Radiogenic isotopes for deciphering terrigenous input provenance in thewestern Mediterranean

M. Rodrigo-Gámiz a,!, F. Martínez-Ruiz a, M. Chiaradia b, F.J. Jiménez-Espejo a,c, D. Ariztegui ba Instituto Andaluz de Ciencias de la Tierra (IACT), CSIC-Universidad de Granada, Granada, Spainb Department of Earth Sciences, University of Geneva, Geneva, Switzerlandc Japan Agency for Marine-Earth Science and Technology, Yokosuka, Japan

a b s t r a c ta r t i c l e i n f o

Article history:Received 15 January 2015Received in revised form 4 June 2015Accepted 5 June 2015Available online 11 June 2015

Keywords:Marine sedimentsWesternmost MediterraneanLast Glacial MaximumRadiogenic isotopesTerrigenous provenanceTransport mechanisms

Radiogenic isotopic signatures in marine sediments can be used to trace terrigenous source areas and transportmechanisms, which are in turn related to climate variability. To date, most of the published studies using thisapproach have been focused on eastern Mediterranean sediments. In contrast, we study here the terrigenousinput provenance in the westernmost Mediterranean (Alboran Sea basin) by using radiogenic isotope proxiesand Nd model ages in a marine record spanning the last 20 ka. Nd, Sr and Pb isotopes, obtained fromcarbonate-free samples from the b37 !msize fraction,were used to characterize terrigenous variations, includingeolian input. Substantial shifts in Pb isotopic signatures throughout the studied time interval reveal a change fromNorth African dominated sources during the glacial period to European dominated sources during the Holocene.Nd and Sr shifts likewise indicate twomain short-term changes in sediment provenance, during the last Heinrichevent and the early–middle Holocene transition (ca. 8.9 ka cal. BP). Nd model ages over 1.45 Ga also support acontribution of an older component in the terrigenous source, likely Archaean material from the presentSenegal region, during both periods. Conversely, terrigenous material mainly shows a dominant provenancefrom present-day Morocco, Mali, Mauritania, Niger, and Algeria, mixed with material from southern Iberia andsouthern France. Source variations in the westernmost Mediterranean were mainly driven by !uctuations inwind intensity and !uvial discharges. These !uctuations seem to have been modulated by the African monsoonsystem further conditioned by the ITCZ migrations and the position of the North Atlantic anticyclone system.

© 2015 Elsevier B.V. All rights reserved.

1. Introduction

The composition of terrigenous constituents in sediments of marginalmarine basins mainly derives from riverine and eolian inputs,thus re!ecting climatic conditions over adjacent continental regions(e.g., Kolla et al., 1979; Jeandel et al., 2007). In the case of theMediterranean region, its proximity to the Sahara desert makes thisarea a key location for tracking terrigenous provenance and variationsin source areas. In the western Mediterranean, the main sources ofterrigenous material are the eolian fraction from arid and semi-aridregions in North Africa, the suspended !uvial particles from thesouthwestern Europe and North African runoff, and the suspendedparticulate matter from the Atlantic and Mediterranean waters(Grousset et al., 1988; Bergametti et al., 1989a, 1989b; Loÿe-Pilot andMartin, 1996; Molinaroli, 1996; Prospero, 1996; Stumpf et al., 2011).The Western Sahara dry land in North Africa is recognized as a majoreolian dust source (Prospero, 1999; Goudie and Middleton, 2001;

Stuut et al., 2009; Prospero and Mayol-Bracero, 2013). Numericalsimulations conclude that North Africa is the largest single sourceof dust on Earth, with up to 8 ! 1014 g/a of atmospheric mineral dust,providing 50–70% of total dust emissions (Goudie and Middleton,2001; Laurent et al., 2008). Extrapolating this "gure to the total westernMediterranean basin area (840,000 km2) yields an average total atmo-spheric !ux of 10.9 ± 0.6 ! 1012 g/a (Bergametti et al., 1989a, 1989b;Loÿe-Pilot et al., 1989). The most important source areas of Saharandust are located in Western Sahara, Mauritania and Senegal, northernMali, Atlas Mountains through Morocco, Algeria and Tunisia, centraland eastern Libya, western Chad (Bodélé depression), southern Egypt,and northern Sudan (Molinaroli, 1996; Moreno et al., 2006; Laurentet al., 2008; Stuut et al., 2009; Scheuvens et al., 2013). Dust source areasover Africa have changed over the past depending on the boundaries ofdifferent wind systems, their intensities, and the palaeo-positioning andmigrations of the Inter-Tropical Convergence Zone (ITCZ) in responseto factors such as orbital-induced summer insolation or changes incross-equatorial temperature gradients (e.g., Nicholson, 2009; McGeeet al., 2014). Furthermore, at the scale of Quaternary climatic oscillations,eolian dust !uxes to the ocean may have been higher during stadial(Greenland Stadial, GS) than interstadial periods (Greenland Interstadial,GI: nomenclature based on INTIMATE group, Lowe et al., 2008), as the

Chemical Geology 410 (2015) 237–250

! Corresponding author at: Avda. de Las Palmeras 4, 18100 Armilla, Granada, Spain.E-mail addresses: martarodrigo@ugr.es (M. Rodrigo-Gámiz), fmruiz@ugr.es

(F. Martínez-Ruiz), Massimo.Chiaradia@unige.ch (M. Chiaradia), fjjspejo@jamstec.go.jp(F.J. Jiménez-Espejo), Daniel.Ariztegui@unige.ch (D. Ariztegui).

http://dx.doi.org/10.1016/j.chemgeo.2015.06.0040009-2541/© 2015 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Chemical Geology

j ourna l homepage: www.e lsev ie r .com/ locate /chemgeo

result of a general southward displacement of the ITCZ during borealsummer (e.g., Reader et al., 1999; Elenga et al., 2000). At present, therelative amount of dust input from the Sahara and Sahelian regions tothe western Mediterranean is low and occurs in winter (Bergamettiet al., 1989b; Dulac et al., 1992), while in the eastern Mediterranean it isfavoured during stronger summer westerlies (Jilbert et al., 2010). Theterrigenous particles de!ated from the surface depend on several factorsincluding wind speed, atmospheric instability, height of the source area,particle size, particle exposure, soil moisture, vegetative cover, andmineralogical composition (e.g., deMenocal, 1995, 2004; Morenoet al., 2006; Mulitza et al., 2008).

The determination of terrigenous particle provenance, of both eolianor riverine input, in Mediterranean marine sediments has, in general,been based on geochemical and mineralogical data (e.g., Coude-Gaussenet al., 1987; Foucault and Mélières, 2000; Wehausen and Brumsack,2000; Caquineau et al., 2002; Weldeab et al., 2002a, 2002b, 2003;Díaz-Hernández et al., 2011; Formenti et al., 2011) as well as on remotesensing methods, analysis of surface dust observations, back-trajectoryanalysis, and the use of mineral tracers (e.g., Goudie and Middleton,2001; Laurent et al., 2008). Radiogenic isotopes (Sr, Nd, Pb) are likewisepowerful tracers for identifying and characterizing source areas ofterrigenous material, which may in turn give us additional informationabout the provenance and transport mechanisms (e.g., Grousset et al.,1988, 1992, 1998; Revel et al., 1996; Tütken et al., 2002; Grousset andBiscaye, 2005; Jullien et al., 2007; Cole et al., 2009; Box et al., 2011;Meyer et al., 2011; Stumpf et al., 2011; Scheuvens et al., 2013).

Previous research in a West–East Mediterranean transect character-ized the spatial distribution of detrital !ux and terrigenous provenancein Late Pleistocene andHolocene sediments, obtaining asmain radiogenicend members Saharan dust and Nile particulate matter (e.g., Krom et al.,1999a, 1999b;Weldeab et al., 2002a, 2002b, 2003; Revel et al., 2010; Boxet al., 2011; Blanchet et al., 2013). In the westernmost Mediterraneanarea, however, less work has focused on the identi"cation of terrigenousprovenance and transport patterns. Thus, here we provide a novel anddetailed characterization of the terrigenous material and eolian inputprovenance using radiogenic signatures in marine sediments from asediment record in the Alboran Sea basin. Previous mineralogical,geochemical and sedimentological analyses from this record have beenused to reconstruct the climate variability in terms of atmospheric andoceanic responses as well as to identify eolian input variations since theLast Glacial Maximum (LGM) (Rodrigo-Gámiz et al., 2011, 2014a,b).However, the terrigenous provenance had not been described yet. Inthis study, we use Sr, Nd, and Pb isotopes and Nd model ages (TDM) toconstrain for the "rst time the geographic provenance of terrigenousmaterial and eolian dust, and their transport mechanisms during thelast 20 ka.

1.1. Radiogenic isotopes as proxies for unravelling chemical weathering orterrigenous source

The variability in radiogenic isotope composition of rocks, whichare ultimately the source of the particulate matter suspended andtransported by winds or rivers, is essentially the result of chemicalfractioning between radioactive parents and radiogenic daughtersoperated by large-scale geological and ageing processes (Frank, 2002).Consequently, rocks of different ages and provenance in terms of largetimescale reservoirs, e.g. mantle vs. crust, have signi"cantly andmeasurably different isotopic compositions. The isotope composition ofthe terrigenous material that is the weathering product of the sourcerocksmay coincide or notwith that of the source rock, as discussed below.

Nd is a relatively immobile element during weathering andtherefore the Nd isotope composition in weathering products isstable during changes from wet to dry periods (e.g., Nesbitt et al.,1980; Dickin, 1997; Braun et al., 1998; Frank, 2002). In contrast,the Sr isotope composition is associated with changes in weatheringintensity (Nesbitt et al., 1980; Frank, 2002), and thus radiogenic Sr is

preferentially removed from the source region, leaving a residuewith low Sr isotope ratios (Blum and Erel, 1997). During prolongedchemical weathering the radiogenic Sr fraction of a rock is extractedfrom the source faster than the non-radiogenic Sr fraction, due to thepreferential breakdown of Rb-rich phases such as mica and K-feldspar(Nesbitt et al., 1980; Frank, 2002). Therefore the combination ofSr–Nd isotope variations could re!ect two different signatures, i.e.changes in the source area, and intense chemical weathering periodsinvolving enhanced rainfall and riverine runoff (Frank, 2002). Theextent of chemical weathering is strongly related to prevailing environ-mental conditions in the source area, such as cover vegetation or aridity.Furthermore, while the Nd isotope signature is unaffected by grain sizevariations (cf. Goldstein et al., 1984; Grousset et al., 1992), Sr isotopeshave shown a relationship with grain size, i.e. Sr isotope ratios increasewith decreasing grain size, as a result of the high Rb/Sr in "ner sediments(cf. Biscaye and Dasch, 1971; Wehausen and Brumsack, 2000).

Additionally, the Pb isotope composition provides information onsource provenances since natural Pb can be transported in the detritalmaterial or in eolian dust for long distances (Grousset et al., 1994;Grousset and Biscaye, 2005; Kylander et al., 2005). However, anthropo-genic Pb contamination can overprint the natural Pb isotope signal(Grousset et al., 1994; Kylander et al., 2005). A particular advantage ofthe Th–U–Pb system is that binary mixtures form straight linear arraysin Pb–Pb isotope space, and deviations from such arrays implymixturesinvolving more than two components. Therefore, linear Pb isotopearrays are consistent with binary mixing and imply the existence ofmultiple and different contributions of Pb sources.

Further information can be obtained from the Nd model ages, TDM,which provide an isotopic "ngerprint of the crustal source, in terms oftiming and processes of crust formation (Arndt and Goldstein, 1987).Since the investigated sediments are a mixture of terrigenous materialderived from different source areas, i.e. North Africa and South Europe,we expect to obtain “mixed” Nd model ages, which lie between themodel ages of the surrounding potential source materials (Fig. 1a).

