Clim.

Past, 10, 1967–1981, 2014

www.clim-past.net/10/1967/2014/
doi:10.5194/cp-10-1967-2014
© Author(s) 2014. CC Attribution 3.0 License.

Hydroclimate variability of the northwestern Amazon Basin near

the Andean foothills of Peru related to the South American
Monsoon System during the last 1600 years
J. Apaéstegui1,3,4 , F. W. Cruz2 , A. Sifeddine3,4,12 , M. Vuille5 , J. C. Espinoza1,6 , J. L. Guyot8 , M. Khodri3,7 , N. Strikis2 ,
R. V. Santos9 , H. Cheng10,11 , L. Edwards11 , E. Carvalho9 , and W. Santini8
1 Instituto Geofísico del Perú, Lima, Peru
2 Instituto de Geociências, Universidade de São Paulo, São Paulo, Brazil
3 LMI “PALEOTRACES” (URD/UFF/Uantof-Chili), Departamento de Geoquimica-UFF, Niterói-RJ, Brazil
4 Departamento de Geoquimica, Universidade Federal Fluminense, Niterói-RJ, Brazil
5 University of Albany, SUNY, Albany, NY, USA
6 Universidad Agraria La Molina, Lima, Peru
7 UMR LOCEAN (IRD/UPMC/CNRS/MNHN), Paris-Jussieu, France
8 UMR GET (IRD) Géosciences Environnement Toulouse, CNRS-IRD-UPS, OMP, Toulouse, France
9 Instituto de Geociências, Universidade de Brasilia, Brasilia, DF, Brazil
10 Institute of Global Environmental Change, Xi’an Jiaotong University, Xi’an, China
11 Department of Geology and Geophysics, University of Minnesota, Twin Cities, Minneapolis, Minnesota, USA
12 LOCEAN (CNRS, IRD, MNHN, UPMC), Bondy, France

Correspondence to: J. Apaéstegui ([email protected])

Received: 3 January 2014 – Published in Clim. Past Discuss.: 10 February 2014

Revised: 9 October 2014 – Accepted: 13 October 2014 – Published: 19 November 2014

Abstract. In this paper we explore a speleothem δ 18 O record to evaluate the influence of the AMO on the Palestina cave
from Palestina cave, northwestern Peru, at a site on the east- δ 18 O and other δ 18 O-derived SASM reconstructions, allow
ern side of the Andes cordillera, in the upper Amazon Basin. insight into the spatial footprints of the AMO over tropi-
The δ 18 O record is interpreted as a proxy for South Ameri- cal South America and highlight differences between records
can Summer Monsoon (SASM) intensity and allows the re- during key studied periods. This work also reveals that repli-
construction of its variability during the last 1600 years. Two cating regional climate signals from different sites, and using
periods of anomalous changes in the climate mean state cor- different proxies is absolutely essential for a comprehensive
responding to the Medieval Climate Anomaly (MCA) and understanding of past changes in SASM activity.
the Little Ice Age (LIA) periods identified in the Northern
Hemisphere are recognized in the record, in which decreased
and increased SASM activity, respectively, have been doc-
umented. Variations in SASM activity between the MCA 1 Introduction
and the LIA seem to be larger over the northern part of the
continent, suggesting a latitudinal dependence of the MCA South American paleoclimate reconstructions over the last
footprint. Our results, based on time series, composite and millennium based on δ 18 O from different proxies such as
wavelet analyses, suggest that the Atlantic Multidecadal Os- speleothems, lake sediments and ice cores have shown re-
cillation (AMO) plays an relevant role for SASM modula- gionally coherent patterns of change during distinct events
tion on multidecadal scales (∼ 65 years), especially during recognized as the Medieval Climate Anomaly (MCA;
dry periods such as the MCA. Composite analyses, applied AD 900–1250) and the Little Ice Age (LIA; AD 1400–1850)
(e.g., Thompson et al., 1986, 2013; Rabatel et al., 2008;

Published by Copernicus Publications on behalf of the European Geosciences Union.

1968 J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years

Reuter et al., 2009; Jomelli et al., 2009; Bird et al., 2011a; southern Brazil speleothems do not clearly document this dry
Vuille et al., 2012; Novello et al., 2012; Kanner et al., 2013). period.
These climate anomalies are primarily manifested as changes Aside from the lack of records to adequately map the re-
in temperature over Northern Hemisphere oceanic and conti- gional climate response to the perturbed climate states during
nental areas but appear more strongly related to variations in the LIA and MCA when compared to the Current Warm Pe-
hydroclimatic conditions over tropical South America (e.g., riod (CWP; Bird et al., 2011a), the dynamical mechanisms
Bird et al., 2011a; Vuille et al., 2012). Both the LIA and involved in changing the SASM mean state and its spatial
the MCA have been identified in documentary and proxy footprint are still not fully understood. Recent studies suggest
records from the Northern Hemisphere where warm/dry con- that teleconnections between the Pacific and Atlantic oceans
ditions were inferred from several proxies during the MCA affect SASM intensity and/or rainfall distribution over South
(e.g., Graham et al., 2010 and references cited therein) while America at different timescales (e.g., Kanner et al., 2013;
cold/wet conditions have been inferred from glaciers advanc- Novello et al., 2012). How these modes of variability and
ing through the LIA period (e.g., Mann et al., 2002 and ref- their interactions change in response to changes in radiative
erences cited therein). forcing is of major importance for understanding past and
In the South American Andes, one of the first climate re- future SASM variations. In fact, there is considerable con-
constructions for the last 1500 years was produced in the cern that the SASM dynamics will be significantly affected
1980s based on the oxygen isotopic signal (δ 18 O) from the by increasing greenhouse gas concentrations in the 21st cen-
Quelccaya ice core (Thompson et al., 1986). The LIA was tury (Seth et al., 2010). Hence, there is an urgent need to
recognized in this record as a period with more negative val- better document and understand the causes of monsoon vari-
ues of ice δ 18 O, originally interpreted as reflecting cold pe- ations in response to natural forcing during the most recent
riods. Other studies based on speleothem δ 18 O records from past (Vuille et al., 2012). Yet modeling studies, which could
the eastern Andes suggested that rainfall during the LIA in- help diagnose past changes in SASM dynamics, still suffer
creased by approximately 20 % compared to the 20th cen- greatly from a lack of proxy data to validate their simula-
tury (Reuter et al., 2009). Moreover, a recently published tions.
study based on authigenic calcite δ 18 O deposited in annu- Here we explore new high-resolution δ 18 O records
ally laminated lacustrine sediments of a high-altitude lake of well-dated speleothems PAL3 and PAL4 collected in
on the eastern flank of the Andes postulated an intensifica- Palestina cave on the eastern side of the Andes Cordillera
tion of the South American Summer Monsoon (SASM) dur- (northeastern Peru) and discuss them in the light of exist-
ing the LIA period and diminished SASM activity during ing paleoclimate proxies along the Andes, which have simi-
the MCA interval (Bird et al., 2011a). The climatic controls larly been interpreted as recording the intensity of the SASM.
on the δ 18 O signal in Andean ice core records are consid- The Palestina records span the last millennium with ∼ 5-year
ered to be the same as those proposed from the carbonate temporal resolution, allowing us to explore the SASM vari-
records from speleothems and lake records within the SASM ability as recorded at this site from sub-decadal to centennial
domain (Vuille et al., 2012) because all these records pri- time-scales. Time series and wavelet analyses are performed
marily reflect the isotopic composition of monsoonal rainfall in order to identify oceanic and atmospheric modes and their
(Vuille and Werner, 2005; Vimeux et al., 2005). However, the teleconnections affecting past SASM activity. Additionally,
actual regional-scale response to changes in radiative forc- composite techniques are applied to explore possible mecha-
ing during the MCA and LIA is still poorly understood due nisms that could explain the observed spatiotemporal SASM
to the lack of sufficient highly resolved and chronologically variability across the tropical South American continent over
well-constrained isotopic proxy records from the region. For the past 1600 years.
example, it remains unclear how changes in rainfall parti-
tioning between the Amazon lowlands and the Andean high-
elevation sites may have affected the isotopic response (e.g., 2 Study area and modern climatology
Bird et al., 2011b). Previous studies have also detected differ-
ences in the timing of MCA and LIA in tropical South Amer- The Palestina cave (Nueva Cajamarca, San Martín, Peru)
ica (Bird et al., 2011a; Vuille et al., 2012), but it is debatable was explored over a length of 2380 m by the Bristol Ex-
to what extent these offsets are real or related to chronologi- ploration Club (BEC, UK) in 2003, and mapped by the
cal errors. Several studies have also pointed toward a distinct French–Peruvian GSBM–ECA team in 2011. The cave is
antiphased regional rainfall response during the LIA, with located in northeastern Peru (5.92◦ S, 77.35◦ W), in the up-
increased precipitation over the tropical Andes and southern per Amazon Basin along the eastern margin of the Andes
Brazil (Oliveira et al., 2009; Vuille et al., 2012), while drier (870 m a.s.l.) in a Triassic–Jurassic limestone–dolomitic for-
conditions were recorded over northeastern Brazil (Novello mation (INGEMET-Peru; Fig. 1). The present-day climate is
et al., 2012). Similarly a relatively dry climate is documented tropical humid with a mean annual temperature of 22.8 ◦ C,
during the MCA over the Andes and in NE Brazil, while and mean annual precipitation around 1570 mm, measured
∼ 15 km from the cave, in the meteorological station of Rioja

