Vegetation, climate and fire regime changes in the Andean region of southern Chile (38°S) covaried with centennial-scale climate anomalies in the tropical Pacific over the last 1500 years

Quaternary Science Reviews 46 (2012) 46e56

Contents lists available at SciVerse ScienceDirect

Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev

Vegetation, climate and fire regime changes in the Andean region of southern Chile (38
S) covaried with centennial-scale climate anomalies in the tropical Pacific over the last 1500 years Michael-Shawn Fletcher a, *, Patricio Iván Moreno a, b a b

Institute of Ecology and Biodiversity, University of Chile, Santiago, Chile Department of Ecological Sciences, University of Chile, Santiago, Chile

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 January 2012 Received in revised form 21 April 2012 Accepted 23 April 2012 Available online 2 June 2012

Pollen and charcoal analysis from Laguna San Pedro (38
260 S, 71
190 W), a small closed-basin lake located within the present-day distribution of Araucaria-Nothofagus forest in the Temperate-Mediterranean Transition zone in the Andes of Chile (35.5e39.5
S), reveal centennial-scale changes in vegetation, climate and fire regime since 1500 cal yr BP. We interpret periods of relatively low growing season (summer) moisture and increased fire activity between 1500e1300 and 1000e725 cal yr BP, the latter period is also characterised by remarkably rapid bulk sediment accumulation and we infer prolonged annual sedimentation resulting from a decrease in the duration of lake freezing under a warmer climate. Relatively moist conditions during summer and low fire activity occurred between 1300e1000 and 725e121 cal yr BP, with slow bulk sediment accumulation through the latter phase in particular implying a cool and wet climate. Our results suggest that the Medieval Climate Anomaly chronozone was relatively warm and dry, followed by a cool-wet climate during the Little Ice Age chronozone, before a substantial modification of the vegetation landscape by Europeans occurred in the mid 1800’s. The timing and direction of changes in the Laguna San Pedro record bear a striking resemblance to multiple independent tropical Pacific precipitation reconstructions, areas where precipitation is governed by the El NiñoSouthern Oscillation (ENSO), and with the modern influence of ENSO over the climate in the region. We conclude that ENSO was the main driver of changes in growing season moisture in this part of the Temperate-Mediterranean Transition in south-central Chile over the last 1500 years. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: Nothofagus-Araucaria forests South-central Chile Temperate-Mediterranean Transition El Niño-Southern Oscillation (ENSO) Medieval Climate Anomaly Little Ice Age

1. Introduction It is of vital importance that we understand the nature, timing, impacts and regional expression of past periods of abrupt climate change if we are to place current and future climate trends in to a meaningful historical context. A significant body of evidence now documents widespread evidence for decadal to centennial-scale climate intervals over the last 1500 years, most notably manifest in the so called Little Ice Age (LIA), Medieval Climate Anomaly (MCA) and the Dark Ages Cold Period (DACP) (Seager et al., 2007; Solomon et al., 2007; Mann et al., 2009; Graham et al., 2010). To date, there is a heavy bias towards northern hemisphere records and comparatively little empirical evidence for concurrent climate intervals from southern hemisphere landmasses. Available studies

* Corresponding author. Tel.: þ56 29787420. E-mail addresses: michael.fl[email protected] (M.-S. Fletcher), pimoreno@ uchile.cl (P.I. Moreno). 0277-3791/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.quascirev.2012.04.016

from the southern hemisphere indicate evidence of both regionalscale trends over the last 1500 years (e.g. Seager et al., 2007; Mann et al., 2009; Graham et al., 2010; Neukom et al., 2010a, 2010b) and a high degree of spatial heterogeneity in the nature, direction and timing of climate trends (Masiokas et al., 2009). South America is the only landmass that virtually transects the entire meridional gradient from the tropics to the sub-Antarctic latitudes and is, thus, an important location for the discourse over global climate dynamics. Southern South America spans the entire zone influenced by the sub-tropical and extra-tropical climate domains and, as such, is well situated to track changes in two globally important components of the climate system: the subtropical South Pacific Ocean high pressure system and the extratropical southern westerly winds (SWW), as well as their leading inter-annual modes (the El Niño-Southern Oscillation (ENSO) and Southern Annular Mode (SAM), respectively). Moreover, the Andes Cordillera, a lofty and glaciated mountain range that runs in a north-south direction along the entire western sea-board of southern South America, intercepts the Earth’s zonal tropospheric

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

flow (Garreaud et al., 2009), resulting in sharp environmental gradients that make this region well suited for tracking the effects of past climate change on terrestrial environments. Despite this critical geographical setting, the plethora of fast sedimentaccumulating lakes and long-lived climatically sensitive tree species, there are remarkably few continuous palaeoclimatic reconstructions of sufficient resolution to allow an adequate appraisal of change over the last 1500 years (e.g. Stine, 1994; Villalba, 1994; Jenny et al., 2002; Ariztegui et al., 2007; Boës and Fagel, 2008; Moy et al., 2009; von Gunten et al., 2009; Neukom et al., 2010b; Jara and Moreno, 2012). Available evidence from sub- and extra-tropical southern South America suggests that the broadly defined LIA chronozone (ca 600100 calendar years before present [cal yr BP]) was characterised by: (i) an expansion of virtually all studied Andean glaciers (Luckman and Villalba, 2001; Masiokas et al., 2009) and (ii) relatively low temperatures between 20 and 55
S (Villalba, 1994; Villalba et al., 2003; Neukom et al., 2010b). Despite these broad-scale trends, there is a substantial degree of spatial heterogeneity in the timing and nature of environmental responses to climate change through the LIA chronozone, with, for example, substantial differences in the timing of maximum ‘LIA’ glacier expansion across southern South America (Luckman and Villalba, 2001; Masiokas et al., 2009). Likewise, the MCA (ca 1050e600 cal yr BP) broadly encompasses a period of climatic change in southern South America that is variously manifested in both temperature and precipitation changes (Jenny et al., 2002; Bertrand et al., 2005; Mann et al., 2009; Moreno et al., 2009; Moy et al., 2009; Graham et al., 2010; Neukom et al., 2010a, 2010b; e.g. Villalba, 1994; von Gunten et al., 2009). Here we present detailed pollen, charcoal and sedimentary data from a lake sediment core retrieved from Laguna San Pedro, southcentral Chile, that span the last 1500 years. Laguna San Pedro is a small closed-basin lake located close to the precipitation dependent forest-steppe ecotone in the transitional zone between the sub-tropical and temperate regions of southern South America, the Temperate-Mediterranean Transition (TMT). The site, which was the focus of a previous palynological study of sediments extracted from a profile in an adjacent wetland (Rondanelli-Reyes, 2000), is located in Valle Lonquimay (Fig. 1), near several explosive stratovolcanoes. Precipitation in the TMT is derived entirely from the SWW and the main driver of inter-annual variability in the modern climate is ENSO (Montecinos and Aceituno, 2003; Garreaud, 2007; Garreaud et al., 2009). The Laguna San Pedro data allows assessment of the following questions: (1) what was the nature of environmental change in this region over the last 1500 years?; (2) are there discernible biological/climatic anomalies during the LIA, MCA and DACP chronozones?; and (3) what were the main driver(s) of environmental change in this region over the last 1500 years?