1.2. Present-day climate over the westernmost Mediterranean

Modern climate conditions over the western Mediterranean andnorthwestern Africa areas are governed by the Azores high-pressuresystem linked to the North Atlantic climate variability and by Africanmonsoonal dynamics. Summertime climates are usually dry and hotdue to the in!uence of the atmospheric subtropical high-pressure belt(Sumner et al., 2001). During winter the subtropical high shifts to thesouth, allowing mid-latitude storms to enter the region from the openAtlantic and bringing enhanced amounts of rainfall to the westernMediterranean. This humidity regime in the western Mediterranean ismainly modulated by the North Atlantic Oscillation (NAO) (Hurrell,1995; Trigo et al., 2002). A high (positive) NAO index causes a morenortherly position of the North Atlantic depression, stronger thanusual westerlies and warm and wet winters over Northern Europeand dry and cold winters over southern Europe, the Mediterraneanand northern Africa (Wanner et al., 2001; Trigo et al., 2002). Conversely,a low (negative) mode of the NAO leads to opposite conditions.

In addition, local climatology of Northern Africa is seasonallymodulated by the latitudinal shift of the ITCZ and hence the Africanmonsoon front. The ITCZ directly controls the location of precipitationover North Africa; changes in its position affect how much rainfalloccurs and therefore the degree of river runoff. The review by Nicholson(2009) suggests a complex ITCZ pattern, especially for the boreal summersituation. In winter, the equator-ward displacement of the ITCZ (10°N)(Fig. 1a) causes a southward shift of dry subtropical air masses and isassociated with the development of strong easterly Saharan Air Layer(SAL) winds, with a northern branch (NSAL). Consequently, a dustplume, usually generated by low-pressure systems, is transportedbetween 15° and 25°N along an E–W axis over the tropical AtlanticOcean (Holz, 2004) and over the Mediterranean (Moulin et al., 1997).

238 M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

During boreal summer, the dry subtropical air shifts northward and theITCZ is located around 20°N (Fig. 1a), representing the onset of therainy season (summer monsoon) in North Africa with heavy rainfalland changes in atmospheric circulation (Peyrillé and Lafore, 2007).Complex parameters and mechanisms actually modulate precipitationover West Africa during summer (Nicholson, 2009). For instance, awarming in the Mediterranean Sea basin could favour the northwardexpansion of the monsoon in summer (Rowell, 2003; Hall and Peyrillé,2006).

2. Material and analytical techniques

2.1. Marine setting, sediment record description and sample selection

The Alboran Sea, located in thewesternmostMediterranean Sea, is asemi-closed basin surrounded by the Iberian Peninsula in the North andtheNorth Africamargin in the South. A gravity core (293G)was taken inthe East Alboran basin (Fig. 1a) (36°10.414!N, 2°45.280!W, depth1840 m, length 402 cm) during the oceanographic cruise Training

8.2k

a

7.4k

a PC

AI

YDAller d gnill

B

H1 LGM

0 2 4 6 8 10 12 14 16 18 20

lA/r

Z

b)

Algeria

Spain

France Switzerland

293G

Ebro

Rhône

Guadalquivir

Libya

Egypt

GuadalfeoGuadalhorce

Nd

DM

LanguedocTDM

(0)Nd

Nd

TDM87 86

Mauritania MaliNiger

Chad Sudan

Guinea

winter-January

0º

30ºN

20ºN

10ºN

40ºN

30ºN

20ºN

10ºN

40ºN

10ºE10ºW20ºW 20ºE

Atlas MontainsMorocco

WesternSahara

summer-JulySenegal

DM

TDM

Algeria-NigeriaTDM

TDM

a)

206 204

207 204

208 204

208 206

206 204

207 204

208 204

Têt

Nd

87 86

Tunisia

Nd

87 86

Nd

87 86

Nd

87 86

Nd

87 86

Nd

87 86

Guadiana

26

24

22

20

18

16

14

TH

ME

GS-1 GI-1 GS-22b1c 1e1a 1b 1d 2a

0 2 4 6 8 10 12 14 16 18 20

Holocene

Fig. 1. a) Map of the studied area showing the location of the marine sediment record 293G retrieved in the westernmost Mediterranean. Potential terrigenous source areas from NorthAfrica, southern Iberia and southwestern Europewith radiogenic signatures and TDM compiled from the literature are included. Dashed red and blue lines indicate the summer (July) andwinter (January) positions of the ITCZ, respectively; b) Zr/Al pro!le fromRodrigo-Gámiz et al. (2011) as proxy for eolian inputwas used to select 37 samples for radiogenic measurementscovering the last 20 ka. (For interpretation of the references to color in this !gure legend, the reader is referred to the web version of this article.)

239M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

Through Research-12 (R/V Professor Logachev) (Comas and Ivanov,2003).

The selected sediment record contains homogeneous green-brownishhemipelagicmud-clays with some nannofossils. High-resolution geo-chemical, mineralogical and sedimentological analyses at about 75 a/1.5 cm (n = 267 samples) from this sediment record have been re-ported in Rodrigo-Gámiz et al. (2011). Conventional X-Ray Fluores-cence (XRF) was used to obtain the elemental geochemistry ofmajor elements and Zirconium (Zr). A previously developed agemodel based on linear interpolation between ten calibrated AMS14C-dates (see Rodrigo-Gámiz et al., 2011, for a description) and twolater dates (Rodrigo-Gámiz et al., 2014b)during critical intervals indicatesthat thismarine core spans the last 20 kawith amean sedimentation rateof ca. 20 cm/ka.

A total of 37 samples of potential eolian material were selectedcovering the last 20 ka on the basis of eolian proxies such as Zr/Alratio (Fig. 1b), quartz and palygorskite contents and Si/Al and Ti/Alratios (Supplementary Fig. S1; Rodrigo-Gámiz et al., 2011). Becausechanges in grain size can have a substantial in!uence in Sr-isotopecompositions used as provenance indicators and in order to avoid anyhemipelagic contribution, we consistently worked with the samegrain size fraction. Based on the fact that most of the potential Saharandust particles are "ne silts and clay-sizeminerals (Grousset et al., 1998;Grousset and Biscaye, 2005; Stuut et al., 2009) and the grain-sizedistribution of this material showed mean sizes below 6 !m (Fig. 3d)(Rodrigo-Gámiz et al., 2011), in this study we investigated the "ne siltfraction, i.e. b37 !m.

2.2. Sediment digestion procedure, element separation and isotope analysis

Chemical extractions for Sr, Nd and Pb isotopes were carried out atthe Department of Earth Sciences of the University of Geneva(Switzerland). Each sample was treated to leach the carbonate fractionwith 1 N acetic acid solution followed by a triple rinse with water. TheN37 !m fraction was removed by dry sieving. Around 100–150 mg ofthe dried alumino-silicate residual fraction was mineralized in pureacids using Te!on bombs heated on an electric hotplate (140 °C) in atwo-step procedure: at "rst adding a mixture of 1 ml HNO3 15 M with4 ml HF at 140 °C for 7 days, and secondly after evaporation adding3 ml HNO3 15 M at 140 °C for 2 days. After evaporation, the sampleswere "nally diluted in 1.9 ml HNO3 1 M for chemical separation.

Sr, Nd and Pb separations from the prepared solutions were carriedout using cascade columns with Sr-Spec, TRU-Spec and Ln-Spec resinsfollowing a modi"ed method after Pin et al. (1994). Pb was furtherpuri"ed by anion exchange chromatography using an AG-MP1-M cleanresin in hydrobromic medium and small volume columns (0.08 ml).

Pb was loaded on Re "laments using the silica gel technique(Gerstenberger and Haase, 1997) and Pb isotope ratios of all samplesand standards were measured in static mode on Faraday cups on amulticollector Thermo TRITON mass spectrometer at a pyrometercontrolled temperature of 1220 °C. Pb isotope ratios were correctedfor instrumental fractionation by a factor of 0.1% per amu based onmore than 100 measurements of the SRM981 standard and using thestandard values of Todt et al. (1996). Procedural blanks were b200 pg.External reproducibilities (2") of the standard ratios are 0.05% for206Pb/204Pb, 0.08% for 207Pb/204Pb, 0.10% for 208Pb/204Pb, 0.006% for206Pb/207Pb, 0.007% for 208Pb/207Pb and 0.008% for 208Pb/206Pb.

Sr was loaded on single Re "laments with a Ta oxide solution andmeasured on the multicollector Thermo TRITON mass spectrometer ata pyrometer controlled temperature of 1480 °C in static mode usingthe virtual ampli"er design to cancel out biases in gain calibrationamong ampli"ers. 87Sr/86Sr values were internally corrected forfractionation using a 88Sr/86Sr value of 8.375209. Raw valueswere furthercorrected for external fractionation by a value of+0.03‰, determined byrepeated measurements of the SRM987 standard (87Sr/86Sr = 0.710250;

McArthur et al., 2001). External reproducibility (1") of the SRM987standard is b7 ppm.

Nd was loaded on double Re "laments with 1 M HNO3 andmeasured on a 7-collector Finnigan MAT 262 thermal ionizationmass spectrometer with extended geometry and stigmatic focusing.143Nd/144Nd ratios were measured in dynamic mode (quadruplecollector) and internally corrected for fractionation using a 146Nd/144Ndvalue of 0.721903 and corrected for external fractionation usingthe JNdi1 standard (143Nd/144Nd = 0.512115 ± 7; Tanaka et al.,2000). Our mean of 19 replicated analyses of this standard indynamic mode was 0.512097 ± 3 ! 10!6 (2") during the periodof analysis. The 143Nd/144Nd ratio is commonly expressed as#Nd(0) = [(143Nd/144Nd)sample " (143Nd/144Nd)CHUR ! 1] ! 104

with the present day Chondritic Uniform Reservoir (CHUR) being0.512638 (Wasserburg et al., 1981).

2.3. Nd model ages

Additionally, another portion (around 100–150 mg) of 35 out of37 dried alumino-silicate residual sediments was used to calculatethe distribution of Nd model ages relative to depleted mantlemodel ages, TDM, according to DePaolo (1981) (147Sm/144Nd =0.21378 and 143Nd/144Nd=0.513155). The alumino-silicate residualsamples were acid leached using a digestion of HNO3 + HF inPFA-lined pressure vessels in a microwave "eld following the methodby Montero and Bea (1998). The 147Sm/144Nd ratio was determinedby inductively coupled-plasma mass spectrometry (ICP-MS) in theCentre for Scienti"c Instrumentation (CIC) at the University of Granada(Spain). The ICP-MS was a quadrupole-based Perkin-Elmer SciexELAN-5000a equipped with a conventional cross-!ow nebulizer andspray-chamber. Measurements were taken in triplicatewith a precisionbetter than 1.2% (2") and usingRb, Sr, Sm,Ndand Rh standard solutionsof various concentrations as internal standards.

3. Results

3.1. Sr, Nd and Pb isotope composition

The Sr and Nd isotope values, Sm andNd concentrations, 147Sm/144Ndratio and TDM of the carbonate-free b37 !m sediment fraction are shownin Table 1, and the Pb isotope values are presented in Table 2.

Nd model ages in the marine sediment yield a nearly Gaussiandistribution with the most frequent TDM value centred at 1.4 Ga with aSD of 0.5 (Fig. 2). The temporal evolution of Nd model ages for the last20 ka in the westernmost Mediterranean shows the geological agevariations of the source regions (Fig. 3a). Maximum TDM values of 1.48and 1.60 Ga are reached at 16.2 and 10.2 ka cal. BP respectively, followedby 1.46 Ga recorded at 8.9 and 8.7 ka cal. BP. Minimum TDM values ca.1.34–1.36 Ga are observed at 14.3, 14.1, 13.3, 11.1, 9.4 and 3.4 ka cal. BP(Fig. 3a).

The Nd isotope signature ranges between #Nd(0) !11.7 and !10.5(Fig. 3b) with more negative #Nd(0) values, i.e. less radiogenic143Nd/144Nd ratio, at 17.6, 16.2–16.0, 14.6, 8.9–8.7, 5.6 and 2.2 ka cal.BP (Fig. 3b). Some of these shifts in the #Nd(0) signature are alsoevidenced in the 87Sr/86Sr ratio, showing several periods with lessradiogenic values recorded at 18.7, 17.1–16.7, 14.3–14.1, 13.5, 13.1, 10.2,8.9, 5.6 and 1.1 ka cal. BP (Fig. 3c).

