Petrographic and isotopic evidence for Holocene long-term climate change and shorter-term...

Palaeogeography, Palaeoclimatology, Palaeoecology xxx (2011) xxx–xxx

PALAEO-05716; No of Pages 13

Contents lists available at ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology

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

Petrographic and isotopic evidence for Holocene long-term climate change andshorter-term environmental shifts from a stalagmite from the Serra do Courel ofnorthwestern Spain, and implications for climatic history across Europe andthe Mediterranean

L. Bruce Railsback a,⁎, Fuyuan Liang b, Juan Ramón Vidal Romaní c, Aurora Grandal-d'Anglade c,Marcos Vaqueiro Rodríguez c, Luisa Santos Fidalgo c, Daniel Fernández Mosquera c,Hai Cheng d,e, R. Lawrence Edwards e

a Department of Geology, University of Georgia, Athens, GA 30602-2501, USAb Department of Geography, Western Illinois University, 1 University Circle, Macomb, IL 61455, USAc Universidade da Coruña, Instituto Universitario de Xeoloxia, Coruna, Spaind Institute of Global Environmental Change, Xi'an Jiaotong University, Xi'an, Shaanxi 710049, Chinae Department of Geology and Geophysics, University of Minnesota, Minneapolis, 55455, USA

⁎ Corresponding author. Tel.: +1 706 542 3453; fax:E-mail address: [email protected] (L.B. Railsback).

0031-0182/$ – see front matter © 2011 Elsevier B.V. Aldoi:10.1016/j.palaeo.2011.02.030

Please cite this article as: Railsback, L.B., etenvironmental shifts from a stalagmite..., P

a b s t r a c t
a r t i c l e i n f o
Article history:Received 17 November 2010Received in revised form 22 February 2011Accepted 24 February 2011Available online xxxx

Keywords:HoloceneStalagmitePalaeoclimateIsotopesPetrologySpain

A small stalagmite from Cova da Arcoia in the Serra do Courel of northwestern Spain provides a record both ofgeneral climate change across the Holocene and of shorter-term shifts between wetter and drier conditions.Nineteen U-series dates from 9.3 ka to the present provide a chronology of the stalagmite with uncertaintiesof commonly only decades and at the most only 200 years. These dates constrain the age of 18 petrographicsurfaces that are of two types, one (Type E) inferred to result from dissolution during enhanced flow of waterand the other (Type L) inferred to result from lesser flow. Aragonite is more abundant in the earlier parts ofthe stalagmite and calcite is more abundant in the later part. Values of δ13C range from −5.6 to −1.3‰relative to VPDB but are greatest beneath Type L surfaces. Values of δ18O range from −5.4 to −4.4‰ relativeto VPDB and have a weak but statistically significant correlation with δ13C.Greater abundance of aragonite, greater values of δ13C, and lesser frequency of Type E surfaces early in thestalagmite suggest a drier and perhaps warmer early to middle Holocene, whereas greater abundance ofcalcite and frequent Type E surfaces higher in the stalagmite suggest wetter conditions later in the Holocene.δ18O data are most compatible with this general conclusion. However, a prominent Type E surface records the8.2 ka event as an isolated wetter episode in the otherwise-dry early Holocene.Much of the stalagmite's aragonite and calcite suggests normal stalagmite growth in moderate climaticconditions. However, the near or complete cessation of growth at Type L surfaces and greater δ13C, greaterδ18O, and lesser fluorescence of aragonite beneath those surfaces combine to suggest at least six century-scaleepisodes of strikingly dry conditions. On the other hand, evidence of dissolutional erosion at Type E surfacessuggests at least ten distinct episodes of much wetter conditions. These combine to suggest highly variableclimate across the Holocene at the scale of decades to a few centuries.The record from this stalagmite is similar to records from Greenland ice cores and from pollen and/orstalagmite records from France, northernmost Spain, and more broadly from northern or Atlantic-facingEurope. In contrast, it suggests trends nearly opposite those in records from eastern and southern Spain,suggesting that the Serra do Courel is north of, but perhaps not far from, the boundary between Atlantic andMediterranean regions of Holocene climate trends in the Iberian Peninsula. That general antiphasalrelationship between northwestern Spain and regions to the south and east extends to the specificobservation that four of the most pronounced wetter events in the stalagmite's record coincide with periodsof widespread drought and/or cultural collapse in the Mediterranean and Middle East.

+1 706 542 2425.

l rights reserved.

al., Petrographic and isotopic evidence for Hoalaeogeogr. Palaeoclimatol. Palaeoecol. (2011

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

At the broadest scale, the Holocene is considered a period ofrelative climate stability, in contrast to the variability in climate from

locene long-term climate change and shorter-term), doi:10.1016/j.palaeo.2011.02.030

http://dx.doi.org/10.1016/j.palaeo.2011.02.030

mailto:[email protected]


http://www.sciencedirect.com/science/journal/00310182


2 L.B. Railsback et al. / Palaeogeography, Palaeoclimatology, Palaeoecology xxx (2011) xxx–xxx

interglacial to glacial maximum in the 100,000 years that precededthe Holocene. However, significant climate change has occurred in theHolocene (Roberts, 1998; Oldfield, 2003), and recent work hasdocumented significant temporal and geographic variation of majorclimatic intervals within the Holocene (e.g., Davis et al., 2003; Fisherand Koerner, 2003; Mayewski et al., 2004). Furthermore, work of thelast few decades has documented that some climate change has beenremarkably abrupt (e.g., Berger and Labeyrie, 1987; National ResearchCouncil, 2002).

As a contribution to the emerging picture of climate change in theHolocene, this paper reports on a stalagmite from northwestern Spainthat spans the last ten thousand years. Stalagmites have manycharacteristics from which palaeoclimate can be inferred (Gascoyne,1992). These include radiometric age (Harmon et al., 1975; Edwardset al., 1987; Lauritzen, 2003), variation of C and O stable isotopes(McDermott, 2004), trace element chemistry (Verheyden, 2004;Fairchild and Treble, 2009), UV-stimulated fluorescence (Shopovet al., 1994; Baker et al., 1998), pollen content (McGarry andCaseldine, 2004), varying detrital content (Railsback et al., 1999;Webster et al., 2007), petrographic relationships (Railsback, 2000;Turgeon and Lundberg, 2001; Frisia et al., 2002) and, perhaps mostfundamentally, changing mineralogy (González and Lohmann, 1988;Railsback et al., 1994; Frisia et al., 2002). With regard to the twoprincipal CaCO3 minerals, aragonite and calcite, many studies havelinked precipitation of aragonite to more extensive evaporation(Murray, 1954; Pobeguin, 1965; Siegel, 1965; Siegel and Dort, 1966;Thrailkill, 1971; Cabrol and Coudray, 1982) and to higher temperature(Burton and Walter, 1987; Morse et al., 1997) than the conditions inwhich calcite forms. The stalagmite described herein contains bothcalcite and aragonite in layers bounded by critical petrographic

Fig. 1. A. Map of the Iberian Peninsula showing two major rivers mentioned in text and locStalagmite ESP03 was taken.

Please cite this article as: Railsback, L.B., et al., Petrographic and isotopicenvironmental shifts from a stalagmite..., Palaeogeogr. Palaeoclimatol. P

surfaces and characterized by abrupt shifts in C isotope composition.These characteristics, combined with 19 radiometric ages, provide arecord of both long-term trends and shorter shifts in climate in theAtlantic-facing highlands of northwestern Spain during the Holocene.

2. Setting

The stalagmite described here, which was designated StalagmiteESP03 in the Sedimentary Geochemistry Laboratory of the Universityof Georgia (U.S.A.), comes from Cova da Arcoia in the Serra do Courelin the westernmost Cordillera Cantabrica (Fig. 1). Cova da Arcoia is at42° 36′ 44.39″ N latitude and 7° 05′ 06.32″ W longitude, near thevillage of Céramo in the province of Lugo, in the easternmost portionof the autonomous community of Galicia, in northwestern Spain. Covada Arcoia sits at 1240 masl, whereas the summits upslope from thecave have elevations of 1500 to 1600 m, and the valley floors beloware at 900 to 1100 m. The cave is on a north-facing slope.

Mean annual precipitation in the region around Cova da Arcoia isabout 2500 mm (Leira and Santos, 2002), with most precipitation inthe cooler months (Walter and Lieth, 1960). The native vegetationrepresents a transition from Eurosiberian to Mediterranean (IzcoSevillano et al., 1982, cited in Leira and Santos, 2002), although thepresent landscape is largely scrubland and pasture (Leira and Santos,2002).

