at SciVerse ScienceDirect

Quaternary Science Reviews 53 (2012) 1e23

Contents lists available

Quaternary Science Reviews

journal homepage: www.elsevier .com/locate/quascirev

Invited review

A review of Glacial and Holocene paleoclimate records from southernmostPatagonia (49e55�S)

Rolf Kilian a,*, Frank Lamy b

aGeologie, FBVI, Universität Trier, Behringstr. 16, 54826 Trier, GermanybAlfred Wegener Institut für Polar- und Meeresforschung, Am Alten Hafen 26, 27568 Bremerhaven, Germany

a r t i c l e i n f o

Article history:Received 10 January 2012Received in revised form11 July 2012Accepted 18 July 2012Available online

Keywords:South AmericaPaleoclimatePalynologyGlaciologySpeleothemSouthern hemispheric westerliesSSTHoloceneGlacialQuaternary

* Corresponding author.E-mail address: kilian@uni-trier.de (R. Kilian).

0277-3791/$ e see front matter � 2012 Elsevier Ltd.http://dx.doi.org/10.1016/j.quascirev.2012.07.017

a b s t r a c t

Southern South America is the only landmass intersecting the southern westerly wind belt (SWW) thatinfluences the large-scale oceanography and controls for example the outgassing of CO2 in the SouthernOcean. Therefore, paleo-reconstructions from southernmost Patagonia are of global interest and anincreasing number of paleoclimate records have been published during the last decades. We provide anoverview on the different records mostly covering the Holocene but partly extending into the Late Glacialbased on a large variety of archives and proxies. We particularly discuss possible reasons for regionallydiverging palaeoclimatic interpretations and summarize potential climate forcing mechanisms. TheDeglacial and Holocene temperature evolution of the region including the adjacent Pacific Ocean indi-cates “Antarctic” pattern and timing consistent with glacier re-advances during the Antarctic ColdReversal. Some records indicate a significant accumulation control on the glacier fluctuations related tochanges in SWW strength and/or position. Reconstructions of Holocene changes in the SWW behaviourprovide partly inconsistent and controversially discussed pattern. While records from the hyperhumidside point to a stronger or southward displaced SWW core during the Early Holocene thermal maximum,records from the lee-side of the Andes show either no long term trend or the opposite, suggestingenhanced westerlies during the late Holocene “Neoglacial”. Likewise, centennial-scale global or hemi-spheric cold intervals, such as the Little Ice Age, have been interpreted in terms of enhanced and reducedSWW strength. Some SWW variations can be linked to changes in the El Niño-Southern Oscillation(ENSO) consistent with instrumental climate data-sets and might be ultimately forced by solar vari-ability. Resolving these inconsistencies in southernmost Patagonian SWW records is a prerequisite forimproving hemispheric comparisons and links to atmospheric CO2 changes.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

The southernmost tip of South America is of particular interestfor paleoclimate reconstruction, since it is the only land-massintersecting the core of the southern westerly wind belt (SWW)at latitudes between 49 and 53�S (Fig. 1). On a hemispheric scale,SWW changes substantially contribute to the forcing of the deepand vigorous Antarctic Circumpolar Current (ACC). Wind-inducedupwelling within the ACC in the Southern Ocean raises largeamounts of deep water to the ocean’s surface in this circumpolarbelt affecting the global thermohaline circulation (e.g. Marshall andSpeer, 2012) and atmospheric CO2 contents (e.g. Toggweiler et al.,2006). Therefore, the SWW exerts a strong control on globalclimate and oceanography.

All rights reserved.

The first paleoclimate records from Patagonia were based onglacier advance reconstructions (Caldenius, 1932; Mercer, 1965)and on pollen records from soil and peat records (Auer, 1933, 1958,1960, 1974; Heusser, 1971). Multi-proxy reconstructions based onlake and fjord sediment cores as well as stalagmites initiated onlyduring the past 10 years and constitute now a growing number ofpartially high resolution records. The number of paleoclimate-related publications particularly increased since ca 2005 (to morethan 25 publications/year; Fig. 2). Despite this large number ofpublications and the global implications of the Patagonian paleo-climate, reconstructions of Glacial and Holocene SSW changes arepartly inconsistent and discussed controversially. Contrastinginferences have been derived from proxy records in southernPatagonia for example on multi-millennial time-scales during theHolocene (e.g. Mayr et al., 2007a,b; Lamy et al., 2010; Moreno et al.,2010; Waldmann et al., 2010; Fletcher and Moreno, 2011) but alsoregarding shorter term climate variations on centennial time-

Fig. 1. A) Southernmost South America with average annual precipitation New et al., 2002) and the modern annual mean sea surface temperature distribution of the surroundingoceans (data from the NOAA-CIRES Climate Diagnostics Center http://www.cdc.noaa.gov/index.html). The Antarctic Circumpolar Current (ACC) and Cape Horn Current (CH) areindicated. B) December to February zonal wind distribution over the Southern Hemisphere based on NCEP/NCAR reanalysis data (Kalnay et al., 1996) and C) correlation between850 hPa zonal wind and precipitation (Garreaud et al., in press).

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e232

scales, e.g. during the globally known Little Ice Age (LIA; e.g. Moyet al., 2008; Schimpf et al., 2011).

Besides SWW reconstructions which are primarily based onprecipitation proxies and their correlation to wind changes, pale-otemperature reconstructions in southernmost Patagonia may helpto understand atmospheric and ocean temperature behaviour inthe Southern Hemisphere. These are for example essential forevaluating coupled atmospheric-ocean circulation changes as wellas timing and dynamics of interhemispheric climate changesincluding the bipolar seesaw (e.g. Stocker and Johnsen, 2003; Lamyet al., 2007; Barker et al., 2009). Furthermore, the southern Pata-gonian Ice field (PIF) is of special interest, because it constitutes the

Fig. 2. Number of published paleoclimate studies in southernmost Patagonia over thepast decades.

largest continental ice-sheet outside the polar regions (Casassaet al., 2000) which is highly sensitive to temperature as well asprecipitation changes (Warren and Sudgen, 1993) and contributessignificantly to global sea level changes (Rignot et al., 2003; Glasseret al., 2011).

As with the paleoclimate data, also modelling studies on theposition and strength of the SWW that have been performed inparticular for the Last Glacial Maximum (LGM) are not yetconclusive (Rojas et al., 2009). Holocene changes of the SWW havealso been addressed by for example mid-Holocene and preindus-trial simulations (Wagner et al., 2007; Rojas and Moreno, 2011) andmodelling studies that focus on the impact of solar variabilityduring the past 3000 years (Varma et al., 2011).

Instrumental climate and weather data collected over the pastca 50 years show that SSW extends more than 3000 kmnorthesouth with significant latitudinal variations on seasonal todecadal times-scales (Fig. 1A and B). Due to the fewweather stationdata in the Southern Hemisphere, it is likely that the SWW vari-ability is spatially more complex than presently known (Garreaud,2007; Garreaud et al., in press). Regarding paleo-SWW recon-structions, it is therefore important to note that past changes of theSWW cannot be estimated from a record of a single site. Furthercomplicating is the west-east distribution of sites across the Pata-gonian Andes, where the SSW forces in one of themost pronouncedclimate divide on earth (Fig. 1C). This strong precipitation gradientproduces different, highly sensitive ecosystems at the hyperhumidwestern and evaporation controlled arid eastern side of the Andes.Paleoclimate archives from such contrasting environments oftenrequire different and not directly comparable proxies. Therefore,local climate characteristics are very important in southern Pata-gonia as well as proxy calibration and proxy monitoring which hasonly started within the past few years (e.g. Schimpf et al., 2011). Afurther complication of paleoclimate reconstructions from Pata-gonia is the fact that some proxy records may have been affected by

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 3

Glacial to Holocene coastline changes related to a regionallydistinct isostatic, eustatic, and neotectonic development of theregion. These factors result for example in a regionally differenttiming of marine transgression into proglacial lakes and lowlandaround the southern tip of South America (e.g. Anderson andArcher, 1999; Kilian et al., 2007a).

This paper provides an overview on Glacial and Holocenepalaeoclimatic reconstructions based on a large variety of marineand terrestrial archives as well as diverse proxies. We focus ona compilation of records from both sides of the Andean climatedivide and discuss possible reasons for partly contrasting palae-oclimatic interpretations primarily regarding past changes of theSWW. We conclude with hemispheric and global implications aswell as potential climate forcing mechanisms.

2. Characteristics of the present day climate and vegetation

The present-day latitudinal distribution of the SWW is moni-tored by abnormal high precipitation along the west coast of SouthAmerica and in the Andes between latitudes ca 30�S and >55�S(Fig. 1A). During the last century, the annual variations of the SWWstrength and related precipitation have been recorded at some far-off located regional weather stations (Schneider et al., 2003;Garreaud, 2007; Rasmussen et al., 2007; Carrasco et al., 2008;Aravena and Luckman, 2009), but only few of these stations aresituated near to the Andean climate divide.

World-wide weather station data and numerical weatherprediction model output data since 1948 have been used toconstruct an interpolated global climate data-set (NCEP/NCARreanalysis data: Kalnay et al., 1996; Fig. 1B and C). In summer, theNCEP/NCAR data show the highest wind velocities across the SouthAmerican continent between 49 and 53�S. This marks the core ofthe SWW (Fig. 1B) characterised also by a precipitation maximumat the west coast and the Andes during summer (Schneider et al.,2003). During winter, the SWW is broader and slightly displacednorthward, while the core section is relatively weaker. Interpolatedweather station data indicate a clear positive correlation betweenprecipitation and SSW strength at the continental margin and inthe Andes (r ¼ 0.4 to 0.8; Fig. 1C; Garreaud et al., in press). To theeast of the Andes this correlation soon becomes weaker (r ¼ 0.4 to0.2) and negatively at the eastern steppe region towards theAtlantic coast. In the lee of the Andes and in the Patagonian steppethe dry winds produce high evaporation and negative waterbalance, in particular during summer (Endlicher, 1991). Thereforemany lagoons dry out in summer and regional run-off is veryrestricted in glacier-free eastern areas between 50� and 55 �C.

The NCEP/NCAR data also show the relationship betweenchanges of Intertropical Convergence Zone (ITCZ), El Nino SouthernOscillation (ENSO) and the SWW. During El Niño events, whichtypically occur between December and March, the ITCZ is generallyweaker and/or southward displaced while the core section of SWWis reduced and westerly winds are enhanced north of ca 45�S(Schneider and Gies, 2004; Garreaud, 2007; Boucher et al., 2011).

The overall air temperature of southernmost South America ispredominately controlled by the SWW which in turn largelyreflects SST’s within the Antarctic Circumpolar Current (ACC).Occasionally, an atmospheric blocking of the SWW over Patagonia(<20% of weather situations; Schneider et al., 2003; Garreaud et al.,in press) enables moisture-rich easterly winds from the SouthAtlantic to migrate over the Patagonian Steppe until the easternrange of the Andes. Furthermore, northward advection of coldAntarctic air masses can reach the southern tip of the continent andsometimes could move further north reaching the subtropics/topics. In general such weather situations are more likely duringwinter. In the north eastern sector of Patagonia (north of 50�S) the

climate becomes more continental characterized by an increasedseasonality in temperatures of up to 20 �C.

The Cape Horn Current branches from the northern ACC ataround 45�S southward and transports relatively warmer watermasses along the Pacific coast toward the Drake Passage (Fig. 1A;Chaigneau and Pizarro, 2005). At 52�S, the open marine coastalannual SST’s (ca 8.5 �C) are ca 2 �C warmer than air temperaturesmeasured at the coastal weather station Evangelistas (Fig. 4). Thisstation shows an average annual temperature of 6.5 �C similar toother weather stations across the continent at this latitude (e.g.Punta Arenas; Schneider et al., 2003; Garreaud et al., in press),indicating that the air temperatures are not significantly increasedby the abnormal warm Cape Horn Current. However, the relativelywarm coastal water enters the fjord system as bottomwater belowthe superficial freshwater which is comparatively cold duringsummer due to snow and glacier melting. This produces a strongthermohaline layering in the fjords (Kilian et al., 2007a).

Distinct vegetational pattern occur in southernmost SouthAmerica denoting the extremely different moisture conditions overthe region. In the Andes and the western fjords, the extremely highprecipitation (>4000 mm annual precipitation) and comparativelylow evaporation causes stagnant water in most soils, in particularduring summer. These conditions limit the plant developmentwhich is characterised by the occurrence of species-poor plantcommunities (e.g. Pisano, 1977; Moore, 1979, 1983; Kleinebeckeret al., 2007). Climatic and non-climatic factors controlling thecomposition and distribution of plant communities within terres-trial ecosystems are poorly investigated. Steep slopes of fjords andregions with tectonic fracture zones, both with very good run-offconditions as well as sediment bars and moraines with coarseclastic sediment cover, and efficient drainage conditions enablelocally better conditions for forest growth. These observations alsoindicate that the relative quantity of trees versus peat bog speciesand/or hygrophytes could reflect besides climatic factors (precipi-tation and temperature) also changes of regional coast lines whichare controlled by eustasy, isostasy and tectonics. Dendroecologicalstudies indicate a significant ecesis after the glacier retreat (Kochand Kilian, 2005), highlighting the role of local conditions (e.g.nutrient deficit) and/or ecological features (e.g. dispersionsyndromes) in the plant colonization and succession dynamics.Furthermore, disturbance events such as earthquake-related massflows (Waldmann et al., 2011) or volcanic ash deposition canproduce important vegetational perturbances (Kilian et al., 2006).

To the east of the climate divide (leeward side of the Andes),a strong reduction in soil moisture is induced by a substantialdecrease in annual precipitation (<500mm/yr) and higher summertemperatures (>12 �C) in combination with strong winds whichcause enhanced evaporation (Fig. 3). The result is an open andtreeless landscape known as the Patagonian steppe characterisedby summer dry stress for most plants (e.g. Endlicher and Santana,1988; Endlicher, 1991).

Between these dissimilar environments, the hyper-humid in thewest and semi-desert in the east, a Sub Antarctic deciduous forestgrows in a transitional area which is characterized by annualprecipitation of 500e1000 mm/yr (Heusser, 1995). This forest ispartly strongly affected by mechanical wind-stress (Armesto et al.,1992), in particular when more open parkland was developed. Themoderate climate conditions enable the development of morespecies-rich vegetal communities (Heusser, 1995). Charcoal recordsindicate that fires shaped this vegetation sequence during theHolocene (Heusser, 1995; Huber et al., 2004; Whitlock et al., 2007;Markgraf and Huber, 2010). However it remains highly disputedhow far the fires are of natural or anthropogenic origin. A clearanthropogenic impact was produced by the activity of farmersduring the last 130 years.

Fig. 3. WeE Transect across the Andes at 52�S with annual precipitation, evaporation,and annual wind velocities as well as the correlation between wind velocities andprecipitation (compiled from data of Schneider et al., 2003; Aravena and Luckman,2009; Garreaud et al., in press).

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e234

3. Regional distribution of archives and used proxies

The locations of investigated sites and possible future archivesdepend on the natural distribution of glaciers, ice fields, fjords,lakes, peat land and forestwhich is predominately controlled by themorphology and the strong precipitation gradient in southernmostPatagonia (Figs. 3 and 4). An additional more practical factor is theaccessibility of sites by roads which is only possible on the easternside of the Andes and in the steppe region. The hyperhumidwestern area with its steep fjord and mountain belt as well as theextended westernmost island zone is only accessible by ship.

Fig. 4. Regional map of southernmost Patagonia with site locations and

Terrestrial explorations in this totally unpopulated area are muchmore difficult and require an extended logistic support. Thesegeographic differences explain, why 85% of published paleoclimatestudies are from regions to the east of the Andean climate divide.Major sites discussed in the text are listed in Table 1 and theirlocation is shown in Fig. 4.

Mapping of Glacial and Holocene glacier extent also primarilyconcerns the eastside of the Andes: the Cordillera Darwin(Kuylenstierna et al., 1996), Central Strait of Magellan (e.g.Clapperton et al., 1995; McCulloch et al., 2005; Kaplan et al., 2008;Sugden et al., 2009), the glacial lake areawithin the Torres del Painearea (e.g. Moreno et al., 2009b) as well as the Lago Argentino andLago Viedma areas (e.g. Wenzens, 2005; Ackert et al., 2008; Kaplanet al., 2011; Strelin et al., 2012). Only few investigations documentglacier fluctuations from the central and western range of theAndes. These include the Isla Santa Ines (Aravena, 2007) and GranCampo Nevado area (Koch and Kilian, 2005; Schneider et al., 2007;Möller and Schneider, 2010). In the western area, many morainesare subaquatic and therefore have only been locally explored bymultibeam and echo sounding (Kilian et al., 2007b; Breuer et al., inpress-a).

The Northern and Southern Patagonian Ice Fields (PIF) mayprovide suitable sites for ice drilling. However, only few areas ofthese ice fields are elevated enough to prevent summer melting ofthe ice/snow surface related firn formation which complicates thepalaeoclimatic analyses. Frequent and extended eolian redistribu-tion of snow further hinders palaeoclimatic exploration of ice-cores. Even if there are some areas appropriate areas with low iceflow rates, the extremely high snow accumulation across the ice

numbering (used in the text and Table 1) and regional geography.

Table 1Locations of paleoclimate archives and their stratigraphic range as well as regional weather stations.