Clim. Past, 10, 1967–1981, 2014 www.clim-past.net/10/1967/2014/

 J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years 1969

east of the Amazon region (Ronchail et al., 2002; Espinoza

et al., 2009, 2013) as a result of adjustments in the location
and strength of the Walker and Hadley circulations to the un-
derlying Pacific sea surface temperature anomalies (SSTA;
Ronchail et al., 2002; Garreaud et al., 2009; Marengo et al.,
2012). The influence of Atlantic SSTA on the interannual
variability of the SASM has not been studied in similar detail
as the influence of Pacific SSTA. Warm SSTA in the tropical
North Atlantic region have been associated with drought con-
ditions over the southern (e.g., Yoon and Zeng, 2010; Ron-
chail et al., 2002; Marengo et al., 2008) and western Ama-
zon Basin (Espinoza et al., 2011; Lavado et al., 2012). For
instance, the 2005 and 2010 droughts over Amazonia were
not related to ENSO but to anomalously warm SST in the
tropical North Atlantic (Espinoza et al., 2011; Marengo et
al., 2008, 2012).
On decadal and longer timescales SASM activity is less
well understood, mainly because of limitations in the instru-
mental data. The Pacific Decadal Oscillation (PDO) is con-
1
Figure 1. Geopotential height (contours, in geopotential meters – sidered an important modulator of South American precipi-
2 gpm) and total wind at 850 hPa from ERA-40 for the 1975–2002 tation, with a very similar, albeit slightly weaker, spatial foot-
3 Fig. 1: Geopotential height (contours, in gpm) and total wind at 850 hPa from ERA-40 for the 1975-
4 period and mean daily rainfall from CMAP data for the 1979–2002
2002 period and mean daily rainfall from CMAP data for the 1979 - 2002 period. A) During DJF print over South America as ENSO, and with teleconnection
5 period.
season. Panel
B) During JJA(a) shows
season. during
Limit of DJFBasin
the Amazon season; (b) during JJA sea-
is shown. mechanisms that are also similar to those observed for ENSO
6 Numbers in figure of
son. Limits indicate locations of other
the Amazon Basin proxy
arerecords
shown. in South
TheAmerica
numbers1. Palestina
in therecord
7 (this study); 2) Cascayunga Cave record (Reuter et al., 2009); 3) Pumacocha Lake record (Bird et on interannual timescales (Garreaud et al., 2009). Linkages
8 figure indicate locations of other proxy records in South America.
al., 2011a); 4) Huagapo Cave record (Kanner et al., 2013); 5) Quelccaya ice cap (Thompson et al., to the Atlantic Multidecadal Oscillation (AMO), with peri-
9 Point
1986); (1) Cave
6) DV2 Palestina recordet(this
record (Novello study);
al., 2012); (2)Cave
7) Cristal Cascayunga cave
Record (Taylor, record
2010).
10
odicities of ∼ 64 years have also been documented as an im-
(Reuter et al., 2009); (3) Pumacocha lake record (Bird et al., 2011a);
(4) Huagapo cave record (Kanner et al., 2013); (5) Quelccaya Ice
portant factor driving SASM variability, based on evidence
Cap (Thompson et al., 1986); (6) DV2 cave record (Novello et al., from continental proxy records (Novello et al., 2012), marine
2012); (7) Cristal cave Record (Taylor, 2010). sediments (Chiessi et al., 2009) and model simulations (e.g.,
Parsons et al., 2014). Yet previous paleoclimatic studies per-
formed at high and lowland sites in the Andes (e.g., Reuter
et al., 2009; Bird et al., 2011a) did not perform statistical
during the period 1963–2012. Rainfall displays a bimodal analyses of the stable isotope data to confirm the presence of
distribution featuring a rainy season between October and decadal to multidecadal changes in monsoon intensity. Thus,
March, which represents around 63 % of the total annual there is still virtually no information available on the climate
rainfall. A tropical circulation regime allowing precipitation response to multidecadal forcing over the Amazon region.
during the winter and early spring (April–September) is re- 28 Oxygen isotopes (δ 18 O) in rainfall over South America
sponsible for the remaining 37 % of total rainfall. This pre- suggest that fractionation processes are to some extent re-
cipitation regime results from the seasonal march of convec- lated to the amount effect (between −0.4 and −0.8 ‰ per
tive activity over the tropical continent associated with the 100 mm increase in the mean annual precipitation; Reuter et
onset, mature phase and demise of the SASM (Marengo and al., 2009). Moreover, recent studies based on isotope-enabled
Nobre, 2001; Espinoza et al., 2009; Garreaud et al., 2009). numerical models demonstrate that the degree of moisture
On interannual timescales, changes in SASM intensity are recycling and rainout upstream over the Amazon Basin as-
partially related to sea surface temperature (SST) anomalies sociated with the intensity of the SASM is a significant first-
in the tropical Pacific associated with El Niño–Southern Os- order controlling factor (Vuille and Werner, 2005; Vuille et
cillation (ENSO) (Nogués-Paegle and Mo, 2002; Garreaud et al., 2012; Kanner et al., 2013). The SASM is considered to
al., 2009; Grimm, 2010 and references cited therein). During be the main contributor to seasonal rainfall in the study area
the warm or positive phase of ENSO (El Niño), below aver- (63 %), although precipitation related to residual equatorial
age precipitation during the summer wet season is observed rainfall during austral winter is still significant (37 %) and
in northern South America, especially over the northeast of isotopically different from the SASM δ 18 O signature.
the Amazon Basin (Ronchail et al., 2002), the tropical An-
des (Garreaud et al., 2003) and occasionally in the western
Amazon (Espinoza et al., 2011). The opposite conditions are
observed during cold or negative phases (La Niña), where
abundant rainfall and flooding occurs in the north and north-