47

material and lithosols derived from lava, scoria, pumice and tephra prevail across the landscape (Peralta, 1980). Four stratovolcanos (active over the historic period) lie less than 45 km west of the site: Volcán Llaima (2920 m a.s.l., 44 km); Volcán Tolhuaca (2806 m a.s.l., 31 km); Volcán Sierra Nevada (2554 m a.s.l., 27 km); and Volcán Lonquimay (2865 m a.s.l, 21 km) (Fig. 1). 2.2. Climate

2. Study region

The Andes Cordillera induces a dramatic drop in annual rainfall from a maximum of ca 3000 mm on the western slopes of the Andes (Las Raíces climate station, Chile: 38
320 S, 71
320 W, 1010 m a.s.l.) to ca 200 mm in Provincia del Neuquén (Las Lajas meteorological station, Argentina: 38
320 S, 70
230 W, 713 m a.s.l.) over a distance of less than 100 km (Peralta, 1980; Paez et al., 1997). The study site lies w35 km west of the continental divide and receives 1919 mm of annual precipitation (73% falling during autumn/ winter; Lonquimay climate station, Chile: 38
270 S, 71
220 W, 900 m a.s.l.). The thermal amplitude is considerable, reflecting the continental location, with winter (JJA) temperatures at the site as low as 20
C and summer (DJF) temperature as high as 35
C (although sub-zero summer temperatures are not uncommon) (Peralta, 1980). Average annual temperature recorded at Lonquimay is 8.3
C with an average annual thermal amplitude of 13.2
C, while east of the site in the sub-humid region (Las Lajas climate station) annual thermal amplitude reaches 29.3
C as continentality increases (Peralta, 1980; Paez et al., 1997). Laguna San Pedro freezes in winter and the broader region is blanketed in over 2 m of snow (Rondanelli-Reyes, 2000). The growing season is short, ca 120 days (between December 10-April 10), and summer (JFM) precipitation is a key determinant of plant growth (Peralta, 1980). Inter-annual climate variation in this part of Chile is predominantly associated with ENSO and SAM (Montecinos and Aceituno, 2003; Garreaud, 2007; Garreaud et al., 2009). While there is an overall increase (decrease) in precipitation and temperature over the region during El Niño (La Niña) events, much of the additional rain falls in winter and spring (SON) as snow. The most salient feature in the seasonal breakdown of the influence of ENSO over the climate of the study region is the tendency for warm (cool) and dry (wet) summers resulting from a southward (northward) displacement of the cool moisture-bearing SWW related to a strengthened (weakened) South Pacific sub-tropical high pressure system during El Niño (La Niña) events (Montecinos and Aceituno, 2003; Garreaud et al., 2009). Oscillations of the SAM, the leading inter-annual mode of the extra-tropical SWW, impact primarily on spring rainfall in the study region, with positive (negative) SAM resulting in decreased (increased) rainfall over the region as the SWW migrate southward (northward) (Garreaud, 2007; Garreaud et al., 2009).

2.1. Physical setting

2.3. Vegetation

This study is centred on the mountainous Región de la Araucanía (36e39.5
S) in south-central Chile. The study area lies in the core of the Andes Cordillera which in this sector attains average elevation of ca 3000 m a.s.l., hosting several glaciers and active volcanoes. Here we sampled and analysed sediment cores from Laguna San Pedro (913 m a.s.l.; 38
260 S, 71
190 W; Fig. 1), a small closed-basin lake that lies within the formerly glaciated Valle Lonquimay that is situated 3 km east of the town of Lonquimay. The valley is a relatively flat southwest-northeast trending glaciofluvial plain that flows northeast from Volcán Sierra Nevada and forms the headwaters of Alto Río Bío-Bío. The bedrock in the region is composed of predominantly andesitic and basaltic volcanic

Much of the vegetation of the flatter arable areas within Valle Lonquimay has been converted to exotic pasture. The native vegetation of the broader region is considered ‘transitional’, reflecting the influence of the steep precipitation and thermal gradients imparted by the Andes: the west to east precipitation gradient exerts the greatest influence over vegetation composition, with temperature (along the altitudinal gradient) inducing a secondary influence (Michell, 1980). The region is characterised by a double treeline, with Nothofagus pumilio forming the lower treeline at ca 1500 m a.s.l. and Araucaria araucana forming the upper treeline at ca 1800 m a.s.l. (Mc Queen, 1977; Michell, 1980). Michell (1980) identified seven plant communities occurring in the

48

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

Fig. 1. Map of the study area showing the location of sites mentioned in text.

Lonquimay region today: Discaria-Colletia spiny shrub (arbustos espinosos); high altitude Nothofagus obliqua (Roble de altura); hard pasture (pastos duros-Mallines); Nothofagus antarctica; N. pumilioA. araucana; Pure A. araucana (Pehuén sensu Michell, 1980) and High Andean Steppe. Fig. 2 is a schematic that shows the location of each of these communities along altitudinal and moisture (eastwest) gradients. Also shown for reference to the broader region in Fig. 2 are the locations in the schematic of communities not currently found in the Lonquimay area: N. obliqua of the Central Valley; N. obliquaeN. alpina; and Nothofagus dombeyi (sensu Michell, 1980). 2.4. Fire and climate A wealth of literature points towards the importance of fire at the forest-steppe ecotone of southern South America and the present study region is no exception (Gonzalez et al., 2005; González and Veblen, 2009; e.g. Kitzberger et al., 1997; Veblen et al., 1999). The frequency and magnitude of fires in this region are significantly correlated with ENSO (Veblen et al., 1999; Kitzberger et al., 2001), with widespread fires occurring in response to both the winter-spring moisture deficit experienced during strong La Niña events, and the warm dry summers experienced during El Niño events (Veblen et al., 1999). Given that the study region is covered in snow and is effectively in a state of dormancy during winter and much of spring, we expect that the warm dry

summers experienced during El Niño events would exert the greatest influence over inter-annual trends in fire regime. 3. Methods 3.1. Field operations and sediment analysis Two offset and overlapping cores were retrieved from the deepest part of Laguna San Pedro with a 7.5 cm diameter plexiglass piston corer that preserves the integrity of the wateresediment interface: MSF0907SC1(0e100 cm) and MSF0907SC2 (50e123 cm). The deepest part of the lake was located with the aid of a bathymetric map we created using an echo-sounder. We extruded and sub-sampled the cores at 1 cm-thick contiguous intervals on-site. Loss-on-ignition analysis was performed on contiguous 1 cm-thick samples to determine organic content (550
C), carbonate content (925
C) and siliciclastic content (residual), after overnight drying at 105
C (Heiri et al., 2001). 3.2. Chronology The chronology of the sediment cores is controlled by six AMS radiocarbon dates of bulk organic lake sediment and 210Pb analysis of the upper 19 cm of the lake sediments. Radiocarbon dating of bulk sediments is prone to carbon-reservoir effects in which ‘old carbon’ stored in the environment and subsequently delivered in to