The whole zone has been affected during the Pleistocene by theQuaternary glaciations, developing the main karstic systems underthe ice accumulations and helped by the subglacial melt waters. Thefauna that occupied the territory during the Upper Pleistocene toHolocene, including the human species, is known in detail. TheHolocene fossil record in Arcoia and surrounding caves includes

ation of area in Part B. B. Map of the area around Cova da Arcoia, the cave from which

evidence for Holocene long-term climate change and shorter-termalaeoecol. (2011), doi:10.1016/j.palaeo.2011.02.030


3L.B. Railsback et al. / Palaeogeography, Palaeoclimatology, Palaeoecology xxx (2011) xxx–xxx

Brown bear (Ursus arctos) and aurochs (Bos primigenius) in the earlyHolocene, along with Epipaleolithic human remains (Grandal-d'Angladeet al., 2010). Domestic faunas (mainly cattle, sheep and goat) are themost common finds of Middle to Late Holocene age in the region. Brownbear have been found in Arcoia and Sima Paleira caves, both at highaltitude, since the Pleistocene. There is a hiatus in the macromammalfossil record during the glacial maximum. Remains of Holocene brownbears have been found in several caves in the area (Purruñal, Tarelo,Sima Paleira and A Tara), but they only reoccupied caves at highaltitude (Sima Paleira) in the climatic optimum, after the cold episodeat 8.2 ka (Grandal-d'Anglade et al., 1997; Vidal Romaní et al., 2010;García-Vázquez et al., in press).

The stalagmite was located about 50 m from the present caveentrance, at the back of a room about 8 m by 10 m in size. It was lessthan 10 m from an original cave entrance that was sealed by a collapsethat took place more than 40 kyr BP. The stalagmite had grown on thetop of one of the limestone blocks that had fallen in the collapse of theoriginal entrance. That collapse buried the bones of Ursus arctos, thebrown bear dated at more than 40 kyr (Grandal-d'Anglade et al.,2010), which yields the minimum age for the collapse of the originalentrance, long before the deposition of Stalagmite ESP03. Thethickness of the rock over the cave at the stalagmite's location is atmost 5 m.

The pH of the drip water onto the stalagmite at the time ofcollectionwas 8.25, and the temperature of that waterwas 8.1 °C. Dripwater concentrations of Mg and Ca were 0.32 and 47.9 mg/kgrespectively, yielding a molar Mg/Ca ratio of 0.011. Meteoricprecipitation in the region has values of δ18O that range frommonthlyaverages of −9 to −6‰ relative to SMOW during cooler and wettermonths and averages of −6 to −3‰ in warmer and drier months(Walter and Lieth, 1960; IAEA, 2006), suggesting a weighted averageof −6 to −7. Domínguez-Villar et al. (2008) showed that δ18O ofrainwater increases with temperature in the northwestern IberianPeninsula, and most relevantly at the inland stations of Léon andBragança, each of which is about 130 km from Cova da Arcoia. TheirTable 3 also shows that δ18O of rainwater decreases with increasingrainfall in the region and thus shows a typical “amount effect”.

Table 1Uranium and thorium isotopic compositions and 230Th ages for Stalagmite ESP03.

mm fromtop

SampleID

Weightg

238Uppb

232Thppt

δ234Umeasureda

0.9 ESP03-T 0.168 157±1 12,700±100 40±61.8 ESP03-S 0.158 10.0±0.1 635±5 −221±265.2 ESP03-R 0.131 22.4±0.1 1560±10 −81±88.8 ESP03-Q 0.102 518±2 1594±8 59±411.3 ESP03-N 0.090 1170±6 1436±9 69±515.2 ESP03-M 0.155 808±4 1826±6 64±518.5 ESP03-6 0.113 2200±14 2897±9 94±723.3 ESP03-H 0.055 684±4 1375±13 106±624.8 ESP03-G 0.129 18.2±0.1 987±6 83±930.7 ESP03-F 0.152 297±1 1115±5 95±632.8 ESP03-E 0.143 1659±13 4410±20 125±738.8 ESP03-C 0.150 1117±8 2449±9 118±744.2 ESP03-5 0.164 2351±14 1759±6 140±649.3 ESP03-4 0.187 2104±14 1604±5 98±756.2 ESP03-B 0.177 1178±8 6570±30 121±758.0 ESP03-3 0.075 1010±5 5090±40 101±659.6 ESP03-2 0.080 1697±11 4230±10 93±760.7 ESP03-A 0.148 1827±17 5250±19 101±866.4 ESP03-1 0.080 3505±27 3610±20 102±6

Analytical errors are 2σ of the mean. U and Th decay constants are the same as those descra δ234U=([234U/238U]activity−1)×1000. δ234Uinitial (corrected) was calculated based onb [230Th/238U]activity=1−e−λ230T+(δ234Umeasured/1000)[λ230/(λ230−λ234)](1−e−(λ

c The degree of detrital 230Th contamination is indicated by the [230Th/232Th] atomic ratid Age corrections were calculated using an average crustal 230Th/232Th atomic ratio of 4.4

crustal 232Th/238U value of 3.8. The errors are arbitrarily assumed to be 50%.


3. Methods

Nineteen samples of ca. 100 mg eachwere drilled along the centralgrowth axis of the ESP03 stalagmite and dated by inductively coupledplasma mass spectrometry (ICP-MS) in the Stable Isotope Laboratoryat the University of Minnesota following procedures described inEdwards et al. (1987) and Cheng et al. (2000). The dating results werecalculated using half-lives determined by Cheng et al. (2000) andreported with analytical errors of 2σ of the mean (Table 1). Wecorrected for initial 230Th, using a generic “bulk earth” atomic 230Th/232Thratio of 4.4×10−6.We have assumed an error of ±50% in this value. Theerror in the “corrected” age incorporates this uncertainty. This errorestimate is robust toward lower 230Th/232Th values, as zero is thetheoretical limit, and documented values under 2×10−6 are rare.Whereas 230Th/232Th values higher than our upper limit have beendocumented, our estimate in this direction is reasonable.

For C and O stable isotope analyses, 104 powdered samplesnumbered from 007 to 110 from lowest to highest were taken fromthe stalagmite with a dental handpiece and fine dental bur. X-raydiffraction showed that four of these samples contained mixtures ofcalcite and aragonite, and so they were discarded. 100 samples thusremained and are reported here. The layers of the stalagmite are toosmall to allow drilling of samples that can be confidently assumed tobe coeval, and so no Hendy test (Hendy, 1971) was performed.

C and O stable isotope analyses were performed by a methodmodified from McCrea (1950). Samples weighing roughly 5 mg werereacted under vacuum in 100% phosphoric acid at 50 °C. The resultingCO2 was extracted on a conventional vacuum line and analyzed onFinnigan MAT Delta E and MAT 252 mass spectrometers. Laboratorystandards were prepared and analyzed with each batch of samples.These standards have been calibrated to NBS-19 (δ13C=+1.95,δ18O=−2.2‰ relative to VPDB) and NBS-18 (δ13C=−5.0 andδ18O=−23.0‰ relative to VPDB). Sample isotopic results arenormalized to the lab standards using a two-point scale, so that allδ13C and δ18O values herein are reported relative to VPDB. The 2-sigma error of the combined extraction and analysis is 0.04‰ for δ13Cand 0.05‰ for δ18O.

[230Th/238U]activityb

[230Th/232Th]ppmc

Ageuncorrected

Agecorrectedd

δ234Uinitial

correcteda

0.0215±0.0047 4±1 2300±500 −10±1300 40±60.0270±0.0099 7±3 3900±1400 1450±1900 −222±260.0275±0.0062 7±1 3300±800 1100±1400 −82±80.0145±0.0003 78±2 1500±30 1420±50 59±40.0250±0.0003 335±4 2580±30 2540±40 70±50.0306±0.0003 223±2 3180±40 3120±50 64±50.0351±0.0003 439±3 3550±40 3520±40 95±70.0377±0.0004 309±4 3780±50 3730±60 107±60.0467±0.0056 14±2 4800±600 3300±900 84±90.0487±0.0005 214±2 4960±60 4860±80 96±60.0539±0.0006 334±3 5350±70 5280±80 127±70.0582±0.0006 437±4 5820±80 5760±80 120±70.0694±0.0005 1528±8 6840±70 6820±70 143±60.0703±0.0005 1521±7 7210±70 7190±80 100±70.0788±0.0009 233±2 7940±100 7800±130 123±70.0760±0.0022 249±7 7800±240 7660±250 103±70.0838±0.0007 555±3 8700±100 8620±100 95±80.0872±0.0010 500±4 9000±120 8910±130 103±80.0903±0.0008 1450±10 9300±110 9280±110 105±6

ibed in Cheng et al. (2009a, 2009b).230Th age (T), i.e., δ234Uinitial=δ234Umeasured×eλ234⁎T, and T is corrected age.

230−λ234) T), where T is the age.o instead of the activity ratio.×10−6±2.2×10−6. Those are the values for a material at secular equilibrium, with the



Fig. 3. Exterior of Stalagmite ESP03. Arrows point to rills inferred to be the result ofdissolution. Dashed line indicates plane along which stalagmite was sectioned forfurther analysis.


