Vegetation and climate changes during the last 22,000yr from a marine core near Taitao Peninsula, southern Chile

Palaeogeography, Palaeoclimatology, Palaeoecology 369 (2013) 335–348

Contents lists available at SciVerse ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology

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

Vegetation and climate changes during the last 22,000 yr from a marine core nearTaitao Peninsula, southern Chile

Vincent Montade a,⁎, Nathalie Combourieu Nebout b, Catherine Kissel b, Simon G. Haberle c,Giuseppe Siani d, Elisabeth Michel b

a Laboratoire des Sciences du Climat et de l'Environnement (LSCE), UMR 8212 CNRS/UVSQ/CEA, Orme des Merisiers, Bâtiment 701, 91191 Gif-sur-Yvette Cedex, Franceb Laboratoire des Sciences du Climat et de l'Environnement (LSCE), UMR 8212 CNRS/UVSQ/CEA, Avenue de la Terrasse, 91198 Gif-sur-Yvette Cedex, Francec Department of Archaeology & Natural History, College of Asia and the Pacific, Australian National University, Canberra, Australian Capital Territory 0200, Australiad Laboratoire des Interactions et Dynamique des Environnements de Surface (IDES), UMR 8148, CNRS, Bâtiment 504, 91405 Orsay Cedex, France

⁎ Corresponding author.E-mail addresses: [email protected] (V. M

[email protected] (N. Combourieu Nebout), ca(C. Kissel), [email protected] (S.G. Haberle), gi(G. Siani), [email protected] (E. Michel).

0031-0182/$ – see front matter © 2012 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.palaeo.2012.11.001

a b s t r a c t
a r t i c l e i n f o
Article history:Received 30 November 2011Received in revised form 31 October 2012Accepted 4 November 2012Available online 9 November 2012

Keywords:Southern hemispherePatagoniaMarine palynologyVegetationPalaeoenvironmentLast deglaciationHoloceneAntarctic Cold ReversalSouthern Westerly Wind belt

High-temporal resolution of pollen analyses from marine core MD07-3088 (46°S) documents regional andcoastal vegetation changes in the mid-latitude of southern Chile during the last 22 kyr BP. The coastal marginwas partly ice-free during the last glacial period with the presence of scattered vegetation. After~17.6 kyr BP,the expansion of Nothofaguswoodland highlights a warming trend and marks the beginning of the last degla-ciation. The deglacial forest expansion is interrupted by the development of Magellanic moorland simulta-neously with the Antarctic Cold Reversal. This vegetation change illustrates a pause in the warming trendwith an abrupt increase in precipitation which is related to strong intensity of Southern Westerly Winds.After 12.8 kyr BP, the retreat of Magellanic moorland illustrates a second-step warming that is strengthenedafter 11.5 kyr at the onset of the Holocene. Expansion of heliophytic taxa illustrates warmest and driest con-ditions of the Holocene from 11 to 7.4 kyr BP. Later, the decrease of heliophytic taxa shows a return to coolerand wetter conditions, illustrated by the presence of the North Patagonian rainforest that reaches moderncondition after 5 kyr. Climate changes inferred from the marine pollen record are consistent with geochem-ical data changes from the same samples and illustrate a clear synchronicity with δD variations of EPICADome C ice core during the last deglaciation and Holocene. Comparison between vegetation changes fromthe marine core and regional palaeoclimatic records highlights southward-northward shifts of the SouthernWesterly Wind belt and supports a close link between Southern Westerly Wind belt and atmospheric CO2

variability during the last deglaciation.© 2012 Elsevier B.V. All rights reserved.

1. Introduction

Due to the Andean Cordillera, southern South America is a majortopographic constraint on atmospheric and oceanographic circulation.It represents the only continental landmass intercepting the entireSouthernWesterlyWind belt (SWW) in the southern hemisphere. Cli-mate simulations have suggested that the latitudinal position of theSWW belt has changed during the past and that these changes couldhave played a key role in CO2 degassing throughwesterlies-driven up-welling in the Southern Ocean. During the last deglaciation the shift ofSWWbelt over the Southern Oceanmay have strengthened upwellingand thus allowed more CO2 degassing to the atmosphere. Prolonged

ontade),[email protected]@u-psud.fr

rights reserved.

westerly effects over the Southern Ocean would have contributed toraising atmospheric CO2 above the threshold necessary to sustain in-terglacial conditions (Toggweiler et al., 2006). In southern Chile, thepresent-day SWWextends northward up to 30°S in the austral winterand up to 45°S in the austral summer and generates a high gradient ofrainfall which increases from north to south along the Chilean coastwith maxima over 4000 mm.year−1 south of 48°S (Garreaud et al.,2009). The southern Chilean coast is therefore a critical region tostudy past climate and precipitation changes in order to understandthe role of the SWW, their linkages with CO2 degassing from SouthernOcean, and their role on global climate changes (e.g. Lamy et al., 2010;Moreno et al., 2010). Palaeoclimatic data showed that the SWW beltwas north of 41°S during the Last Glacial Maximum, then shiftedsouthward during the last deglaciation (e.g. Villagrán, 1988a; Lamyet al., 1999; Moreno et al., 1999; McCulloch et al., 2000). At 41°S insouthern Chile, vegetation changes during the last deglaciation fromthe oceanic core ODP 1233 illustrate a warming trend interrupted bya brief climate reversal contemporaneous with the Antarctic Cold

http://dx.doi.org/10.1016/j.palaeo.2012.11.001

mailto:[email protected]







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

336 V. Montade et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 369 (2013) 335–348

Reversal (ACR) (Heusser et al., 2006b). This two-step warming trendappears comparable to Antarctic ice cores trends (e.g. Jouzel et al.,1995). At the same latitude, the continental pollen records show aslightly different timing with a deglaciation interrupted by a cooland humid phase between 13.8 and 11.5 kyr, thus covering a part ofthe ACR period and the following northern hemisphere YoungerDryas reversal period (Moreno, 2004). Further south, lake sedimentsfrom Taitao Peninsula (Lumley and Switsur, 1993; Bennett et al.,2000) and Chonos Archipelago (Bennett et al., 2000; Haberle andBennett, 2004) indicate an increase in precipitation generally contem-poraneous with the ACR and no reversal trend in temperature duringthe last deglaciation. Although climate and precipitation changesprobably related to SWW belt changes occurred during the last degla-ciation, there is no clear evidence for the timing and the amplitude ofthe westerlies latitudinal displacements during the last deglaciation.Was there a progressive southward shift of the westerlies during thelast deglaciation (McCulloch et al., 2000) or was this southward shiftinterrupted by northward shifts (Kaiser et al., 2008) during abruptevents such as the ACR? Answering this question will help to under-stand the role of the SWWduring the last deglaciation and thus on cli-mate at more global scale.

Marine cores can be used to reconstruct regional vegetation changesand, most significantly, they provide the opportunity to compare andcorrelate marine and continental indicators of palaeoclimatic changes(e.g. Heusser and Balsam, 1977; Combourieu Nebout et al., 2002;Sánchez Goñi et al., 2002; Dupont and Wyputta, 2003; Heusser et al.,2006a,b; Hooghiemstra et al., 2006; Desprat et al., 2007). In this context,we present a high-temporal resolution pollen record from marine coreMD07-3088 located in southern Chile (46°S). Although located in theSWW belt, this marine core is located north of the main core of theSWW belt where precipitation is highest and "Magellanic moorland"vegetation has developed. As shown by previous studies, Magellanicmoorland is highly sensitive to SWW belt changes and its expansionwhich indicates wet conditions related to the increasing intensity ofSWW (Villagrán, 1988a; Moreno et al., 1999). We expect therefore toillustrate regional vegetation changes including coastal vegetation andtheir links to climate and SWW belt changes during the key-period ofthe last deglaciation.

2. Environmental setting

Themarine coreMD07-3088was collected in southern Chile, near tothe coast of Taitao Peninsula extending from latitude~44° to 46°S andlongitude~73° to 75°W (Fig. 1A). At these latitudes, the coastal rangereaches atmost 1000 m above sea level (asl) and intense glacial erosionhas shaped complex systems of numerous scattered islands with deepfjords during the last glaciation (Clapperton and Sugden, 1988). East-ward, the Andean Cordillera rarely exceeds 3000 m asl with numerousglaciers including twomajor ice caps: the “Hielo Patagonico Norte” andthe “Hielo Patagonico Sur” (Fig. 1B). In southern Chile, the SWW repre-sent one of the main climatic factors controlling precipitation on thewest side of the Andes (Garreaud, 2007; Garreaud et al., 2009). Thewesterlies are strongest during the austral summer, peaking between45° and 55°S. During the austral winter, the northern margin of theSWW belt expands northward up to 30°S, but the westerlies are weak-ened particularly at 50°S (Garreaud et al., 2009). In the region of TaitaoPeninsula, climate is cool-temperate and hyper-humid with low sea-sonality in temperature cycles and high precipitation throughout theyear (Fig. 1C). At the border of maximum intensity of SWW the precip-itation is between 2000 and 3000 mm.year−1. Southward, the precipi-tation increases with SWW intensity and can reach more than7000 mm.year−1 (Fig. 1A and C). The mean annual temperature nearto the coast of Taitao Peninsula reaches~8 °C and decreases southwardandwith increasing altitude. The regional ocean circulation patterns areorganized into several surface currents strongly influenced by the pre-vailing westerly winds. The Antarctic Circumpolar Current (ACC)

generates two currents (Fig. 1A): (1) an equatorward surface currentcalled the Peru-Chile Current (PCC) flowing from 38°S in the australwinter and from 45°S in the austral summer and, (2) a rather diffusepoleward surface current called the Cape Horn Current (CHC) flowingfrom 38°S in austral winter and from 46°S in austral summer (Strubet al., 1998). The deeper currents include the Antarctic IntermediateWater (AAIW) that moves northward between~500 and~1200 mwater depth and the Pacific Central Water (PCW) an abyssal currentthat flows southward below~1200 m water depth (Tsuchiya andTalley, 1998).