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

49

resolution (9 yr cm1) to produce equally-spaced intervals and then calculated charcoal accumulation rates (CHAR ¼ particles cm2 yr1). Charcoal peaks, a statistically robust proxy for local fire episodes (Higuera et al., 2010), were identified as the positive residuals exceeding the 95th percentile threshold of a locally fitted Gaussian mixture CHAR background model. Charcoal peak magnitude, possibly related to fire size, fire intensity and/or taphonomic processes, represents the total accumulation of charcoal for statistically significant charcoal peaks (Higuera et al., 2007). The signal-to-noise ratio, a metric of CHARAnalysis performance was also calculated. 4. Results 4.1. Bathymetry, stratigraphy and chronology

Fig. 2. A plot of vegetation community versus their key environmental determinants adopted from Michell (1980): altitude (read: temperature) and moisture.

a lake can result in erroneously old ages, an effect that results in offsets between 0 and 8000 year depending on-site context (Grimm et al., 2009). Given that the site is a precipitation fed, small and shallow closed-basin lake around which there is no carbon-rich bedrock and whose sediments contain a negligible amount of carbonate (Fig. 4), it is unlikely that the chronology suffers from a carbon-reservoir effect. Bayesian modelling was employed to generate an age-depth model using Oxcal 4.1 (Bronk Ramsey, 2010), as recommended by Blaauw (2010). Calibrations between 14C years before present (14C yr BP) and cal yr BP were performed using the Southern Hemisphere calibration curve (McCormac et al., 2004). We consider tephra layers to be instantaneous events and subtracted their thickness from the depth scale prior to fitting of the age-depth model.

We produced a bathymetric map (Fig. 3) that reveals a simple and planar lake basin with a maximum depth of 450 cm. We matched the overlapping sediment cores (MSF0907SC1 and MSF0907SC2) on the basis of their loss-on-ignition data and produced a composite 123 cm-long record presented in Fig. 4. The sediments consist of brown organic lake mud (gyttja), pyroclastic material (tephra) and fine silt. The 123 cm composite core contains tephra layers at 78e88 cm and 98e114 cm, that are clearly evident in the inorganic density curve (Fig. 4) and which provided a good tie point between the cores. Apart from an inorganic layer between 10 and 11 cm that consists of fine silt sized particles (Fig. 4) and the afore mentioned tephra layers, there is no evidence of sudden influxes of allochtonous sediment in to the lake. Organic content of the gytta varies between 20 and 30%, while carbonate values consistently below 4% (material lost after the 925
C combustion) fall within the background error of the method and thus suggest an overall absence of carbonates from the lake system.

3.3. Pollen analysis We processed contiguous 1 cm-thick samples for pollen and spore content, using standard pollen extraction procedures (Faegri and Iversen, 1989). The basic pollen sum for each level includes at least 300 pollen grains of terrestrial origin. We calculated percentages of aquatic and fern taxa from a super-sum that included the basic pollen sum, the sum of aquatic taxa and the sum fern spores. We also tallied algal cysts (Botryococcus spp. and Pediastrum spp.) which we express as accumulation-rate data. 3.4. Charcoal We analysed the macroscopic charcoal content of 1-cm thick contiguous sediment samples (2 cc) to document the local fire history. The sediment samples were disaggregated in a 10% KOH solution, and sieved using 106 and 212 mm mesh diameters (Whitlock and Larsen, 2001). We performed time series analysis of the charcoal data using the CharAnalysis software (Higuera et al., 2008) by interpolating charcoal counts to the median sample

Fig. 3. Bathymetric map of Laguna San Pedro showing the location of the coring site at 4.5 m water depth. The contour interval is 1 m.

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

te nt

(g /c bo na

ni c

te

de

co n

ns i

ty

)

C ar

In or ga

Ra di oc De arb on pt h (c Da m te Li ) s th ol og y O rg an ic co nt en t( % ) O rg an ic de ns ity (g /c c

(% )

c)

50

0 10 20

740 ± 35 755 ± 35

30 40 50 60

965 ± 30 950 ± 35

70 80

1275 ± 35

90 100 110

1515 ± 35 120

10

20

30

Gyttja

0.04

Tephra

0.80

1.60

2.5

5.0

Silt

Fig. 4. Results of the Loss on Ignition analysis on the sediments of Laguna San Pedro. The dashed horizontal line shows the splice point between MSF0907SC1 (0e97 cm) and MSF0907SC2 (98e123 cm) Radiocarbon dates are in 14C yr BP.

The chronology is constrained by 16 210Pb activity determinations obtained from MSF0807SC1 (Table 1) and six radiocarbon assays obtained from both MSF0907SC1 and MSF0907SC2 (Table 2). The 210Pb age model was derived using the constant rate of supply model. An age-depth model was fitted to the 210Pb and calibrated 14 C ages using Bayesian modelling after subtraction of the two tephra layers between 78e88 cm and 98e114 cm using Oxcal 4.1 (Bronk Ramsey, 2010). Four of the radiocarbon dates displayed a binomial distribution on the calibration curve (Fig. 5) and we selected the option in Oxcal 4.1 to use intercept values as chronological tie points to produce an age-depth model devoid of age reversals (Fig. 5). Ages were interpolated to each sample depth

based on a linear interpolation between median values at each chronological tie point. As with all age-depth models, each interpolated value represents one of a range of possible values for its depth and our interpretations must be considered within this context. Furthermore, given the possibility of an over estimate of ages within the radiocarbon-dated section, as discussed in Section 3.2, we consider the age-depth model as our best estimate for the age of the sediments. Given these caveats, the bulk sediment deposition time is highly variable, ranging from ca 2.25 years/cm in rapidly accumulating sections and ca 39.75 years/cm in slow accumulating sections of the record. 4.2. Palynology

Table 1 Results of 210Pb determinations performed on core MSF0907SC1. Age interpolations are based on the constant rate of supply model. Depth (cm)

210

0 1 2 3 4 5 6 7 8 9 12 15 18 21 24 30

0.076 0.069 0.079 0.061 0.053 0.055 0.056 0.053 0.056 0.031 0.022 0.015 0.012 0.007 0.005 0.002

Pb (Bq/g)

Age (years AD)

Age (cal yr BP)

Error (cal yr BP)

2009 2007 2004 1999 1996 1992 1987 1982 1976 1967 1937 1914 1883

59.31 56.77 53.58 49.33 45.61 41.88 37.28 32.23 25.79 16.57 12.87 36.04 66.78 Background Background Background

0 0 1 1 1 2 3 4 4 6 10 20 35

The pollen and spore content of the Laguna San Pedro sediment core consists of 97 continuous contiguous samples (excluding subtracted tephras). We identified five pollen zones (Fig. 6), the location of which were guided (but not defined) by a stratigraphically constrained CONISS ordination (Grimm, 1987). The pollen record is overwhelmingly dominated by N. dombeyi type and Poaceae, which display a clear and consistent inverse relationship with each other. Following is a brief description of the principal trends in each defined pollen zone (pollen zone depths are expressed as modified depths [tephras removed] and follow depths in Fig. 6; percent values in parentheses indicate either maximum values or ranges). 4.2.1. Pollen zone SP 1 (96e88 cm; ca 1500e1370 cal yr BP) This zone is dominated by N. dombeyi type (50e70%) and Poaceae (15e25%), with the latter (former) peaking (dipping) at ca 1400 cal yr BP. N. obliqua type gradually declines from 18% to 10% towards the top of the zone. This zone contains a relatively high abundance of Botryococcus spp. cysts.