Mineralogy of powdered samples was determined using a BrukerD8 X-ray Diffractometer in the Department of Geology of theUniversity of Georgia. Samples were scanned from 20 to 65°2Θusing CoKα radiation with scan times of 7 min.

4. Stalagmite characteristics

4.1. General description

Stalagmite ESP03 is a small, nearly hemispherical stalagmite only69 mm tall (Fig. 2). The brown outer surface of the stalagmite isincised by dissolutional rills descending from its top to its sides(Fig. 3), indicating that it was not precipitating when it was collectedand that in fact it had undergone a significant period of dissolutionsince its most recent deposition. X-ray diffraction analysis indicatesthat the stalagmite consists of both spelean aragonite and calcite, witha petrography suggesting that both the calcite and aragonite areprimary and lack the characteristics noted in secondary speleanminerals (Railsback et al., 1997, 2002). The stalagmite also containsminor detrital clay and Fe oxides or hydroxides in particles as much as0.07 mm in size (in the size range of very fine sand), a few grains ofdetrital calcite 0.16–0.34 mm in diameter (in the size range of fine- tomedium-grained sand), and at least one grain of quartz that is 0.2 mmin diameter (in the size range of fine sand). The size of these particlesis suggestive of deposition by water moving over the stalagmite,rather than as an aerosol. The stalagmite also contains at least onephosphatic grain that is likely a piece of bone.

4.2. Layer-bounding surfaces or discontinuities

Layers of calcite and aragonite in Stalagmite ESP03 range inthickness from 0.04 to 0.28 mm and are cut by at least 18 layer-bounding surfaces or discontinuities, which can generally becategorized in two types (Fig. 4). One type, designated “Type E” for“erosion”, consists of surfaces beneath which layers are truncatedlaterally as the result of removal of spelean CaCO3. This removal iscommonly greatest, or most observable, at the crest of the stalagmite,so that layers present on the stalagmite's side are not present towardthe center (Fig. 5A and B). At one surface, designated Surface 15because it is the fifteenth from the bottom of the stalagmite, thisremoval includes material beneath an overhanging remnant of thetruncated layer and thus constitutes a cavity that is the microscopicanalog of a cave (Fig. 5A). Its morphology thus cannot have originatedby physical abrasion and instead is seemingly the result of dissolution.At another notable surface, Surface 10, straight-sided remnants of thetruncated layer project upward from, and locally hang over, the

Fig. 2. A. Scanned image of Stalagmite ESP03 in reflected light. B. Photograph of Stalagmitgeneralized intervals of aragonite and calcite, respectively. C. Detail of B, with numbers indicain Figs. 7 and 8.


eroded surface and thus constitute features that are the microscopicanalogs of mesas or hoodoos (Fig. 5C). Type E surfaces are commonlycoated with detrital clays and/or Fe oxides or hydroxides (Table 2),suggestive of enhanced flow sufficient to carry detrital particles. AtSurface 2, a layer of calcite is preserved beneath a detrital layer(Fig. 5F), but that calcite layer has been removed where the detritallayer is not present, further suggesting that removal was caused bydissolving waters that were ineffective where a sealing layer of non-carbonate detritus blocked the attack of water running over thestalagmite. All of these considerations suggest that Type E surfacesresulted from greater flow of water across the stalagmite.

The second type of layer-bounding surface, designated “Type L” for“lesser deposition”, consists of surfaces beneath which layers areevident near the crest of the stalagmite but do not extend to the flanksof the stalagmite (Figs. 4B, 5D and E). These layers of lesser lateralextent thin gradually to the sides of the stalagmite, rather than beingabruptly truncated laterally. Type L surfaces thus have, in the mostgeneral sense, a geometry opposite that of Type E surfaces. Theincomplete layers in a succession beneath a Type L surface typicallybecome laterally smaller upwards and are capped by one overlyinglayer that extends from one side of the stalagmite to the other

e ESPO3 when fluorescing in response to ultraviolet radiation. “Ar” and “Cc” indicateting layers correlative with stable isotope samples. Numbers are the same as those used



Fig. 4. Generalized sketches of layer-bounding surfaces in Stalagmite ESP03. A. A Type Esurface, for which the defining characteristic is lateral truncation of underlying layers,commonly toward the crest of stalagmite. Many such surfaces have a coating of detritalclay (suggested here by the dark layer). B. A Type L surface, for which definingcharacteristics are underlying layers that thin toward the side of the stalagmite and thatare smaller in area in the progression upward to the layer-bounding surface. Shading ofthis surface is only to emphasize its location, not detrital material. Fig. 5 shows theillustrations of both kinds of surfaces.

Fig. 5. Mosaics of a total of roughly 85 photomicrographs of selected layer-boundingsurfaces of discontinuity in Stalagmite ESP03. A and B: Right and left sides, respectively,of Surface 15, a Type E surface. Arrows point to top of truncated aragonite. Two thickarrows point to void space at right where removal of aragonite has generated amicroscopic cave in the stalagmite. C. Surface 10, a Type E surface. Arrows point to topof truncated calcite. Note mesa-like erosional remnants and dark detrital coating.D. Surfaces 7, a Type L surface, and 8, a Type E surface. For Surface 8, arrow points to topof truncated calcite. Doubly arrowed line at right points to interval missing to left at


(Fig. 4B). Type L surfaces are thus boundaries below which thestalagmite became thinner before overlying layers resumed growth atthe more typical width of the stalagmite. This is compatible withdiminishing flow of water onto the stalagmite (Dreybrodt, 1999), as isthe observation that Type L surfaces commonly have no detritalcoating, or at most very minor ones (Table 2). Thus Type E and Type Lsurfaces seem to be opposite not only in their geometry but also intheir implications regarding passage of water across the stalagmite.Furthermore, Type L surfaces are found only in aragonite, whereasType E surfaces are commonly (but not always) associated withcalcite (Table 2).

A Type L surface is not necessarily a surface at which record ismissing at the crest of the stalagmite. The geochronological data arecompatible with the assumption that there is no missing record atType L surfaces. By contrast, the geochronological data imply missingrecord at most if not all of the Type E surfaces, as discussed below.

Surface 7. Three circled numbers indication positions correlative with stable isotopesamples, using the same number system as in Figs. 7 and 8. E. Surfaces 4, a Type Lsurface, and 5, a Type E surface. Upward-pointing small arrows beneath Surface 5 pointto incised portions of Surface 5 cut lower than rest of surface, which is marked withdownward-pointing arrows. Circled numbers indicate stable isotope samples labeled inFig. 7. F. Surface 2, a Type E surface. Arrows point to top of truncated calcite. Note darkcoating of detrital clay on surface. Sketch between C and D indicates positions offeatures shown.

5. Construction of a chronology

The nineteen radiometric dates and their associated estimates oferror provide a record in which ages are consistently youngerupwards through Stalagmite ESP03 (Fig. 6). Five ages (ESP03-3,G, R,S, and T) have exceptionally large estimates of error, which areassociated with small U concentrations in all five cases, small 230Th/232Th ratios in the last four, and a large 232Th concentration in the last,and thus uppermost, case (Table 1). Of the other fourteen ages, thethirteen oldest have error estimates less than 2% of the age, and theyoungest of the fourteen (ESP03-Q) has a similarly small estimateerror, and it has a large percentage error only because of its young age(Table 1).

Three intervals in which growth rates can be unambiguouslyestimated (Intervals PI1, PI2, and PI4) yield growth rates of 0.010,0.012, and 0.012 mm/yr. Application of a growth rate of 0.011 mm/yrto the rest of the stalagmite (Fig. 6) yields an internally consistentscheme, with no assignment of two intervals of the stalagmite to oneage. This scheme also yields and requires unrepresented intervals oftime beneath each Type E surface, and thus beneath each surface atwhich petrography predicts an unrepresented interval of timebecause of dissolutional erosion of previously deposited speleanCaCO3. As Fig. 6 shows, growth rate is well constrained early in thestalagmite's history. It is less well constrained later, but for the sake of


consistency, and for lack of a better estimate, the same rate is used inFig. 6 throughout the stalagmite.

The scheme in Fig. 6 implies eleven preserved intervals in thestalagmite, which are designated PI1 to PI11. The timing of PI1, PI2,PI4, PI5, PI7, PI8, and PI9 are tightly constrained by the data. Thetiming of PI3 and PI6 have uncertainties in the range of decades andabout 200 years, respectively, because they are so thin that no datecould be obtainedwithin them. The timing of the end of PI10 and boththe beginning and end of PI11 have uncertainties of as much as200 years (2.1% of the time of the entire ESP03 record) because of thelarge errors in ages ESP03-R, S, and T, the three uppermost ages in thestalagmite (Fig. 6).

6. Stable isotope data

Values of δ13C of CaCO3 in the 100 samples from Stalagmite EPSO3have a large range, from −5.6 to −1.3‰ relative to VPDB. Values of



Table 2Layer-bounding surfaces (discontinuities) in Stalagmite ESP03.