www.clim-past.net/10/1967/2014/ Clim. Past, 10, 1967–1981, 2014

1970 J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years

3 Materials and methods vice, mounted for XYZ motion control, in order to obtain
a high resolution for the most recent period recorded in the
A monitoring program was setup between May 2011 and speleothem. Between 16 and 39 mm, a 0.3 mm sampling in-
September 2013 in order to evaluate the isotopic signature of terval was used for comparison with δ 18 O values of PAL4
precipitation and cave seepage waters related to speleothem samples. These samples were also analyzed in the Labo-
formation. Water samples for isotope analyses (15 mL bot- ratório de Estudos Geodinamicos e Ambientais at the Uni-
tles) were collected twice a month in the isotope rainfall col- versity of Brasilia (UnB) using a Kiel IV carbonate device
lector installed close to the cave entrance, and also in the coupled to a Finnigan MAT 253 mass spectrometer (Thermo
Palestina spring that forms the Jordan river (which forms Fisher Scientific). This device allows optimizing the sample
Palestina cave). Additional samples from cave drip waters mass for analyses (≤ 20 µg of carbonate), and the analyti-
were taken for comparison with the calcite signature ob- cal precision obtained is ±0.06 ‰ for δ 13 C and ±0.09 ‰
tained from the speleothems collected. Unfortunately, logis- for δ 18 O.
tical problems lead to a loss of most samples during year Spectral analysis techniques were performed on the annu-
1 of the monitoring program. Therefore data availability is ally interpolated composite record of PAL4 and PAL3 δ 18 O
sparse for 2011, but continuous for most of the variables dur- time series. Wavelet analysis (Torrence and Compo, 1998)
ing the period January 2012–October 2013. Water analyses is used to display the frequency variability of the δ 18 O time
were performed in the Laboratorio de Estudos Geodinâmi- series, while cross-wavelet analysis (Grinsted et al., 2004) is
cos e Ambientais at the University of Brasilia (UnB) using used to test the similarity and coherence in periodicities be-
a Picarro L2120-i water isotope analyzer which allows mea- tween our δ 18 O records and other reconstructed indices on
surements at an analytical precision of 0.03 ‰ and reports multidecadal timescales. In order to quantify the AMO influ-
to δ 18 O ‰ relative to Vienna Standard Mean Ocean Water ence on South American paleo-records during the common
(SMOW) standard. period between AD 734 and 1922, δ 18 O composites were
Two speleothems were collected in the Palestina cave at created for both positive and negative phases of the AMO,
a distance of 600 and 700 m from the entrance and ∼ 80 m using the 25th and 75th percentiles as thresholds to define
below the surface. Stalagmites PAL3 and PAL4 are ∼ 10 the negative (AMO−) and positive (AMO+) phases, respec-
and ∼ 17 cm tall. The stable isotope profiles presented in tively. A Student’s t test was employed to test whether de-
this study were obtained from the first 40 mm (PAL3) and partures in the δ 18 O composites were significantly differ-
80 mm (PAL4) of these speleothems, respectively. The age ent from the long-term mean in each proxy. Regional maps
model developed for PAL4 is constrained by 13 U–Th ages for δ 18 O during AMO+ and AMO− were created to assess
and the one for PAL3 by 6 U–Th ages. The U–Th analyses whether spatial patterns emerge that would characterize the
were performed at the Minnesota Isotope Laboratory, Uni- AMO influence on δ 18 O records.
versity of Minnesota, using an inductively coupled plasma
mass spectrometry (ICP-MS) technique (Cheng et al., 2013).
The chemical procedures used to separate uranium and tho- 4 Results and discussion
rium fractions for 230 Th dating are similar to those described
by Edwards et al. (1987), where most dates present errors 4.1 Stable isotopic composition of precipitation
of 2σ < 1 %, representing a mean value of ∼ 15 years (Ta- and speleothems
bles S1 and S2 in the Supplement). The chronological model
was developed for each speleothem by linearly interpolating The local meteorological water line (LMWL) obtained from
ages between dates. 40 values of δ 18 O and δD pairs from each rainfall sam-
Analyses of δ 18 O were performed with a Finnigan Delta pled nearby Palestina cave has a slope of ∼ 8.2, which is
Plus Advantage (Thermo Fisher Scientific) mass spectrome- very close to the global meteorological water line (GMWL;
ter, in the Laboratorio de Estudos Geodinâmicos e Ambien- Rozanski et al., 1992; Fig. S3 in the Supplement). Rainwater
tais at the University of Brasilia (UnB). For δ 18 O, error anal- samples exhibit a range of δ 18 O from −0.61 to −15.12 ‰
yses were 0.1 ‰. Records are reported as δ 18 O ‰ relative to and co-vary with seasonal changes in precipitation. Still, the
the Vienna Pee Dee belemnite (PDB) standard. 264 samples amount effect does not explain the total variance in the iso-
were collected over a length of 8 cm, along the growth axis topic composition of precipitation (R 2 = 0.3, p < 0.01). Ad-
of PAL4 stalagmite, sampling at an interval of 0.3 mm, using ditional influences, mostly related to the degree of rainfall
a Sherline micro-drill model 5400, coupled to an automated upstream over the Amazon Basin, affect the isotopic signal
XYZ Stage. This sampling approach provides a temporal res- (Bird et al., 2011a; Vuille et al., 2012; Kanner et al., 2013).
olution between 2 and 10 years (∼ 5 years). During the austral summer season, a higher percentage of
The oxygen isotope profile of PAL3 is based on 200 anal- isotopically depleted air masses originating over the Atlantic
yses using a 100 µm sampling interval for the first 16 mm Ocean and transported across the Amazon Basin reach the
using a micro-milling machine system with a high-speed pre- cave site. These more distal air masses have been affected by
cision drill and stereomicroscope with a charged coupled de- the loss of heavy isotopes during periods of strong convective

Clim. Past, 10, 1967–1981, 2014 www.clim-past.net/10/1967/2014/

J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years 1971

activity over the core SASM region. During austral winter, and in the water infiltrating the cave environment, most likely
on the other hand, enriched values of δ 18 O characterize the related to SASM precipitation.
residual tropical rainfall in association with diminished rain- A composite oxygen isotope (δ 18 O) time series was cre-
out upstream due to weaker convective activity. ated from the PAL4 and PAL3 records (Fig. 3a). The com-
Rainwater exhibits relatively large variations in compari- posite δ 18 O record, hereafter referred to as the Palestina
son with river water (from −6.47 to −11.91 ‰, n = 22) and record, indicates major changes in the oxygen isotope ra-
drip waters (from −5.42 to −7.42 ‰, n = 7). Speleothem tio over different periods. Heavier values of δ 18 O are
calcite δ 18 O presents mean values of −7.11 ‰ (PDB), which recorded between AD 421–580, 722–820 and during the
is consistent with mean values recorded for river water MCA (AD 920–100) which is marked by a double peak
(−7.9 ‰, SMOW) and rain water (−6.73 ‰, SMOW). The of enriched δ 18 O values centered at AD ∼ 940 (∼ −6.3 ‰)
limited range of values in the river and drip waters indicates and at AD ∼ 1025(∼ −5.8 ‰). Lighter or more negative val-
attenuation of the seasonal isotopic variations in rainfall re- ues are observed between AD 580–720, 820–920 and dur-
lated to mixing during meteoric water recharge. This effect ing the period 1325–1820, which corresponds to the Little
might be triggered by variations in the hydraulic head de- Ice Age in the Palestina record. The LIA period is charac-
pending on rainfall intensity as suggested for extreme values terized by a gradual but substantial decrease in δ 18 O val-
in δ 18 O from the river waters, which allows observations of ues with minimum values observed between AD 1400 and
variations in relation to rainfall events. In this sense, intense 1593 (−7.6 ‰; Fig. 3a), after which values increase gradu-
recharge during the humid monsoon season is what triggers ally toward AD 1820, representing the end of the LIA in the
the infiltration and drip inside the cave, and hence degassing Palestina record.
and deposition of calcite inside the cave. Spectral analysis of our record indicates significant peri-
odicities centered on 70, 44 and 29 years, within 95 % sta-
tistical confidence (Fig. S5). Wavelet analyses support these
4.2 Description of the Palestina record
results, indicating statistically significant superimposed peri-
odicities (Fig. S6). During most of the MCA low-frequency
The PAL3 and PAL4 calcite speleothems present micro- variability dominates, centered at ∼ 70 and 48 years. During
crystalline fabric and no apparent “hiatus” was detected. the LIA, decadal variability (∼ 10 years) appears to be more
The twenty U–Th dates in PAL4 and PAL3 reveal that significant and persistent.
samples cover the period from AD 413 ± 10 to 1824 ± 16
years and from AD 1096 ± 15 to 1925 ± 8 years, respec- 4.3 Comparison with other regional isotopic proxies
tively. Mean growth rates for PAL4 and PAL3 are 0.056 and
0.049 mm yr−1 , respectively. Sampling spacing of 0.3 mm A comparison between the Palestina cave isotope record and
for PAL4 and 0.01 mm for PAL3 yields a data resolution other high-resolution proxies of SASM activity in the An-
between 2 and 8 years (∼ 5 years) for both speleothems des is shown in Fig. 3. There is a striking resemblance be-
(Fig. 2). For the climate interpretation of spectral signals, tween the Palestina and Pumacocha lake records at decadal
however, we chose a conservative estimate and refrain from to centennial timescales (Bird et al., 2011a) throughout most
interpreting signals at frequencies below 16 years (twice the of the record (Figs. 3c and S7), confirming the regional-scale
lowest resolution of the record). nature of the δ 18 O signal and that the mean state changes
Stable oxygen isotopes for the entire records in PAL4 and of the SASM during the MCA and LIA periods also af-
PAL3 stalagmites present absolute mean values of −7.14 and fected Palestina cave. Between AD 430 and 900, however,
−7.08 ‰, respectively. These values are in agreement with the Palestina cave isotope record differs from the Pumacocha
other results obtained from speleothems in the same region record. Afterwards similarities between the two records be-
(Van Breukelen et al., 2008; Reuter et al., 2009). Stable oxy- come more evident, suggesting a stronger regional control
gen and carbon isotope ratios (δ 18 O and δ 13 C) are poorly of the SASM during the past millennium. Interestingly, two
correlated along the growth axis for both speleothems (PAL4, distinctive decadal-scale peaks of more positive values cen-
R 2 = 0.26, n = 252 and PAL3, R 2 = 0.01, n = 195, Fig. S4), tered at AD ∼ 940 and ∼ 1020 are recorded in both prox-
which indicates that calcite deposition occurred close to equi- ies during the MCA. Furthermore, a gradual decrease of
librium conditions with drip water (Hendy, 1971). Further δ 18 O values starting at AD ∼ 1325 can be identified in both
tests to rule out kinetic fractionation were not carried out due records, but whereas minimum values are observed between
to the difficulty of extracting samples from exactly the same AD 1480 and 1600 in the Palestina record, they occur more
layer. Moreover, even though the speleothems have different than a hundred years later in the Pumacocha record. Nonethe-
sampling resolutions and age models, the two oxygen isotope less, the Palestina record confirms many characteristic cli-
records are nearly identical both in terms of the mean values matic features from the past millennium first detected in the
and the magnitude of change through the overlapping ∼ 900- Pumacocha record, which is quite remarkable given the dif-
year interval (Fig. 2). These similarities suggest that a com- ferent origin of the calcite (speleothem vs. lake sediments)
mon factor governs the isotopic signal in both speleothems, obtained from very different Andean environments (low-