The vegetation (Fig. 1B), characterized by the North Patagonianrainforest, expands in the lowlands between 43° and 47–48°S(e.g. Schimithüsen, 1956; Veblen et al., 1983; Gajardo, 1987; Haberleand Bennett, 2001; Heusser, 2003; Luebert and Pliscoff, 2004). This veg-etation is mainly represented by evergreen broadleaves and conifertaxa such as species of Nothofagus, Pilgerodendron uviferum, Podocarpusnubigenus, Drimys winteri, Weinmannia trichosperma, with other plantsthat form the understorey such as Blechnummagellanicum, Pseudopanaxlaetevirens, Lomatia ferruginea, Philesia magellanica and several speciesof Myrtaceae (Haberle and Bennett, 2001; Heusser, 2003). In the archi-pelago, Nothofagus dombeyi grows preferentially eastward in the conti-nental forest whereas N. betuloides and N. nitida dominate the coastalforest. Poorly drained sites generate more humid conditions and openforests develop with trees such as Pilgerodendron uviferum and N. nitida(Haberle and Bennett, 2001). With increasing altitude, forest diversitydecreases. Above 400 m asl, long-lived heliophytic taxa, such asWeinmannia trichosperma, Pseudopanax laetevirens and someMyrtaceaedrop out and open forest similar to the Subantarctic rainforest developswith N. betuloides and Pilgerodendron uviferum (Fig. 1B). Above thetree-line at about 650–700 m asl and from 44° to 46°S, the typicalcushion plants characteristic of the Magellanic moorland developwith Astelia pumila, Donatia fascicularis and Oreobolus obtusangulus(Haberle and Bennett, 2004). Magellanic moorland grows more fre-quently southward along the coast from approximately 48° to 56°S(Fig. 1B), where low temperatures, high rainfall (≥4000 mm.year−1)and gale-force winds characterize the rigorous conditions associatedwith the SWW (Godley, 1960; Moore, 1983; Heusser, 1995). Therefore,a northward/southward expansion of Magellanic moorland would indi-cate northward/southward shift of the core of the SWW belt as alreadyshown during the Last Glacial Maximum (Villagrán, 1988a; Morenoet al., 1999).

In southern Chile, a study has shown that the pollen input in marinesurface sediment samples reflect the local vegetation from the nearbycontinental area (Montade et al., 2011). Today, the fine-grained terrig-enous material (including pollen) deposited off the southern Chile con-tinental slope is considered to be derived primarily from detrital origindue to the strong precipitation rate and high river discharge (Stuutet al., 2007). In core MD07-3088, the clay mineralogical assemblage insediments published by Siani et al. (2010) illustrate a clear sedimentprovenance from the surrounding continental rocks by river and/or gla-ciers linked to their expansion during the past. Under high precipitation,mountainous topography and numerous islands are the source ofnumerous short rivers which limit the long distance inputs of sedimentby rivers in core MD07-3088. The pollen study of Montade et al. (2011)shows that long distance drifting of pollen by oceanic currents remainsminor in this region. This low influence of oceanic currents could be re-lated to the high marine productivity that quickly aggregated pollen infilamentous algae or fecal pellets accelerating pollen sinking and to pol-len diagenesis by oxic degradation that reduces pollen drifting (e.g. Keilet al., 1994; Hooghiemstra et al., 2006). As it has been observed in off-shore sites off Patagonia (Heusser et al., 2006a,b; Montade et al.,2011), aeolian inputs of pollen remain minimal due to the year-roundonshorewinds. Accordingly, all these conditions limit long distance pol-len transport and suggest that the pollen content of core MD07-3088most likely represents the local to regional vegetation over Taitao Pen-insula area.

Patagonian SteppeSubantarctic Deciduous ForestMagellanic MoorlandSubantarctic RainforestNorth Patagonian RainforestValdivian RainforestLowland Deciduous Forest

CHC Cap Horn CurrentPCC Peru-Chile CurrentACC Antactic Circumpolar Current

Oceanic circulation

Vegetation

5 m/s

Wind

AAIW Antarctic Intermediaite Water

Precipitation

mm/yr

PCW Pacific Central Water

C

A

GMT0 200 400

km

N

W

S

E

350

500

500

350

10001500

2000

3000

40005000

60007000

6000

5000

40003000

20002500

20002500

2500

2000

2500

4000

3000

2000

1500

1000

4000

-80 -75 -70 -65

GMT0 200 400

km

N

W

S

E

-55

-50

-40

-45

B

ACC

CH

C

PCC

AA

IW

MD 07 3088

-80 -75 -70 -65

Peninsulade Taitao

Archipelagode Chonos

PacificOcean Isla grande de

tierra del Fuego

AtlanticOcean

HieloPatagonico

Sur

Hielo Patagonico

Norte

Cordillera de Darwin

Glacial limitduring LGM

PC

W

(1)

(2)

(1) Cabo Raper(46°49’S/75°34’W/40 m)

Precipitation Temperature

(2) Isla San Pedro(47°42’S/74°54’W/22 m)

0

100

200

300

400

500

02040

Mean annual temperature: 8.1°CMean annual precipitation: 4486 mm

J F M A M J J A S O N D

(°C) (mm)

0

50

100

150

200

250

010203040

J F M A M J J A S O N D

Mean annual temperature: 8.2°CMean annual precipitation: 2027 mm

(°C) (mm)

Fig 1. Climate and vegetation of Patagonia: (A) Precipitation map (McCulloch et al., 2000) with annual average of winds (Kalnay et al., 1996). (B) Vegetation map (afterSchimithüsen, 1956) with location of core MD07-3088 (red point), Patagonian icefield (gray area) and the main oceanic currents represented by arrows (Strub et al., 1998; Pizzaro,2004). (C) Ombrothermic diagrams (1 and 2) of the nearby meteorological stations (data from New_LocClim software; Grieser et al., 2006).

337V. Montade et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 369 (2013) 335–348

3. Materials and methods

3.1. Lithology and chronology

CoreMD07-3088 (46°04′S, 75°41′W; 18.91 m length; 1536 mwaterdepth) is located in the open ocean close to the coast of Taitao Peninsula(~50 km). This core was collected during the IMAGES-MD159-PACHIDERME cruise on board the R.V. “Marion Dufresne II” with the

Calypso giant piston corer. The sedimentation consists in a fairly uniformsuccession of olive black to grayish olive diatom and silt-bearing clayfor the upper 6 m corresponding to Holocene sedimentation (Fig. 2).For the remaining 12.9 m the sediments are dominated by grayisholive to gray silty clays characterized by frequent intercalations ofnormally graded silty/sandy layers of uncertain origin (Kissel and theshipboard scientific party, 2007). The age model detailed in Siani et al.(2010) has been establishedwith 11 AMS 14Cmeasurements performed

0 5 10 15 20 250

200

400

600

800

1000

1200

1400

1600

1800

2000

Depth (cm) Age cal kyr BP

tephra layers

silty/sandy layers

clay

HW7

HW3HW2

HW1

300 cm/kyr

60 cm/kyr

Fig. 2. Lithology and chronology of core MD07-3088 (after Siani et al., 2010).


on foraminifera and 4 tephrochronological markers (Table 1, Fig. 2).Calibrated age for the level 0–3 cm performed on epibenthic forami-nifera integrates a correction of 1545 yr for the deep water 14C age at1500 m according to Geosecs radiocarbon data (Stuiver et al., 1983).For the other dated horizons, a marine reservoir of ca. 600 yr has beentaken into account (Taylor and Berger, 1967; Ingram and Southon,1996). Tephra layers provide further independent chronologicalmarkers and are attributed to the Hudson volcano activity (Haberleand Lumley, 1998; Naranjo and Stern, 1998; Carel et al., 2011). In thisstudy, hereafter, all the ages will be discussed as calibrated (cal) yrBefore Present (BP).