Surface Type Mineralogya Comments

Below Above

18 E C – Top of stalagmite: dissolutional rills17 E C C No detrital coating16 E C A followed

by CLocally thick detrital coating

15 E A C Much topography, with a microscale “cave”14 L A A No detrital coating13 L A A Very minor partial detrital coating12 E A A Minor and partial detrital coating11 L A A No detrital coating10 E C A Thick detrital coating on pronounced surface9 E C C Partial detrital coating8 E A C Local thin detrital coating, with phosphate

grain lying on surface7 L A A No detrital coating6 E A A Very localized thin detrital coating with two

detrital carbonate grains lying on surface5 E A A Evidence of dissolution at only two points4 L A A No detrital coating3 L A A No detrital coating2 E C and A A Thick detrital coating where calcite survives1 L A A No detrital coating

a C = calcite; A = aragonite.


δ13C in aragonite span that entire range, whereas values of δ13C incalcite have a smaller range, from −5.4 to −4.6‰ vs. VPDB.

Values of δ13C have pronounced relationships to the petrographicsurfaces described above. Perhaps most strikingly, greatest values ofδ13C are found below Type L surfaces. For example, Stable IsotopeSample 59, which has the greatest value of δ13C, is just below SurfaceS7, and Stable Isotope Samples 20, 23 to 25, and 39 to 41 are from justbelow Surfaces S1, S3, and S4 respectively (Fig. 7). On the other hand,

Fig. 6. Diagram used to develop chronology of Stalagmite ESP03. Data regarding U-series dexamples are shown in Fig. 5. Growth rate is well constrained in preserved intervals (PI) 1,used in the later intervals as well. Light- and dark-shaded intervals are shown in Fig. 7.


values of δ13C beneath Type E surfaces are moderate to low (forexample, Sample 94 beneath Surface S15, Samples 72 and 71 beneathSurface S10, Sample 70 beneath Surface S9, Sample 60 beneathSurface S8, and Sample 46 beneath Surface S5). The smallest values ofδ13C are found in the middles of intervals of aragonite (e.g., Samples85, 52, and 31) (Fig. 7).

Values of δ18O of the same 100 samples of CaCO3 in StalagmiteEPSO3 have a small range, only from−5.4 to−4.4‰ relative to VPDB.Values of δ18O in aragonite range from −5.3 to −4.4, and values ofδ18O in calcite have a similar range, from−5.4 to−4.6. Values of δ18Oare only weakly correlative with values of δ13C: among samples ofcalcite r2=0.25, and among samples of aragonite r2=0.15. However,both correlations are statistically significant in that they have lowprobabilities of arising randomly (for calcite p=0.01 and foraragonite p=0.0005). Among samples from aragonite, this relation-ship results in part from the absence of any samples with small δ18Oand large δ13C. Furthermore, local maxima in δ18O coincide with largevalues of δ13C found beneath Type L surfaces in Samples 59, 25, and20, and local minima in δ18O correspond with small values of δ13C inSamples 31, 28, and 19 (Fig. 7).

7. Discussion, part I: inferences about palaeoclimate in the Serrado Courel

7.1. Long-term trends

Times of preserved deposition and times of dissolutional erosionare not distributed evenly through the ESP03 record. For example, thelongest interval of uninterrupted deposition (PI3) comes early in thehistory of the stalagmite, from 8000 to 6500 years BP and thus duringthe Holocene Climatic Optimum (Figs. 6 and 7). The next two longestintervals of uninterrupted deposition are PI1 and PI4, likewise in thefirst half of the Holocene. However, PI1 and PI2 were separated by an

ates are given in Table 1. Type E and Type L surfaces are defined in text and in Fig. 4;2, and 4; both for consistency and for lack of better knowledge, the same growth rate is



Fig. 7. Data from Galician stalagmite ESP03 in the context of data from other palaeoenvironmental records. A. Plot of abundance of ice-rafted debris in the North Atlantic (Bond et al.,2001). B. Plot of stable isotope data fromGreenland ice cores from Johnsen et al., 2001. C. Markers for preserved intervals from Fig. 6, where each is designated PI1, PI2, etc. D. Markersfor surfaces listed in Table 2 and represented in Fig. 6. Thicker bars, rather than lines, indicate surfaces at which a significant hiatus is inferred. “E”s indicate Type E surfaces; longervertical black lines also represent Type L surfaces. E. Stable isotope data from ESP03. Small numerals next to symbols for stable isotope data indicate numbers of specific samplesmentioned in text; these numbers are also used in Figs. 2, 5, and 8. Vertical bars showing intervals of calcite with lesser values of δ18O at Kaite Cave are from Domínguez-Villar et al.(2009). F. Qualitative interpretation of changing climatic conditions, as suggested by Fig. 8. G. Selected pollen data from Laguna Lucenza, a lake near the cave from which StalagmiteESP03 was taken (see Fig. 1).


extensive period of erosion represented by Surface 2 and correlativewith the 8.2 ka event (Dansgaard, 1987; Johnsen et al., 2001).

On the other hand, intervals of non-deposition and/or erosion arecommon later in the history of the stalagmite. For example, over thepast 5100 years, the combined time of hiatuses in ESP03 is greaterthan the combined time represented by the preserved intervals. Innumber, Type L surfaces dominate over the interval prior to2500 years ago, but in the interval from 2500 years ago to the present,the last four layer-bounding surfaces are all Type E surfaces at whichdissolutional erosion is inferred (Fig. 8), and they include Surface S15,the most pronounced surface of dissolution (Fig. 5A and B). The mostrecent Type E surface (S18) is the dissolutionally rilled outer surface ofthe stalagmite at the time of its collection (Fig. 3).


Spelean mineralogy is likewise not evenly distributed throughoutStalagmite ESP03. Aragonite dominates as the CaCO3 mineral over theinterval prior to 2500 years ago and is present only as one minor layerthereafter (Fig. 6). Calcite, on the other hand, is more abundant inlater parts of the stalagmite and, across the interval prior to5000 years ago, is present only as one thin and partial layer belowSurface 2.

These trends in deposition, dissolution, and mineralogy, com-bined with previous observations that aragonite commonly formsunder warmer and/or more evaporative conditions and that calciteforms under cooler and/or less evaporative conditions, support thenotion of environmental change inside Cova da Arcoia that weassume reflects changes in rainfall, or more precisely in “effective



Fig. 8. Schematic cartoons illustrating three inferred modes of deposition or modification of Stalagmite ESP03. A. “Normal” precipitation of stalagmite CaCO3. B. Dissolution inferredto have occurred during periods of greater passage of water through cave. C. Decrease or cessation of growth inferred to have occurred during periods of lesser passage of waterthrough cave. Numbers of stable isotope samples shown at bottom are numbers used in Figs. 2, 5, and 7; numbers of surfaces shown at bottom are numbers used in Figs. 5, 6, and 7and in Table 2.


rainfall”, in the region. Specifically, they suggest a warmer and/ordrier early Holocene and a cooler and/or wetter later Holocene,consonant with trends in the Greenland ice cores (as summarized byJohnsen et al., 2001; Fisher and Koerner, 2003). The greaterfrequency of dissolutional erosional events later in the Holocenealso suggests a wetter climate than that during the Holocene ClimaticOptimum. Furthermore, changes in the ESP03 δ18O data, althoughsmall, are compatible with warmer and more evaporative conditionsduring aragonite deposition, as opposed to cooler and wetterconditions, broadly similar to those of the present, during calcitedeposition (Fig. 9). The conclusion of a warmer early to middleHolocene from the ESP03 record is in accord with abundance ofQuercus (oak) and Corylus (hazel) pollen in sediments from nearbyLaguna Lucenza (Santos et al., 2000), because those are consideredthermophilic taxa. On the other hand, wetter and/or cooler condi-tions in at least the last two thousand years accord with increasedabundance of Ericaceae (heather and heath) and Betula (birch)pollen in the same record.


