CPClimate of the PastCPClim. Past1814-9332Copernicus PublicationsGöttingen, Germany10.5194/cp-13-879-2017The last glacial termination on the eastern flank of the central Patagonian
Andes (47∘ S)HenríquezWilliam I.Villa-MartínezRodrigoVilanovaIsabelDe Pol-HolzRicardoMorenoPatricio I.pimoreno@uchile.clhttps://orcid.org/0000-0002-1333-6238Victoria University of Wellington, Wellington, New ZealandInstituto de Ecología y Biodiversidad, Departamento de Ciencias
Ecológicas, Universidad de Chile, Casilla 653, Santiago, ChileGAIA-Antártica, Universidad de Magallanes, Avda. Bulnes 01855,
Punta Arenas, ChileMuseo Argentino de Ciencias Naturales Bernardino
Rivadavia, Avda. Angel Gallardo 470, Buenos Aires, ArgentinaPatricio I. Moreno (pimoreno@uchile.cl)14July20171378798952September201614September20162June20177June2017This work is licensed under the Creative Commons Attribution 3.0 Unported License. To view a copy of this licence, visit https://creativecommons.org/licenses/by/3.0/This article is available from https://cp.copernicus.org/articles/13/879/2017/cp-13-879-2017.htmlThe full text article is available as a PDF file from https://cp.copernicus.org/articles/13/879/2017/cp-13-879-2017.pdf
Few studies have examined in detail the sequence of events during the last
glacial termination (T1) in the core sector of the Patagonian Ice Sheet
(PIS), the largest ice mass in the Southern Hemisphere outside of Antarctica.
Here we report results from Lago Edita (47∘8′ S,
72∘25′ W, 570 m a.s.l.), a small closed-basin lake located in a
valley overridden by eastward-flowing Andean glaciers during the Last Glacial
Maximum (LGM). The Lago Edita record shows glaciolacustrine sedimentation
until 19 400 yr BP, followed by organic sedimentation in a closed-basin lake and a mosaic of cold-resistant hygrophilous
conifers and rainforest trees, along with alpine herbs between
19 400 and 11 000 yr BP. Our data suggest that the PIS retreated at least
∼ 90 km from its LGM limit between ∼ 21 000 and 19 400 yr BP
and that scattered, low-density populations of cold-resistant hygrophilous
conifers, rainforest trees, high-Andean and steppe herbs thrived east of the
Andes during the LGM and T1, implying high precipitation levels and southern
westerly wind (SWW) influence at 47∘ S. The conifer
Podocarpus nubigena increased between 14 500 and 13 000 yr BP,
suggesting even stronger SWW influence during the Antarctic Cold
Reversal, after which it declined and persisted until 11 000 yr BP. Large
increases in arboreal pollen at ∼ 13 000 and ∼ 11 000 yr BP
led to the establishment of forests near Lago Edita between
10 000 and 9000 yr BP, suggesting a rise in the regional tree line along the
eastern Andean slopes driven by warming pulses at ∼ 13 000 and
∼ 11 000 yr BP and a subsequent decline in SWW influence at
∼ 11 000 yr BP. We propose that the PIS imposed a regional cooling
signal along its eastern, downwind margin through T1 that lasted until the
separation of the northern and southern Patagonian ice fields along the Andes during
the Younger Dryas period. We posit that the withdrawal of glacial and
associated glaciolacustrine environments through T1 provided a route for the
dispersal of hygrophilous trees and herbs from the eastern flank of the
central Patagonian Andes, contributing to the afforestation of the western
Andean slopes and pacific coasts of central Patagonia during T1.
Sketch map of the study area showing the location of central-western
Patagonia, the position of Valle Chacabuco relative to the Río Blanco,
María Elena, Lago Columna (LC) and Lago Posada (LP) ice limits east
Lago Cochrane, and the northern Patagonian ice field and the Taitao Peninsula to
the west. We also included Sierra Colorado, Lago Esmeralda and Cerro Oportus
for reference. The lower portion of the figure shows a detail on the Cerro
Tamango area and the portion of Valle Chacabuco where Lago Edita and Lago
Augusta are located. Also shown are palynological sites discussed in the
main text (Canal de la Puntilla, Huelmo, Mallín Lago Shaman,
Mallín Pollux, Lago Stibnite, Lago Augusta).
Introduction
The Patagonian Ice Sheet (PIS) was the largest ice mass in the Southern
Hemisphere outside of Antarctica during the last glacial maximum (LGM). Outlet
lobes from the PIS flowed westward into the Pacific coast south of
43∘ S and eastward toward the extra-Andean Patagonian plains,
blanketing a broad range of environments and climatic zones across and along
the Andes. Land biota from formerly ice-free sectors underwent local
extinction or migrated toward the periphery of the advancing PIS during the
last glaciation until its culmination during the LGM. The PIS then underwent
rapid recession and thinning through the last glacial termination
(termination 1 = T1: between ∼ 18 000 and 11 000 yr BP) toward the
Andes as illustrated by stratigraphic, geomorphic and radiocarbon-based
chronologies from northwestern Patagonia (39–43∘ S) (Denton et al.,
1999; Moreno et al., 2015). These data, along with the Canal de la
Puntilla-Huelmo pollen record (∼ 41∘ S) (Moreno et al., 2015)
(Fig. 1), indicate abandonment from the LGM margins in the lowlands at
17 800 yr BP, abrupt arboreal expansion and accelerated retreat that
exposed Andean cirques located above 800 m a.s.l. within 1000 years or less
in response to abrupt warming. Similarly, glaciers from Cordillera Darwin
(54–55∘ S), the southernmost ice field in South America, underwent
rapid recession from their LGM moraines located in central and northern
Tierra del Fuego prior to 17 500 yr BP and led to ice-free conditions by
16 800 yr BP near the modern ice fronts (Hall et al., 2013). Sea surface
temperature records from the SE Pacific (Caniupán et al., 2011) are
consistent with these terrestrial records; however, their timing, structure,
magnitude and rate of change may be overprinted by the vicinity of former ice
margins and shifts in marine reservoir ages (Caniupán et al., 2011; Siani
et al., 2013).
In contrast, in the Andean sector of
central-western Patagonia (45–48∘ S) very few studies have been conducted about the timing of glacial
advances near the end of the LGM or the structure and chronology of
glacial retreat and climate changes during T1. Recent chronologies include
cosmogenic radionuclides of terminal moraines of the Río Blanco,
recessional moraines deposited by the Lago Cochrane ice lobe (LCIL) (Boex et
al., 2013; Hein et al., 2010) (Fig. 1) and optically stimulated luminescence
dating of glaciolacustrine beds associated with the glacial Lake Cochrane (GLC)
(47∘ S) (Glasser et al., 2016). These studies reported ages between
29 000 and 19 000 yr BP for the final LGM advance and drainage of GLC toward
the Pacific between 13 000 and 8000 yr BP caused by the breakup of the northern and
southern Patagonian ice fields during the final stages of T1 (Turner et al.,
2005). Palynological interpretations from the Lago Shaman
(44∘26′ S, 71∘11′ W, 919 m a.s.l.) and Mallín
Pollux (45∘41′ S, 71∘50′ W, 640 m a.s.l.) sites
(de Porras et al., 2012; Markgraf et al., 2007), located east of the Andes
(Fig. 1), indicate the predominance of cold and dry conditions during T1 and
reduced southern westerly wind (SWW) influence. The validity and regional
applicability of these stratigraphic, chronological and palynological
interpretations, however, awaits replication with detailed
stratigraphic and geomorphic data from sensitive sites constrained by precise
chronologies.
Paleoclimate simulations (Bromwich et al., 2004, 2005) and stratigraphic
studies (Kaufman et al., 2004) in the periphery of the Laurentide Ice Sheet
in North America have detected that large ice sheets exerted important
impacts on the thermal structure and atmospheric circulation on regional,
continental and zonal scales from the LGM to the early Holocene. This aspect
has remained largely unexplored in the PIS region and might be a factor of
importance for understanding the dynamics of the SWW and
climatic and biogeographic heterogeneities through T1 on a regional scale. Progress
in this field requires an understanding of the deglacial chronology of the PIS and
a suite of sensitive paleoclimate sites across and along the residual ice
masses through the last transition from extreme glacial to extreme
interglacial conditions.