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

51

Table 2 Information on the radiocarbon ages from the Laguna San Pedro site. Bayesian age-depth modelling was performed in Oxcal 4.1 (Bronk Ramsey, 2010) using the southern hemisphere calibration curve (McCormac et al., 2004). Displayed modelled ages (and upper and lower ranges) were calculated in Oxcal 4.1 using intercept values. All depths are modified to match the composite 123 cm core. * Depths in parentheses correspond to the original uncorrected depths of core MSF0907SC2. Depth (cm)* 31 43 64 73 78 (12) 88 (22)

14

C Age (years) 740 755 965 950 1275 1515

Error (14C years)

Median modelled age (cal yr BP)

Upper modelled age (cal yr BP)

Lower modelled age (cal yr BP)

Thickness

Core

35 35 35 35 35 35

558 633 790 866 1009 1295

598 715 885 929 1222 1395

577 665 833 899 1107 1336

1 1 1 1 1 1

MSF0907SC1 MSF0907SC1 MSF0907SC1 MSF0907SC1 MSF0907SC2 MSF0907SC2

4.2.2. Pollen zone SP 2 (88e79 cm; ca 1370e1100 cal yr BP) This zone directly overlies the lowermost tephra (88e89 cm, Fig. 7; 99e114 cm, Fig. 4), for which we obtained a close maximum limiting age of ca 1350 cal yr BP. The zone starts with a reversal of the trends in N. dombeyi type and Poaceae observed during the previous zone, with N. dombeyi type increasing markedly (70%) and Poaceae falling to the lowest values of the entire

record (5%) at ca 1100 cal yr BP. N. obliqua type, A. araucana and Misodendron spp. all display some variability with no clearly identifiable trend. Cyperaceae increases slightly and the accumulation rates of Botryococcus spp. are low. 4.2.3. Pollen zone SP 3 (79e49 cm; ca 1100e725 cal yr BP) This zone directly overlies the uppermost tephra (79e80 cm, Fig. 7; 79e87 cm, Fig. 4) for which we obtained a close maximum age of ca 1100 cal yr BP. The upper part of this zone features a gradual and slight reversal of these trends described in N. dombeyi type and Poaceae. The accumulation rate of Pediastrum spp. increases sharply at ca 740 cal yr BP and remain abundant until the end of this zone. This zone starts with a brief rise that lead to a peak in A. araucana (7%), a steady increase in Poaceae that culminated at 23% by ca 800 cal yr BP, and a decrease of N. dombeyi type to a low (55%) at ca 800 cal yr BP. 4.2.4. Pollen zone SP 4 (49e19 cm; ca 725e121 cal yr BP) Poaceae continues to decline to a low at the upper boundary of this zone, while N. dombeyi type increases to a maximum (78%). A. araucana declines to its lowest values (1%) at ca 300 cal yr BP. N. obliqua type values display some variability, but remain fairly constant throughout. Misodendron spp. peaks (4%) at ca 500 cal yr BP, concomitant with a marked increase in Potamogeton spp. (18%). This pollen type dips between 400 and 300 cal yr BP, before increasing steadily to the upper boundary of the zone. Botryococcus spp. and Pediastrum spp. peak initially before decreasing towards the base of the zone are low throughout. 4.2.5. Pollen zone SP 5 (19e0 cm; ca 121 cal yr BP to present [e59 cal yr BP]) This zone begins with a sharp decline in N. dombeyi type from 78% to 50% in less than 40 years and a further decline to 30% towards the present. Conversely, Poaceae increases markedly to 40% towards the present, as do Rumex acetocella (14%) and Plantago lanceolata (7%) and finally Pinus spp. and Cupressaceae (8%) after ca 0 cal yr BP (1950AD). Potamogeton spp. continues to increase (28%) towards ca 100 cal yr BP, before declining to the present. Botryococcus spp. increases sharply after ca 10 cal yr BP (1960AD). 4.3. Charcoal

Fig. 5. Results of the Bayesian age-depth model fitted through the 210Pb and 14C analyses showing the range of ages (in blue), probability distributions of each calibrated radiocarbon age and the location of the median (vertical line intersecting the probability distributions) age used to interpolate the age-depth model employed in this analysis. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The results of the charcoal analysis are presented in Figs. 7 and 8 as charcoal accumulation rates, calculated using the median values of the Bayesian age-depth model. Initially, charcoal accumulation rates are low, sharply rising between 1422 and 1355 cal yr BP. The period between 1357 and 898 cal yr BP is characterised by low charcoal accumulation, increasing markedly between 898 and 713 cal yr BP, with a prominent peak at 891 cal yr BP. Charcoal accumulation rates, whilst remaining relatively high overall, vary somewhat over the next century, with prominent peaks at 658, 621 and 599 cal yr BP. Low charcoal accumulation rates characterise the

52

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

Fig. 6. Percentage pollen diagram from Laguna San Pedro showing selected taxa. Note the changes in x-scale. The y-axis is plotted according to depth with volcanic ash layers (tephras) removed. Pollen zones are delineated by dashed lines. Grey lines show the location of the extracted tephra layers. The unit of measurement for Macroscopic CHAR (charcoal accumulation rate) is particles cm2 yr1; Botryococcus spp. AR (accumulation rate) is cysts cm2 yr1; and Pediastrum spp. AR is 1000 cysts cm2 yr1 Radiocarbon dates are in 14C yr BP.

remainder of the record, with a 40 year-long peak in charcoal accumulation commencing at 40 cal yr BP. The CHAR interpolations produced by CharAnalysis reveal an initial peak at 1381 cal yr BP and high CHAR values between 1381 and 1349 cal yr BP indicate a period of high relative importance of fire in the landscape (Fig. 7). Low CHAR values prevail between (1349 and 893 cal yr BP), punctuated by a two low-magnitude peaks at 1213 and 1117 cal yr BP. CHAR values increase dramatically

after 893 cal yr BP and remain high until 717 cal yr BP. The initial charcoal peak in this interval (at 893 cal yr BP) was substantial, registering the highest magnitude peak of the last 1500 years and is followed by an overall decline in CHAR values through to 701 cal yr BP. CHAR values are variable over the next 100 years, with a moderate peak at 621 cal yr BP, followed by a substantial decrease in CHAR values to a low at 557 cal yr BP. Low CHAR values and two low-magnitude CHAR peaks (477 and 149 cal yr BP) characterise the next 460 years. A moderate CHAR peak occurs at 40 cal yr BP and CHAR values decline from 13 cal yr BP to the present. 5. Discussion 5.1. Palaeoenvironmental reconstruction

Fig. 7. Charcoal accumulation rate (CHAR) for the last 1500 years at Laguna San Pedro. Accumulation rates were decomposed into background CHAR (the slowly varying solid curve overlying the accumulation rate curve; window-width ¼ 300 yr) and peaks or fire episodes. The signal-to-noise ratio is shown. Fire episodes are charcoal events that exceed the locally defined threshold (the slowly varying dashed curve overlying the accumulation rate curve). Peak magnitude (particles cm2 peak1) measures fire size, fire intensity and/or charcoal delivery.