7.2. Shorter-term shifts

Variations in δ13C in speleothems can result from many environ-mental changes (McDermott, 2004), of which the most fundamentalis change in the abundance of vegetation driving input of 12C-enrichedCO2 to soil gas via root respiration and/or decay of dead organicmatter (e.g., Hesterberg and Siegenthaler, 1991; Baldini et al., 2005).The large range of δ13C values in Stalagmite ESP03 and the presence ofthe greatest values of δ13C beneath Type L surfaces suggests episodesin which climate became sufficiently dry that vegetative input of 12C-enriched CO2 was minimized (giving comparatively large values ofδ13C) and little water dripped onto the stalagmite (resulting in layersof diminished lateral extent) (Fig. 8C). Several other lines of evidencesupport this interpretation. For example, the association of Type Lsurfaces with aragonite, rather than calcite, is compatible with theirdevelopment during periods of greater evaporation, and the scarcityof detrital clay and absence of larger detrital grains at these surfaces islikewise compatible with their development when flow of water was



Fig. 9. Curves showing temperatures and values of δ18Ow of precipitating water in isotopic equilibrium with aragonite (A) and calcite (C) with average values of δ18OCaCO3 (thickercurves) andmaximum and minimum values (thinner curves) observed in aragonite and calcite of Stalagmite ESP03. Curves were generated using the fractionation equations of Kimand O'Neil (1997) and Kim et al. (2007). Purpose of diagram is not to infer specific temperatures of precipitation through time or specific values of δ18Ow through time, but instead toevaluate the feasibility of general scenarios shown.


not great. Also, UV-induced fluorescence in the aragonite of ESP03 isleast in intervals of greater δ13C (Figs. 2 and 7). Because greaterfluorescence is in general associated with more soil organic matterand wetter conditions (Shopov et al., 1994; Baker et al., 1998; vanBeynen et al., 2001), the inverse relationship between fluorescenceand δ13C further supports the inference that the greatest values of δ13Cin Stalagmite ESP03 represent dry periods in which vegetative inputof soil CO2 diminished. Coincidences in local maxima in δ18OCaCO3 withgreatest values of δ13C, as noted above, also suggest less rainfall and/ormore evaporation. These thoughts combine to suggest that the Type Lsurfaces are indicators of at least six distinct episodes of strikinglydrier climate, each lasting periods on the order of decades to roughlytwo hundred years. These episodes, represented by Surfaces 1, 3, 4, 7,11, and 14, occurred approximately 8600, 7600, 6700, 5200, 3600, and2500 years ago.

The opposite condition is seemingly represented by Type Esurfaces at which spelean CaCO3 was removed (Fig. 8B). The mostlikely condition for such dissolution would be rapid passage of waterthrough the overlying soil (allowing insufficient time for chargingwith vegetatively generated CO2), rapid passage through the bedrock


above the cave (allowing insufficient time for dissolution of and thusequilibrium with bedrock CaCO3), and rapid passage of water acrossthe stalagmite (allowing insufficient time for degassing of CO2 thatwould bring the water to equilibriumwith, or to supersaturation withrespect to, CaCO3). The association of both coatings of detrital clay andindividual larger detrital grains with Type E surfaces also suggestsrapid movement of water through conduits in the overlying bedrockand onto the stalagmite. The association of Type E surfaces with calciteis also compatible with wetter conditions before and/or after theirformation. In this respect, Surface 2 may be a particularly instructiveexample, in that sustained deposition of aragonite briefly gave way todeposition of calcite and isolated detrital grains (Fig. 5F), presumablyduring progressively wetter and/or cooler conditions, before dissolu-tional erosion removed spelean CaCO3 to give Surface 2.

If the petrography of Type E surfaces suggests that they formedduring wet periods, comparison of the timing of Type E surfaces withthe NorthGRIP ice core (Johnsen et al., 2001) supports the notion thatType E surfaces represent episodes of cooler climate as well. Moststrikingly, Surface 2 developed during the 8.2 ka event (Fig. 7), whichis a negative and thus presumably colder excursion in the Greenland




ice cores (Dansgaard, 1987; Johnsen et al., 2001) and which has beencorrelatedwith cool conditions in Europe (e.g., Alley et al., 1997; Prosset al., 2009). Similarly, the beginning and end of the hiatusesassociated Surfaces 5 and 15 coincide with less prominent negativeexcursions in the NorthGRIP record (Fig. 7). The period over the last2400 years, during which only Type E surfaces have formed, is theperiod in whichminima in δ18O in the NorthGRIP core reached−36‰relative to SMOW and thus were comparable to the data from the8.2 ka event. Thus Type E surfaces seemingly represent periods thatwere both wetter and cooler, consonant with the presently greatermeteoric precipitation during cooler seasons in the region aroundCova da Arcoia.

Between many Type E and/or Type L surfaces are intervals ofcomparatively moderate to small values of δ13C. These presumablyrepresent periods of sufficient meteoric precipitation to give vegeta-tion that contributed more 12C-rich CO2 to soil gas above the cave andto give waters dripping on the stalagmite sufficiently frequently todeposit CaCO3 across the entire stalagmite (Fig. 8A). These periods ofrelatively normal stalagmite deposition are perhaps best exemplifiedby the small δ13C values of Stable Isotope Samples 31, 52, and 85,which are neither just above nor just below Type E or Type L surfaces.These samples come from intervals that have the bright fluorescence(Fig. 2), further suggesting greater development of vegetation than atother times.

The presence of both Type L surfaces (representing periods of solittle precipitation that stalagmite deposition dwindled to nearly zero)and Type E surfaces (representing periods of so much flow thatprevious spelean CaCO3 was dissolved by undersaturated water)suggests broad shifts in climate in the Holocene. Furthermore, thebrief time between onset of deposition after dissolution above Type Esurfaces (the end of a cool/wet period) and precipitation of aragonitewith large values of δ13C suggests that the change from exceptionallywet conditions to exceptionally dry conditions was ecologicallyabrupt. For example, the interval from Surface 2 to Isotope Sample23 represents at most about 200 years, and the same is true of theinterval from Surface 10 to Isotope Sample 75. Thus our results fromStalagmite ESP03 indicate not only that there was broad change froma drier and likely warmer early Holocene to a wetter and likely coolerlater Holocene in northwestern Spain, but also that across shortertime scales there were decadal-to-century-scale shifts from periods sodry that vegetation dwindled to periods so cool and wet that watergushed through the soil and limestone too quickly to reach saturationwith respect to CaCO3 (Fig. 8).

8. Discussion, part II: comparison with other records

8.1. Comparison with other Iberian records: an Atlantic vs.Mediterranean contrast

The ESP03 record suggests that the longest single period ofwarmer and drier conditions was during PI2 and thus from 8.0 to6.5 ka. This period coincides with the greatest values of δ18O in theNorthGRIP ice core (Johnsen et al., 2001) (Fig. 7B). More broadly, PI1,PI2, PI3, and PI4, which include the greatest values of δ13C inStalagmite ESP03 and most of its aragonitic intervals, coincide withthe broader early Holocene peak in the NorthGRIP core andspecifically with the intervals in which the 1 ka low-pass filteredversion of the NorthGRIP data exceed −34.8‰ relative to SMOW(Fig. 7A). Thus the warmer and drier periods inferred from StalagmiteESP03 coincide with warmer periods in the NorthGRIP core andgenerally with the Holocene Climatic Optimum. In contrast, coolerand wetter periods inferred from the ESP03 core broadly correlatewith lesser values in the 1 ka low-pass filtered version of theNorthGRIP data. The change from drier and perhaps warmerconditions to wetter and perhaps cooler conditions at roughly 5 ka


corresponds with themajor change in the ice-rafting record of Bond etal. (2001) (Fig. 7A).

The positive correlation of the ESP03 record with the NorthGRIPice-core data is paralleled by a positive correlation with pollen datafrom southern to eastern France. Guiot et al. (1993) compiled changein the palaeobioclimatic component of three pollen sequences andfound consistent evidence of warming from the Last Glacial Maximumto a warmest period at 8 to 5 ka and then, relative to that warm early-to-middle Holocene, a cooler later Holocene from 5 ka to present. Theperiod from 8 to 5 ka in their compilation would include the greatestvalues of δ13C in Stalagmite ESP03 (those below Surfaces 3, 4, and 7)and thus the latter's warmest and/or driest episodes.

Episodes in the record from Stalagmite ESP03 also coincide withthe results of studies of stalagmites from Kaite Cave on northern coastof Spain reported by Domínguez-Villar et al. (2008, 2009). Forexample, the erosional episode inferred from Surface 15 in ESP03developed during the period at 1.6 to 2.4 ka for which Domínguez-Villar et al. (2008) inferred more rainfall at Kaite Cave. Likewise, theerosional episode inferred from Surface 2 in ESP03 correlates with thetwo intervals of low δ18OCaCO3 interpreted by Domínguez-Villar et al.(2009) to represent enhanced rainfall at Kaite Cave (Fig. 7). It alsocoincides with the cold interval interpreted by Leira and Santos(2002) from diatom and pollen data in a lake in the Courel Sierra andthus near ESP03.

These positive correlations with Holocene records to the north andwest extend to comparisonwithmore generalized records. Davis et al.(2003) used pollen records from across Europe to reconstructtemperature variation through the Holocene and arrived at sixgeneralized curves for eastern and western halves of northern,central, and southern zones of Europe. The ESP03 record is not aperfect match with any of these six, but it is much more similar to thecurves for central and northern Europe than to those for southwesternor southeastern Europe.