www.clim-past.net/10/1967/2014/ Clim. Past, 10, 1967–1981, 2014

1972 J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years

1
Figure 2. Stable isotope time series of δ 13 C and δ 18 O for PAL4 (blue line) and PAL3 (cyan line), respectively. U–Th dates and corresponding
2
error bars are represented by red and black dots for PAL4 and PAL3 stalagmites, respectively.
3 Fig. 2: Stable isotope time series of δ13C and δ18O for PAL4 (blue line) and PAL3 (cyan line)
4 respectively. U/Th dates and corresponding error bars are represented by red and black dots for
5 PAL4 and PAL3 stalagmites respectively
lands vs. high Andes), ∼ 500 km away and 3400 m elevation a comparison with the variations observed by the Palestina
difference. The6 fact that the Palestina record shows evidence record at the beginning of the last millennium.
of many of the7 same features detected in Pumacocha (e.g., When compared with the Huagapo cave record (Fig. 3d),
the double-peak structure of the MCA), highlights the value an anti-phased relationship can be observed between AD 400
of data replication and of establishing dense regional proxy and 900. The Huagapo cave record also does not show an
networks. Only by replicating such signals with independent abrupt migration to heavier values during the MCA as is ob-
records can they unambiguously be attributed to climate forc- served during the period AD 900–1100 in the Palestina and
ing as opposed to non-climatic effects. Pumacocha records. Afterwards, the δ 18 O values at Huagapo
Other records based on speleothems of the Andean re- cave gradually decrease between AD 1365 and 1820, which
gion are represented by δ 18 O series from the Cascayunga is in agreement with the LIA period in the Palestina record.
and Huagapo caves. Comparison with the Cascayunga The onset of the gradual decrease in δ 18 O at Huagapo is
speleothem record (Reuter et al., 2009; Fig. 3b), which is lo- delayed by ∼ 40 years when compared with Palestina and
cated close to Palestina cave (∼ 20 km away), shows gener- Pumacocha, but this time frame is close to chronological un-
ally similar variations in δ 18 O during the last 900 years. Two certainty. However, the LIA period, defined in the Huagapo
periods of heavier δ 18 O values stand out in the Cascayunga record as between AD 1450 and 1850 (Kanner et al., 2013),
record: AD 1278–1320 and 1485–1570. These features are to is shorter than for the other records in the Andean region pre-
some extent also apparent in the Palestina record; especially sented earlier.
the latter period, which is less evident in all other Andean The Quelccaya ice core (Thompson et al., 1986; Fig. 3e)
δ 18 O reconstructions. Nevertheless, these features are shared also shows similar variations in the mean state of the SASM
with a T /P index of upslope convective activity based on during the past 1500 years. This record also confirms the re-
the ratio between cloud transported pollen from the Andean gional coherence in the δ 18 O signal along the Andes; how-
forest to the bog (T ) and Poaceae pollen frequencies, related ever, there are several offsets in the timing of key periods
to the edaphic moisture of the Paramo (P ) record obtained such as the MCA and LIA. Recent definitions of the MCA
from Papallacta in the Andes of Ecuador (Ledru et al., 2013). (AD 1100–1300) and LIA (early LIA 1520–1680, late LIA
It should be noted however, that the Cascayunga record lacks 1681–1880) events in the Quelccaya record 29 (Thompson et
chronological control between ∼ 1090 and 1450. If the δ 18 O al., 2013) differ from the timing of these events as detected
record from Cascayunga is adjusted within the age uncer- in carbonate records presented above. Moreover, a double
tainties as suggested in Fig. 3b, the discrepancies between peak of heavier δ 18 O values is observed at AD 990 and
proxies could at least partially be resolved. Unfortunately, the 1080, which might be related to the double peak observed
Cascayunga record does not extend back far enough to allow in the Palestina and Pumacocha records. Centennial-scale

Clim. Past, 10, 1967–1981, 2014 www.clim-past.net/10/1967/2014/

J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years 1973

2
Figure 3. Comparison between Fig.Palestina
3: Comparison between record
composite Palestina(gray
composite
linerecord
in the(gray line in the background)
background) and other and other Andean records with their re-
eastern
3 eastern
spective chronological controls Andeanbars.
and error records
From with top
theirtorespective
bottom:chronological controls
(a) Palestina and (this
record error study);
bars. From
(b)topCascayunga
to record (Reuter et al.,
2009); (c) Pumacocha lake4 record (Bird
bottom: et al., 2011a);
a) Palestina record (this (d) Huagapo
study); cave record
b) Cascayunga record(Kanner
(Reuter et et
al.,al., 2013);
2009); (e) Quelccaya Ice Cap (Thompson et
c) Pumacocha
al., 1986). Shaded background
5 represents
Lake MCA
record (Bird (red)
et al., andd)LIA
2011a); periods
Huagapo Cave(light
recordblue).
(Kanner et al., 2013); e) Quelccaya ice
6 cap (Thompson et al., 1986). Shaded background represents MCA (red) and LIA periods (light
7 blue).
variability of diminished δ 18 O values during the LIA pe- An interesting finding is revealed by comparisons with
riod and timing of the demise of this period is consistent and 30
speleothem records from northeastern South America, col-
roughly contemporaneous with the other proxies discussed. lected at Diva de Moura cave (DV2 record) in the south
Comparison with other high-resolution proxies of the of the state of Bahia in Brazil (Fig. 4b). During the MCA,
SASM far from the Andes Cordillera also shows coherent defined in this record as the period between AD 890 and
fluctuations of δ 18 O values during the LIA. Over southeast- 1154, heavier values in δ 18 O are observed, resembling the
ern tropical South America at Cristal cave in Brazil (Fig. 4c), records from Palestina and Pumacocha. It is quite remark-
a period of lighter values is observed during the LIA, albeit able to see the similarities between these records, all featur-
with some delay in the onset of the minimum values when ing the same structure of a double peak during the MCA,
compared with Andean records (Taylor, 2010; Vuille et al., which is not evident in other proxies in South America. Dur-
2012). Minimum values of δ 18 O in Cristal cave are observed ing the LIA, however, variations in the DV2 isotope record
during a relatively short period from AD 1520 to 1736 only. are the opposite of what is seen in other records under the
Hence, the start of the LIA in Cristal cave occurs about one influence of the SASM. Moreover, the onset of a trend to-
hundred years later than the onset of the LIA in the Palestina ward heavier values in the DV2 record occurs around the
record and ends a hundred years earlier than the demise of year AD 1333, which is contemporaneous with the onset of
this period in other Andean records. the LIA in many Andean records, supporting the notion of

www.clim-past.net/10/1967/2014/ Clim. Past, 10, 1967–1981, 2014

1974 J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years

2
Figure 4. Comparison between Palestina cave δ 18 O record (gray line in the background) and high-resolution proxies of the SASM from
non-Andean records: (a) Palestina
3
record (this study), (b) DV-2 cave record (Novello et al., 2012) and (c) Cristal cave record (Taylor, 2010;
Fig. 4: Comparison between Palestina cave δ18O record (gray line in the background) and high-
Vuille et al., 2012).
4 resolution proxies of the SASM from non-Andean records: a) Palestina record (this study), b) DV-2
5 cave record (Novello et al., 2012); c) Cristal Cave record (Taylor, 2010; Vuille et al., 2012).
6
existingteleconnectionsbetween both regions. At the same ity. A very distinctive double peak structure is observed on
time the DV2 record shows similar periodicities affecting decadal timescales, centered at AD ∼ 934 and ∼ 1039, re-
summer precipitation as detected in the Palestina record, sulting in the periods of lowest SASM activity in the en-
such as of 76, 65, 40, 22 and 15 years (Novello et al., 2012). tire record. Interestingly these features are shared with other
These statistically significant periodicities suggest a com- records in South America such as Pumacocha lake in the
mon large-scale mechanism along the east–west domains of eastern Andes, the DV2 record over northeastern Brazil and
the SASM; probably related to similar climate mechanisms, peaks of Ti (%) in the Cariaco sediment record, indicative
which are most evident at multidecadal timescales. of a displacement of the Intertropical Convergence Zone
(ITCZ) to a more northerly position (Haug et al., 2001). The
4.4 Climatic interpretation coherence among records from different sites and proxies
suggests that a common mechanism triggered these isotopic
The Palestina oxygen isotope record is interpreted here as a variations across much of the northern part of the continent.
function of SASM activity, consistent with previous studies In addition, intense humid events recorded in the Chaac sta-
in the eastern Andes (Bird et al., 2011a; Vuille et al., 2012; lagmite on the Yucatan Peninsula are coincident with those
Kanner et al., 2013). It shows marked centennial-scale vari- peaks (Medina et al., 2010). The sum 31 of all these records sug-
ability spanning the entire record of 1600 years. Two peri- gests that these periods could have been characterized by a
ods with reduced SASM activity stand out from AD 410 to reduced moisture influx in to the SASM system during aus-
570 and from AD 720 to 820. Both periods were followed tral summer due to a more northerly position of the ITCZ,
by relatively abrupt changes in enhanced SASM intensity and hence more intense rainfall over the tropical Northern
from AD 580 to 720 and between AD 820–920. Higher over- Hemisphere during boreal summer. At Palestina cave, the re-
all variability in the δ 18 O record from Palestina, when com- duced SASM would have lead to less rainout upstream over
pared with other Andean δ 18 O records, would suggest higher the Amazon Basin and therefore more enriched δ 18 O val-
overall precipitation variability at this site during these above ues. It is also possible that reduced summer precipitation, and
periods. It is worth noting that variations in δ 18 O values for therefore a relatively higher proportion of austral winter pre-
these periods are on the same order of magnitude but more cipitation with an enriched signature of δ 18 O could partially
sudden than during the MCA and LIA. explain the magnitude of the two double peaks during the
The Medieval Climate Anomaly, spanning the period be- MCA.
tween AD ∼ 920 and 1100, shows diminished SASM activ-