In this study we report high-resolution pollen and macroscopic charcoal
records from sediment cores we collected from Lago Edita (47∘8′ S,
72∘25′ W, ∼ 570 m a.s.l.), a medium-sized closed-basin
lake (radius ∼ 250 m) located in Valle Chacabuco ∼ 16 km
northeast of the Cochrane township, east of the central Patagonian Andes
(Fig. 1). The relevant source area for pollen from lakes of this size is
about 600–800 m from the lake's edge, according to numerical simulations
using patchy vegetation landscapes (Sugita, 1994). Stratigraphic and
chronological results from Valle Chacabuco are important for elucidating the
timing and rates of deglaciation in this core region of the PIS because this
valley is located approximately 90 km upstream from the LGM
moraines deposited by LCIL east of Lago Cochrane relative to the modern ice
fronts, and its elevation spans the highest levels of GLC during T1. The Lago
Edita data allow the assessment of vegetation and fire-regime and climate changes
during the last global transition from extreme glacial to extreme
interglacial conditions in central-western Patagonia. The aim of this paper is
to contribute to (1) the development of a recessional chronology of the
LCIL and (2) regressive phases of GLC, (3) document the composition and
geographic shifts of the glacial and deglacial vegetation, (4) understand
the tempo and mode of vegetation and climate changes during T1 and the early
Holocene, (5) constrain the regional climatic influence of the PIS during
T1 in terrestrial environments, and (6) improve our understanding of the
biogeography of the region, including the identification of possible
dispersal routes of tree taxa characteristic of modern evergreen forests in
central-western Patagonia during T1.
Study area
Central Chilean Patagonia, i.e., the Aysén region
(43∘45′–47∘45′ S), includes numerous channels, fjords,
islands and archipelagos along the Pacific side, attesting to tectonic
subsidence of Cordillera de la Costa and intense glacial erosion during the
Quaternary. The central sector features an intricate relief associated with the
Patagonian Andes with summits surpassing 3000 m a.s.l., deep valleys, lakes
of glacial origin, and active volcanoes such as Hudson, Macá, Cay,
Mentolat and Melimoyu (Stern, 2004). The Andes harbors numerous glaciers and
the northern Patagonian ice field (Fig. 1), which acted as the source for
multiple outlet glacier lobes that coalesced with glaciers from the southern
Patagonian ice field to form the PIS during Quaternary glaciations, that blocked
the drainage toward the Pacific, funneling large volumes of glacial meltwater
toward the Atlantic (Turner et al., 2005). Farther to the east the landscape
transitions into the back-arc extra-Andean plains and plateaus.
Patagonia is ideal for studying the paleoclimate evolution of the southern
midlatitudes, including past changes in the SWW, because it is the sole
continental landmass that intersects the low and mid-elevation zonal
atmospheric flow south of 47∘ S. Orographic rains associated with
storms embedded in the SWW enhance local precipitation caused by the ascent
of moisture-laden air masses along the western Andean slopes, giving way to
subsidence and acceleration of moisture-deprived winds along the eastern
Andean slopes (Garreaud et al., 2013). This process accounts for a steep
precipitation gradient across the Andes, illustrated by the annual
precipitation measured in the coastal township of Puerto Aysén
(2414 mm year-1) and the inland Balmaceda (555 mm year-1)
(http://explorador.cr2.cl/), localities separated by ∼ 80 km
across the west to east axis of the Andes. The town of Cochrane, located
∼ 15 km south of our study site features annual precipitation of
680 mm year-1 and a mean annual temperature of 7.8 ∘C (Fig. 1).
Weather station and reanalysis data along western Patagonia show positive
correlations between zonal wind speed and local precipitation, a
relationship that extends to sectors adjacent to the eastern slopes of the
Andes (Garreaud et al., 2013; Moreno et al., 2014). Therefore, changes in
local precipitation in the Aysén region are good diagnostics for
atmospheric circulation changes associated with the frequency and intensity of
storms embedded in the SWW over a large portion of the southeast Pacific.
This relationship can be applied to paleoclimate records from central
Chilean Patagonia for inferring the behavior of the SWW on the basis of past
changes in precipitation or hydrologic balance.
The steep precipitation gradient, in conjunction with adiabatic cooling and
enhanced continentality toward the east, influences the distribution and
composition of the vegetation, inducing altitudinal, latitudinal and
longitudinal zonation of plant communities throughout the Patagonian Andes.
Physiognomic and floristic studies (Gajardo, 1994; Luebert and Pliscoff,
2006; Pisano, 1997; Schmithüsen, 1956) have recognized five units or
communities, which we characterize succinctly in the following sentences:
Magellanic moorland is a unit that occurs in maritime sectors with high
precipitation (3000–4000 mm year-1 and low seasonality) along the
islands, fjords and channels. It is dominated by cushion-forming plants such
as Donatia fascicularis, Astelia pumila and
Tetroncium magallanicum. Also present are the hygrophilous
cold-resistant trees Nothofagus betuloides and the conifers
Pilgerodendron uviferum, Lepidothamnus fonkii and
Podocarpus nubigena.
Evergreen rainforest is present in humid, temperate (1500–3000 mm year-1;
< 600 m a.s.l.) sectors of Aysén. This unit is characterized by the
trees Nothofagus nitida, N. betuloides, Drimys winteri and P. nubigena, along with P. uviferum in waterlogged
environments.
Winter deciduous forests are located in cooler and/or drier
sectors with higher seasonality (400–1000 mm year-1;
500–1180 m a.s.l.). The dominant tree is Nothofagus pumilio,
which intermingles with N. betuloides at western sites and the
Patagonian steppe eastward. In the latter N. pumilio forms
monospecific stands and presents a species-poor understory. A study of the
spatial and temporal variation in N. pumilio growth at the tree line
along its latitudinal range (35∘40′–55∘ S) in the Chilean
Andes (Lara et al., 2005) showed that temperature has a spatially larger
control on tree growth than precipitation and that this influence is
particularly significant in the temperate Andes (> 40∘ S). These
results suggest that low temperatures are the main limiting factor for the
occurrence of woodlands and forests at high elevations in the Andes,
considering that precipitation increases with elevation at any given latitude
(Lara et al., 2005). The modern tree line near Cochrane is dominated by
N. pumilio and lies between 800 and 1180 m a.s.l.
Patagonian steppe occurs in substantially drier (< 500 mm year-1)
lowland areas with heightened continentality. This unit is dominated by herbs
of the families Poaceae (Festuca, Deschampsia, Stipa, Hordeum, Rytidosperma, Bromus, Elymus) and Rubiaceae (Galium) and shrubs of
the families Apiaceae (Mulinum), Rosaceae (Acaena), Fabaceae
(Adesmia) and Rhamnaceae (Discaria).
High-Andean
desert occurs in the windswept montane environments above the tree line
(> 1000 m a.s.l.) under cold conditions, a high-precipitation regime and
prolonged snow cover throughout the year. This vegetation unit is represented
by herbs of the families Poaceae (Poa, Festuca), Asteraceae
(Nassauvia, Senecio, Perezia), Berberidaceae
(Berberis), Brassicaceae (Cardamine), Santalaceae
(Nanodea), Rubiaceae (Oreopolus) Apiaceae
(Bolax) and
Ericaceae (Gaultheria, Empetrum), along with Gunnera magellanica and Valeriana, with occasional patches of
Nothofagus antarctica.
Materials and methods
We collected overlapping sediment cores over the deepest sector of Lago Edita
(8 m water depth) from an anchored coring rig equipped with 10 cm diameter
aluminum casing tube, using a 5 cm diameter Wright piston corer and a
7.5 cm diameter sediment–water interface piston corer with a transparent
plastic chamber. We characterized the stratigraphy through visual
descriptions, digital X radiographs to identify stratigraphic structures and
loss on ignition to quantify the amount of each organic, carbonate and
siliciclastic component in the sediments (Heiri et al., 2001).
The chronology of the record is constrained by accelerator mass spectrometry (AMS) radiocarbon dates on bulk
sediment and chronostratigraphic correlation of the H1 tephra from Mount
Hudson (Stern et al., 2016). The radiocarbon dates were calibrated to
calendar years before present (yr BP) using the CALIB 7.0 program. We
developed a Bayesian age model using the Bacon package for R (Blaauw and
Christen, 2011) to assign interpolated ages and confidence intervals for each
level analyzed.