A more loosely positive correlation of the ESP03 record of wetterand drier periods exists with the record of floods in west-centralSpain. Benito et al. (2003) used deposits along the Tagus River inwest-central Spain to estimate the timing and extent of floods of theTagus. The most general summary of their results is that periods ofepisodic flooding early in the Holocene (from 11 to 7.5 ka) and late inthe Holocene (from 1.2 ka to the present) bracketed a middleHolocene period without floods. They attributed the periods offlooding to enhanced winter rainfall. The dry period in the Tagusflooding record begins and ends later than the seemingly driest part ofthe ESP03 record. However, the broad pattern of wetter conditions inthe earliest Holocene (prior to first deposition of ESP03 and during the8.2 ka event) and in later Holocene (during development of Surfaces15 to 18 in ESP03) that bracket a drier period inferred from ESP03 isanalogous to the Tagus record.

In contrast with the positive correlations with records to the northand west discussed above, comparison of the ESP03 record with datain eastern and southern Spain yields few similarities. For example,Davis and Stevenson (2007) concluded that the 8.2 ka event was, inthe Ebro Valley of northeastern Spain, a time of more extensive firesand generally drier climate, opposite the inferred wetter periodinferred from the ESP03 record. Also in northeastern Spain, Morellónet al. (2009) inferred high lake levels in the early to middle Holoceneand two arid phases in the later Holocene, opposite our inferencesfrom the ESP03 record. Similarly, but in southeastern Spain, Carrion(2002) studied lake levels and, from those results and from work byothers in southeastern Spain and northern Africa, inferred a mid-Holocene wet period from 7500 to 5200 BP, in contrast to drierconditions earlier and later in the Holocene — likewise opposite theinference from ESP03 above that period from 8.0 to 6.5 ka was asustained moderate to dry period and the period from 8.0 to 5.2 wasgenerally drier with two lesser wet periods represented by Surfaces 5and 6. Pons and Reille (1988) likewise inferred a “hot and humid




Holocene optimum” from pollen from a peat bog in southern Spain,opposite the inferences above from ESP03. More broadly, Davis andStevenson (2007) generalized that the early to middle Holocene insouthern Europe (and more specifically the Mediterranean region)was, in contrast to northern Europe, a time of cooler and wetterclimate than that of today, consistent with generalized curves forsouthwestern and southeastern Europe generated by Davis et al.(2003).

In summary, the ESP03 Holocene record from the north-facingslopes of the Serra do Courel and, more broadly, from the north-facingslopes of the Cordillera Cantabrica, shows considerable similarity torecords to the north andwest, from France to Greenland. In contrast, itshows much less similarity to records from eastern and southernSpain and, more generally, from the Mediterranean. This result from acave in southern Europe may illustrate that, in the effort to generalizeabout the Holocene history of Europe, one may be wiser to speak ofAtlantic and Mediterranean regions (as did McDermott et al., 1999)than of “Northern Europe” and “Southern Europe”. Furthermore, theboundary between the Atlantic and Mediterranean in Spain wassomewhere (but perhaps not far) to the south of the crest of the Serrado Courel.

8.2. Correlations with climatic and cultural events at the global scale

Four of the most prominent Type E surfaces in Stalagmite ESP03coincide in time with droughts in other regions and/or with crises inOld-World human civilizations to the east. The earliest example isSurface 2, which is a pronounced Type E surface representing one ofthe longest hiatuses, which implies dissolution of exceptionalintensity or duration. Surface 2 developed during the 8.2 ka Event,which Alley et al. (1997) showed was correlative with dry climate inAsia, Africa, and North America. The great age of Surface 2 and the8.2 ka Event preclude their association with any known crisis inhuman civilization.

Surface 5 represents one of the longest hiatuses in the ESP03record (Fig. 6) and developed at roughly 6 ka. It thus coincides withthe “5.9 ka Event” in which the Ubaid Period of Middle Easterncivilization (Carter and Philip, 2010) abruptly ended as climatebecame drier in the Arabian Peninsula (Parker et al., 2006).

Surface 10, perhaps the most visually evident Type E surface(Fig. 5C), developed about 3.9 ka and thus coincides with the 4.2 kaEvent. During the 4.2 ka Event, the Egyptian Old Kingdom collapsed inresponse to drier climate (Stanley et al., 2002), the Akkadian Empirein Mesopotamia collapsed in a period of aridification (Weiss et al.,1993; Cullen et al., 2000), the Harrapan or Indus Valley Civilizationcame to an end in a time of drought (Staubwasser et al., 2003), and theLongshan Culture in the central and lower Yellow River Valley inChina declined during a period of colder and drier climate (Zhanget al., 2010). Evidence of drought at roughly 4.2 ka is also known fromdust in ice cores from Kilimanjaro (Thompson et al., 2002),geochemical evidence from a flowstone in an Italian cave (Drysdaleet al., 2006), and from a variety of evidence of drier climate in mid-continent North America (Booth et al., 2005).

Surface 15, a visually striking surface of dissolutional erosion(Fig. 5A and B) and the surface representing the hiatus of greatestduration (Fig. 6), formed about 1.7 ka and thus at the end of theRoman Warm Period (McDermott et al., 2001). Like Surfaces 2, 5, and10, Surface 15 corresponds with a local minimum in δ18O ofNorthGRIP glacial ice, and it further corresponds to one of thatrecord's transitions from greater to lesser values of δ18O. Developmentof Surface 15 occurred as the Roman Empire declined. Given thedependence of the city of Rome on supplies of grain from northernAfrica (Rickman, 1980) and evidence of drying of the North Africanclimate since Roman times (Grove, 1972) to its present dry condition(e.g., Knapp, 1970), one might speculate that yet another drying of


western Asian and Mediterranean climate at roughly 1.7 ka mighthave contributed to yet another noted Old-World cultural decline.

9. Conclusions

Stalagmite ESP03 from Cova da Arcoia in the Serra do Courelpreserves a record of Holocene climate suggesting a drier and perhapswarmer early to mid Holocene and a wetter and perhaps cooler laterHolocene (Fig. 7) in that area. This general pattern is similar to thatobserved in Holocene records from Greenland and northern Europe,and it contrasts sharply with records from eastern and southern Spainthat seemingly represent a more Mediterranean, rather than Atlantic,climatic regime.

Within this Holocene record are distinct intervals of shorter-term(decades to a few centuries) periods of distinctly dry climate in whichstalagmite deposition nearly stopped and, in contrast, intervals ofwetter climate in which enhanced flow over the stalagmite causeddissolution of previously precipitated CaCO3 (Fig. 8). These strikinglydrier andwetter shifts punctuating otherwisemore normal stalagmitedeposition suggest a Holocene in which climate change was notmonotonic but instead featured changes in climate that were abruptwhen compared to ecologic time scales. Several of the wetter and/orcooler events can be correlated with local minima in the NorthGRIPoxygen isotope record from Greenland glacial ice (Fig. 7), and theycoincide with periods of widespread drought and/or cultural collapsein the Mediterranean and Middle East.

Finally, with regard to palaeoclimate studies using stalagmites, thesignificance of interpreting dissolution features, bounding surfaces,and detrital material to understanding Stalagmite ESP03 illustratesthe potential significance of stalagmite petrography to palaeoclimaticresearch. In this case, the synthesis of petrography, mineralogy, and Cand O isotope geochemistry generated a far richer understanding ofclimate change than would have been possible using only O isotopedata.

Acknowledgments

This is a contribution to the Research Project BTE-CGL-2006-08996of the Spanish Ministry of Science and Technology. An early draft ofthe manuscript was improved by the comments of Dr. GuglielmoAngelo Caddeo, and the submitted manuscript was improved by thecomments of two anonymous Palaeo-3 reviewers and Palaeo-3Editor-in-Chief Dr. Thierry Corrège.

References

Alley, R.B., Mayewski, P.A., Sowers, T., Stuiver, M., Taylor, K.C., Clark, P.U., 1997.Holocene climatic instability: a prominent, widespread event 8200 yr ago. Geology25, 483–486.

Baker, A., Genty, D., Smart, P.L., 1998. High-resolution records of soil humification andpaleoclimatic change from variations in speleothem luminescence excitation andemission wavelengths. Geology 26, 903–906.

Baldini, J.U.L., McDermott, F., Baker, A., Baldini, L.M., Mattey, D.P., Railsback, L.B., 2005.Biomass effects on stalagmite growth and isotope ratios: a 20th century analoguefrom Wiltshire, England. Earth and Planetary Science Letters 240, 486–494.

Benito, G., Sopeña, A., Sánchez-Moya, Y., Machado, M.J., Pérez-González, A., 2003.Palaeoflood record of the Tagus River (Central Spain) during the Late Pleistoceneand Holocene. Quaternary Science Reviews 22, 1737–1756.

Berger, W.H., Labeyrie, L.D., 1987. Abrupt Climatic Change. D. Reidel PublishingCompany, Dordrecht. 425 pp.

Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M.N., Showers, W., Hoffmann, S.,Lotti-Bond, R., Hajdas, I., Bonani, G., 2001. Persistent solar influence on NorthAtlantic climate during the Holocene. Science 294, 2130–2136.

Booth, R.K., Jackson, S.T., Forman, S.L., Kutzbach, J.E., Bettis III, E.A., Kreigs, J., Wright, D.K.,2005. A severe centennial-scale drought inmidcontinental North America 4200 yearsago and apparent global linkages. Holocene 15, 321–388.

