interglacial changes of Southern Hemisphere wind circulation from the geochemistry of South American dust

JID:EPSL

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.1 (1-12)

Earth and Planetary Science Letters ••• (••••) •••–•••

Contents lists available at ScienceDirect

Earth and Planetary Science Letters www.elsevier.com/locate/epsl

Glacial/interglacial changes of Southern Hemisphere wind circulation from the geochemistry of South American dust Stefania Gili a , Diego M. Gaiero a,∗ , Steven L. Goldstein b,c , Farid Chemale Jr. d , Jason Jweda b,c , Michael R. Kaplan b , Raúl A. Becchio e , Edinei Koester f a

CICTERRA-CONICET/FCEFyN, Universidad Nacional de Córdoba, Argentina Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY, USA c Department of Earth and Environmental Sciences, Columbia University, Palisades, NY, USA d Universidade do Vale do Rio dos Sinos, São Leopoldo, Brazil e Universidad Nacional de Salta, Salta, Argentina f Instituto de Geociências, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil b

a r t i c l e

i n f o

Article history: Received 14 December 2016 Received in revised form 23 March 2017 Accepted 2 April 2017 Available online xxxx Editor: M. Frank Keywords: Southern westerly winds dust REE radiogenic isotopes South America Antarctica

a b s t r a c t The latitudinal displacement of the southern westerlies and associated climate systems is a key parameter for understanding the variations of Southern Hemisphere atmospheric circulation during the Late Quaternary Period. To increase understanding of past atmospheric circulation and of the paleoenvironmental conditions associated with continental dust sources, we dig deeper into dust provenance in paleo-archives of the Southern Hemisphere. We present here a Sr–Nd isotopic and rare earth element study of surface sediments collected along a ∼4000 km latitudinal band from arid and semi-arid terrains in southern South America. Findings from terrains that served as paleo-dust suppliers are compared with modern dust collected from monitoring stations along the same latitudinal band, which affords a test on how actual present-day aeolian compositions compare to those of the past potential source areas. Moreover, the comparison between past and present-day datasets is useful for understanding presentday atmospheric circulation. Armed with a new comprehensive dataset, we revise previous interpretations of the provenance of dust trapped in the Antarctic ice and sediments deposited in the South Atlantic sector of the Southern Ocean. These comparisons support multiple source regions in southern South America that changed with climates. The findings reveal that, although Patagonia plays an important role in contributing dust to the higher latitudes, central Western Argentina and (to a lesser extent) the southern Puna region also emerge as potentially important dust sources during glacial times. The southern Altiplano appears to be a major contributor during interglacial periods as well. We rely in part on an understanding of modern wind–dust activities to conclude that the possible presence of southern South America source regions – other than Patagonia – in East Antarctic ice is consistent with an overall equatorward displacement during glacial times of both the mid-latitude westerlies and the subtropical jet stream. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Southern South America (SSA) is of particular interest for paleoclimate studies because it is the only land-mass intersecting the zonal circulation of both the southern westerly winds (SWW, ∼50◦ S), and the high altitude subtropical jet stream (STJ, ∼30◦ S). The reconstruction of the position and strength of the SWW has particular importance given its interaction with the Southern

*

Corresponding author. E-mail address: [email protected] (D.M. Gaiero).

http://dx.doi.org/10.1016/j.epsl.2017.04.007 0012-821X/© 2017 Elsevier B.V. All rights reserved.

Ocean is a major driver of regional and global climate (e.g. Toggweiler et al., 2006; Anderson et al., 2014). Also important are changes in location, intensity, or altitude of the STJ, which can promote variations in the frequency and intensity of storms (Archer and Caldeira, 2008), and thus can modulate the wetter–dry cycles of specific regions such as the Puna–Altiplano region. During glacial/interglacial cycles, these wind systems change their strength and latitudinal positions, affecting their capacity for erosion and transport of mineral dust from their sources to depositional areas in downwind marine and terrestrial environments, where dust is archived.

JID:EPSL

AID:14402 /SCO

2

[m5G; v1.213; Prn:24/04/2017; 15:51] P.2 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

The chemical/isotopic/mineralogical characterization of mineral dust in these paleo-climatic archives (e.g. the Southern Ocean, Antarctica) enables us to learn about its provenance. Such knowledge sheds light on strengthened wind erosion in the source areas and/or the weakening of the hydrological cycle in specific regions (e.g. Maher et al., 2010). Past studies of the provenance of dust deposited in the Southern Ocean (e.g. Walter et al., 2000; Lamy et al., 2014) and Antarctica (e.g. Vallelonga et al., 2010; Delmonte et al., 2010) have led to an increased understanding of past atmospheric circulation in the Southern Hemisphere. Revisiting the literature on this topic, it became apparent that an important limitation, hampering increased understanding of the specific areas contributing past and present dust to the Southern Hemisphere atmosphere, is the lack of systematic studies at different potential source areas (PSAs) of the continents. Having this in mind, we carried out the first comprehensive north-to-south study of the rare earth element (REE) and Sr–Nd isotopic geochemistry of surface sediments and dusts from the Puna–Altiplano (PAP) to Patagonia, where we know the geomorphic context and age of the potential dust source deposits. Selected topsoil samples were collected from desert lands along a 4000 km long latitudinal band. This transect is more or less perpendicular to the main zonal atmospheric circulation (e.g. SWW and STJ), which ensures that we captured possible dust sources and transport to the easternmost settings such as the Pampas, the Southern Ocean (SO), and Antarctica. Furthermore, a series of dust monitoring sites located downwind from these desert areas were set up in order to capture the modern geochemical fingerprints of SSA potential source regions. 2. Settings of potential dust source areas The rain shadow effects caused by the last Andean uplift (26–28 Ma) created the ‘South American arid diagonal’ (Blisniuk et al., 2005). This is a long and narrow region that extends from ∼2◦ S in the Gulf of Guayaquil to ∼53◦ S on the northern Tierra del Fuego island, following the coast in Ecuador to northern Chile, crossing into Argentina north and east of Santiago, and continuing southward through Patagonia. The region has an average annual precipitation of roughly 250 mm (Blisniuk et al., 2005). In central South America (∼15 to 30◦ S), along the west coast, the arid diagonal is dominated by desert areas, whereas towards the east it is characterized by the seasonally dry Chaco forest, and subtropical grasslands where the moisture mainly comes from the Atlantic. The major present-day dust source areas in SSA are located in a continuous N–S band of arid and semi-arid terrains coinciding with the arid diagonal (Prospero et al., 2002). We singled out three primary persistent source areas: Patagonia, central-western Argentina (CWA) and the Puna/Altiplano Plateau (PAP) (Fig. 1). 2.1. The Puna–Altiplano Plateau (PAP) The main features of the PAP region are summarized in Gaiero et al. (2013 and references therein). Briefly, the PAP is a high elevation basin (∼4000 m a.s.l.) located in the central portion of the Andes and is over 1000 km long and ∼200 km wide. The area consists of extensive, internally drained depocenters flanked by N–S oriented mountain ranges, often between 5000 and 6000 m elevation (Strecker et al., 2007). The region consists of large areas covered by salt lake-beds, including the Salar of Uyuni (∼10,000 km2 ) in Bolivia, as well as some smaller ones, for example, the Salinas Grandes (∼200 km2 ), Salar del Hombre Muerto (500 km2 ), Salar de Arizaro (1600 km2 ), Salar de Antofalla (500 km2 ) in Argentina. The South American summer monsoon (SASM) promotes intense convective storms supplying about 80% of precipitation in the austral summer (November to March), while the STJ (westerly winds prevailing in middle and upper troposphere) causes