The Pb isotope composition, in particular 206Pb/204Pb, 207Pb/204Pb,and 208Pb/204Pb ratios, shows almost the same variability pattern, rangingfrom 18.8275 to 18.6929, 15.7547 to 15.5947, and 39.0807 to 38.6815,respectively (Fig. 4a–c). Higher Pb isotope ratio values during the last20 ka are observed at 17.6, 13.5, 11.1 and 4.3–3.4 ka cal. BP (Fig. 4a–c).Conversely, lower values of Pb isotope ratios are recorded at 18.7, 16.7,12.8, 11.5, 8.9 and 2.2 ka cal. BP (Fig. 4a–c). These latest periods are alsorepresented by less negative values of #Nd(0), i.e. more radiogenic143Nd/144Nd ratio, and lower values of 87Sr/86Sr (Fig. 3b–c). Finally, the

240 M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

207Pb/206Pb values range between 0.83788 and 0.83286, the208Pb/206Pb ratio shows variations between 2.07794 and 2.06768,and the 208Pb/207Pb ratio between 2.48654 and 2.47854 (Fig. 4d–f),displaying more radiogenic signatures at 18.7, 16.0, 13.5, 11.1, 9.4and 4.3–2.2 ka cal. BP.

4. Discussion

4.1. Radiogenic isotope variations in the westernmost Mediterraneanduring the last 20 ka

Radiogenic isotopes have provided an additional tool for recon-structing palaeoenvironmental conditions in the westernmostMediterranean. For the last 20 ka, the 87Sr/86Sr ratio showed lowvalues during the LGM, at 18.7 ka cal. BP, and the last Heinrichevent (H1), ca. 16.7–16.2 ka cal. BP (Fig. 3c). This could suggest therelease of radiogenic Sr during chemical weathering, leaving a lowradiogenic signature in the source rock (Blumand Erel, 1997). However,during the Greenland Stadial 2 (GS-2, from 20 to 14.7 ka cal. BP)generally arid conditions and semi-desert vegetation cover prevailedover the continental areas surrounding the western Mediterraneanbasin (e.g., Fletcher and Sánchez Goñi, 2008; Combourieu Neboutet al., 2009), and no prolonged weathering processes have been regis-tered. In addition, contemporaneous Nd and Pb isotopic !uctuations(Figs. 3b, 4) suggest that the Sr isotope signature indeed re!ectsvariations in the terrigenous provenance rather than a weatheringsignal. The Sr isotope signature and !Nd(0) values for these period arecomparable with the signal of Sahara material (Jeandel et al., 2007;Abouchami et al., 2013). Although TDM model ages do not show major

shifts until 16.2 ka cal. BP (with a small peak value of ca. 1.48 Ga;Fig. 3a), the source variation in the terrigenous material is furthersupported by the increased values of eolian proxies such as Zr/Al, Si/Al,and Ti/Al ratios, major quartz content and the presence of palygorskite(Supplementary Fig. S1; Rodrigo-Gámiz et al., 2011), which is inagreement with cold and arid conditions during both time periods(Bout-Roumazeilles et al., 2007; Jullien et al., 2007). Model resultsand palaeoclimate records have documented southward shifts ofthe ITCZ during boreal summer with an increase in eolian dust inputduring these cold periods (e.g. Steager et al., 2011; Arbuszewski et al.,2013; McGee et al., 2013, 2014).

Thereafter, the Bølling–Allerød (B–A) period (from ca. 14.7 to12.9 ka cal. BP) showed rapid shifts to likely weathered conditionswith preferential leaching of radiogenic Sr at 14.3–14.1, 13.6, and13.1 ka cal. BP (Fig. 3c). Assuming the onset of the African Humid Period(AHP) in North Africa around 14.5 ka cal. BP (deMenocal et al., 2000),the progressive increase of tropical vegetation and major humid condi-tions over northern African and southern European areas would havefavoured major weathering and the release of radiogenic Sr from theterrigenous source. Typical !Nd(0) values from!uvial sediments derivedby the RhôneRiver in thewesternMediterranean range between!10.8and !9.7 (Henry et al., 1994), whereas the typical average Saharan–Sahelian isotopic !Nd(0) signature ranges from !14 to !11 (seesynthesis in Scheuvens et al., 2013), and the average Europeanatmospheric input has !Nd(0) values between!12.2 and!10.2 (Henryet al., 1994). The !Nd(0) values at 14.3 and 13.3 ka cal. BP (!11.0 and!10.9; Fig. 3b) are close to both the riverine input signature and theEuropean atmospheric input (Henry et al., 1994), suggesting that theB–A period was characterized by a mixture of terrigenous material.

Table 1Sr concentrations, 87Sr/86Sr and 143Nd/144Nd ratios, !Nd(0) values, Sm and Nd concentrations, 147Sm/144Nd ratio and TDM on the carbonate-free fraction below 37 "m from themarine core293G. 147Sm/144Nd = 0.21378 and 143Nd/144Nd = 0.513155 value ratios used to calculate TDM according to DePaolo (1981).

Samples Depth (cm) Age (ka cal. BP) Sr (ppm) 87Sr/86Sr 1# " 10!6 143Nd/144Nd 1# " 10!6 !Nd(0) Sm (ppm) Nd (ppm) 147Sm/144Nd TDM (Ga)

293G 1 16.5–18 17.25 1.1 449.325 0.716771 2 0.512088 2 !10.7 4.99 27.82 0.108 1.38293G 1 34.5–36 36.75 2.2 565.668 0.717964 2 0.512068 3 !11.1 3.92 22.26 0.106 1.38293G 1 55.5–57 56.25 3.4 610.58 0.717332 2 0.512081 2 !10.9 4.85 27.7 0.104 1.34293G 2 12–13.5 70.75 4.3 557.265 0.717776 2 0.512096 2 !10.6 4.71 26.46 0.108 1.38293G 2 33–34.5 91.75 5.6 538.196 0.716461 2 0.512066 4 !11.2 4.77 26.66 0.108 1.43293G 2 54–55.5 112.75 6.8 581.792 0.717616 2 0.512087 8 !10.7 4.60 26.06 0.107 1.38293G 3 16.5–18 133.75 8.0 628.359 0.717816 5 0.512077 2 !10.9 4.97 27.39 0.110 1.43293G 3 28.5–30 145.75 8.7 652.7198 0.717795 2 0.512037 8 !11.7 4.66 26.10 0.108 1.46293G 3 33–34.5 150.25 8.9 642.647 0.714397 2 0.512041 6 !11.6 4.27 23.96 0.108 1.46293G 3 42–43.5 159.25 9.4 651.014 0.717152 3 0.512099 16 !10.5 4.41 25.04 0.106 1.36293G 4 6–7.5 181.25 10.2 645.088 0.716134 1 0.512095 23 !10.6 3.92 21.48 0.110 1.41293G 4 28.5–30 203.75 10.8 581.759 0.717591 2 0.512071 14 !11.1 3.07 15.43 0.120 1.60293G 4 42–43.5 217.25 11.1 565.881 0.717651 2 0.512078 2 !10.9 3.90 22.65 0.104 1.36293G 5 0–1.5 232.75 11.5 553.416 0.717110 3 0.512059 7 !11.3 4.39 25.35 0.105 1.39293G 5 12–13.5 244.75 12.0 556.243 0.717639 2 0.512057 6 !11.3 4.26 24.16 0.107 1.42293G 5 22.5–24 255.25 12.7 582.626 0.716945 3 0.512068 1 !11.1 3.91 22.65 0.104 1.37293G 5 30–31.5 262.75 12.8 596.618 0.716280 2 0.512062 7 !11.2 4.25 23.57 0.109 1.44293G 5 39–40.5 271.75 13.1 581.398 0.715134 3 0.512068 1 !11.1 4.22 23.59 0.108 1.44293G 5 46.5–48 279.25 13.3 561.848 0.717187 3 0.512079 7 !10.9 2.79 16.38 0.103 1.34293G 5 54–55.5 286.75 13.5 588.682 0.716166 1 0.512064 4 !11.2 4.03 22.56 0.108 1.43293G 6 1.5–3 292.25 13.6 579.871 0.717347 3 0.512052 18 !11.4 4.14 23.60 0.106 1.42293G 6 7.5–9 298.25 13.8 543.292 0.717056 3 0.512064 1 !11.2 – – – –293G 6 12–13.5 302.75 13.9 526.542 0.716427 2 0.512053 6 !11.4 4.42 25 0.107 1.43293G 6 19.5–21 310.25 14.1 503.957 0.715899 2 0.512062 7 !11.2 4.03 23.68 0.103 1.36293G 6 28.5–30 319.25 14.3 445.874 0.716152 2 0.512076 2 !11.0 4.39 25.81 0.103 1.34293G 6 34.5–36 325.25 14.6 429.905 0.717414 2 0.512060 25 !11.3 4.04 22.65 0.108 1.43293G 6 39–40.5 329.75 15.2 403.880 0.717045 4 0.512080 2 !10.9 3.73 20.82 0.108 1.41293G 6 48–49.5 338.75 16.0 385.348 0.716053 2 0.512047 5 !11.5 ! ! ! !293G 6 51–52.5 341.75 16.1 425.231 0.716376 3 0.512062 1 !11.2 4.49 24.92 0.109 1.44293G 6 52.5–54 343.25 16.2 456.444 0.715867 2 0.512044 3 !11.6 4.33 23.79 0.110 1.48293G 7 1.5–3 351.25 16.7 447.815 0.713110 3 0.512067 1 !11.1 4.07 23.73 0.104 1.37293G 7 7.5–9 357.25 17.1 423.367 0.715785 2 0.512057 8 !11.3 4.44 25.11 0.107 1.42293G 7 15–16.5 364.75 17.6 375.639 0.717949 2 0.512041 9 !11.6 3.82 22.19 0.104 1.41293G 7 18–19.5 367.75 17.8 414.314 0.717095 3 0.512060 2 !11.3 4.02 23.29 0.104 1.38293G 7 24–25.5 373.75 18.2 539.771 0.718042 2 0.512057 26 !11.3 2.71 15.40 0.106 1.42293G 7 33–34.5 382.75 18.7 641.713 0.714939 2 0.512074 3 !11.0 3.19 18.06 0.107 1.40293G 7 40.5–42 390.25 19.2 603.560 0.717840 2 0.512060 14 !11.3 4.13 24 0.104 1.38

241M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

Table2

Pbisotop

icda

tameasu

redon

thecarbon

ate-free

fractio

nbe

low

37!m

from

themarineco

re29

3G.2

06Pb

/204Pb

,207Pb

/204Pb

,208Pb

/204Pb

,207Pb

/206Pb

,208Pb

/206Pb

,and

208 Pb/

207 Pbratio

san

dtheirresp

ectiv

eerrors.