3.2. Pollen analysis

All sediments were sieved through a 150 μmmesh screen, after dry-ing and volume measurement. Pollen samples were treated usingHCl and HF following a standard chemical protocol from Faegri and

Table 1Radiocarbon ages obtained on monospecific planktonic foraminifera and tephrochronologi(Pelletron 3MV) AMS facilities at the French CNRS-CEA Saclay and calibration of radiocarboby using CALIB 5 software (Stuiver and Reimer, 1993; Stuiver et al., 1998). Tephra layers w

Laboratorynumber

Core depth(cm)

Species

SacA 10551 0–3 Benthic foraminifera159–161 Tephra HW7

SacA 12815 280–281 G. bulloidesSacA 12816 360–362 G. bulloidesSacA 12817 500–501 G. bulloidesSacA 10553 539–541 G. bulloidesSacA 12819 610–611 G. bulloides

660–661 Tephra HW3750–751 Tephra HW2800–801 Tephra HW1

SacA 12822 870–871 G. bulloidesSacA 12825 1040–1041 G. bulloidesSacA 10555 1170–1172 G. bulloidesSacA 12827 1710–1711 G. bulloidesSacA 12828 1890–1891 G. bulloides

Iversen (1975). The residues were sieved through a 5 μm mesh screenand mounted on palynological slides with glycerine. Pollen grainswere counted and determined using a light microscope (OlympusBX51) at×400 or×1000 magnification and several pollen atlases havebeen used to assist with pollen identification (Heusser, 1971;Markgraf and d' Antoni, 1978; Villagrán, 1980). Prior to chemical treat-ment, a known amount of Lycopodiummarker (in calibrated tablet) hasbeen added to each sample in order to calculate the pollen concentra-tions (grains.cm−3). 137 pollen samples were counted and provide anaverage time resolution of~150 yr. A total of 92 pollen taxa were deter-mined and pollen percentages are based on sums ≥200 pollen grains.Percentages were calculated on the total pollen sum (excluding un-known, exotic pollen grains and spores). Spore percentages were calcu-lated on the total pollen sum (excluding unknown and exotic pollengrains). The pollen records were plotted using Psimpoll (Bennett,1994) and divided into zones on the basis of Constrained Cluster Anal-ysis by Sum of Squares analysis (CONISS) with pollen taxa that reachpercentages ≥1% (Grimm, 1987). Principal Component Analysis (PCA)was performed on the major pollen taxa (excluding pteridophytes)that reach percentages ≥5%.

3.3. Magnetic properties

U-channels were sampled in the central part of the half core formagnetic analyses. The measurements were done every 2 cm using755R-2G cryogenic magnetometers equipped with high resolutionpick-up coils and placed in the mu-metal shielded room of “Laboratoiredes Sciences du Climat et de l'Environnement”. The spatial resolution ofthe results is about 4 cm. Anhysteretic remanent magnetization (ARM)was acquired using a 1 cm/s translation speed in a 0.1 T AF field and a50 μT DC field. The isothermal remanent magnetization (IRM) wasobtained stepwise up to 1 T using a 160 cm long solenoid from 2G com-pany. To check their stability, both magnetizations were demagnetizedstepwise using alternating fields up to 80 mT and steps varyingbetween 5 and 10 mT.

4. Results

4.1. Pollen zones

The pollen record for core MD07-3088 is divided into 10 signifi-cant zones described and summarized in Table 2. The pollen percent-ages and concentrations are illustrated by the 2 pollen diagrams inFigs. 3 and 4, respectively.

cal markers (Siani et al., 2010). Radiocarbon dating were performed by UMS-ARTEMISn ages were performed according to INTCAL04 calibration curve (Reimer et al., 2004)ere correlated with on land tephra (Haberle and Lumley, 1998).

14C AMS age(yr BP)

± Error(yr)

Calibrated yr BP(range 1σ)

1595 30 36–681560

4475 30 4399–44395975 35 6212–62458430 30 8631–87289080 30 9519–954410,050 35 10,712–10,792

11,06013,89014,560

14,970 80 17,107–17,59415,755 45 18,589–18,69916,320 45 18,930–19,02617,610 50 20,108–20,30018,800 60 21,597–21,984

Table 2Pollen zones from core MD07-3088, indicating their depth, age range and their pollen signature linked to the relative climatic conditions.

Period Pollen zone Pollen signature Climate

Holocene Late PT-9b Steady representation in Nothofagus (50%) Modern conditions0–305 cm Increase in Misodendrum (5%), Drimys (5–10%), Lophosoria (10%) Temperature decrease and

humidity increase0–5 kyr Slight decrease in Myrtaceae (b5%, except for the 2 lasts samples)Middle PT-9a Increase in Nothofagus (40→50%)

305–425 cm Slight increase in Pilgerodendron (5%) and Lophosoria (5–10%)5–7.4 kyr Decrease in Myrtaceae (5%) and low percentages in Tepualia (1%), Pseudopanax (1%)

Early PT-8425–555 cm7.4–9.8 kyr

Regular increase in Nothofagus dombeyi type (30→40%) Warm and dryTemperature increase andhumidity decrease

Oscillations in Myrtaceae (5–10%) and Podocarpus (10%)Slight increase (1–5%) in Pseudopanax, Tepualia and LophosoriaDecrease in Astelia (b5%)

PT-7555–680 cm9.8-11.5 kyr

Oscillations in Nothofagus dombeyi type (20–30%)Slight decrease in Podocarpus (20→15%)Increase in Myrtaceae (5→10%) and Drimys (1–5%)Regular presence of Tepualia, Pseudopanax, Lomatia–GevuinaLow percentages in herbs and decrease in Astelia (5–10%)Regular presence in Cystopteris, slight decrease in Lycopodium (b5%)

Last Glacial InterglacialTransition

PT-6680–710 cm11.5–12.8 kyr

Increase in Nothofagus dombeyi type (20–30%) and Temperature increase buthigh humidity persistPodocarpus (15–20%)

Slight decrease in Astelia (10–15%), Gunnera (5–10%),Cyperaceae and Ericales (b5%)

Antarctic ColdReversal

PT-5710–755 cm12.8–14.1 kyr

Nothofagus dombeyi type (20%)High and abrupt humidity increase and pause in the warming trendHigher abundance in Astelia (15–25%) and slight increase inRanunculaceae (1–5%) and Podocarpus (5→10%)Peak in Ericales (10%) and Lepidothamnus (5%)

PT-4 Abrupt decrease in Nothofagus dombeyi type (40→20%)755–795 cm Abrupt increase in Astelia (10→20%)14.1–14.5 kyr Increase in Cyperaceae (5–10%) and peak in Gunnera (15%)PT-3 Regular decrease in Nothofagus dombeyi type (50→40%) Temperature795–820 cm Increase in Astelia (3–8%),14.5–15.4 kyr Cyperaceae (4–7%) and Gunnera (10%)PT-2820–900 cm15.4–17.6 kyr

Regular decrease in Nothofagus dombeyi type (60→50%)Slight decrease in Poaceae (5–10%), Ericales (1–6%),Apiaceae and Ranunculaceae (1%)Increase in Astelia and Cyperaceae (2→6%)

Late Periglacial PT-1900–1890 cm>17.6 kyr

Nothofagus dombeyi type abundant (50–60%) ColdPodocarpus (5–10%), Myrtaceae, Pilgerodendron and Drimys (b5%)Poaceae (5–15%), Gunnera (5–10%), Ericales (1–8%), Apiaceae (1–7%)Oscillations in Astelia, Asteraceae subfs asteroideae, Cyperaceae,Ranunculaceae and Lycopodium (1–5%)


4.2. PCA analysis

Principal Component Analysis (PCA) is represented by twodimensional-plots or biplots between pollen taxa and each pollensamples. Fig. 5A shows the eigenvectors for major pollen taxa in thisstudy and PCA plot for Axis 1 and 2 are represented in Fig. 6. TheAxis 1mainly contrastsDrimys, Pilgerodendron, Myrtaceae, Podocarpuswith Apiaceae, Poaceae, Asteraceae (asteroideae), Gunnera, Ericales,Cyperaceae. This first component seems to be related to forest cover:closed-canopy forest with positive values versus open vegetationunder cooler conditions with negative values. Axis 2 mainly contrastsNothofagus dombeyi type with Astelia pumila. N. dombeyi type corre-sponds to several tree species with wide ecological range andA. pumila is a cushion plant characteristic of the Magellanic moorlandthat generally grows under more humid conditions than Nothofagustree species. This second component seems to be related to precipita-tion influence: Nothofagus woodland-less humid with positive valuesversus Magellanic moorland-more humid with negative values.Fig. 5B plots the pollen samples according to the 10 pollen zones.Their succession in the scheme reflects the vegetation changes in thepollen record from the glacial to the present-day conditions.

5. Palaeoenvironment history from core MD07-3088

5.1. Late Glacial Period (22 to 17.6 kyr)

Before 17.6 kyr (TP-1, 900–1890 cm), pollen concentrations showvery low values with samples frequently barren of pollen (Figs. 4

and 6). At this time, the average sedimentation rate (300 cm/kyr) ishigher than those recorded during the deglaciation and Holocene(60 cm/kyr) and clay-mineral assemblages suggest an increase oferosion processes by the Coastal Range glaciers (Siani et al., 2010).In addition, fluctuations in the magnetic properties indicate abruptlarge amplitude coarsening of the magnetic grains (lower values ofthe ARM/IRM ratio) accompanied by decreases in the concentrationof magnetic particles (decrease in ARM) (Fig. 6). This suggests ice-rafted detrital discharges ("magnetic IRD"; Kissel, 2005). Together,these results indicate predominant inputs of sediments from massivedischarges of debris generated by glaciers which were widely ex-tended over the region of Taitao Peninsula during the last glacialperiod (Heusser, 2002). Sediments carried by glaciers have very loworganic matter content (including pollen) due to high dilution bythe large amounts of inorganic matter from glaciers (Silva, 2008)and the presence of Patagonian ice cap over the continent has re-stricted vegetation development reducing pollen productivity. Theseprocesses explain the low pollen concentrations recorded before17.6 kyr.