extreme dry conditions from May to October (Garreaud et al., 2009). In addition, a subtropical high-pressure region (the Bolivian High) and atmospheric subsidence dominate the area, fostering extremely arid conditions between about 15◦ S and 27◦ S, comparable to the deserts at the same latitudes in western Africa and Australia (Strecker et al., 2007). During winter, the interannual seasonal change in the tropospheric temperature gradient between low and mid-latitudes supports a stronger STJ. This prevents regional moisture from reaching the eastern flank of the Andes, promoting a dry winter climate over the PAP (Prohaska, 1976). Also during this season, associated winds with gusts over 100 ms−1 have been recorded (Milana, 2009) leading to the development of extensive dust storms (Gaiero et al., 2013). 2.2. The central-western Argentina (CWA) Central-western Argentina is located between ∼27◦ and ∼39◦ S, and extends from the eastern flank of the Andes to the western slope of the Sierras Pampeanas (Fig. 1). Covering ∼600,000 km2 , it contains varied and complex geological settings. Temperate climatic conditions characterize this part of South America. In the region, wind conditions are controlled by the subtropical high pressure cells (Pacific and Atlantic anticyclones), the intensity of the quasi-stationary low in the Gran Chaco, and the mid-latitudes westerlies. This sector consists mainly of long, wide longitudinal valleys and short, narrow valleys that cut Andean and extraAndean geomorphologic features. The area is also crossed by several rivers and temporary streams that reach the Andean foothills to form extensive deposits of sand, marshes and saline lakes. The whole area is drained by the Bermejo–Desaguadero–Curacó hydrographic system that covers an extension of ∼250,000 km2 . At present, due to the dominant desert climate this hydrographic network is poorly integrated and almost inactive (Iriondo and Krohling, 1995). The atmospheric circulation is dominated by northeasterly surface winds in summer and northwesterly winds in winter (Prohaska, 1976). From May to August (austral autumn–winter) katabatic winds (locally called Zonda) with a dominant west to east component are observed (Norte et al., 2008). These are hot and dry winds that occur over most central parts of western Argentina, being more prominent between 28◦ –37◦ S. These zonal winds are likely to occur farther south, in areas where strong westerlies cross the Andes, resulting in vigorous down slope flow, drying the lower troposphere to the east of the mountains (Garreaud et al., 2013), promoting deflation of northern Patagonia. On its way down in elevation the Zonda can break through the boundary layer and inject dust into the middle troposphere, which then is transported aloft by the jet streams (Norte, personal communication). As we make a case in this paper, CWA provides an unappreciated but persistent source of dust between 27◦ –36◦ S and 67◦ –70◦ W and along the eastern flank of the Andes. 2.3. Patagonia The main features of the Patagonian region as a dust source are summarized by Gaiero et al. (2003, 2004, 2007 and references therein). Briefly, Patagonia is a large and diverse region in southern South America that covers an area of over 900,000 km2 , extending from ∼39◦ S down to the southern tip of the continent (∼55◦ S), including Tierra del Fuego in the southernmost part. The topography of the Argentinean side of Patagonia is dominated to the west and south by the Andes, and by dissected plateaus (often largely volcanic in nature) and low plains to the east. Throughout the year the area is situated in the core of the mid-latitudes and the climate is controlled by the dynamics of the strong SWW. This wind system blows from the Pacific Ocean, and given the presence of the Andes, most of its moisture is discharged on the west

JID:EPSL

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.3 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

3

Fig. 1. Potential source areas (PSAs) of dust in southern South America (SSA). The black contours represent days per month (numbers in italics) with dust activity (Prospero et al., 2002). Sample numbers are in bold font close to the symbols, and data are in Table S1. Triangles are the modern dust monitoring stations. The white dotted line shows the southern and eastern limits of the Puna–Altiplano Plateau. N-CWA, M-CWA and S-CWA refers to Northern, Middle and Southern Central Argentina respectively.

side, and then it continues as dry winds to the east. Hence, the regional climate is strongly affected by the SWW, and the Andes produces a strong, east–west gradient with annual precipitation of ∼4000–7000 mm in the west to ∼200 mm per year in the east; in the latter rainfall tends to occur in the austral autumn and winter seasons (Garreaud et al., 2009).

3. Sampling and methodology 3.1. Surface sediments collection The locations and top soil samples presented in this work are intended to represent the most important potential dust sources in

JID:EPSL

AID:14402 /SCO

4

[m5G; v1.213; Prn:24/04/2017; 15:51] P.4 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