Samples

Dep

th(cm)

Age

(kacal.BP

)20

6Pb

/204Pb

1"!10

"4

207Pb

/204Pb

1"!10

"4

208Pb

/204Pb

1"!10

"4

207Pb

/206Pb

1"!10

"5

208Pb

/206Pb

1"!10

"5

208Pb

/207Pb

1"!10

"5

293G

116

.5–1

817

.25

1.1

18.733

0–

15.688

0–

38.909

0–

0.83

754

–2.07

703

–2.48

018

–29

3G134

.5–3

636

.75

2.2

18.694

71

15.664

02

38.823

95

0.83

788

02.07

673

12.47

854

129

3G155

.5–5

756

.25

3.4

18.803

974

15.723

162

39.057

215

80.83

607

42.07

706

132.48

395

1029

3G212

–13.5

70.75

4.3

18.803

017

615

.712

814

739

.069

535

60.83

572

82.07

794

142.48

654

1829

3G233

–34.5

91.75

5.6

18.777

64

15.669

45

38.887

914

0.83

447

12.07

094

32.48

174

229

3G254

–55.5

112.75

6.8

18.773

710

415

.679

587

38.901

621

40.83

512

42.07

190

102.48

079

1429

3G316

.5–1

813

3.75

8.0

––

––

––

––

––

––

293G

328

.5–3

014

5.75

8.7

18.763

23

15.668

53

38.876

310

0.83

507

12.07

193

32.48

117

229

3G333

–34.5

150.25

8.9

18.742

91

15.667

31

38.864

93

0.83

591

02.07

359

12.48

064

129

3G342

–43.5

159.25

9.4

18.756

471

15.683

858

38.927

814

40.83

620

32.07

553

62.48

211

929

3G46–

7.5

181.25

10.2

18.770

33

15.682

83

38.929

37

0.83

551

02.07

395

12.48

228

129

3G428

.5–3

020

3.75

10.8

18.787

918

15.677

815

38.932

837

0.83

444

12.07

216

32.48

328

429

3G442

–43.5

217.25

11.1

18.823

012

415

.754

710

739

.080

726

40.83

701

62.07

673

222.48

085

1429

3G50–

1.5

232.75

11.5

18.752

420

15.649

216

38.834

341

0.83

457

12.07

092

22.48

140

329

3G512

–13.5

244.75

12.0

18.779

93

15.675

33

38.914

910

0.83

468

12.07

214

32.48

255

229

3G522

.5–2

425

5.25

12.7

18.757

069

15.663

257

38.891

414

30.83

506

32.07

334

62.48

275

729

3G530

–31.5

262.75

12.8

18.709

511

715

.612

094

38.749

023

70.83

442

42.07

104

72.48

191

1029

3G539

–40.5

271.75

13.1

––

––

––

––

––

––

293G

546

.5–4

827

9.25

13.3

18.735

82

15.667

52

38.847

44

0.83

623

02.07

342

12.47

947

129

3G554

–55.5

286.75

13.5

18.781

434

15.723

231

39.022

283

0.83

714

32.07

786

122.48

212

829

3G61.5–

329

2.25

13.6

18.769

93

15.668

13

38.873

510

0.83

475

12.07

105

32.48

105

229

3G67.5–

929

8.25

13.8

18.799

021

15.687

819

38.937

847

0.83

454

12.07

161

72.48

229

529

3G612

–13.5

302.75

13.9

18.768

52

15.666

62

38.871

75

0.83

473

02.07

109

12.48

116

129

3G619

.5–2

131

0.25

14.1

18.769

952

15.676

343

38.905

010

50.83

535

32.07

293

42.48

151

629

3G628

.5–3

031

9.25

14.3

––

––

––

––

––

––

293G

634

.5–3

632

5.25

14.6

18.752

53.1

15.670

33.1

38.870

69.3

0.83

564

02.07

285

22.48

054

129

3G639

–40.5

329.75

15.2

18.788

019

15.677

416

38.893

339

0.83

444

12.07

012

22.48

080

329

3G648

–49.5

338.75

16.0

18.780

320

15.672

317

38.876

339

0.83

448

12.06

999

32.48

060

329

3G651

–52.5

341.75

16.1

18.757

041

15.689

132

38.906

982

0.83

646

22.07

435

82.47

984

729

3G652

.5–5

434

3.25

16.2

18.755

72

15.660

32

38.843

55

0.83

496

02.07

102

12.48

039

129

3G71.5–

335

1.25

16.7

18.692

015

115

.594

712

438

.681

530

80.83

433

62.06

950

102.48

048

1529

3G77.5–

935

7.25

17.1

18.785

21.7

15.655

71.8

38.855

35.8

0.83

340

02.06

839

12.48

187

129

3G715

–16.5

364.75

17.6

18.827

579

15.687

266

38.953

216

40.83

323

42.06

894

62.48

311

929

3G718

–19.5

367.75

17.8

18.806

513

15.663

110

38.885

327

0.83

286

12.06

768

22.48

261

229

3G724

–25.5

373.75

18.2

18.785

077

15.661

065

38.874

216

10.83

375

32.06

953

62.48

228

829

3G733

–34.5

382.75

18.7

18.738

410

415

.641

189

38.793

621

80.83

464

42.07

029

72.48

050

929

3G740

.5–4

239

0.25

19.2

18.787

52.1

15.661

82.4

38.858

37.7

0.83

363

02.06

830

22.48

108

1

242 M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

The Younger Dryas (YD) cold period (from ca. 12.9–11.7 ka cal. BP)has been associated with a sharp increase in regional aridity thatinterrupted the AHP in subtropical African records (e.g., deMenocalet al., 2000; Gasse, 2000). In contrast, the Sr, Nd, and Pb isotope compo-sitions present almost constant values (Figs. 3b–c, 4). Previous geo-chemical and mineralogical data from this record (see SupplementaryFig. S1 and Rodrigo-Gámiz et al., 2011) showed a palaeoenvironmentalscenario involving a combinationof twophases, a!rst dry phase followedby a secondwithmorehumid conditions over poorly vegetated continen-tal areas. Palynological records from the western Mediterranean regionalso highlight an early period with very dry conditions and an increasein semi-desert taxa, followed by a more humid period with a slightincrease in forest vegetation (e.g., Combourieu Nebout et al., 2009).Similar dry–humid conditions have been reported for this period basedon marine records from offshore NW Africa, suggesting a shift of thenorthern limit of the African rain belt and associated wind systems(Meyer et al., 2011). However, these two phases are not re"ected in the87Sr/86Sr ratio, probably because it was dominated by sporadic runoffswithout suf!cient time to have a prolonged weathering effect. Therefore,the absence of substantial differences in the isotope composition and

1

2

3

4

5

6

0

d)

dN

(0)

1.301.351.401.451.501.551.601.651.70

)aG(

TM

D

c)

b)

a)

8786

0.7120.7130.7140.7150.7160.7170.7180.719

-9.5-10.0-10.5-11.0-11.5-12.0-12.5

GS-12b1a 1b 1d 2a

Holocene

8.2k

a

7.4k

a PC

AI

YDAller d gnill

B

H1 LGM

0

GI-1 GS-21c 1e

0 2 4 6 8 10 12 14 16 18 20

2 4 6 8 10 12 14 16 18 20

TH

ME

Fig. 3. 87Sr/86Sr ratio (a), !Nd(0) (b) and TDM (c) plotted against age. Error bars are smaller than symbol size if they are not shown. Data from African (Scheuvens et al., 2013) and Europeansources (Henry et al., 1994) are showed. Grey vertical bars show periods with less radiogenic 87Sr/86Sr values. Upper panel: light red vertical bars indicate the Allerød (A), and Bølling(B) warm time intervals. Light blue vertical bars indicate main cold periods, such as the last Heinrich event (H1), the Older Dryas and Younger Dryas (YD) time intervals, and the 8.2 ka coldevent. Short dashed red vertical bars show the early–middle Holocene transition (EMHT) at ca. 8.5–8.4 ka cal. BP and the demise of the African Humid Period (AHP) at 7.4 ka cal. BP. Lowerpanel: Greenland Stadials (GS-1, GS-2) and Interstadials (GI-1) and Holocene time intervals according to the event stratigraphy timing proposed by the INTIMATE group (Lowe et al., 2008).Black squares indicate twelve 14C-AMS dates with two-sigma probability interval using Calib 6.0.2 software (Stuiver and Reimer, 1993) and the Marine09 calibration curve (Reimer et al.,2009). (For interpretation of the references to color in this !gure legend, the reader is referred to the web version of this article.)

Fig. 2. Distribution of Nd model ages, TDM, in marine core 293G, calculated according toDePaolo (1981).

243M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

comparable signatures and Nd model ages to the B–A period (Fig. 3a–c)likely suggests similar terrigenous source areas.

During the Holocene, major Sr–Nd–Pb isotope excursions are datedat ca. 11.5, 10.2, 8.9–8.7, 5.6, 2.2 and 1.1 ka cal. BP (Figs. 3b–c, 4). Highand low TDM values of 1.60 Ga at 10.8 ka cal. BP and of 1.34 Ga at3.4 ka cal. BP (Fig. 3a), respectively, suggest variations in the sourcerock contributions, from pre-Pan-African and Pan-African and Iberianrocks. However, the Sr, Nd, or Pb isotope compositions did not showany substantial excursion at older TDM values around 10.8 ka cal. BP. Incontrast, previous mineralogical data and typically !uvial-derivedgeochemical ratios from this record have shown an increase in eolianinput signalled by higher palygorskite and quartz contents over!uvial-derived material at the early Holocene (Supplementary Fig. S1;Rodrigo-Gámiz et al., 2011).

The end of the AHP over northwestern Africa has been documentedby terrigenous records and Sr isotopes between 6.0 and 4.0 ka cal. BP(Gasse and Van Campo, 1994; Swezey, 2001; Gasse, 2002; Kuhlmannet al., 2004; Cole et al., 2009). However, the Holocene environmental

conditions in the western vs. eastern Africa have shown to be differentand complex. For instance, Sr isotope ratios from theArabianSea re!ecteda "rst aridi"cation step at 8.5 ka cal. BP followed by a second phase at6.0 ka cal. BP that ceased at 3.8 ka cal. BP when the modern-day dryclimate over North Africa was established (Jung et al., 2004). Moreover,a palaeolake in northern Kenya recorded a dry trend just before the8.2 ka cal. BP cooling event (Junginger et al., 2014 and references therein).In contrast, in western and northern Africa water-level variations weredescribed around 10.5 to 8.5–8.0 and 7.5–4.5 ka cal. BP (Gasse and VanCampo, 1994; Swezey, 2001; Gasse, 2002; Lézine et al., 2011).

In the Iberian Mediterranean region, pollen sequences re!ected asigni"cant environmental variability during the Holocene (Carrión,2002; Combourieu Nebout et al., 2009; Carrión et al., 2010; Fletcheret al., 2010). The major transition from a mesophytic optimum (warmand humid conditions) with relatively high lake levels to xerophyticconditions and high "re activity is traced to the period from 7.5 to5.0 ka cal. BP (Carrión et al., 2001a, 2001b). Major Saharan dust inputsin a southern Iberian alpine lake have been documented to increase

a)

c)

e)

b)

d)

f)

206

204

204

208

208

206

208

207

207

206

207

204

38.638.738.838.939.039.1

18.65

18.70

18.75

18.80

18.85

15.55

15.60

15.65

15.70

15.75

15.80

2.4762.4782.4802.4822.4842.4862.4882.4902.066

2.0682.0702.0722.0742.0762.0782.080 0.832

0.8330.8340.8350.8360.8370.8380.839

8.2k

a

7.4k

a PC

AI

YD

gnillB

H1 LGM

0Aller d

2 4 6 8 10 12 14 16 18 20

TH

ME

GS-12b1c 1e1a 1b 1d 2a

GI-1 GS-2

0 2 4 6 8 10 12 14 16 18 20

Holocene

Fig. 4. 206Pb/204Pb (a), 207Pb/204Pb (b), 208Pb/204Pb (c), 207Pb/206Pb (d), 208Pb/206Pb (e), and 208Pb/207Pb (f) ratios inmarine core 293G (b37 !m fraction) plotted against age. Error bars aresmaller than symbol size if they are not shown. Grey bars show periods with more unradiogenic Pb ratios.

244 M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

from 7.0 to 6.0 ka cal. BP and increased since then until present(Jiménez-Espejo et al., 2014).

Abrupt changes in !uxes preceding the 8.2 ka cold event by 0.2 and1.0 ka with North African eolian dust inputs over central Europe havealso been described in a peat bog record (Le Roux et al., 2012). Inparticular, a signi"cant period of increased dust deposition described inthe ombrotrophic bog during the early–middle Holocene transition(EMHT), at ca. 8.5–8.4 ka cal. BP, was recorded with a decrease of!Nd(0) signature up to!12.5, which is comparable with old continentalshields from the Sahara (Le Roux et al., 2012). During the sameperiod, en-hanced!uxes of Saharan dust deposition have also been found in lake andice records from Africa (Gasse, 2000; Thompson et al., 2002). Further-more, changes in !uvial regime were interrupted by a short-term aridevent at 8.5–7.3 ka cal. BP in marine records from the eastern Mediterra-nean (Blanchet et al., 2013). Thus, an arid signal from North Africa seemsconsistent with the Sr–Nd excursion and !Nd(0) values of !11.7 de-scribed at 8.9–8.7 ka cal. BP, likely corresponding to the EMHT (Table 1,Fig. 3b–c).

During the late Holocene, a strong reduction in tropical trees andSahelian grassland cover in northern Chad allowed large-scale dustmobilization from 4.3 ka cal. BP (Kröpelin et al., 2008). Around3.0–2.0 ka cal. BP, a widespread phase of !uvial–lacustrine depositionand eolian stabilization is dated in the Sahara region (Swezey, 2001),while around 2.7 ka cal. BP, a desert ecosystem was established withperiods of very severe droughts especially at 2.0–1.2 ka cal. BP (Gasse,2002; Kröpelin et al., 2008). We have identi"ed major Sr–Nd–Pb shiftsat 5.6, 2.2 and 1.1 ka cal. BP (Figs. 3c, 4), but there is not enoughcorrespondence with African source signatures.