Core Location Lat. South Long. West Core length [m] Water depth [m] Stratigr range [ka] References

Marine sediment recordsMD07-3124 Concepción F. 50�310 74�580 22.2 564 0e12.0 Kissel et al. (2007)MD07-3128 Pacific coast 52�400 75�34 33.0 1030 0e60.0 Caniupán et al. (2011)MD07-3132 Whiteside Channel 53�440 70�190 21.7 301 0e15.0 Kissel et al. (2007)Vo1 Estero Vogel 52�490 72�330 4.6 25 0e13.3 Kilian et al. (2007b)Es1 Seno Skyring 52�340 72�160 4.1 78 0e>5.0 Baeza (2005)Sk1 Seno Skyring 52�370 71�420 4.8 62 0e18.0 Kilian et al. (2007b)BA-1 Bahia Arevalo 52�420 73�240 9.9 5 0e>6.0 Breuer et al. (in press-b)LO-1 Bahia Lobo 52�450 73�160 8.2 83 0e0.8 Breuer et al. (in press-b)LU-1 Quenca Luca 52�460 73�190 1.6 320 0e1.2 Breuer et al. (in press-b)Sg-1 Seno Glaciar 52�470 73�230 0.6 520 0e0.6 Breuer et al. (in press-b)PALM-2 Isla Parlamento 52�470 73�390 8.9 44 0e16.0 Lamy et al. (2010)TM1 Cabo Tamar 52�540 73�470 4.8 31 0e15.0 Lamy et al. (2010)JPC Marinelli fjord 54�250 69�300 13.5 30e200 0e16.0 Boyd et al. (2008)Lake sediment coresART-1 Arthuro 53�300 72�560 0.42 42 0e2.1 Breuer et al. (in press-a)CH-1 Chandler 52�490 72�540 6.5 16 0e12.5 Kilian et al. (2006)TML-1 Tamar 52�540 73�480 7.6 20 0e15.0 Lamy et al. (2010)LMP-I Muy Profundo 52�440 73�130 2.0 205 0e4.2 Breuer et al. (in press-a)HB-KL I-III Hambre 53�360 70�57 12.2 17 0e17.3 Hermanns and Biester (2011)P8 Guanaco 50�520 72�520 6.0 16 0e16.0 Moreno et al. (2010)Various cores Potrok Aike 51�580 70�220 <122 76 0e52.0 Recasens et al. (2011)CAK 99 (various) Cardiel 48�550 71�150 15 76 0e15.0 Gilli et al. (2005)LF06 (various) Fagnano 54�350 68�000 8.0 200 0e12.0 Waldmann et al. (2011)Peat recordsP14 Puerto Eden 49�080 74�250 2.5 0e10.0 Ashworth et al. (1991)P13 Cerro Frias 50�260 72�430 0e9.0 Mancini et al. (2005)P12 Lago Argentino 50�350 72�550 3.0 0e13.5 Wille and Schäbitz (2009)P11 Meseta La Torre 1-2 50�310 72�030 0e8.5 Schäbitz (1991)P10 Torres del Paine 50�590 72�400 8.6 0e13.0 Heusser (1995)P9 Nandu Vega 50�560 72�460 0e12.5 Villa-Martínez and Moreno (2007)P7 Rio Rubens 52�040 71�310 7.0 0e17.0 Huber et al. (2004)P6 Punta Arenas 53�090 70�570 7.5 0e16.8 Heusser (1995)P5 Estancia Esmeralda 53�300 70�350 3.0 0e17.0 McCulloch and Davis (2001)P4 Puerto del Hambre 53�360 70�550 8.5 0e17.3 McCulloch and Davis (2001)P3 Passo Garibaldi 54�53 66�100 2.8 0e13.0 Huber et al. (2004)P2 Haberton 54�540 67�10 10.2 0e17.0 Markgraf (1993)P1 Isla de los Estados 54�500 64�400 3.5 0e15.0 Ponce et al. (2011)GC1 Gran Campo Nevado 52�470 72�570 1.7 0e2.5 Biester et al. (2002)GC2 Gran Campo Nevado 52�480 72�560 2.7 0e14.3 Lamy et al. (2010)Sky1 Skyring 52�310 72�080 2.8 0e6.5 Biester et al. (2003)Pbr2 Puerto del Hambre 53�380 70�580 6.5 0e15.0 Biester et al. (2003)

Cave and Stalagmites Location Lat. South Long. West Core length [m] Altitude [m a.s.l.] Stratigr range [ka] References

MA Cave (MA1) Bahia Arevalo 52�41’7 73�2303 0.3 20 0e4.5 Schimpf et al. (2011)GC Cave (GC 1) Seno Glaciar 52�48’1 73�16’3 0.3 35 2.4e8.0

Automatic Weather Station Location Lat. South Long. West Annual range of precip [m/yr] Altitude [m a.s.l.] Remarks References

Evangelistas Pacific coast 52�240 75�060 1.5e5.5 30 Since AD 1990 Aravena and Luckman (2009)Felix Str. of Magellan 52�570 74�040 3.2e7.0 15 Since AD 1910Skyring Seno Skyring 52.570 71�510 0.6e1.1 8 Since AD 2001 Schneider et al. (2003)Passo Northern GCN 52�450 73�010 7.2e9.8 380 Since AD 2000 Schneider et al. (2003)Bahamondes Ba. Bahamondes 52�480 72�560 3.3e7.1 26 Since AD 1999 Schneider et al. (2003)Arevalo Bahia Arevalo 52�410 73�160 4.8 90 Since AD 2007Punta Arenas Punta Arenas 53�080 70�530 0.2e0.8 30 Since AD 1888 Aravena and Luckman (2009)

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 5

fields (reaching >10,000 mm water equivalent per year (Rivera,2004), strongly limits the possible length of ice records. So-faronly at the 4100 m high Mt. San Valentin within the Northern PIFup to 122m long ice cores have been drilled successfully. This site isunaffected by summer melting and the record includes at least thelast 150 years up to 85 m core depth. Still undated older ice of thelowermost core represents an earlier Holocene snow accumulation(Vimeux et al., 2008, 2009).

In SE Patagonia, lake sediment records have been obtained fromnumerous sites including Lago Fagnano (e.g. Waldmann et al., 2010;Moy et al., 2011), Lago Hambre (e.g. Hermanns and Biester, 2011;Breuer et al., in press-a), Lago Guanaco (Moreno et al., 2010), LagunaPotrok Aike (e.g. Mayr et al., 2007a; Haberzettl et al., 2009) andLago Cardiel (e.g. Stine and Stine, 1990; Gilli et al., 2005; Fig. 4). Asclose to the Andes all lakes haven been covered by glaciers,

paleoclimate records only start after the ice retreat between 17 and14 ka BP (e.g. Lamy et al., 2010). Exceptions are records from thenon-glaciated Laguna Potrok Aike that reach back to ca 50 ka BP(e.g. Recasens et al., 2011; Lisé-Pronovost et al., 2012). On thewestern side of the Andes at around 52�300S Late Glacial to Holo-cene records have been published from Lake Tamar, Lake Arthuro,Lake Muy Muy and Lake Chandler (Fig. 4; e.g. Kilian et al., 2007b;Lamy et al., 2010; Breuer et al., in press-b). Lake sediment studiesare mostly based on multiple palaeoclimatic proxies, typicallyincluding pollen, and geochemistry and isotopy of organic materialas well as granulometry and geochemistry of siliciclastic sedimentcomponents. XRF core scanner records of some lake sediment coresreach up to yearly-resolution (e.g. Lake Fagnano and Tamar).

Late Glacial and Holocene peat cores have been obtainedfrom>17 sites at both sides of the Andes (Sites in Fig. 4 and Table 1;

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e236

e.g. Isla de los Estados, Haberton, Garibaldi, Puerto del Hambre,Punta Arenas, Rio Rubens; Gran Campo Nevado). The records oftenstart during the Late Glacial with a minerogenic peat compositionwhich partly evolved to exclusively ombrogenic composition in theHolocene (only aerial deposition above ground water level;McCulloch and Davis, 2001; Biester et al., 2003). They have beeninvestigated in particular with respect to pollen and microscopiccharcoal abundances (e.g. Pendall et al., 2001 and Chapter 6.2.). Thecomplex regional distribution of vegetation communities andresulting modern pollen assemblages is not yet fully understood insouthern Patagonia (see 2; Heusser, 1995). Therefore, palae-oclimatic inferences from past pollen records are often complicatedby local conditions at different sites (e.g. soil-types, drainage, slope,facing) and plant ecology (e.g. dispersal, disturbance history)(McCulloch and Davis, 2001; Markgraf et al., 2003; Huber et al.,2004). In addition, climate implications based on peat decompo-sition rates, anthropogenic Hg and Pb accumulation (Biester et al.,2002, 2003), sea spray-related halogenide and heavy metalcontents (Biester et al., 2006) as well as dust deposition (Sapkotaet al., 2007) have been investigated in peat cores.

27 tree ring sites have been studied in southern Patagonia(Fig. 4; Villalba et al., 2009), including a relatively large number ofsites in the forested area around the Beagle Channel (e.g. Villalbaet al., 2003). From these and further records located close PuntaArenas, in Torres del Paine area, and at Lago Argentino, regionalpaleotemperature reconstructions covering the past severalhundred years have been recently compiled by Neukom et al.(2010a). From the hyperhumid western side, only few tree-ringrecords of the cypress Pilgerodendron Uviferum have been pre-sented from the Gran Campo area (Koch and Kilian, 2002), IslaNavarino (Aravena et al., 2002) and near to the Pio XI Glacier to thewest of the southern PIF (Rigozo et al., 2007). However, it remainsstill unclear, how far temperature or frequent stagnant water“water stress” controls the tree ring growth in this hyperhumidarea.

Several open marine and fjord sediment cores have been drilledbetween 1989 and 1995 by RV Polar Duke and the Italian RV O.G.S.Explora. Anderson and Archer (1999) investigated sediment coresfrom Bahia Inutil in Tierra del Fuego and the White Side Channel(Fig. 4) which give implications for marine transgression, but onlylimited paleoclimate information. To our knowledge, only few coresfrom the RV O.G.S. Explora cruises have been dated and paleo-climatologically investigated in more detail (e.g. Brambati, 2000;Marinoni et al., 2008). In the marine records the Uk37-index ofalkenones formed by cocolithophorides provides calibratedtemperatures (e.g. Prahl and Wakeham, 1987; Prahl et al., 1988)which typically reflect annual SSTs but may contain rather a springor summer SST signal at higher latitudes as alkenone producingalgae do not grow year-round (e.g. Prahl et al., 2010).

More recently, during 15 cruises of RV Gran Campo II since 2002,sediment basins of the western fjord zone between latitudes 49 to54�S and former proglacial lakes such Seno Skyring and SenoOtway have been explored systematically by echo sounding andmultibeam (e.g. Breuer et al., in press-b) and numerous sedimentcores have been recovered in the western fjord region (Fig. 4).These cores have been investigated applying diverse terrigenous(e.g. accumulation of terrestrial organic carbon and siliciclasticmaterial) and marine proxies (e.g. aquatic organic carbon, biogeniccarbonate, and opal) and provide records of continental paleo-climate and marine paleoproductivity changes.

During the IMAGES XV-MD159 Pachiderme cruise of RV MarionDufresne in 2007, several long “Calypso” sediment cores have beentaken from the fjords and Pacific continental slope between 49 and53�S (Kissel et al., 2007; unpublished cruise report). Among thosethe open ocean core MD07-3128 provides the first regional SST and

IRD-related ice extent reconstruction back to 60 ka BP (Caniupánet al., 2011).

4. Age control

Age control largely depends on the palaeoclimatic archive andwe here only shortly review the major methods applied to Pata-gonian records. The stratigraphic background of most archives isbased on 14C ages which have been typically calibrated to calendarages in publications of the last decade. However, only some recordsconsider the Southern Hemisphere calibration curve of McCormacet al. (2004) which gives 50e80 years younger Holocene ages. Inparticular lake sediments often include leaves and macro plantremnants for reliable radiocarbon dating. Moy et al. (2011)demonstrate that 14C ages of bulk organic material may be 4to >6 ka older than coevally deposited macro plant remnants andpollen concentrates. In sediment core ES1 located in Seno Skyring,14C of bulk organic material is around 8 ka older than a well datedtephra layer (Baeza, 2005). Problems could also occur, if dated plantremnants have been deposited by terrestrial mass flows whichoften include older plant material from eroded peaty soils. In peatcores, root activity has often modified the structure and layering oforganic material which alter the depth-relationship of 14C ages ofmacro remnants.

14C-based stratigraphies of marine and in particular fjord sedi-ment cores have to deal with variable marine reservoir ages in therange of 200e800 years. Shallow and freshwater-rich coastal sitesexhibit low reservoir effects of 200e300 years (Kilian et al., 2007a).For the Puerto Natales region, a mollusc shell collected AD 1939(pre-bomb) gave a reservoir age of 530 years (Ingram and Southon,1996), but no exact information have been published for its origin.Deeper sediment core sites (>300 m water depth) include signifi-cantly “older” Pacific water which may cause higher reservoir agesfrom 550 to 800 years (Caniupán et al., 2011). Since the degree ofwind-induced mixing of superficial freshwater with older fjordbottomwater may have changed during the Holocene, we supposethat there are also still unknown long-term changes in the reservoireffect.

Only some piston sediment cores include dating of the upper-most sediment section by the 210Pb method (e.g. Appleby andOldfield, 1992) in order to prove a sub-recent sediment surfacewhich was also applied successfully for dating of the superficialpart of peat cores (Biester et al., 2002).

Tephrochronology represents an important stratigraphic tool forthe Magellan region, since Late Glacial and Holocene tephra layersare relatively well dated here. Best known are eruptions from theReclus volcano atw15.0 ka BP, Mt. Burney atw9.0 ka BP, Hudson atw7.2 ka BP, Mt. Burney at w4.2 ka BP, Aguilera at w3.0 ka BP andMt. Burney at w2.0 ka BP (Stern, 1990, 1992, 2008; McCulloch andDavis, 2001; Kilian et al., 2003; McCulloch et al., 2005; Moy et al.,2008). Volcanic glasses of the different volcanoes can be clearlydistinguished due to their characteristic K and Ti contents (Sternand Kilian, 1996; Stern, 2008). Due to anoxic conditions in thesediment, most lakes have well preserved even mm-thick tephrafall deposits (Kilian et al., 2003). In many peat cores thin tephralayers are often smeared by the root activity which makes theiridentification difficult. Some sediment records exhibit dissemi-nated tephra or thin tephra layers after stronger erosion events,which could be reworked tephra. Tephra particles originating fromthe 4.2 and 2.0 ka BP Mt. Burney eruptions have also been foundwithin stalagmite layers which grew within an open cave. Thesehave been used as independent control for the detritus correctionof Th/U ages (Schimpf et al., 2011).

Several Patagonian moraines have been dated during the lastdecade by using cosmogenic nuclide ages which improves the

Fig. 5. Glacier extent in southern Patagonia compared to SST and Antarctic tempera-ture changes during the last Glacial. A: Glacial stages in the Magellan Strait region(Sugden et al., 2009). B: Non-sea-salt calcium from EPICA Dome C (EDC) (a proxy fordust content changes in Antarctic ice cores (Fischer et al., 2007). C: Percentage of>150 mm carbonate-free sediment fraction as a proxy for IRD in core MD07-3128 offthe Pacific entrance of the Magellan Strait (Caniupán et al., 2011); D: Timing of glacialice-sheet variations in the Magellan Strait region based on 10Be ages (ice retreattowards the lower left; Kaplan et al., 2008). E: Alkenone SST record of core MD07-3128(Caniupán et al., 2011); F: Alkenone SST record of ODP Site 1233 at the Pacific conti-nental slope at 41�S (Kaiser and Lamy, 2010). G: Antarctic surface temperature changes(deviation from mean of the last millenium; Jouzel et al., 2007) plotted on the newLemieux-Dundon time-scale (Lemieux-Dudon et al., 2010). Grey bars indicatesynchronous positive temperature excursions.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 7

chronologies of glacier fluctuations, in particular for the LateGlacial. However, most recent comparisons between 14C andcosmogenic nuclide constrains for moraine ages suggest anunderestimation of the radionuclide ages by 1e2 ka related tohigher Late Glacial production rates (Putnam et al., 2010; Kaplanet al., 2011).

Finally, a first study applying high resolution paleomagneticdating methods have been applied to Holocene and Glacial sedi-ment sequences at Laguna Potrok Aike (Lisé-Pronovost et al., 2012).This study suggests a large potential of paleomagnetic datingapplicable in both terrestrial and marine archives.

5. Glacial records

5.1. Full glacial (60e20 ka BP)

Paleoclimate records reaching back to the last full glacial periodin southern Patagonia are scarce because of the coverage of vastareas by the Patagonian ice-sheet. One example is Laguna PotrokAike where up to 122 m long lake sediment cores reach back to ca50 ka BP (Recasens et al., 2011). A preliminary pollen record fromthis site suggests significantly colder glacial temperatures andprobably reduced humidity. More detailed paleoenvironmentalstudies based on the glacial records from Laguna Potrok Aike arecurrently in progress.

A second example includes marine sediments from the Pacificcontinental margin. The only published records from this area arebased on sediment coreMD07-3128 recovered at 53�S off the Pacificentrance of the Strait of Magellan (Caniupán et al., 2011). This well-dated core reaches back to ca 60 ka BP (Fig. 5) and provides detailedinformation on surface water changes and ice-rafted debris (IRD)deposition. Alkenone-derived sea surface temperatures (SST) wereup to 8 �C lower during the full glacial compared to the earlyHolocene and reveal substantial millennial-scale fluctuations thatare partly similar to those observed in SSTchanges at ODP Site 1233at the Pacific margin further north (41�S) and temperature recon-structions from Antarctic ice cores (Fig. 5E). The strong glacialcooling at site MD07-3128 implies a substantial northward expan-sion of polar water masses and the Southern Ocean fronts with thesub-Antarctic front probably located close to the site during the fullglacial. An interesting feature of the SST record is a long-termwarming trend of þ2 �C from ca 50 to 25 ka BP and a successive3 �C cooling trend culminating at ca 19 ka BP when the coldest SSTwere recorded. These trends have been related to variable supply ofcold melt water from the adjacent Patagonian ice-sheet (Caniupánet al., 2011). A more direct proxy for changes in the extent of thePatagonian ice-sheet comes from the IRD (Fig. 5B). This recordshows pronounced IRD pulses between ca 30 and 18 ka BP thatpartly overlap with glacial advances reconstructed on the easternside of the ice-sheet in southern Patagonia (Kaplan et al., 2008;Sugden et al., 2009) and further north at the Pacific margin of thenorthern ice-sheet (Kaiser and Lamy, 2010). Patagonian ice-sheetadvances have been mechanistically linked to dust maxima recor-ded in Antarctic ice-cores (Caniupán et al., 2011; Kaiser and Lamy,2010; Sugden et al., 2009; Fig. 5A) since the fine-grained glacialdetritus from the eastern Patagonian outwash plains was detectedas themajor dust source area for Antarctica. Interestingly, terrestrialreconstructions indicate a maximum ice-extent in southern Pata-gonia at around 25 ka BP and a slightly retreating ice extent alreadybefore the global LGM (Fig. 5; Kaplan et al., 2008). This regionalretreat was explained by reduced precipitation during the coldestglacial interval due to northward displaced westerlies. A weakersouthern SWW margin would have enabled colder superficial meltwater tomigratewestward. This could explain the observed coolingtrend from25 to 19 Ka BP at theMD07-3128. Such an accumulation-

driven early glacier retreat is consistent with ice-sheet modellingresults that suggest reduced precipitation in southern Patagoniaduring the Last Glacial Maximum (Hulton et al., 2002).

5.2. Deglaciation (20e10 ka BP)

Along the central Strait of Magellan rapid glacier retreat startedafter the last major set of moraines were built prior to 17.5 ka BP(Moraine stage D: McCulloch and Bentley, 1998; McCulloch et al.,2005; Kaplan et al., 2008; Sugden et al., 2009) leaving the Puertodel Hambre site around 80 km southward ice-free by 17.3 ka BP(McCulloch and Davis, 2001). Ice retreat in the Seno Skyring regionoccurred largely in phase (Fig. 6). However, there are indications ofa slightly earlier response in this area before 18 ka BP, most likelybecause the Seno Skyring glacier catchment was more sensitive toa small initial elevation increase of the equilibrium-line altitude(Kilian et al., 2007b). Deglacial warming in the Southeast Pacificstarted likewise earlier (between 18 and 19 ka BP) both at siteMD07-3128 directly offshore the Strait of Magellan (Caniupán et al.,2011) and further north at ODP Site 1233 (Lamy et al., 2007). In both

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e238

SST records, the initial warming lasted until ca 15 ka BP and wasfollowed byplateau and a secondwarming step starting at ca 12.5 kaBP leading into the pronounced early Holocene warming period(Fig. 6). The SST plateau largely coincides with the Antarctic ColdReversal (ACR) known from ice-cores. Along the Strait of Magellanpollen assemblages indicatemore humid conditions after 16.9 ka PBuntil at least 14.6 ka BP which may have caused a restricted re-advance forming moraine stage E at the Northern Isla Dawson(McCulloch and Davis, 2001). This limited and partly accumulation-driven re-advance partly overlaps with the ACR. The more than100 km ice retreat from moraine stage E towards the Cordillera

Fig. 6. Deglaciation history from 22 to 8 ka BP: A: d18O record from NGRIP (NorthGreenland Ice Core Project Members, 2004); B: Ice retreat phases at Lago Argentino(Kaplan et al., 2011), C: Glacier retreat stages C to E and suggested retreat lengths alongthe Central Strait of Magellan (McCulloch et al., 2005) as well as retreat constrains (redsquares) by Boyd et al. (2008) with a small early Holocene re-advance (CD-RE). D:Retreat lengths from the Seno Skyring (Kilian et al., 2007b). E: Alkenone SST record ofcore MD07-3128 (Caniupán et al., 2011); F: Antarctic surface temperature changes(deviation from mean of the last millenium; Jouzel et al., 2007) plotted on the newLemieux-Dundon time-scale (Lemieux-Dudon et al., 2010). G: Global sea level esti-mates compiled by Siddall et al. (2003). The marine transgression at the westernentrance of the Strait of Magellan (MT West, Kilian et al., 2007b) and its easternAtlantic entrance (MT East, McCulloch and Morello, 2009) are indicated.