Burton, E.A., Walter, L.M., 1987. Relative precipitation rates of aragonite and Mg calcitefrom seawater: temperature or carbonate ion control? Geology 15, 111–114.

Cabrol, P., Coudray, J., 1982. Climatic fluctuations influence the genesis and diagenesisof carbonate speleothems in southwestern France. National Speleological SocietyBulletin 44, 112–117.




Carrion, J.S., 2002. Patterns and processes of Late Quaternary environmental change in amontane region of southwestern Europe. Quaternary Science Reviews 21,2047–2066.

Carter, R.A., Philip, G. (Eds.), 2010. Beyond the Ubaid: Transformation and Integration inthe Late Prehistoric Societies of the Middle East (Studies in Ancient OrientalCivilization 63). The Oriental Institute of the University of Chicago, Chicago. 396 pp.

Cheng, H., Edwards, R.L., Hoff, J., Gallup, C.D., Richards, D.A., Asmerom, Y., 2000. Thehalf-lives of uranium-234 and thorium-230. Chemical Geology 169, 17–33.

Cheng, H., Fleitmann, D., Edwards, R.L., Wang, X.F., Cruz, F.W., Auler, A.S., Mangini, A.,Wang, Y.J., Burns, S.J., Matter, A., 2009a. Timing and structure of the 8.2 ky eventinferred from δ18O records of stalagmites from China, Oman and Brazil. Geology 37,1007–1010.

Cheng, H., Edwards, R.L., Broecker, W.S., Denton, G.H., Kong, X.G., Wang, Y.J., Zhang, R.,Wang, X.F., 2009b. Ice Age terminations. Science 326, 248–252.

Cullen, H.M., deMenocal, P.B., Hemming, S., Hemming, G., Brown, F.H., Guilderson, T.,Sirocko, F., 2000. Climate change and the collapse of the Akkadian empire: evidencefrom the deep sea. Geology 28, 379–382.

Dansgaard, W., 1987. Ice core evidence of abrupt climatic changes. In: Berger, W.J.,Labeyrie, L.D. (Eds.), Abrupt Climatic Change: Evidence and Implications. Reidel,Dordrecht, The Netherlands, pp. 223–233.

Davis, B.A.S., Stevenson, A.C., 2007. The 8.2 ka event and Early–Mid Holocene forests,fires and flooding in the Central Ebro Desert, NE Spain. Quaternary Science Reviews26, 1695–1712.

Davis, B.A.S., Brewer, S., Stevenson, A.C., Guiot, J., Contributors, Data, 2003. Thetemperature of Europe during the Holocene reconstructed from pollen data.Quaternary Science Reviews 22, 1701–1716.

Domínguez-Villar, D., Wang, X., Cheng, H., Martín-Chivelet, J., Edwards, R.L., 2008. Ahigh-resolution late Holocene speleothem record from Kaite Cave, northern Spain:δ18O variability and possible causes. Quaternary International 187, 40–51.

Domínguez-Villar, D., Fairchild, I.J., Baker, A., Wang, X., Edwards, R.L., Cheng, H., 2009.Oxygen isotope precipitation anomaly in the North Atlantic region during the8.2 ka event. Geology 37, 1095–1098.

Dreybrodt,W., 1999. Chemical kinetics, speleothemgrowth and climate. Boreas 28, 347–356.Drysdale, R., Zanchetta, G., Hellstrom, J., Maas, R., Fallick, A., Pickett, M., Cartwright, I.,

Piccini, L., 2006. Late Holocene drought responsible for the collapse of Old Worldcivilizations is recorded in an Italian cave flowstone. Geology 34, 101–104.

Edwards, R.L., Chen, J.H., Wasserburg, G.J., 1987. 238U–234U–230–232Th systematics andthe precise measurement of time over the past 500000 years. Earth and PlanetaryScience Letters 81, 175–192.

Fairchild, I.J., Treble, P.C., 2009. Trace elements in speleothems as recorders ofenvironmental change. Quaternary Science Reviews 28, 449–468.

Fisher, D.A., Koerner, R.M., 2003. Holocene ice-core climate history — a multi-variableapproach. In: Mackay, A., Battarbee, R., Birks, J., Oldfield, F. (Eds.), Global Change inthe Holocene. Arnold, London, pp. 281–293.

Frisia, S., Borsato, A., Fairchild, I.J., McDermott, F., Selmo, E.M., 2002. Aragonite–calciterelationships in speleothems (Grotte de Clamouse, France): environment, fabrics,and carbonate geochemistry. Journal of Sedimentary Research 72, 687–699.

García-Vázquez, A., Grandal-d'Anglade, A., Vaqueiro Rodríguez, M., Vidal Romaní, J.R., inpress. On the relation of cave and brown bears in O Courel Sierra (Galicia, NWSpain). Quaternaire.

Gascoyne, M., 1992. Palaeoclimate determination from cave calcite deposits. QuaternaryScience Reviews 11, 609–632.

González, L.A., Lohmann, K.C., 1988. Controls onmineralogy and composition of speleancarbonates: Carlsbad Caverns, New Mexico. In: James, N.P., Choquette, P.W. (Eds.),Paleokarst. Springer-Verlag, New York, pp. 81–101.

Grandal-d'Anglade, A., Pérez Rama, M., García Vázquez, A., Vidal Gorosquieta, A.,González Fortes, G., 2010. In: López, Díaz, A.J., Ramil, Rego, E. (Eds.), Geocaracter-ización de yacimientos arqueológicos: 765 conservación de restos paleontológicosy obtención de datos biogeoquímicos, vol. 4. Monografías doMuseo de Prehistoria eArqueoloxía de Vilalba, Vilalba, pp. 57–70.

Grandal-d'Anglade, A., López González, F., Vidal Romaní, J.R., 1997. Condicionantes en ladistribución de macromamíferos en Galicia (NW P. I.) durante el Cuaternariosuperior. Cadernos do Laboratorio Xeolóxico de Laxe, 23, pp. 43–66.

Grove, A.T., 1972. Climatic change in Africa in the last 20,000 years. Les Problèmes deDéveloppement du Sahara Septentrional, vol. 2. Colloque de Ouarghla, Algiers.

Guiot, J., de Beaulieu, J.L., Cheddadi, R., David, F., Ponel, P., Reille, M., 1993. The climate inWestern Europe during the last glacial–interglacial cycle derived from pollen andinsect remains. Palaeogeography, Palaeoclimatology, Palaeoecology 103, 73–93.

Harmon, R.S., Thompson, P., Schwarcz, H.P., Ford, D.C., 1975. Uranium-series dating ofspeleothems. National Speleological Society Bulletin 37, 21–33.

Hendy, C.H., 1971. The isotopic geochemistry of speleothems—I: the calculations of theeffects of different modes of formation on the isotopic composition of speleothemsand their applicability as paleoclimate indicators. Geochimica et CosmochimicaActa 35, 801–824.

Hesterberg, R., Siegenthaler, U., 1991. Production and stable isotope composition of CO2

in a soil near Bern, Switzerland. Tellus 43B, 197–205.IAEA, 2006. Isotope Hydrology Information System. The ISOHIS DatabaseAccessible at

http://www.iaea.org/water.Izco Sevillano, J., Amigo Vázquez, J., Guitián Rivera, J., 1982. El Caurel. Análisis

sincorológico de una comarca en la frontera Mediterráneo-Eurosiberiana. IIJornadas de Fitosociología, Santiago de Compostela, Spain.

Johnsen, S.J., Dahl-Jensen, D., Gundestrup, N., Steffensen, J.P., Clausen, H.B., Miller, H.,Masson-Delmotte, V., Sveinbjörnsdottir, A.E., White, J., 2001. Oxygen isotope andpalaeotemperature records from six Greenland ice-core stations: Camp Century,Dye3, GRIP, GISP2, Renland and NorthGRIP. Journal of Quaternary Science 16,299–307.


Kim, S.T., O'Neil, J.R., 1997. Equilibrium and non-equilibrium oxygen isotope effects insynthetic carbonates. Geochimica et Cosmochimica Acta 61, 3461–3475.

Kim, S.T., O'Neil, J.R., Hillaire-Marcel, C., Mucci, A., 2007. Oxygen isotope fractionationbetween synthetic aragonite and water: influence of temperature and Mg2+

concentration. Geochimica et Cosmochimica Acta 71, 4704–4715.Knapp, W., 1970. Tunisia. Thames and Hudson, London. 224 pp.Lauritzen, S.-E., 2003. Reconstructing Holocene climate records from speleothems. In:

Mackay, A., Battarbee, R., Birks, J., Oldfield, F. (Eds.), Global Change in the Holocene,London. Arnold, London, pp. 242–263.

Leira, M., Santos, L., 2002. An early Holocene short climatic event in the northwestIberian Peninsula inferred from pollen and diatoms. Quaternary International93–94, 3–12.