We processed and analyzed continuous and contiguous sediment samples
(2 cm3) for pollen and fossil charcoal. The samples were processed
using a standard procedure that includes 10 % KOH, sieving with a
120 µm mesh, 46 % hydrofluoric acid (HF) and acetolysis (Faegri and Iversen, 1989).
We counted between 200 and 300 pollen grains produced by trees, shrubs and herbs
(terrestrial pollen) for each palynological sample and calculated the percent
abundance of each terrestrial taxon relative to this sum. The percentage of
aquatic plants was calculated in reference to the total pollen sum
(terrestrial plus aquatic pollen) and the percentage of ferns from the total
pollen and spores sum. Zonation of the pollen record was aided by a
stratigraphically constrained cluster analysis on all terrestrial pollen taxa
with ≥ 2 %, after recalculating sums and percentages.
Stratigraphic column, radiocarbon dates and loss-on-ignition data
from the Lago Edita record. The labels on the right indicate the identity
and stratigraphic span (dashed horizontal lines) of each core segment.
We identified the palynomorphs based on a modern reference collection housed
at the laboratory of Quaternary paleoecology of Universidad de Chile, along
with published descriptions and keys (Heusser, 1971). In most cases the
identification was done at family or genus level, in some cases at the
species level (Podocarpus nubigena, Drimys winteri, Gunnera magellanica, Lycopodium magellanicum). The palynomorph
Nothofagus dombeyi type includes the species N. antarctica,
N. pumilio, N. betuloides and N. dombeyi. The
morphotype Fitzroya/Pilgerodendron includes the cupressaceous
conifers Fitzroya cupressoides and Pilgerodendron uviferum.
We calculated running means of selected pollen taxa using a triangular
weighing function of values along seven adjacent levels.
We tallied microscopic (< 120 µm) and macroscopic
(> 106 µm) charcoal particles to document regional and local
fire events, respectively. Microscopic particles were counted from each
pollen slide, while macroscopic charcoal was counted from 2 cm3
sediment samples obtained from 1 cm thick and
continuous and contiguous
sections. The samples were prepared using a standard procedure that involves
deflocculation in 10 % KOH and careful sieving through 106 and
212 µm diameter meshes to avoid rupture of individual particles,
followed by visual inspection on a Zeiss KL 1500 LCD stereoscope at
10× magnification. These results were analyzed by a time series
analysis to detect local fire events using the CharAnalysis software (Higuera
et al., 2009), interpolating samples at regular time intervals based in the
median time resolution of the record. We deconvoluted the CHAR signal into a
peaks and background component using a lowess robust-to-outlier smoothing
with a 100-year window width. We calculated locally defined thresholds to
identify statistically significant charcoal peaks or local fire events (99th
percentile of a Gaussian distribution).
Results
The sediment stratigraphy (Fig. 2) reveals a basal unit of blue-grey mud
between 1726 and 819 cm, horizontally laminated for the most part, in some
sectors massive and sandier with small amounts of granule and gravel immersed
in a clayey matrix (segment PC0902AT9). These inorganic clays are overlain by
organic silt between 819 and 678 cm and organic-rich lake mud (gyttja) in
the topmost 678 cm. We found laminated authigenic carbonates between
794–759 and 394–389 cm (range: 5–20 %); for the remainder of the
record carbonate values are negligible or null (< 5 %). The record
includes two tephras between 630–628 and 661–643 cm, which exhibit sharp
horizontal contacts with the over and underlying mud and, consequently, we
interpret them as aerial fallout deposits from explosive events originating
from Mount Hudson (H1 tephra) and from Volcán Mentolat (M1 tephra)
based on geochemical data (Stern et al., 2016).
Age model of the Lago Edita record, with the blue zones representing the
probability distribution of the calibrated radiocarbon dates and the grey zone
representing the calculated confidence interval of the Bayesian age model.
Radiocarbon dates from the Lago Edita core. The radiocarbon dates
were calibrated to calendar years before present using the CALIB 7.0
program.
The radiocarbon results show an approximately linear increase in age with
depth between 19 000 and 9000 yr BP (Fig. 3), which, in conjunction with
the sediment stratigraphy, suggests undisturbed in situ pelagic deposition of
lake mud and tephras in the Lago Edita basin. This study focuses on the
interval between 19 000 and 9000 yr BP (Fig. 2, Table 1) and consists of
155 contiguous palynological and macroscopic charcoal levels with a median
time step of 65 years between analyzed samples.
Pollen stratigraphy
We divided the record in six zones based on conspicuous changes in the pollen stratigraphy and a
stratigraphically constrained cluster analysis to facilitate its description and
discussion (Fig. 4). The following
section describes each pollen zone, indicating the stratigraphic and
chronological range, and the mean abundance of major taxa in parentheses.
Zone Edita-1 (795–780 cm; 19 000–18 100 yr BP) is co-dominated by
Poaceae (33 %) and Empetrum (32 %). This zone starts with a
gradual increase in Empetrum, attaining its maximum abundance
(∼ 53 %) at the end of this zone. Asteraceae subfamily Asteroideae
(7 %), Acaena (4 %), Caryophyllaceae (3 %) and
Cyperaceae (9 %) decrease, while Poaceae shows fluctuations in its
abundance between 2 and 16 % over the entire interval. Other herbs and
shrubs such as Ericaceae (3 %), Phacelia (∼ 2 %),
Valeriana (1 %), Gunnera magellanica (∼ 2 %),
Apiaceae (< 1 %), and Asteraceae subfamily Cichorioideae
(< 1 %) remain relatively steady. The arboreal taxa N. dombeyi type (10 %), Fitzroya/Pilgerodendron (2 %),
P. nubigena (< 1 %) and D. winteri (< 1 %)
are present in low abundance, as well as the ferns L. magellanicum
(∼ 1 %) and Blechnum type (5 %) and the
green microalgae Pediastrum (2 %).
Percentage pollen diagrams from the Lago Edita core. The labels on
the right indicate the identity and stratigraphic span (dashed horizontal
lines) of each pollen assemblage zone. The black dots indicate the presence of
Drimys winteri pollen grains, normally < 2 %.
Zone Edita-2 (780–758 cm; 18 100–16 800 yr BP) begins with a decline
in Empetrum (30 %) and an increase in Poaceae (34 %),
followed by its decrease until the end of this zone. N. dombeyi type
(15 %), Caryophyllaceae (5 %) and Asteraceae subfamily Asteroideae
(5 %) show a rising trend in this zone, while other arboreal taxa
(Fitzroya/Pilgerodendron (3 %), P. nubigena
(< 1 %) and D. winteri (< 1 %) and most of the herbs
maintain an abundance similar to that of the previous zone. L. magellanicum
(2 %) and Pediastrum (4 %) rise slightly, and high
variability in Cyperaceae (7 %) is shown.
Zone Edita-3 (758–701 cm; 16 800–13 200 yr BP) is characterized by a
sharp rise in Poaceae (45 %) and a declining trend in Empetrum
(15 %). The conifer P. nubigena (2 %) starts a sustained
increase, while N. dombeyi type (13 %) and
Fitzroya/Pilgerodendron (3 %) remain relatively invariant.
D. winteri (< 1 %) and Misodendrum (< 1 %), a
mistletoe that grows on the Nothofagus species, appear in low abundance
in an intermittent manner. Pediastrum (30 %) shows a rapid
increase until 15 600 yr BP, followed by considerable variations in its
abundance until the end of this zone (between 19 and 55 %). L. magellanicum (3 %) shows a steady increase, while Blechnum type
(6 %) remains invariant and Cyperaceae (7 %) exhibits large
fluctuations superimposed upon a declining trend.
Zone Edita-4 (701–681 cm; 13 200–11 600 yr BP) starts with an increase
in N. dombeyi type (29 %) and a minor rise in
Misodendrum (1 %). P. nubigena (5 %) starts this
zone with variability and stabilizes toward the end, concurrent
with Fitzroya/Pilgerodendron (3 %) and traces of D. winteri (< 1 %). Poaceae (38 %) shows a steady decrease, while
Empetrum (6 %) continues with a declining trend that started
during the previous zone. Asteraceae subfamily Asteroideae (5 %) and
Caryophyllaceae (2 %) decrease, L. magellanicum (3 %),
Cyperaceae (4 %) and Pediastrum (24 %) decline gradually
with considerable fluctuations, while Blechnum- type (11 %)
shows modest increases.