Clim. Past, 10, 1967–1981, 2014 www.clim-past.net/10/1967/2014/

J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years 1975

Figure 5. (a) Comparison between the reconstructed AMO index (Mann et al., 2009; brown line) and the Palestina record (blue line); (b
and c) Departure of δ 18 from long-term mean during periods when AMO was above the 75th percentile (AMO+, left) and below the 25th
percentile (AMO−, right). Significant departures from long-term mean at p < 0.1 based on a Student’s t test are indicated by red (increase
in δ 18 O and blue (decrease in δ 18 O) circles. Yellow circles indicate records that do not exhibit any statistically significant changes.

Precipitation over western Amazonia is especially sensi- cal South America, hydroclimate variability during the MCA
tive to changes in tropical Atlantic SSTs, with reduced pre- might be related to leading modes of North Atlantic climate
cipitation when positive SST anomalies are observed in the variability. Indeed, previous works have already documented
tropical North Atlantic (Espinoza et al., 2011, 2012). There the influence of the Atlantic modes in rainfall distribution and
is evidence pointing to unusually warm sea surface temper- intensity over tropical South America (Reuter et al., 2009;
ature anomalies (SSTA) in the North Atlantic sector during Bird et al., 2011a; Vuille et al., 2012).
the MCA (Keigwin et al., 1996), with characteristics that re- Above comparisons between the AMO index (Mann et al.,
semble the positive phase of the AMO (Feng et al., 2008). In- 2009) and the Palestina record revealed strong similarities.
deed, there are indications that the positive phase of the AMO The influence of the AMO on South American δ 18 O records
leads to reduced SASM intensity at multidecadal timescales, is further evaluated by comparing departures in δ 18 O val-
linked to a northward migration of the ITCZ (Chiessi et ues during positive (+) and negative (−) phases of the AMO
al., 2009; Strikis et al., 2011; Bird et al., 2011a; Novello et (Fig. 5b). Statistically significant negative isotopic anomalies
al., 2012). Periodicities observed in the Palestina cave series are observed during AMO− over the southern part of tropical
highlight a multidecadal influence over the MCA in concor- South America, which would suggest an enhancement of the
dance with the AMO frequency of ∼ 65 years, as diagnosed SASM during the negative phase of the AMO, affecting pre-
in both model and instrumental data (Knight et al., 2006; dominantly the southern parts of the SASM domain. During
Fig. S6). The reconstructed AMO index published by Mann AMO+ phases, the isotopic response is spatially less coher-
et al. (2009) shows similar characteristics as the Palestina ent, with a mixture of non-significant and significant posi-
record (Fig. 5a) and also documents a period of persistent tive anomalies scattered across the region. Nonetheless, it is
positive AMO anomalies during the MCA period, including worth noting that there are no significant negative anoma-
the same double peak structure, albeit with some lag in the lies in any of the records during the AMO+ phase. While
time series. This observation suggests that in northern tropi- most records show a linear response with anomalies of the

www.clim-past.net/10/1967/2014/ Clim. Past, 10, 1967–1981, 2014

1976 J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years

opposite sign during the two phases of the AMO, this is not SASM activity during the MCA and opposite conditions
the case at all sites. The DV2 record from northeastern South during the LIA. Both periods were initially recognized
America, in particular, stands out prominently with more en- in the Northern Hemisphere where the median of recon-
riched values during both positive and negative phase of the structions document mostly warm conditions from about
AMO. It is worth noting that the Palestina record does not AD 950 to about 1250 (MCA) and colder conditions at about
exhibit significant anomalies (p ≤ 0.1) during either phase AD 1450–1850 (LIA; IPCC, 2013). In tropical South Amer-
of the AMO despite featuring strong similarities with the ica, however, the definition of the MCA and LIA events
AMO index. Most likely other factors such as changes in is quite variable between studies and proxies considered,
the relative precipitation contribution during different sea- and discrepancies are observed in relation to the intensity
sons also play a relevant role. This feature is coherent with and temporal definition of these events (between studies and
recent works that documented the role of Pacific and Atlantic proxies considered and discrepancies, e.g., Rabatel et al.,
oceans on rainfall variability during the austral summer and 2008; Vuille et al., 2012). Indeed, this variability does not
winter, respectively (Yoon and Zeng, 2012; Espinoza et al., seem to be related simply to chronological errors of individ-
2011, Marengo et al., 2013). ual records but may also show a latitudinal migration of the
The transition period from the MCA to the LIA (AD ∼ climatic perturbation and its isotopic footprint in and out of
1100 to 1325) shows mean values of −7.1 ‰ in δ 18 O char- this region along the Andes. In order to quantify variations
acterized by persistent decadal-scale variability. The LIA pe- between both events along the Andean region, the average
riod itself is characterized by a substantial increase in SASM δ 18 O difference between MCA and LIA periods as defined
activity from AD ∼ 1325 to 1820, reaching its maximum in- in the Palestina record (AD 900–1100 and 1325–1826) was
tensity between AD 1400 and 1593 (−7.6 ‰; Fig. 2). In- calculated for all Andean records and divided by their re-
creased SASM activity during the LIA is synchronous with spective standard deviation to make them comparable across
cold events in the Northern Hemisphere (e.g., Gray et al., proxies.
2006; Mann et al., 2009). These cold conditions forced a Larger differences in δ 18 O values between the MCA and
southward migration of the ITCZ (Haug et al., 2001; Reuter the LIA can be observed for the records located in the north-
et al., 2009; Bird et al., 2011a; Vuille et al., 2012; Novello ern region of the Andes (< 11◦ S), when compared with prox-
et al., 2012) as documented by diminished Ti concentrations ies located in more southern latitudes (Fig. 6). Taking into
in the Cariaco Basin during the LIA (Haug et al., 2001) and account that the δ 18 O signature is sourced from the Atlantic
a significant decrease in SSTs over the tropical North At- Ocean and that even the altitudinal location of the prox-
lantic (Black et al., 2007). Since the ITCZ serves as the ma- ies does not erase the SASM signal (Vuille et al., 2012),
jor moisture source fueling the SASM, an intensification of it is plausible to suggest that larger MCA–LIA differences
this system is to be expected from a more southerly posi- at northern locations might be caused by influences other
tion (Vuille et al., 2012), and consistent with the more neg- than summer precipitation. In Fig. 6 we also show the sea-
ative δ 18 O values observed in different proxies across An- sonal variation coefficient of rainfall (SVC), which is the ra-
dean records. The prominent exception to this rule is the tio between the standard deviation of monthly rainfall values
DV2 record in northeastern South America, which shows the (1963–2003) and the mean of monthly rainfall values, con-
exact opposite behavior during the LIA with more enriched sidering 756 stations in the Amazon Basin (Espinoza et al.,
δ 18 O values suggesting dry conditions. This antiphased be- 2009). Indeed, high SVC display a strong rainfall seasonal-
havior, however, might reflect the intensification of the Bo- ity (i.e., large difference between the wet and dry season),
livian high–Nordeste low pressure system, with increased while low SVC values represent weak rainfall seasonality
SASM activity and related convective heating over the south- (i.e., small difference between the wet and dry season). The
western portion of the Amazon region (Lenters and Cook, SVCs show reduced seasonality over the northern part of the
1997), being balanced by enhanced subsidence and drying Andean region. Hence, it is reasonable to assume that during
over the northeast of the continent (Novello et al., 2012). periods of a weakened SASM, the relative contribution of
This mechanism, associated with increased upper level con- austral winter precipitation linked to heavier isotope values
vergence, subsidence and a deficit in summer precipitation would be enhanced. Such an increased contribution of win-
over northeastern Brazil during periods of enhanced SASM ter precipitation synchronous with a weakened SASM during
activity, has been invoked to explain the antiphasing between the MCA would provide an alternative mechanism that could
precipitation in Nordeste rainfall and most of tropical South cause larger differences between both periods in the Andean
America on orbital timescales (Cruz et al., 2009). Neverthe- records, but it might preferentially affect more northern sites.
less, this mechanism may be nonlinear, and dependent on In order to better understand how variations in the SASM
oceanic boundary conditions (e.g., the state of the tropical over a broader area of South America relate to the Atlantic
Pacific), which could explain why this antiphased behavior Multidecadal Oscillation, cross-wavelet and coherence anal-
occurs during the LIA but not during the MCA. yses were developed for both Palestina and DV2 records.
The Palestina record is consistent with other δ 18 O records In both analyses, results suggest significant periodicities of
along the Andes in the sense that it indicates diminished ∼ 64–80 years as being the most important feature in the time