Pollen spectra show poorly diversified vegetation and the domi-nance of the Nothofagus dombeyi type (Fig. 3). In fact, this pollentype includes several tree species which cannot be distinguished bytheir pollen morphology and which are often over-represented inthis region. In addition, their high pollen productivity and their longdistance pollen transport by winds (e.g. Heusser, 1971; Haberle andBennett, 2001) explain the high pollen percentages of N. dombeyitype (60%) recorded in core MD07-3088. Herbaceous pollen taxasuch as Poaceae and Apiaceae are well represented and reach their



maximum percentages between 22 and 17.6 kyr (Fig. 3). Thesepollen taxa are associated with relatively high values of Gunnera,Ranunculaceae, Asteraceae and Ericales suggesting the presence ofopen vegetation developed under cold and relative dry conditions,also highlighted by the maximum values of PCA Axis 1 (Fig. 6). Thisvegetation might correspond to the Subantarctic Parkland character-ized by Nothofagus-Poaceae communities which was widespread onglacial and glacio-fluvial deposits (Heusser et al., 1996). In such vege-tation, Nothofagus trees expanded in the wind-protected areas whilethe windy lowlands are largely occupied by Poaceae, Asteraceae,Empetrum-Ericales, and other non-arboreal taxa. Consequently, pollenproductivity prior to 17.6 kyr (except for Nothofagus dombeyi type)could originate from scattered vegetation that developed along theChilean coast on the margins and moraines of the south Patagonianice sheet. Planktonic δ18O values recorded in core MD07-3088 reflectglobal sea water δ18O signal and local regional sea surface tempera-ture and salinity changes (Fig. 6). During the last glacial period,short-term variability observed in the planktonic δ18O values andalso recorded by clay-mineral assemblages seems to be attributed tothe glacier dynamic (Siani et al., 2010). Low pollen concentrations be-fore 17.6 kyr and the different transport mechanisms of pollen incomparison with present-day conditions limit the interpretation ofpollen data to highlight such a short-term variability. However, before17.6 kyr, summer sea surface temperature from foraminifera assem-blages are 5 °C lower than the modern one (Siani et al., submittedfor publication) which is consistent with pollen results from thesame core and sea surface temperature recorded further north(41°S) in site ODP 1233 (Lamy et al., 2004; Kaiser et al., 2005,2008).

5.2. Late Glacial Interglacial Transition (17.6 to 11.5 kyr)

At 17.6 kyr (TP-2, 820–900 cm), the pollen concentration increasesabruptly and the sedimentation rate drops (Fig. 4). Initially, this pollenconcentration increase can be explained by the sedimentation rate de-crease. However, the total pollen concentration continues to increaseafter the sedimentation rate change occurring at around 17.3 kyr and870 cm depth. Simultaneously clay-mineral assemblages indicate amajor change (Siani et al., 2010) and magnetic properties vary muchmore smoothly, most likely illustrating a predominant transport ofsediment by river inputs (Fig. 6). This is not an apparent smoothingresulting from lower sedimentation rates because, although reduced,the latter still give us a time resolution in the magnetic propertiesof about 60 years. These changes mark the glacial retreat and thesediment transport mechanisms could be considered as similar toHolocene/present-day conditions, with a predominance of fluvial sedi-ment input from the surrounding continental area.

The continuous increase in pollen concentrations thus illustrates thecolonization of vegetation on land linked to the ice cap retreat. Percent-ages ofNothofagus dombeyi type, always overrepresented, decrease pro-gressively due to the development of other taxa (Fig. 3). Nevertheless,N. dombeyi type concentrations continue to rise until 15.4 kyr whichsuggests the expansion of Nothofaguswoodland (Fig. 4). This forest ex-pansion is associated with the glacial retreat and marks the first-stepwarming of the last deglaciation at~17.6 kyr, also recorded by a majordecrease of the δ18O values (Fig. 6). This decrease from 17.6 to14.5 kyr of isotopic values is mainly driven by a regional warming of3 °C (Siani et al., submitted for publication). After 15.4 kyr (TP-3,795–820 cm), N. dombeyi type percentage decreases and wetland taxaincrease slightly. This pattern is strengthened between 14.5 and12.8 kyr (TP 4, 755–795 cm and TP 5, 710–755 cm) by the abrupt in-crease of Astelia (A. pumila) that coincides with the drop of N. dombeyitype in percentages and concentrations (Figs. 3 and 4). Astelia is a

Fig. 3. Pollen percentage diagram from core MD07-3088 divided into 10 pollen zones with chshows the sedimentation rate change from the glacial period to the deglacial and Holocene

typical plant of the Magellanic moorland that grows frequently southof 48°S in the core of SWW belt under cool and very wet conditionswith precipitation ≥4000 mm.year−1 (Godley, 1960; Moore, 1983;Heusser, 1995). Such vegetation changes illustrate a reversal trend inthe forest development with the expansion of theMagellanic moorlandat the expense of the Nothofaguswoodland. This suggests an abrupt in-crease in precipitation and a slight decrease in temperature or at least apause in the warming trend with an increasing influence of SWW. ThePCA axis also underlines this pattern, with a major shift in values ofAxis 2 related to the Astelia expansion and with a reversal trend invalues of Axis 1 between 15.5 and 13 kyr (Fig. 6). Shortly after Asteliapeaks, at around 14 kyr Podocarpus increase gradually. Between 12.8and 11.5 kyr (TP-6, 680–710 cm), the decline of Astelia and theNothofagus dombeyi type re-increase, associated with the maximumpercentages of Podocarpus, points to the Magellanic moorland retreatand the expansion of North Patagonian rainforest. These vegetationchanges illustrate a second-step warming, although with rather cooland humid conditions related to the Astelia still maintaining high per-centages. Therefore, the pollen record illustrates a deglaciation withtwo-step warming interrupted by a reversal trend also evidenced bythe δ18O record (Fig. 6) and summer sea surface temperature from thesame core (Siani et al., submitted for publication).

5.3. Holocene (11.5 kyr to present)

After 11.5 kyr (TP-7, 555–680 cm),Astelia decrease significantly andthe North Patagonian rainforest continues to expand from 11 to 10 kyrwith Myrtaceae, Pseudopanax and Tepualia (Fig. 3). The PCA Axis 1shows also the increase of the closed-canopy forest that reaches nega-tive values after 11 kyr (Fig. 6). Such a change in vegetation indicatesa decrease in precipitation and, therefore, in the influence of SWWwith higher temperatures than present-day ones highlighted by themaximum expansion of Myrtaceae and Tepualia. These conditionsmark the Pleistocene/Holocene transition recorded by the lowest δ18Ovalues, indicating the warmest period, before a trend towards highervalues from about 10 kyr (Fig. 6). After 7.4 kyr (TP-9a, 305–425 cm),Myrtaceae, Pseudopanax and Tepualia decrease and a slight expansionoccurs for Pilgerodendron, Nothofagus dombeyi type and Lophosoria.The tree Pilgerodendron (P. uviferum) is a conifer typical of poorlydrained sites of the North Patagonian rainforest (Haberle and Bennett,2001). This tree expansion indicates cooler and wetter climate condi-tions from 7.4 kyr. The rainforest reaches modern conditions at about5 kyr until the present (TP-9b, 0–305 cm) with a stable representationof the main taxa of trees such as Nothofagus dombeyi type, Podocarpus,Myrtaceae, Pilgerodendron and Drimys, suggesting no major changes inclimate conditions up to present-day climate conditions. Short-termchanges in clay-mineral composition and δ18O record suggestingsea-surface temperature and/or salinity variability during the Holocenecould be indicative of short glacier advance events (Siani et al., 2010).Short-term in variability in sedimentation is also depicted by magneticgrain size (Fig. 6), however these events are not observed in pollenspectra. This could be due to the lower temporal resolution of pollenanalyses during the Holocene and/or the permanent closed-canopyrainforest in the region of Taitao Peninsula was less sensitive to rapidclimate changes.

6. Large scale comparison

During the late glacial period, the Patagonian ice cap was extendedover 1880 km along the Andes (Fig. 1) and covered the relief from theArgentina Andean piedmont to the Pacific shoreline (Hollin andShilling, 1981). However, the glacial limit on the Pacific side remains

ronostratigraphy indicated on the right and calibrated ages on the left. The dashed lineperiods.


Nothofagus dombeyi type

PodocarpusERICALES

Pilgerodendron

MYRTACEAE

Drimys

POACEAE

Gunnera

APIACEAE

ASTERACEAE asteroideae

CYPERACEAE

Astelia pumila

-8-7-6-5-4-3-2-10123456

-9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12

Axi

s 2

(21.

97 %

)

Axis 1 (32.73 %)

A

Open Vegetation

Forest

Warm Cool Wet

-8-7-6-5-4-3-2-10123456

-9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12

Axi

s 2

(21.

