (MMi12b, from 7.84 to 7.24Ma). The first common occurrence (FCO) of the G. conomiozea

group is not documented in the stratigraphic record of the basin,therefore excluding a Messinian age for the marine infill.

Fig. 3. Stratigraphy of the Upper Miocene units of the Granada Basin, indicating the two main planktonic foraminifer bioevents (influx of the G. conomiozea group and FO of

G. suterae) used for the correlation among the three sections studied. Additional data of small-mammal localities and continental biozones are included. For interpretation of

references to colours, see the web version of this article.

H. Corbı et al. / Geobios 45 (2012) 249–263252

3. Material and methods

Our biostratigraphical study focuses on the La Malaha section,in the centre of the Granada Basin (Fig. 2). In this section, the top ofUnit I (Subunit Ib) and Units II and III crops out. Two additionalsections (Genil River and Agron; Figs. 2 and 3) were also consideredin order to correlate the main bioevents throughout the GranadaBasin. The Genil River section lies on the eastern basin margin,

where Unit II shows coarse-grained deltaic facies. The Agronsection is located on the southern basin margin, where evaporitesfrom Unit III reach their maximum thickness.

The marly sediments of the La Malaha, Genil River, and Agronsections have been collected with a sample spacing of less than1 m. For planktonic foraminiferal analyses (Appendix A), the125 mm sieve fraction of washed residues were used to determinethe presence and relative abundance of Upper Miocene marker

H. Corbı et al. / Geobios 45 (2012) 249–263 253

species (G. conomiozea group and G. suterae). Coiling directions inN. acostaensis were also determined. Preservation is fairly good forthe Upper Miocene planktonic foraminifera; no abrasion ordissolution of the tests is visible, discarding reworking ordownslope transport processes during faunal accumulation. Weconsidered the most recent astronomically calibrated biozonationcharts, such as those proposed by Sprovieri et al. (1999), Lourenset al. (2004), Iaccarino et al. (2007), and Husing et al. (2009). The LaMalaha section has also been analysed for calcareous nannoplank-ton (Appendix B). The smear slides were processed following themethod proposed by Lancis (1998), which, by a repeated procedureof centrifugation-sonification, increases the nannofossil/silt ratio.The marker species shown in the zonal scheme by Okada and Bukry(1980) were determined. Calibration data are from Raffi et al.(2006).

Concerning the planktonic foraminiferal biostratigraphy, theG. conomiozea group (Zachariasse, 1979 in Krijgsman et al., 1995),also known as the Globorotalia miotumida group (Sierro et al.,1993), is crucial for the Late Miocene zonal schemes. The FirstCommon Occurrence (FCO) of this group correlates with theTortonian/Messinian boundary (Lourens et al., 2004). Modernastronomical calibrations indicate an age of 7.24 Ma for thisboundary. The FCO of the G. conomiozea group is preceded by aninflux of the same group during the late Tortonian. This influx iscalibrated in the sections of Metochia (Gavdos island, Greece),Kastelli (Crete), and Gibliscemi (Sicily) at 7.89 Ma (Hilgen et al.,1995). More recently, the same influx has been documented in theLorca Basin (eastern Betics) by Krijgsman et al. (2000) and atMonte dei Corbi (northern Apennines) by Husing et al. (2009). Abioevent that immediately followed the G. conomiozea groupinflux is the First Occurrence (FO) of G. suterae, which has beenastronomically dated to 7.84 Ma in the Gibliscemi and Falconarasections in Sicily (Sprovieri et al., 1999). The frequency ofoccurrence of G. suterae in the late Tortonian sediments fromthe Betic basins (Gonzalez Donoso, 1977–1978; Serrano, 1979;Sierro, 1984; Corbı, 2010; among others) makes it useful inregional-scale stratigraphic correlations. This species has beenconsidered a zonal marker in the Mediterranean by Iaccarino et al.(2007).

4. Studied sections and biostratigraphy

4.1. La Malaha section

4.1.1. Lithostratigraphy

In this section, the top of Unit I (Subunit Ib) and Units II (pre-evaporitic) and III (evaporitic) are exceptionally well exposed(Fig. 4). Only the topmost 15 m of Unit I are represented, arrangedas a fining-upward succession with breccias at the base and greysandy silt in the upper part. These silts gradually pass upwards toUnit II (Fig. 5), represented by 25 m of yellow marls. In both units,planktonic and benthic foraminifera, as well as calcareousnannoplankton, are abundant.

The evaporites forming Unit III conformably overlie Unit II.Within these evaporites, a lower interval of gypsiferous marls (7m-thick) is followed by an upper gypsum interval (40 m-thick;Fig. 6). This gypsum interval contains 13 sedimentary packagesaveraging 5 m in thickness, each defined by two facies associa-tions: a gypsum laminite/marl fine alternation (Fig. 6A–C) and amicrocrystalline nodular gypsum with chickenwire texture(Fig. 6E). Two additional features are noteworthy within thenodular gypsum strata. One is the presence of irregular reddenedsurfaces, marked by the accumulation of iron oxides (Fig. 6D). Theother is the appearance, at the top of some beds, of laminatedcarbonates with stromatolite-like undulations (Fig. 6F). In short,

the persistent alternation of beds of gypsum laminite/marl, andnodular gypsum, which jointly form the gypsum interval of Unit III,is interpreted as the result of oscillations of the water table in thecontext of a shallow-water evaporitic basin. Adopting the model ofKendall (1979), the gypsum laminite and marl representssubaqueous deposition, whereas nodular gypsum records a coastal(supratidal) sabkha sedimentation, where episodic pulses ofemersion generate irregular iron-coated surfaces. The upwarddecrease in marine microfossil content recorded in the marl layerspoints towards a progressive restriction in the shallow-marineevaporitic basin.

The top of Unit III is an irregular erosion and/or karstificationsurface underlying Unit IV. This latter unit is represented by asuccession more than 70 m-thick composed of a regular alterna-tion of silts and sandstones, with subordinate laminated gypsumlevels. Near the top of the unit is a noticeable red interval (6 m-thick), without sandstones, made up of silts and gypsum. Thesedimentological features of this succession were illustrated byDabrio et al. (1972), who, on the basis of the presence of Boumasequences in the sandstone layers, proposed a model of shallow-water turbidites in a basin isolated from the sea. The washedresidues from the silt layers indicate a continental origin, inagreement with the lacustrine origin assumed in other works(Dabrio et al., 1982). Actually, all the samples analysed areexclusively siliciclastic (quartz and mica), without microfossils.

4.1.2. Planktonic foraminifera

Units I and II have a varied association of planktonicforaminifera, represented by the genera Catapsydrax, Dentoglobi-

gerina, Globigerina, Globigerinella, Globigerinita, Globigerinoides,Globorotalia, Globoturborotalita, Neogloboquadrina, Orbulina, Tur-

borotalita, and Sphaeroidinellopsis (adopting the taxonomiccriteria of Iaccarino et al., 2007). With respect to the markerspecies, sinistral Neogloquadrina acostaensis is present and/orabundant in the 59 samples from Units I and II; only in 13 samples,dextral forms of this species (less than 10%) coexist with sinistralforms (Fig. S1). G. menardii group I (sensu Sierro et al., 1993), whichincludes sinistral-coiling forms of Globorotalia merotumida,Globorotalia plesiotumida, and G. menardii form 4 (Tjalsma,1971), appears regularly from the bottom to the top of thesuccession. Another species commonly used as a marker isG. extremus, which has been recognized in some intervals(samples 7 to 58; Fig. S1). Subunit Ib (samples 1 to 4) showstypical specimens of the G. conomiozea group (Globorotalia

mediterranea and G. miotumida). From sample 4 upwards,Sphaeroidinellopsis seminulina is sporadically found. G. suterae

appears regularly in the transition zone between Units Ib and II(from sample 18 on; Fig. S1). In sample 24 (Unit II), an influx ofdominant dextral Neogoloquadrina acostaensis has been detected.In samples 41 to 44, in a short stratigraphic interval from themiddle-upper part of Unit II, specimens of G. menardii group II(sensu Sierro et al., 1993) dominate, represented by dextral-coiling forms of Globorotalia merotumida, G. plesiotumida, andG. menardii form 5 (Tjalsma, 1971).