Therefore, although we have to consider some uncertainties in thechronological control of the different archives, Sr, Nd and Pb isotopevariations and Nd model ages obtained in the marine sediment fromthe western Mediterranean have shown main recognized shifts at 16.7–16.2 (corresponding with the H1) and 8.9–8.7 ka cal. BP (ca. EMHT).These changes seem related to variations in terrigenous material fromdifferent source areas associated with the main phases of eolian dustinput in accordance with other nearby palaeorecords.

volcanicend-member10

5

0

-5

-10

-15

-20

-25

-30

-35

0.700 0.705 0.710 0.715 0.720 0.725 0.730 0.735 0.740

(Mauritania)

Mauritania

Senegal

MalidN

(0)

87 86Sr/ Sr

-12.5

-12.0

-11.5

-11.0

-10.5

-10.0

Europeanatmosphericinput

0.713 0.714 0.715 0.716 0.717 0.718

Mauritania

dN

(0)

87 86Sr/ Sr

Fig. 5. !Nd(0) distribution vs. 87Sr/86Sr ratiomodi"ed afterGrousset et al. (1998). Error bars are smaller than symbol size (Table 1). Data fromAfrican (Scheuvens et al., 2013) and Europeansources (Henry et al., 1994) are showed.

245M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

18.8638.6

38.7

38.8

38.9

39.0

39.1

39.2

206 204Pb/ Pb

402802

bP

/bP

206 204Pb/ Pb

207 206Pb/ Pb

(EMHT)

15.58

15.60

15.62

15.64

15.66

15.68

15.70

15,72

15.74

15.76

N

E

E

N

402702

bP

/bP

18.76 18.78 18.80 18.82 18.84

18.76 18.78 18.80 18.82 18.84 18.86

0.835 0.836 0.837 0.838

18.66 18.68 18.70 18.72 18.74

18.66 18.68 18.70 18.72 18.74

0.832 0.833 0.8342.066

2.068

2.070

2.072

2.074

2.076

2.078y=2.2279x+0.2119R =0.8242

602802

bP

/bP

a)

b)

c)

(EMHT)

(EMHT)

Fig. 6. (a) 207Pb/204Pb vs. 206Pb/204Pb, (b) 208Pb/204Pb vs. 206Pb/204Pb and (c) 208Pb/206Pb vs. 207Pb/206Pb ratios for 293G core (b37 !m fraction). Error bar values are shown in Table 2.Dashed contour represents Pb isotope composition of granitic K-feldspars from North Africa (N) and European sources (E) (Juteau et al., 1986; Fagel et al., 2004).

246 M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

4.2. Terrigenous provenance

We have compiled Sr, Nd, and Pb isotopic data as well as TDM fromavailable literature as possible source areas of eolian and riverinederived terrigenous input to the westernmost Mediterranean (Fig. 1a).

Based on published literature, the Sr–Nd isotopic composition of theeolian dust from the Saharan region can be explained by a mixture ofthreemain end-members: an almost constant contribution of a volcanicend-member that can be considered as the background, and two crustalend-members, i.e. theArchean (2.3-to-3Ga) andpost-Archean “Birimian”(2.3-to-1.7 Ga) geological formations (Grousset and Biscaye, 2005). Thus,the “North Africa domain” approximately includes a mixture of thesethree end-member domains. Three sub-provinces can be also distin-guished according to the geology of the source areas: Northern sources(Morocco, Algeria, Mauritania and Mali); Eastern/Southern sources(Libya, Chad, Guinea, Senegal) and the Archean Saharan shield, withits main outcrops located in Mauritania (Grousset and Biscaye, 2005).

The combination of 87Sr/86Sr and !Nd(0) values (Fig. 5) shows adistribution of samples in two sub-clusters, i.e., Fields I and II (Fig. 5).Field I comprises northern African regions such as Morocco, Mauritania,and Mali, as well as more remote regions such as Niger and Guinea, andis the main source cluster over the time period studied. Conversely,Field II corresponds to isotopic signatures from the present Senegal region(Grousset et al., 1998). Speci!cally, those sediment samples clustering inField II are dated at 16.7 ka (last H1 event) and 8.9 ka cal. BP, supportingthat the shifts in the 87Sr/86Sr ratio at these periods (Figs. 3c, 5) mainlyrecorded variations in source areas rather than weathering effects.

Combining TDM data compiled (Fig. 1a), Field I will be characterizedby TDM around 1.0–1.8 Ga, and Field II will have higher values of ca.2.5 Ga (Bea et al., 2010). However, the two data points included inField II do not show particularly old TDM values, i.e. 1.37–1.46 Ga(Table 1 and Fig. 3a). These two TDMdatamay suggest highmobilizationand contribution of mixedmaterial from North Africa, corresponding toField I, and southern European areas (Nägler, 1990), although someolder material transported from Senegal (Field II) cannot be completelydiscarded (Fig. 1a). TDM values are generally lower than 1.46 Ga, exceptat 10.8 ka cal. BP that is recorded the oldest Nd model age (1.60 Ga,Table 1 and Fig. 3a), pointing out to a main contribution of materialfrom the Anti-Atlas African granitoids from Field I and IberianCambro-Ordovician rocks. Low !Nd(0) values at 10.2, 9.4, 6.8, 4.3 and1.1 ka cal. BP (Fig. 3b) may be also related to typical values of materialderived from the Rhône and Têt Rivers in the western Mediterranean(Henry et al., 1994; Stille and Schaltegger, 1996), with a relative contri-bution of southern Iberian rocks from Ossa Morena due to the input ofsouthern Iberian rivers (López-Guijarro et al., 2008) (Fig. 1a).

Concerning Pb isotope ratios, obtained values re"ect in general atypical eolian Saharan dust signature (Grousset et al., 1994; Kylanderet al., 2005; Erel and Torrent, 2010). Cross-plots of 207Pb/204Pb vs.206Pb/204Pb (Fig. 6a) and 208Pb/204Pb vs. 206Pb/204Pb (Fig. 6b) showthat unradiogenic Pb signatures correspond to a typical composition ofgranites from North Africa (206Pb/204Pb = 18.74, 207Pb/204Pb = 15.66,and 208Pb/204Pb = 38.80; Juteau et al., 1986), whereas the signaturesoutside this group, however, could be associated with a Europeangranite source (Juteau et al., 1986). A linear correlation in 207Pb/204Pbvs. 206Pb/204Pb (Fig. 6a) and 208Pb/204Pb vs. 206Pb/204Pb (Fig. 6b) canbe drawn, suggesting the presence of two distinct Pb components ofsimilar age, but different time-integrated Th/U ratios. According topotential European Pb sources described by Fagel et al. (2004), the Pbsignature could correspond to upper and post-Paleozoic materialsrelated to the Variscan orogeny from Western Europe. Moreover, thecross-plot of 208Pb/206Pb vs. 207Pb/206Pb shows a good linear correlation(R2 = 0.824, Fig. 6c), with some outliers during the late Holocenecorresponding to 4.3, 3.4 and 2.2 ka cal. BP. These events are alsocharacterized with high Sr isotopic values over the last 20 ka (Fig. 3c)and minimum TDM values, between 1.34 and 1.38 (Fig. 3a), as result ofa contribution of Iberian and Anti-Atlas sediments. More radiogenic Sr

and Pb isotopic signatures have been described in sediments depositedin the Gulf of Cadiz and the Portuguese margin, implying major contri-butions of eroded material from Iberia with respect to a North Africansource (Stumpf et al., 2011). Moreover, differences between Pb signalsfrom the Saharan dust and from ancient Iberian mining and smeltingca. 3.2 ka BP were identi!ed in a northern Iberian peat bog (Kylanderet al., 2005). Similarly, some southern Iberian records have shown asigni!cant anthropogenic lead pollution signal by early metallurgysince 3.9 ka cal. BP (García-Alix et al., 2013). However, we do not haveclear evidences in our record for this kind of potential lead pollutionduring these periods. Therefore, natural Pb signatures in the marinesediment record from the westernmost Mediterranean can also beexplained in terms of simple binary mixing between North Africa andEuropean sources (Fig. 6c). Variations along the mixing line re"ectchanges in the relative mixing proportions of the two-natural Pbend-member sources. The more radiogenic Pb source corresponds to aEuropean source, and the less radiogenic end member to a North Africaprovenance. Enhanced signatures of unradiogenic Pb occurred duringcolder periods such as the last H1 event (16.7 ka cal. BP), while moreradiogenic Pb signatures dominated during most of the time period,particularly during warmer time intervals such as the Allerød period(13.5 ka cal. BP) and some Holocene intervals (11.1, 9.4, 1.1 ka cal. BP).

Thus, the combination of the three isotopic systems (Nd, Sr, Pb)and Nd model ages in the marine record displays broadly concordantvariations, providing a !rst approach of terrigenous input provenancein the westernmost Mediterranean. Even though further Sr, Nd and Pbsignatures and TDM data from some potential source areas are neededto provide a more accurate source discrimination, results mainly re"ecta mixture of terrigenous material from North African and SouthEuropean sources linked with prevailing climate conditions ratherthan a strong in"uence of a particular area.

4.3. Transport mechanisms

Traditional interpretations about transport mechanisms of terrige-nous material based on grain size distribution indicated that smallgrain sizes (b6 "m) are indicative of riverine runoff, and coarse grainsizes (N6 "m) are in general associated with eolian input (Sarntheinet al., 1981). However, we cannot project different transports based onthe physical sorting since the grain size distribution in the marinerecord showed always values lower than 6 "m (Fig. 3d; Rodrigo-Gámizet al., 2011). Interestingly, major sizes (N3 "m; Fig. 3d) are almostcontemporaneous with major shifts in Sr isotopes, i.e. less radiogenic87Sr/86Sr ratio (Fig. 3c), which could suggest to be an indicator for highhumidity in the source area. Nevertheless, some of these shifts, like at16.70 ka cal. BP have occurred during less humid periods, and thereforewe cannot establish a strong relationship between grain size and Srcomposition variations.

The terrigenous material deposited in the westernmost Mediterra-nean has been derived by either atmospheric inputs or river runofffrom South European and North African sources. The main southernEuropean rivers, due to the scarcity of rivers along North Africa, are(Fig. 1a): (1) the Rhône River draining detrital sediments from theAlps (Juteau et al., 1986; Nägler, 1990; Stille and Schaltegger, 1996),which were further transported from the Gulf of Lion to the AlboranSea by the oceanic gyre (Millot, 1999); (2) the Guadalhorce andGuadalfeo Rivers from southern Iberia that discharge material in theAlboran Sea as torrential or sporadic runoff due to their non-permanent feature; and (3) the Guadiana and Guadalquivir Rivers, lo-cated in the southwestern Iberia, which "uvial discharge can transporthigh amounts of detrital particulate matter from the Guadalquivirbasin (Erel and Torrent, 2010) facilitating the sediment distributioninto the western Mediterranean by Atlantic in"ow waters (Groussetet al., 1998).

Regarding atmospheric transport, some pressure con!gurations havebeen described to cause Saharan dust input to the Iberian Peninsula:

247M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

(1) low pressures over West and/or Southwest Portugal; (2) highpressure over the East or Southeast Iberian Peninsula; and (3) thecombination of both low and high pressure systems (Avila et al., 1998;Rodríguez et al., 2001). Interannual variations in dust transport to theAtlantic Ocean have been correlated with the climatic variabilityde!ned by the NAO (Hurrell, 1995; Moulin et al., 1997). Furthermore,terrigenous input is largely controlled by the migrations of the ITCZ,which also determines the position of the subtropical Azores anticyclone.ITCZ migrations are associated with alternating dry and wet periodsaccording to the monsoon activity "uctuations. These are accompaniedby changes in the wind regime and intensity, as well as variable coverageof land by vegetation, and variations in rainfall amount and riverdischarge (Clemens and Prell, 1990; Gasse and Van Campo, 1994;Gasse, 2000).