97 %

)

Axis 1 (32.73 %)

B

TP-1TP-2TP-3TP-4TP-5TP-6TP-7TP-8TP-9aTP-9b

Fig. 5. Principal Component Analysis (PCA) between pollen percentages and pollensamples with biplot centered on the origin. The sum of the two Axes reaches 54.71%of the original variability (Axis 1=32.73% and Axis 2=21.97%). (A) Species biplotshowing major pollen taxa in pollen analysis ≥5%. (B) Plot of the pollen sampleswith the colored polygons that represent the 10 pollen zones. The black arrows under-line the succession of the pollen zones.


unknown and the existence of unbroken cover of ice is conjectural(Heusser, 2002). The glaciers were at their maximum extent between34 and 26 kyr (Clapperton et al., 1995; Denton et al., 1999; Glasseret al., 2004; Kaplan et al., 2004) and persisted in Taitao Peninsula, atLaguna Stibnite (46°S), with a local ice cap up to 17.4 kyr (Lumley andSwitsur, 1993; Heusser, 2002). In the Chonos Archipelago (44°S) closeto Taitao Peninsula, the rapid expansion of forest after the deglaciationsuggests the presence of forest refugia during the last glacial period onthe western coastal margin probably partly ice-free (Haberle andBennett, 2004). The presence of such forest refugia could explain thepresence of pollen in core MD07-3088 with low diversity before17.6 kyr.

Rapid expansion of Nothofagus woodland and decrease of δ18Ovalues mark the temperature increase and the glacial retreatat~17.6 kyr. These palaeoenvironment data from core MD07-3088illustrate the onset of the last deglaciation and are consistent with thefirst-step warming trend recorded by δD from EPICA Dome C (EDC)ice core (Fig. 6; Lemieux-Dudon et al., 2010; Stenni et al., 2011). Theonset of the last deglaciation in the studied region is generally recordedbetween 18 and 17 kyr according to the locations of the numerouspaleoenvironmental records from southern Chile (e.g. Heusser, 1966;

Fig. 4. Pollen concentration diagram from core MD07-3088 with the pollen zones and chroshows the sedimentation rate change from the glacial period to the deglacial and Holocen

Villagrán, 1990; Heusser et al., 1995, 1996; Bennett et al., 2000;Moreno et al., 2001; Massaferro and Brooks, 2002; Abarzúa et al.,2004; Moreno, 2004). At Laguna Stibnite, the nearest site to coreMD07-3088 (Lumley and Switsur, 1993; Bennett et al., 2000),Nothofagus woodland expansion occurs at~17.3 kyr, almost simulta-neously with Nothofagus woodland expansion from the marine core(Fig. 6). In the continental pollen records from the Chonos Archipelagothis forest expansion occurs later between 17 and 15 kyr, which high-lights different time-lags depending on the site (Fig. 7 and SM 1;Bennett et al., 2000; Haberle and Bennett, 2004). Three tephra layersprovide further independent chronological markers during the last de-glaciation in coreMD07-3088 andminimize the uncertainties ofmarinereservoir age changes as the origin of these differences. However, it re-mains difficult to determinewhether the differences observed betweenmarine and terrestrial records are the result of radiocarbon dating un-certainties and/or time-lags of vegetation response according to sitelocations.

Between 14.5 and 12.8 kyr, the Magellanic moorland expansion(mainly Astelia) in core MD07-3088 interrupts the forest expansionand marks an increase in precipitation and a slight cooling or atleast a pause in the warming trend which is consistent with geochem-ical data from the same core (Fig. 6). This event is well constrained bytephrochronological markers and occurs simultaneously with the ACRrecorded in EDC ice core (e.g. Jouzel et al., 1995; Stenni et al., 2011).Other records suggest that this climate reversal has affected themid-latitudes of the southern hemisphere as illustrated by vegetationchanges and glacial advances in Patagonia (Heusser et al., 2006b;Lamy and Kaiser, 2009; Moreno et al., 2009) and New Zealand(Vandergoes et al., 2008; Putnam et al., 2010).

The lacustrine record at Laguna Stibnite does not show any signif-icant occurrence of Magellanic moorland but a progressive warmingtrend through the last deglaciation with no temperature reversal(Lumley and Switsur, 1993; Bennett et al., 2000). However,Magellanic moorland expansion depicted in core MD07-3088 occurssimultaneously with Pilgerodendron expansion from Laguna Stibnite(Fig. 6). Pilgerodendron is a tree growing in poorly drained sites(Haberle and Bennett, 2001), and its expansion also underlines amoisture increase correlated with Magellanic moorland increase re-lated to precipitation increases recorded in core MD07-3088. In addi-tion, in the Chonos Archipelago, burning is common before~14.5 kyr,then becomes absent between 14.5 and 12.5 kyr, which may also il-lustrate precipitation increase (Haberle and Bennett, 2004). There-fore, the Magellanic moorland would have preferentially developedon the coastal area of Taitao Peninsula where it is more sensitive tooceanic influence, which could explain why the marine pollen dataalone show this expansion.

This moisture increase during the ACR could be related to an inter-mediate position of the SWW belt between their full northward gla-cial position and their southward Holocene position explaining theMagellanic moorland expansion in core MD07-3088. During the lastglacial period, Magellanic moorland was developed in northern Pata-gonia (~41°S) suggesting the northward position of the SWW belt(Villagrán, 1988a; Moreno et al., 1999). Then, until the Holocene,Magellanic moorland would have progressively developed southwardalong the coast, driven by a shift of the SWW belt from Reloncavi re-gion toward the south. However, Magellanic moorland has not beenreported north of Taitao Peninsula in the most coastal sites of ChonosArchipelago or Isla de Chiloé (Fig. 7 and SM 1; Villagrán, 1988b;Bennett et al., 2000; Haberle and Bennett, 2004). Therefore, theassumption of a progressive southward shift of SWW belt does notexplain the successive events of vegetation changes during the lastdeglaciation.

nostratigraphy indicated on the right and calibrated ages on the left. The dashed linee periods.


Fig. 7. Schematic changes of vegetation during the last deglaciation to illustrate significant expansion of Magellanic moorland, according to different records of pollen near TaitaoPeninsula. The dashed line corresponds to the hypothetical limit between Magellanic moorland mixed with forest. The numbers correspond to the site locations discussed in thetext: 1 Canal de la Puntilla, Lago Condorito (Moreno et al., 1999; Moreno, 2004); 2 Río Negro (Villagrán, 1988a); 3 Laguna Soledad, Laguna Chaiguata, Puerto Carmen (Villagrán,1988b); 4 Laguna Oprasa (Haberle and Bennett, 2004); 5 Laguna Facil (Haberle and Bennett, 2004); 6 Laguna Lofel (Bennett et al., 2000); 7 Laguna Lincoln (Bennett et al.,2000); 8 Laguna Stibnite (Lumley and Switsur, 1993; Bennett et al., 2000); 9 Laguna Six Minute (Bennett et al., 2000); 10 Puerto Eden (Ashworth et al., 1991). The star denotedthe location of marine core MD07-3088.


A northward shift and/or expansion of the SWW belt couldexplain the Magellanic moorland expansion in the Taito Peninsuladuring the ACR. However, the northern expansion and/or location ofthe SWW belt would have been of less importance than during theLast Glacial Maximum since Magellanic moorland are absent at thenorthern sites during the ACR. We consider this scheme to be a betterexplanation for the regional vegetation changes than a progressivesouthward shift of the SWW belt. Consequently, this implies a rapidsouthward shift of the SWW belt at the onset of the last deglaciationwhich has not permitted a significant development of the Magellanicmoorland vegetation southward along the coast. In addition, a rapidsouthward shift at the onset of the last deglaciation is consistentwith the increase of westerlies driven-upwelling in Southern Oceandepicted from opal flux record (Anderson et al., 2009). During theACR, a decrease of westerlies driven-upwelling in the SouthernOcean as depicted by the opal flux record could be related to intensitydecrease or northward shift of the SWWbelt. Our results illustrating a

Fig. 6. Isotopic, magnetic, PCA axis and pollen records with pollen zones from core MD07-30CO2 from EDC ice core (Monnin et al., 2001; Lemieux-Dudon et al., 2010; Stenni et al., 2011bulloides (Siani et al., 2010); ARM (IRM) are for anhysteretic (isothermal) remanent magnetand ARM/IRM ratio is for the average magnetic grain size (high ARM/IRM values corresponfrom cooler open vegetation (positive values) with closed-forest (negative values), the secocipitation increase (negative values); pollen concentration and simplified pollen diagram.Switsur, 1993; Bennett et al., 2000).

local intensity increase of the SWW during the ACR, which suggests anorthward shift of the SWW belt rather than a global intensity de-crease of the SWW. This regional scheme with a rapid southwardshift of the SWW belt at the onset of the last deglaciation and itsnorthward shift during the ACR is synchronous, respectively, withthe increase and the interruption of the atmospheric CO2 changesregistered in EDC ice core (Fig. 6; Monnin et al., 2001). Despite uncer-tainties about SWW inducing CO2 degassing shown by simulations(Tschumi et al., 2008), the results presented here support the ideathat prolonged westerly effects over the Southern Ocean wouldhave contributed to raising atmospheric CO2 above the threshold nec-essary to sustain interglacial conditions (Toggweiler et al., 2006) andstrengthen the link between SWW belt and CO2 changes during thelast deglaciation.