In Unit III, the first sample of the gypsiferous marls (59) has ahigh planktonic content, with diversified foraminifera of normalsize (Globigerina, Globigerinella, Globigerinoides, Globorotalia, Glo-

boturborotalita, Neogloboquadrina, Orbulina, and Turborotalita). Thesecond sample (60) marks a significant change with respect to theprevious one; planktonic foraminifera are very scarce, small(dwarf), and have very low diversity (Globigerina, Globigerinita, andTurborotalita). In the laminated marl interbeds of the uppergypsum interval, the microfossil record, though well preserved, isdwarf, scarce, and variable from one level to another. The generaltrend is a gradual impoverishment of the microfauna from thebottom to the top of this gypsum interval. The lowest levels contain

Fig. 4. Panoramic views of the La Malaha section. Note the conformable boundary between Unit II (pre-evaporitic) and Unit III (evaporitic), and the irregular contact,

interpreted as a palaeokarst, at the top of Unit III. Overlying this irregular surface appear the first continental deposits (Unit IV). Both outcrops are used to construct the

stratigraphic sections illustrated in Figs. 5 and 6.

H. Corbı et al. / Geobios 45 (2012) 249–263254

up to five genera of planktonic foraminifera (predominantlyGlobigerina, in addition to Globigerinoides, Globorotalia, Neoglobo-

quadrina, and Turborotalita) and up to five genera of benthonicforaminifera (mainly Cibicides, accompanied by Elphidium, Melonis,

Oridorsalis, and Uvigerina). In the levels of the middle part, onlythree planktonic genera were recognized (Globigerina, Globorotalia,and Neogloboquadrina) and two benthic genera (Cibicides andGyroidina). The upper levels of the unit contain no microfauna.

Fig. 5. Stratigraphy and biostratigraphy of Units I, II, and III (pro parte) in the La Malaha section, indicating the position of the astronomically calibrated bioevents (PF1-CN1 to

PF4).

H. Corbı et al. / Geobios 45 (2012) 249–263 255

Based on these data, four events of planktonic foraminifera canbe established (PF1 to PF4; Fig. S1) and correlated withastronomically-dated events in other sections of the Mediterra-nean (Fig. 5). The PF1 event is defined in sample 4 by the jointpresence of G. menardii group I (sinistral), G. mediterranea andG. miotumida (G. conomiozea group), and S. seminulina. Thisassociation occurred at 7.892 Ma in a brief influx of theG. conomiozea group illustrated by Hilgen et al. (1995) in Metochia(Crete) and Gibliscemi (Sicily), by Krijgsman et al. (2000) in Lorca(Spain), and by Husing et al. (2009) in Monte dei Corbi (Apennines).The PF2 event is marked by the FCO of G. suterae (sample 18), anevent dated at 7.84 Ma (Sprovieri et al., 1999; Iaccarino et al.,2007). In the PF3 event (sample 24), dextral N. acostaensis

predominates over sinistral forms of this species. This event canbe correlated with the first and most dominant influx of dextral

N. acostaensis detected by Krijgsman et al. (2000) in Lorca (Spain),which coincides with the limit between the C4n.2n and C4n.1rchrons (7.697 Ma). The PF4 event (samples 40 to 44) ischaracterized by an abundant influx of the G. menardii group II(dextral forms). It is interesting to note that, in sample 40,Turborotalita multiloba (Fig. S2) coexists abundantly with dextralG. menardii. This influx of T. multiloba is not documented in otherMediterranean basins; generally, this species is reported in ayounger interval of the Messinian (Hilgen and Krijgsman, 1999;Sierro et al., 2001). The PF4 event coincides with the first influx(FO) of the G. menardii group II during the latest Tortonian,calibrated at 7.355 Ma by Hilgen et al. (1995) in Faneromeni andMetochia (Crete), between ca. 7.36 and 7.37 Ma by Krijgsman et al.(1997) in Monte del Casino (Apennines), and at 7.419 Ma by Morigiet al. (2007) in Pissouri (Cyprus).

Fig. 6. Stratigraphy of Unit III (evaporites) in the La Malaha section, represented mainly by gypsum cycles. A–C. Details of the gypsum laminite/marls fine alternation.

D. Irregular reddish (iron oxides) surfaces inside the microcrystalline nodular gypsum. E. Nodular gypsum showing chickenwire texture. F. Nodular gypsum capped by

laminated carbonates with stromatolite-like undulations.

H. Corbı et al. / Geobios 45 (2012) 249–263256

In the biostratigraphy of Unit III, the most diversifiedassociations of planktonic foraminifera in the lower part of theunit have little chronological value. Nevertheless, the dominantdextral-coiling Globorotalia scitula and the absence of theG. conomiozea group point to the latest Tortonian, bracketedbetween 7.34 and 7.24 Ma (in agreement with the data of Morigiet al., 2007 in the Pissouri section in Cyprus).

4.1.3. Calcareous nannoplankton

Units I and II contain, from bottom to top, an association ofnannoflora characterized by the genera Coccolithus, Reticulofenestra

(between 5 and 7 mm), Dictyococcites, Helicosphaera, Calcidiscus,Pontosphaera, Geminilitella, Amaurolithus, and Discoaster. Reticulo-

fenestra pseudoumbilicus greater than 7 mm is absent or inpercentages of less than 1%. In contrast, small reticulofenestrids(Dictyococcites productus and Reticulofenestra minuta) abound in allthe samples. From samples 1 and 4, Discoaster berggrenii andD. surculus, respectively, regularly appear. At the top of Unit 2, thelocal presence of Reticulofenestra rotaria (sample 52) was detectedas well as Amaurolithus primus (sample 57), with morphologiescharacteristic of initial forms.

From a biostratigraphic standpoint (Fig. 5), the absence ofR. pseudoumbilicus greater than 7 mm indicates that Units I and IIare within the paracme of this long-range species, the bottom and

top of which are calibrated at 8.7 and 7.1 Ma, respectively (Lourenset al., 2004), that is, basically the late Tortonian. This paracme hasbeen recognized by Krijgsman et al. (2000) in the pre-evaporiticmarine sediments of the late Tortonian of Lorca (Spain). The onlyevent with precise chronological significance is the CN1 event(sample 4), with the FCO of D. surculus, dated at 7.88 Ma accordingto the astronomical ages from Raffi et al. (2006). The occasionalpresence of R. rotaria in sample 52 gives an approximate age for thetop of Unit 2 as younger than 7.4 Ma. In Pissouri (Cyprus), Morigiet al. (2007) located the FO of this species at 7.435 Ma, an age closeto the 7.41 Ma proposed by Lourens et al. (2004). The occasionalpresence of A. primus in sample 57 (only 2.5 m above sample 52)points to an age younger than 7.4 Ma. Actually, Morigi et al. (2007)dated the first occurrence of this species at 7.449 Ma, a value closeto the 7.422–7.362 Ma given by Lourens et al. (2004) and Raffi et al.(2006).

4.1.4. Standard biozonations

For planktonic foraminifera, we adopted the Mediterraneanzonal scheme proposed by Iaccarino et al. (2007), in which the agesand magnetochronological calibrations of biozones are based onLourens et al. (2004). According to this zonal scheme, the lateTortonian corresponds to the MMi 12 biozone or G. extremus

interval zone, defined by the FO of this species at 8.35 Ma and the

H. Corbı et al. / Geobios 45 (2012) 249–263 257

FCO of the G. miotumida group at 7.24 Ma (Tortonian/Messinianboundary). The entire La Malaha succession was assigned to thisbiozone, given that the marker species was recognized almost fromthe bottom (sample 7) to the top (sample 58) of the succession.This assignation is supported by the age of the PF1 event (7.892Ma) and of the CN1 event (7.88 Ma), the two coinciding in sample4. In the zonal scheme of Iaccarino et al. (2007), within the MMi 12biozone, a separation has been made between the MMi 12a andMMi 12b subzones, the limit of which is defined by the FO ofG. suterae at 7.84 Ma. In the La Malaha succession, the FO of thisspecies was located in sample 18, a finding that enables us to assignthe majority of the succession to the MMi12b biozone. It bearsnoting that the MMi 13 biozone (Messinian), defined by the FCO ofthe Goborotalia miotumida group (or G. conomiozea group), is notrepresented in the La Malaha succession, nor in the rest of theGranada Basin successions.