Darwin was probably very fast, since marine sedimentation in theMarinelli Fjord started at around 14 ka BP (Boyd et al., 2008).

A second warming step occurred during the Northern Hemi-sphere Younger Dryas period. The possible occurrence of a YoungerDryas (YD) cold period in southern South America has been inten-sively debated over the past decade as pollen records in the ChileanLake District and on Chiloé Island have been interpreted in terms ofa YD cooling (Moreno et al., 1999, 2001) whereas pollen recordsfrom the Taitao peninsula and the Chonos archipelago, revealed nosuch cooling during the YD (Bennett et al., 2000). Further south atLago Argentino, Ackert et al. (2008) reconstructed advancingglaciers during the YD that culminated around 11 ka BP. This glacieradvance on the lee-side of the Andes was attributed an increasedprecipitation from easterly sources and does not imply colderconditions. However, a recent recalculation of the cosmogenicnuclide dates fromLagoArgentino suggests that this LagoArgentinoglacier advance took place earlier andmost of the advance falls intothe ACR (Fig. 6; Kaplan et al., 2011). This would be consistent withACR glacier advances in the region of the Strait of Magellan Strait(Sugden et al., 2005) and in the Torres del Paine area (Moreno et al.,2009b). Taken together, theDeglacial records confirm the “Antarctictiming” of paleoclimate changes in southern Patagonia.

Pollen records indicate the development of a treeless and openlandscape (tundra-like) during the deglaciation, dominated byeurythermal herbs and shrubs like Acaena, Empetrum, Gunnera andPoaceae, followed by the expansion of woody elements (e.g.Nothofagus) in response to an increase in temperature and mois-ture (e.g. Markgraf, 1993; Heusser, 1995, 2003; Huber et al., 2004;Villa-Martínez and Moreno, 2007; Mancini, 2009; Moreno et al.,2009a, 2010; Markgraf and Huber, 2010; Ponce et al., 2011).Interesting is the dissimilar timing and structure in the forestexpansion between northern and southern sites located along theeast of the Andes between latitudes 49 to 55�S. We observe that innorthern sites (Cerro Frias, Brazo Sur, Vega Ñandu, Lago Guanaco,Rio Rubens and Punta Arenas; Fig. 7), the Nothofagus expansionoccurs between >16 and 11 ka BP and gradually, whereas in moresouthern locations (Puerto del Hambre, Estancia Esmeralda,Puerto Haberton, Paso Garibaldi, Isla de los Estados) its transitionsoccurs much later (between 11.0 and 9 ka BP) and more abrupt.The humidity-sensitive and primarily thermophilous nature ofNothofagus suggests: (i) Favourable (unfavourable) moisture and/or temperature conditions during the deglaciation and earliestHolocene in northern (southern) sites, or (ii) non-climatic features(underlying mechanisms) controlling the forest expansiondynamics. Based on genetical evidences, Premoli et al. (2010)suggest that Nothofagus trees survived full glacial conditionswithin ecological niches on Tierra del Fuego at latitudes 54�S.Similar conditions must be occurred around the latitude 52�Swhere Nothofagus pollen present in glacial sediments of the PotrokAike record suggest small sheltered patches of Nothofagussurviving at the foot of the Andes (Wille et al., 2007; Recasenset al., 2011). Therefore northern and southern locations have thesame Nothofagus dispersal potential, and explain their dissimili-tude in these terms is not likely. Nevertheless at some sites, thedeglacial Nothofagus expansion was probably delayed hundreds orthousands of years with respect to the deglacial climate change(Heusser, 1995; Pisano, 1977) due to long term soil evolution andconsolidation within metastable glacial detritus (Heusser, 1995;Pisano, 1977; Ballantyne, 2002). Therefore if climate was the mainresponsible mechanism in the dynamic of the Nothofagus expan-sion, it may indicate a northerly position of the SWW core until theearliest Holocene (Lamy et al., 2010).

During the Late Glacial glacier retreat, a stepwise marinetransgression occurred into the proglacial lakes, forming anextended Patagonian fjord system. Kilian et al. (2007a) determined

Fig. 7. Nothofagus pollen records from sites within a precipitation range of400e1200 mm/year at the eastern side of the Andes (Isla de los Estados: Ponce et al.,2011; Haberton and Passo Garibaldi: Pendall et al., 2001; Puerto del Hambre I: Heusser,1995; Puerto del Hambre II: McCulloch and Davis, 2001; Punta Arenas, Rio Rubens:Huber et al., 2004; Lago Guanaco in the Torres del Paine Park: Moreno et al., 2010). Therelative amount of Nothofagus, the dominant tree species, indicates the vicinity to thetransition between Magellan forest and open parkland and/or steppe vegetation.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 9

a marine transgression to the western Strait of Magellan at around14.3 ka BP, related to melt water pulse 1A (Fig. 6) and nearly coevalto first marine influences in the Marinelli fjord (Boyd et al., 2008).The Atlantic transgression to the Strait of Magellan occurred later ataround 9 ka BP when the further global sea level rise passed theshallower eastern sills (McCulloch and Morello, 2009). The marinetransgression cross-cut the Andes and reached the Seno Otway ataround 14 ka BP and Seno Skyring as well as Seno Ultimo Esperanzaat around 10.3 ka BP (Kilian et al., 2007a; Stern et al., 2011).

6. Holocene

6.1. Paleovegetation

The climate transition zone to the east of the Andeswith annual precipitation of 500e1000 mm/yr was probably

characterised by rather open Nothofagus pumilio-dominatedsubantarctic deciduous forest during most of the Holocene (e.g.Heusser, 1995; McCulloch and Davis, 2001). Its local evolutionand eastward expansion towards the steppe may have beencontrolled by available moisture, in particular during summer(e.g. Pisano, 1977). In this area the available moisture depends onthe complex interplay between precipitation and amount ofevaporation which is controlled by wind velocities, humidity,temperature as well as on site-specific soil-types and drainageconditions of underlying substrate. In particular, the climatecontrol on the degree of evaporation at different sites have beenpoorly quantified and therefore remains a major problem forpalaeoclimatic interpretations based on pollen assemblages inthis region.

Eight Nothofagus pollen records from this humidity-sensitivetransition zone to the east of the Andes between latitudes 49� to55�S are summarised in Fig. 7. The records of the southernmostlocations Isla de los Estados (54�S, Ponce et al., 2011) and Haberton(54�S; Pendall et al., 2001) show a relatively late and gradual forestexpansion between 11.0 and 5.0 ka BP. High values of Ericaceae(>80%) and low values of Nothofagus (>20%) on Isla del Estadosuggest an open and perturbed vegetation, in particular between10.3 and 8.3 ka BP, which has been related to dry and warm climaticconditions (Unkel et al., 2010).

The Puerto del Hambre and Passo Giribaldi records (e.g. Heusser,1995; Heusser et al., 2000; McCulloch and Davis, 2001) show a fullevolved forest (with >75% Nothofagus pollen) first after 8.5 ka BPwhen regional temperatures dropped by around 1.5 �C (Chapter6.2.) and the SWWstrength probably decreased (Chapter 6.3.; Lamyet al., 2010). Previously, between w12.0 and 9.0 ka BP, the Puertodel Hambre and Estancia Esmeralda records show increased pollendegradation as well as humified peat deposits indicating very loweffective moisture (McCulloch and Davis, 2001). This could indicateless precipitation and/or higher evaporation rates possibly relatedto stronger westerlies combined with higher temperatures (Lamyet al., 2010, see Chapter 6.3.).

The Hambre record shows a significant reduction of tree pollenduring the last 5.5 ka BP (Figs. 7 and 13F; see also Chapter 6.6. onpossible human impact), whereas the pollen records from Isla delos Estados (Ponce et al., 2011), Haberton, Passo Giribaldi (Pendallet al., 2001), and Punta Arenas (Heusser, 1995) show evolvedforest vegetation after 5.5 ka BP and persisting throughout the LateHolocene. The Estancia Esmeralda site on the northern Isla Dawsonshows only restricted Nothofagus pollen (<40%) reflecting thesignificantly lower precipitation at this site situated further eastalong the regional precipitation gradient (Figs. 1 and 2).

Records from the more northern sites at Punta Arenas (Heusser,1995) and Rio Rubens (Huber et al., 2004; Markgraf and Huber,2010) as well as Lago Guanaco (Moreno et al., 2010) and LagunaVedga Ñandu (Villa-Martínez and Moreno, 2007) in the Torres delPaine area and locations close to Brazo Sur of Lago Argentino(Mancini, 2009; Wille and Schäbitz, 2009) show a more gradualevolution to a denser forest starting between 16 and 14 ka BP andlasting throughout the Holocene. This has been interpreted also asan eastward migration of the forestesteppe transition. The rela-tively high Nothofagus pollen (up to 50%) at the Punta Arenas sitebetween 16 and 12 ka BP are extraordinary and difficult to explain.In this case the Late Glacial section of the peat/sediment core couldhave been contaminated by Holocene material during drilling witha Hiller corer.

The above discussed pollen records does not exhibit a synchro-nous behaviour during phases of regionally known climateperturbations, like the 8.5 to 7.5 ka BP period, the beginning of the“Neoglacial” at ca 5.5 ka BP, the 3.5 to 2.5 ka BP period, theMedievalClimate Anomaly (MCA), and the LIA, which have been found in

Fig. 8. Pollen records from the hyperhumid western side of the Andes: (A) GC1 site near the Gran Campo Nevado (Fesq-Martin et al., 2004; Lamy et al., 2010) and (B) TM1 site onTamar Island (Lamy et al., 2010).

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e2310

many other records (e.g. Figs. 10 and 12). This individual behaviourmay be related to local microclimates and individual sensitivities ofthe ecosystems as well as locally distinct available moisturedepending on wind exposition and local soil types.

Pollen records frommore eastern sites in the Patagonian steppehave been published from Laguna Potrok Aike (e.g. Mayr et al.,2007a, 2009) and from Lago Cardiel (Markgraf et al., 2003). Thesepollen assemblages are partly controlled by a long distance trans-port of Nothofagus from Andean sites and/or by local moistureconditions which are slightly inversely correlated to westerly windstrength due to regular precipitation events derived from SouthAtlantic sources (Fig. 10D). Therefore their interpretation on thelocal paleovegetation and their climate relationship is notstraightforward. The SWW-relationship of the Laguna Potrok Aike

tree pollen record related to long distance pollen transport fromAndean sources is discussed in Chapter 6.3.

Only two well-dated pollen records have been presented fromthe hyperhumid side of the southern Andes up to now. They includethe Gran Campo Nevado area (Fesq-Martin et al., 2004) and LakeTamar (Lamy et al., 2010) along the western Strait of Magellan (sitesGC2 and TML in Fig. 4). Since the climate sensitivity of thesehyperhumid ecosystems is poorly investigated, the pollen varia-tions must be evaluated with care. In general the precipitation-dependent height of stagnant water in the peaty soils as well aslocal drainage conditions control to what extent peat bogs or,alternatively, forest evolved.

In the GC2-record of the Gran Campo area the amount ofhumidity sensitive hygrophyte pollen was interpreted within the

Fig. 9. Holocene paleotemperatures (and precipitation): A: Combined alkenone SSTrecord from core GeoB 3313-1 (Lamy et al., 2002) and ODP Site 1233 (Lamy et al., 2007)at the Pacific continental slope at 41�S (the published GeoB 3313-1 SST data have beencorrected by �0.7 �C in order to match the overlapping period with the ODP 1233record). B: Tree ring-based temperature reconstruction for southernmost SouthAmerica (Neukom et al., 2010a); C: Alkenone SST record of core MD07-3128 (Caniupánet al., 2011) D: Diatom-based SST-record from core PS2102-2 in the South Atlantic at53�S (Bianchi and Gersonde, 2004); E: Terrestrical organic carbon accumulation in coreTML-1 from Lake Tamar as a record for paleo-precipitation (Lamy et al., 2010). F:Antarctic surface temperature changes (deviation from mean of the last millenium;Jouzel et al., 2007) plotted on the new Lemieux-Dundon time-scale (Lemieux-Dudonet al., 2010).

Fig. 10. Precipitation and SWW-related records: A: Lake level reconstruction of LagoCardiel (Stine and Stine, 1990; Ariztegui et al., 2010); B: Magnetic susceptibility recordof core CAR 99-9P from Lago Cardiel as an indicator for westerly wind related driftdeposition in the lake (Gilli et al., 2005). C: Pollen-based reconstruction of precipita-tion near Brazo Sur (Lago Argentino; Tonella et al., 2009). D: Andean forest taxa ina sediment core of Laguna Potrok Aike as wind indicator (Mayr et al., 2007a, 2009); E:Pollen record of Laguna Guanaco in the Paine National Park as proxy for humiditychanges related to the SWW (Moreno et al., 2010). F and G: Clay/silt ratio and illitecontent of core SK1in the eastern Seno Skyring as proxy for wind-induced longdistance eastward sediment transport (Lamy et al., 2010). H: Terrestrial organic carbonaccumulation in Tamar Lake sediment core TML-1 as a paleo-precipitation proxy (Lamyet al., 2010). I: C/N ratios in sediment core PC-18 of Lago Fagnano as indicator forprecipitation-dependent terrestrial plant supply (Moy et al., 2008) and J: Fe contentrecord of core LF06-16 from Lago Fagnano as proxy for precipitation-related siliciclasticsediment-input (Waldmann et al., 2010). Thick line is 80-point running average.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 11

context of other multiproxy records to reflect humidity andrelated SWW-strength during the Holocene (Fig. 8A; Lamy et al.,2010). Cyperacae and Astelia which characterize plants of thepeat and wetland are more frequent in the Early Holocenebetween 11.2 and 8.5 ka BP and decrease towards the “Neoglacial”.Nothofagus growth which is hampered by high stagnant waterlevels became more frequent during the Middle and Late Holo-cene. Nothofagus pollen show a strong perturbation at the GC-2site associated with a large peak of the pioneer plant Gunnerafrom 4.1 to 2.9 ka BP. This was probably produced by local forestdecay and millennial scale regional perturbation of plantcommunities due to long-term sulphur release and additionalacidification of soils after deposition of a sulfur-bearing tephralayer from the 4.15 ka Mt. Burney eruption (Kilian et al., 2006).During the last millennium, a higher amount of peat land versustree pollen could indicate increased moisture and/or may reflectregionally up to the 1.5 �C lower temperatures after 0.7 ka BP(Neukom et al., 2010a,b).

The Lake Tamar (Site TML in Fig. 4) pollen record showsa distinctive forest succession (Fig. 8B; Lamy et al., 2010), rangingfrom a Nothofagus-dominated forest (w13.2e9.3 ka BP), toNothofaguseDrimys forest (w9.3e5.2 ka BP), and leading to the

Fig. 11. Compilation of selected paleoclimate records and Holocene glacier advances inthe southernmost Andes: A: Paleoprecipitation based on tree pollen from LakeGuanaco in Paine National Park 50�S (Moreno et al., 2010), B: Alkenone SST record ofODP Site 1233 at the Pacific continental slope at 41�S (Kaiser et al., 2005; Lamy et al.,2007); C: Alkenone SST record of core MD07-3128 (Caniupán et al., 2011), D) TerrestrialCorg accumulation in core TML1 from Lake Tamar as a precipitation proxy (Lamy et al.,2010), E: d18O values of stalagmite MA1 in the Southern Andes at 53�S as indicator forprecipitation (Schimpf et al., 2011), F: Glacier advance phases in the southernmostAndes from 1: Wenzens and Wenzens (1998), 2: Aniya (1995), 3: Mercer (1982), 4:Masiokas et al. (2009a) and from the Antarctic Peninsula and adjacent islands (Hall,2009). GI: Magnetic susceptibility as indicator for glacial clay in core JPC67 corefrom the Marinelli fjord next to Cordillera Darwin (Boyd et al., 2008) and H: Ice rafteddebris (IRD) in the South Atlantic at 53�S (Hodell et al., 2001).

Fig. 12. High resolution records interpreted in terms of SWW behaviour, ENSOchanges, sun activity, position, strength of the ITCZ, and/or Antarctic temperatures: A:10Be-based sun activity reconstruction (Steinhilber et al., 2009); B: Ti content record ofODP Site 1002 from the Cariaco basin, indicating position and strength of the ITCZ(Haug et al., 2001); C: Grey scale variations in a sediment core from Laguna Pallcacochain the Ecuadorian Andes as a proxy for ENSO events (Moy et al., 2002); D: Colour-scalevariations as indicator for ENSO-related flood-derived lithic grains in sediment core106KL from the Peruvian coast (Rein et al., 2005); E: Fe content changes in core inLF06-16 from Lago Fagnano as indicator for SWW-related precipitation (Waldmannet al., 2010). F and G: d18O values and Y content in the stalagmite MA1 in theSouthern Andes at 53�S as indicator for drip rates and SWW strength (Schimpf et al.,2011). H: Antarctic surface temperature changes (deviation from mean of the lastmillennium; Jouzel et al., 2007).

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e2312

establishment of mixed evergreen forest (w5.2 ka BP to present).The establishment of NothofaguseDrimys forest and finally themixed evergreen forest was interpreted as an important precipi-tation decrease towards the late Holocene. High relative percentageof Misodendrum, a hemiparasite that infests several Nothofagusspecies (Markgraf et al., 1981), during the early Holocene(w11.3e9.3 ka BP) has been interpreted as Nothofagus forestgrowing under limited ecological conditions (e.g. swamp soils). Thispalaeoclimatic interpretation is consistent with high values of C/Nand accumulation rates of terrestrial organic carbon in Lake Tamartriggered by enhanced precipitation during this period (Lamy et al.,2010). Nevertheless high values in the total pollen concentrationand influx parameters during the early Holocene suggest denserplant cover than during the late Holocene.

6.2. Paleotemperature

Tree-ring widths (e.g. Villalba et al., 2003; Neukom et al., 2010a,b), hydrogen isotopes of mosses (Pendall et al., 2001), C and O

isotopes of stalagmites (Mühlinghaus et al., 2008, 2009), and theUK37 index of alkenones (e.g. Caniupán et al., 2011) in marinesediments were used to calculate calibrated Holocene tempera-tures. While most of these records show relatively low tempera-tures for the LIA after 0.7 ka BP, there are no clear trends orfluctuations for the whole Holocene except for the open oceanrecord of Caniupán et al. (2011) which reveals a well-defined earlyHolocene SST maximum. In the following we discuss the differentpaleotemperature reconstructions and possible reasons forinconsistencies.