SSA. The location of samples and site characteristics are in Fig. 1 and Table S1. In this study, we extended the original data set of 17 top soil samples in Gaiero et al. (2004, 2007, 2013) to 43 samples, covering the distance between the southern Altiplano (∼21◦ S) and Tierra del Fuego (∼53◦ S). Using plastic scoops and polythene sample bags, about 500 g of the upper 5 cm of loose sediments were collected from the surface of arid and semi-arid terrain. The sampling procedure was carried out following the methodology in Gaiero et al. (2003, 2013). Samples were taken from inter-montane closed basins with alluvial fans, and from the desiccated edge of lakes (Table S1). The choice of sampling sites was also supported by satellite imagery showing dust activity (see Figs. S1–S6 in Appendix A). Chunks of salt containing dust were collected at different points inside the Salar de Uyuni and Salinas Grandes (details are in Gaiero et al., 2013). Our study assumes the sediments collected at these salar pans as representative of the chemical/isotopic composition of aeolian dust lifted from these source areas. About 100–200 g of this material was dissolved in 1 L of MilliQ® water and then filtered through 0.45 μm Millipore® filters. As winds mix and sort dust grains during transport, it is important to understand the grain size dependence of REE and isotopic tracers, as it could impact interpretations of geochemical constraints on the origin of the dust in paleo-records (e.g. Basile et al., 1997). For example, it has been observed that during long range transport, atmospheric dust particles of up to ∼70 μm can be carried over long distances (>2000 km) from the source (e.g. Gaiero et al., 2013). On the other hand, it was also observed that the <5 μm grain-size fraction of sediments is usually found in the most remote paleo-climatic archives such as the ones studied on the Antarctic Plateau or in the Southern Ocean (e.g. Walter et al., 2000; Delmonte et al., 2004). To evaluate grain size effects, we focused on two different size fractions, a finer (<5 μm) and a coarser (<63 μm), for chemical and isotopic analyses. Accordingly, samples were sieved with a 63 μm plastic mesh. In addition, some of the samples were separated further by settling in order to obtain a finer fraction (<5 μm), using Stoke’s Law. The particle size distributions were confirmed with a Horiba LA950 particle size analyzer, which confirmed that over 90% of the total mass was finer than 5 μm. 3.2. Dust collection In order to gain the modern perspective on dust provenance, dust samples (n = 57) were collected from six monitoring stations (Fig. 1 and Table S2) between 2004 and 2010. Information about the stations and their setting has been discussed previously (Gaiero et al., 2013; Simonella et al., 2015). Briefly, the monitoring sites are equipped with pyramidal receptacles suited to measure vertical fluxes and to obtain enough material for subsequent analyses (e.g., Skonieczny et al., 2011; Gaiero et al., 2013). The stations ran automatically for about 14–30 days, after which an operator collected the samples using a vacuum pump to filter (through dry or wet medium) and retain particles on a pre-weighed 0.45 μm membrane. In some cases dust samples were taken immediately after ground or satellite observations indicated that a dust plume was impacting or travelling toward a monitoring site (e.g. Gaiero et al., 2013). To avoid collecting local saltation material, the traps were set up on poles at heights ∼5 m above the ground. They were located within the campuses of different national institutions of Argentina far away from urban settlements. The northernmost station La Calderilla (10 km north from Salta city) is located at the foot of the eastern slope of the Andes and is suited for monitoring dust activity originating from the PAP sector. The Marcos Juárez station is located in a dust cross-corridor and depositional area (the central

Pampas) and is suited for monitoring dust activity on this region and the frequent dust storm outbreaks originating to the west from the arid diagonal. The Bahía Blanca station is suited for monitoring dust activity on the northern edge of Patagonia and also to record the intermittent outbreaks of Zonda winds and dust activity of the southern Pampas. Further south, three stations were placed along the Atlantic coast (e.g. Trelew, San Julián and Río Grande) with the aim of monitoring dust activity in Patagonia. Grain-size analyses were performed in random dust samples from each monitoring station (Table S3). 3.3. Chemical and isotope analyses REE and radiogenic Sr and Nd isotopes were analyzed on the surface sediments and dust samples (Table S1 and Table S2 respectively). Samples were treated with 30% H2 O2 (analytical grade) to eliminate the presence of organic matter, and dilute HCl to eliminate carbonates. The REE were analyzed at a commercial laboratory (Actlabs, Canada). The Sr and Nd isotopic compositions were analyzed at the Universidade Federal do Río Grande do Sul, Brazil (LGI-UFRGS) and at Lamont–Doherty Earth Observatory (LDEO), Columbia University, USA. At both labs, Sr and Nd ratios were normalized to 87 Sr/86 Sr = 0.1194 and 146 Nd/144 Nd = 0.7219, respectively. For convenience, Nd isotope data are also reported here in units of (ε Nd(0) = ([143 Nd/144 Nd (meas.)/143 Nd/144 Nd (CHUR)] − 1] ∗ 104 ), where the CHUR (Chondritic Uniform Reservoir) value is 0.512638 (Jacobsen and Wasserburg, 1980). Additional details of the analytical methodology are provided in the supplementary material (Appendix A). 4. Results and discussion 4.1. Rock sources controlling the composition of SSA surface sediments We evaluate the regional and temporal representativeness of the selected top soil samples by comparing and contrasting their chemical and isotopic compositions with the main bedrock units or outcrops near each PSA. Further discussion of this data is provided in the Supplementary material. In summary, the data presented for each area indicates that the samples we focused on are similar to the composition of the locally outcropping rocks and hence they can be used with confidence in dust provenance studies. 4.2. Grain-size control on REE composition and Sr–Nd isotopic ratios Differences in the REE and Sr–Nd isotopic compositions of fine and coarse fractions of sediments have important implications for provenance tracing studies that use such information, as only certain particle sizes can be transported regionally versus long distance or even globally. Hence, for provenance studies it is important to analyze the right size fraction in the source region for accurate comparisons with dust studied at different paleo-archives downwind, such as ice cores. In this context, the conclusions of previous studies on the grainsize control of REE contents of sediments have been controversial. Studies of Asian surface sands have shown that REE concentrations vary greatly with the size fraction (e.g. Honda et al., 2004 and references therein). However, dust collected in the Chinese deserts and loess-paleosol sequences in the Chinese loess plateau have shown similar REE concentrations independent of the grain-size fraction of sediments (e.g. Kanayama et al., 2005). For SSA sediments, data indicate that the mean REE concentrations of the fine fractions (<5 μm) are significantly lower than the coarse fraction (Figs. 2a, b).

JID:EPSL

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.5 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