As for the calcareous nannoplankton, the biozonation of Okadaand Bukry (1980) has been used, their data having been calibratedastrobiochronologically by Raffi et al. (2006). According to thiszonal scheme, the late Tortonian corresponds to biozone CN9a andthe lower part of biozone CN9bA. Biozone CN9a is defined betweenthe lowest occurrence of D. berggrenii at 8.294 Ma and the lowestoccurrence of A. primus at 7.424–7.362 Ma. Within this biozone,the lowest consistent occurrence of D. surculus is significant at 7.88Ma. The CN9bA biozone spans to the lowest occurrence ofNicklithus amplificus at 6.909 Ma. In the La Malaha section,D. berggrenii appears from sample 1, and A. primus is present insample 57. This indicates that almost all the section corresponds tobiozone CN9a, and only the end part can be assigned to biozoneCN9bA.

4.2. Other complementary sections in the Granada Basin

Units I and II are represented in the Genil River section;due to the position of this section on the basin margin, Unit III isabsent (Fig. 3). Unit I consists of a lower interval of skeletalsandstones and conglomerates (Subunit Ia) that graduallyevolved upwards to light-grey marls with abundant planktonicorganisms (Subunit Ib). Above these marls, Unit II consists of13 thickening- and coarsening-upward stacked deltaicsequences, each of which begins with dark micaceous siltsand ends with matrix-supported breccias and conglomerates(according to Garcıa-Garcıa, 2003). The fine sediments at thebottom of the sequences are rich in planktonic and benthicorganisms, as well as other marine fossils (echinoderms,ostreids, and pectinids). In the transition zone between SubunitsIa and Ib, characterized by an alternation of skeletal sandstonesand grey marls, the influx of the G. conomiozea group wasdetected. This influx is represented predominantly by specimensof G. miotumida, though some morphotypes have a very inflatedventral side, resembling G. mediterranea. At the bottom of thelight-grey marls that form Subunit Ib, the FCO of G. suterae wasrecognized, with this species extending upwards throughoutUnit II.

In the Agron section, Subunit Ia lies directly over the basement,dominated by skeletal sandstones and conglomerates. Theoverlying Subunit Ib is represented by an alternation of thin-bedded sandstones and light-coloured sandy marls. Unit II, as inthe La Malaha section, consists of marls rich in planktonic andbenthic foraminifers. In Agron, the Unit III evaporites reach theirmaximum thickness in the basin. At the bottom of the Unit II marlsis the FCO of G. suterae. The influx of the G. conomiozea group wasnot detected below the bioevent of G. suterae due both to themeagre content in planktonic foraminifera of the sandy marls ofSubunit Ib as well as to the scant development of the transitionzone between Subunit Ib and Unit II.

5. Discussion

5.1. Chronology of palaeogeographic events in the Granada Basin

Our biostratigraphic results are crucial for dating the mainchanges that occurred during the marine sedimentation in theGranada Basin. Unit I, which marks the entry of the sea into thebasin, is dated in Subunit Ib, where the influx of the G. conomiozea

group occurred at 7.892 Ma. In the Genil River section, where themarls of this subunit are thickest, this bioevent is followed by theFO of G. suterae. This finding indicates that the transgressionculminated at 7.84 Ma. The age of the beginning of thetransgression cannot be constrained because Subunit Ia (coastalto shallow-marine sandstones) is very poor in biostratigraphicmarkers. Nevertheless, the presence of G. extremus andN. acostaensis (sinistral forms, according to our own data)illustrated by Gonzalez Donoso (1977–1978) points to an ageyounger than 8.35 Ma.

The persistent occurrence of G. suterae in Unit II suggests thatthe maximum marine extension in the basin took place between7.84 Ma and 7.35–7.41 Ma (PF4 event). Additional nannofossil data(presence of A. primus and R. rotaria at the top of Unit II) confirmthat the fully marine conditions persisted until at least 7.4 Ma. Theabsence of the FCO of the G. conomiozea group indicates that thissituation did not last until 7.24 Ma, and thus, the top of Unit II isrestricted to the latest Tortonian.

This date allows us to approximate the age of the bottom of UnitIII (evaporites) and, consequently, the onset of the marine-restriction phase prior to the establishment of continentalenvironments. As commented above, the association of planktonicforaminifera recorded at the bottom of the evaporites (dextralforms of G. scitula) point to a latest Tortonian age. Unfortunately,the marl interbeds in the rest of the evaporitic succession do notcontain biostratigraphic markers to determine precisely the age ofthe top of Unit III. A latest Tortonian age for the top of the evaporiticunit has been proposed by Garcıa-Alix et al. (2008). Thischronological assignation is supported by the presence of smallmammals that date the upper part of the Middle Turolian (MN12zone) at the bottom of the first continental unit (Unit IV, in thiswork) of the Granada Basin. We have reservations about this datingfor the top of the evaporites since a lower Messinian age cannot beruled out. The limits of the MN12 zone are highly imprecise (Agustıet al., 2001: fig. 2). According to these authors, the absolute age forthe lower boundary of this zone is debatable between 7.5 Ma,which they propose, and 8.0 Ma (Opdyke et al., 1997). The upperlimit of the MN12 zone is truly problematic, being situated in anuncertainty interval of 7.2 to 6.8 Ma. Until a more accurateabsolute calibration of this upper limit is provided, the precise ageof the onset of continental sedimentation in the Granada Basin willremain unresolved.

5.2. Regional correlation

The evaporites of the Late Miocene prior to the MessinianSalinity Crisis can be considered one of the best markers of themarine-restriction phase of the basins that formed the Beticseaway. These evaporites were recognized in three interior basinsof the Betic Cordillera: the Granada Basin, located in the centralsector, and the Lorca and Fortuna basins, both in the eastern sector(Fig. 7). In these two latter basins, the marine-restriction phasethat led to the evaporitic precipitation has been defined as theTortonian Salinity Crisis (TSC; Krijgsman et al., 2000). Theevaporites related to the Messinian Salinity Crisis (MSC) appearonly in the marginal basins, very close to the Mediterranean, of theeastern end of the Betic Cordillera (Bajo Segura, Sorbas, and Nıjarbasins).

Fig. 7. Biostratigraphic-based correlation between the Granada, Lorca, Fortuna, and Bajo Segura basins. Note the diachrony in the restriction age (defined by the onset of evaporite deposition) in the first three basins, and the lack of

coeval evaporites in the Bajo Segura Basin. For interpretation of references to colours, see the web version of this article.

H.

Co

rbı

et a

l. /

Geo

bio

s 4

5 (2

01

2)

24

9–

26

32

58

H. Corbı et al. / Geobios 45 (2012) 249–263 259

The diverse astronomically calibrated bioevents considered inorder to establish the chronology of the stratigraphic record of theGranada Basin have been applied to the Lorca and Fortuna basins(Fig. 7). These biostratigraphic criteria are key for establishing theprecise onset of the TSC in the three interior Betic basins, andconsequently for discussing the isochronous vs. diachronouscharacter of the TSC at a regional scale. Additionally, the pre-MSC sequence of the Bajo Segura Basin was analysed in order toillustrate the stratigraphic manifestation of the TSC in marginalbasins, without evaporites before the MSC.