Zone Edita-5 (681–674 cm; 11 600–11 100 yr BP) shows marked declines
in N. dombeyi type (27 %) and Poaceae (33 %) in concert with
a noticeable increase in the conifers Fitzroya/Pilgerodendron
(12 %) and P. nubigena (9 %), which reach their peak abundance
in the record. The abundance of herbs and shrubs decreases or remains steady,
with the exception of an ephemeral increase in Phacelia (3 %).
Blechnum type (39 %) shows a remarkable increase to its peak
abundance in the entire record, while L. magellanicum (3 %),
Cyperaceae (8 %) and Pediastrum (17 %) rise slightly.
Zone Edita-6 (674–640 cm; 11 100–8940 yr BP) is characterized by an
abrupt increase in N. dombeyi type (62 %) and
Misodendrum (2 %), along with a noticeable decline in
Fitzroya/Pilgerodendron (2 %) and P. nubigena (2 %)
at the beginning of this zone. Poaceae (26 %) shows a downward trend over
this period, while other herbs and shrubs (Empetrum, Ericaceae,
Caryophyllaceae, Asteraceae subfamily Asteroideae, Acaena,
Phacelia, Valeriana, Gunnera magellanica, Apiaceae
and Asteraceae subf. Cichorioideae) show their lowest abundance in the
record. Blechnum type (7 %) drops sharply, followed by a gradual
decline in concert with L. magellanicum (1 %). Cyperaceae
(7 %) and Pediastrum (6 %) show initial declines followed by
increases toward the end of this zone.
Charcoal stratigraphy
The record from Lago Edita shows absence of macroscopic charcoal particles
between 19 000 and 14 300 yr BP followed by an increase in charcoal
accumulation rate (CHAR) that led to a variable plateau between 13 200 and
12 000 yr BP, a 1000-year-long decline, and a sustained increase led to
peak abundance at 9700 yr BP. Charcoal values then declined rapidly to
intermediate levels by 9000 yr BP. We note a close correspondence between
the arboreal pollen abundance (as a percentage) and the CHAR, suggesting that charcoal
production was highly dependent upon quantity and spatial continuity of
coarse woody fuels in the landscape (Fig. 5).
Time series analysis of the macroscopic charcoal record revealed 11
statistically significant peaks we interpret as local fire events within the
Lago Edita watershed (Fig. 5). The temporal structure of these events
indicates a sequence of millennial-scale peaks in fire frequency, with maxima
at 14 100, 13 100, 12 000, 10 900 and 9600 yr BP. We observe a steady
increase in the fire frequency maxima from 14 100 to 10 900 yr BP
(Fig. 5).
Macroscopic charcoal record from the Lago Edita core and results
of CharAnalysis: the blue line is the background component, the red line is the locally defined
threshold, triangles are statistically significant charcoal peaks and magnitude
is the
residual abundance that supersedes the threshold. CHAR is the charcoal accumulation rate.
Selected palynomorph abundance of the Lago Edita record shown in
the timescale domain. The red lines correspond to weighted running means of
seven adjacent samples with a triangular filter. The taxa shown in the left
panel are characteristic of humid environments currently found in sectors
adjacent to the Pacific coast and/or the Andean tree line in the study area.
The taxon Nothofagus dombeyi type, which includes multiple species with contrasting climatic
tolerances, is also found in (relatively) humid sectors east of the Andes.
The herbs and shrubs shown in the right panel are either cosmopolitan or
present in the Patagonian steppe and sectors located at or above the Andean
tree line in central-western Patagonia.
DiscussionPaleovegetation and paleoclimate
Given the size of Lago Edita (radius ∼ 250 m) its pollen record is
adequate to reflect local vegetation within 600–800 m from the lake's edge.
An extra-local component is also present considering that species of the
genus Nothofagus also produce large quantities of pollen grains
susceptible to long-distance transport (Heusser, 1989). These attributes
suggest that the Lago Edita fossil pollen record might be a good sensor of
the vegetation located on the western end of Valle Chacabuco and the Lago
Cochrane basin. The record (Figs. 4, 6) documents dominance of herbs and
shrubs (chiefly Poaceae, Empetrum and Asteraceae, accompanied by
Caryophyllaceae, Acaena, Ericaceae, Phacelia,
Valeriana, and Apiaceae in lower abundance) found above the modern
tree line and the Patagonian steppe between 19 000 and 11 000 yr BP,
followed by increasing Nothofagus. We interpret this as the establishment
of scrubland (∼ 13 000–11 000 yr BP), woodland
(∼ 11 000–10 500 yr BP) and forest
(∼ 10 500–9000 yr BP). Within the interval dominated by
non-arboreal taxa, we distinguish an initial phase with abundant
Empetrum between 19 000 and 16 800 yr BP, followed by
diversification of the herbaceous assemblage and predominance of Poaceae
during the interval ∼ 16 800–11 000 yr BP (Figs. 4, 6). This
change is contemporaneous with a sustained rise in P. nubigena and
the mistletoe Misodendrum coeval with conspicuous increases in
Lycopodium magellanicum and the green microalgae Pediastrum.
We emphasize the continuous presence of the arboreal Nothofagus and
Fitzroya/Pilgerodendron in low but constant abundance (∼ 15
and ∼ 3 %, respectively) between 19 000 and 13 000 yr BP, along
with traces (< 2 %) of hygrophilous trees (Podocarpus nubigena, Drimys winteri) and herbs (Gunnera magellanica,
Lycopodium magellanicum) accounting, in sum, for a persistent
∼ 25 % of the pre-13 200 yr BP pollen record (Figs. 4, 6). We
note that the Nothofagus parkland on the western end of Valle
Chacabuco and the Lago Cochrane basin must have approached the vicinity of
Lago Edita at 16 800 yr BP, judging from the appearance of
Misodendrum during that time (Figs. 4, 6) under relatively constant
mean Nothofagus abundances.
The conifer Podocarpus nubigena remained in low abundance
(< 2 %) prior to ∼ 14 500 yr BP in the Lago Edita record,
increased between 14 500 and 13 000 yr BP, experienced a variable decline
between 13 000 and 11 800 yr BP, reached a maximum between 11 800 and
11 200 yr BP, and declined between 11 200 and 10 200 yr BP
(Figs. 4, 6). This cold-resistant hygrophilous tree is commonly found in
temperate evergreen rainforests along the Pacific coast of central Patagonia
and is currently absent from the eastern Andean foothills at the same
latitude. Its presence and variations in the Lago Edita record suggest an
increase in precipitation relative to the pre-14 500 yr BP conditions,
with millennial-scale variations starting at ∼ 13 000 yr BP. The
variable decline in P. nubigena at 13 000 yr BP coincided with an
increase in Nothofagus that led to a variable plateau of
∼ 30 % between 13 000 and 11 200 yr BP, which we will discuss in the
following paragraphs.
The mixed palynological assemblage between ∼ 19 400 and 11 000 yr BP
has no modern analogues in the regional vegetation (Luebert and Pliscoff,
2006; Mancini, 2002). Possible explanations for its development involve
(a) downslope migration of high-Andean vegetation driven by snow line and
tree line lowering associated with intense glaciation in the region, coupled
with (b) the occurrence of scattered, low-density populations of hygrophilous
trees and herbs along the eastern margin of the PIS during the LGM and T1. We
rule out the alternative explanation that pollen grains and spores of
hygrophilous trees and herbs in Lago Edita represent an advected signal
through the Andes from ice-free humid Pacific sectors harboring these species
because (i) no empirical basis is currently available for ice-free
conditions and the occurrence of cold-resistant hygrophilous taxa along the
western Andean slopes or the Pacific coast of central Patagonia during the
LGM. In fact, the oldest minimum limiting dates for ice-free conditions in
records from the Taitao Peninsula and the Chonos Archipelago yielded ages of
14 335 ± 140 and 13 560 ± 125 14C yr BP (median age
probability, MAP: 17 458 and 16 345 yr BP), respectively (Haberle and
Bennett, 2004; Lumley and Switsur, 1993). (ii) The appearance of
Fitzroya/Pilgerodendron and Podocarpus nubigena at
∼ 15 000 and ∼ 14 000 yr BP, respectively, occurred
4000–5000 years later in coastal Pacific sites relative to the Lago Edita
record (Fig. 7), and (iii) background levels of Nothofagus between 15
and 20 % in Lago Edita predate the appearance and expansion of this taxon
in coastal Pacific sites and, once realized, its abundance in Lago Edita
cannot be attributed to long-distance transport from the western Pacific
coast (Fig. 7).