During summer a northward shift in the ITCZ position locatedaround 20–30°N leads to a more northerly position of the monsoonrain belt system, decreasing the aridity over North Africa and causingnorthward expansion of the Sahelian savannah into the Sahara region(e.g., Moulin et al., 1997; Bergametti et al., 1989a; Rodríguez et al.,2001; Torres-Padrón et al., 2002). This situation, together with lowwind intensity, would reduce the dust emission from the southernand southwestern African regions, favouring terrigenous transportfrom the northern African regions to the western and central partsof the Mediterranean (Moulin et al., 1997). In general, most of theterrigenous sediment cluster in the northern African (Western Sahara,Morocco, Mauritania, Mali, Niger, and Algeria), southern Iberia andsouthern France areas (Fig. 1a).

In contrast, the southernmost position of the ITCZ during winterleads to a more southerly position of the anticyclone centre and themonsoon rain belt system, increasing the aridity in North Africaand decreasing the cover vegetation (e.g., Matthewson et al., 1995;Arbuszewski et al., 2013; McGee et al., 2013, 2014). This con!gurationaccompanied by major wind intensity gives rise to greater eolian dustmobilization from latitudes of 10–20°N and transport toward thewesternmost Mediterranean. This situation could explain a secondterrigenous source area represented by more unradiogenic terrigenousmaterial like old Archaean rocks from the present day Senegal regionrecorded at 16.7–16.2 (H1) and 8.9–8.7 ka cal. BP (EMHT) (Fig. 1a).

Therefore, a combination of changes in wind intensity, the displace-ment and distance of the anticyclone centres and ITCZ migrations aswell as the availability of the eolianmaterial according to the vegetationcover, are the main factors controlling the different transport mecha-nisms of terrigenous material deposited in the western Mediterraneanduring the last 20 ka.

5. Conclusions

Radiogenic isotope (Sr, Nd, Pb) signatures and Nd model ages haveallowed us to reconstruct source areas and transport mechanisms ofterrigenous material in the westernmost Mediterranean since the LastGlacial Maximum. Nd, Sr and Pb isotope compositions from carbonate-free sediments of the b37 !m size fraction and Nd model ages havebeen integrated with previous geochemical and mineralogical data forsuch reconstruction. Lower 87Sr/86Sr signatures re"ect either the in"u-ence of chemical weathering or variations in terrigenous provenance.Contemporaneous shifts in the 87Sr/86Sr ratio and "Nd(0) during the H1(at 16.7–16.2 ka cal. BP) and the EMHT (at 8.9–8.7 ka cal. BP) suggestvariations in the provenance of terrigenous material during these aridand cold periods. The isotopic signatures in the terrigenous componentshowed an older contribution of likely old Archaean rocks from thepresent-day Senegal region. In contrast, warmer periods displayradiogenic signatures that indicate terrigenous provenance from thenorthern Africa (Western Sahara, Morocco, Mauritania, Mali, Niger,and Algeria), southern Iberia and southern France areas. Furthermore,although Pb ratios do not discriminate among potential sources inNorth Africa or South Europe, they re"ect the mixture between two-

natural Pb sources, i.e., a radiogenic European source, transported byriver runoff or atmospheric input mainly during the Holocene, and aless radiogenic North Africa source during the glacial period. Variationsin TDM further support the mixture of pre-Pan-African and Pan-Africanand Iberian derived-material in the western Mediterranean marinesediments. Migrations of the ITCZ along with changes in the intensityand regime of the wind system and the African monsoon, and theposition of the subtropical Azores anticyclone seem to be the maintransport mechanisms controlling terrigenous provenance.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.chemgeo.2015.06.004.

Acknowledgments

This work was supported by the European Regional DevelopmentFund (ERDF)-co!nanced grants CGL2012-32659, CGL2009-07603(Secretaría de Estado de Investigación, Desarrollo e Innovación,MINECO), Project RNM-5212 and Research Group RNM-179 (Junta deAndalucía). We are also grateful to the oceanographic cruise TrainingThrough Research Programme (UNESCO-Moscow State University).We would like to thank F. Grousset for his helpful comments on anearly version of the manuscript and P. Montero and F. Bea for the TDMdata provided. We likewise thank M. Senn-Gerber and D. Fontignie fortheir laboratory assistance. J. L. Sanders post-edited the English style.Comments and suggestions from two anonymous reviewers havesubstantially improved the !nal version of the manuscript.

References

Abouchami, W., Näthe, K., Kumar, A., Galer, S.J.G., Jochum, K.P., Williams, E., Horbe, A.M.C.,Rosa, J.W.C., Balsam,W., Adams, D., Mezger, K., Andreae,M.O., 2013. Geochemical andisotopic characterization of the Bodélé Depression dust source and implications fortransatlantic dust transport to the Amazon Basin. Earth Planet. Sci. Lett. 380, 112–123.

Arbuszewski, J.A., deMenocal, P.B., Cléroux, C., Bradtmiller, L., Mix, A., 2013. Meridionalshifts of the Atlantic intertropical convergence zone since the Last Glacial Maximum.Nat. Geosci. 6, 959–962.

Arndt, N.T., Goldstein, S.L., 1987. Use and abuse of crust-formation ages. Geology 15,893–895.

Avila, A., Alarcón, M., Queralt, I., 1998. The chemical composition of dust transported inred rains— its contribution to the biochemical cycle of a holm oak forest in Catalonia(Spain). Atmos. Environ. 32, 179–191.

Bea, F., Montero, P., Talavera, C., Abu, Anbar M., Scarrow, J.H., Molina, J.F., Moreno, J.A., 2010.The palaeogeographic position of Central Iberia in Gondwana during the Ordovician:evidence from zircon chronology and Nd isotopes. Terra Nova 22, 341–346.

Bergametti, G., Dutot, A.L., Buat-Ménard, P., Remoudaki, E., 1989a. Seasonal variability ofthe elemental composition of atmospheric aerosol particles over the northwesternMediterranean. Tellus 41B, 553–561.

Bergametti, G., Gomes, L., Remoudaki, E., Desbois, M., Martin, D., Buat-Ménard, P., 1989b.Present transport and deposition patterns of African dusts to the north-westernMediterranean. In: Leinen, M., Sarnthein, M. (Eds.), Paleoclimatology andPaleometeorology: Modern and Past Patterns of Global Atmospheric Transport.NATO ASI Series, Mathematical and Physical Sciences vol. 282. Kluwer, Boston,pp. 227–250.

Biscaye, P.E., Dasch, E.J., 1971. The rubidium, strontium, strontium-isotope system indeep-sea sediments: Argentine Basin. J. Geophys. Res. 76, 5087–5096.

Blanchet, C.L., Tjallingii, R., Frank, M., Lorenzen, J., Reitz, A., Brown, K., Feseker, T.,Brückmann, W., 2013. High- and low-latitude forcing of the Nile River regime duringtheHolocene inferred from laminated sediments of theNile deep-sea fan. Earth Planet.Sci. Lett. 364, 98–110.

Blum, J.D., Erel, Y., 1997. Rb–Sr-isotope systematics of a grantic soil chronosequence: theimportance of biotite weathering. Geochim. Cosmochim. Act 61, 3193–3204.

Bout-Roumazeilles, V., Combourieu, Nebout N., Peyron, O., Cortijo, E., Landais, A.,Masson-Delmotte, V., 2007. Connection between South Mediterranean climate andNorth African atmospheric circulation during the last 50,000 yr BP North Atlanticcold events. Quat. Sci. Rev. 26, 3197–3215.

Box, M.R., Krom, M.D., Cliff, R.A., Bar-Matthews, M., Almogi-Labin, A., Ayalon, A., Paterne,M., 2011. Response of the Nile and its catchment to millennial-scale climatic changesince the LGM from Sr isotopes andmajor elements of EastMediterranean sediments.Quat. Sci. Rev. 30, 431–442.

Braun, J.J., Viers, J., Dupre, B., Polve, M., Ndam, J., Muller, J.-P., 1998. Solid/liquid REEfractionation in the lateritic system of Goyoum, East Cameroon: the implication forthe present dynamics of the soil covers of the humid tropical regions. Geochim.Cosmochimi. Act 62, 273–299.

Caquineau, S., Gaudichet, A., Gomes, L., Legrand, M., 2002. Mineralogy of Saharandust transported over northwestern tropical Atlantic Ocean in relation with sourceregions. J. Geophys. Res. 107, 4251.

248 M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

Carrión, J.S., 2002. Patterns and processes of Late Quaternary environmental change in amontane region of southwestern Europe. Quat. Sci. Rev. 21, 2047–2066.

Carrión, J.S., Andrade, A., Bennett, K.D., Navarro, C., Munuera, M., 2001a. Crossing forestthresholds: inertia and collapse in a Holocene sequence from southcentral Spain.The Holocene 11, 635–653.

Carrión, J.S., Munuera, M., Dupré, M., Andrade, A., 2001b. Abrupt vegetation changes inthe Segura mountains of southern Spain throughout the Holocene. J. Ecol. 89,783–797.

Carrión, J.S., Fernández, S., Jiménez-Moreno, G., Fauquette, S., Gil-Romera, G.,González-Sampériz, P., Finlayson, C., 2010. The historical origins of aridity and vegetationdegradation in southeastern Spain. J. Arid Environ. 74, 731–736.

Clemens, S.C., Prell, W.L., 1990. Late Pleistocene variability of Arabian Sea summer monsoonwinds and continental aridity: eolian records from the lithogenic component ofdeep-sea sediments. Paleoceanography 5, 109–145.

Cole, J.M., Goldstein, S.L., deMenocal, P.B., Hemming, S.R., Grousset, F.E., 2009. Contrastingcompositions of Saharan dust in the eastern Atlantic Ocean during the last deglaciationand African Humid Period. Earth Planet. Sci. Lett. 278, 257–266.

Comas, M.C., Ivanov, M.K., 2003. Alboran Basin (Leg 3). In: Kenyon, H., Ivanov, M.K.,Akhmetzhanov, A.M., Akhmanov, G.G. (Eds.), Interdisciplinary Geoscience Researchon the North East Atlantic Margin, Mediterranean Sea and Mid-Atlantic Ridge. IOCTechnical Series. UNESCO, Paris, pp. 51–71.

Combourieu Nebout, N., Peyron, O., Dormoy, I., 2009. Rapid climatic variability in thewestMediterranean during the last 25 000 years from high resolution pollen data. Clim.Past 5, 503–521.

Coude-Gaussen, G., Rognon, P., Bergametti, G., Gomes, L., Strauss, B., Gros, J., Le Coustumer,M., 1987. Saharan dust on Fuerteventura island (Canaries): chemical and mineralogicalcharacteristics, air mass trajectories, and probable sources. J. Geophys. Res. 92,9753–9771.

deMenocal, P.B., 1995. Plio-Pleistocene African climate. Science 270, 53–59.deMenocal, P.B., 2004. African climate change and faunal evolution during the

Pliocene–Pleistocene. Earth Planet. Sci. Lett. 220, 3–24.deMenocal, P.B., Ortiz, J., Guilderson, T., Adkins, J., Sarnthein, M., Baker, L., Yarusinsky, M.,

2000. Abrupt onset and termination of the African Humid Period: rapid climateresponses to gradual insolation forcing. Quat. Sci. Rev. 19, 347–361.

DePaolo, D.J., 1981. Neodymium isotopes in the Colorado Front Range and implicationsfor crust formation and mantle evolution in the Proterozoic. Nature 291, 193–197.

Díaz-Hernández, J.L., Martín-Ramos, J.D., López-Galindo, A., 2011. Quantitative analysis ofmineral phases in atmospheric dust deposited in the southeastern Iberian Peninsula.Atmos. Environ. 18, 3015–3024.

Dickin, A., 1997. Radiogenic Isotope Geology. second ed. Cambridge University Press,Cambridge.

Dulac, F., Tarnè, D., Bergametti, G., Burat-Ménard, P., Desbois, M., 1992. Assessment of theAfrican airborne dust mass over the westernMediterranean Sea using Meteosat data.J. Geophys. Res. 97, 2489–2506.

Elenga, H., Peryon, O., Bonne!lle, R., Prentice, I.C., Jolly, D., Cheddadi, R., Guiot, J., Andrieu,V., Bottema, S., Buchet, G.L., de Beaulieu, J.L., Hamilton, A.C., Maley, J., Marchant, R.,Perez-Obiol, R., Reille, M., Riollet, G., Scott, L., Straka, H., Taylor, D., Van Campo, E.,Vincens, A., Laarif, F., Jonson, H., 2000. Pollen based reconstruction for SouthernEurope and Africa 18,000 years ago. J. Biogeogr. 27, 621–634.