5.2.1. Lorca Basin

The Lorca Basin has been studied by a great number ofresearchers who have examined its evaporitic sequence in more orless detail (Geel, 1976; Guillen-Mondejar et al., 1995; Michalzik,1996; Rouchy et al., 1998; Krijgsman et al., 2000; Taberner et al.,2000; among many others). In our study, the planktonicforaminiferal biostratigraphy has been carried out in a section ofthe basin centre. The position of this section roughly coincides withsection B1 of Geel (1976), the reason why the lithostratigraphicnomenclature proposed by this author was adopted. The mostcharacteristic feature of the section is a thick succession of marine,plankton-rich marls (Hondo Formation). This formation is cappedby a marly-evaporitic succession (Serrata Formation), whichcomprises a lower member (Varied Member) and an upper one(Gypsum Member). The Varied Member is characterized by analternation of diverse lithologies, primarily laminated marls rich inorganic matter, carbonates, diatomites, silexites, gypsum, andsandstones. The Gypsum Member is dominated by laminated andchickenwire fabrics. The top of the succession is represented bylutites of continental origin.

Rouchy et al. (1998) hold that the Varied Member of the SerrataFormation (Tripoli Unit according to these authors) was depositedin a shallow-water environment subjected to hypersaline episodes,marking the episode of marine restriction that culminated with theaccumulation of the thick Gypsum Unit of the top of the SerrataFormation (including a halite unit detected by boreholes). Rouchyet al. (1998) reported the presence of G. conomiozea at the top of theHondo Formation, leading them to date the Serrata Formation asearly Messinian. Afterwards, Krijgsman et al. (2000) argued thatsuch a bioevent would correspond to the influx of the G. miotumida

group (or G. conomiozea group) of the late Tortonian, at 7.892 Ma.Based on the astronomical calibration of the cycles from the top ofthe Hondo Formation, these authors conclude that the Hondo/Serrata transition (i.e., the onset of the marine-restriction episode),occurred at 7.80 Ma. In addition, from the calibration of the cyclesof the Varied Member, the Gypsum Member dates to about 7.6 Ma.These datings have been the key for defining the Tortonian SalinityCrisis of the eastern Betics as having a duration of ca. 200 kyr(Krijgsman et al., 2000).

Our biostratigraphic results indicate that the totality of theHondo Formation corresponds to the late Tortonian (Fig. 7) giventhat the FO of G. extremus (8.35 Ma, according to Iaccarino et al.,2007) has been identified at its bottom. Close to the top of theHondo Formation, the influx of the G. conomiozea group (whichincludes typical forms of G. mediterranea) has been recognized inthe same position as illustrated by Rouchy et al. (1998) andKrijgsman et al. (2000). This influx can be correlated with the PF1event (7.892 Ma) of the Granada Basin, located at the transitionbetween Unit I (skeletal sandstones) and Unit II (basinal marls). Asa novelty of our study, in the last levels of the Hondo Formation,immediately under the Serrata Formation, the appearance ofG. suterae has been identified, permitting a correlation with the PF2event (7.84 Ma) of the Granada Basin, located at the bottom of theUnit II marls. Finally, in the lower part of the Serrata Formationappear dominant dextral forms of N. acostaensis as reported by

Rouchy et al. (1998), and Krijgsman et al. (2000). This latterbioevent fits with the PF3 event (7.69 Ma) of the Granada Basin,located within the Unit II marls. The foraminifer content in theupper part of the Varied Member of the Serrata Formation,especially below and within the Gypsum Member, is very poor;most of the samples studied are barren or contain no biostrati-graphic markers. For this reason, due to the less than favourablepalaeoecological conditions of the evaporitic basin for thedevelopment of planktonic organisms, the PF4 event of theGranada Basin has not been detected (influx of G. menardii groupII [dextral forms]).

These data point to a diachrony of the marine-restriction event(Tortonian Salinity Crisis) in the Granada and Lorca basins. In theGranada Basin, the beginning of the evaporitic conditions (Unit III)is younger than the PF4 event (7.37 Ma). In the Lorca Basin, theVaried Member of the Serrata Formation (which contains the firstevaporites in the basin) began at 7.80 Ma.

5.2.2. Fortuna Basin

The stratigraphic succession that best characterizes thesedimentary fill of the Fortuna Basin, where the evaporiticsequence appears, crops out along the Chicamo River. Thissuccession is dominated by a thick unit of marine marls calledthe Fortuna Marls (Montenat et al., 1990a), which are overlain bythe basin’s evaporitic suite. This suite contains three evaporiticunits (Santisteban, 1981; Lancis et al., 2010). The first is known asthe Lower Gypsum (the First Evaporitic Group; Ortı et al., 1993).This lies over a basal conglomerate that separates the FortunaMarls and is overlain by a unit of cyclically-arranged marls andsandstones (Sanel Marls) characterized by their high content inplanktonic and benthic microfossils. The second, which representsthe main and thickest evaporitic unit of the basin, contains a basalassemblage of laminated gypsum (Tale Gypsum) overlain by analternation of gypsum and marine diatomites (Chicamo Cycles).Over these cycles, a noticeable conglomerate layer (WichmannBed) marks the discontinuity bounding the third evaporitic unit(Rambla Salada Gypsum), of continental origin.

The present study examines the planktonic foraminifera of theentire succession of the Chicamo River, including most of theFortuna Marls and the rest of the overlying units. The regularpresence of sinistral N. acostaensis has been identified throughoutthe succession, in addition to the first appearance of G. extremus

immediately above a conspicuous interval of sandstones called theSerratilla Bed. This latter bioevent permits a correlation of theFortuna Marls with the Hondo Formation of the Lorca Basin as wellas with Unit I of the Granada Basin. With respect to the age of theevaporitic events, the foraminifera do not resolve the problem.According to nannoflora data from Lancis et al. (2010), incombination with the magnetostratigraphy given by Dinares-Turell et al. (1999) and Krijgsman et al. (2000), the Tortonian/Messinian boundary is located in the Sanel Marls/Tale Gypsumtransition.

Following Tent-Manclus et al. (2008), the Lower Gypsumrepresents the first restriction event in the Fortuna Basin, that is,the onset of the Tortonian Salinity Crisis. Assuming the lithostrati-graphic correlation argued by Ortı et al. (1993), we can concludethat the Lower Gypsum in the Fortuna Basin is equivalent to theGypsum Member of the Serrata Formation in the Lorca Basin, andthus, an age of around 7.6 Ma can be assigned. This chronologicalassignment is consistent with the FO of A. primus (7.45 Ma;according to Morigi et al., 2007 in Cyprus) within the overlyingSanel Marls. The second and main basin-restriction event occurredwith the Tale Gypsum, the age of which coincides with theTortonian/Messinian boundary at 7.24 Ma, immediately before thefirst occurrence of Amaurolithus delicatus (FO at 7.22 Ma, accordingto Lourens et al., 2004; see Lancis et al., 2010 for the

H. Corbı et al. / Geobios 45 (2012) 249–263260

biomagnetostratigraphic argument). Finally, the base of theRambla Salada Gypsum, which represents the onset of continentalsedimentation in the Fortuna Basin, fits with the limit of the chronsC3Ar and C3An.2n at 6.73 Ma (ATNTS 2004), in accord with the firstoccurrence of N. amplificus (FO at 6.68 Ma; Lourens et al., 2004)immediately underneath, on top of the Chicamo Cycles.

The above argument points to the notion that the first marine-restriction event in the Fortuna Basin (at ca. 7.6 Ma) was diachronicwith respect to the Lorca Basin (at 7.8 Ma). There was nocorrespondence between the second restriction event of theFortuna Basin (Tale Gypsum, at 7.24 Ma) and the main evaporiticphase of the Lorca Basin (Gypsum Member of the SerrataFormation, at ca. 7.6 Ma). This chronological divergence contra-dicts the isochrony of the Tortonian Salinity Crisis proposed byKrijgsman et al. (2000) for these basins. In relation to the GranadaBasin, the basin-restriction event (Unit III; evaporites) is youngerthan 7.37 Ma and thus, cannot be correlated with the firstrestriction event of the Fortuna Basin (7.6 Ma). More reliable is theequivalence of the evaporitic phase of the Granada Basin and thesecond evaporitic unit of the Fortuna Basin (Tale Gypsum andChicamo Cycles). In good agreement with this proposal, A. primus

was found to be present immediately before the aforementionedevaporites in both basins.