After~12.8 kyr (Fig. 6), the Magellanic moorland declines whenPodocarpus show highest values. This illustrates a slight climaticshift suggesting a precipitation decrease and a temperature increase

88 against EDC ice core and Laguna Stibnite records during the last 22 kyr. (A) δD and). (B) Variations of oxygen isotope (δ18O) from the planktonic foraminifera Globigerinaizations. The ARM changes illustrate variations in the concentration of magnetic grainsd to finer magnetic grains); PCA plot for Axis 1 and Axis 2, the first Axis contrasts taxand Axis contrasts Nothofagus woodland (positive values) with moorland linked to pre-(C) Simplified pollen diagram from Taitao Peninsula, at Laguna Stibnite (Lumley and


consistent with the second-step warming recorded by δD values fromEDC ice core (Fig. 6; Lemieux-Dudon et al., 2010; Stenni et al., 2011).At Laguna Stibnite, Podocarpus develops rapidly at the expense ofPilgerodendron forest (Fig. 6; Lumley and Switsur, 1993; Bennettet al., 2000). Marine and continental vegetation records highlightthe same climatic trends with temperature increase and precipitationdecrease also confirmed by increase in fire activity after 12.5 kyr inthe Chonos Archipelago (Haberle and Bennett, 2004). Togetherthese results illustrate a decrease in the SWW belt influence thathad probably shifted southward synchronous with the second-stepincrease in the atmospheric CO2 registered in EDC ice core (Fig. 6;Monnin et al., 2001).

This warming trend is strengthened at~11.5 kyr with an abruptdecline in Magellanic moorland that moves southward at 49°S(Fig. 7) near to the present-day location of the core of the SWWbelt (Ashworth et al., 1991). These vegetation changes mark intergla-cial conditions of the Holocene and it is consistent with δ18O values inthe same core, continental records of southern Chile (e.g. McCullochand Davies, 2001; Abarzúa et al., 2004; Haberle and Bennett, 2004)and δD values from EDC ice core (Fig. 6; Lemieux-Dudon et al.,2010; Stenni et al., 2011). At 11 kyr up to 7.4 kyr, pollen resultsfrom core MD07-3088 show an increase of heliophytic trees, indi-cated by Tepualia and Myrtaceae, and mark the warmest and driestconditions of the Holocene. These results fit well with vegetationchanges recorded at Laguna Stibnite (Fig. 6), especially concerningTepualia expansion that also points out the warmest and driest condi-tions of the Holocene (Lumley and Switsur, 1993; Bennett et al.,2000). After 7.4 kyr, the vegetation shows more variable representa-tion of heliophytic taxa up to 5 kyr, after which modern values arereached illustrating the dominance of the present North Patagonianrainforest under cooler and wetter conditions. At Laguna Stibnite,Tepualia decrease at~8 kyr marking the end of the warmest anddriest conditions. Depending on the site locations, the end of this cli-maticmaximumfluctuates between8 and 6 kyr andmodern conditionsare reached generally after 6–5 kyr (Villagrán, 1988b; Ashworth et al.,1991; Bennett et al., 2000; Haberle and Bennett, 2004). These differ-ences could reflect age uncertainties, but also time-lag response of veg-etation changes to a progressive northward shift or expansion of theSWW belt from the mid-Holocene.

During the late Holocene Pilgerodendron show an abrupt drop atLaguna Stibnite and Six Minutes (Fig. 6 and SM 1; Bennett et al.,2000). These vegetation changes are not observed in other lacustrineand marine pollen records. However, lacustrine records from theChonos Archipelago (Laguna Oprasa and Laguna Facil) indicate an in-crease in fire-related disturbance after~2.8 kyr. Pilgerodendron de-crease seems to be related to local disturbances such as volcaniceruptions, anthropogenic activity and local fire disturbance and notrelated to a global climate change.

7. Conclusion

A new high-temporal resolution of marine pollen analyses fromcore MD07-3088 represents a continuous record of vegetationchanges in Taitao Peninsula region during the last 22 kyr. Comparisonbetween the marine pollen record and geochemical records from thesame core illustrates a consistent pattern with close timing betweenthe ocean and land responses to climate change. Timing of vegetationchanges from the marine core are similar to those observed in conti-nental pollen records. Despite short time-lags dependent to the sitelocations, this comparison allows us to reconstruct the regional pat-tern of vegetation changes and to differentiate local environmentalchanges from regional environmental changes.

Forest expansion from coreMD07-3088 between~17.6 and 14.5 kyrillustrate glacial retreat and the first-step warming of the last deglacia-tion simultaneously depicted by geochemical data in the same core andby terrestrial pollen record at the same latitude (Laguna Stibnite).

Coincident with the ACR, forest development is interrupted by anabrupt expansion of Magellanic moorland. The expansion ofMagellanicmoorland along the coast marks at least a pause in the regionalwarming trend and an abrupt increase in precipitation that we relateto local increase in SSW intensity. In a broader context, this suggests arapid southward shift of the SWW belt followed by a slight northwardshift of SWW belt during the ACR. After 12.8 kyr, Magellanic moorlandretreat and the North Patagonian rainforest expansion show second-step warming of the last deglaciation with a slight precipitation de-crease related to decreased influence of the SWWbelt as it shifts south-ward. These trends are strengthened at~11.5 kyr at the onset ofHolocene, though it is not until around 5 kyr that vegetation reachesmodern conditions in the region of Taitao Peninsula. The pollen recordfrom core MD07-3088 illustrates a clear synchronicity with climatechanges recorded in EDC ice core and correspond to an important steptowards constraining the position and intensity of the SWW duringthe last deglaciation. These new data demonstrate that the ACR affectedthe mid-latitudes of the south-eastern Pacific area modifying the SWWbelt and supporting the link between SWWbelt and CO2 degassing dur-ing the last deglaciation.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.palaeo.2012.11.001.

Acknowledgments

This research was supported by Laboratoire des Sciences du Climatet de l'Environnement (LSCE), Centre National de la RechercheScientifique, Commissariat à l'Energie Atomique, Université VersaillesSt Quentin and Université Paris Sud 11. Financial support for thisstudy was provided by the French INSU-LEFE-Pachiderme projectand this study has been conducted in the framework of the interna-tional IMAGES program. We would like to thank Institut PolaireFrançais Paul Emile Victor and the shipboard scientific party for theMD159/PACHIDERME cruise to collect the core material and data onboard of the French vessel “Marion Dufresne II”. We express ourthanks to K.D. Bennett for providing continental pollen record dataand to UMS-ARTEMIS group for AMS radiocarbon dating. We arealso grateful for the very constructive comments from the editorand two anonymous referees in improving this paper. This is LSCEcontribution no. 4932.

References

Abarzúa, A.M., Villagrán, C., Moreno, P.I., 2004. Deglacial and postglacial climate historyin east-central Isla Grande de Chiloé, southern Chile (43°S). Quaternary Research62, 49–59.

Anderson, R.F., Ali, S., Bradtmiller, L.I., Nielsen, S.H.H., Fleisher, M.Q., Anderson, B.E.,Burckle, L.H., 2009. Wind-driven upwelling in the southern ocean and the deglacialrise in atmospheric CO2. Science 323, 1443–1448.

Ashworth, A.C., Markgraf, V., Villagrán, C., 1991. Late Quaternary climatic history of theChilean Channels based on fossil pollen and beetle analyses, with an analysis of themodern vegetation and pollen rain. Journal of Quaternary Science 6, 279–291.

Bennett, K.D., 1994. “psimpoll” version 2.23: a C program for analysing pollen data andplotting pollen diagrams. INQUA Commission for the study of the Holocene:Working group on data-handling methods. Newsletter 11, 4–6.

Bennett, K.D., Haberle, S.G., Lumley, S.H., 2000. The Last Glacial–Holocene transition inSouthern Chile. Science 290, 325–328.

Carel, M., Siani, G., Delpech, G., 2011. Tephrostratigraphy of a deep-sea sediment se-quence off the south Chilean margin: new insight into the Hudson volcanic activitysince the last glacial period. Journal of Volcanology and Geothermal Research 208,99–111.

Clapperton, C.M., Sugden, D.E., 1988. Holocene glacier fluctuations in South Americaand Antarctica. Quaternary Science Reviews 7, 185–190.

Clapperton, C.M., Sugden, D.E., Kaufman, D.S., McCulloch, R.D., 1995. The Last Glaciationin Central Magellan Strait, Southernmost Chile. Quaternary Research 44, 133–148.

Combourieu Nebout, N., Turon, J.L., Zahn, R., Capotondi, L., Londeix, L., Pahnke, K., 2002.Enhanced aridity and atmospheric high-pressure stability over the westernMediterranean during the North Atlantic cold events of the past 50 k.y. Geology30, 863–866.

Denton, G.H., Lowell, T.V., Heusser, C.J., Schluchter, C., Andersen, B.G., Heusser, L.E.,Moreno, P.I., Marchant, D.R., 1999. Geomorphology, stratigraphy, and radiocarbonchronology of Llanquihue drift in the area of the southern Lake District, SenoReloncavi, and Isla Grande de Chiloe, Chile. Geografiska Annaler A81, 167–229.




Desprat, S., Sánchez Goñi, M.F., Naughton, F., Turon, J.-L., Duprat, J., Malaizé, B., Cortijo, E.,Peypouquet, J.-P., 2007. 25. Climate Variability of the Last Five Isotopic Interglacials:Direct Land-sea-ice Correlation from the Multiproxy Analysis of North-WesternIberianMargin Deep-sea Cores. The Climate of Past Interglacials. Elsevier, pp. 375–386.