Pendall et al. (2001) presented a Late Glacial and Holocenetemperature reconstruction which is based on a hydrogen isotoperecord of mosses from a peat bog at Tierra del Fuego. This recorddoes not show clear Holocene fluctuations within the given errorrange and the effect of likely changes in the rain water isotopyremains unclear. The d13C record from bulk organic material in lakesediments of the Lago Fagnano was interpreted to reflect partlytemperature-controlled aquatic bioproductivity indicating

Fig. 13. Relationship between human occupation in the Magellan region and Tierra delFuego and paleoclimate records: A: Antarctic surface temperature changes (deviationfrom mean of the last millennium; Jouzel et al., 2007) plotted on the new Lemieux-Dundon time-scale (Lemieux-Dudon et al., 2010). B: Terrestrial organic carbon accu-mulation in Tamar Lake sediment core TML-1 as paleo-precipitation proxy for theAndean area (Lamy et al., 2010). C and D: Y content and C/N ratios in a sediment corefrom Lake Hambre as a proxy for terrestrial erosion around the lake (Hermanns andBiester, 2011). E: Human occupation phases at Tierra del Fuego (Morello et al., 2012).F: Charcoal as well as Graminae and Nothofagus pollen Puerto del Hambre (Heusser,1995). G: Black arrows indicate human occupation phases (for details and referencessee text) and global sea level curve after Siddall et al. (2003) and present day waterdepths at Primera and Segund Angostura of the Strait of Maggellan.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 13

a Holocene “Hypsithermal” between 7.0 and 5.0 ka BP (Moy et al.,2011). However, this record does not cover the Early Holocene.

For southernmost Patagonia tree-ring based temperatures havebeen reconstructed for the last 1.2 ka BP (Villalba et al., 2003;Neukom et al., 2010a,b) indicating a significant temperaturedecrease of about 1.5 �C between 0.7 and 0.5 ka BP (Fig. 9).However, due to the limited length of most tree ring chronologies,only the last 0.6 ka BP have smaller error bars and are more reliable(Neukom et al., 2010a,b). Though average annual temperatures (atthe same altitude) do not change significantly across the Andeandivide, tree-ring based temperatures from the eastern slopes of theAndes, in particular north of 52�S, may respond to the highersummer temperatures (higher seasonality). Furthermore, tree-ring

chronologies from the hyperhumid southern Andes do not showa temperature-relationship when compared with 100 year weatherstation data from Evangelistas (Koch and Kilian, 2002; Rigozo et al.,2007).

Based on the distinct temperature dependence of the kineticallycontrolled C and O isotope fractionation in two stalagmites for thesouthernmost Andes at latitude 52�S (Ma1 and MA2 records inMühlinghaus et al., 2008, 2009), paleotemperatures have beencalculated for the past 5 ka BP. However, this reconstruction doesnot consider temperature dependent changes in the isotopiccomposition of rain water (Schimpf et al., 2011), in particular,during periods of significantly lower temperature, like during theLIA (1.5 �C; see tree-ring temperatures in Fig. 9).

An alkenone-temperature record was obtained from site MD07-3128 at the continental margin off the Pacific entrance of theMagellan Strait (Caniupán et al., 2011). This record provides hightime resolution for the Last Glacial between 55 and 13 ka BP (Fig. 5;Chapter 5), whereas the Holocene record has low time resolution.However, the record shows an early Holocene temperaturemaximum between 11.5 and 8.5 ka BP (Fig. 9) similar to openmarine records off Chile further north (e.g. ODP Site 1233, 41�S;Kaiser et al., 2005). Holocene fjord SST records from southernPatagonia have not yet been published. Fjord alkenone SST recordsfrom the northern Patagonian fjords (Sepulveda et al., 2009) andon-going studies in the south (Magaly Caniupan and Claudia Ara-cena, personal communication) suggest SST cooling between 2.5and 3.5 ka BP and the LIA period and hint to the large potential ofthis method for regional paleotemperature reconstructions. The2.5e3.5 ka cooling has also previously been suggested based onpeat records in southern Patagonia (Van Geel et al., 2000;Chambers et al., 2007) and might be a global event (Wanner et al.,2011).

6.3. Paleoprecipitation, water mass balance and possiblerelationship to SWW strength

A major focus of paleoclimate research in southern Patagonia isthe reconstruction of changes in the intensity and latitudinalposition of the SWW (e.g. Gilli et al., 2005; Villa-Martínez andMoreno, 2007; Mayr et al., 2007a, b, 2009; Moy et al., 2008, 2011;Moreno et al., 2009a, 2010; Kastner et al., 2010; Lamy et al., 2010;Waldmann et al., 2010; Fletcher and Moreno, 2011; Fletcher andMoreno, 2012). The SWW crosses the entire southern tip of SouthAmerica (Fig. 1B; e.g. Garreaud et al., in press) and therefore purewind proxies should give similar results from different sites.However, such proxies (e.g. wind-related transport of pollen orspecific clay-types in superficial waters) are rare and their relationto wind strength is not always straightforward. Therefore, mostSWW reconstructions are based on the relationship betweenprecipitation and SWW strength. As shown in Fig. 1C, the westernside of the Andes the western fjord system exhibit a good corre-lation of r ¼ 0.5e0.8 between precipitation and westerly windstrength while the eastern slopes of the Andes (e.g. Punta Arenasand Torres del Paine area) exhibit a weaker correlation ofr ¼ 0.2e0.4. In this area the wind-related effect on evaporation forthe local water balance becomes important, in particular duringsummer. Further to the east, in the steppe region, westerly windstrength is slightly negatively correlated with precipitation due toincreased easterly moisture from the Atlantic, in particular whenthe SWW strength is weak. This applies for example to Lago Cardieland Laguna Potrok Aike.

Already 20 years ago (Stine and Stine, 1990), lake level changesof Lago Cardiel (Site in Fig. 4) have been used to depict the regionalwater balance during the Late Glacial and Holocene (Fig. 10A).These reconstructions are consistent with more recent studies (e.g.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e2314

Ariztegui et al., 2010; Cusminsky et al., 2011) on precipitation/evaporation ratios deduced from lacustrine microfossil assem-blages. However, the lake catchment receives precipitation origi-nating from the westerlies during summer months, but alsosignificant moisture form Atlantic sources in particular duringwinter. The lake level increased significantly from �80 m(compared to present day) after 11.5 ka BP and reacheda maximum of þ60 m at around 10 ka BP (Fig. 10B). After 8 ka BP,only minor changes of �20 m are recorded. Gilli et al. (2005)suggested that the magnetic susceptibility of late and middleHolocene sediment records in this lake reflect westerly windstrength. Between 8 and 5 ka BP, this record indicates higherwesterly wind intensities which afterwards decrease until thepresent (Fig. 10B).

Tonella et al. (2009) calibrated a pollen record from a site closeto Lago Argentino (Site location see in Fig. 4) with respect toprecipitation. For this propose, they investigated surface pollen ina W-E transect and compared them with the present day precipi-tation gradient. Their record indicates a successive precipitationincrease from 12.0 to 3.5 ka BP (from ca 300e600 mm/year) anda subsequent general decrease superimposed by a secondarymaximum around the Medieval Climate Anomaly anda significant decrease during the LIA (Fig. 10C). However, the cali-bration has a large error range of about�150mm/yr and the impactof evaporation on the available moisture in the soils has not beenconsidered clearly.

Mayr et al. (2007a, 2009) show the relative contribution ofAndean Forest Taxa (AFT) during the past 12 ka BP based ona sediment record from Laguna Potrok Aike (Site location in Fig. 4)(Fig. 10D), situated at 52�S around 80 km east of the Andean forestbelt. They interpreted the AFT content as predominately wind-dependent eastward pollen transport from forested Andean sites.The highly resolved record shows pronounced decadal to centen-nial variability, but no significant long term trend throughout theHolocene. Amarked increase of AFT associated with a low lake levelis registered between 8.5 and 7.5 ka BP (Fig. 10D). This period wasinterpreted as the strongest Holocene SWWphase characterised bylow precipitation and pronounced evaporation. For the sameperiod, pollen records from Lago Guanaco (Moreno et al., 2010,Fig. 10E) and Vega Ñandu (Villa-Martínez and Moreno, 2007) (bothTorres del Paine region) show the lowest tree pollen content duringthe past 12 ka BP. In contrast to the results of Mayr et al. (2007a,2009), the period from 8.5 to 7.5 ka BP was interpreted as theweakest SWW phase of the Holocene. The systematic increase oftree pollen in both Torres del Paine records has been interpreted asincreasing westerly wind strength during the Holocene, culmi-nating during the LIA (Villa-Martínez andMoreno, 2007; Moy et al.,2008; Moreno et al., 2010). This interpretation is based on anassumed positive (Fig. 1C) of tree pollen, precipitation and SWWstrength. More importantly, the impact of wind velocities andtemperature on evaporation and related available moistureremains unclear. Lago Guanaco is relatively alkalic (Moy et al.,2008), suggesting that evaporation is an important factor in thisarea which dries out soils and plants, especially during the windysummer periods (see discussion in Lamy et al., 2010). An alternativeinterpretation of the Torres del Paine pollen records could be thatthe available moisture in the soils of the hilly catchment wasstrongly reduced during windier and warmer periods (e.g. betweenca 7.5 and 8.5 ka BP interval) or more generally during the early andmiddle Holocene. Such an alternative interpretation would beconsistent with previous pollen interpretation in this region(Heusser, 1995) and records west of the climate divide (Lamy et al.,2010; see below) (Fig. 10E).

Further to the south at Tierra del Fuego, sediment records fromLago Fagnano (Fig. 4) have been used for westerly wind

reconstructions (Waldmann et al., 2010; Moy et al., 2011). Thoughless pronounced than the western fjord region (see below), thisarea is characterised by a clear positive correlation of precipitationand wind strength (Fig. 1C). Waldmann et al. (2010) used the ironcontent record of a sediment core from the lake as a proxy forprecipitation-dependent siliciclastic sediment input (Fig. 10J). Thisultra-high resolution record shows pronounced decadal and somecentennial changes, but may also depict single extreme storm and/or precipitation periods. There is however no clear long term trendin this record. Moy et al. (2011) use C/N ratios from another sedi-ment core retrieved from Lago Fagnano to characterize the possiblyrain-dependent supply of terrestrial organic matter. In contrast tothe above described iron record, the continuously increasing C/Nratios since 8 ka BP are interpreted to reflect a continuous increasein precipitation and SSW strength throughout the Holocene,culminating in the LIA. As such this record would be compatiblewith the tree pollen interpretations based on the above discussedTorres del Paine records (Villa-Martínez and Moreno, 2007;Moreno et al., 2010). However, the millennial-scale changes of theVega Ñandu pollen record are also not visible in the C/N recordfrom Lago Fagnano. An increased accumulation of terrestrialorganic matter in the lake sediment could alternatively also beexplained by increasing thicknesses of peaty soils throughout theHolocene which would enable more terrestrial organic carbonsupply during the Late Holocene, even during similar or lessintensive precipitation and runoff. Pollen records from Puerto delHambre (McCulloch and Davis, 2001) within a similar climaticsetting at the lee-side of the Andes which have been recentlyinterpreted to reflect regional precipitation and SWW strength(McGlone et al., 2010) do not clearly show a correlation with the C/N record of the nearby Lake Hambre (Fig. 13B; Hermanns andBiester, 2011).

Lamy et al. (2010) present a Holocene multiproxy compilationfrom sites within the hyperhumid Andes where the present-daycorrelation between precipitation and SWW strength is highest(Fig. 1C). An increase in precipitation and SWW strength wasdeduced from humidity sensitive pollen (Fig. 8), precipitation-dependent terrestrial organic carbon accumulation in lake andfjord sediments, salinity dependent decrease in biogenic carbonateaccumulation and the >80 km long-distance eastward transport ofillite-rich Andean clay from the Patagonian Batholith towards theeastern sector of the former proglacial Lake Seno Skyring (two ofthe proxy records are shown in Fig. 10FeH). These records indicatea stronger SWW from 12.5 to 8.5 ka BP, a transitional weaker SWWphase from 8.5 to 5.5 ka BP and a subsequent interval (“Neoglacial”)with a more variable but in general lower SSW strength. Theseresults contrast especially with the above described pollen inter-pretation form the Torres del Paine area (Villa-Martínez andMoreno, 2007; Moreno et al., 2010) and the C/N record from LagoFagnano given by Moy et al. (2011).

Taken together, the above presented records (most of themshown in Fig. 10) of Holocene westerly wind strength changes dopresently not provide a consistent picture concerning the behav-iour of the SWW during the Holocene. While pollen interpretationsfrom the Torres del Paine region and the C/N record of Lago Fagnanosuggest weak westerlies in the Early Holocene and stronger west-erlies during the Neoglacial (culminating during the LIA), recordsfrom the hyperhumid western side of the Andes in general showthe reverse pattern. The only possibility to explain these contrast-ing trends would be subtle latitudinal shifts of the westerliescausing contrasting wind intensities over short distances (ca150 km between Torres del Paine and the Gran Campo area).Considering the modern extension of the SWW core, such small-scale shifts for explaining contrasting precipitation trends insouthern Patagonia appear rather unlikely.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 15

Other records such as the AFT from Potrok Aike and the Feintensities from Lago Fagnano indicate high short term vari-ability, but only subtle long term trends. Independent of theindividual SWW related interpretation of precipitation records,the general trends allow to distinguish characteristic periods.These periods include the latest Glacial to early Holocene from12.5 to 8.5 ka BP and a period from 8.5 to 7.5 ka BP which ispossibly related to global climate reorganization related to the8.2 ka BP event which was observed at many sites of theNorthern Hemisphere and also at lower latitudes between 8.4and 7.9 ka BP (e.g. Rohling and Pälike, 2005). From 7.5 to 5.5 kaBP most records reveal a transitional phase leading into a distinctlate Holocene interval. Pronounced perturbations are alsoobvious in the period from 3.5 to 2.5 ka BP and during the LIAafter 0.65 ka BP (Fig. 12).

6.4. Glacier fluctuations

On a global scale, Holocene glacier fluctuations are oftenascribed to temperature and related ablation changes (Andersonet al., 2010), especially during the past few centuries (Oerlemans,2005). In temperate very humid mountain ranges such as theNew Zealand alps and the Norwegian mountains, winter snowaccumulation and thus precipitation related to atmospheric circu-lation changes may play a major role (e.g. Rother and Shulmeister,2006; Nesje et al., 2008; Anderson et al., 2010). Due to theextremely high precipitation, glacier advances may thus alsolargely be accumulation driven in southern Patagonia as suggestedalready by Warren and Sudgen (1993) who also speculated thatablation may play a major role on the drier eastern side of theAndes. In the following, we summarise the available information onHolocene glacier advances and discuss their forcing in the contextof the paleoclimate background data (i.e. paleotemperature andpaleoprecipitation reconstructions).

In Fig. 11F, we provide a compilation of Holocene glacieradvances in the southernmost Andes (e.g. Mercer, 1982; Aniya,1995; McCulloch et al., 2000; Glasser et al., 2004; Wenzens,2005; Masiokas et al., 2009a) and the Antarctic Peninsula (Hall,2009) located at the southern margin of the SWW. Only few EarlyHolocene advances (mostly around ca 8 ka BP) have so-far recon-structed from southern Patagonia, for example from the LagoArgentino and Viedma area (Wenzens and Wenzens, 1998;Wenzens, 1999, 2005). These studies are consistent with recon-structed glacier advances at the NPIF (Douglass et al., 2005) theAntarctic Peninsula (Hall, 2009). However, the early Holocenerecords are primarily based on moraine dating and subsequentmore extended late Holocene glacier advances might have over-printed or eroded earlier moraines. The only early Holocene glacieradvance record not based on moraine dating, is a glacier claydeposition record from a fjord core north of Cordillera Darwin(Fig. 11G) which reveals a minor advance at ca 8 ka BP (Boyd et al.,2008).

The next dated glacier advances (from ca 4.9e5.4 ka BP) at thebeginning of the “Neoglacial” are recorded at several sites in thesouthern Andes (Mercer, 1982; Aniya, 1995; Glasser et al., 2004;Wenzens, 2005) and coincide with glacier advances on a globalscale (Porter, 2000; Magny and Haas, 2004) and increased icerafted detritus recorded in the South Atlantic (Hodell et al.,2001). A subsequent minor advance between ca 3.9 and 4.3 kaBP occurred at several locations in the southern Andes, butprobably not at Cordillera Darwin (Kuylenstierna et al., 1996).However, an IRD peak in the South Atlantic also indicatesincreased Antarctic drift ice during this period (Fig. 11H; Hodellet al., 2001). At many sites at the eastern (Fig. 11F) and westernside of the Andes, glacier advances were recorded for the interval

between ca 2.0e2.7 ka BP (e.g. Mercer, 1982; Glasser et al., 2004).This advance period appears also in the glacial clay depositionrecord of the Cordillera Darwin (Boyd et al., 2008, Fig. 11G). Atsome sites at the western side of the Andes further advancesoccurred during the period from 0.9 to 1.2 ka BP which roughlycoincides with the global MCA (Masiokas et al., 2009a). At othereastern sites, these moraines may have been overrun by moreextended advances during the global LIA. The LIA advance periodin southern Patagonia ranged from 0.6 to 0.05 ka BP. Mostmoraines are well preserved at nearly all glacier systems. Duringthis period, many glaciers reached their maximum extension ataround 0.55 to 0.4 ka BP and afterwards experienced a successiveretreat until the last century (Koch and Kilian, 2005, Kilian et al.,2007a; Masiokas et al., 2009a,b).

The few available Holocene temperature reconstructions (see6.2.) suggest that the early Holocene from 11.5 to 8.5 ka BP wascharacterised by 1.5e2 �C higher temperatures compared to theLate Holocene (e.g. Fig. 11B and C). Thus early Holocene advancesrequire significantly higher precipitation than during the Middleand Late Holocene. Such accumulation-driven advances, wouldonly be conceivable under very humid conditions and mighttherefore support the western Andean paleoprecipitation recon-structions by Lamy et al. (2010) (Fig. 11D) compared with thecontrasting interpretations from the Torres del Paine region(Fig. 11A; Moreno et al., 2010). Also the “Neoglacial” advancesappear to coincide with short-term comparatively warm andhumid late Holocene intervals as recorded in stalagmites (Schimpfet al., 2011) superimposed on a long-term cooling (Fig. 11BeD).Tree-ring data suggest substantially, ca 1e1.5 �C colder tempera-tures around the LIA which might have forced southern PatagonianLIA glacier advances in spite of probably more arid conditions(Fig. 11E) (Schimpf et al., 2011).