Clim. Past, 10, 1967–1981, 2014 www.clim-past.net/10/1967/2014/

 J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years 1977

Figure 6. Standardized LIA–MCA difference in the δ 18 O of Andean records plotted with the altitudinal locations of each proxy (left panel);
seasonal variability coefficient (SVC) of precipitation calculated for the Amazon Basin (right panel); from Espinoza et al. (2009).

1
Figure 7. Left panels: cross-wavelets between the AMO index and Palestina record (top) and coherence analysis (bottom) between time
2 Fig
series.7:Right
Left Panel:
panels: Cross-wavelets
cross-wavelet between DV2between AMO
record in Bahia index
and AMO and
index Palestina
(top); record
coherence analysis (top)
is also and
shown coherence
(bottom).

3 analysis (bottom) between time series. Right Panel: Crosswavelet between DV2 record in Bahia and
4 series and with stronger amplitudes during the MCA (Fig. 7).
AMO lower frequencies of ∼ 25 years and ∼ 60 years are super-
index (top); coherence analysis is also shown (bottom).
During the LIA, however, even though this low-frequency imposed at different periods. Based on these results, we hy-
signal persists, the wavelet analyses suggest that the two re- pothesize that the lower-frequency signals are related to the
5 gions of the continent were governed by different frequencies Atlantic Multidecadal Oscillation, while the slight signal of a
(Fig. 7). While in DV2 (NE Brazil) ∼ 64 years is the most ∼ 25-year frequency (between AD 1600 and 1800, in Fig. 7)
persistent frequency band, over the eastern Andes higher to could be consistent with one of the most energetic signals of

www.clim-past.net/10/1967/2014/ Clim. Past, 10, 1967–1981, 2014

1978 J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years

the Pacific Decadal Oscillation. It is hard to disentangle the dominant mode of variability is characterized by frequen-
superimposed influences of the Pacific and Atlantic at differ- cies around 65 years in both Palestina and DV2, which is
ent timescales, especially when both oceans trigger distinct consistent with the documented influence of the AMO on
modes that may dampen or reinforce one another and lead to SASM precipitation. During the LIA, however, there is no
nonlinear impacts on rainfall over the South American con- correspondence in the signal between both areas. In fact,
tinent. Nevertheless, based on marine evidence (Sifeddine et decadal variability suggests that other influences may have
al., 2008; Gutierrez et al., 2009; Salvattecci et al., 2014), it is contributed to decoupling the two regions.
possible to assume that during the LIA period, an enhanced Statistical analyses suggest the influence of the AMO in
role of the Pacific would increase rainfall variability over the several South American records, with a generally stronger
eastern Andes. and spatially more coherent response during the negative
phase of the AMO (AMO−). This analysis also highlights
that the AMO influence is not always linear, as shown, for
5 Conclusions example, for the case of the DV2 record, which exhibits sig-
nificant positive departures during both phases of the AMO.
The Palestina record reveals important details of changes Nevertheless, a stronger focus on studying multidecadal vari-
in the mean state of the SASM during the past 1600 years ability of the SASM during both recent and past periods
and confirms most of the major results from previous pale- seems warranted in order to enhance our understanding of
oclimate reconstructions in the eastern Andes. During the this unique monsoon system and to better prepare society
last millennium, SASM activity exhibited departures from for future changes associated with anthropogenic climate
the mean state, which are relatively synchronous with ma- change.
jor periods of global climate change, recognized as MCA
and LIA. The Palestina record adds to the growing network
of high-resolution paleoclimate reconstructions for this An- The Supplement related to this article is available online
dean region and allows replication and thereby confirmation at doi:10.5194/cp-10-1967-2014-supplement.
of the veracity of key climate signals from this region, such
as the double peak of SASM failure during the peak MCA
period. This study highlights that replicating regional cli-
mate signals from different sites and using different proxies
Acknowledgements. We thank Luis Mancini and Ana Carolina Mi-
is absolutely essential for fully understanding past changes
randa for their support during the stable isotope data acquisition at
in SASM activity, but also in separating climate signals from the Universidade de Brasília and Osmar Antunes for his support
proxy-specific non-climatic effects. In this sense, the corre- at Universidade de São Paulo. We also thank Augusto Auler,
spondence between records in the Andes and other regions Daniel Menin and the Espeleo Club Andino (ECA Perú) for their
under influence of the SASM is quite remarkable and high- support during the field work. This work was undertaken as part of
lights the ability of δ 18 O to integrate climate signals across the the PALEOTRACES project (IRD-UFF-UANTOF), HYBAM
large spatial scales affected by common mechanisms such as Project (IRD) and PRIMO cooperative project (CNPq-IRD), and
the SASM. supported by the Fundação de Amparo a Pesquisa do Estado
Comparison between records reveals stronger and more de Rio de Janeiro, Brazil (FAPERJ grant E-26/100.377/2012),
coherent variability during the LIA in the Andean region Fundação de Amparo a Pesquisa do Estado de São Paulo, Brazil
(FAPESP grants 2011/39450394 to F. W. Cruz and 2012/50260-6
and over southeastern South America. Moreover, variations
and University of São Paulo through INCLINE research group) and
in SASM activity between the MCA and the LIA seem to
also by grants 2013CB955902, CNSF 41230524, US NSF grants
be larger over the northern part of the continent, suggesting 0502535 and 3961103404 to L. Edwards and H. Cheng and NSF
a latitudinal dependence of the MCA footprint. Diminished grants 1003690 and 1303828 to M. Vuille. We are also grateful to
SASM activity and an increase in the relative contribution of the anonymous reviewers, whose comments helped to significantly
winter precipitation would be a plausible mechanism for ex- improve the manuscript. We wish to recognize their effort and
plaining higher isotopic values during the MCA in the north- time spent to help put this record into the regional paleoclimate
ern tropical sector of the Andes during this period. Since framework of tropical South America.
modern records reveal reduced seasonality of precipitation
in the northern region of the Amazon Basin, it is reasonable Edited by: E. Zorita
to assume that more positive δ 18 O values could be related to
a differential contribution of seasonal precipitation.
References
In-phase variations between Palestina and DV2 records
during the MCA and opposite conditions during the LIA re- Bird, B. W., Abbott, M. B., Vuille, M., Rodbell, D. T., Stansell, N.
veal that the influence of oceanic modes over the tropical D., and Rosenmeier, M. F.: A 2300-year-long annually resolved
continent is being modulated in different ways depending record of the South American summer monsoon from the Peru-
on the prevailing boundary conditions. During the MCA the vian Andes, P. Natl. Acad. Sci., 108, 8583–8588, 2011a.