Dupont, L.M., Wyputta, U., 2003. Reconstructing pathways of aeolian pollen transportto the marine sediments along the coastline of SW Africa. Quaternary ScienceReviews 22, 157–174.

Faegri, K., Iversen, J., 1975. In: John Wiley & Sons (Ed.), Textbook of Pollen Analysis(London).

Gajardo, R., 1987. La Végétation naturelle du Chili: proposition d'un système de classi-fication et représentation de la distribution géographique.

Garreaud, R.D., 2007. Precipitation and circulation covariability in the extratropics.Journal of Climate 20, 4789–4797.

Garreaud, R.D., Vuille, M., Compagnucci, R., Marengo, J., 2009. Present-day South Americanclimate. Palaeogeography, Palaeoclimatology, Palaeoecology 281, 180–195.

Glasser, N.F., Harrison, S., Winchester, V., Aniya, M., 2004. Late Pleistocene andHolocene palaeoclimate and glacier fluctuations in Patagonia. Global and PlanetaryChange 43, 79–101.

Godley, E.J., 1960. The Botany of Southern Chile in Relation to New Zealand and theSubantarctic. Proceedings of the Royal Society of London 152, 457–475.

Grieser, J., Giommes, R., Bernardi, M., 2006. New_LocClim — the Local Climate Estima-tor of FAO. Geophysical Research Abstracts 8, 08305.

Grimm, E., 1987. CONISS: a FORTRAN 77 program for stratigraphically constrained clusteranalysis by the method of incremental sum of squares. Computers & Geosciences 13,13–35.

Haberle, S.G., Bennett, K.D., 2001. Modern pollen rain and lake mud-water interfacegeochemistry along environmental gradients in southern Chile. Review ofPalaeobotany and Palynology 117, 93–107.

Haberle, S.G., Bennett, K.D., 2004. Postglacial formation and dynamics of North Patago-nian Rainforest in the Chonos Archipelago, Southern Chile. Quaternary ScienceReviews 23, 2433–2452.

Haberle, S.G., Lumley, S.H., 1998. Age and origin of tephras recorded in postglacial lakesediments to the west of the southern Andes, 44°S to 47°S. Journal of Volcanologyand Geothermal Research 84, 239–256.

Heusser, C.J., 1966. Late-Pleistocene pollen diagrams from the Province of Llanquihue,Southern Chile. Proceedings of the American Philosophical Society 110, 269–305.

Heusser, C.J., 1971. In: The University of Arizona Press (Ed.), Pollen and Spores of Chile(Tuscon, Arizona).

Heusser, C.J., 1995. Palaeoecology of a Donatia-Astelia cushion bog, MagellanicMoorland-Subantarctic evergreen forest transition, southern Tierra del Fuego,Argentina. Review of Palaeobotany and Palynology 89, 429–440.

Heusser, C.J., 2002. On glaciation of the southern Andes with special reference to thePenínsula de Taitao and adjacent Andean cordillera (~46°30's). Journal of SouthAmerican Earth Sciences 15, 577–589.

Heusser, C.J., 2003. Ice age Southern Andes - A chronicle of paleoecological events,Developments in Quaternary Science. ed. Elsevier, Amsterdam.

Heusser, L., Balsam, W.L., 1977. Pollen distribution in the northeast Pacific Ocean.Quaternary Research 7, 45–62.

Heusser, C.J., Denton, G.H., Hauser, A., Andersen, B.G., Lowell, T.V., 1995. Quaternarypollen records from the Archipiélago de Chiloé in the context of glaciation andclimate. Revista Geologica de Chile 22, 25–46.

Heusser, C.J., Lowell, T.V., Heusser, L.E., Hauser, A., Andersen, B.G., Denton, G.H., 1996.Full-glacial–late-glacial palaeoclimate of the southern Andes: evidence from thepollen, beetle and glacial records. Journal of Quaternary Science 11, 173–184.

Heusser, L., Heusser, C., Mix, A., McManus, J., 2006a. Chilean and Southeast Pacificpaleoclimate variations during the last glacial cycle: directly correlated pollenand δ18O records from ODP Site 1234. Quaternary Science Reviews 25, 3404–3415.

Heusser, L., Heusser, C., Pisias, N., 2006b. Vegetation and climate dynamics of southernChile during the past 50,000 years: results of ODP Site 1233 pollen analysis.Quaternary Science Reviews 25, 474–485.

Hollin, J.T., Shilling, D.H., 1981. Late Wisconsin–Weichselianmountain Glaciers and SmallIce Caps. In: Denton, G.H., Hughes, T. (Eds.), The Last Great Ice Sheets, pp. 179–220(New York).

Hooghiemstra, H., Lézine, A.-M., Leroy, S.A.G., Dupont, L., Marret, F., 2006. Late Quater-nary palynology in marine sediments: a synthesis of the understanding of pollendistribution patterns in the NW African setting. Quaternary International 148,29–44.

Ingram, B.L., Southon, J.R., 1996. Reservoir ages in eastern pacific coastal andestuarinewaters. Radiocarbon 38, 573–582.

Jouzel, J., Vaikmae, R., Petit, J.R., Martin, M., Duclos, Y., Stievenard, M., Lorius, C., Toots,M., Mélières, M.A., Burckle, L.H., Barkov, N.I., Kotlyakov, V.M., 1995. The two-stepshape and timing of the last deglaciation in Antarctica. Climate Dynamics 11,151–161.

Kaiser, J., Lamy, F., Hebbeln, D., 2005. A 70-kyr sea surface temperature record of southernChile (Ocean Drilling Program Site 1233). Paleoceanography 20, 1–15.

Kaiser, J., Schefuß, E., Lamy, F., Mohtadi, M., Hebbeln, D., 2008. Glacial to Holocenechanges in sea surface temperature and coastal vegetation in north central Chile:high versus low latitude forcing. Quaternary Science Reviews 27, 2064–2075.

Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., Iredell, M., Saha,S., White, G., Woollen, J., Zhu, Y., Leetmaa, A., Reynolds, R., Chelliah, M., Ebisuzaki,W., Higgins, W., Janowiak, J., Mo, K.C., Ropelewski, C., Wang, J., Jenne, R., Joseph, D.,1996. The NCEP/NCAR 40-Year Reanalysis Project. Bulletin of the AmericanMeteorological Society 77, 437–471.

Kaplan, M.R., Ackert, R.P., Singer, B.S., Douglass, D.C., Kurz, M.D., 2004. Cosmogenicnuclide chronology of millennial-scale glacial advances during O-isotope stage 2in Patagonia. Geological Society of America Bulletin 116, 308–321.

Keil, R.G., Hu, F.S., Tsamakis, E.C., Hedges, J.I., 1994. Pollen in marine sediments as anindicator of oxidation of organic matter. Nature 369 (6482), 639–641.

Kissel, C., 2005. Magnetic signature of rapid climatic variations in glacial North Atlantic,a review. Comptes Rendus Geosciences 337, 908–918.

Kissel, C., the shipboard scientific party, 2007. In: Institut Paul- Emile Victor (Ed.),MD159-PACHIDERME-IMAGES XV, Cruise report, Les rapports de campagne à lamer, Réf. OCE/2007/01.

Lamy, F., Kaiser, J., 2009. Glacial to Holocene Paleoceanographic and ContinentalPaleoclimate Reconstructions Based on ODP Site 1233/GeoB 3313 Off SouthernChile. In: Vimeux, F., Sylvestre, F., Khodri, M. (Eds.), Past Climate Variability in SouthAmerica and Surrounding Regions. Springer Netherlands, Dordrecht, pp. 129–156.

Lamy, F., Hebbeln, D., Wefer, G., 1999. High-resolution marine record of climaticchange in mid-latitude Chile during the last 28,000 years based on terrigenoussediment parameters. Quaternary Research 51, 83–93.

Lamy, F., Kaiser, J., Ninnemann, U., Hebbeln, D., Arz, H.W., Stoner, J., 2004. Antarctictiming of surface water changes off Chile and Patagonian ice sheet response.Science 304, 1959–1962.

Lamy, F., Kilian, R., Arz, H.W., Francois, J.-P., Kaiser, J., Prange, M., Steinke, T., 2010.Holocene changes in the position and intensity of the southern westerly windbelt. Nature Geoscience 3, 695–699.

Lemieux-Dudon, B., Blayo, E., Petit, J.-R., Waelbroeck, C., Svensson, A., Ritz, C., Barnola,J.-M., Narcisi, B.M., Parrenin, F., 2010. Consistent dating for Antarctic and Green-land ice cores. Quaternary Science Reviews 29, 8–20.

Luebert, F., Pliscoff, P., 2004. Classification de pisios de vegetatacion y anlysis derepresentatividad ecologica de areas propuesta para la proteccion en la ecoregionvaldivia. In: Serie de publicationes WWF Chile Programa Ecoregion Valdivia(Ed.), (Santiago).

Lumley, S.H., Switsur, R., 1993. Late quaternary chronology of the Taitao Peninsula,southern Chile. Journal of Quaternary Science 8, 161–165.

Markgraf, V., d' Antoni, H.L., 1978. In: The University of Arizona Press (Ed.), Pollen Floraof Argentina (Tucson, Arizona).