6.5. High resolution records and event stratigraphy

High resolution records (1e10 yr resolution) have been obtainedfrom Laguna Potrok Aike (e.g. Habelzettl et al., 2007, 2009), LagunaVizcachas (Fey et al., 2009), Lake Tamar (Lamy et al., 2010), LakeHambre (Fig. 13; Hermanns and Biester, 2011), Lake Fagnano(Waldmann et al., 2011) and from staglamites of the Marcelo Are-valo (MA) cave (e.g. MA1 stalagmite record; Schimpf et al., 2011).These records may be able to depict single storm events (e.g. LakeTamar, Lamy et al., 2010) or can show mass flows related to earthquakes which appeared at Lago Fagnano throughout the Holocenewithin recurrence times of 0.2e1.0 ka (Waldmann et al., 2011).Although all the above named records are situated at the southernmargin of the present-day core of the SWW (Figs. 1 and 4) at lati-tudes 52�30’S to 54�S, correlations on a decadal to centennial scaleare often difficult. However, the precipitation-dependent ironconcentrations in the LF06-16 sediment core from Lago Fagnano(Fig. 13E; Waldmann et al., 2010) and the drip rate/precipitation-dependent d18O values as well as Y contents of the MA1 stalag-mite (MA cave near Gran Campo Nevado) exhibit surprisinglysynchronous pattern on a centennial and decadal scale (Fig. 12FeG)although all three proxies have largely unknown sensitivities andthreshold levels. A major difference between both records concernsthe phase between 3.5 and 2.5 ka BP for which a relatively weekSWW is indicated in the stalagmite record. At Lago Fagnano situ-ated east of the climate divide a generally higher precipitation isrecorded during this period which can be explained by a higheramount of easterly-derived precipitation due to a weaker SWW.However, during the last 5 ka both records show seven pronouncedprecipitation and SWW maxima with durations of 0.1e0.25 ka aswell as SWW minima (1e7a and b in Fig. 12E and F). The patternof both SWW-related records also show considerable correlations

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e2316

with the total solar irradiance (Fig. 12A; Steinhilber et al., 2009;Vieira et al., 2011), the latitudinal position and/or intensity of theITCZ (Fig. 12B; Haug et al., 2001) as well as the only two existingcontinuous and highly resolved records of El Niño events repre-sented by colour intensity variations in a sediment core fromLaguna Pallcacocha in Ecuador (Fig. 12C; Moy et al., 2002) and theamount of lithic grains in a sediment core from the Peruvian coast(Fig. 12D; Rein et al., 2005). Both cores show in common severalphases with an increased number of El Niño events during theperiod between 3.5 and 2.5 ka BP as well as during several shorterphases. In the Peruvian coastal record the MCA was identified asa period of dominant La Niña situations while the Laguna Pallca-cocha record shows increased El Niño events after around 1.2 ka BP.However, the youngest age control point in the Pallcacocha recordis at 1.32 ka BP and low sedimentation rates during more frequentLa Niña could imply a later onset of this El Niño event period in thisrecord as shown in Fig. 12CeD.

6.6. Possible anthropogenic interferences in paleoclimate records

The possible human impact on ecosystems and paleoclimateproxies is difficult to evaluate, because quantitative estimates ofthe population and there regional distribution are not well known.Nevertheless, we summarise in the following the available infor-mation on the inhabitation history and their possible relation topaleoclimate. Thereafter, we discuss potential anthropogenicperturbations in the paleoclimate records. The steppe region ofsouthernmost Patagonia was populated by terrestrial hunter-gatherer after ca 12.5 ka BP (e.g. Borrero, 1999; Legoupil, 2009;Orquera et al., 2011) when most glacier systems of the Magellanesregion have been retreated nearly to present-day extension (e.g.McCulloch and Bentley, 1998; Kilian et al., 2007a; Boyd et al.,2008). These humans occupied also Tierra del Fuego soon after12.0 ka BP when the lower global sea level allowed for a landbridge in the sector of Primera and Segunda Angostura of the Straitof Magellan (McCulloch and Morello, 2009; Morello et al., 2012,Figs. 4 and 13G). This early Holocene period between 11.5 and8.5 ka BP was probably very arid in the steppe region and partly atthe eastern slopes of the Andes, characterised by high charcoalpresence within regional pollen records (Fig. 13F; e.g. Heusser,1995; McCulloch and Davis, 2001) (see also Chapters 6.2. and6.3.). This arid climate (with strong wind-induced drying of plants)abets the expansion of fires, but a possibly stronger SWW(following Lamy et al., 2010) (Fig. 13B) decreases the probability ofthunderstorms which could cause natural fires. Heusser (1995)suggested that at least some fires of this period are related tohuman activity. However, the causes for natural versus anthro-pogenic fire-activity remain complex and therefore stronglydisputed, especially for open forest regions to the east of the Andes(e.g. Heusser, 1994; Huber et al., 2004; Whitlock et al., 2007;Markgraf and Huber, 2010), whereas charcoal is nearly absent inrecords of the hyperhumid Patagonian rain forest (Fesq-Martinet al., 2004).

The period from 8.5 to 5 ka BP was probably less arid to the eastof the Andes and pollen records from the Puerto del Hambre as wellas Estancia Esmeralda on the Northern Isla Dawson indicate a moreexpanded and stable forest ecosystem with restricted fire activity(Figs. 7 and 13F; Heusser, 1995; McCulloch and Davis, 2001). Onlyfew remnants of inhabitants have been found on Tierra del Fuegoduring this period. This may be related to its island location afterthe formation of the Western Strait of Magellan at around 9 ka BPcaused by the global sea levels rise (Fig.13G; McCulloch et al., 2005;McCulloch and Morello, 2009). Pollen records from Puerto delHambre also indicate a stable forest system between 10.0 and5.5 Ka BP (Fig. 13F). After around 6.5 ka BP, an increased number of

settling ages of maritime hunteregatherer were reported fromAndean fjords and the southern sector of the Strait of MagellanStrait near Puerto el Hambre and Santa Ana (Prieto et al., 2009;Torres, 2009; Orquera et al., 2011). This occupation period withcanoeing Indians may explain a renewed inhabitation of Tierra delFuego after 6.5 ka BP (Morello et al., 2012). However significantanthropogenic impacts on proxy records of this period are notclearly documented.

A reduced and more open Nothofagus forest together withmore regular charcoal events occur during the last 5.5 ka BP inthe pollen record of Puerto del Hambre although palaeoclimaticbackground data only show moderate climate variations duringthis period (Fig. 13A and B) and other regional pollen recordsfrom e.g. Punta Arenas, Haberton and Paso Giribaldi do not showa significant forest reduction (Fig. 7). Therefore it can be ques-tioned whether these changes in the forest ecoton near to Puertodel Hambre may reflect anthropogenic perturbation. The LakeHambre sediment record indicates a pronounced increase of LateHolocene erosion characterised by very high siliciclastic accu-mulation rates (high Y-content of the sediment; Fig. 13E) and veryhigh deposition rates of terrestrial plants (high C/N ratios;Hermanns and Biester, 2011), in particular from 5.5 to 4.3 ka BP.This coincides with higher charcoal contents in the Puerto delHambre record (Fig. 13F; Heusser, 1995) and frequent radiocarbonages of settling remnants in this area (Prieto et al., 2009; Torres,2009; Orquera et al., 2011). This settling activity may haveprovoked more frequent anthropogenic fires with forestdestruction and related erosion.

7. Hemispheric/interhemispheric linkages and climateforcing mechanisms

7.1. Insolation

Insolation changes are regarded as the major forcing factorand pacemaker for glacial-interglacial changes in global climate(e.g. Hays et al., 1976). In the traditional Milankovic theory,primarily summer insolation at 65�N and its impact on NorthernHemisphere glaciation has been involved for explaining globalice-ages (Imbrie et al., 1992). For the Southern Hemisphere (SH)where summer insolation changes are out-of-phase to the north,changes in austral spring insolation have been recently proposedto explain the warming during the Last Glacial termination (e.g.Stott et al., 2007). Modelling studies suggest that these springinsolation changes combined with variations in greenhouse gasare sufficient to explain most of the SH deglacial warming andthe cooling during the course of the Holocene (Renssen et al.,2005; Timmermann et al., 2009). Though small in amplitude,changes in mean annual insolation may additionally play a rolein particular during the Holocene as suggested by climatemodelling studies (Liu et al., 2003). The effect on orbital-scaleinsolation forcing on the SWW during the Holocene has alsobeen recently addressed in modelling studies (Wagner et al.,2007; Varma et al., 2011). Wagner et al. (2007) suggest forexample enhanced austral summer and reduced austral winterwesterlies in the middle Holocene over southernmost Patagoniainduced by insolation-driven meridional changes in temperaturegradients.

During the Deglacial, the impact of insolation changes may bepartly obscured by the antiphased temperature excursions of theACR and Younger Dryas, probably related to the bipolar seesaweffect (e.g. Stocker, 1998; Stenni et al., 2011; Chapter 7.3.). HoloceneSST records within the SWW and ACC belt off Chile, including thesouthern Patagonian site MD07-3128 and the northern PatagonianODP Site 1233, are compared with orbital-scale insolation changes

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 17

in Fig. 14. These records show an Early Holocene thermal maximumbetween 11.5 and 8.5 ka BP likewise known from Antarctica(Masson et al., 2000) which may be related to the maximum inaverage annual insolation at ca 10 ka BP (Fig. 14H). However, thismaximum represents only 0.4 W/m2 higher insolation andmodelling results suggest that annual insolation may account foronly 0.5 �C early Holocene warming in the Southern Ocean (Liuet al., 2003). Summer insolation reaches a pronounced maximumduring the Late Holocene, opposite to the general slight cooling inthe SST records whereas Antarctic temperature records show noclear trends during the middle and late Holocene (Fig. 14). Theformer considerations suggest that the link between proxy data andinsolation changes in the Southern Hemisphere is notstraightforward.

Fig. 14. Relationship between insolation and hemispheric temperatures: A: Summerinsolation at 50�N (Berger and Loutre, 1991). B: d18O record from the NGRIP ice-core asindicator for the Northern Hemisphere temperature evolution (North Greenland IceCore Project Members, 2004); C: Alkenone SST record of ODP Site 1233 at the Pacificcontinental slope at 41�S (Kaiser and Lamy, 2010); D: Alkenone SST record of coreMD07-2120 near New Zealand at 46�S (Pahnke and Sachs, 2006): E: Alkenone SSTrecord of core MD07-3128 (Caniupán et al., 2011); F: Diatom-based SST-record insediment core PS2102-2 from the South Atlantic at 53�S (Bianchi and Gersonde, 2004);G: Summer insolation at 50�S (Berger and Loutre, 1991) and H: Average annual inso-lation at 50�S (Berger and Loutre, 1991).

7.2. Sun activity

Total solar irradiance (TSI) changes have been recently recon-structed for the whole Holocene based on changes of the Earth’smagnetic field and the 14C (Vieira et al., 2011) as well as 10Bebehaviour (Steinhilber et al., 2009). The TSI variations are in therange of �1 W/m2 which is higher than the above discussedHolocene changes of the average annual insolation (�0.4 W/m2).Phases of low TSI have been related with Holocene North Atlanticdrift ice events (Bond et al., 2001) and also to global climateperturbations (Wanner et al., 2008, 2011). In southern Patagoniasuch probably sun-related temperature depressions occurredespecially during the period between from 3.5 to 2.5 ka BP, inparticular around 2.7 ka BP (Chambers et al., 2007; Van Geel et al.,2000; Mühlinghaus et al., 2009). As discussed in Chapter 6.2., treering-based temperature reconstructions (Neukom et al., 2010a, b)and all SST records (Fig. 9) show a temperature depression of1.0e1.5 �C during the LIA period which follows in general the well-known sun activity minima.

The evaluation of early and mid-Holocene influences of the TSIis more difficult. Fjord SST reconstruction have to be consideredwith care during this time interval, since high melt water fluxesduring spring and summer may have depressed the SST’s signifi-cantly. Furthermore wind-induced changes in transport rates ofdrift ice in the fjords after major calving events may have impactedSSTs. Furthermore, most TSI changes appeared on a scale of somedecades to a century which is relatively short compared to thetime resolution of fjord sediment cores. Nevertheless, in Fig. 11 wecompare SWW-sensitive high resolution lake and stalagmiterecords with the TSI values of the last 5 ka BP (Fig. 11). There isa clear relationship between reconstructed SWW strength and theTSI record. Since SWW changes are driven especially by latitudinalSST gradients on the Pacific and vertical atmospheric temperaturegradients, we expect that the sun has a clear influence on thethermal structure of the Southern Hemisphere within a timeresolution of decades to centuries. This linkage is also confirmedby sun related cyclicities (11, 22, 90 and 210 yrs) which have beendetected in tree ring records near the Pio XI glacier on the westernside of the southern Patagonian Ice Field (Rigozo et al., 2007) andin a stalagmite record from the Andes at 53�S (Schimpf et al.,2011).

7.3. Ocean currents

The maritime climate of Southern Patagonia is strongly affectedby oceanographic changes especially in the Southern Pacific. Thedirectly off the Strait of Magellan located site MD07-3128 revealstemperature changes that are largely following an “Antarctictiming” as known from ice-core records (e.g. EPICA CommunityMember, 2006) and other marine records offshore Chile (Lamyet al., 2004; Kaiser et al., 2005) and New Zealand (Pahnke et al.,2003). This includes millennial-scale temperature changes duringthe Last Glacial as well as the timing and structure of the Deglacialwarming. Terrestrial records from southern Patagonia partlyconfirm the “Antarctic timing”, e.g. through the existence of glacieradvances during the Antarctic Cold Reversal (e.g. Sugden et al.,2005; Moreno et al., 2009b). In general, millennial-scale tempera-ture changes in Antarctica over the Last Glacial may be consistentlyexplained by the bipolar seesaw concept that suggests an out-of-phase millennial-scale climate pattern between the Northern andSouthern Hemisphere during the Last Glacial (e.g. Stocker, 1998;Stenni et al., 2011). Therefore, changes in ocean currents providea substantial forcing for southern Patagonia. Also for the earlyHolocene warm period, large-scale global circulation changesinvolving seesaw-like SST pattern have been proposed (e.g. Masson

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e2318

et al., 2000) that contributed to the strengthening of the core SWWin southern Patagonia (Lamy et al., 2010). During the late Holocene,warming in the Eastern Tropical Pacific and cooling further southpossibly enhanced the mid- to low-latitude SST gradient resultingin stronger winds at the northern margin of the SWW (Lamy et al.,2010).

7.4. Atmospheric circulation

Atmospheric circulation changes within the SWW have beeninvolved as a key forcing mechanism both on glacialeinterglacialtime-scales as well as during the Glacial and Deglacial millennial-scale climate events (e.g. Toggweiler et al., 2006; Lamy et al.,2007; Toggweiler and Joellen Russell, 2008). Based on a simpli-fied general circulation model, Toggweiler et al. (2006) suggestedthat an equator-ward shifted SWWduring the glacial allowedmorerespired CO2 to accumulate in the deep ocean. During Glacialterminations, the southward moving SWW reduced polar stratifi-cation and enhanced upwelling of deep water masses aroundAntarctica that would then have released large amounts of thestored CO2 to the atmosphere. Though the glacial northward shift ofthe SWW is not clearly confirmed by more sophisticated modelstudies (e.g. Rojas et al., 2009), it is consistent with paleoceano-graphic and palaeoclimatic reconstructions from the SoutheastPacific (e.g. Heusser, 1989; Lamy et al., 1999, 2004, 2007; Morenoet al., 2001). However, most studies of glacial westerlies were sofar based only on records from the northern margin of the SWW.Based on opposing rainfall trends across the northern margin andthe core of the SWW belt during the Holocene, Lamy et al. (2010)suggested that glacial westerlies may resemble more the presentwinter situationwith a northward expansion of westerlywinds andnot necessarily a latitudinal shift of the wind belt as a whole.

Furthermore, the SWW plays a major role for coupling of bothhemispheres despite out-of-phase summer insolation signals. Forthe last deglaciation, Denton et al. (2010) suggested that thecombined influence of the oceanic bipolar seesaw and the south-ward displacement of the SWW allowed the high southern lati-tudes to warm as a result of melting and collapse of NH ice sheetsinto the North Atlantic. These authors further proposed that theITCZ plays a major role in transferring the Northern Hemisphereclimate signals southward to the SWW.

Though it is generally assumed that the ITCZ was located furthersouth during Northern Hemisphere cold phases including the LGM,a reconstruction of the Equatorial Front-ITCZ system off northwestSouth America over glacial-interglacial cycles reveals a northwardITCZ position (Rincón-Martínez et al., 2010). The authors suggestthat this northward position during the Last Glacial may have beencontrolled by enhanced northward advection of cold water masseswith the Humboldt Current system and was restricted to theeastern tropical Pacific. The same study likewise suggestspredominance of La Niña-like conditions during the Glacial,whereas other authors propose that average conditions during LastGlacial were more similar to modern El Niño years (e.g. Koutavaset al., 2002). Based on the modern climatology, more La Niña-likeconditions during the Last Glacial would however imply enhancedwinds in the core SWW over Southern Patagonia and a reductionnorth of ca 45�S (Schneider and Gies, 2004) which is not consistentwith the available paleodata (see 5.1.).

Modelling studies (e.g. Cane, 2005) suggest that the ITCZ waslocated north of its present position and the frequency and strengthof El Niño events were strongly reduced during the early Holocene,consistent with reconstructions based on paleo-data from theeastern tropical Pacific and adjacent subtropical South America (e.g.Moy et al., 2002). Assuming that reduced El Niño events are con-nected with more La Niña-like conditions this would imply

a stronger SWW over Southern Patagonia during the early Holo-cene (Schneider and Gies, 2004) consistent with enhanced rainfallon the western hyperhumid side of the Andes at these latitudes(Lamy et al., 2010). Since ca 8 ka BP, the number of El Niño eventsrecorded in Laguna Pallcacocha in Ecuador (Moy et al., 2002) largelyincreased in particular during the past ca 3.5 ka. This middle andlate Holocene increase in El Niño strength and frequency has beenrelated to orbital-scale seasonal insolation changes close to theEquator (Clement et al., 2000; Cane, 2005).

The frequency and amplitude of El Niño events also varied onsub-orbital time-scales during the Holocene and were linked tochanges in the SWW strength and the TSI (see Chapter 7.2.). Sevenphases of increased and decreased strength of the SWW core havebeen detected in high-resolution Patagonian stalagmite and lakesediment records from latitudes 52�300 to 54�300S. Phases of lowsun activity are related to low SWW strength and increased ElNiño situations and vice versa. Rein et al. (2005) suggested forexample a pronounced reduction of El Niño events during theMCA consistent with higher rainfall and thus a stronger SWW asrecorded in both records (Fig. 11). A reduced precipitation andSWW strength occurs especially between 2.5 and 3.5 ka BP andfrom 0.7 to 0.1 ka BP. when increase El Niño events have beenrecorded at the Peruvian coast (Rein et al., 2005) and in theLaguna Pallcacocha record (Fig. 11), again consistent with themodern El NiñoeSWW relationship in the Southern Patagonia(Schneider and Gies, 2004).

Taken together, atmospheric circulation changes within theSWW are the major forcing factor for paleoclimate changes inSouthern Patagonia. Past SWW changes are strongly linked tochanges in the tropical climate system especially latitudinal shiftsof the ITCZ and changes in the ENSO system. During the Holocene,latitudinal shifts of the ITCZ and related SWW changes werecontrolled by millennium-scale Northern Hemispheric insolationchanges as well as centennial scale changes of sun activity (TSI).