In-situ pelagic sedimentation of organic rich sediment has prevailed in Laguna San Pedro during the last 1500 years, punctuated by deposition of two thick coarse pyroclastic layers at ca 1350 and 1100 cal yr BP that reflect volcanic eruptions, and a silt layer at 121 cal yr BP. The pollen record from Laguna San Pedro reveals a dynamic vegetation landscape. Broadly speaking, the changes are characterised by either a trend towards increased forest importance (increasing N. dombeyi type pollen) or towards increased grass importance (increasing Poaceae pollen). To simplify these trends, we calculated a log-normalised Nothofagus/Poaceae index (NPI) in which positive (negative) excursions reflect a trend towards forest (grass) dominance. Nothofagus species that produce N. dombeyi type pollen (N. dombeyi, N. antarctica and N. pumilio only in this region) dominate vegetation communities upslope (i.e. where temperature is lower) from Laguna San Pedro up to the climatic treeline and downslope (and west) through the more humid zone (Fig. 2). There is little change in the percent abundance of N. dombeyi type in modern pollen spectra sampled along an altitudinal gradient in the local region up to 1750 m a.s.l. (the climatic treeline), above which alpine grass-dominated plant communities prevail and Poaceae pollen increases in importance (Paez et al., 1997). We can surmise from the modern pollen analysis of Paez et al. (1997), that a temperature-driven shift in the NPI would require a lowering of the climatic treeline to elevation of Laguna San Pedro today (a lowering of w800 m), which, employing an

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

53

Fig. 8. Summary plot of (a) Laguna Pallcacocha, Ecuadorian Andes alluvial events (Moy et al., 2002); (b) reconstructed SOI index (Yan et al., 2011); (c) Laguna San Pedro NPI (this study); (d) Laguna San Pedro depositional time (this study); (e) tree-ring based climate reconstruction from Northern Patagonia (Villalba, 1994); (f) pooled mean calibrated 14C ages for drought termination events synchronous between Argentina and California (Stine, 1994); (g) Laguna San Pedro statistically significant fire episodes (in red) and CHAR curve (n.b. this curve is the CHAR raw data, not the interpolated CHAR values produces by CHARAnalysis and presented in Fig. 7). Black triangles represent Laguna San Pedro radiocarbon assays. Key chronozones (LIA, MCA, DACP) are indicated. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

environmental lapse rate of 0.61
C per 100 m1 for this region (Cavieres et al., 2000), would require a temperature depression in the order of 5e6
C. In contrast, the steep west-east precipitation gradient exerts a strong influence over the present vegetation mosaic in the Lonquimay region. This precipitation gradient results in the juxtaposition of xerophytic grass-dominated plant communities against mesophytic Nothofagus-dominated plant communities (Michell, 1980) over relatively short distances and suggests that the NPI will be sensitive to shifts in moisture balance driving changes in plant community composition along a spectrum from mesophytic to xerophytic. Moreover, Chusquea species, a grass taxon that is common after disturbance, is not listed as an important component of the vegetation growing near to the Lonquimay area (Michell, 1980) and it is unlikely that disturbance-driven

changes in pollen input from species of this genus (which is characteristically large and was not well represented in the pollen record) exert a significant influence over the NPI. Given these considerations, we interpret the NPI as a proxy for moisture balance, which most likely reflects summer rainfall considering the short summer growing season, winter dormancy, and physiological drought (Roig et al., 2008) imposed by the extreme winter cooling. Importantly, no steppe or sub-desert indicator pollen types, such as Mulinum, Ephedra or Colliguaja (Paez et al., 1997), occur in any section of the Laguna San Pedro record and it is evident that the range of vegetation change around the site over the last 1500 years was mesophytic forests to the forest-steppe ecotone. In addition to the NPI, we have calculated depositional time (years cm1) over the last 1500 years at Laguna San Pedro. The

54

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

frigid temperatures and deep winter snow-lie in the study region result in lake freezing and system (and depositional) stasis through winter and we interpret changes in depositional time (Fig. 8), as a proxy for the duration of organic lake sediment accumulation over the annual cycle in the lake basin. We expect that a decrease (increase) in the duration of lake freezing will result in increased (decreased) annual lake sediment accumulation, a factor that is most likely a function of temperature. The period between ca 1500e1350 cal yr BP is characterised by negative NPI values (reflecting an increase in Poaceae relative to Nothofagus) and high charcoal content (with a significant local fire event at 1381 cal yr BP), which we interpret as relatively low moisture during the growth season (summer). A contemporary peak of Botryococcus spp. is consistent with this interpretation, as this taxon displays a propensity for warm, stagnant and shallow pelagic environments (Guy-Ohlson, 1992; Berrio et al., 2000; Medeanic, 2006; Martínez et al., 2008). A substantial drop in CHAR followed after 1349 cal yr BP is concomitant with a steep trend towards positive NPI values that indicate a prevalence of mesophytic Nothofagus-dominated plant communities over xerophytic communities in which grass taxa are more important, a trend that likely reflects a response to an increase in growing season moisture. Importantly, the positive trend in the NPI commences before and continues after the deposition of the volcanic ash layer dated at 1350 cal yr BP. Indeed this volcanic eruption had no discernible impact on the terrestrial or aquatic environment (as observable through pollen analysis). The positive trend in the NPI and increasing importance of forest under a higher relative moisture regime continues for three centuries and culminates at ca 1150 cal yr BP. An increase (decrease) in the abundance of Poaceae (Nothofagus) occurs over the next ca 300 years, reflecting a response of the vegetation to a decrease in growing season moisture (a trend to negative NPI values). Again, this trend commences prior to and continues after the tephra dated at 1100 cal yr BP, with the only apparent impact of the eruption being a small peak in A. araucana pollen ca 50 years after the deposition of the ash. The relatively slow deposition time through this period suggests a prolonged winter stasis of the pelagic environment and we infer a trend towards cool and dry conditions over the region. The sediment accumulation rate increases after 890 cal yr BP and remains well above the 1500-year long mean between 890 and 570 cal yr BP. We interpret this period of rapid sediment accumulation as reflecting a warm phase that reduced the duration of snow-lie, lake freezing and pelagic system stasis, thus allowing a substantially longer period of organic lake sediment production and accumulation through the annual cycle. The negative NPI values suggest low relative growing season moisture through this warm period, consistent with the high CHAR values (Fig. 7). A positive trend in the NPI commences at 800 cal yr BP, with values becoming strongly positive after 590 cal yr BP, implying a dense forest cover in response to increased growing season moisture, consistent with the drop in CHAR values after 589 cal yr BP. Low CHAR and increasing N. dombeyi type pollen through approximately the next four centuries (587e121 cal yr BP) implies a persistence of high relative growing season moisture. The marked increase in Potamogeton spp., an aquatic plant fond of moderate water depth and tolerant of turbid conditions (Van den Berg et al., 1999), between ca 500e50 cal yr BP (with a decline centred on ca 325 cal yr BP), suggests a possible increase in lake level and possible turbid lake conditions brought by wind-driven sediment mixing (sensu Howick and Wilhm, 1985) and/or increased runoff (the latter is not supported by an increase in inorganic content [Fig. 4]). Together, these changes suggest cool, wet and possibly windier conditions in the Lonquimay region between 587 and 121 cal yr BP.