5

In our surface sediments, Zr concentrations are systematically lower in the fine fractions compared to the coarse fractions (Table S1). As zircons have high HREE abundances, (e.g. Taylor and McLennan, 1985) its presence or absence impacts the REE concentrations and patterns of sediments. Our surface samples, with the exception of fine sediments from Patagonia, are characterized by enrichments of the middle REE over LREE and HREE (Fig. 2b) similar to patterns observed in most of fine sediments of rivers around the Earth (Bayon et al., 2015). In general, both the coarse and the fine fractions of SSA sediments show positive Eu/∗ Eu (>1) values (Fig. 2c). Most of the samples show lower mean Eu/∗ Eu values for the finer fraction, but La/Yb do not show systematic variations according to grain-size. Interestingly, the fine fractions of the northern and middle CWA samples (hereafter N-CWA and M-CWA) have REE patterns similar to the fine material of Puna (Fig. 2b). This suggests that N-CWA and M-CWA may receive fine aeolian sediments from the highland areas located to its southwest (section 4.3 and Fig. S1), in addition to contributions from the Sierras Pampeanas. The grain size dependence of Sr and Nd isotopic compositions in sediments has been discussed by numerous studies (e.g. Goldstein et al., 1984; Smith et al., 2003; Delmonte et al., 2004; Gaiero, 2007). The new data set for central and southern South America (Fig. 2d) conforms to the longstanding observation that there are statistically significant differences between 87 Sr/86 Sr ratios, but much less so for ε Nd(0) values, especially when comparing the <5 and <63 μm grain-size fractions of sediments (Fig. 2d). Patagonia and S-CWA samples show the smallest differences between Sr isotopic compositions of the fine and the coarse fractions (87 Sr/86 Sr ∼0.0013 and 0.0012 units respectively), consistent with previous results (Gaiero, 2007). Delmonte et al. (2010) indicated that dust recovered from the Illimani ice core (Bolivia) has larger differences between the Sr isotopic ratios of coarse and fine dust (87 Sr/86 Sr ∼0.0025 units). Similarly, our data for PAP sediments indicate differences between grain-size fractions of ∼0.0022 units. The surface sediments from the M-CWA have the highest Sr isotopic differences (87 Sr/86 Sr ∼ 0.0099 units). For all of SSA regions, the mean difference between ε Nd(0) values for the coarse and fine fractions of sediments is <1.5ε units and is not significant. In general, our results are in agreement with the observed small differences of ∼1–2ε units between clay and silt for grain-size fractions in river sediments (e.g. Bayon et al., 2015). A key finding is that any differences in REE and isotopic data between the fine and the coarse grain-size of sediments are small compared to the overall variability between different regions, and thus the chemical and isotopic signatures for the different PSAs of SSA can be used to fingerprint provenance. 4.3. The fingerprint of surface sediments from the PSAs of SSA

Fig. 2. Effect of grain-size fractionation on the chemical (REE, linear concentration scale) and isotopic (86 Sr/87 Sr and ε Nd(0)) compositions of surface sediments representing the PSAs of SSA. Acronyms are defined in Table S1 in the supplementary material. Eu/∗ Eu and La/YbN here are based on normalizing to upper continental crust (UCC, Taylor and McLennan, 1985).

There are clear differences between the different areas representing the PSAs of SSA in terms of both isotopic and REE compositions (Fig. 3) (Table S2). In general, the data indicate that REE are less useful for distinguishing sediments from CWA and Puna (Fig. 3a), while the isotopic data are better for distinguishing the sediments from the three sub-regions within the PAP region (Fig. 3b). An important observation is that more “mafic-like” compositions of Patagonian sediments (reflected by lower LREE/HREE, higher 143 Nd/144 Nd, lower 87 Sr/86 Sr) contrast with more “crustallike” compositions of material from the PAP area. There is a slight overlap between the compositions of sediments from Patagonia and the S-CWA, and between the isotopic data from S-CWA and the southern Puna. The REE signature of surface sediments from Tierra del Fuego is not distinguishable from similar data representing continental Patagonia, while some samples have lower Nd isotope ratios (Figs. S3d and S3b, supplementary material).

JID:EPSL

6

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.6 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

pas is blocked by the presence of the Sierras Pampeanas (Zárate and Tripaldi, 2012) (Fig. 1). As a consequence, when Zonda winds reach the foothills of the Andes, its W to E component is deflected to the SSE along with the general NNW–SSE orientation of the Sierras Pampeanas (Fig. S5). As Zonda loses energy, atmospheric dust is deposited towards the southeast, and eventually it can reach the latitude of the S-CWA area. In other cases, intense southern winds associated with polar fronts transport sediments northward and eventually dust is deposited in the northern CWA sectors, as for example, in the foothill of the Andes (Fig. S6). Based on these observations, we will not discuss the possibility of contributions of aeolian materials from N-CWA and M-CWA to the eastern environments. 4.4. SSA atmospheric circulation from the geochemical fingerprints of modern dust

Fig. 3. Surface sediment geochemical data: (a) Relationships between Eu/∗ Eu and La/YbN , and (b) 86 Sr/87 Sr vs ε Nd(0) of surface sediments representing PSAs of SSA. CWA refers to Central West Argentina. SS14 and SS18 represent samples from Middle CWA (Table S1 in supplementary material). Eu/∗ Eu and La/YbN here is based on normalizing to upper continental crust (UCC, Taylor and McLennan, 1985).

Due to the similarities in the chemical and Sr–Nd isotopic compositions of sediments from the N-CWA and M-CWA, we grouped them as a single compositional area (Figs. 3a, b). These data fall in a compositional sector encircled by the composition of PAP materials; in addition a group of these samples have isotopic composition similar to sediments from the southern Puna and southern Altiplano. We observed similarities between REE compositions of the fine sediments of Puna and those from N-CWA and M-CWA (Text S2). Atmospheric transport of dust from the southern Puna to the northern and middle sectors of CWA is often observed on satellite images (Fig. S4). As noted in Text S2, the compositions of samples 14 and 18 are ambiguous because their geographic positions and the atmospheric circulation over the Altiplano prevent significant contributions of aeolian sediments from the latter to the M-CWA sector (Fig. 3b). Atmospheric circulation is not consistent with the M-CWA and N-CWA areas being a significant supplier of dust to the eastern environments (e.g. the central Pampas). As discussed in Section 2.2, the main atmospheric circulation over the CWA is from north to south and no evidence exists in the present day for dust storms associated with this wind system. Indeed, dust storms in the region are frequently associated with Zonda winds, and to a lesser extent, to the outbreaks of polar fronts. As already noted, Zonda events have a mainly west to east component and frequently impact the latitudes between 28◦ S to 37◦ S in the S-CWA (Lassig et al., 1999; Norte, 2015). At ∼30◦ S, the direct transport of dust to the Pam-