Erel, Y., Torrent, J., 2010. Contribution of Saharan dust to Mediterranean soils assessed bysequential extraction and Pb and Sr isotopes. Chem. Geol. 275, 19–25.

Fagel, N., Hillaire-Marcel, C., Humblet, M., Brasseur, R., Weis, D., Stevenson, R., 2004. Ndand Pb isotope signatures of the clay-size fraction of Labrador Sea sediments duringthe Holocene: implications for the inception of the modern deep circulation pattern.Paleoceanography 19 (PA3002).

Fletcher, W.J., Sánchez Goñi, M.F., 2008. Orbital- and sub-orbital-scale climate impacts onvegetation of the western Mediterranean basin over the last 48.000 yr. Quat. Res. 70,451–464.

Fletcher, W.J., Sánchez Goñi, M.F., Peyron, O., Dormoy, I., 2010. Abrupt climate changes ofthe last deglaciation detected in a Western Mediterranean forest record. Clim. Past 6,245–264.

Formenti, P., Schütz, L., Balkanski, Y., Desboeufs, K., Ebert, M., Kandler, K., Petzold, A.,Scheuvens, D., Weinbruch, S., Zhang, D., 2011. Recent progress in understandingphysical and chemical properties of African and Asian mineral dust. Atmos. Chem.Phys. 11, 8231–8256.

Foucault, A., Mélières, F., 2000. Palaeoclimatic cyclicity in central Mediterranean Pliocenesediments: the mineralogical signal. Palaeogeogr. Palaeoclimatol. 158, 311–323.

Frank, M., 2002. Radiogenic isotopes: tracers of past ocean circulation and erosional input.Rev. Geophys. 40, 1–38.

García-Alix, A., Jimenez-Espejo, F.J., Lozano, J.A., Jiménez-Moreno, G., Martinez-Ruiz, F.,García, Sanjuán L., Aranda, Jiménez G., García, Alfonso E., Ruiz-Puertas, G., Scott,Anderson R., 2013. Anthropogenic impact and lead pollution throughout the Holocenein Southern Iberia. Sci. Total Environ. 449, 451–460.

Gasse, F., 2000. Hydrological changes in the African tropics since the Last Glacial Maximum.Quat. Sci. Rev. 19, 189–211.

Gasse, F., 2002. Diatom-inferred salinity and carbonate oxygen isotopes in Holocenewaterbodies of the western Sahara and Sahel (Africa). Quat. Sci. Rev. 21, 737–767.

Gasse, F., Van Campo, E., 1994. Abrupt post-glacial climate events in West Asia and NorthAfrica monsoon domains. Earth Planet. Sci. Lett. 126, 435–456.

Gerstenberger, H., Haase, G., 1997. A highly effective emitter substance for massspectrometric Pb isotope ratio determinations. Chem. Geol. 136, 309–312.

Goldstein, S.L., Onions, R.K., Hamilton, P.J., 1984. A Sm–Nd isotopic study of atmo-spheric dusts and particulates from major river systems. Earth Planet. Sci. Lett.70, 221–236.

Goudie, A.S., Middleton, N.J., 2001. Saharan dust storms: nature and consequences. Earth-Sci.Rev. 56, 179–204.

Grousset, F.E., Biscaye, P.E., 2005. Tracing dust sources and transport patterns using Sr, Ndand Pb isotopes. Chem. Geol. 222, 149–167.

Grousset, F.E., Joron, J.L., Biscaye, P.E., Latouche, C., Treuil, M., Maillet, N., Faugères,J.C., Gonthier, E., 1988. Mediterranean out"ow through the Strait of Gibraltarsince 18,000 years B.P.: mineralogical and geochemical arguments. Geo-Mar.Lett. 8, 25–34.

Grousset, F.E., Biscaye, P.E., Revel, M., Petit, J.R., Pye, K., Joussaume, S., Jouzel, J., 1992.Antarctic (Dome C) ice-core dust at 18 k.y. B.P.: isotopic constraints on origins. EarthPlanet. Sci. Lett. 111, 175–182.

Grousset, F., Quétel, C.R., Thomas, B., Buat-Ménard, P., Donard, O.F.X., Bucher, A., 1994.Transient Pb isotopic signatures in the western European atmosphere. Environ. Sci.Technol. 28, 1605–1608.

Grousset, F.E., Parra, M., Bory, A., Martínez, P., Bertrand, P., Shimmield, G., Ellam, R.M.,1998. Saharan wind regimes traced by the Sr–Nd isotopic composition of subtropicalAtlantic sediments: Last Glacial Maximum vs today. Quat. Sci. Rev. 17, 395–409.

Hall, N.M.J., Peyrillé, P., 2006. Dynamics of the west African monsoon. J. Phys. IV 139,81–99.

Henry, F., Jeandel, C., Duprè, B., Minster, J.-F., 1994. Particulate and dissolved Nd in thewestern Mediterranean Sea: sources, fate and budget. Mar. Chem. 45, 283–305.

Holz, C., 2004. Climate-induced Variability of Fluvial and Aeolian Sediment Supplyand Gravity-driven Sediment Transport Off Northwest Africa (Ph. D. dissertation).Bremen Univ.

Hurrell, J.W., 1995. Decadal trend in the North Atlantic Oscillation: regional temperaturesand precipitations. Science 269, 676–679.

Jeandel, C., Arsouze, T., Lacan, F., Téchiné, P., Dutay, J.-C., 2007. Isotopic Nd compositionsand concentrations of the lithogenic inputs into the ocean: a compilation, with anemphasis on the margins. Chem. Geol. 239, 156–164.

Jilbert, T., Reichart, G.-J., Aeschlimann, B., Günther, D., Boer, W., de Lange, G., 2010.Climate-controlled multidecadal variability in North African dust transport to theMediterranean. Geology 38, 19–22.

Jiménez-Espejo, F.J., García-Alix, A., Jiménez-Moreno, G., Rodrigo-Gámiz, M., Anderson, R.S.,Rodríguez-Tovar, F.J., Martínez-Ruiz, F., Giralt, S., Delgado, Huertas A., Pardo-Igúzquiza,E., 2014. Saharan aeolian input and effective humidity variations over western Europeduring the Holocene from a high altitude record. Chem. Geol. 374–375, 1–12.

Jullien, E., Grousset, F.E., Malaizé, B., Duprat, J., Sanchez Goñi, M.F., Eynaud, F., Charlier, K.,Schneider, R., Bory, A., Bout-Roumazeilles, V., Flores, J.A., 2007. Low-latitude “dustyevents” vs. high-latitude “icy Heinrich events”. Quat. Res. 68, 379–386.

Jung, S.J.A., Davies, G.R., Ganssen, G.M., Kroon, D., 2004. Stepwise Holocene aridi!cation inNE Africa deduced from dust-borne radiogenic isotope records. Earth Planet. Sci. Lett.221, 27–37.

Junginger, A., Roller, S., Olaka, L.A., Trauth, M.H., 2014. The effects of solar irradiation changeson the migration of the Congo Air Boundary and water levels of paleo-Lake Suguta,Northern Kenya Rift, during the African Humid Period (15–5 ka BP). Palaeogeogr.Palaeoclimatol. 396, 1–16.

Juteau, M., Michard, A., Albarede, F., 1986. The Pb–Sr–Nd isotope geochemistry of somerecent circum-Mediterranean granites. Contrib. Mineral. Petrol. 92, 331–340.

Kolla, V., Biscaye, P.E., Hanley, A.F., 1979. Distribution of quartz in late Quaternary Atlanticsediments in relation to climate. Quat. Res. 11, 261–277.

Krom, M.D., Cliff, R.A., Eijsink, L.M., Herut, B., Chester, R., 1999a. The characterization ofSaharan dusts and Nile particulate matter in surface sediments from the Levantinebasin using Sr isotopes. Mar. Geol. 155, 319–330.

Krom, M.D., Michard, A., Cliff, R.A., Strohle, K., 1999b. Sources of sediment to the IonianSea and western Levantine basin of the Eastern Mediterranean Sea during S-1sapropel times. Mar. Geol. 160, 45–61.

Kröpelin, S., Verschuren, D., Lézine, A.-M., Eggermont, H., Cocquyt, C., Francus, P., Cazet, J.-P.,Fagot, M., Rumes, B., Russell, J.M., Darius, F., Conley, D.J., Schuster, M., von Suchodoletz,H., Engstrom, D.R., 2008. Climate-driven ecosystem succession in the Sahara: the past6000 years. Science 320, 765–768.

Kuhlmann, H., Meggers, H., Freudenthal, T., Wefer, G., 2004. The transition of themonsoonal and the N Atlantic climate system off NW Africa during the Holocene.Geophys. Res. Lett. 31, L22204.

Kylander, M.E., Weiss, D.J., Martínez, Cortízas A., Spiro, B., Garcia-Sanchez, R., Coles,B.J., 2005. Re!ning the pre-industrial atmospheric Pb isotope evolution curvein Europe using an 8000 year old peat core from NW Spain. Earth Planet. Sci.Lett. 240, 467–485.

Laurent, B., Marticorena, B., Bergametti, G., Léon, J.F., Mahowald, N.M., 2008. Modelingmineral dust emissions from the Sahara desert using new surface properties andsoil database. J. Geophys. Res. 113, D14218.

Le Roux, G., Fagel, N., De Vleeschouwer, F., Krachler, M., Debaille, V., Stille, P., Mattielli, N., vander Knaap, W.O., van Leeuwen, J.F.N., Shotyk, W., 2012. Volcano- and climate-drivenchanges in atmospheric dust sources and "uxes since the Late Glacial in CentralEurope. Geology 40, 335–338.

Lézine, A.-M., Hély, C., Grenier, C., Braconnot, P., Krinner, G., 2011. Sahara and Sahelvulnerability to climate changes, lessons from Holocene hydrological data. Quat. Sci.Rev. 30, 3001–3012.

López-Guijarro, R., Armendáriz, M., Quesada, C., Fernández-Suárez, J., Murphy, J.B., Pin, C.,Bellido, F., 2008. Ediacaran–Palaeozoic tectonic evolution of the Ossa Morena andCentral Iberian zones (SW Iberia) as revealed by Sm–Nd isotope systematics.Tectonophysics 461, 202–214.

Lowe, J.J., Rasmussen, S.O., Bjoerck, S., Hoek, W.Z., Steffensen, J.P., Walker, M.J.C., Yu, Z.C.,INTIMATE group, 2008. Synchronisation of palaeoenvironmental events in the NorthAtlantic region during the Last Termination: a revised protocol recommended by theINTIMATE group. Quat. Sci. Rev. 27, 6–17.

Loÿe-Pilot, M.D., Martin, J.M., 1996. Saharan dust input to the western Mediterranean: aneleven years record in Corsica. In: Guerzoni, S., Chester, R. (Eds.), The Impact of DesertDust Across the Mediterranean. Kluwer Academic, Dordrecht, pp. 191–199.

249M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250

Loÿe-Pilot, M.D., Martin, J.M., Morelli, J., 1989. Atmospheric input of particulate matterand inorganic nitrogen to the northwestern Mediterranean. Water Pollut. Res. 13,368–376.

Matthewson, A.P., Shimmield, G.B., Kroon, D., Fallick, A.E., 1995. A 300 kyr high-resolutionaridity record of the North African continent. Paleoceanography 10, 3.

McArthur, J.M., Howarth, R.J., Bailey, T.R., 2001. Strontium isotope stratigraphy: LOWESSversion 3: best !t to the marine Sr-isotope curve for 0–509 Ma and accompanyinglook-up table for deriving numerical age. J. Geol. 109, 155–169.

McGee, D., deMenocal, P.B., Winckler, G., Stuut, J.B.W., Bradtmiller, L.I., 2013. The magnitude,timing and abruptness of changes in North African dust deposition over the last20,000 yr. Earth Planet. Sci. Lett. 371–372, 163–176.

McGee, D., Donohoe, A., Marshall, J., Ferreira, D., 2014. Changes in ITCZ location andcross-equatorial heat transport at the Last Glacial Maximum, Heinrich Stadial 1,and the mid-Holocene. Earth Planet. Sci. Lett. 390, 69–79.