5.2.3. Bajo Segura Basin

The Bajo Segura is a typical marginal Mediterranean basinlocated at the eastern end of the Betic Cordillera. According to thecomplete biostratigraphic study by Corbı (2010) and Corbı et al.(2010) using planktonic foraminifera, the age of their stratigraphicrecord spans the early Tortonian to early Pliocene (Fig. 7). For thistime interval, all the modern, astronomically calibrated biozoneshave been recognized, together with the specific intrazonalbiohorizons (Lourens et al., 2004; Iaccarino et al., 2007). Thisbasin presents an evaporitic unit (San Miguel Formation) related tothe Messinian Salinity Crisis (Soria et al., 2008). This unit can becorrelated (based on biomagnetostratigraphic criteria) with theevaporites of the Sorbas and Nıjar basins, the beginnings of whichhave been astronomically calibrated at 5.96 Ma (Krijgsman et al.,1999). The pre-evaporitic deposits (Torremendo Formation) of theBajo Segura Basin are equivalent in age as well as in cyclicarrangement to the corresponding ones in the Sorbas and Nıjarbasins (Abad Member). Contrary to the situation in the Granada,Lorca, and Fortuna basins, no marine-restriction event leading toearly evaporite precipitation (late Tortonian) occurred in the BajoSegura. Instead, the Bajo Segura Basin maintained marineconditions during the Tortonian, as witnessed by the depositionof the Columbares and Pujalvarez formations. This basin isespecially significant to illustrate the validity and usefulness forregional correlations of the PF1 bioevent (influx of theG. conomiozea group) and PF2 (FO of G. suterae), among otherbioevents, detected in the Granada Basin.

In the Bajo Segura Basin, the limit between the ColumbaresFormation (below) and the Pujalvarez Formation (above) coincideswith the FO of G. extremus. This allows Unit I (Granada Basin), theHondo Marls (Lorca Basin), and the Fortuna Marls (Fortuna Basin)to be correlated with the Pujalvarez Formation (Bajo Segura Basin).Towards the middle of the Pujalvarez Formation appear the firstspecimens of the G. conomiozea group, including typical forms ofG. mediterranea. This group appears in a stratigraphic interval of30 m, in association with specimens of the G. menardii group I(sinistral forms). The joint presence of these two groups ofgloborotalids defines bioevent PF1 (7.89 Ma) in the same manneras in the Granada Basin. In the upper part of the PujalvarezFormation, the FO of G. suterae (PF2; 7.84 Ma) was detected.

In short, the complex history of early marine restriction in theGranada, Lorca, and Fortuna interior basins occurred while marine

sedimentation continued (Pujalvarez Formation) in the BajoSegura marginal basin. The only restriction event in this basin(evaporites of the San Miguel Formation) is related to theMessinian Salinity Crisis.

5.3. The evaporitic phase in the palaeogeographic evolution of the

Betic seaway

The palaeogeographic evolution of the Betic seaway is difficultto elucidate, as a great part of the stratigraphic record has beeneliminated by erosion, due to the general uplift of the BeticCordillera since the Late Miocene (Braga et al., 2003; Sanz deGaldeano and Alfaro, 2004). There is no consensus amongresearchers concerning the configuration of the Betic seaway;some trace its shape in relation to present-day outcrops, whereasothers extrapolate the configuration to broader areas, resulting innotable differences in the palaeogeographic schemes proposed(Serrano, 1979; Rodrıguez Fernandez, 1982; Rodrıguez Fernandezand Sanz de Galdeano, 1992; Sanz de Galdeano and Vera, 1992;Soria, 1994; Esteban, 1996; Sanz de Galdeano and RodrıguezFernandez, 1996; Soria et al., 1999; Braga et al., 2003).

The palaeogeographic scheme assumed in our study (Fig. 8)incorporates the data of the above-mentioned authors. Themaximum extension of the Betic seaway corresponds to the highsea-level phase during the late Tortonian. Biostratigraphically,this phase occurred after the FO of G. extremus (at 8.35 Ma) in theGranada, Lorca, Fortuna, and Bajo Segura basins. The highstandunits of these basins are correlated with other basins of the BeticCordillera from both the Atlantic and the Mediterranean domain(Serrano, 1979; Rodrıguez Fernandez, 1982), indicating that anample marine connection existed between the two domains. Thepalaeogeography corresponding to the restriction of the Beticseaway has been documented only where lowstand depositsformed, as in the Guadix Basin (Soria, 1994; Soria et al., 1999;among others) and in the Fortuna Basin (Tent-Manclus et al.,2008).

In the Granada Basin, the lowstand evaporites occupy the basincentre. We propose a connection of the evaporites with theMediterranean through a narrow tectonic corridor located south ofthe basin. This corridor inherits its position from a larger seawaybetween the Granada Basin and the Mediterranean (Braga et al.,2003). The existence of a network of NW-SE extensional faults(among other sets) to the south of the basin, active starting in theTortonian (Sanz de Galdeano and Lopez-Garrido, 2000), points totectonic control in the genesis of the restricted seaway with whichthe Granada Basin evaporites are associated.

The Guadix Basin occupies a crucial position for understandingthe Atlantic-Mediterranean connection during the Tortonian (Soriaet al., 1999). Its stratigraphic record is comparable to that of theGranada Basin (Fernandez et al., 1996). In both basins, marine unitsI and II show a similar stratigraphic expression and age. However,in the Guadix Basin, Unit III, the last marine unit, haspredominantly shallow marine and coastal clastic facies (Soriaet al., 2003) as opposed to the evaporitic character of Unit III in theGranada Basin. In the Guadix Basin, Unit III was deposited in therestricted Betic seaway, which would have been open bothtowards the Atlantic domain (Guadalquivir Basin, to the north)as well as towards the Mediterranean domain (Lorca Basin, to theeast). The age of Unit III has been assigned to the last biozone of theTortonian (G. suterae biozone; Soria, 1994). This age is confirmed inthe recent magnetobiostratigraphic study by Husing et al. (2010),which states that the bottom of the lowstand deposits is situated atca. 7.87 Ma. This dating is an important support for the diachronyof the Betic seaway restriction event between the Guadix Basin andthe Granada Basin, as in the latter this event was younger than 7.37Ma (this study).

Fig. 8. Simplified palaeogeography of the Betic seaway. The maximum extent of the seaway corresponds to the high sea-level phase of the Betic basins during the late

Tortonian (after 8.35 Ma; FO of G. extremus). The restricted Betic seaway is marked by evaporite deposition in the Granada, Lorca, and Fortuna basins), and by clastic shallow-

marine and coastal sedimentation in the Guadix Basin. The different restriction age in the Betic seaway is related to the local tectonic setting of these basins. For interpretation

of references to colours, see the web version of this article.

H. Corbı et al. / Geobios 45 (2012) 249–263 261

The Lorca Basin is one of the basins with Mediterranean affinity,as expressed in the palaeogeographic scheme of Fig. 8. Actually, themarine marls of the Hondo Formation present the samestratigraphic position as the marine marls of the Columbaresand Pujalvarez formations in the Bajo Segura Basin. In addition, thebiostratigraphic data presented in this study (Fig. 7) confirm thecorrelation of these formations. Tectonically, the genesis andevolution of the Lorca Basin is related to the Alhama de Murciaaccident (Montenat et al., 1990b), which forms part of the Trans-Alboran Shear Zone (de Larouziere et al., 1988). This largetranscurrent fault zone was active from the Tortonian onwardsand was the tectonic element that controlled the early marinerestriction (Tortonian Salinity Crisis) of the eastern Betics (Krijgs-man et al., 2006).