In general, the mean REE patterns and the isotopic signatures of SSA dust obtained from the different monitoring stations can be distinguished from each other (Fig. 4) (Table S2). The new data from modern dust traps in Patagonia also confirm the general uniformity of the chemical and isotopic signature of dust emitted from this region, supporting the findings from the composition of surface sediments discussed above. From Trelew to Río Grande (Tierra del Fuego), dusts typically have more mafic-like compositions, with light REE depletions and enrichments in heavy REE relative to average upper crust. The mean composition of dust collected in Marcos Juárez (central Pampas) has high REE concentrations. Further north, dust samples from La Calderilla (Salta) show relatively flatter REE patterns (Fig. 4a). The Sr–Nd isotopic compositions of dusts from continental Patagonia are slightly different from dust collected in Tierra del Fuego. As indicated for surface sediments, dusts from Tierra del Fuego have higher 87 Sr/86 Sr ratios (0.707–0.708) and more negative ε Nd(0) values compared with dust from Patagonia. Dust from continental Patagonia has variable ε Nd(0) values (4.7 to −3.0) and fairly homogeneous 87 Sr/86 Sr ratios of ∼0.706 (Fig. 4c). At the northern edge of Patagonia (Bahía Blanca), dusts exhibit variable compositions (for example, ranging from 87 Sr/86 Sr ∼ 0.705 to 0.709 and ε Nd(0) ∼ 1.0 to −4; Fig. 4c). In general, a Patagonian-like composition (87 Sr/86 Sr ∼ 0.706 to 0.707 and ε Nd(0) ∼ −2.4 to −1.4) prevails during summer time when westerly winds are intense at ∼50◦ S (Garreaud et al., 2009). Within this group of samples, the more positive ε Nd-values are probably derived from a nearby dusty area located SE from this monitoring site (Johnson et al., 2010) (Fig. S7). This area is frequently flooded by the Colorado and Negro Rivers, and the deflated sediments have the composition of the material transported by these rivers (Gaiero et al., 2007). A second group of dust samples has more crustal-like compositions (i.e., lower Eu/∗ Eu, higher La/Yb ratios, 87 Sr/86 Sr ∼ 0.709 and ε Nd(0) ∼ −3.8), and shows similarities with the chemical and isotopic compositions of surface sediments from the S-CWA area (Figs. 5a, b). The transport mechanism of dust from S-CWA to Bahía Blanca can be associated with the Zonda winds, which frequently impact northwest Patagonia at ∼39◦ S (Lassig et al., 1999). This agrees with the collection of these samples during winter time (JAS), when Zonda events are more frequent (Norte, 2015). Furthermore, at these latitudes the zonal circulation of Zonda is not interrupted by mountain or topographic high areas, and dust transport to Bahía Blanca can occur (Ramsperger et al., 1998). The most crustal-like signature found among the dust samples is observed in those collected at La Calderilla (Fig. 4). The REE compositions of dusts from this site show near uniform Eu/∗ Eu, close to ∼1.0, but variable La/Yb ratios of 0.65 to 1.0. Similarly, the isotopic compositions show variable 87 Sr/86 Sr ra-

JID:EPSL

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.7 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

7

Fig. 5. Geochemical compositions of surface sediments, and dust from monitoring stations (Fig. 4b). Also shown are compositions of two dust events recorded at Buenos Aires. (a) Eu/∗ Eu vs La/YbN and (b) 86 Sr/87 Sr vs ε Nd(0) isotopic ratios. The solid black lines indicate the isotopic mixing trend lines between S-CWA, Southern Altiplano and Northern Puna (Table S4). Sample ADPdeB is from Delmonte et al. (2004). The acronyms next to the symbols are defined in Table S1 in supplementary material. Eu/∗ Eu and La/YbN here are based on normalizing to upper continental crust (UCC, Taylor and McLennan, 1985).

Fig. 4. Geochemical signature of modern dust from southern South America (a) Mean REE patterns. (b) Relationship between Eu/∗ Eu and La/YbN ratios and, (c) 86 Sr/87 Sr vs ε Nd(0) isotopic ratios. Puerto Madryn, Comodoro Rivadavia and some of the Bahía Blanca dust composition were taken from Gaiero et al. (2007). The acronyms next to the symbols are defined in Table S1 in Supplementary material. Isotopic data for dust event ADPdeB is from Delmonte et al. (2004). Eu/∗Eu and La/YbN here are based on normalizing to upper continental crust (UCC, Taylor and McLennan, 1985).

tios (∼0.711–0.718) but near uniform ε Nd(0) values (−10 to −12) (Fig. 4b). The isotopic results indicate a mix of northern Puna and southern Altiplano as the dominant signature of dusts deposited in this sector (Figs. 5a, b). Interestingly, the isotope ratios of samples collected during dust storms occurred on July 2009 (sample ADLC2) and July 2010 (sample ADLC6), show a dominant southern Altiplano signature (Fig. 5b) (Gaiero et al., 2013).

Dust collected in the central Pampas (Marcos Juárez) show the most variable chemical and Sr–Nd isotopic compositions, with intermediate values between dust from Bahía Blanca and La Calderilla (Figs. 4b, c). It is worth noting that in this area, as in most of the Pampean region, the agriculture activity is intense and perturbation of the local soils could bias the signatures of the collected dusts. Nevertheless, ‘no-till farming’ has been adopted since the early 1970’s and this has led to a clear reduction of soil erosion (Peiretti and Dumanski, 2014). As no-till farming is done only when soil humidity is favorable, and considering that most of the farmers employ direct seeding systems, soil disturbance is reduced almost year-round. Accordingly, data indicate that only two of the twelve dust samples analyzed from this site might show the signature of the local soils (e.g. sample LLS in Table S1). This agrees with ground and satellite observations showing that the dust deposited at Marcos Juárez is a combination of material transported from mid- to distant sources (Gaiero et al., 2013). The S-CWA-like signature found in dust from this site can be linked to the arrival of strong surface winds associated with polar front systems, which deflate soils of the southern Pampas and/or the northern Patagonia (Fig. 5 and Figs. S7, S8). Nonetheless, a significant number of samples from this site have intermediate isotopic compositions aligned on a compositional trend defined by the S-CWA and southern Alti-

JID:EPSL

8

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.8 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