8. Conclusions

The southern tip of South America represents a key area for theunderstanding of global paleoclimate and therefore more than 150paleoclimate reconstructions have been published only during thelast 5 years. These records give new implications concerningpaleotemperature as well as changes in SSW strength and/or lat-itudinal shifts. However, when comparing these records manyinconsistencies or partly opposite interpretations appear which arerelated to restricted proxy understanding and/or unavailable cali-bration. This concerns also the individual sensitivity and thresholdof each proxy with respect to climate change and an often non-linear proxy-climate relationship. Some proxies are highly sensi-tive to even short term changes or occasional climate eventswhereas others have high natural variability often unrelated toclimate or do not reach a threshold for registering paleoclimatefluctuations of a given amplitude.

The few existing full glacial records indicate that the PatagonianIce Sheet experienced significant oscillations between 60 and 12 kaBP which have been primarily controlled by paleotemperatures andthe intensity and latitudinal position of the SWW. The paleo-temperatures of this period are clearly related to those recon-structed from Antarctic Ice cores. The Deglacial also followsAntarctic patternwith an Antarctic Cold Reversal and no significantcooling during the Younger Dryas. During the Holocene mostglacier advances occurred after 5.5 ka BP and were probablyprimarily driven by short term (a few hundreds of years) increasedSWW strength and related increased precipitation/accumulation.Only the LIA advances likely controlled by reduced ablation andthus temperature.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 19

SSW air masses circulate within few weeks around the globeand have a clear oceanic control. Therefore Holocene temperaturesare best represented by oceanic SST records which show similaroverall pattern at various sites within the northern ACC and SWWbelt. These are characterised by an Early Holocene thermalmaximum (11.5e8.5 ka BP) and relatively constant 1.5 �C lowertemperatures during the last 8 ka BP. Therefore regional terrestrialtemperature anomalies in southernmost Patagonia may be veryweak. Southern Hemisphere insolation has only a minor or noinfluence on this pattern, but periods of low sun activity between3.5 and 2.5 ka BP and especially the LIA are characterised bya significant drop in temperature up to 1.5 �C in the fjord SSTrecords and in terrestrial paleotemperature reconstructions.

Different reconstructions of the SWW strength and relatedprecipitation show an inconsistent or partly opposite behaviour.Some records indicate a weak SWW core in the early Holocene anda strengthening towards the Late Holocene, whereas others showthe reverse pattern or do not reveal a clear trend. These inconsis-tencies may partly be related to unclear relations between precip-itation and SWW strength. Furthermore, restricted knowledge ofthe seasonally variable wind- and temperature-related effect ofevaporation on the available moisture in soils and/or the generalwater balance might have let to conflicting results. Resolving theseinconsistencies in southernmost Patagonian SWW records isa prerequisite for improving hemispheric comparisons with e.g.New Zealand records (e.g. Fletcher and Moreno, 2012) and links toatmospheric CO2 changes (Hodgson and Sime, 2010).

For the last 5 ka BP high resolution records show clear linkageson a decadal and centennial scale between SWW strength (at thesouthern margin of its present-day core at latitudes 52�300 to54�300S) and the behaviour of ENSO, the ITCZ and sun activity. Lowsun activity is related to weak and/or northward displaced SWWand to more frequent El Niño events as well as a southward dis-placed and/or weaker ITCZ.

A denser grid of sediment core records is required to bettercover the whole precipitation gradient across the Andean climatedivide in southernmost Patagonia. In this respect systematic tran-sects across the climate gradient with comparable archives andproxies would be desirable. These studies should be accompaniedby more extensive long-term climate and ocean monitoring,particularly in the western fjord region.

Acknowledgements

We thank Helge Arz, Rene Garreaud, Jeróme Kaiser, ChristopherMoy, Michael Mayr, Nicolas Waldmann, Luis Vieira, and RodrigoVilla-Martínez for sharing data and discussions. Jean Pierre Francoisprovided very valuable comments on pollen issues. Oscar Baeza andFrancisco Rios gave important technical support. We furtheracknowledge the constructive reviews by Robert McCulloch and ananonymous reviewer that helped to improve the manuscript. TheGerman Research Foundation is thanked for research grants Ki 456/8 to Ki 456/12, LA 1273/3-2, LA1273/5-1, and LA1273/7-1.

References

Ackert Jr., R.P., Becker, R.A., Singer, B.S., Kurz, M.D., Caffee, M.W., Mickelson, D.M.,2008. Patagonian glacier response during the Late GlacialeHolocene transition.Science 321 (5887), 392e395.

Anderson, D.M., Archer, R.B., 1999. Preliminary evidence of early deglaciation insouthern Chile. Palaeogeography, Palaeoclimatology, Palaeoecology 146 (1e4),295e301.

Anderson, B., MacKintosh, A., Stumm, D., George, L., Kerr, T., Winter-Billington, A.,Fitzsimons, S., 2010. Climate sensitivity of a high-precipitation glacier in NewZealand. Journal of Glaciology 56, 114e128.

Aniya, M., 1995. Holocene glacial chronology in Patagonia -Tyndall and Upsalaglaciers. Arctic and Alpine Research 27 (4), 311e322.

Appleby, P.G., Oldfield, F., 1992. Application of lead-210 to sedimentation studies.Uranium series disequilibrium: applications to environmental problems. In:Ivanovich, M., Harmon, R.S. (Eds.), Uranium Series Disequilibrium: Applicationsto Environmental Problems. Clarendon Press, Oxford, pp. 731e778.

Aravena, J.-C., 2007. Reconstructing climate variability using tree rings and glacierfluctuations in southernmost Chilean Andes. PhD thesis, University of WesternOntario, London, Ontario, Canada.

Aravena, J.C., Luckman, B.H., 2009. Spatio-temporal rainfall patterns in southernsouth America. International Journal of Climatology 29, 2106e2120.

Aravena, J.C., Lara, A., Wolodarsky-Franke, A., Villalba, R., Cuq, E., 2002. Tree-ringgroth patterns and temeparture reconstruction from Nathofagus pumilio(Fagaceae) en el límete arbores superior de la Patagonia austral de Chile. RevistaChilena de Historia Natural 75, 361e375.

Ariztegui, D., Gilli, A., Anselmetti, F.S., Goni, R.A., Belardi, J.B., Espinosa, S., 2010.Lake-level changes in central Patagonia (Argentina): crossing environmentalthresholds for Lateglacial and Holocene human occupation. Journal of Quater-nary Science 25 (7), 1092e1099.

Armesto, J.J., Casassa, I., Dollenz, O., 1992. Age structure and dynamics of Patagonianbeech forests in Torres del Paine National Park, Chile. Vegetatio 98, 13e22.

Ashworth, A.C.,Markgraf, V., Villagran, C.,1991. LateQuaternaryclimatic historyof theChilean Channels based on fossil pollen and beetle analysis, with an analysis ofthemodernvegetation andpollen rain. Journal of Quaternary Science 6, 279e291.

Auer, V., 1933. Verschiebungen der Wald-und Steppengebiete Feuerlands in post-glazialer Zeit. Acta Geographica 5, 1e313.

Auer, V., 1958. The Pleistocene of Fuego-Patagonia. Part II. The history of the floraand vegetation. Annales Academiae Scientiarum Fennicae III. Geologica-Geo-graphica 50, 1e239.

Auer, V., 1960. The Quaternary history of Fuego-Patagonia. Proceedings of the RoyalSociety of London. Series B, Biological Sciences. A Discussion on the Biology ofthe Southern Cold Temperate Zone 152 (949), 507e516.

Auer, V., 1974. The isorhythmicity subsequent to Fuego-Patagonian and Fenno-scandian ocean level transgressions and regressions of the latest glaciation.Annales Academiae Scientiarum Fennicae III. Geologica-Geographica 115, 1e88.

Baeza, O., 2005. Lake and fjord sediments as Late Glacial to Holocene environmentaland climate archives of the southernmost Andes at 53�S, Chile. Ph.D. thesis,Albert-Ludwigs-Universität Freiburg, Germany.

Ballantyne, C.K., 2002. Paraglacial geomorphology. Quaternary Science Reviews 21,1935e2017.

Barker, S., Diz, P., Vautravers, M.J., Pike, J., Knorr, G., Hall, I.R., Broecker, W.S., 2009.Interhemispheric Atlantic seesaw response during the last deglaciation. Nature457 (7233), 1097e1102.

Bennett, K.D., Haberle, S.G., Lumley, S.H., 2000. The Last GlacialeHolocene transi-tion in southern Chile. Science 290, 325e328.

Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10 millionyears. Quaternary Science Reviews 10, 297e317.

Bianchi, C., Gersonde, R., 2004. Climate evolution at the last deglaciation: the roleof the southern ocean. Earth and Planetary Science Letters 228 (3e4),407e424.

Biester, H., Kilian, R., Franzen, C., Woda, C., Mangini, A., Schöler, H.F., 2002.Elevated mercury accumulation in a peat bog of the Magellanic Moorlands,Chile (53�S) e an anthropogenic signal from the Southern Hemisphere. Earthand Planetary Science Letters 201 (3e4), 609e620.

Biester, H., Martinez-Cortizas, A., Birkenstock, S., Kilian, R., 2003. Effect of peatdecomposition and mass loss on historic mercury records in peat bogs fromPatagonia. Environmental Science Technology 37 (1), 32e39.

Biester, H., Selimovíc, D., Hemmerich, S., Petri, M., 2006. Halogens in pore water ofpeat bogs e the role of peat decomposition and dissolved organic matter.Biogeosciences 3, 53e64.

Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M.N., Showers, W., Hoffmann, S.,Rusty Lotti-Bond, R., Hajdas, I., Bonani, G., 2001. Persistent solar influence onnorth Atlantic climate during the Holocene. Science 294 (5549), 2130e2136.

Borrero, L.A., 1999. Human dispersal and climatic conditions during Late Pleistocenetimes in Fuego- Patagonia. Quaternary International 53/54, 93e99.

Boucher, É., Guiot, J., Chapron, E., 2011. A millennial multi-proxy reconstruction ofsummer PDSI for Southern South America. Climate of the Past 7, 957e974.http://dx.doi.org/10.5194/cp-7-957-2011.

Boyd, B., Anderson, J., Wellner, J., Fernández, R., 2008. The sedimentary record ofglacial retreat, Marinelli Fjord, Patagonia: regional correlations and climate ties.Marine Geology 255 (3e4), 165e178.

Brambati, 2000. Paleoclimatic and Paleoenvironmental Records in Sediments fromthe Southern Ocean (Strait of Magellan and Ross Sea), Terra Antarctica Reports4, pp. 1e41.

Breuer, S., Kilian, R., Baeza, O., Lamy, F., Arz H. Holocene denudation rates from thesuperhumid southernmost Patagonian Andes (53�S) deduced from lake sedi-ment budgets. Geomorphology, in press-a.

Breuer, S., Kilian, R., Weinrebe, W. Schörner, D., Behrmann, J. Glacier and tectoniccontrol on fjord morphology and sediment deposition in the Magellan region(53�S). Marine Geology, in press-b.

Caldenius, C., 1932. Las glaciaciones cuaternarias de la Patagonia y Tierra del Fuego.Geografiska Annaler 14, 1e164.

Cane, M.A., 2005. The evolution of El Niño, past and future. Earth and PlanetaryScience Letters 230 (3e4), 227e240.

Caniupán, M., Lamy, F., Lange, C.B., Kaiser, J., Arz, H.W., Kilian, R., Baeza-Urrea, O.,Aracena, C., Hebbeln, D., Kissel, C., Laj, C., Mollenhauer, G., Tiedemann, R., 2011.Millennial-scale sea surface temperature and Patagonian Ice Sheet changes off

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e2320

southernmost Chile (53�S) over the past w60 kyr. Paleoceanography 26(PA3221), 1e10.

Carrasco, J.F., Osorio, R., Casassa, G., 2008. Secular trend of the equilibrium-linealtitude on the western side of the southern Andes, derived from radiosondeand surface observations. Journal of Glaciology 54 (186), 538e550.

Casassa, G., Rivera, A., Aniya, M., Naruse, R., 2000. Características glaciológicas delCampo de Hielo Patagónico Sur. Anales del Instituto de la Patagonia, SerieCiencias Naturales 28, 5e22.

Chaigneau, A., Pizarro, O., 2005. Surface circulation and fronts of the South PacificOcean, east of 120� W. Geophysical Research Letters 32 (L08605).

Chambers, F., Mauquoy, D., Brain, S., Blaauw, M., Daniell, J., 2007. Globallysynchronous climate change 2800 years ago: proxy data from peat in SouthAmerica. Earth and Planetary Science Letters 253 (3e4), 439e444.

Clapperton, C.M., Sugden, D.E., Kauffman, D., McCulloch, R.D., 1995. The last glaci-ation in central Magellan Strait, southernmost Chile. Quaternary Research 44,133e148.

Clement, A.C., Seager, R., Cane, M.A., 2000. Suppression of El Niño during theMid-Holocene by changes in the Earth’s orbit. Paleoceanography 15 (6),731e737.

Cusminsky, G., Schwalb, A., Pérez, A.P., Pineda, D., Viehberg, F., Whattley, R.,Markgraf, V., Gilli, A., Aritztegui, D., Anselmetti, F.S., 2011. Late Quarternaryenvironmental changes in Patagonia as inferred from lacustrine fossil andextant ostracods. Biological Journal of the Linnean Society 103, 397e408.

Denton, G.H., Anderson, R.F., Toggweiler, J.R., Edwards, R.L., Schaefer, J.M.,Putnam, A.E., 2010. The Last Glacial termination. Science 328 (5986),1652e1656.

Douglass, D.C., Singer, B.S., Kaplan, M.R., Ackert, R.P., Mickelson, D.M., Caffee, M.W.,2005. Evidence of early Holocene glacial advances in southern South Americafrom cosmogenic surface-exposure dating. Geology 33, 237e240.

Endlicher, W., 1991. Patagonien e klima- und agrarökologische Probleme an derMagellanstraße. Geographische Rundschau 43, 143e151.

Endlicher, W., Santana, A., 1988. El clima del sur de la Patagonia y sus aspectosecológicos. Un siglo de Mediciones Climatológicas en Punta Arenas. AnalesInstituto Patagonia, Serie Ciencias Naturales 18, 57e86.

EPICA community members, 2006. One-to-one coupling of glacial climate vari-ability in Greenland and Antarctica. Nature 444, 195e198.

Fesq-Martin, M., Friedmann, A., Peters, M., Behrmann, J., Kilian, R., 2004. Late-glacialand Holocene vegetation history of the Magellanic rain forest in southwesternPatagonia, Chile. Vegetation History and Archaeobotany 13 (4), 249e255.

Fey, M., Korr, C., Maidana, N.L., Carrevedo, M.L., Corbello, H., Dietrich, S.,Haberzettl, T., Kuhn, G., Lücke, A., Mayr, C., Ohlendorf, C., Paez, M.A.,Quintana, F.A., Schäbitz, F., Zolitschka, B., 2009. Paleoenvironmental changesduring the last 1600 years inferred from the sediment record of a cirque lake inSouthern Patagonia (Laguna Viscachas, Argentina. Paleogeography, Paleocli-matology, Paleoecology 281 (3e4), 363e375.

Fischer, H., Fundel, F., Huybrechts, P., Rybak, O., Oerter, H., Twarloh, B., Wegener, A.,Ruth, U., Udisti, R., Becagli, S., Castellano, E., Fattori, I., Severi, M., Wolff, E.,Littot, G., Röthlisberger, R., Mulvaney, R., Hutterli, M., Kaufmann, P., Federer, U.,Lambert, F., Bigler, M., Stauffer, B., Hansson, M., Jonsell, U., de Angelis, M.,Gabrielli, P., Boutron, C., Siggaard-Andersen, M.-L., Steffensen, J.P., Barbante, C.,Gaspari, V., Wagenbach, D., 2007. Reconstruction of millennial changes in dustemission, transport and regional sea ice coverage using the deep EPICA ice coresfrom the Atlantic and Indian Ocean sector of Antarctica. Earth and PlanetaryScience Letters 260, 340e354.

Fletcher, M.S., Moreno, P.I., 2011. Zonally symmetric changes in the strength andposition of the Southern Westerlies drove atmospheric CO2 variations over thepast 14 k.y. Geology 39 (5), 419e422.

Fletcher, M.S., Moreno, P.I., 2012. Have the Southern Westerlies changed in a zonallysymmetric manner over the last 14,000 years? A hemisphere-wide take ona controversial problem. Quaternary International 253, 32e46.

Garreaud, R., 2007. Precipitation and circulation covariability in the extratropics.Journal of Climate 20 (18), 4789e4797.

Garreaud, R., Lopez, P., Minvielle, M., Rojas, M. Large scale control on the PatagoniaClimate. Journal of Climate, in press.

Gilli, A., Ariztegui, D., Anselmetti, F., Mckenzie, J., Markgraf, V., Hajdas, I.,Mcculloch, R., 2005. Mid-Holocene strengthening of the Southern Westerlies inSouth America e Sedimentological evidences from Lago Cardiel, Argentina(49�S). Global and Planetary Change 49 (1e2), 75e93.

Glasser, N.F., Harrison, S., Winchester, V., Aniya, M., 2004. Late Pleistocene andHolocene palaeoclimate and glacier fluctuations in Patagonia. Global andPlanetary Change 43 (1e2), 79e101.

Glasser, N.F., Harrison, S., Jansson, K.N., Anderson, K., Cowley, A., 2011. Global sea-level contribution from the Patagonia Icefields since the Little Ice Agemaximum. Nature Geoscience 4 (5), 303e307.

Habelzettl, T., Corbella, H., Fey, M., Janssen, S., Lücke, A., Mayr, C., Ohlendorf, C.,Schäbitz, F., Schleser, G.H., Wille, M., Wulf, S., Zolitschka, B., 2007. Late glacialand Holocene wetedry cycles in southern Patagonia: chronology, sedimen-tology and geochemistry of a lacustrine record from Laguna Potrok Aike,Argentina. The Holocene 17 (3), 297e310.

Haberzettl, T., Anselmetti, F.S., Bowen, S.W., Fey, M., Mayr, C., Zolitschka, B.,Ariztegui, D., Mauz, B., Ohlendorf, C., Kastner, S., Lücke, A., Schäbitz, F., Wille, M.,2009. Late Pleistocene dust deposition in the Patagonian steppe e extendingand refining the paleoenvironmental and tephrochronological record fromLaguna Potrok Aike back to 55 ka. Quaternary Science Reviews 28 (25e26),2927e2939.

Hall, B.L., 2009. Holocene glacial history of Antarctica and the sub-Antarctic islands.Quaternary Science Reviews 28, 2213e2230.

Haug, G.H., Hughen, K.A., Sigman, D.M., Peterson, L.C., Röhl, U., 2001. Southwardmigration of the Intertropical Convergence Zone through the Holocene. Science293, 1304e1308.

Hays, J.D., Imbrie, J., Shackleton, N.J., 1976. Variations in the Earth’s orbit: pacemakerof the Ice Ages. Science 194 (4270), 1121e1132.

Hermanns, Y.-M., Biester, H., 2011. A Holocene record of mercury accumulation ina pristine lake in Sothernmost South America (53�S) e climatic and environ-mental drivers. Biogeosciences Discussions 8 (4), 6555e6588.

Heusser, C.J., 1971. Pollen and Spores of Chile, Modern Types of the Pteridophyta,Gymnospermae, and Angiospermae. The University of Arizona Press, USA,176 pp.

Heusser, C.J., 1989. Late Quaternary vegetation and climate of southern Tierra delFuego. Quaternary Research 31 (3), 396e406.