Mayewski, P.A., Rohling, E.E., Stager, J.C., Karlén, W., Maasch, K.A., Meeker, L.D.,Meyerson, E.A., Gasse, F., van Kreveld, S., Holmgren, K., Lee-Thorp, J., Rosqvist, G.,Rack, F., Staubwasser, M., Schneider, R.R., Steig, E.J., 2004. Holocene climatevariability. Quaternary Research 62, 243–255.

McCrea, J.M., 1950. On the isotopic chemistry of carbonates and a paleotemperaturescale. The Journal of Chemical Physics 18, 849–857.

McDermott, F., 2004. Palaeo-climate reconstruction from stable isotope variations inspeleothems: a review. Quaternary Science Reviews 23, 901–918.

McDermott, F., Frisia, S., Huang, Y., Longinelli, A., Spiro, B., Heaton, T.H.E., Hawkesworth,C.J., Borsato, A., Keppens, E., Fairchild, I.J., van der Borg, K., Verheyden, S., Selmo, E.,1999. Holocene climate variability in Europe: evidence from δ18O, textural andextension-rate variations in three speleothems. Quaternary Science Reviews 18,1021–1038.

McDermott, F., Mattey, D.P., Hawkesworth, C., 2001. Centennial-scale Holocene climatevariability revealed by a high-resolution speleothem δ18O record from SW Ireland.Science 294, 1328–1331.

McGarry, S.F., Caseldine, C., 2004. Speleothem palynology: an undervalued tool inQuaternary studies. Quaternary Science Reviews 23, 2389–2404.

Morellón, M., Valero-Garcés, B., Vegas-Vilarrúbia, T., González-Sampériz, P., Romero, O.,Delgado-Huertas, A., Mata, P., Moreno, A., Rico, M., Corella, J.P., 2009. Lateglacial andHolocene palaeohydrology in the western Mediterranean region: the Lake Estanyarecord (NE Spain). Quaternary Science Reviews 28, 2582–2599.

Morse, J.W., Wang, Q., Tsio, M.Y., 1997. Influences of temperature and Mg:Ca ratio onCaCO3 precipitates from seawater. Geology 25, 85–87.

Murray, J.W., 1954. The deposition of calcite and aragonite in caves. Journal of Geology62, 481–492.

National Research Council, 2002. Abrupt Climate Change. National Academy Press,Washington, D.C.

Oldfield, F., 2003. Introduction: the Holocene, a special time. In: Mackay, A., Battarbee, R.,Birks, J., Oldfield, F. (Eds.), Global Change in the Holocene. Arnold, London,pp. 1–9.

Parker, A.G., Goudie, A.S., Stokes, S., White, K., Hodson, M.J., Manning, M., Kennet, D.,2006. A record of Holocene climate change from lake geochemical analyses insoutheastern Arabia. Quaternary Research 66, 465–476.

Pobeguin, Th., 1965. Sur les concretions calcaires Observées dans la Grotte de Moulis(Ariège). Société Géologique de la France, Compte Rendu 241, 1791–1793.

Pons, A., Reille, M., 1988. The Holocene and upper Pleistocene pollen record from Padul(Granada, Spain). A new study. Palaeogeography, Palaeoclimatology, Palaeoecology66, 243–263.

Pross, J., Kotthoff, U., Müller, U.C., Peyron, O., Dormoy, I., Schmiedl, G., Kalaitzidis, S.,Smith, A.M., 2009. Massive perturbation in terrestrial ecosystems of the EasternMediterranean region associated with the 8.2 kyr B.P. climatic event. Geology 37,887–890.

Railsback, L.B., 2000. An Atlas of Speleothem Microfabrics. http://www.gly.uga.edu/speleoatlas/SAIndex1.html.

Railsback, L.B., Brook, G.A., Chen, J., Kalin, R., Fleischer, C.J., 1994. Environmentalcontrols on the petrology of a Late Holocene speleothem from Botswana withannual layers of aragonite and calcite. Journal of Sedimentary Research A64,147–155.

Railsback, L.B., Sheen, S.-W., Rafter, M.A., Brook, G.A., Kelloes, C., 1997. Diageneticreplacement of aragonite by aragonite in speleothems: criteria for its recognititonfrom Botswana and Madagascar. Speleochronos 8, 3–11.

Railsback, L. Bruce, Brook, G.A., Webster, J.W., 1999. Petrology and paleoenvironmentalsignificance of detrital sand and silt in a stalagmite from Drotsky's Cave, Botswana.Physical Geography 20, 331–347.

Railsback, L.B., Dabous, A.A., Osmond, J.K., Fleisher, C.J., 2002. Petrographic andgeochemical screening of speleothems for U-series dating: an example fromrecrystallized speleothems from Wadi Sannur Cavern, Egypt. Journal of Cave andKarst Studies 64, 108–116.

Rickman, G., 1980. The Corn Supply of Ancient Rome. Clarendon Press, Oxford.Roberts, N., 1998. The Holocene: An Environmental History, 2nd ed. Blackwell, Malden,

MA.Santos, L., Vidal Romani, J.R., Jalut, G., 2000. History of vegetation during the Holocene

in the Courel and Queixa Sierras, Galicia, northwest Iberian Peninsula. Journal ofQuaternary Science 15, 621–632.

Shopov, Y.Y., Ford, D.C., Schwarcz, H.P., 1994. Luminescence microbanding inspeleothems: high resolution chronology and paleoclimate. Geology 22, 407–410.

Siegel, F.R., 1965. Aspects of calcium carbonate deposition in Great Onyx Cave,Kentucky. Sedimentology 4, 285–299.

Siegel, F.R., Dort Jr., W., 1966. Calcite–aragonite speleothems from a hand-dug cave innortheast Kansas. International Journal of Speleology 2, 165–169.

Stanley, J.D., Krom,M.D., Cliff, R.A.,Woodward, J.C., 2002. Nile flow failure at the end of theOld Kingdom, Egypt: strontium isotopic and petrologic evidence. Geoarchaeology 18,395–402.


http://www.iaea.org/water

http://www.gly.uga.edu/speleoatlas/SAIndex1.html

http://www.gly.uga.edu/speleoatlas/SAIndex1.html



Staubwasser, M., Sirocko, F., Grootes, P.M., Segl, M., 2003. Climate change at the 4.2 kaBP termination of the Indus valley civilization and Holocene south Asian monsoonvariability. Geophysical Research Letters 30, 1425.

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.

Thrailkill, J., 1971. Carbonate deposition in Carlsbad Caverns. Journal of Geology 79,683–695.

Turgeon, S., Lundberg, J., 2001. Chronology of discontinuities and petrology ofspeleothems as paleoclimatic indicators of the Klamath Mountains, southwestOregon, USA. Carbonates and Evaporites 16, 153–167.

van Beynen, P., Bourbonniere, R., Ford, D., Schwarcz, H., 2001. Causes of colour andfluorescence in speleothems. Chemical Geology 175, 319–341.

Verheyden, S., 2004. Trace elements in speleothems: a short review of the state of theart. International Journal of Speleology 33, 95–101.


Vidal Romaní, J.R., Sanjurjo Sánchez, J., Grandal-d'Anglade, A., Vaqueiro Rodríguez, M.,Fernández Mosquera, D., 2010. Geocaracterización de yacimientos arqueológicosen medios sedimentarios: cronología absoluta y relativa. In: López Díaz, A.J., RamilRego, E. (Eds.), Arqueoloxía: Ciencia e Restauración, vol. 4. Monografías do Museode Prehistoria e Arqueoloxía de Vilalba, Vilalba, pp. 7–19.

Walter, H., Lieth, H., 1960. Klimadiagramm-weltatlas. Gustav Fischer Verlag, Jena. 2 volumes.Webster, J.W., Brook, G.A., Railsback, L.B., Cheng, H., Edwards, R.L., Alexander, C., Reeder,

P.R., 2007. Stalagmite evidence from Belize indicating significant droughts at thetime of Preclassic Abandonment, the Maya Hiatus, and the Classic Maya Collapse.Palaeogeography, Palaeoclimatology, Palaeoecology 250, 1–17.

Weiss, H., Courty, M.-A., Wetterstrom, W., Guichard, F., Senior, L., Meadow, R., Curnow,A., 1993. The genesis and collapse of Third Millennium North Mesopotamiancivilization. Science 261, 995–1004.

Zhang, G., Zhu, C., Wang, J., Zhu, G., Ma, C., Zheng, C., Zhao, L., Li, Z., Li, L., Jin, A., 2010.Environmental archaeology on Longshan Culture (4500–4000 aBP) at YuhuicunSite in Bengbu, Anhui Province. Journal of Geographical Sciences 20, 455–468.



Petrographic and isotopic evidence for Holocene long-term climate change and shorter-term...

Documents

Transcript of Petrographic and isotopic evidence for Holocene long-term climate change and shorter-term...