plano end-members (Fig. 6b). Although most of dust from Marcos Juárez are well constrained by this mixing trend, the role played by the southern Puna cannot be ruled out. For example, one dust sample collected during August 2004 lies on a different mixing trend (Fig. 5b, sample ADMJ8) and is consistent with the occurrence of dust storms in the southern Puna during the same period (Fig. S4). Similar to the data from La Calderilla, the two samples collected during the PAP dust storms that occurred on July 2009 and 2010 show a dominant southern Altiplano signature (Fig. 5b). Notably, for some sampling periods, no dust storms were observed or reported in the southern Altiplano, but still some dust samples collected at Marcos Juárez station (just to the south east) have the isotopic signature from this region (Table S2). After 15–30 days of sampling, dusts at the monitoring sites represent a mix of materials from the different SSA sources. However, a sample from the 2010 PAP dust storm was collected by local people at Buenos Aires city just a day after the start of the event (Fig. 5b; details in Gaiero et al., 2013). This valuable sample shows isotopic compositions between the northern Puna and southern Altiplano, which is similar to the composition of a dust sample collected at the Buenos Aires province during a dust event recorded in 1997 (Delmonte et al., 2004). This evidence is also useful for supporting data obtained at the Marcos Juárez site and confirms previous assumptions that the central Pampas received a mix of materials deflated from the main dust sources of SSA (e.g. Smith et al., 2003; Gaiero, 2007). 4.5. Past atmospheric circulation and environmental conditions inferred from SSA aeolian materials The new data have important implications for identifying the sources of sediments deposited in the Atlantic sector of the Southern Ocean (ASO) and the East Antarctic Plateau (EAP) (Fig. 6). In general, the isotopic compositions are more diagnostic than the REE for identifying the glacial/interglacial dust provenance in paleo-archives (Figs. 6b, c, d), while at the same time the REE data add critical support. An important finding is that the isotopic compositions of glacial/interglacial sediments from the South Atlantic and its sector of the Southern Ocean and the EAP can be explained by contributions of two main sources; a more maficlike end-member represented by Patagonia and/or S-CWA and a common more crustal-like end-member defined by the southern Altiplano and northern Puna (Figs. 6b, d). During glacial times, data from Antarctica matches well with Patagonia (including Tierra del Fuego), S-CWA and southern Puna PSAs, although there could be a possible contribution of 10–20% from Australia (Revel-Rolland et al., 2006; Delmonte et al., 2010). The new Sr–Nd isotopic data indicate that South America can explain all the sources of EAP dust, without any necessity for Australia as an additional end-member. Fig. 6c shows that some outFig. 6. Comparison of the REE and the Sr–Nd isotopic compositions of materials representing PSAs (e.g. SSA and Australia) and paleo-climatic records from the Southern Hemisphere (East Antarctica, Southern Ocean and Illimani Glacier). (a) Eu/∗ Eu vs La/YbN and (b, c, d) 86 Sr/87 Sr vs ε Nd(0). Numbers and the H (Holocene) next to the symbols refer to “marine isotopic stages” (MIS). The red lines represent the mixing lines between different PSAs of SSA (Table S4). The chemical and isotopic compositions of deep sea core sediments from the Atlantic sector of the Southern Ocean: WSA, western south Atlantic; CSA, central south Atlantic and, ESA, eastern south Atlantic are shown (Walter et al., 2000; Noble et al., 2012) (see Fig. 7). In (c) the Australian PSAs are indicated by an orange dashed line (Gingele and De Deckker, 2005; Revel-Rolland et al., 2006). The bi-directional error bars indicate the isotopic variability from Australian PSAs; Lake Eyre (LE), Murray and Darling Basin. The modern Australian dust collected at the SE coast (De Deckker et al., 2014) could represents the mean isotopic composition of dust transported by westerly winds. In (d) also shown are the isotopic compositions of modern dust from La Calderilla and Marcos Juárez. Data from Berkner Island and Illimani Glacier are from Bory et al. (2010) and Delmonte et al. (2010). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

JID:EPSL

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.9 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

9

Fig. 7. For glacial periods, paleo-data indicate a substantial decrease of the core of the southern westerly winds (SWW, larger red arrows indicate higher intensity) over the southern tip of South America and the strengthening of the northern margin of this wind system around 33◦ –40◦ S (Lamy et al., 2015). During glacial times, increased deflation over the S-CWA was probably powered by an overall strengthening of Zonda winds (curved red arrows crossing the Andes) induced by the more vigorous SWW over this latitudes. Similarly, during colder periods an equatorward movement of the subtropical jet stream (SJT, blue arrows), could increase deflation over the southern and northern Puna (SP and NP) but weaken dust emission from the southern Altiplano (SP), where high lake levels prevailed. The figure also shows the position of sediment cores from the Atlantic sector of the Southern Ocean; WSA, west South Atlantic; CSA, central South Atlantic; ESA, east South Atlantic. Map from http://www.geomapapp.org. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

liers among the Australian PSAs match the Antarctic dust data, and in particular, only samples from the Eyre Lake could explain the composition of a few Antarctic samples. However, Australian PSAs are located in a narrow latitudinal band but extended over wide longitudinal regions. After severe dust storms powered by westerly winds, dust samples collected in SE Australia – which include the main PSAs including the Eyre Lake sediments – showed a near uniform Nd isotope ratios (−6.89 to −7.44) and a narrow range of Sr isotope ratios (0.71612 to 0.7192) (De Deckker et al., 2014), thus contributing with a relatively well-mixed “homogeneous” Australian signature (Fig. 6c). Similar observations were recently reported based on Pb isotopes (Gili et al., 2016) and are fully supported by model simulations (Albani et al., 2012). The good agreement between REE and isotopic data highlight the importance of Patagonia and S-CWA as active dust sources to East Antarctic and the Southern Ocean during the last climatic cycles. Moreover, considering that the geochemical characteristics of some East Antarctic dust samples match well with data from the S-CWA, this suggests that this particular area could have also acted as an important dust supplier during glacial times. The dominance of the S-CWA signature over the Patagonian one is also indicated by the composition of interglacial samples from the Southern Ocean (Fig. 6b) and modern dust from Marcos Juárez (Fig. 6b), which fall within the mixing line between S-CWA and southern Altiplano/northern Puna end-members. Furthermore, several lines of geomorphological evidence and atmospheric simulations support an important role for S-CWA during the late Quaternary. Firstly, the S-CWA aeolian system extended to the SE from its present position, reaching northern Patagonia (Zárate and Tripaldi, 2012). Secondly, paleo-climatic studies indicate that during MIS 2, the S-CWA experienced a major climatic shift to drier and windier conditions with significant aeolian ac-

tivity (Tripaldi et al., 2011). Thirdly, the presence of SW-to-NE and W-to-E oriented aeolian landforms suggest that the loess deposited to the east from S-CWA is the result of deflation of alluvial deposits from the Bermejo–Desaguadero–Curacó fluvial system (Iriondo and Krohling, 1995). Finally, for the LGM atmospheric models show increased dust emission between 37◦ S and 42◦ S, roughly corresponding to the south of S-CWA and northern Patagonia (Albani et al., 2012). In agreement with the isotopic data, this model also indicates a weaker dust source in southernmost Patagonia (e.g., Tierra del Fuego). These observations are also supported by paleodata indicating a substantial decrease of the core of the SWW over the southern tip of South America during colder periods and, at the same time, the strengthening of the northern margin of this wind system around 33◦ –40◦ S (Lamy et al., 2015). The activation of the S-CWA was probably powered by an overall strengthening of Zonda winds induced by more vigorous SWW during glacials, while shifted northward closer to the equator (Fig. 7). A 50% dust contribution from Australia has been suggested to account for the slightly more crustal-like signature observed in some older glacial Antarctic samples (Revel-Rolland et al., 2006) (Fig. 6c). However, the new SSA data presented here suggest that some EAP dust having higher 87 Sr/86 Sr values (MIS, 6, 8, 10 and 12) can be also explained by a southern, and to a lesser extent, northern Puna origin. The relative importance of a glacial-age dust component from Puna would have been due probably to a latitudinal displacement and intensification of the subtropical westerly jet stream (STJ). Currently, the core of STJ is centered at ∼30◦ S, moving poleward during summer and equatorward during winter (Archer and Caldeira, 2008). We infer that the isotopic data reported here indicate an equatorward movement of the STJ during older glacial periods. In the southern Puna (∼25◦ S), intense deflation is evidenced by the presence of large basins of pri-