Heusser, C.J., 1994. Paleoindians and fire during the late Quaternary in southernSouth America. Revista Chilena de Historia Natural 67 (4), 435e443.

Heusser, C.J., 1995. Three Late Quaternary pollen diagrams from southern Patagoniaand their paleoecological implications. Palaeogeography, Palaeoclimatology,Palaeoecology 118 (1e2), 1e24.

Heusser, C.J., 2003. Ice Age Southern Andes e a Chronicle of Paleoecological Events.In: Developments in Quaternary Science, vol. 3. Elsevier, Amsterdam.

Heusser, C., Heusser, E., Lowell, T., Moreira, A., Moreira, S., 2000. Deglacial palae-oclimate at Puerto del Hambre, subantarctic Patagonia, Chile. Journal ofQuaternary Science 15 (2), 101e114.

Hodell, D.A., Kanfoush, S.L., Shemesh, A., Crosta, X., Charles, C.D., Guilderson, T.P.,2001. Abrupt cooling of Antarctic surface waters and sea ice expansion in theSouth Atlantic sector of the Southern Ocean at 5000 cal yr B.P. QuaternaryResearch 56, 191e198.

Hodgson, A., Sime, L.C., 2010. Southern westerlies and CO2. Nature Geoscience 3,66e667.

Huber, U.M., Markgraf, V., Schäbitz, F., 2004. Geographical and temporal trends inlLate Quaternary pre histories of Fuego-Patagonia, South America. QuaternaryScience Reviews 23 (9e10), 1079e1097.

Hulton, N.R.J., Purves, R.S., Mcculloch, R.D., Sugden, D.E., Bentley, M.J., 2002. TheLast Glacial maximum and deglaciation in southern South America. QuaternaryScience Reviews 21 (1e3), 233e241.

Imbrie, J., Boyle, E.A., Clemens, S.C., Duffy, A., Howard, W.R., Kukla, G., Kutzbach, J.,Martinson, D.G., McIntyre, A., Mix, A.C., Molfino, B., Morley, J.J., Peterson, L.C.,Pisias, N.G., Prell, W.L., Raymo, M.E., Shackleton, N.J., Toggweilerl, J.R., 1992. Onthe structure and origin of major glaciation cycles: 1. Linear responses toMilankovic forcing. Paleoceanography 7 (6), 701e738.

Ingram, B.L., Southon, J.R., 1996. Reservoir ages in eastern Pacific coastal andestuarine waters. Radiocarbon 38 (3), 573e582.

Jouzel, J., Masson-Delmotte, V., Cattani, O., Dreyfus, G., Falourd, S.,Hoffmann, G., Minster, B., Nouet, J., Barnola, J.M., Chappellaz, J., Fischer, H.,Gallet, J.C., Johnsen, S., Leuenberger, M., Loulergue, L., Luethi, D., Oerter, H.,Parrenin, F., Raisbeck, G., Raynaud, D., Schilt, A., Schwander, J., Selmo, E.,Souchez, R., Spahni, R., Stauffer, B., Steffensen, J.P., Stenni, B., Stocker, T.F.,Tison, J.L., Werner, M., Wolff, E.W., 2007. Orbital and millennial Antarcticclimate variability over the past 800,000 years. Science 317 (5839),793e796.

Kaiser, J., Lamy, F., 2010. Links between Patagonian Ice Sheet fluctuations andAntarctic dust variability during the Last Glacial period (MIS 4-2). QuaternaryScience Reviews 29 (11e12), 1464e1471.

Kaiser, J., Lamy, F., Hebbeln, D., 2005. A 70-kyr sea surface temperature record offsouthern Chile (Ocean Drilling Program Site 1233). Paleoceanography 20,PA4009. http://dx.doi.org/10.1029/2005PA001146.

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. Bulletinof the American Meteorological Society 77 (3), 437e470.

Kaplan, M.R., Fogwill, C.J., Sugden, D.E., Hulton, N.R.J., Kubik, P.W., Freeman, S.P.H.T.,2008. Southern Patagonian glacial chronology for the Last Glacial period andimplications for Southern Ocean climate. Quaternary Science Reviews 27 (3e4),284e294.

Kaplan, M.R., Strelin, J.A., Schaefer, J.M., Denton, G.H., Finkel, R.C., Schwartz, R.,Putnam, A.E., Vandergoes, M.J., Goehring, B.M., Travis, S.G., 2011. In-situ cosmo-genic 10Be production rate at Lago Argentino, Patagonia: implications for late-glacial climate chronology. Earth and Planetary Science Letters 309 (1e2), 21e32.

Kastner, S., Ohlendorf, C., Haberzettl, T., Lücke, A., Mayr, C., Maidana, N.I.,Schäbitz, F., Zolitschka, B., 2010. Southern hemispheric westerlies controlspatial distribution of modern sediments in Laguna Potrok Aike, Argentina.Journal of Paleolimnology 44, 887e902.

Kilian, R., Hohner, M., Biester, H., Wallrabe-Adams, H., Stern, Ch, 2003. Holocenepeat and lake sediment tephra record from the southernmost Chilean Andes(53-55�S). Revista Geológica de Chile 30 (1), 47e64.

Kilian, R., Biester, H., Behrmann, J., Baeza, O., Fesq-Martin, M., Hohner, M.,Schimpf, D., Friedmann, A., Mangini, A., 2006. Millennium-scale volcanic impacton a superhumid and pristine ecosystem. Geology 34 (8), 609e612.

Kilian, R., Baeza, O., Steinke, T., Arévalo, M., Ríos, C., Schneider, C., 2007a. LatePleistocene to Holocene marine transgression and thermohaline control onsediment transport in the western Magellanes fjord system of Chile (53�S).Quaternary International 161 (1), 90e107.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 21

Kilian, R., Schneider, C., Koch, J., Fesq-Martin, M., Biester, H., Casassa, G., Arévalo, M.,Wendt, G., Baeza, O., Behrmann, J., 2007b. Palaeoecological constraint on LateGlacial and Holocene ice retreat in the Southern Andes (53�S). Global andPlanetary Change 59 (1e4), 49e66.

Kissel, C., Leau, H., The shipboard Scientific party, 2007. MD159-PACHIDERME(IMAGES XV) Cruise Report. Ref OCE2007/0, 2007. In: Série “Les rapports decampagnes à la Mer, IPEV”.

Kleinebecker, T., Hölzel, N., Vogel, A., 2007. Gradients of continentallty and moisturein South Patagonian ombrotrophic peatland vegetation. Folia Geobotanica 42,363e382.

Koch, J., Kilian, R., 2002. Dendroecological potencial of common tree species alonga transect across the southernmost Andes, Chile (53�). Anales Instituto Pata-gonia, Serie Ciencias Naturales 30, 123e132.

Koch, J., Kilian, R., 2005. Little Ice Age glacier fluctuations, Gran Campo Nevado,southernmost Chile. The Holocene 15 (1), 20e28.

Koutavas, A., Lynch-Stieglitz, J., Marchitto, T.M., Sachs, J., 2002. El Niño-like patternin Ice Age tropical Pacific sea surface temperature. Science 297, 226e230.

Kuylenstierna, J.L., Rosqvist, G.C., Holmlund, P., 1996. Late-Holocene glacier varia-tions in the Cordillera Darwin, Tierra del Fuego. Holocene 6 (3), 353e358.

Lamy, F., Hebbeln, D., Wefer, G., 1999. Highz resolution marine record of climatechange in mid-latitude Chile during the last 28,000 years based on terrigenoussediment vparameters. Quaternary Research 51, 83e93.

Lamy, F., Rühlemann, C., Dierk Hebbeln, D., Wefer, G., 2002. High and low latitudeclimate control on the position of the southern Peru-Chile Current during theHolocene. Paleoceanography 17 (2). http://dx.doi.org/10.1029/2001PA000727.

Lamy, F., Kaiser, J., Arz, H., Ninnemann, U., Hebbeln, D., 2004. Antarctic timing ofsurface water changes off Chile and Patagonian Ice Sheet response. Science 304(5679), 1959e1962.

Lamy, F., Kaiser, J., Arz, H., Ninnemann, U., Hebbeln, D., Timm, O., Timmermann, A.,Toggweiler, J., 2007. Modulation of the bipolar seesaw in the southeast pacificduring Termination 1. Earth and Planetary Science Letters 259 (3e4), 400e413.

Lamy, F., Kilian, R., Arz, H., Francois, J., Kaiser, J., Prange, M., Steinke, T., 2010.Holocene changes in the position and intensity of the southern westerly windbelt. Nature Geoscience 3 (10), 695e699.

Legoupil, D., 2009. Cerro Castillo rockshelter occupation: a residential site ina panoramic viewing-point. Magallania 37 (1), 47e60.

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 Antarcticand Greenland ice cores. Quaternary Science Reviews 29 (1e2), 8e20.

Lisé-Pronovost, Agathe, St-Onge, G., Gogorza, C., Haberzettl, T., Preda, M., Kliem, P.,Francus, P., Zolitschka, B., The PASADO Science Team, 2012. High-resolutionpaleomagnetic secular variations and relative paleointensity since the LatePleistocene in southern South America. Quaternary Science Reviews. http://dx.doi.org/10.1016/j.quascirev.2012.05.012.

Liu, Z., Brady, E., Lynch-Stieglitz, J., 2003. Global ocean response to orbital forcing inthe Holocene. Paleoceanography 18 (2), 1041. http://dx.doi.org/10.1029/2002PA000819.

Magny, M., Haas, J.N., 2004. A major widespread climatic change around 5300 cal.yr BP at the time of the Alpine Iceman. Journal of Quaternary Science 19 (5),423e430.

Mancini, M.V., 2009. Holocene vegetation and climate changes from a peat pollenrecord of the forest-steppe ecotone, Southwest of Patagonia (Argentina).Quaternary Science Reviews 28 (15e16), 1490e1497.

Mancini, M., Paéz, M., Prieto, A., Stutz, S., Tonello, M., Vilanova, I., 2005. Mid-Holocene climatic variability reconstruction from pollen records (32e52�S,Argentina. Quaternary International 132 (1), 47e59.

Marinoni, L., Setti, M., Salvi, C., López-Galindo, A., 2008. Clay minerals in lateQuaternary sediments from the south Chilean margin as indicators of prove-nance and palaeoclimate. Clay Minerals 43, 235e253.

Markgraf, V., 1993. Paleoenvironments and paleoclimates in Tierra del Fuego andsouthernmost Patagonia, South America. Palaeogeography, Palaeoclimatology,Palaeoecology 102 (1e2), 53e68.

Markgraf, V., Huber, U.M., 2010. Late and postglacial vegetation and fire history inSouthern Patagonia and Tierra del Fuego. Palaeogeography, Palaeoclimatology,Palaeoecology 297 (2), 351e366.

Markgraf, V., D’Antoni, H., Ager, T., 1981. Modern pollen dispersal in Argentina.Palynology 5, 235e254.

Markgraf, V., Bradbury, J.P., Schwalb, A., Burns, S.J., Stern, C., Ariztegui, D., Gilli, A.,Anselmetti, F.S., Stine, S., Maidana, N., 2003. Holocene palaeoclimates ofsouthern Patagonia: limnological and environmental history of Lago Cardiel,Argentina. The Holocene 13 (4), 581e591.

Marshall, J., Speer, K., 2012. Closure of the meridional overturning circulationthrough Southern Ocean upwelling. Nature Geoscience. http://dx.doi.org/10.1038/ngeo1391.

Masiokas, M.H., Rivera, A., Espizua, L.E., Villalba, R., Delgado, S., Aravena, J.C., 2009a.Glacier fluctuations in extratropical South America during the past 1000 years.Palaeogeography, Palaeoclimatology, Palaeoecology 281 (3e4), 242e268.

Masiokas, M., Luckman, B., Villalba, R., Delgado, S., Skvarca, P., Ripalta, A., 2009b.Little Ice Age fluctuations of small glaciers in Monte Fitz Roy and Lago delDesierto areas, south Patagonian Andes, Argentina. Palaeogeography, Palae-oclimatology, Palaeoecology 281, 351e362.

Masson, V., Vimeux, F., Jouzel, J., Morgan, V., Delmotte, M., Ciais, P., Hammer, C.,Johnsen, S., Lipenkov, V.Y., Mosley-Thompson, E., Petit, J.-R., Steig, E.J.,Stievenard, M., Vaikmae, R., 2000. Holocene climate variability in Antarcticabased on 11 ice-core isotopic records. Quaternary Research 54 (3), 348e358.

Mayr, C., Wille, M., Haberzettl, T., Fey, M., Janssen, S., L}ucke, A., Ohlendorf, C.,Oliva, G., Schäbitz, F., Schleser, G.H., Zolitschka, B., 2007a. Holocene variability ofthe southern hemisphere westerlies in Argentinean Patagonia (52�S). Quater-nary Science Reviews 26 (5e6), 579e584.

Mayr, C., Lücke, A., Stichler, W., Trimborn, P., Ercolano, B., Oliva, G., Ohlendorf, C.,Soto, J., Fey, M., Haberzettl, T., 2007b. Precipitation origin and evaporation oflakes in semi-arid Patagonia (Argentina) inferred from stable isotopes (d18O,d2H). Journal of Hydrology 334, 53e63.

Mayr, C., Lücke, A., Maidana, N.I., Wille, M., Haberzettl, T., Corbella, H., Ohlendorf, C.,Schäbitz, F., Fey, M., Janssen, S., Zolitschka, B., 2009. Isotopic fingerprints onlacustrine organic matter from Laguna Potrok Aike (southern Patagonia,Argentina) reflect environmental changes during the last 16,000 years. Journalof Paleolimnology 42 (1), 81e102.

McCormac, F.G., Hogg, A.G., Blackwell, P.G., Buck, C.E., Higham, T.F.G., Reimer, P.J.,2004. SHCal04 Southern Hemisphere calibration 0e11.0 cal Kyr BP. Radiocarbon46 (3), 1087e1092.

McCulloch, R., Bentley, M.J., 1998. Late-glacial ice advances in the Strait of Magellan,southern Chile. Quaternary Science Reviews 17, 775e787.

McCulloch, R., Davis, S., 2001. Late-glacial and Holocene palaeoenvironmentalchange in the central Strait of Magellan, southern Patagonia. Palaeogeography,Palaeoclimatology, Palaeoecology 173 (3e4), 143e173.

McCulloch, R., Morello, F., 2009. Evidencia glacial y paleoecológica de ambientestardiglaciales y del Holoceno temprano. Implicaciones para el poblamientotemprano de Tierra del Fuego. In: Salemme, M., Santiago, F., Àlvarez, M.,Piana, E., Vázquez, M., Mansur, M.E. (Eds.), Arqueología de Patagonia: unamirada desde el ultimo confin. Editorial Utopías, Ushuaia, pp. 119e136.

McCulloch, R., Bentley, M., Purves, S., Hulton, N., Sudgen, R., Clapperton, Ch, 2000.Climatic inferences from glacial and palaeoecological evidence at the LastGlacial termination, southern South America. Journal of Quaternary Science 15(4), 409e417.

McCulloch, R., Fogwill, C., Sugden, D., Bentley, M., Kubik, P., 2005. Chronology of thelast glaciation in central Strait of Magellan and Bahía Inútil, southernmostSouth America. Geografiska Annaler: Series A, Physical Geography 87 (2),289e312.

McGlone, M.S., Turney, C.S.M., Wilmshurst, J.M., Renwick, J., Pahnke, K., 2010.Divergent trends in land and ocean temperature in the Southern Ocean over thepast 18,000 years. Nature Geoscience 3 (9), 622e626.

Mercer, J.H., 1965. Glacier variations in southern Patagonia. Geogr. Rev. 55, 390e413.Mercer, J.H., 1982. Holocene glacial variations in southern South America. Striae 18,

35e40.Möller, M., Schneider, C., 2010. Calibration of glacier volumeearea relations from

surface extent fluctuations and application to future glacier change. Journal ofGlaciology 56 (195), 33e40.

Moore, D.M., 1979. Southern oceanic wet-heathlands (including Magellanic Moor-land). In: Specht, R.L. (Ed.), Heathlands and Related Shrublands, Ecosystems ofthe World 9A. Elsevier, Amsterdam, pp. 489e497.

Moore, B.M., 1983. Flora of Tierra del Fuego. Anthony Nelson, Oswesty.Morello, F., Borrero, L., Massone, M., Stern, C., García-Herbst, A., McCulloch, R.,

Arroyo-Kalin, M., Calás, E., Torres, J., Prieto, A., Martinez, I., Bahamonde, G.,cárdenas, P., 2012. Hunter-gatherers, biogeographic barriers and the develop-ment of human settlement in Tiera del Fuego. Antiquity 86, 71e87.

Moreno, P.I., Lowell, T.V., Jacobson Jr., G.L., Denton, G.H., 1999. Abrupt vegetationand climate changes during the Last Glacial Maximum and last termination inthe Chilean Lake District: a case study from Canal de la Puntilla (41�S). Geo-grafiska Annaler: Series A 81, 285e311.

Moreno, P.I., Jacobson, G.L., Lowell, T.V., Denton, G.H., 2001. Interhemisphericclimate links revealed by a late-glacial cooling episode in southern Chile. Nature409, 804e808. http://dx.doi.org/10.1038/35057252.

Moreno, P.I., François, J.P., Villa-Martínez, R.P., Moy, C.M., 2009a. Millennial-scalevariability in Southern Hemisphere westerly wind activity over the last 5000years in SW Patagonia. Quaternary Science Reviews 28 (1e2), 25e38.

Moreno, P.I., Kaplan, M.R., Francois, J.P., Villa-Martinez, R., Moy, C.M., Stern, C.R.,Kubik, P.W., 2009b. Renewed glacial activity during the Antarctic cold reversaland persistence of cold conditions until 11.5 ka in Southwestern Patagonia.Geology 37 (4), 375e378.

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 (8),727e730.

Moy, C.M., Seltzer, G.O., Rodbell, D.T., Anderson, D.M., 2002. Variability of El Nino/Southern oscillation activity at millennial timescales during the Holoceneepoch. Nature 420, 162e165.

Moy, C.M., Dunbar, R.B., Moreno, P., Francois, J.P., Villa-Martinez, R., Guilderson, T.P.,Garreaud, R.D., 2008. Isotopic evidence for hydrologic change related to thewesterlies in SW Patagonia, Chile, during the last millennium. QuaternaryScience Reviews 27 (13e14), 1335e1349.

Moy, C.M., Dunbar, R.B., Guilderson, T.P., Waldmann, N., Mucciarone, D.A.,Recasens, C., Ariztegui, D., Austin, J.A., Anselmetti, F.S., 2011. A geochemical andsedimentary record of high southern latitude Holocene climate evolution fromLago Fagnano, Tierra del Fuego. Earth and Planetary Science Letters 302 (1e2),1e13.

Mühlinghaus, C., Scholz, D., Mangini, A., 2008. Temperature and precipitationrecords from stalagmites grown under disequilibrium conditions: a firstapproach. PAGES News 16 (3), 19e20.

Mühlinghaus, C., Scholz, D., Mangini, A., 2009. Modelling fractionation of stableisotopes in stalagmites. Geochimica et Cosmochimica Acta 73, 7275e7289.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e2322

Nesje, A., Bakke, J., Dahl, S.O., Lie, O., Matthews, J.A., 2008. Norwegian mountainglaciers in the past, present and future. Global and Planetary Change 60, 10e27.