Comparison of selected tree pollen recorded in Lago Fácil,
Lago Oprasa, Lago Stibnite (Lumley and Switsur, 1993) and Lago Edita. The
red line corresponds to a weighted running mean in each record of seven
adjacent samples with a triangular filter. The lower panels show the curves
from all sites expressed in a common percent scale (Lago Fácil is the purple
line, Lago Oprasa is the blue line, Lago Stibnite is the black line and Lago Edita is the red
line).
Previous palynological studies from bogs located east of the central
Patagonian Andes (de Porras et al., 2012; Markgraf et al., 2007) (Mallín
Lago Shaman and Mallín Pollux, Fig. 1) interpreted dry conditions prior
to ∼ 12 000 yr BP, based on the premise that low abundance of
arboreal taxa and predominance of herbs and shrubs were indicative of
Patagonian Steppe communities. The glacial-to-interglacial vegetation change
in those studies was interpreted as a westward shift of the forest-steppe
boundary brought by lower-than-present SWW influence at 44–46∘ S,
followed by a rise in temperature and precipitation at the end of the last
glaciation. In contrast, the Lago Augusta site (located in Valle Chacabuco
∼ 7 km northeast of Lago Edita) (Fig. 1) shows a pollen assemblage
prior to 15 600 yr BP dominated by high-Andean herbs and shrubs, along
with taxa characteristic of hyper-humid environments along the Pacific coasts
of central Patagonia (Nothofagus, Fitzroya/Pilgerodendron, Podocarpus nubigena, Saxegothaea conspicua, Drimys winteri, Dysopsis glechomoides and
the ferns Blechnum, Hymenophyllaceae and Cystopteris)
(Villa-Martinez et al., 2012). It appears then that floristic elements of
modern Patagonian forests were present in low abundance and in a
discontinuous manner along the eastern flank of the PIS between
44 and 47∘ S. The data from Lago Edita shown in this paper, along with
the results from Lago Augusta, suggest that Valle Chacabuco harbored cryptic
refugia (Bennett and Provan, 2008) of rainforest trees and herbs during the
interval 19 000–11 000 yr BP. Therefore, the interpretation of
lower-than-present precipitation of SWW origin in previous studies (de Porras
et al., 2012; Markgraf et al., 2007) is not applicable to the Valle
Chacabuco area over this time interval. Plant colonization of Valle Chacabuco
must have started from the LGM limits located east of Lago Cochrane and
followed the shrinking ice masses to the west once the newly deglaciated
sectors were devoid of glaciolacustrine influence through T1.
Declines in and the virtual disappearance of the cold-resistant hygrophilous trees
Fitzroya/Pilgerodendron and Podocarpus nubigena along with the
herbs Gunnera magellanica and Lycopodium magellanicum took
place at ∼ 11 000 yr BP in the Lago Edita record (Figs. 4, 6) in
response to a sudden decline in precipitation relative to the
∼ 14 500–11 000 yr BP interval. These changes were contemporaneous
with a sustained rise in Nothofagus, decreases in all other shrubs
and herbs and a major increase in macroscopic charcoal (Fig. 5), signaling
an increment in arboreal cover, higher spatial continuity of coarse fuels and
forest fires. We interpret this arboreal increase and fire-regime shift as
driven by warming, which might have triggered a tree line rise and favored the
spread and/or densification of woody species and coarse fuels (Figs. 4, 5, 6).
Possible ignition agents for the beginning of fire activity at
14 300 yr BP in the Lago Edita record include the incendiary effects of
explosive volcanic activity, lightning strikes and human activity. We rule
out volcanic disturbance as a driving factor, considering the lack of
contemporary tephras in the stratigraphy of the Lago Edita sediment cores,
and we cannot support nor reject other ignition agents considering the current
lack of stratigraphic proxies to constrain their likely influence in the
Valle Chacabuco area. Finally, Nothofagus forests (∼ 70 %
abundance) were established near Lago Edita between 10 000 and 9000 yr BP.
Glacial recession in Valle Chacabuco and the Lago Cochrane basin
Stratigraphic and chronological results from Lago Edita are key for deciphering
the evolution of Valle Chacabuco and for constraining the timing and rates of
deglaciation in this core region of the PIS. Previous studies (Hein et al.,
2010) indicate that Valle Chacabuco was overridden by the LCIL during the LGM and deposited the Río Blanco moraines
∼ 90 km downstream from Lago Edita, distal to the eastern end of Lago
Cochrane in Argentina (Argentinian name: Pueyrredón). Cosmogenic
radionuclide dating of three main moraine limits by Hein et al. (2010)
yielded cosmogenic 10Be exposure ages, recently recalculated by Kaplan
et al. (2011) at ∼ 21 100, ∼ 25 100 and
∼ 28 700 yr BP. This was followed by glacial recession starting at
19 600 ± 800 yr BP, formation of GLC and
stabilization and deposition of the Lago Columna and Lago Posada moraines
before 17 600 ± 900 yr BP ∼ 55 km upstream from the Río
Blanco moraines (Hein et al., 2010; Kaplan et al., 2011) (Fig. 1). Further
glacial recession led to the westward expansion and lowering of GLC until the
LCIL stabilized and deposited moraines in Lago Esmeralda between 13 600 and
12 800 yr BP ∼ 60 km upstream from the Lago Columna and Lago Posada
moraines (Turner et al., 2005). Recession from this position led to sudden
drainage of GLC toward the Pacific Ocean via Río Baker, once the
continuity between the northern and southern Patagonian ice fields was breached by
glacial recession and thinning. These data suggest that Valle Chacabuco may
have been ice-free and devoid of glaciolacustrine influence after
∼ 17 600 yr BP. More recently, Boex et al. (2013) reported a
cosmogenic radionuclide-based reconstruction of vertical profile changes of
the LCIL through the LGM and T1 that reveals deposition of (i) the Sierra
Colorado lower limit by 28 980 ± 1206 yr BP, which can be traced to
the Río Blanco moraines; (ii) the highest summits of the Cerro Oportus and
Lago Columna moraines by 18 966 ± 1917 yr BP; and (iii) the
María Elena moraine by 17 088 ± 1542 yr BP. According to these
data, Valle Chacabuco may have been ice-free after ∼ 17 000 yr BP.
Lago Edita is a closed-basin lake located ∼ 11 km east of the Cerro
Tamango summit along the ridge that defines the southern edge of the Valle
Chacabuco watershed (Fig. 1). Lacustrine sedimentation in Lago Edita started
when ice-free conditions developed in Valle Chacabuco, as the LCIL snout
retreated eastward to a yet unknown position. The Lago Edita cores show 9 m
of blue-gray clays with millimeter-scale laminations, interrupted by
sporadic intervals of massive pebbly mud appreciable in X radiographs and
the LOI550 record as increases in the inorganic density data (Fig. 2).
We also found exposed glaciolacustrine beds and discontinuous fragments of
lake terraces in the vicinity of Lago Edita, attesting for a large lake that
flooded Valle Chacabuco in its entirety. Differential GPS measurements of
570 m a.s.l. for the Lago Edita surface and 591 m a.s.l. for a
well-preserved terrace fragment located ∼ 150 m directly south of Lago
Edita provide minimum-elevation constraints for GLC during this stage. The
Lago Augusta site (Villa-Martinez et al., 2012), located ∼ 7 km
northeast of Lago Edita on the Valle Chacabuco floor at 444 m a.s.l.
(Fig. 1), shows 8 m of basal glaciolacustrine mud (Fig. 2), lending support
to our interpretation.
Glaciolacustrine sedimentation persisted in Lago Edita and Lago Augusta until
the surface elevation of GLC dropped below 570 and 444 m a.s.l.,
respectively, and the closed-basin lakes developed. The chronology for this
event is constrained by statistically identical AMS dates of
16 250 ± 90 and 16 020 ± 50 14C yr BP (UCIAMS-133418 and
CAMS-144454, respectively) (Table 1) from the same level in the basal portion
of the organic sediments in the Lago Edita record; this estimate approaches
the timing for the cessation of glaciolacustrine influence in Lago Augusta,
radiocarbon-dated at 16 445 ± 45 14C yr BP (CAMS-144600)
(Table 1). Because we observe approximately the same age for the transition
from glaciolacustrine to organic-rich mud in both stratigraphic layers, we
interpret the weighted mean age of those three dates
(16 254 ± 63 14C yr BP, MAP: 19 426 yr BP, two different
laboratories) as a minimum-limiting age for ice-free conditions and nearly
synchronous glaciolacustrine regression from elevations 591 and
444 m a.s.l. in Valle Chacabuco. We acknowledge that Villa-Martínez et
al. (2012) excluded the age of date CAMS-144600 from the age model of the
Lago Augusta palynological record because it was anomalously old in the
context of other radiocarbon dates higher up in the core.