JID:EPSL

10

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.10 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

mary aeolian origin, likely of Pliocene to early Quaternary age (Goudie and Wells, 1995) (Fig. S2). It is possible that the equatorward movement of the STJ reached even lower latitudes than southern Puna (Fig. 7), however, high-stand lake levels prevailed in more northerly regions during glacial periods (Baker and Fritz, 2015), which would have reduced the mobilization of dust from the southern Altiplano. The same atmospheric mechanism that explains the composition of modern dust deposited in the central Pampas (e.g. Marcos Juárez), provides a key insight that can be used to understand changes in the provenance of modern/interglacial dust (Fig. 6d). East Antarctic interglacial samples can be explained by a main southern Altiplano source, although there is no clear dominance of any of the three sub-areas from PAP. Activation of the PAP source during interglacials is also suggested by data from the Illimani glacier in the Bolivian Andes (Fig. 6d). While during modern/interglacial periods the core of the STJ is centered at 30◦ S, during winter seasons this wind system is more intense over the Altiplano–northern Puna, leading to the development of large dust storms (Gaiero et al., 2013). This phenomenon is intensified during El Niño years, when the meridional thermal gradient between the tropics and subtropics becomes stronger, and the incoming of easterly moisture is reduced, promoting extreme dry conditions over the area (e.g., Garreaud et al., 2009). Modern atmospheric simulations show significant dust transport from subtropical latitudes to lower latitudes like the South Atlantic and EAP (Li et al., 2008; Albani et al., 2012) and this is also inferred from observation of the modern STJ (http://squall.sfsu.edu/scripts/shemjet_archloop. html) (Gaiero, 2007) (Fig. S9). Moreover, PAP sources are consistent with some of the isotopic compositions of dust deposited on Berkner Island, which cannot be ascribed to the isotopic compositions of the most common potential Antarctic source rocks (Gaiero et al., 2007; Bory et al., 2010; Noble et al., 2012). Hence, the East Antarctica isotopic data are consistent with paleo-data indicating drier conditions in the Altiplano during the Holocene (Fornace et al., 2014) and older interglacials (e.g. MIS 5.5) (Baker and Fritz, 2015). The other samples from Berkner Island having more negative Nd values are better explained by local Antarctic dust sources (Bory et al., 2010). Moreover, our interpretation of the EAP dust provenance matches well with the isotopic data from South Atlantic sediments (Fig. 6b, Fig. 7) and agrees with observations (e.g., Anderson et al., 2014) and atmospheric models indicating important modern and glacial dust contributions from SSA to this oceanic basin (Li et al., 2008; Albani et al., 2012; Martínez-García et al., 2014). Glacial/interglacial sediments deposited on the western side of the South Atlantic Ocean (WSA) have a clear Patagonian signature and, as observed above for EAP dust, the isotopic mixing line also singles-out the S-CWA as an important dust supplier. Notably, the Tierra del Fuego PSAs match well with the isotopic composition of glacial samples from the Scotia Sea (WSA) and highlight the importance and value of the new data for provenance studies in the region. Further east, central Atlantic Ocean sediments (CSA) confirm the activation of the southern Altiplano source during interglacials, and the contribution of S-CWA during glacials. For the easternmost ocean sediments (ESA), the interpretation of the isotopic data should be made with caution; the southernmost samples from this sector have clear SSA fingerprints and agree with previous observations (e.g., Anderson et al., 2014). Further north however, the high detrital fluxes recorded in the cores are likely associated with material derived from southern Africa (Noble et al., 2012), with compositions similar to the northernPuna.

5. Final remarks In this paper the “dust provenance” proxy for SSA has been refined in order to better constrain the interpretation of paleo-dust records in the Southern Hemisphere. Combined REE and Sr–Nd isotopes are shown to be powerful tools for discriminating the main dust source areas of SSA. The REE data successfully differentiate Patagonia from S-CWA and PAP signatures, while they are similar for the CWA and Puna regions. Moreover, the isotopes are diagnostic for separating the PSAs of SSA; they can be used to distinguish between the signatures of continental Patagonian sediments from Tierra del Fuego and they clearly distinguish sediments from the PAP region. Isotopic and chemical compositions of dust samples combined with satellite information have proven to be a useful tool for improving knowledge of modern atmospheric circulation over SSA and the interpretation of past dust provenance. In particular, modern dust data single out the Pampas as region where the provenance of dust is sensitive to changes in the main atmospheric circulations over the region (e.g. the surface westerly winds and the subtropical westerly jet stream). By better distinguishing the source signatures in paleo-dust archives of the Southern Hemisphere, we can better understand the major atmospheric zonal circulation patterns during the last glacial–interglacial cycles. The strong northern Patagonia and SCWA signatures in the isotopic data of East Antarctica dust support an equatorward migration of the Southern Hemisphere westerlies during glacials and suggest that central/southern Patagonia and Tierra del Fuego may have had a lesser role than considered until now as a dust supplier to East Antarctica. Specifically, by analogy to present circulation and dust storms, a higher frequency of East Antarctic glacial dust showing signatures from northern Patagonia/S-CWA suggests that the equatorward displacement of the westerlies probably increased the frequency of Zonda winds. Also by comparing East Antarctica and PAP data we infer a possible intensification and an equatorward displacement of the subtropical westerly jet stream during older glacial periods (e.g. MIS 4 and earlier) as well as a major role of the southern Altiplano as a dust supplier during interglacial periods. Acknowledgements This work was financially supported by grants awarded to D.M.G.: Antorchas, IAI, the Weizmann Institute, SECyT/UNC, FONCyT (PICT-0625 and 0525). It was also partly supported by the Storke Endowment of the Department of Earth and Environmental Sciences, Columbia University. We thank to J. Arce (INTA, Marcos Juárez), S. Bidart (UNS, Bahía Blanca), J. Busto and J. González (Base Aeronaval Cte. Zar, Trelew), O. Cabrera (UNPSJB, San Julián) and J. Balderramas (Río Grande) for help with the activities in the monitoring stations. We thank Louise Bolge for her efforts running the LDEO ICP-MS Lab. We are particularly thankful to the three reviewers and the Editor M. Frank, for helping to improve the paleo-data discussions of the manuscript. This is LDEO Contribution # 8108. Appendix A. Supplementary material Supplementary material related to this article can be found online at http://dx.doi.org/10.1016/j.epsl.2017.04.007. References Albani, S., Delmonte, B., Maggi, V., Baroni, C., Petit, J.R., Stenni, B., Mazzola, C., Frezzotti, M., 2012. Interpreting last glacial to Holocene dust changes at Talos Dome (East Antarctica): implications for atmospheric variations from regional to hemispheric scales. Clim. Past 8, 741–750. http://dx.doi.org/10.5194/cp-8-741-2012.