A sharp and rapid decrease in Nothofagus dombeyi type (>20% decline), an increase in Poaceae and the occurrence of European pastoral weeds (R. acetocella and P. lanceolata) after 121 cal yr BP collectively signal deforestation of the landscape by Europeans. An increase in CHAR values at this time, with a major fire episode at 121 cal yr BP, suggests that this process was achieved with the aid of deliberate burning of the landscape during the European conquest of the local indigenous population. Interestingly, Potamogeton spp. decrease in the upper ca 50 years, possibly reflecting trend towards a less turbid (windy?) environment probably related to a southward displacement of the SWW detected during the twentieth century (Villalba et al., 2003). This notion is also supported by the further increase (decrease) in Poaceae (Nothofagus) at this time, although this may reflect further anthropogenic disturbance. The trend towards rapid sediment accumulation in the upper section of the record (Fig. 8) reflects lower degrees of compaction towards the sedimentewater interface and, possibly, increased organic productivity driven by relatively high temperatures over the twentieth century. 5.2. Centennial-scale climate trends The palaeoenvironmental history inferred from Laguna San Pedro provides important palaeoclimatic information for this part of southern South America that is poorly represented in the palaeoclimate literature. Our reconstruction includes relatively low summer moisture (negative standardised NPI) between 1500e1300 and 1000e725 cal yr BP. The latter period, while bearing in mind the constraints on the chronology, is contemporaneous with the early phase of the MCA (1050e600 cal yr BP) and supports the notion originally put forward by Stine (1994) that the MCA was manifest variously as either temperature and/or moisture signals around the globe. The earlier phase (1500e1300 cal yr BP) is contemporaneous with the latter part of the Roman Warm Period (RWP). Conversely, periods of positive NPI, reflecting relatively moist conditions during summer, occur between 1300 and 1000 cal yr BP (broadly consistent with the DACP: 1450e1050 cal yr BP) and 725e121 cal yr BP (broadly consistent with the LIA: 600e100 cal yr BP). The short summer growing season in the Lonquimay region affords a unique opportunity to reconstruct centennial-scale trends in warm-season moisture over this region. As previously discussed (Section 2.2), inter-annual trends in summer moisture in this part of southern South America over the historic period are governed by ENSO, with a seasonal breakdown revealing a summer moisture deficit (surplus) occurring during El Niño (La Niña) events. Given this clear synoptic-scale relationship between inter-annual precipitation and ENSO in the study region, we postulate that the sustained periods of low relative growing season moisture at Laguna San Pedro between 1500e1300 and 1000e725 cal yr BP reflect phases where an El Niño-like climate state persisted over the region. Likewise, the sustained periods of relatively high growing season moisture identified between 1300e1000 and 725e121 cal yr BP at Laguna San Pedro can be visualized as persistence of La Niña-like conditions in the region. If the modern synoptic-scale relationship between ENSO and the climate regime of the Lonquimay region has been stationary over the last 1500 years, and our reconstruction of summer moisture balance from Laguna San Pedro represents a regional expression of ENSO activity, then persistent periods of low (high) summer moisture should correlate with evidence of persistent El Niño (La Niña)-like synoptic patterns elsewhere in the ENSO domain. Two such reconstructions are offered by Yan et al. (2011), who present a lake-sediment-based reconstruction of the Southern Oscillation Index (SOI), and Moy et al. (2002), who present a high-

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

resolution reconstruction of heavy (El Niño driven) rainfalltriggered alluvial events recorded in Laguna Pallcacocha, in the Ecuadorian Andes. A comparison of our data with these ENSO reconstructions reveals (Fig. 8) that periods of low relative summer moisture and high CHAR at Laguna San Pedro (1500e1300 and 1000e725 cal yr BP) correspond remarkably well with sustained periods of (a) negative (El Niño) trends in the reconstructed SOI (Yan et al., 2011) and (b) higher frequency of alluvial events into Laguna Pallcacocha resulting from heavy rains during sustained periods of El Niño-like conditions (Moy et al., 2002). Furthermore, periods of relatively high summer moisture and low CHAR between 1300e1000 and 725e121 cal yr BP at Laguna San Pedro correspond remarkably well with (a) positive (La Niña) SOI trends and (b) lower frequency of alluvial events in the Laguna Pallcacocha record, suggesting low (or absent) El Niño activity. This close correspondence between our reconstructed summer moisture regime at Laguna San Pedro and two independent tropical ENSO reconstructions is entirely consistent with the modern climatology, and strongly suggests that ENSO has been the dominant hydroclimatic driver on the in the Temperate-Mediterranean Transitional (TMT) zone of central Chile over the last 1500 years. It should be noted that while our results are corroborated by the high-resolution terrestrial proxy ENSO reconstructions of Yan et al. (2011) and Moy et al. (2002), and a number of other proxy (e.g. Newton et al., 2006; Oppo et al., 2009; Tierney et al., 2010) and model-based palaeoclimate reconstructions (e.g. Held and Soden, 2006; Vecchi et al., 2006) from within the ENSO domain, they contradict the dominant paradigm of ENSO activity over the last 1500 years as inferred predominantly from sea-surface temperature and modelling data: namely that the LIA (MCA) was characterised by persistent El Niño (La Niña) like conditions (e.g. Seager et al., 2007; Mann et al., 2009; Graham et al., 2010). The evidence for warming (increased sedimentation) between 890 and 570 cal yr BP at Laguna San Pedro is synchronous with the MCA and is remarkably consistent with a nearby tree-ring summer temperature reconstruction from 41
S in northern Patagonia that found evidence for medieval warming between 870 and 700 cal yr BP (Villalba, 1994). Furthermore, Villalba (1994) found evidence for (1) cool and wet conditions between 680 and 290 cal yr BP (during the LIA), contemporaneous with increased summer moisture and cool temperatures that we have inferred at Laguna San Pedro between 570 and 121 cal yr BP (Fig. 8); and (2) cold conditions between 1050 and 880 cal yr BP (during the DACP) that is also consistent with our inference of cool temperature influencing the depositional time at Laguna San Pedro at this time (Fig. 8). One possible driver of these shifts in the thermal regime may lie in the influence of ENSO on the temperature regime over this part of southern South America: El Niño (La Niña) is significantly correlated with warmer (cooler) temperatures over the study region (Garreaud et al., 2009). This relationship is consistent with the trends identified in our hydrological reconstruction of warm El Niño-like conditions predominating between 1000 and 725 cal yr BP and cool La Niña-like conditions prevailing between 1300e1000 and 725e121 cal yr BP. Furthermore, high-magnitude statistically significant charcoal peaks at 893 and 621 cal yr BP in the Laguna San Pedro record (Fig. 7), which we interpret as local fire episodes, are contemporaneous (within the error envelopes) with the radiocarbon-dated drought terminations identified by Stine (1994) in several sites in Argentina and California (Fig. 8). These droughts constitute extreme centennial-scale variations in hydroclimatic conditions during medieval times in areas currently teleconnected to ENSO, augmenting the notion that low-frequency ENSO-like variations have driven climate change in the TMT zone during the last 1500 years.