Neukom, R., Luterbacher, J., Villalba, R., Küttel, M., Frank, D., Jones, P.D., Grosjean, M.,Wanner, H., Aravena, J.C., Black, D.E., Christie, D.A., D’Arrigo, R., Larra, A.,Morales, M., Soliz-Gamboa, C., Srur, A., Urrutia, R., von Gunten, L., 2010a. Mul-tiproxy summer and winter surface air temperature field reconstructions forsouthern South America covering the past centuries. Climate Dynamics 37(1e2), 35e51.

Neukom, R., Luterbacher, J., Villalba, R., Küttel, M., Frank, D., Jones, P.D., Grosjean, M.,Esper, J., Lopez, L., Wanner, H., 2010b. Multi-centennial summer and winterprecipitation variability in southern South America. Geophysical ResearchLetters 37, L14708.

New, M., Lister, D., Hulme, M., Makin, I., 2002. A high-resolution data set of surfaceclimate over global land areas. Climate Research 21, 1e25.

North Greenland Ice Core Project Members, 2004. High resolution climate record ofthe northern hemisphere reaching into the Last Interglacial period. Nature 431,147e151.

Oerlemans, J., 2005. Extracting a climate signal from 169 glacier records. Science308 (5722), 675e677.

Orquera, L.A., Legoupil, D., Piana, E.L., 2011. Littoral adaptation at the southern endof South America. Quaternary International 239, 61e69.

Pahnke, K., Sachs, J.P., 2006. Sea surface temperatures of southern midlatitudes0e160 kyr BP. Paleoceanography 21 (2).

Pahnke, K., Zahn, R., Elderfield, H., Schulz, M., 2003. 340,000-year centennial-scalemarine record of Southern Hemisphere climatic oscillation. Science 301 (5635),948e952.

Pendall, E., Markgraf, V., White, J.W.C., Dreier, M., 2001. Multiproxy record of LatePleistoceneeHolocene climate and vegetation changes from a peat bog inPatagonia. Quaternary Research 55 (2), 168e178.

Pisano, E., 1977. Fitogeografía de Fuego-Patagonia Chilena. 1: Comunidades vege-tales entre las latitudes 52� y 56�S. Anales del Instituto de la Patagonia 8,121e150.

Ponce, J.F., Borromei, A.M., Rabassa, J., Martinez, O., 2011. Late Quaternary palae-oenvironmental change in western Staaten island (54.5� S, 64� W), FuegianArchipelago. Quaternary International 233 (2), 89e100.

Porter, S.C., 2000. Onset of Neoglaciation in the Southern Hemisphere. Journal ofQuaternary Science 15 (4), 395e408.

Prahl, F.G., Wakeham, S.G., 1987. Calibration of unsaturation patterns in long-chain ketone compositions for paleotemperature assessment. Nature 330,367e369.

Prahl, F.G., Muehhausen, L.A., Zahnle, D.L., 1988. Further evaluation of long-chainalkenones as indicators of paleoceanographic conditions. Geochimica et Cos-mochimica Acta 52, 2303e2310.

Prahl, F.G., Rontani, J.-F., Zabeti, N., Walinsky, S.E., Sparrow, M.A., 2010. Systematicpattern in UK’37-Temperature residuals for surface sediments from high lati-tude and other oceanographic settings. Geochimica et Cosmochimica Acta 74(1), 131e143.

Premoli, A.C., Mathiasen, P., Kitzberger, T., 2010. Southernmost Nothofagus tressenduring ice ages: genetic and ecological niche retrodiction reveal high latitude(54�S) glacial refugia. Palaeogeography, Paleoclimatology, Paleoecology 298,247e256.

Prieto, A., Calás, E., Morello, F., Torres, J., 2009. Sitios arqueologico Myren2, Tierradel Fuego. Magallania 35 (2), 83e103.

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., Schlüchter, C., 2010. Glacier advancein southern middle-latitudes during the Antarctic Cold Reversal. Nature Geo-science 3, 700e703.

Rasmussen, L.A., Conway, H., Raymond, C.F., 2007. Influence of upper air condi-tions on the Patagonia icefields. Global and Planetary Change 59 (1e4),203e216.

Recasens, C., Ariztegui, D., Gebhardt, C., Gogorza, C., Haberzettl, T., Hahn, A., Liesé-Pronovost, Kliem, Lücke, A., Maidana, N., Maryr, C., Ohlendorf, C., Schäbitz, F., St-Onge, G., Wille, M., Zolitschka, B., POSADA Science Team, 2011. New insightsinto paleoenvironmental changes in Laguna Potrok Aike, southern Patagonia,since the Late Pleistocene: the Posada multiproxy record. The Holocene. http://dx.doi.org/10.1177/0959683611429833.

Rein, B., Lückge, A., Reinhardt, L., Sirocko, F., Wolf, A., Dullo, W.-C., 2005. El Niñovariability off Peru during the last 20,000 years. Paleoceanography 20, PA4003.http://dx.doi.org/10.1029/2004PA001099.

Renssen, H., Goosse, H., Fichefet, T., Brovkin, V., Driesschaert, E., Wolk, F., 2005.Simulating the Holocene climate evolution at northern high latitudes usinga coupled atmosphere-sea ice-ocean-vegetation model. Climate Dynamics 24(1), 23e43.

Rignot, E., Rivera, A., Casassa, G., 2003. Contribution of the Patagonia icefields ofsouth America to sea level rise. Science 302 (5644), 434e437.

Rigozo, N.R., Nordemann, D.J.R., da Silva, H.E., Souza Echer, M.P., Echer, E., 2007.Solar and climate signal records in tree ring width from Chile (AD 1587e1994).Planetary and Space Science 55 (1e2), 158e164.

Rincón-Martínez, D., Lamy, F., Contreras, S., Leduc, G., Bard, E., Saukel, C., Blanz, T.,Mackensen, A., Tiedemann, R., 2010. More humid interglacials in Ecuadorduring the past 500 kyr linked to latitudinal shifts of the equatorial front andthe Intertropical Convergence Zone in the eastern tropical Pacific. Paleo-ceanography 25, PA2210. http://dx.doi.org/10.1029/2009PA001868.

Rivera, A., 2004. Mass balance investigations at Glaciar Chico, Southern PatagoniaIcefield, Chile. PhD thesis, Bristol University, UK. 303 pp.

Rohling, E.J., Pälike, H., 2005. Centennial-scale climate cooling with a sudden coldevent around 8,200 years ago. Nature 434 (7036), 975e979.

Rojas, M., Moreno, P., 2011. Atmospheric circulation changes and neoglacialconditions in the Southern Hemisphere mid-latitudes: insights from PMIP2simulations at 6 kyr. Climatre Dynamics 37, 357e375.

Rojas, M., Moreno, P., Kageyama, M., Crucifix, M., Hewitt, C., Abe-Ouchi, A.,Ohgaito, R., Brady, E.C., Hope, P., 2009. The Southern Westerlies during the LastGlacial Maximum in PMIP2 simulations. Climate Dynamics 32 (4), 525e548.http://dx.doi.org/10.1007/s00382-008-0421-7.

Rother, H., Shulmeister, J., 2006. Synoptic climate change as a driver of lateQuaternary glaciations in the mid-latitudes of the Southern Hemisphere.Climate of the Past 2, 11e19.

Sapkota, A., Cheburkin, A.K., Bonani, G., Shotyk, W., 2007. Six millenia of atmo-spheric dust deposition in southern south America (Isla Navarino, Chile). TheHolocene 17 (5), 561e572.

Schimpf, D., Kilian, R., Kronz, A., Simon, K., Spötl, C., Wörner, G., Deininger, M.,Mangini, A., 2011. The significance of chemical, isotopic, and detrital compo-nents in three coeval stalagmites from the superhumid southernmost Andes(53� S) as high-resolution palaeo-climate proxies. Quaternary Science Reviews30 (3e4), 443e459.

Schäbitz, F., 1991. Holocene vegetation and climate in southern Santa Cruz,Argentina. Bamberger Geographische Schriften 11, 235e244.

Schneider, C., Gies, D., 2004. Effects of El Niñoesouthern oscillation on southern-most South America precipitation at 53�S revealed from NCEPeNCAR rean-alyses and weather station data. International Journal of Climatology 24 (9),1057e1076.

Schneider, C., Kilian, R., Santana, A., Butorovic, N., Casassa, G., 2003. Weatherobservations across the southern Andes at 53�S. Physical Geography 24 (2),97e119.

Schneider, C., Schnirch, M., Acúna, C., Casassa, G., Kilian, R., 2007. Glacier inventoryof the Gran Campo Nevado Ice Cap in the Southern Andes and glacier changesobserved during the recent decades. Global and Planetary Change 59 (1e4),87e100.

Sepúlveda, J., Pantoja, S., Hughen, K., Bertrand, S., Figueroa, D., León, T., Drenzek, N.,Lange, C., 2009. Late Holocene sea-surface temperature and precipitationvariability in northern Patagonia, Chile (Jacaf Fjord, 44�S). Quaternary Research.http://dx.doi.org/10.1016/j.yqres.2009.06.010.

Siddall, M., Rohling, E.J., Almogi-Labin, A., Hemleben, Ch., Meischner, D., Schmelzer,Smeed, D.A., 2003. Sea level fluctuations during the Last Glacial cycle. Nature423, 853e858.

Steinhilber, F., Beer, J., Fröhlich, C., 2009. Total solar irradiance during the Holocene.Geophysical Research Letters 36, L19704. http://dx.doi.org/10.1029/2009GL040142.

Stenni, B., Burion, 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-Delmonte, V.,Mazzola, C., Minster, B., Montagnat, M., Mulvaney, R., Narcisci, B., Oerter, H.,Parrenin, F., Petit, J.R., Ritz, C., Scarchilli, C., Schilt, A., Schüpach, S., Schwander, J.,Selmo, E., Sereri, M., Stocker, T.F., Udisti, R., 2011. Expression of the bipolar see-saw in Antarctic climate records durinmg the last deglaciation. Nature Geo-sience 4, 46e50.

Stern, C.R., 1990. Tephrochronology of southernmost Patagonia. NationalGeographic Research 6 (1), 110e126.

Stern, C.R., 1992. Tefrocronologia de Magellanes: nuevos datos e implicaciones.Anales del Instituto de la Patagonia 21, 129e141.

Stern, C.R., 2008. Holocene tephrochronology record of large explosive eruptions inthe southernmost Patagonia Andes. Bulletin of Volcanology 70 (4), 435e454.

Stern, C.R., Kilian, R., 1996. Role of the subducted slab, mantle wedge and conti-nental crust in the generation of adakites from the Andean Austral VolcanicZone. Contributions to Mineralogy and Petrology 123 (3), 263e281.

Stern, C.R., Moreno, P.I., Villa-Martínez, R., Sagredo, E.A., Prieto, A., Labarca, R., 2011.Evolution of ice-dammed proglacial lakes in Última Esperanza, Chile: implica-tions from the late-glacial R1 eruption of Reclús volcano, Andean AustralVolcanic Zone. Andean Geology 38 (1), 82e97.

Stine, S., Stine, M., 1990. A record from Lake Cardiel of climate change in southernSouth America. Nature 345, 705e708.

Stocker, T.F., 1998. Climate change: the seesaw effect. Science 282 (5386), 61e62.Stocker, T.F., Johnsen, S.J., 2003. A minimum thermodynamic model for the bipolar

seesaw. Paleoceanography 18 (4), 1087. http://dx.doi.org/10.1029/2003PA000920.

Stott, L., Timmermann, A., Thunell, R., 2007. Southern Hemisphere and deep-seawarming led deglacial atmospheric CO2 rise and tropical warming. Science318 (5849), 435e438.

Strelin, J.A., Denton, G.H., Vandergoes, M.J., Ninnemann, U.S., Putnam, A.E., 2012.Radiocarbon chronology of the late-glacial Puerto Bandera moraines, southernPatagonian Icefield, Argentina. Quaternary Science Reviews. http://dx.doi.org/10.1016/j.quascirev. 2011.05.004.

Sugden, D., Bentley, M., Fogwill, M., Hulton, N., McCulloch, R., Purves, R., 2005. Late-glacial glacier events in southernmost South America: a blend of ‘northern’ and‘southern’ hemispheric climatic signals? Geografiska Annaler: Series A, PhysicalGeography 87 (2), 273e288.

Sugden, D.E., McCulloch, R.D., Bory, A.J.M., Hein, A.S., 2009. Influence of Patagonianglaciers on Antarctic dust deposition during the Last Glacial period. NatureGeoscience 2 (4), 281e285.

R. Kilian, F. Lamy / Quaternary Science Reviews 53 (2012) 1e23 23

Timmermann, A., Timm, O., Stott, L., Menviel, L., 2009. The roles of CO2 and orbitalforcing in driving southern hemispheric temperature variations during the last21,000 years. Journal of Climate 22 (7), 1626e1640.

Toggweiler, J.R., Joellen Russell, J., 2008. Ocean circulation in a warming climate.Nature 451, 286e288.

Toggweiler, J.R., Russell, J.L., Carson, S.R., 2006. Midlatitude westerlies, atmosphericCO2, and climate change during the ice ages. Paleoceanography 21, PA2005.

Tonella, M.S., Mancini,M.V., Seppä, H., 2009. Quantitative reconstruction of Holoceneprecipitation changes in southern Patagonia. Quaternary Research 72, 410e420.

Torres, J., 2009. La pesca entre los cazadores recolectores terrestres de la Isla Grandede Tierra del fuego, desde la prehistoria a tiempos etnográficos. Magallania 37(2), 109e138.

Unkel, I., Fernandez, M., Björk, S., Ljung, K., Wohlfarth, B., 2010. Records of envi-ronmental changes during the Holocene from Isla de los Estados (54.4�S),southeastern Tierra del Fuego. Global and Planetary Change 74 (3e4), 99e113.

Van Geel, B., Heusser, C.J., Renssen, H., Schuurmans, C.J.E., 2000. Climatic change inChile at around 2700 BP and global evidence for solar forcing: a hypothesis. TheHolocene 10, 659e664.

Varma, V., Prange, M., Lamy, F., Merkel, U., Schulz, M., 2011. Solar-forced shifts of thesouthern hemisphere westerlies during the Holocene. Climate of the Past 7 (2),339e347. doi:10.5194/cp-7-339-2011.

Vieira, L.E.A., Solanki, S.K., Krivova, N.A., Usoskin, I., 2011. Evolution of the solarirradiance during the Holocene. Astronomy and Astrophysics 531, A6.

Villa-Martínez, R., Moreno, P.I., 2007. Pollen evidence for variations in the southernmargin of the westerly winds in SW Patagonia over the last 12,600 years.Quaternary Research 68 (3), 400e409.

Villalba, R., Lara, A., Boninsegna, J.A., Masiokas, M., Delgado, S., Aravena, J.C.,Roig, F.A., Schmelter, A., Wolodarsky, A., Ripalta, A., 2003. Large-scale temper-ature changes across the southern Andes: 20th-century variations in thecontext of the past 400 years. Climatic Change 59 (1e2), 177e232.

Villalba, R., Grosjean, M., Kiefer, T., 2009. Long-term multi-proxy climate recon-structions and dynamics in South America (LOTRED-SA): state of the art andperspectives. Palaeogeography, Palaeoclimatology, Palaeoecology 281, 175e179.

Vimeux, F., De Angelis, M., Ginot, P., Magand, O., Casassa, G., Pouyaud, B., Falourd, S.,Johnsen, S., 2008. A promising location in Patagonia for paleoclimate andpaleoenvironmental reconstructions revealed by a shallow firn core fromMonte San Valentín (Northern Patagonia Icefield, Chile). Journal of GeophysicalResearch 113, 1e20.

Vimeux, F., Ginot, P., Schwikowski, M., Vuille, M., Hoffmann, G., Thompson, L.G.,Schotterer, U., 2009. Climate variability during the last 1000 years inferred fromAndean ice cores: a review of methodology and recent results. Palae-ogeography, Palaeoclimatology, Palaeoecology 281 (3e4), 229e241.

Wagner, S., Widmann, M., Jones, J., Haberzettl, T., Lucke, A., Mayr, C., Ohlendorf, C.,Schabitz, F., Zolitschka, B., 2007. Transient simulations, empirical reconstruc-tions and forcing mechanisms for the Mid-Holocene hydrological climate insouthern Patagonia. Climate Dynamics 29 (4), 333e355.

Waldmann, N., Ariztegui, D., Anselmetti, F.S., Austin, J.A., Stern, C., Moy, C.M.,Recasens, C., Dunbar, R., 2010. Holocene climatic fluctuations and positioning ofthe Southern Hemisphere westerlies in Tierra del Fuego (54�S), Patagonia.Journal of Quaternary Science 25 (7), 1063e1075.

Waldmann, N., Anselmetti, F.S., Austin, J.A., Pirouz, M., Moy, C.M., Dunbar, R., 2011.Holocene mass-wasting events in Lago Fagnano, Tierra del Fuego (54�S):implications for paleoseismicity of the Magallanes-Fagnano transform fault.Basin Research 23 (2), 171e190.

Wanner, H., Beer, J., Bütikofer, J., Crowley, T.J., Cubasch, U., Flückiger, J., Goosse, H.,Grosjean, M., Joos, F., Kaplan, J.O., Küttel, M., Müller, S., Prentice, I.C.,Solomina, O., Stocker, T.F., Tarasov, P., Wagner, M., Widmann, M., 2008. Mid-to late Holocene climate change: an overview. Quaternary Sci. Rev. 27,1791e1828.

Wanner, H., Solomina, O., Grosjean, M., Ritz, S., Jetel, M., 2011. Structure andorigin of Holocene cold events. Quaternary Science Reviews 30 (21e22),3109e3123.

Warren, C., Sudgen, D.E., 1993. The Patagonian Icefields: a Glaciological review.Arctic and Alpine Research 25 (4), 316e331.

Wenzens, G., 1999. Fluctuations of outlet and valley glaciers in the southern Andes(Argentina) during the past 13,000 years. Quaternary Research 51 (3), 238e247.

Wenzens, G., 2005. Glacier advances east of the Southern Andes between the LastGlacial Maximum and 5,000 BP compared with lake terraces of the endorrheicLago Cardiel (49�S, Patagomgonia, Argentina). Zeitschrift fur GeomorphologieN.F 49 (4), 433e454.

Wenzens, G., Wenzens, E., 1998. Late glacial and Holocene glacier advances in thearea Lago Viedma (Patagonia, Argentina). Zentralblatt Geologie Paläontologie.Teil I 1997 (3e6), 593e608.

Whitlock, C., Moreno, P.I., Bartlein, P., 2007. Climatic controls of Holocene firepatterns in southern South America. Quaternary Research 68 (1), 28e36.

Wille, M., Schäbitz, F., 2009. Late-glacial and Holocene climate dynamics at thesteppe/forest ecotone in southernmost Patagonia, Argentina: the pollen recordfrom a fen near Brazo Sur, Lago Argentino. Vegetation History and Archae-obotany 18 (3), 225e234.

Wille, M., Maidana, N.I., Schäbitz, F., Fey, M., Haberzettl, T., Janssen, S., Lücke, A.,Mayr, C., Ohlendorf, C., Schleser, G.H., Zolitschka, B., 2007. Vegetation andclimate dynamics in South America: the microfossil record of Laguna PotrokAike, Santa Cruz, Argentina. Review of Paleobotany and Palynology 146,234e246.