Comparison of the radiocarbon-dated stratigraphy from the Lago Edita record with
the exposure-age-dated glacial geomorphology from Lago
Cochrane (Pueyrredón), Valle Chacabuco and the surrounding mountains reveals
the following:
The geochronology for the innermost (third) belt of Río Blanco
moraines (∼ 21 100 yr BP) (Hein et al., 2010; Kaplan et al., 2011),
glacial deposits on the highest summits of Cerro Oportus and the Lago Columna
moraines (18 966 ± 1917 yr BP) (Boex et al., 2013) is compatible
(within error) with the onset of organic sedimentation in Lago Edita and Lago
Augusta at 19 426 yr BP in Valle Chacabuco. If correct, this indicates a
∼ 90 km recession of the LCIL from its LGM limit within
∼ 1500 years.
The dates of Hein et al. (2010) for the final LGM limit and Lago
Columna and Lago Posada moraines, as well as the chronology of Boex et al. (2013) for the María Elena
moraine, should be considered as minimum-limiting
ages. This is because cosmogenic radio nuclide ages for these landforms
postdate the onset of organic sedimentation in Lago Edita and Lago Augusta,
despite being morphostratigraphically distal (older) than Valle Chacabuco.
As shown in Fig. 1, Lago Edita is located along a saddle that establishes
the southern limit of the Río Chacabuco catchment and the northern limit
of the Lago Cochrane basin. According to Hein et al. (2010) the drainage
divide on the eastern end of the Lago Cochrane (Pueyrredón) basin is located at
475 m a.s.l.; therefore, the presence of this perched glacial lake with a
surface elevation of 591 m a.s.l. requires the presence of ice dams located in the Valle
Chacabuco and the Lago Cochrane basin. This suggests that both valleys
remained partially ice covered and that enough glacier thinning and recession
early during T1 enabled the development of a topographically constrained
glacial lake that covered Valle Chacabuco up to the aforementioned saddle.
The high stand of GLC at 591 m a.s.l. lasted for less than 1500 years during
the LGM and was followed by a nearly instantaneous lake-level lowering of at
least ∼ 150 m at ∼ 19 400 yr BP in Valle Chacabuco. The
abrupt large-magnitude drainage event of this predecessor lake was
recently recognized by Bourgois et al. (2016), but its chronology and
hydrographic and climatic implications have not been addressed in the
Quaternary literature.
Biogeographic and paleoclimatic implications
The persistence of scattered, low-density populations of rainforest trees and
herbs east of the Andes during the LGM and T1 (Figs. 4, 6) implies that
precipitation delivered by the SWW must have been substantially higher than
at present (680 mm year-1 measured at the Cochrane meteorological
station). Because local precipitation in western Patagonia is positively and
significantly correlated with low-level zonal winds (Garreaud et al., 2013;
Moreno et al., 2010, 2014), we propose that the SWW influence
at 47∘ S was stronger than present between 19 000 and
11 000 yr BP, in particular between 16 800 and 11 000 yr BP.
Subsequent increases in arboreal vegetation, chiefly Nothofagus, at
∼ 13 000 and ∼ 11 000 yr BP led to the establishment of
forests near Lago Edita between 10 000 and 9000 yr BP (Figs. 4, 6). We
interpret these increases as episodes of tree line rise driven by warming pulses
coupled with a decline in SWW strength at 47∘ S (relative to the
∼ 14 500–11 000 yr BP interval), as suggested by the disappearance
of cold-resistant hygrophilous trees and herbs at ∼ 11 000 yr BP. We
speculate that the warm pulse and decline in SWW influence at
∼ 11 000 yr BP might account for the abandonment of early Holocene
glacier margins in multiple valleys in central Patagonia (Glasser et al.,
2012).
Comparison of the percent sum of arboreal pollen (AP) in records
from Lago Edita, Lago Stibnite (Lumley and Switsur, 1993) and the spliced
Canal de la Puntilla-Huelmo time series (Moreno et al., 2015), as proxies for
local rise in tree line driven by deglacial warming. These data are compared
to the delta Deuterium
record from the EPICA Dome Concordia (EDC) ice core (Stenni et al.,
2010), and hydrologic estimates from northwestern Patagonia. The latter
consist of the percent abundance of Magellanic moorland species found in the
spliced Canal de la Puntilla-Huelmo record (Moreno et al., 2015), indicative
of a hyper-humid regime, and the percent abundance of the littoral macrophyte
Isoetes savatieri from Lago Lepué (Pesce and Moreno, 2014),
indicative of low lake level (LL) during the earliest stages of T1 and the
early Holocene (9000–11 000 yr BP). The vertical dashed lines constrain
the timing of the early Holocene SWW minimum at 41–43∘ S
(9000–11 000 yr BP) (Fletcher and Moreno, 2011), a low-precipitation
phase during the early termination at 41–43∘ S
(16 800–17 800 yr BP) associated with a southward shift of the SWW
(Pesce and Moreno, 2014), the final LGM advance of piedmont glacier lobes
(17 800–19 300 yr BP) and the final portion of the Varas interestade
(19 300–21 000 yr BP) in the Chilean Lake District (Denton et al., 1999;
Moreno et al., 2015). The dashed green horizontal lines indicate the mean AP
of each pollen record prior to their increases during T1 (Lago Edita:
17 %, Lago Stibnite: 2 %, spliced Canal de la Puntilla-Huelmo:
31 %). The ascending oblique arrow represents a northward shift of the
SWW the descending arrow represents a southward shift of the SWW at the beginning of
T1.
Five salient aspects of the Lago Edita record are relevant for deciphering
the pattern and rates of climate change and dispersal routes of the
vegetation in Central Patagonia (47∘ S) during T1:
There is an absence of stratigraphically discernable indications of deglacial warming
between 19 400 and 13 000 yr BP, in contrast to northwestern Patagonian
records (the Canal de la Puntilla–Huelmo record, Fig. 1) (Moreno et al.,
2015), which show that 75–80 % of the glacial–interglacial temperature
recovery was accomplished between 17 800 and 16 800 yr BP (Fig. 8). The
record from Lago Stibnite (46∘26′ S, 74∘25′ W), located
in central-western Patagonia upwind from the PIS and Lago Edita (Fig. 1), shows
a rapid increase in arboreal pollen from ∼ 2 % to > 80 % in
less than 1000 years starting at 16 200 yr BP (Fig. 8). We posit that cold
glacial conditions lingered along the periphery of the shrinking PIS during
T1, affecting adjacent downwind sectors such as Valle Chacabuco. According to
Turner et al. (2005) the LCIL stabilized and deposited moraines in Lago
Esmeralda, located ∼ 10 km upstream along the glacier flow line and
∼ 240 m lower in elevation than Lago Edita, between 13 600 and
12 800 yr BP. We propose that the climatic barrier for arboreal expansion
vanished in downwind sectors such as Valle Chacabuco once glacial recession
from the Lago Esmeralda (Fig. 1) margin breached the continuity of the
northern
and southern Patagonian ice fields along the Andes. Thus, we propose that
regional cooling induced by the PIS along its eastern margin through T1
accounts for the delayed warming in Valle Chacabuco relative to records
located in the western and northwestern sectors (Fig. 8).
Cold and
wet conditions prevailed between 19 400 and 16 800 yr BP, followed by an
increase in precipitation at 16 800 yr BP. The latter event is
contemporaneous with the onset of a lake-level rise in Lago Lepué
(43∘ S, central-east Isla Grande de Chiloé) (Fig. 8), which
Pesce and Moreno (2014) interpreted as a northward shift of the SWW as it
recovered from a prominent southward shift from latitude
∼ 41 to 43∘ S (Fig. 8) following the onset of T1 (Moreno et al.,
2015).