55

6. Conclusion Our palaeoenvironmental reconstruction from Laguna San Pedro in south-central Chile (38
S) suggest that periods of high relative growing season (summer) moisture and cool temperatures correspond well with evidence for persistent cool/La Niña ENSO states (1300e1000 and 725e121 cal yr BP; DCAP and LIA, respectively). We also infer periods of low relative growing season moisture and warmer temperature that correspond well with evidence for persistent warm/El Niño ENSO states (1500e1300 and 1000e125 cal yr BP; RWP and MCA, respectively). There is a striking correlation between regional ENSO proxy reconstructions and the inferred multi-decadal to centennial-scale trends in moisture balance and temperature regime in Laguna San Pedro. Finally, we reveal a dramatic landscape alteration associated with the arrival of exotic taxa and an increase in burning, which we attribute to European colonisation of the area after 121 cal yr BP (1829 AD). These results contribute high-quality data to a region where there is a dearth of palaeoclimate information spanning the last 1500 years. Acknowledgements M.S.F. is funded by Fondecyt 3110180 and the Institute of Ecology and Biodiversity, Chile, and P.I.M by Fondecyt 1110612, and Iniciativa Científica Milenio P05-002, contract PFB-23. Our appreciation goes to Oscar Pesce and Bree Fletcher for help in the field and to Loreto Hernández for processing the palynological samples and for conducting LOI and macroscopic charcoal analyses, and to Carmen Miranda for drafting Fig. 1. References Ariztegui, D., Bosch, P., Davaud, E., 2007. Dominant ENSO frequencies during the Little Ice Age in northern Patagonia: the varved record of proglacial Lago Frías, Argentina. Quaternary International 161, 46e55. Berrio, J.C., Hooghiemstra, H., Behling, H., Van der Borg, K., 2000. Late Holocene history of savanna gallery forest from Carimagua area, Colombia. Review of Palaeobotany and Palynology 111, 295e308. Bertrand, S., Boës, X., Castiaux, J., Charlet, F., Urrutia, R., Espinoza, C., Lepoint, G., Charlier, B., Fagel, N., 2005. Temporal evolution of sediment supply in Lago Puyehue (Southern Chile) during the last 600 yr and its climatic significance. Quaternary Research 64, 163e175. Blaauw, M., 2010. Methods and code for ‘classical’ age-modelling of radiocarbon sequences. Quaternary Geochronology 5, 512e518. Boës, X., Fagel, N., 2008. Relationships between southern Chilean varved lake sediments, precipitation and ENSO for the last 600 years. Journal of Paleolimnology 39, 237e252. Bronk Ramsey, C., 2010. OxCal 4.1 Manual. Cavieres, L.A., Peñaloza, A., Kalin Arroyo, M., 2000. Altitudinal vegetation belts in the high-Andes of central Chile (33 S). Revista Chilena de Historia Natural 73, 331e344. Faegri, K., Iversen, J., 1989. Textbook of Pollen Analysis. Wiley, New York. Garreaud, R.D., Vuille, M., Compagnucci, R., Marengo, J., 2009. Present-day South American climate. Palaeogeography, Palaeoclimatology, Palaeoecology 281, 180e195. Garreaud, R.D., 2007. Precipitation and circulation covariability in the extratropics. Journal of Climate 20, 4789e4797. González, M.E., Veblen, T.T., 2009. Climatic Influences on Fire in Araucaria AraucanaeNothofagus Forests in the Andean Cordillera of South-central Chile. Gonzalez, M.E., Veblen, T.T., Sibold, J.S., 2005. Fire history of Araucaria-Nothofagus forests in Villarrica National Park, Chile. Journal of Biogeography 32, 1187e1202. Graham, N.E., Ammann, C.M., Fleitmann, D., Cobb, K.M., Luterbacher, J., 2010. Support for global climate reorganization during the “Medieval Climate Anomaly”. Climate Dynamics 2010, 1e29. Grimm, E.C., Maher, L.J., Nelson, D.M., 2009. The magnitude of error in conventional bulk-sediment radiocarbon dates from central North America. Quaternary Research 72, 301e308. Grimm, E., 1987. CONISS: a FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Computers & Geosciences 13, 13e35. Guy-Ohlson, D., 1992. Botryococcus as an aid in the interpretation of palaeoenvironment and depositional processes. Review of Palaeobotany and Palynology 71, 1e15.