The Fortuna Basin, like the Lorca Basin, was clearly opentowards the Mediterranean during the maximum extension of theBetic seaway (Fig. 8). Prior to the restriction phase of the Beticseaway, on the eastern end of the Betics, marine marls weredeposited both in the Fortuna Basin as well as in the Bajo SeguraBasin (Columbares and Pujalvarez formations; Fig. 7). According toTent-Manclus et al. (2008), during the latest Tortonian, the FortunaBasin was partially separated from the Bajo Segura Basin by astructural high (Orihuela and Callosa sierras) corresponding to thenorthern segment of the Trans-Alboran Shear Zone. This closureprovoked the marine restriction and precipitation of evaporites inthe Fortuna Basin, while open marine conditions persistedthroughout most of the Bajo Segura Basin.

The above synthesized palaeogeographic data point to tectoniccontrol in the genesis of the evaporitic basins related to therestriction phase of the Betic seaway. The diachrony at the onset ofthe evaporitic phase is associated with the local geodynamicconditions of the Granada, Lorca, and Fortuna basins. Therestriction phase occurred early in the Lorca and Fortuna basins(7.8 and 7.6 Ma, respectively), both located in the eastern sector ofthe Betic Cordillera. Later, the restriction occurred in the Granada

Basin (7.37–7.24 Ma), in the central sector of the Cordillera. Thisdifference can be explained by the dissimilar tectonic behaviour ofthe central and eastern sectors of the Cordillera. In the centralsector, the extensional faults shaping the Granada Basin generatedvertical tectonic movements of greater magnitude on a regionalscale; the late Tortonian marine deposits are currently found at1300 mbsl in the highly subsident zones of the basin interior,whereas the same deposits rise to 1800 masl in the mountain reliefof the Sierra Nevada (Rodrıguez Fernandez and Sanz de Galdeano,2006: fig. 12). By contrast, in the eastern sector, the verticaltectonic movements associated with the Trans-Alboran Shear Zoneare of notably lower magnitude. That is, the late Tortonian marinedeposits are found between 0 and 200 mbsl in the slowly subsidentGuadalentın valley, whereas the maximum elevation is reached inthe Sierra Carrascoy at 950 masl (Sanz de Galdeano and Alfaro,2004). In short, there is a marked correspondence between theduration of the marine restriction phase and the tectonic style andrate of vertical movements. The most long-lived restricted marineseaway is in the Granada Basin, where the highest rates ofsubsidence, associated to the extensional faults of the centralBetics, took place. In contrast, the most short-lived restrictedmarine seaway is in the Lorca and Fortuna basins, bothcharacterized by a low rate of tectonic subsidence in relation tothe transcurrent faults of the eastern Betics.

6. Conclusions

An updated biostratigraphy (planktonic foraminifera andcalcareous nannoplankton) of the Granada Basin has enabledthe recognition of the main late Tortonian bioevents included inthe Astronomically Tuned Neogene Time Scale (Lourens et al.,2004). According to our results, the lowest stratigraphic unit (UnitI) presents two close bioevents in its upper part: the influx of theG. conomiozea group (7.892 Ma) and the FO of G. suterae (7.84 Ma).Unit II, in its upper part, presents the abundant influx of the

H. Corbı et al. / Geobios 45 (2012) 249–263262

G. menardii group II (dextral forms; 7.37 Ma). Additionalnannofossil data (presence of A. primus and R. rotaria at the topof Unit II) confirm that this unit lasted until at least 7.4 Ma, and theabsence of the FCO of the G. conomiozea group indicates that it didnot reach 7.24 Ma. The bottom of the uppermost marine unit (UnitIII) contains dextral forms of G. scitula, indicating a latest Tortonianage. The top of Unit III could not be dated given the absence ofplanktonic fauna and flora.

Unit III records the evaporitic precipitation phase in theGranada Basin. This unit is represented by a cyclic evaporiticsuccession in which two facies associations are repeated: a finegypsum laminite/marl alternation and a microcrystalline nodulargypsum (chickenwire texture). Both facies associations areinterpreted as having formed in a shallow-water setting, wheregypsum laminites and marls represent subaqueous deposition, andnodular gypsum records coastal sabkha (supratidal) sedimenta-tion. Unit III marks the continuous marine restriction of theGranada Basin during the late Tortonian, as indicated by theprogressive decrease in planktonic fauna and flora from the bottomto the top of the unit. The restriction climax occurred with thecomplete isolation of the basin, giving rise to the continentalsedimentation phase.

The main bioevents used in this work constitute a valuable toolfor regional-scale correlations between the stratigraphic units ofthe Granada, Lorca, and Bajo Segura basins. In this respect,particularly significant are the influx of the G. conomiozea groupand the first regular occurrence of G. suterae.

From the standpoint of regional palaeogeography, the evapor-ites of the Granada, Lorca, and Fortuna basins record the restrictionphase of the Betic seaway, i.e., the marine connection between theAtlantic and the Mediterranean through the Betic Cordillera. Ourbiostratigraphic data suggest that the beginning of the restrictionphase was diachronic between the Granada Basin (7.37–7.24 Ma),the Lorca Basin (7.8 Ma), and the Fortuna Basin (7.6 Ma). Thispronounced time lag in the Granada Basin is explained by thedifferent structural style of this basin in relation to the Lorca andFortuna basins. From the late Tortonian onwards, the GranadaBasin has been submitted to a high rate of tectonic subsidence byextensional faults, whereas the other two basins underwentmoderate subsidence in relation to transcurrent faults (Trans-Alboran Shear Zone).

Acknowledgements

Financial aid was provided by Research Projects CGL2007-65832 MEC, CGL2009-07830 MCI, GV04B-629 (GeneralitatValenciana) and the ‘‘Paleoenvironmental Changes’’ Group (UA).The authors thank Dr. Gilles Escarguel (Editor in Chief), Dr. FredericQuillevere (Associate Editor), Dr. Marco Roveri (University ofParma), and an anonymous reviewer for their critical reading of thefirst version of the manuscript, as well as David Nesbitt andChristine Laurin for editing the English text.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.geobios.2011.10.006.

References

Agustı, J., Cabrera, L., Garces, M., Krijgsman, W., Oms, O., Pares, J.M., 2001. Acalibrated mammal scale for the Neogene of Western Europe. State of theart. Earth-Science Reviews 52, 247–260.

Braga, J.C., Martın, J.M., Quesada, C., 2003. Patterns and average rates of lateNeogene – Recent uplift of the Betic Cordillera, SE Spain. Geomorphology 50,3–26.

Corbı, H., 2010. Los foraminıferos de la cuenca neogena del Bajo Segura (sureste deEspana): bioestratigrafıa y cambios paleoambientales en relacion con la Crisisde salinidad del Mediterraneo. Thesis, Universidad de Alicante.

Corbı, H., Pina, J.A., Soria, J.M., 2010. Bioestratigrafıa basada en foraminıferosplanctonicos para el Mioceno superior y Plioceno de la Cuenca del Bajo Segura(Cordillera Betica oriental). Geogaceta 48, 71–74.

D’Onofrio, S., Gianelli, L., Iaccarino, S., Morlotti, E., Romeo, E., Salvatorini, G., Sampo,M., Sprovieri, R., 1975. Planktonic foraminifera of the Upper Miocene from someItalian sections and the problem of the lower boundary of the Messinian.Bolletino della Societa Paleontologica Italiana 14, 177–196.

Dabrio, C.J., Garcia Yebra, R., Conzalez Donoso, J.M., Vera, J.A., 1972. Turbiditasasociadas a evaporitas (Mioceno de La Mala, Depresion de Granada). Cuadernosde Geologıa Universidad de Granada 3, 139–164.

Dabrio, C.J., Martin, J.M., Megias, A., 1982. Signification sedimentaire des evaporitesde la depression de Grenade (Espagne). Bulletin de la Societe geologique deFrance 24 (7), 705–710.

de Larouziere, F.D., Bolze, J., Bordet, P., Hernandez, J., Montenat, C., Ott d’Estevou, P.,1988. The Betic segment of the lithospheric trans-Alboran shear zone during thelate Miocene. Tectonophysics 152, 41–52.