Significant ice recession (∼ 90 km) from the eastern LGM
margin of the LCIL was accomplished between
∼ 21 000 and 19 400 yr BP, at times when northwestern Patagonian
piedmont glacier lobes experienced moderate recession during the Varas
interstade (Denton et al., 1999; Moreno et al., 2015) (Fig. 8). In contrast
to the LCIL, northwestern Patagonian piedmont glacier lobes re-advanced to
their youngest glacial maximum position during a cold episode between 19 300
and 17 800 yr BP that featured stronger SWW influence at
41–43∘ S (Moreno et al., 2015) (Fig. 8). One explanation for this
latitudinal difference might be that northward-shifted SWW between 19 300
and 17 800 yr BP fueled glacier growth in northwestern Patagonia while
reducing the delivery of moisture to central Patagonia, causing the LCIL to
continue the recession it had started during the Varas interstade.
A
mosaic of cold-resistant and hygrophilous trees and herbs currently found
along the humid western slopes of the Andes of central Chilean Patagonia and
cold-resistant shrubs and herbs common to high-Andean and Patagonian steppe
communities developed along the eastern margin of the PIS during the LGM and
T1 (Figs. 4, 6). We posit that glacial withdrawal and drainage of GLC through
T1 provided a route for the westward dispersal of hygrophilous trees and
herbs, contributing to the forestation of the newly deglaciated sectors of
central-western Patagonia.
The cold-resistant hygrophilous conifer
Podocarpus nubigena increased between 14 500 and 13 000 yr BP,
suggesting an increase in precipitation brought by the SWW to the eastern
Andean slopes of central Patagonia. This was followed by a decline, which was
contemporaneous with a rise in the regional Nothofagus–dominated
tree line between 13 000 and 11 200 yr BP. These interpretations
imply stronger SWW influence of the SWW at 47∘ S during the
Antarctic Cold Reversal and warming during Younger Dryas time.
We conclude that warm pulses at ∼ 13 000 and ∼ 11 000 yr BP
and a decline in SWW influence at 47∘ S starting at
∼ 11 000 yr BP brought T1 to an end in central-western Patagonia. The
earliest of these events overlaps in timing with the culmination of
Patagonian (Garcia et al., 2012; Moreno et al., 2009; Strelin et al., 2011;
Strelin and Malagnino, 2000) and New Zealand glacier advances (Kaplan et al.,
2010; Putnam et al., 2010) during the Antarctic Cold Reversal. Our data
suggest that the subsequent warm pulse, which was accompanied by a decline in
SWW strength at ∼ 11 000 yr BP (Moreno et al., 2010, 2012), was the
decisive event that led to the end of T1 in the study area.
Aspects of the record are available upon request to the corresponding author.
The authors declare that they have no conflict of interest.
This article is part of the special issue “Southern perspectives
on climate and the environment from the Last Glacial Maximum through the Holocene:
the Southern Hemisphere Assessment of PalaeoEnvironments (SHAPE) project”. It does not belong to a conference.
Acknowledgements
This study was funded by Fondecyt nos. 1080485, 1121141 and 1151469; ICM grants
P05-002 and NC120066; and a CONICYT MSc Scholarship to William I. Henríquez. We thank
Esteban A. Sagredo, Oscar H. Pesce, Enzo Simi, and Ignacio Jara for assistance during field
work and
Keith D. Bennett and Simon Haberle
for sharing published palynological data. We
thank Cristian Saucedo from Agencia de Conservación Patagónica for
permission to work and collect samples in Hacienda Valle Chacabuco (Parque
Patagonia). We thank the editor and the three anonymous reviewers for their
constructive comments on early versions of this paper. Edited by: Helen Bostock
Reviewed by: three anonymous referees
References
Bennett, K. D. and Provan, J.: What do we mean by “refugia”?, Quaternary
Sci. Rev., 27, 2449–2455, 2008.
Blaauw, M. and Christen, J. A.: Flexible Paleoclimate Age-Depth Models Using
an Autoregressive Gamma Process, Bayesian Anal., 6, 457–474, 2011.Boex, J., Fogwill, C., Harrison, S., Glasser, N. F., Hein, A., Schnabel, C.,
and Xu, S.: Rapid thinning of the late Pleistocene Patagonian Ice Sheet
followed migration of the Southern Westerlies, Sci. Rep.-UK, 3, 2118,
10.1038/srep02118, 2013.Bourgois, J., Cisternas, M. E., Braucher, R., Bourlès, D., and Frutos,
J.: Geomorphic Records along the General Carrera (Chile)–Buenos Aires
(Argentina) Glacial Lake (46–48∘ S), Climate Inferences, and
Glacial Rebound for the Past 7–9 ka, J. Geol., 124, 27–53, 2016.
Bromwich, D. H., Toracinta, E. R., Wei, H., Oglesby, R. J., Fastook, J. L.,
and Hughes, T. J.: Polar MM5 Simulations of the Winter Climate of the
Laurentide Ice Sheet at the LGM, J. Climate, 17, 3415–3433, 2004.
Bromwich, D. H., Toracinta, E. R., Oglesby, R. J., Fastook, J. L., and
Hughes, T. J.: LGM Summer Climate on the Southern Margin of the Laurentide
Ice Sheet: Wet or Dry?, J. Climate, 18, 3317–3338, 2005.Caniupán, M., Lamy, F., Lange, C. B., Kaiser, J., Arz, H., Kilian, R.,
Baeza Urrea, O., Aracena, C., Hebbeln, D., Kissel, C., Laj, C., Mollenhauer,
G., and Tiedemann, R.: Millennial-scale sea surface temperature and
Patagonian Ice Sheet changes off southernmost Chile (53∘ S) over the
past ∼ 60 kyr, Paleoceanography, 26, PA3221, 10.1029/2010PA002049,
2011.de Porras, M. E., Maldonado, A., Abarzúa, A. M., Cárdenas, M. L.,
Francois, J. P., Martel-Cea, A., Stern, C. R., Méndez, C., and Reyes, O.:
Postglacial vegetation, fire and climate dynamics at Central Chilean
Patagonia (Lake Shaman, 44∘ S), Quaternary Sci. Rev., 50, 71–85,
2012.
Denton, G. H., Lowell, T. V., Heusser, C. J., Schluchter, C., Andersen, B.
G., Heusser, L. E., Moreno, P. I., and Marchant, D. R.: Geomorphology,
stratigraphy, and radiocarbon chronology of Llanquihue drift in the area of
the southern Lake District, Seno Reloncavi, and Isla Grande de Chiloe, Chile,
Geogr. Ann. A, 81, 167–229, 1999.
Faegri, K. and Iversen, J.: Textbook of pollen analysis, John Wiley &
Sons, 1989.Fletcher, M. S. and Moreno, P. I.: Zonally symmetric changes in the strength
and position of the Southern Westerlies drove atmospheric CO2 variations
over the past 14 ky, Geology, 39, 419–422, 2011.
Gajardo, R.: La Vegetación Natural de Chile, Clasificación y
Distribución Geográfica, Editorial Universitaria, Santiago, Chile,
1994.
Garcia, J. L., Kaplan, M. R., Hall, B. L., Schaefer, J. M., Vega, R. M.,
Schwartz, R., and Finkel, R.: Glacier expansion in southern Patagonia
throughout the Antarctic cold reversal, Geology, 40, 859–862, 2012.
Garreaud, R., Lopez, P., Minvielle, M., and Rojas, M.: Large-Scale Control on
the Patagonian Climate, J. Climate, 26, 215–230, 2013.
Glasser, N. F., Harrison, S., Schnabel, C., Fabel, D., and Jansson, K. N.:
Younger Dryas and early Holocene age glacier advances in Patagonia,
Quaternary Sci. Rev., 58, 7–17, 2012.Glasser, N. F., Jansson, K. N., Duller, G. A. T., Singarayer, J., Holloway,
M., and Harrison, S.: Glacial lake drainage in Patagonia (13–8 kyr) and
response of the adjacent Pacific Ocean, Sci. Rep.-UK, 6, 21064,
10.1038/srep21064, 2016.
Haberle, S. G. and Bennett, K. D.: Postglacial formation and dynamics of
North Patagonian Rainforest in the Chonos Archipelago, Southern Chile,
Quaternary Sci. Rev., 23, 2433–2452, 2004.
Hall, B. L., Porter, C. T., Denton, G. H., Lowell, T. V., and Bromley, G. R.
M.: Extensive recession of Cordillera Darwin glaciers in southernmost South
America during Heinrich Stadial 1, Quaternary Sci. Rev., 62, 49–55, 2013.