Meyer, I., Davies, G.R., Stuut, J.-B.W., 2011. Grain size control on Sr–Nd isotope provenancestudies and impact on paleoclimate reconstructions: an example from deep-seasediments offshore NW Africa. Geochem. Geophy. Geosyst. 12, Q03005.

Millot, C., 1999. Circulation in the Western Mediterranean Sea. J. Mar. Syst. 20, 423–442.Molinaroli, E., 1996. Mineralogical characterization of Saharan dust with a view to its !nal

destination inMediterranean sediments. In: Guerzoni, S., Chester, R. (Eds.), The Impact ofDesert Dust Across the Mediterranean. Kluwer Academic, Dordrecht, pp. 153–162.

Montero, P., Bea, F., 1998. Accurate determination of 87Rb/86Sr and 147Sm/144Nd ratios byinductively-coupled-plasma mass spectrometry in isotope geoscience: an alternativeto isotope dilution analysis. Anal. Chim. Acta 358, 227–233.

Moreno, T., Querol, X., Castillo, S., Alastuey, A., Cuevas, E., Herrmann, L., Mounkaila, M.,Elvira, J., Gibbons, W., 2006. Geochemical variations in aeolian mineral particlesfrom the Sahara–Sahel Dust Corridor. Chemosphere 65, 261–270.

Moulin, C., Lambert, C.E., Dulac, F., Dayan, U., 1997. Control of atmospheric export of dustfrom North Africa by the North Atlantic Oscillation. Nature 387, 691–694.

Mulitza, S., Prange, M., Stuut, J.-B., Zabel, M., von Dobeneck, T., Itambi, A.C., Nizou, J.,Schulz, M., Wefer, G., 2008. Sahel megadroughts triggered by glacial slowdowns ofAtlantic meridional overturning. Paleoceanography 23 (PA4206).

Nägler, T., 1990. Sm–Nd, Rb–Sr and common lead isotope geochemistry on !ne-grainedsediments of the Iberian Massif. Nr. 9245 n° 9245, ETH ZürichDiss. Naturwiss.SwissFederal Institute of Technology Zurich.

Nesbitt, H.W., Markovics, G., Price, R.C., 1980. Chemical processes affecting alkalisand alkaline earths during continental weathering. Geochim. Cosmochimi. Act 44,1659–1666.

Nicholson, S., 2009. A revised picture of the structure of the “monsoon” and land ITCZover West Africa. Clim. Dyn. 32, 1155–1171.

Peyrillé, P., Lafore, J.P., 2007. An idealized two-dimensional framework to study the WestAfrican Monsoon. Part II: large-scale advection and the diurnal cycle. J. Atmos. Sci. 64,2783–2803.

Pin, R., Briot, D., Bassin, C., Poitrasson, F., 1994. Concomitant separation of strontium andsamarium–neodymium for isotopic analysis in silicate samples, based on speci!cextraction chromatography. Anal. Chim. Acta 298, 209–217.

Prospero, J.M., 1996. The atmospheric transport of particles to the ocean. In: Ittekkot, V.,Schäfer, P., Honjo, S., Depetris, J.P. (Eds.), Particle Flux in the Ocean. SCOPE Report57. John Wiley & Sons, New York, pp. 19–52.

Prospero, J.M., 1999. Long-range transport of mineral dust in the global atmosphere:impact of African dust on the environment of the southeastern United States. Proc.Natl. Acad. Sci. U. S. A. 96, 3396–3403.

Prospero, J.M., Mayol-Bracero, O.L., 2013. Understanding the transport and impact ofAfrican dust on the Caribbean Basin. Bull. Am. Meteorol. Soc. 94, 1329–1337.

Reader, M.C., Fung, I., McFarlane, N., 1999. The mineral dust aerosol cycle during the LastGlacial Maximum. J. Geophys. Res. 104, 9381–9398.

Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Ramsey, C.B., Buck,C.E., Burr, G.S., Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson, T.P., Hajdas, I.,Heaton, T.J., Hogg, A.G., Hughen, K.A., Kaiser, K.F., Kromer, B., McCormac, F.G.,Manning, S.W., Reimer, R.W., Richards, D.A., Southon, J.R., Talamo, S., Turney, C.S.M.,van der Plicht, J., Weyhenmeye, C.E., 2009. INTCAL09 and MARINE09 radiocarbon agecalibration curve, 0–50,000 years cal BP. Radiocarbon 51, 1111–1150.

Revel, M., Sinko, J.A., Grousset, F.E., 1996. Sr and Nd isotopes as tracers of North Atlanticlithic particles: palaeoclimatic implications. Palaeogeogr. Palaeoclimatol. 11, 95–113.

Revel, M., Ducassou, E., Grousset, F.E., Bernasconi, S.M., Migeon, S., Revillion, S., Mascle, J.,Murat, A., Zaragosi, S., Bosch, D., 2010. 100,000 years of Africanmonsoon variability insediments of the Nile margin. Quat. Sci. Rev. 29, 1342–1362.

Rodrigo-Gámiz, M., Martínez-Ruiz, F., Jiménez-Espejo, F.J., Gallego-Torres, D., Nieto-Moreno,V., Romero, O., Ariztegui, D., 2011. Impact of climate variability in the westernMediterranean during the last 20,000 years: oceanic and atmospheric responses. Quat.Sci. Rev. 30, 2018–2034.

Rodrigo-Gámiz, M., Martínez-Ruiz, F., Rodríguez-Tovar, F.J., Jiménez-Espejo, F.J.,Pardo-Igúzquiza, E., 2014a. Millennial- to centennial-scale climate periodicities andforcing mechanisms in the westernmost Mediterranean for the past 20,000 yr. Quat.Res. 81, 78–93.

Rodrigo-Gámiz, M., Martínez-Ruiz, F., Rampen, S.W., Schouten, S., Sinninghe Damsté, J.S.,2014b. Sea surface temperature variations in thewesternMediterranean Sea over thelast 20 kyr: a dual-organic proxy (UK!

37 and LDI) approach. Paleoceanography 29,87–98.

Rodríguez, S., Querol, X., Alastuey, A., Kallos, G., Kakaliagou, O., 2001. Saharan dustcontributions to PM10 and TSP levels in Southern and Eastern Spain. Atmos. Environ.35, 2433–2447.

Rowell, D.P., 2003. The impact of Mediterranean SSTs on the Sahelian rainfall season.J. Clim. 16, 849–862.

Sarnthein, G., Tetzlaff, G., Koopmann, B., Wolter, K., P"aumann, U., 1981. Glacial andinterglacial wind regimes over the eastern subtropical Atlantic and North-West Africa.Nature 293, 193–196.

Scheuvens, D., Schütz, L., Kandler, K., Ebert, M., Weinbruch, S., 2013. Bulk composition ofnorthern African dust and its source sediments — a compilation. Earth-Sci. Rev. 116,170–194.

Steager, J.C., Ryves, D.B., Chase, B.M., Pausata, F.S.R., 2011. Catastrophic drought in theAfro-Asian monsoon region during Heinrich Event 1. Science 331, 1299–1302.

Stille, P., Schaltegger, U., 1996. The lithospheric mantle beneath central Europe: Nd isotopicconstraints for its Late Proterozoic enrichment and implications for early crustalevolution. Geoph. Monog. Series 95, 269–276.

Stuiver, M., Reimer, P.J., 1993. Extended 14C data base and revised CALIB 3.0 14C agecalibration program. Radiocarbon 35, 215–230.

Stumpf, R., Frank, M., Schönfeld, J., Haley, B.A., 2011. Climatically driven changes in sedimentsupply on the SW Iberian shelf since the Last Glacial Maximum. Earth Planet. Sci. Lett.312, 80–90.

Stuut, J.-B., Smalley, I., O'Hara-Dhand, K., 2009. Aeolian dust in Europe: African sourcesand European deposits. Quat. Int. 198, 234–245.

Sumner, G., Homar, V., Ramis, C., 2001. Precipitation seasonality in Eastern and Southerncoastal Spain. Int. J. Climatol. 21, 219–247.

Swezey, C., 2001. Eolian sediment responses to late Quaternary climate changes: temporaland spatial patterns in the Sahara. Palaeogeogr. Palaeoclimatol. 167, 119–155.

Tanaka, T., Togashi, S., Kamioka, H., Amakawa, H., Kagami, H., Hamamoto, T., Yuhara, M.,Orihashi, Y., Yoneda, S., Shimizu, H., Kunimaru, T., Takahashi, K., Yanagi, T., Nakano, T.,Fujimaki, H., Shinjo, R., Asahara, Y., Tanimizu, M., Dragusanu, C., 2000. JNdi-1: aneodymium isotopic reference in consistency with La Jolla neodymium. Chem. Geol.168, 279–281.

Thompson, L.G., Mosley-Thompson, E., Davis, M.E., Henderson, K.A., Brecher, H.H.,Zagorodnov, V.S., Mashiotta, T.A., Lin, P.-N., Mikhalenko, V.N., Hardy, D.R., Beer, J.,2002. Kilimanjaro ice core records: evidence of Holocene climate change in tropicalAfrica. Science 298, 589–593.

Todt, W., Cliff, R.A., Hanser, A., Hofmann, A.W., 1996. Evaluation of a 202Pb–205Pb doublespike for high-precision lead isotope analysis. In: Basu, A., Hart, S.R. (Eds.), Earthprocesses: reading the isotopic code. Geophysical Monograph Series 95. AmericanGeophysical Union, Washington, DC, pp. 429–437.

Torres-Padrón, M.E., Gelado-Caballero, M.D., Collado-Sánchez, C., Siruela-Matos, V.F.,Cardona-Castellano, P.J., Hernández-Brito, J.J., 2002. Variability of dust inputs to theCANIGO zone. Deep-Sea Res. Part. II 49, 3455–3464.

Trigo, R.M., Osborn, T.J., Corte-Real, J.M., 2002. The North Atlantic oscillation in"uence onEurope: climate impacts and associated physical mechanisms. Clim. Res. 20, 9–17.

Tütken, T., Eisenhauer, A., Wiegand, B., Hansen, B.T., 2002. Glacial–interglacial cyclesin Sr and Nd isotopic composition of Arctic marine sediments triggered by theSvalbard/Barents Sea ice sheet. Mar. Geol. 182, 351–372.

Wanner, H., Bronnimann, S., Casty, C., Gyalistras, D., Luterbacher, J., Schmutz, C.,Stephenson, D.B., Xoplaki, E., 2001. North Atlantic Oscillation— concepts and studies.Surv. Geophys. 22, 321–382.

Wasserburg, G.J., Jacousen, S.B., DePaolo, D.J., McCulloch, M.T., Wen, T., 1981. Precisedetermination of Sm/Nd ratios, Sm and Nd isotopic abundances in standard solutions.Geochim. Cosmochimi. Act 12, 2311–2323.

Wehausen, R., Brumsack, H.-J., 2000. Chemical cycles in the Pliocene sapropel-bearing andsapropel-barren eastern Mediterranean sediments. Palaeogeogr. Palaeoclimatol. 158,325–352.

Weldeab, S., Emeis, K.-C., Hemleben, C., Siebel, W., 2002a. Provenance of lithogenic surfacesediments and pathways of riverine suspended matters in the Eastern MediterraneanSea: evidence from 143Nd/144Nd and 87Sr/86Sr isotope. Chem. Geol. 186, 139–149.

Weldeab, S., Emeis, K.-C., Hemleben, C., Schulz, H., Vennemann, T.W., 2002b. Sr, Ndisotope composition of Late Pleistocene sapropels and non-sapropel sediments fromthe Eastern Mediterranean Sea: implications for detrital in"ux and climatic conditionsin the source areas. Geochim. Cosmochimi. Act 66, 3585–3598.

Weldeab, S., Siebel, W., Wehausen, R., Emeis, K.-C., Schmiedl, G., Hemleben, C., 2003. LatePleistocene sedimentation in the Western Mediterranean Sea: implications forproductivity changes and climatic conditions in the catchment areas. Palaeogeogr.Palaeoclimatol. 190, 121–137.

250 M. Rodrigo-Gámiz et al. / Chemical Geology 410 (2015) 237–250