Dinares-Turell, J., Ortı, F., Playa, E., Rossell, L., 1999. Palaeomagnetic chronology ofthe evaporitic sedimentation in the Neogene Fortuna Basin (SE Spain): earlyrestriction preceding the ‘‘Messinian Salinity Crisis’’. Palaeogeography Palaeo-climatology Palaeoecology 154, 161–178.

Esteban, M., 1996. Western Mediterranean Reef Complexes. In: Franseen, E.K.,Esteban, M., Ward, W.C., Rouchy, J.M. (Eds.), Models for carbonate stratigraphyfrom Miocene reef complexes of Mediterranean regions, 5. SEPM Concepts inSedimentology and Paleontology, pp. 55–72.

Fernandez, J., Rodrıguez Fernandez, J., 1989. Facies evolution of nearshore marineclastic deposits during the Tortonian transgression. Granada Basin, Betic Cor-dillera, Spain. Sedimentary Geology 71, 5–21.

Fernandez, J., Soria, J.M., Viseras, C., 1996. Stratigraphic architecture of the Neogenebasins in the central sector of the Betic Cordillera (Spain): tectonic control andbase level changes. In: Friend, P.F., Dabrio, C.J. (Eds.), Tertiary Basins of Spain:The Stratigraphic Record of Crustal Kinematics. Cambridge University Press,Cambridge, pp. 353–365.

Garcıa-Alix, A., Minwer-Barakat, R., Martın, J.M., Martın Suarez, E., Freudenthal, M.,2008. Biostratigraphy and sedimentary evolution of Late Miocene and Pliocenecontinental deposits of the Granada Basin (southern Spain). Lethaia 41, 431–446.

Garcıa-Garcıa, F., 2003. Modelos de sedimentacion deltaica en las cuencas neogenasde la Cordillera Betica (sectores central y oriental). Thesis, Universidad deGranada.

Garcıa-Garcıa, F., Viseras, C., Fernandez, J., 1999. Organizacion secuencial de aba-nicos deltaicos controlados por la tectonica (Tortoniense superior, Cuenca deGranada, Cordillera Betica). Revista de la Sociedad Geologica de Espana 12, 199–208.

Geel, T., 1976. Messinian gypsiferous deposits of the Lorca Basin (province ofMurcia, SE Spain). Memorie della Societa Geologica Italiana 16, 369–385.

Gonzalez Donoso, J.M., 1977–1978. Los materiales miocenicos de la Depresion deGranada. Cuadernos de Geologıa Universidad de Granada 8/9, 191–203.

Guillen-Mondejar, F., Rodrıguez Estrella, T., Arana, R., Lopez Aguayo, F., 1995.Historia geologica de la cuenca de Lorca (Murcia): influencia de la tectonicaen la sedimentacion. Geogaceta 18, 30–33.

Hardie, L., Eugster, H., 1971. Depositional environment of marine evaporites: a casefor shallow, clastic accumulation. Sedimentology 16, 187–220.

Hilgen, F.J., Krijgsman, W., 1999. Cyclostratigraphy and astrochronology of theTripoli diatomite formation (pre-evaporite Messinian, Sicily, Italy). Terra Nova11, 16–22.

Hilgen, F.J., Krijgsman, W., Langereis, C.G., Lourens, L.J., Santerelli, A., Zachariasse,W.J., 1995. Extending the astronomical (polarity) time scale into the Miocene.Earth and Planetary Science Letters 136, 495–510.

Husing, S.K., Kuiper, K.F., Link, W., Hilgen, F.J., Krijgsman, W., 2009. The upperTortonian-lower Messinian at Monte dei Corvi (Northern Apennines, Italy):completing a Mediterranean reference section for the Tortonian Stage. Earthand Planetary Science Letters 282, 140–157.

Husing, S.K., Oms, O., Agustı, J., Garces, M., Kouwenhoven, T.J., Krijgsman, W.,Zachariasse, W.J., 2010. On the late Miocene closure of the Mediterranean-Atlantic gateway through the Guadix basin (southern Spain). PalaeogeographyPalaeoclimatology Palaeoecology 291, 167–179.

Iaccarino, S., Premoli-Silva, I., Biolzi, M., Foresi, L.M., Lirer, F., Turco, E., Petrizzo, M.R.,2007. Practical manual of Neogene planktonic foraminifera. InternationalSchool on Planktonic Foraminifera, 6th course. Universita degli Studi di Perugia.

Kendall, A.C., 1979. Subaqueous evaporites. In: Walker, R.G. (Ed.), Facies Model.Geoscience Canada Reprint Serie 1, 14. pp. 159–174.

Krijgsman, W., Garces, M., Agustı, J., Raffi, I., Taberner, C., Zachariasse, W.J., 2000. The‘‘Tortonian Salinity Crisis’’ of the eastern Betics (Spain). Earth and PlanetaryScience Letters 181, 497–511.

Krijgsman, W., Hilgen, F.J., Langereis, C.G., Santarelli, A., Zachariasse, W.J., 1995. LateMiocene magnetostratigraphy, biostratigraphy and cyclostratigraphy in theMediterranean. Earth and Planetary Science Letters 136, 475–494.

Krijgsman, W., Hilgen, F.J., Negri, A., Wijbrans, J.R., Zachariasse, W.J., 1997. TheMonte del Casino section (Northern Apennines, Italy): a potential TortonianMessinian boundary stratotype. Palaeogeography Palaeoclimatology Palaeoe-cology 133, 27–47.

Krijgsman, W., Hilgen, F.J., Raffi, I., Sierro, F.J., Wilson, D.S., 1999. Chronology, causesand progression of the Messinian salinity crisis. Nature 400, 652–655.

H. Corbı et al. / Geobios 45 (2012) 249–263 263

Krijgsman, W., Leewis, M.E., Garces, M., Kouwenhoven, T.J., Kuiper, K.F., Sierro, F.J.,2006. Tectonic control for evaporite formation in the Eastern Betics (Tortonian;Spain). Sedimentary Geology 188/189, 155–170.

Lancis, C., 1998. El nanoplancton calcareo de las cuencas neogenas orientales de laCordillera Betica. Thesis, Universidad de Alicante.

Lancis, C., Tent-Manclus, J.E., Soria, J.M., Caracuel, J.E., Corbı, H., Dinares-Turell, J.,Estevez, A., Yebenes, A., 2010. Nannoplankton biostratigraphic calibration of theevaporitic events in the Neogene Fortuna Basin (SE Spain). Geobios 43, 201–217.

Lourens, L., Hilgen, F., Shackleton, N.J., Laskar, J., Wilson, D., 2004. The Neogeneperiod. In: Gradstein, F.M., Ogg, J.G., Smith, A.G. (Eds.), A geologic Time Scale.Cambridge University Press, Cambridge, pp. 409–440.

Martın, J.M., Ortega-Huertas, M., Torres-Ruiz, J., 1984. Genesis and evolution ofstrontium deposits of the Granada Basin (southeastern Spain): Evidence ofdiagenetic replacement of a stromatolite belt. Sedimentary Geology 39, 281–298.

Michalzik, D., 1996. Lithofacies, diagenetic spectra and sedimentary cycles ofMessinian (Late Miocene) evaporites in SE Spain. Sedimentary Geology 106,203–222.

Montenat, C., Ott d’Estevou, Ph., Coppier, G., 1990a. Les bassins neogenes entreAlicante et Cartagena. In: Montenat, C. (Ed.), Les Bassins Neogenes du DomaineBetique Oriental (Espagne). Documents et Travaux, 12/13. Institut GeologiqueAlbert de Lapparent, pp. 313–368.

Montenat, C., Ott d’Estevou, P., Delort, T., 1990b. Le Bassin de Lorca. In: Montenat, C.(Ed.), Les Bassins Neogenes du Domaine Betique Oriental (Espagne). Docu-ments et Travaux, 12/13. Institut Geologique Albert de Lapparent, pp. 261–280.