Clim. Past, 10, 1967–1981, 2014 www.clim-past.net/10/1967/2014/

J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years 1979

Bird, B. W., Abbott, M. B., Rodbell, D. T., and Vuille, M.: Holocene Garreaud, R. D., Vuille, M., Compagnucci, R., and Marengo, J.:
tropical South American hydroclimate revealed from a decadally Present-day South American climate, Palaeogeogr. Palaeocl.,
resolved lake sediment δ 18 O record, Earth Planet. Sc. Lett., 310, 281, 180–195, 2009.
192–202, 2011b. Graham, N. E., Ammann, C. M., Fleitmann, D., Cobb, K. M.,
Black, D. E., Abahazi, M. A., Thunell, R. C., Kaplan, A., Tappa, E. and Luterbacher, J.: Support for global climate reorganization
J., and Peterson, L. C.: An 8-century tropical Atlantic SST record during the “Medieval Climate Anomaly”, Clim. Dynam., 37,
from the Cariaco Basin: Baseline variability, twentieth-century 1217–1245, doi:10.1007/s00382-010-0914-z, 2010.
warming, and Atlantic hurricane frequency, Paleoceanography, Gray, S. T., Graumlich, L. J., Betancourt, J. L., and Pederson, G.
22, PA4204, doi:10.1029/2007PA001427, 2007. T.: A tree-ring based reconstruction of the Atlantic Multidecadal
Cheng, H., Edwards, L. R., Shen, C.-C., Polyak, V. J., Asmerom, Oscillation since 1567 A.D, Geophys. Res. Lett., 31, L12205,
Y., Woodhead, J., Hellstrom, J., Wang, Y., Kong, X., Spötl, doi:10.1029/2004GL019932, 2004.
C., Wang, X., and Alexander Jr., E.C.: Improvements in 230 Th Grimm, A. M.: Interannual climate variability in South America?
dating, 230 Th and 234 U half-life values, and U–Th isotopic Impacts on seasonal precipitation, extreme events, and possible
measurements by multi-collector inductively coupled plasma effects of climate change, Stoch. Env. Res. Risk A., 25, 537–554,
mass spectrometry, Earth Planet. Sci. Lett., 371-372, 82–91, doi:10.1007/s00477-010-0420-1, 2010.
doi:10.1016/j.epsl.2013.04.006, 2013. Grimm, A. M. and Zilli, M. T.: Interannual Variability and Seasonal
Chiessi, C., Mulitza, S., Paetzold, J., Wefer, G., and Marengo, Evolution of Summer Monsoon Rainfall in South America, J.
J.: Possible impact of the Atlantic Multidecadal Oscillation on Climate, 22, 2257–2275, 2009.
the South American summer monsoon, Geophys. Res. Lett., 36, Grinsted, A., Jevrejeva, S., and Moore, J.: Application of the cross
L21707, doi:10.1029/2009GL039914, 2009. wavelet transform and wavelet coherence to geophysical time se-
Cruz, F. W., Vuille, M., Burns, S. J., Wang, X., Cheng, H., ries, Nonlinear Proc. Geoph., 11, 561–566, 2004.
Werner, M., Lawrence Edwards, R., Karmann, I., Auler, A. Gutiérrez, D., Sifeddine, A., Field, D. B., Ortlieb, L., Vargas, G.,
S., and Nguyen, H.: Orbitally driven east–west antiphasing Chávez, F. P., Velazco, F., Ferreira, V., Tapia, P., Salvatteci, R.,
of South American precipitation, Nat. Geosci., 2, 210–214, Boucher, H., Morales, M. C., Valdés, J., Reyss, J.-L., Campu-
doi:10.1038/ngeo444, 2009. sano, A., Boussafir, M., Mandeng-Yogo, M., García, M., and
Edwards, R. L., Chen, J. H., and Wasserburg, G. J.: 238 U–234 U– Baumgartner, T.: Rapid reorganization in ocean biogeochemistry
230 Th–232 Th systematics and the precise measurement of time off Peru towards the end of the Little Ice Age, Biogeosciences,
over the past 500 000 years, Earth Planet. Sc. Lett., 81, 175–192, 6, 835–848, doi:10.5194/bg-6-835-2009, 2009.
1987. Haug, G. H., Hughen, K., Sigman, D. M., Peterson, L. C., and Röhl,
Espinoza, J. C., Ronchail, J., Guyot, J. L., Cocheneau, G., Filizola, U.: Southward migration of the intertropical convergence zone
N., Lavado, W., de Oliveira, E., Pombosa, R., and Vauchel, P.: through the Holocene, Science, 293, 1304–1308, 2001.
Spatio – Temporal rainfall variability in the Amazon Basin Coun- Hendy, C. H.: Isotopic geochemistry of speleothems.1, Calculation
tries (Brazil, Peru, Bolivia, Colombia and Ecuador), Int. J. Cli- of effects of different modes of formation on isotopic composi-
matol., 29, 1574–1594, 2009. tion of speleothems and their applicability as palaeoclimatic in-
Espinoza, J. C., Ronchail, J., Guyot, J. L., Junquas, C., Vauchel, P., dicators, Geochim. Cosmochim. Ac., 35, 801–824,1971.
Lavado, W. S., Drapeau, G., and Pombosa, R.: Climate variabil- IPCC 2013: Masson-Delmotte, V., Schulz, M., Abe-Ouchi, A.,
ity and extremes drought in the upper Solimões River (Western Beer, J., Ganopolski, A., González Rouco, J. F., Jansen, E., Lam-
Amazon Basin): Understanding the exceptional 2010 drought, beck, K., Luterbacher, J., Naish, T., Osborn, T., Otto-Bliesner, B.,
Geophys. Res. Lett., 38, L13406, doi:10.1029/2011GL047862, Quinn, T., Ramesh, R., Rojas, M., Shao, X., and Timmermann,
2011. A.: Information from Paleoclimate Archives, in: Climate Change
Espinoza, J. C., Ronchail, J., Guyot, J. L., Junquas, C., Drapeau, 2013: The Physical Science Basis. Contribution of Working
G., Martinez, J. M., Santini W., Vauchel, P., Lavado, W., Or- Group I to the Fifth Assessment Report of the Intergovernmen-
doñez, J., and Espinoza, R.: From drought to flooding: under- tal Panel on Climate Change, edited by: Stocker, T. F., Qin, D.,
standing the abrupt 2010–2011 hydrological annual cycle in the Plattner, G.-K., Tignor, M., Allen, S. K., Boschung, J., Nauels,
Amazonas River and tributaries, Environ. Res. Lett., 7, 024008, A., Xia, Y., Bex, V., and Midgley, P. M.: Cambridge University
doi:10.1088/1748-9326/7/2/024008, 2012. Press, Cambridge, United Kingdom and New York, NY, USA,
Espinoza, J. C., Ronchail, J., Frappart, F., Lavado, W., Santini, 2013.
W., and Guyot, J. L.: The Major Floods in the Amazonas River Jomelli, V., Favier, V., Rabatel, A., Brunstein, D., Hoffmann,
and Tributaries (Western Amazon Basin) during the 1970–2012 G., and Francou, B.: Fluctuations of glaciers in the tropi-
Period: A Focus on the 2012 Flood, J. Hydrometeorol, 14, cal Andes over the last millennium and palaeoclimatic im-
1000–1008, 2013. plications: A review, Palaeogeogr. Palaeocl., 281, 269–282,
Feng, S., Oglesby, R. J., Rowe, C., Loope, D., and Hu, Q.: Atlantic doi:10.1016/j.palaeo.2008.10.033, 2009.
and Pacific SST influences on Medieval drought in North Amer- Kanner, L. C., Burns, S. J., Cheng, H., Edwards, R. L., and Vuille,
ica simulated by the Community Atmospheric Model, J. Geo- M.: High-resolution variability of the South American sum-
phys. Res., 113, D11101, doi:10.1029/2007JD009347, 2008. mer monsoon over the last seven millennia: insights from a
Garreaud, R., Vuille, M., and Clement, A.: The climate of the Al- speleothem record from the central Peruvian Andes, Quaternary
tiplano: Observed current conditions and mechanisms of past Sci. Rev., 75, 1–10, doi:10.1016/j.quascirev.2013.05.008, 2013.
changes, Palaeogeogr. Palaeocl., 194, 5–22, 2003. Keigwin, L.: The Little Ice Age and Medieval Warm Period in the
Sargasso Sea, Science, 274, 1504–1508, 1996.