Massaferro, J., Brooks, S.J., 2002. Response of chironomids to Late Quaternary environ-mental change in the Taitao Peninsula, southern Chile. Journal of Quaternary Science17, 101–111.

McCulloch, R.D., Davies, S.J., 2001. Late-glacial and Holocene palaeoenvironmentalchange in the central Strait of Magellan, southern Patagonia. Palaeogeography,Palaeoclimatology, Palaeoecology 173, 143–173.

McCulloch, R.D., Bentley, M.J., Purves, R.S., Hulton, N.R.J., Sugden, D.E., Clapperton, C.M.,2000. Climatic inferences from glacial and palaeoecological evidence at the lastglacial termination, southern South America. Journal of Quaternary Science 15,409–417.

Monnin, E., Indermühle, A., Dällenbach, A., Flückiger, J., Stauffer, B., Stocker, T.F.,Raynaud, D., Barnola, J.-M., 2001. Atmospheric CO2 concentrations over the LastGlacial termination. Science 291, 112–114.

Montade, V., Nebout, N.C., Kissel, C., Mulsow, S., 2011. Pollen distribution in marinesurface sediments from Chilean Patagonia. Marine Geology 282, 161–168.

Moore, D.M., 1983. In: Missouri Bot Garden (Ed.), Flora of Tierra del Fuego (USA).Moreno, P.I., 2004. Millennial-scale climate variability in northwest Patagonia over the

last 15 000 yr. Journal of Quaternary Science 19, 35–47.Moreno, P.I., Lowell, T.V., Jacobson Jr., George L., Denton, G.H., 1999. Abrupt vegetation

and climate changes during the Last Glacial Maximum and Last Termination in theChilean Lake District: a case study from Canal de la Puntilla (41°S). GeografiskaAnnaler A81, 285–311.

Moreno, P.I., Jacobson, G.L., Lowell, T.V., Denton, G.H., 2001. Interhemispheric climatelinks revealed by a late-glacial cooling episode in southern Chile. Nature 409,804–808.

Moreno, P.I., Kaplan, M.R., Francois, J.P., Villa-Martinez, R., Moy, C.M., Stern, C.R., Kubik,P.W., 2009. Renewed glacial activity during the Antarctic cold reversal and persis-tence of cold conditions until 11.5 ka in southwestern Patagonia. Geology 37,375–378.

Moreno, P.I., Francois, J.P., Moy, C.M., Villa-Martínez, R., 2010. Covariability of theSouthern Westerlies and atmospheric CO2 during the Holocene. Geology 38,727–730.

Naranjo, J.A., Stern, C.R., 1998. Holocene explosive activity of Hudson Volcano, southernAndes. Bulletin of Volcanology 59, 291–306.

Pizzaro, O., 2004. Editorial: Observing the Peru–Chile Current System.Putnam, A.E., Denton, G.H., Schaefer, J.M., Barrell, D.J.A., Andersen, B.G., Finkel, R.C.,

Schwartz, R., Doughty, A.M., Kaplan, M.R., Schluchter, C., 2010. Glacier advance insouthern middle-latitudes during the Antarctic Cold Reversal. Nature Geoscience3, 700–704.

Reimer, P.J., Baillie, M.G., Bard, E., Bayliss, A., Beck, J.W., Bertrand, C.J.H., Blackwell, P.G.,Buck, C., Burr, G., Cutler, K.B., Damon, P.E., Edwards, R.L., Fairbanks, R.G., Friedrich,M., Guilderson, T.P., Hogg, A.G., Hughen, K.A., Kromer, B., McCormac, G., Manning,S., Ramsey, C.B., Reimer, R.W., Remmele, S., Southon, J.R., Stuiver, M., Talamo, S.,Taylor, F.W., van der Plicht, J., Weyhenmeyer, C.E., 2004. IntCal04 terrestrial radio-carbon age calibration, 0–26 cal kyr BP. Radiocarbon 46, 1029–1058.

Sánchez Goñi, M.F., Cacho, I., Turon, J.-L., Guiot, J., Sierro, F., Peypouquet, J.-P., Grimalt,J., Shackleton, N., 2002. Synchroneity between marine and terrestrial responses tomillennial scale climatic variability during the last glacial period in the Mediterra-nean region. Climate Dynamics 19, 95–105.

Schimithüsen, J., 1956. Die räumliche Ordnung der chilenischen Vegetation. BonnerGeographische Abhandlungen 17, 1–86.

Siani, G., Colin, C., Michel, E., Carel, M., Richter, T., Kissel, C., Dewilde, F., 2010. Late Glacialto Holocene terrigenous sediment record in the Northern Patagonian margin:Paleoclimate implications. Palaeogeography, Palaeoclimatology, Palaeoecology 297,26–36.


Siani, G., Michel, E., De Pol-Holz, R., DeVries, T., Lamy, F., Carel, M., Lourantou, A., submittedfor publication. Southern Ocean deglacial reservoir age changes and timing of threeepisodes of upwelling. Nature Geosciences.

Silva, N., 2008. Physical and Chemical Characteristics of the Surface Sediments in theAustral Chilean Channels and Fjords. In: Silva, N., Palma, S. (Eds.), Progress in theOceanographic Knowledge of Chilean Interior Waters, from Puerto Montt to CapeHorn, pp. 69–75 (Valparaíso).

Stenni, B., Buiron, D., Frezzotti, M., Albani, S., Barbante, C., Bard, E., Barnola, J.M., Baroni,M., Baumgartner, M., Bonazza, M., Capron, E., Castellano, E., Chappellaz, J.,Delmonte, B., Falourd, S., Genoni, L., Iacumin, P., Jouzel, J., Kipfstuhl, S., Landais,A., Lemieux-Dudon, B., Maggi, V., Masson-Delmotte, V., Mazzola, C., Minster, B.,Montagnat, M., Mulvaney, R., Narcisi, B., Oerter, H., Parrenin, F., Petit, J.R., Ritz, C.,Scarchilli, C., Schilt, A., Schupbach, S., Schwander, J., Selmo, E., Severi, M., Stocker,T.F., Udisti, R., 2011. Expression of the bipolar see-saw in Antarctic climate recordsduring the last deglaciation. Nature Geoscience 4, 46–49.

Strub, P.T., Mesias, J.M., Montecio, V., Ruttlant, J., Salinas, S., 1998. Coastal ocean circu-lation off western South America.

Stuiver, M., Reimer, P.J., 1993. Extended 14C database and revised CALIB radiocarbonprogram. Radiocarbon 35, 215–230.

Stuiver, M., Quay, P.D., Ostlund, H.G., 1983. Abyssal water carbon-14 distribution andthe age of the world oceans. Science 219, 849–851.

Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B.,McCormac, G., Van Der Plicht, J., Spurk, M., 1998. INTCAL98 radiocarbon age calibra-tion, 24,000–0 cal BP. Radiocarbon 40, 1041–1083.

Stuut, J.-B.W., Kasten, S., Lamy, F., Hebbeln, D., 2007. Sources and modes of terrigenoussediment input to the Chilean continental slope. Quaternary International 161,67–76.

Taylor, R.E., Berger, R., 1967. Radiocarbon content of marine shells from the PacificCoasts of Central and South America. Science 158, 1180–1182.

Toggweiler, J.R., Russell, J.L., Carson, S.R., 2006. Midlatitude westerlies, atmosphericCO2, and climate change during the ice ages. Paleoceanography 21, 1–15.

Tschumi, T., Joos, F., Parekh, P., 2008. How important are Southern Hemisphere windchanges for low glacial carbon dioxide? A model study. Paleoceanography 23, 1–20.

Tsuchiya, M., Talley, L.D., 1998. A Pacific hydrographic section at 88°W: water-propertydistribution. Journal of Geophysical Research 12, 899–918.

Vandergoes, M.J., Dieffenbacher-Krall, A.C., Newnham, R.M., Denton, G.H., Blaauw, M.,2008. Cooling and changing seasonality in the Southern Alps, New Zealand duringthe Antarctic Cold Reversal. Quaternary Science Reviews 27, 589–601.

Veblen, T.T., Schlegel, F.M., Oltremari, J.V., 1983. Temperate Broad-leaved Evergreen Forestsof South America. In: Odvington, J.D. (Ed.), Temperate Broad-leaved Evergreen Forest,vol. 10. Elsevier, Amsterdam, pp. 5–31.

Villagrán, C., 1980. Vegetationsgeschichtliche und pflanzensoziologische Untersuchungenim Vicente Perez Rosales Nationalpark Chile. Dissertationes Botanicae 54, 1–165.

Villagrán, C., 1988a. Expansion of Magellanic Moorland during the late Pleistocene:palynological evidence from northern Isla de Chiloé, Chile. Quaternary Research30, 304–314.

Villagrán, C., 1988b. Late Quaternary vegetation of southern Isla Grande de Chiloé,Chile. Quaternary Research 29, 294–306.

Villagrán, C., 1990. Glacial climates and their effects on the history of the vegetation ofChile: a synthesis based on palynological evidence from Isla de Chiloé. Review ofPalaeobotany and Palynology 65, 17–24.

Documents

Vegetation and climate changes during the last 22,000yr from a marine core near Taitao Peninsula, southern Chile