Morigi, C., Negri, A., Giunta, S., Kouwenhoven, T., Krijgsman, W., Blanc-Valleron, M.-M., Orszag-Sperber, F., Rouchy, J.-M., 2007. Integrated quantitative biostratig-raphy of the latest Tortonian-early Messinian Pissouri section (Cyprus): anevaluation of calcareous plankton bioevents. Geobios 40, 267–279.

Okada, H., Bukry, D., 1980. Supplementary modification and introduction of codenumbers to the low-latitude coccolith biostratigraphic zonation (Bukry, 1973,1975). Marine Micropaleontology 5, 321–325.

Opdyke, N., Mein, P., Lindsay, E., Perez-Gonzalez, A., Moissenet, E., Norton, V.L.,1997. Continental deposits, magnetostratigraphy and vertebrate paleontology,late Neogene of Eastern Spain. Palaeogeography Palaeoclimatology Palaeoecol-ogy 133, 129–148.

Ortı, F., Garcıa-Veigas, J., Rosell, L., Rouchy, J.M., Ingles, M., Gimeno, D., Kasprzyk, A.,Playa, E., 1993. Correlacion litoestratigrafica de las evaporitas messinienses enlas cuencas de Lorca y Fortuna (Murcia). Geogaceta 14, 98–101.

Raffi, I., Backman, J., Fornaciari, E., Palike, H., Rio, D., Lourens, L., Hilgen, F., 2006. Areview of calcareous nannofossil astrobiochronology encompassing the past 25million years. Quaternary Science Reviews 25, 3113–3137.

Rodrıguez Fernandez, J., 1982. El Mioceno del sector central de las CordillerasBeticas. Thesis, Universidad de Granada.

Rodrıguez Fernandez, J., Sanz de Galdeano, C., 1992. Onshore Neogene stratigraphyin the North of the Alboran Sea (Betic Internal Zones): paleogeographic impli-cations. Geo-Marine Letters 12, 123–128.

Rodrıguez Fernandez, J., Sanz de Galdeano, C., 2006. Late orogenic intramontanebasin development: the Granada basin, Betics (southern Spain). Basin Research18, 85–102.

Rouchy, J.M., 1976. Sur la genese de deux principaux types de gypse (finement lite eten chevrons) du Miocene terminal de Sicile et d’Espagne Meridionale. Revue deGeographie Physique et de Geologie Dynamique 2, 347–365.

Rouchy, J.M., Pierre, C., 1979. Donnees sedimentologiques et isotopiques sur lesgypses des series evaporitiques messiniennes d’Espagne meridionale et deChypre. Revue de Geographie Physique et de Geologie Dynamique 21, 267–280.

Rouchy, J.M., Taberner, C., Blanc-Valleron, M.M., Sprovieri, R., Russell, M., Pierre, C.,Di Stefano, E., Pueyo, J.J., Caruso, A., Dinares-Turell, J., Gomis-Coll, E., Wolff, G.,Cespuglio, G., Ditchfield, P., Pestrea, S., Comborieu-Nebout, N., Santisteban, C.,Grimalt, J.O., 1998. Sedimentary and diagenetic markers of the restriction in a

marine basin: The Lorca basin (SE Spain) during the Messinian. SedimentaryGeology 121, 23–55.

Santisteban, C., 1981. Petrologıa y sedimentologıa de los materiales del Miocenosuperior de la cuenca de Fortuna (Murcia), a la luz de la crisis de salinidad.Thesis, Universidad de Barcelona.

Sanz de Galdeano, C., Alfaro, P., 2004. Tectonic significance of the present relief ofthe Betic Cordillera. Geomorphology 63, 175–190.

Sanz de Galdeano, C., Lopez-Garrido, A.C., 2000. Las fallas tortonienses a cuaternar-ias entre Granada y la Costa: el lımite occidental del Nevado–Filabride y de lasunidades alpujarrides inferiores. Revista de la Sociedad Geologica de Espana 13,519–528.

Sanz de Galdeano, C., Rodrıguez Fernandez, J., 1996. Neogene palaeogeography ofthe Betic Cordillera: an attempt of reconstruction. In: Friend, P.F., Dabrio, C.J.(Eds.), Tertiary Basins of Spain: The Stratigraphic Record of Crustal Kinematics.Cambridge University Press, Cambridge, pp. 323–329.

Sanz de Galdeano, C., Vera, J.A., 1992. Stratigraphic record and palaeogeographicalcontext of the Neogene Basins in the Betic Cordillera, Spain. Basin Research 4,21–36.

Serrano, F., 1979. Los foraminıferos planctonicos del Mioceno superior de la Cuencade Ronda y su comparacion con los de otras areas de las Cordilleras Beticas.Thesis, Universidad de Malaga.

Sierro, F.J., 1984. Foraminıferos planctonicos y bioestratigrafıa del Mioceno super-ior-Plioceno del borde occidental de la cuenca del Guadalquivir (SO de Espana).Thesis, Universidad de Salamanca.

Sierro, F.J., Flores, J.A., Civis, J., Gonzalez Delgado, J.A., Frances, G., 1993. LateMiocene globorotaliid event-stratigraphy and biogeography in the NE-Atlanticand Mediterranean. Marine Micropaleontology 21, 143–168.

Sierro, F.J., Hilgen, F.J., Krijgsman, W., Flores, J.A., 2001. The Abad composite (SESpain): a Messinian reference section for the Mediterranean and the APTS.Palaeogeography Palaeoclimatology Palaeoecology 168, 141–169.

Soria, J.M., 1994. Evolucion sedimentaria y paleogeografica durante el Miocenosuperior en el borde norte de la Cuenca de Guadix, Cordillera Betica central.Estudios Geologicos 50, 59–69.

Soria, J.M., Caracuel, J.E., Corbı, H., Dinares-Turell, J., Lancis, C., Tent-Manclus, J.E.,Viseras, C., Yebenes, A., 2008. The Messinian-early Pliocene stratigraphic recordin the southern Bajo Segura Basin (Betic Cordillera, Spain): implications for theMediterranean salinity crisis. Sedimentary Geology 203, 267–288.

Soria, J.M., Fernandez, J., Garcıa, F., Viseras, C., 2003. Correlative lowstand deltaicand shelf systems in the Guadix Basin (Late Miocene, Betic Cordillera, Spain):the stratigraphic record of forced and normal regressions. Journal of Sedimen-tary Research 73, 912–925.

Soria, J.M., Fernandez, J., Viseras, C., 1999. Late Miocene stratigraphy and paleogeo-graphic evolution of the intramontane Guadix Basin Central Betic Cordillera,Spain): implications for an Atlantic-Mediterranean connection. Palaeogeogra-phy Palaeoclimatology Palaeoecology 151, 255–266.

Sprovieri, M., Bellanca, A., Neri, R., Mazzola, S., Bonanno, A., Patti, B., Sorgente, R.,1999. Astronomical calibration of late Miocene stratigraphic events and analy-sis of precessionally driven paleoceanographic changes in the Mediterraneanbasin. Memorie della Societa Geologica Italiana 54, 7–24.

Taberner, C., Cendon, D.I., Pueyo, J.J., Ayora, C., 2000. The use of environmentalmarkers to distinguish marine vs. continental deposition and to quantify thesignificance of recycling in evaporite basins. Sedimentary Geology 137, 213–240.

Tent-Manclus, J.E., Soria, J.M., Estevez, A., Lancis, C., Caracuel, J.E., Dinares-Turell, J.,Yebenes, A., 2008. Tortonian Salinity Crisis and tectonics: a view from theFortuna Basin (SE Spain). Comptes Rendus Geoscience 340, 474–481.

Tjalsma, R.C., 1971. Stratigraphy and Foraminifera of the Neogene of the EasternGuadalquivir Basin, S. Spain. Utrecht Micropaleontological Bulletin 4, 161.

Zachariasse, W.J., 1979. The origin of Globorotalia conomiozea in the Mediterraneanand the value of its entry level in biostratigraphic correlations. Annales Geo-logiques des Pays Helleniques 3, 1281–1292.