56

M.-S. Fletcher, P.I. Moreno / Quaternary Science Reviews 46 (2012) 46e56

Heiri, O., Lotter, A.F., Lemcke, G., 2001. Loss on ignition as a method for estimating organic and carbonate content in sediments: reproducibility and comparability of results. Journal of Paleolimnology 25, 101e110. Held, I.M., Soden, B.J., 2006. Robust responses of the hydrological cycle to global warming. Journal of Climate 19, 5686e5699. Higuera, P.E., Peters, M.E., Brubaker, L.B., Gavin, D.G., 2007. Understanding the origin and analysis of sediment-charcoal records with a simulation model. Quaternary Science Reviews 26, 1790e1809. Higuera, P.E., Brubaker, L.B., Anderson, P.M., Brown, T.A., Kennedy, A.T., Hu, F.S., 2008. Frequent fires in ancient shrub tundra: implications of paleorecords for Arctic environmental change. PloS One 3, e0001744. Higuera, P.E., Gavin, D.G., Bartlein, P.J., Hallett, D.J., 2010. Peak detection in sedimentecharcoal records: impacts of alternative data analysis methods on firehistory interpretations. International Journal of Wildland Fire 19, 996e1014. Howick, G.L., Wilhm, J., 1985. Turbidity in Lake Carl Blackwell: Effects of Water Depth and Wind, 51e57 pp. Jara, I.A., Moreno, P.I., 2012. Temperate rainforest response to climate change and disturbance agents in northwestern Patagonia (41
S) over the last 2600 years. Quaternary Research. Jenny, B., Valero-Garcés, B.L., Villa-Martínez, R., Urrutia, R., Geyh, M., Veit, H., 2002. Early to mid-holocene aridity in Central Chile and the Southern Westerlies: the Laguna Aculeo record (34 S). Quaternary Research 58, 160e170. Kitzberger, T., Veblen, T.T., Villalba, R., 1997. Climatic influences on fire regimes along a rain forest-to-xeric woodland gradient in northern Patagonia, Argentina. Journal of Biogeography 24, 35e47. Kitzberger, T., Swetnam, T.W., Veblen, T.T., 2001. Inter hemispheric synchrony of forest fires and the El Niño Southern Oscillation. Global Ecology and Biogeography 10, 315e326. Luckman, B.H., Villalba, R., 2001. Assessing the Synchroneity of Glacier Fluctuations in the Western Cordillera of the Americas during the Last Millennium. Interhemispheric Climate Linkages. Academic Press, San Diego. 119e140. Mann, M.E., Zhang, Z., Rutherford, S., Bradley, R.S., Hughes, M.K., Shindell, D., Ammann, C., Faluvegi, G., Ni, F., 2009. Global signatures and dynamical origins of the Little Ice Age and Medieval Climate Anomaly. Science 326, 1256. Martínez, M.A., Ferrer, N.C., Asensio, M.A., 2008. Primer registro de algas dulceacuícolas del Paleógeno de la Cuenca de Ñirihuau, Argentina: descripciones sistemáticas y análisis palinofacial. Ameghiniana 45. Masiokas, M.H., Rivera, A., Espizua, L.E., Villalba, R., Delgado, S., Aravena, J.C., 2009. Glacier fluctuations in extratropical South America during the past 1000 years. Palaeogeography, Palaeoclimatology, Palaeoecology 281, 242e268. Mc Queen, D.R., 1977. The ecology of Nothofagus and associated vegetation in South America. Tuatara 22, 245e265. 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. Medeanic, S., 2006. Freshwater algal palynomorph records from Holocene deposits in the coastal plain of Rio Grande do Sul, Brazil. Review of Palaeobotany and Palynology 141, 83e101. Michell, R.G., 1980. Vegetacion de bosque de Araucaria araucana (Mol.) K. Koch en la Cordillera de Los Andes (Lonquimay Prov. Malleco). Universidad de Chile, Facultad de Ciencias Forestales. Montecinos, A., Aceituno, P., 2003. Seasonality of the ENSO-related rainfall variability in central Chile and associated circulation anomalies. Journal of Climate 16, 281e296. Moreno, P.I., Francois, J.P., Villa-Martinez, R.P., Moy, C.M., 2009. Millennial-scale variability in Southern Hemisphere westerly wind activity over the last 5000 years in SW Patagonia. Quaternary Science Reviews 28, 25e38. 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 Holocene. Nature 420, 162e165. Moy, C.M., Moreno, P.I., Dunbar, R.B., Kaplan, M.R., Francois, J.P., Villalba, R., Haberzettl, T., 2009. Climate Change in Southern South America during the Last Two Millennia. In: Past Climate Variability in South America and Surrounding Regions: from the Last Glacial Maximum to the Holocene, vol. 353.

Neukom, R., Luterbacher, J., Villalba, R., Küttel, M., Frank, D., Jones, P.D., Grosjean, M., Esper, J., Lopez, L., Wanner, H., 2010a. Multi centennial summer and winter precipitation variability in southern South America. Geophysical Research Letters 37, L14708. Neukom, R., Luterbacher, J., Villalba, R., Küttel, M., Frank, D., Jones, P.D., Grosjean, M., Wanner, H., Aravena, J.C., Black, D.E., 2010b. Multiproxy summer and winter surface air temperature field reconstructions for southern South America covering the past centuries. Climate Dynamics, 1e17. Newton, A., Thunell, R., Stott, L., 2006. Climate and hydrographic variability in the Indo-Pacific Warm Pool during the last millennium. Geophysical Research Letters 33, L19710. Oppo, D.W., Rosenthal, Y., Linsley, B.K., 2009. 2,000-year-long temperature and hydrology reconstructions from the Indo-Pacific warm pool. Nature 460, 1113e1116. Paez, M.M., Villagrán, C., Stutz, S., Hinojosa, F., Villa, R., 1997. Vegetation and pollen dispersal in the subtropical-temperate climatic transition of Chile and Argentina. Review of Palaeobotany and Palynology 96, 169e181. Peralta, P., 1980. Geomorfología, clima y suelos del tipo forestal Araucaria en Lonquimay. Boletín Técnico, Facultad de Ciencias Forestales, Universidad de Chile. Roig, F.A., Villalba, R., Rabassa, J., 2008. Understanding Climate from Patagonian Tree Rings, Developments in Quaternary Science. Elsevier. 411e435. Rondanelli-Reyes, M.J., 2000. Historia vegetacional del bosque andino de Araucaria araucana (Molina) K. Koch, en la cuenca del Alto Valle del río Biobío, Provincia de Lonquimay, Chile centro-sur, durante el Holoceno. Análisis palinológico del perfil Miraflores 2, 1041e1051. Seager, R., Graham, N., Herweijer, C., Gordon, A.L., Kushnir, Y., Cook, E., 2007. Blueprints for Medieval hydroclimate. Quaternary Science Reviews 26, 2322e2336. Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M., Miller, H.L., 2007. Climate Change 2007: the Physical Science Basis. Cambridge University Press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi. Stine, S., 1994. Extreme and persistent drought in California and Patagonia during medieval time. Nature 369, 546e549. Tierney, J.E., Oppo, D.W., Rosenthal, Y., Russell, J.M., Linsley, B.K., 2010. Coordinated hydrological regimes in the Indo-Pacific region during the past two millennia. Paleoceanography 25, PA1102. Van den Berg, M.S., Scheffer, M., Van Nes, E., Coops, H., 1999. Dynamics and stability of Chara sp. and Potamogeton pectinatus in a shallow lake changing in eutrophication level. Hydrobiologia 408, 335e342. Veblen, T.T., Kitzberger, T., Villalba, R., Donnegan, J., 1999. Fire history in northern Patagonia: the roles of humans and climatic variation. Ecological Monographs 69, 47e67. Vecchi, G.A., Soden, B.J., Wittenberg, A.T., Held, I.M., Leetmaa, A., Harrison, M.J., 2006. Weakening of tropical Pacific atmospheric circulation due to anthropogenic forcing. Nature 441, 73e76. 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 temperature changes across the southern Andes: 20th-century variations in the context of the past 400 years. Climatic Change 59, 177e232. Villalba, R., 1994. Tree-ring and glacial evidence for the medieval warm epoch and the little ice age in southern South America. Climatic Change 26, 183e197. von Gunten, L., Grosjean, M., Rein, B., Urrutia, R., Appleby, P., 2009. A quantitative high-resolution summer temperature reconstruction based on sedimentary pigments from Laguna Aculeo, central Chile, back to AD 850. The Holocene 19, 873. Whitlock, C., Larsen, C.P.S., 2001. Charcoal as a Fire Proxy. Tracking Environmental Change Using Lake Sediments. Kluwer Academic Publishers, Dordrecht, The Netherlands. Yan, H., Sun, L., Wang, Y., Huang, W., Qiu, S., Yang, C., 2011. A record of the Southern Oscillation Index for the past 2,000 years from precipitation proxies. Nature Geoscience 4, 611e614.