Hein, A. S., Hulton, N. R. J., Dunai, T. J., Sugden, D. E., Kaplan, M. R.,and
Xu, S.: The chronology of the Last Glacial Maximum and deglacial events in
central Argentine Patagonia, Quaternary Sci. Rev., 29, 1212–1227, 2010.
Heiri, O., Lotter, A. F., and Lemcke, G.: Loss on ignition as a method for
estimating organic and carbonate content in sediments: reproducibility and
comparability of results, J. Paleolimnol., 25, 101–110, 2001.
Heusser, C. J.: Pollen and Spores from Chile, University of Arizona Press,
Tucson, 1971.
Heusser, C. J.: Late Quaternary Vegetation and Climate of Southern
Tierra-Del-Fuego, Quaternary Res., 31, 396–406, 1989.
Higuera, P. E., Brubaker, L. B., Anderson, P. M., Hu, F. S., and Brown, T.
A.: Vegetation mediated the impacts of postglacial climate change on fire
regimes in the south-central Brooks Range, Alaska, Ecol. Monogr., 79,
201–219, 2009.
Kaplan, M. R., Schaefer, J. M., Denton, G. H., Barrell, D. J. A., Chinn, T.
J. H., Putnam, A. E., Andersen, B. G., Finkel, R. C., Schwartz, R., and
Doughty, A. M.: Glacier retreat in New Zealand during the Younger Dryas
stadial, Nature, 467, 194–197, 2010.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., and Travis,
S. G.: In-situ cosmogenic 10Be production rate at Lago Argentino,
Patagonia: Implications for late-glacial climate chronology, Earth Planet Sc.
Lett., 309, 21–32, 2011.Kaufman, D. S., Ager, T. A., Anderson, N. J., Anderson, P. M., Andrews, J.
T., Bartlein, P. J., Brubaker, L. B., Coats, L. L., Cwynar, L. C., Duvall, M.
L., Dyke, A. S., Edwards, M. E., Eisner, W. R., Gajewski, K., Geirsdottir,
A., Hu, F. S., Jennings, A. E., Kaplan, M. R., Kerwin, M. N., Lozhkin, A. V.,
MacDonald, G. M., Miller, G. H., Mock, C. J., Oswald, W. W., Otto-Bliesner,
B. L., Porinchu, D. F., Ruhland, K., Smol, J. P., Steig, E. J., and Wolfe, B.
B.: Holocene thermal maximum in the western Arctic (0–180∘ W),
Quaternary Sci. Rev., 23, 529–560, 2004.Lara, A., Villalba, R., Wolodarsky-Franke, A., Aravena, J. C., Luckman, B.
H., and Cuq, E.: Spatial and temporal variation in Nothofagus pumilio growth
at tree line along its latitudinal range (35 degrees 40′–55 degrees S) in
the Chilean Andes, J. Biogeogr., 32, 879–893, 2005.
Luebert, F. and Pliscoff, P.: Sinopsis Bioclimática y Vegetacional de
Chile, Editorial Universitaria, Santiago, Chile, 2006.
Lumley, S. and Switsur, R.: Late Quaternary chronology of the Taitao
Peninsula, Southern Chile, J. Quaternary Sci., 8, 161–165, 1993.
Mancini, M. V.: Vegetation and climate during the holocene in Southwest
Patagonia, Argentina, Rev. Palaeobot. Palyno., 122, 101–115, 2002.Markgraf, V., Whitlock, C., and Haberle, S.: Vegetation and fire history
during the last 18,000 cal yr BP in Southern Patagonia: Mallin Pollux,
Coyhaique, Province Aisen (45 degrees 41′30′′ S,
71 degrees 50′30′′ W, 640 m elevation), Palaeogeogr. Palaeocl., 254,
492–507, 2007.
Moreno, P. I., Kaplan, M. R., Francois, J. P., Villa-Martinez, R., Moy, C.
M., Stern, C. R., and Kubik, P. W.: Renewed glacial activity during the
Antarctic cold reversal and persistence of cold conditions until 11.5 ka in
southwestern Patagonia, Geology, 37, 375–378, 2009.Moreno, P. I., Francois, J. P., Villa-Martínez, R., and Moy, C. M.:
Covariability of the Southern Westerlies and atmospheric CO2 during the
Holocene, Geology, 39, 727–730, 2010.
Moreno, P. I., Villa-Martinez, R., Cardenas, M. L., and Sagredo, E. A.:
Deglacial changes of the southern margin of the southern westerly winds
revealed by terrestrial records from SW Patagonia (52 degrees S), Quaternary
Sci. Rev., 41, 1–21, 2012.Moreno, P. I., Vilanova, I., Villa-Martínez, R., Garreaud, R. D., Rojas,
M., and De Pol-Holz, R.: Southern Annular Mode-like changes in southwestern
Patagonia at centennial timescales over the last three millennia, Nat.
Commun., 5, 4375, 10.1038/ncomms5375, 2014.
Moreno, P. I., Denton, G. H., Moreno, H., Lowell, T. V., Putnam, A. E., and
Kaplan, M. R.: Radiocarbon chronology of the last glacial maximum and its
termination in northwestern Patagonia, Quaternary Sci. Rev., 122, 233–249,
2015.Pesce, O. H. and Moreno, P. I.: Vegetation, fire and climate change in
central-east Isla Grande de Chiloé (43∘ S) since the Last
Glacial Maximum, northwestern Patagonia, Quaternary Sci. Rev., 90, 143–157,
2014.
Pisano, E.: Los bosques de Patagonia Austral y Tierra del Fuego chilenas,
Anales del Instituto de la Patagonia – Serie Ciencias Naturales (Chile), 25,
9–19, 1997.
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., and
Schluchter, C.: Glacier advance in southern middle-latitudes during the
Antarctic Cold Reversal, Nat. Geosci., 3, 700–704, 2010.
Schmithüsen, J.: Die raumliche Ordnung der chilenischen Vegetation,
Bonner Geographische Abhandlungen, 17, 1–86, 1956.
Siani, G., Michel, E., De Pol-Holz, R., DeVries, T., Lamy, F., Carel, M.,
Isguder, G., Dewilde, F., and Lourantou, A.: Carbon isotope records reveal
precise timing of enhanced Southern Ocean upwelling during the last
deglaciation, Nat. Commun., 4, 2758, 2013.
Stenni, B., Masson-Delmotte, V., Selmo, E., Oerter, H., Meyer, H.,
Rothlisberger, R., Jouzel, J., Cattani, O., Falourd, S., Fischer, H.,
Hoffmann, G., Iacumin, P., Johnsen, S. J., Minster, B., and Udisti, R.: The
deuterium excess records of EPICA Dome C and Dronning Maud Land ice cores
(East Antarctica), Quaternary Sci. Rev., 29, 146–159, 2010.
Stern, C.: Active Andean volcanism: its geologic and tectonic setting, Rev.
Geol. Chile, 31, 161–206, 2004.Stern, C. R., Moreno, P. I., Henriquez, W. I., Villa-Martinez, R., Sagredo,
E., Aravena, J. C., and De Pol-Holz, R.: Holocene tephrochronology around
Cochrane (∼ 47∘ S), southern Chile, Andean Geol., 43, 1–19,
2016.
Strelin, J. A. and Malagnino, E. C.:Late-Glacial History of Lago Argentino,
Argentina, and Age of the Puerto Bandera Moraines, Quaternary Res., 54,
339–347, 2000.Strelin, J. A., Denton, G. H., Vandergoes, M. J., Ninnemann, U. S., and
Putnam, A. E.: Radiocarbon chronology of the late-glacial Puerto Bandera
moraines, Southern Patagonian Icefield, Argentina, Quaternary Sci. Rev., 30,
2551–2569, 2011.
Sugita, S.: Pollen Representation of Vegetation in Quaternary Sediments:
Theory and Method in Patchy Vegetation, J. Ecology, 82, 881–897, 1994.
Turner, K. J., Fogwill, C. J., McCulloch, R. D., and Sugden, D. E.:
Deglaciation of the eastern flank of the North Patagonian Icefield and
associated continental-scale lake diversions, Geogr. Ann. A, 87, 363–374,
2005.
Villa-Martinez, R., Moreno, P. I., and Valenzuela, M. A.: Deglacial and
postglacial vegetation changes on the eastern slopes of the central
Patagonian Andes (47 degrees S), Quaternary Sci. Rev., 32, 86–99, 2012.