www.clim-past.net/10/1967/2014/ Clim. Past, 10, 1967–1981, 2014

1980 J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years

Knight, J. R., Folland, C. K., and Scaife, A. A.: Climate impacts of Quaternary Res., 70, 198–212, doi:10.1016/j.yqres.2008.02.012,
the Atlantic Multidecadal Oscillation, Geophys. Res. Lett., 33, 2008.
L17706, doi:10.1029/2006GL026242, 2006. Reuter, J., Stott, L., Khider, D., Sinha, A., Cheng, H., and Edwards,
Lavado, W., Labat, D., Ronchail, J., Espinoza, J. C., and Guyot, J. R. L.: A new perspective on the hydroclimate variability in north-
L.: Trends in rainfall and temperature in the Peruvian Amazon ern South America during the Little Ice Age, Geophys. Res. Lett.,
– Andes basin over the last 40 years (1965–2007), Hydrological 36, L21706, doi:10.1029/2009GL041051, 2009.
Processes, 41, 2944–2957 doi:10.1002/hyp.9418, 2012. Ronchail, J., Cochonneau, G., Molinier, M., Guyot, J. L., Goretti de
Ledru, M.-P., Jomelli, V., Samaniego, P., Vuille, M., Hidalgo, S., Miranda Chaves, A., Guimarães, V., and de Oliveira, E.: Rainfall
Herrera, M., and Ceron, C.: The Medieval Climate Anomaly and Variability in the Amazon Basin and SSTs in the tropical Pacific
the Little Ice Age in the eastern Ecuadorian Andes, Clim. Past, and Atlantic oceans, Int. J. Climatol., 22, 1663–1686, 2002.
9, 307-321, doi:10.5194/cp-9-307-2013, 2013. Rozanski, K., Araguás-Araguás, L., and Gonfiantini, R.: Relation
Lenters, J. D. and Cook, K. H.: On the origin of the Bolivian high between long-term trends of oxygen-18 isotope composition of
and related circulation features of the South American climate, J. precipitation and climate, Science, 258, 981–985, 1992.
Atmos. Sci., 54, 656–678, 1997. Sachs, J. P., Sachse, D., Smittenberg, R. H., Zhang, Z., Battisti,
Mann, M. E.: Little Ice Age, Encyclopedia of Global Environmental D. S., and Golubic, S.: Southward movement of the Pacific in-
Change, ISBN 0-471-97796-9, 1, 504–509, 2002. tertropical convergence zone AD 1400–1850, Nat. Geosci, 2,
Mann, M. E., Zhang, Z., Rutherford, S., Bradley, R. S., Hughes, M. 519–525, doi:10.1038/ngeo554, 2009.
K., Shindell, D., Ammann, C., Faluvegi, G., and Ni, F.: Global Salvatteci, R., Gutiérrez, D., Field, D., Sifeddine, A., Ortlieb, L.,
signatures and dynamical origins of the Little Ice Age and Me- Bouloubassi, I., Boussafir, M., Boucher, H., and Cetin, F.: The
dieval Climate Anomaly, Science, 326, 1256–1260, 2009. response of the Peruvian Upwelling Ecosystem to centennial-
Marengo, J. A. and Nobre, C. A.: General characteristics and vari- scale global change during the last two millennia, Clim. Past,
ability of climate in the Amazon basin and its links to the global 10, 715–731, doi:10.5194/cp-10-715-2014, 2014.
climate system. The Biogeochemistry of the Amazon Basin, Ox- Seth, A., Rojas, M., and Rauscher, S. A.: CMIP3 projected changes
ford University Press, Oxford, UK, 17–41, 2001. in the annual cycle of the South American monsoon, Climatic
Marengo, J. A., Nobre, C. A., Tomasella, J., Oyama, M. D., De Change, 98, 331–357, 2010.
Oliveira, G. S., De Oliveira, R., Camargo, H., Alves, L. M., and Sifeddine, A., Gutiérrez, D., Ortlieb, L., Boucher, H., Velazco, F.,
Brown, I. F.: The drought of Amazonia in 2005, J. Climate, 21, Field, D., Vargas, G., Boussafir, M., Salvatteci, R., Ferreira, V.,
495–516, 2008. García, M., Valdés, J., Caquineau, S., Mandeng Yogo, M., Cetin,
Marengo, J. A., Liebmann, B., Grimm, A. M., Misra, V., Silva Dias, F., Solis, J., Soler, P., and Baumgartner, T.: Laminated sediments
P. L., Cavalcanti, I. F. A., Carvalho, L. M. V., Berbery, E. H., from the central Peruvian continental slope: A 500 year record of
Ambrizzi, T., Vera, C. S., Saulo, A. C., Nogues-Paegle, J., Zipser, upwelling system productivity, terrestrial runoff and redox con-
E., Seth, A., and Alves, L. M.: Recent developments on the South ditions, Prog. Oceanogr., 79, 190–197, 2008.
American monsoon system, Int. J. Climatol., 32, 1–21, 2012. Silva, G. A. M., Ambrizzi, T., and Marengo, J. A.: Observational
Medina-Elizalde, M., Burns, S. J., Lea, D. W., Asmerom, Y., Von evidences on the modulation of the South American Low Level
Gunten, L., Polyak, V., Vuille, M., and Karmalkar, A.: High res- Jet east of the Andes according the ENSO variability, Ann. Geo-
olution stalagmite climate record from the Yucatán Peninsula phys., 27, 645–657, doi:10.5194/angeo-27-645-2009, 2009.
spanning the Maya terminal classic period, Earth Planet. Sc. Strikis, N. M., Cruz Jr., F. W., Cheng, H., Karmann, I., Edwards,
Lett., 298, 255–262, 2010. R. L., Vuille, M., Wang, X., de Paula, M. S., Novello, V. F.,
Nogués-Paegle, J. N. and Mo, K. C.: Linkages between summer and Auler, A. S.: Abrupt variations in South American monsoon
rainfall variability over South America and sea surface tempera- rainfall during the Holocene based on a speleothem record from
ture anomalies, J. Climate, 15, 1389–1407, 2002. central-eastern Brazil, Geology, 39, 1075–1078, 2011.
Novello, V. F., Cruz, F. W., Karmann, I., Burns, S. J., Stríkis, Taylor, B. L.: A speleothems-based high resolution reconstruction
N. M., Vuille, M., Cheng, H., Edwards, L. R., Santos, V. R., of climate in southeastern Brazil over the past 4100 years, M. S.
Frigo, E., and Barreto, E. A. S.: Multidecadal climate variabil- thesis, University of Massachusetts, Massachusetts, USA, 2010.
ity in Brazil’s Nordeste during the last 3000 years based on Thompson, L. G., Mosley-Thompson, E., Bolzan, J. F., and Koci,
speleothem isotope records, Geophys. Res. Lett., 39, L23706, B. R.: A 1500-year record of tropical precipitation in ice cores
doi:10.1029/2012GL053936, 2012. from the Quelccaya ice cap, Peru, Science, 229, 971–973, 1985.
Oliveira, S. M., Marques Gouveia, S. S. E., Pessenda, L., Favaro, C. Thompson, L. G., Mosley-Thompson, E., Dansgaard, W., and
R., and Teixeira, D. I.: Lacustrine sediments provide geochemi- Grootes, P. M.: The Little Ice Age as recorded in the stratigra-
cal evidence of environmental change during the last millennium phy of the tropical Quelccaya Ice Cap, Science, 234, 361–364,
in southeastern Brazil, Chemie der Erde, 69, 395–405, 2009. 1986.
Parsons, L. A., Yin, J., Overpeck, J. T., Stouffer, R. J., and Thompson, L. G., Mosley-Thompson, E., Davis, M. E., Zagorod-
Malyshev, S.: Influence of the Atlantic Meridional Overturn- nov, V. S., Howat, I. M., Mikhalenko, V. N., and Lin, P.-
ing Circulation on the monsoon rainfall and carbon balance N.: Annually resolved ice core records of tropical climate
of the American tropics, Geophys. Res. Lett., 41, 146–151, variability over the past ∼ 1800 years, Science, 340, 945–50,
doi:10.1002/2013GL058454, 2014. doi:10.1126/science.1234210, 2013.
Rabatel, A., Francou, B., Jomelli, V., Naveau, P., and Grancher, Torrence, C. and Compo, G. P.: A practical guide to wavelet analy-
D.: A chronology of the Little Ice Age in the tropical Andes of sis, B. Am. Meteorol. Soc., 79, 61–78, 1998.
Bolivia (16◦ S) and its implications for climate reconstruction,

Clim. Past, 10, 1967–1981, 2014 www.clim-past.net/10/1967/2014/

J. Apaéstegui et al.: Hydroclimate variability of the northwestern Amazon Basin during the last 1600 years 1981

Van Breukelen, M., Vonhof, H., Hellstrom, J., Wester, W., and Vuille, M., Burns, S. J., Taylor, B. L., Cruz, F. W., Bird, B. W.,
Kroon, D.: Fossil dripwater in stalagmites reveals Holocene tem- Abbott, M. B., Kanner, L. C., Cheng, H., and Novello, V. F.: A
perature and rainfall variation in Amazonia, Earth Planet. Sc. review of the South American monsoon history as recorded in
Lett., 275, 54–60, 2008. stable isotopic proxies over the past two millennia, Clim. Past, 8,
Vimeux, F., Gallaire, R., Bony, S., Hoffmann, G., and Chiang, J. C. 1309–1321, doi:10.5194/cp-8-1309-2012, 2012.
H.: What are the climate controls on δD in precipitation in the Yoon, J.-H. and Zeng, N.: An Atlantic influence on Amazon rain-
Zongo Valley (Bolivia)? Implications for the Illimani ice core fall, Clim. Dynam., 34, 249–264, doi:10.1007/s00382-009-0551-
interpretation, Earth Planet. Sc. Lett., 240, 205–220, 2005. 6, 2010.
Vuille, M. and Werner, M.: Stable isotopes in precipitation record-
ing South American summer monsoon and ENSO variability
– observations and model results, Clim. Dynam., 25, 401–413,
doi:10.1007/s00382-005-0049-9, 2005

www.clim-past.net/10/1967/2014/ Clim. Past, 10, 1967–1981, 2014