JID:EPSL

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.11 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

Anderson, R.F., Barker, S., Fleisher, M., Gersonde, R., Goldstein, S.L., Kuhn, G., Mortyn, P.G., Pahnke, K., Sachs, J.P., 2014. Biological response to millennial variability of dust and nutrient supply in the Subantarctic South Atlantic Ocean. Philos. Trans. R. Soc. A 372, 20130054. http://dx.doi.org/10.1098/rsta.2013.0054. Archer, C.L., Caldeira, K., 2008. Historical trends in the jet streams. Geophys. Res. Lett. 35, L08803. http://dx.doi.org/10.1029/2008GL033614. Baker, P.A., Fritz, S.C., 2015. Nature and causes of Quaternary climate variation of tropical South America. Quat. Sci. Rev. 124, 31–47. Basile, I., Grousset, F.E., Revel, M., Petit, J.R., Biscaye, P.E., Barkov, N.I., 1997. Patagonian origin of glacial dust deposited in East Antarctica (Vostok and Dome C) during glacial stages 2, 4 and 6. Earth Planet. Sci. Lett. 146, 573–589. Bayon, G., Toucanne, S., Skonieczny, C., André, L., Bermell, S., Cheron, S., Dennielou, B., Etoubleau, J., Freslon, N., Gauchery, T., Germain, Y., Jorry, S.J., Ménot, G., Monin, L., Ponzevera, E., Rouget, M.-L., Tachikawa, K., Barrat, J.A., 2015. Rare earth elements and neodymium isotopes in world river sediments revisited. Geochim. Cosmochim. Acta 170, 17–38. Blisniuk, P.M., Stern, L.A., Chamberlain, C.P., Idleman, B., Zeitler, P.K., 2005. Climatic and ecologic changes during Miocene surface uplift in the southern Patagonian Andes. Earth Planet. Sci. Lett. 230, 125–142. http://dx.doi.org/10.1016/ j.epsl.2004.11.015. Bory, A.E., Wolff, E., Mulvaney, R., Jagoutz, E., Wegner, A., Ruth, U., Elderfield, H., 2010. Multiple sources supply eolian mineral dust to the Atlantic sector of coastal Antarctica: evidence from recent snow layers at the top of Berkner Island ice sheet. Earth Planet. Sci. Lett. 291, 138–148. http://dx.doi.org/10.1016/ j.epsl.2010.01.006. De Deckker, P., Munday, C.I., Brocks, J., O’Loingsigh, T., Allison, G.E., Hope, J., Norman, M., Stuut, J.-B.W., Tapper, N.J., van der Kaars, S., 2014. Characterisation of the major dust storm that traversed over eastern Australia in September 2009: a multidisciplinary approach. Aeolian Res. 15, 133–149. http://dx.doi.org/10.1016/j.aeolia.2014.07.003. Delmonte, B., Basile-Doelsch, I., Petit, J.R., Maggi, V., Revel-Rolland, M., Michard, A., Jagoutz, E., Grousset, F., 2004. Comparing the EPICA and Vostok dust records during the last 220,000 years: stratigraphical correlation and provenance in glacial periods. Earth-Sci. Rev. 66, 63–87. Delmonte, B., Andersson, P.S., Schoberg, H., Hansson, M., Petit, J.R., Delmas, R., Gaiero, D., Maggi, V., Frezzotti, M., 2010. Geographic provenance of Aeolian dust in East Antarctica during Pleistocene glaciations: preliminary results from Talos Dome and comparison with East Antarctic and new Andean ice core data. Quat. Sci. Rev. 29, 256–264. Fornace, K.L., Hughen, K.A., Shanahan, T.M., Fritz, S.C., Baker, P.A., Sylva, S.P., 2014. A 60,000-year record of hydrologic variability in the Central Andes from the hydrogen isotopic composition of leaf waxes in Lake Titicaca sediments. Earth Planet. Sci. Lett. 408, 263–271. Gaiero, D.M., Probst, J.L., Depetris, P.J., Bidart, S.M., Leleyter, L., 2003. Iron and other transition metals in Patagonian riverborne and windborne materials: geochemical control and transport to the southern South Atlantic Ocean. Geochim. Cosmochim. Acta 67 (19), 3603–3623. Gaiero, D.M., Depetris, P.J., Probst, J.L., Bidart, S.M., Leleyter, L., 2004. The signature of river- and windborne materials exported from Patagonia to the southern latitudes: a view from REEs and implications for paleoclimatic interpretations. Earth Planet. Sci. Lett. 219 (3–4), 357–376. Gaiero, D.M., 2007. Dust Provenance in Antarctic Ice during Glacial Periods: from where in Southern South America? Geophys. Res. Lett. 34, L17707. http:// dx.doi.org/10.1029/2007GL030520. Gaiero, D.M., Brunet, F., Probst, J.L., Depetris, P.J., 2007. A uniform isotopic and chemical signature of dust exported from Patagonia: rock sources and occurrence in southern environments. Chem. Geol. 238, 107–120. Gaiero, D.M., Simonella, L., Gassó, S., Gili, S., Stein, A.F., Sosa, P., Becchio, R., Arce, J., Marelli, H., 2013. Ground/satellite observations and atmospheric modeling of dust storms originating in the high Puna–Altiplano deserts (South America): implications for the interpretation of paleo-climatic archives. J. Geophys. Res., Atmos. 118, 1–15. http://dx.doi.org/10.1002/jgrd.50036. Garreaud, R.D., Vuille, M., Compagnucci, R., Marengo, J., 2009. Present-day South American climate. Palaeogeogr. Palaeoclimatol. Palaeoecol. 281, 180–195. Garreaud, R., Lopez, P., Minvielle, M., Rojas, M., 2013. Large scale control on the Patagonia climate. J. Climate 26, 215–230. Gili, S., Gaiero, D.M., Goldstein, S.L., Chemale, F. Jr., Koester, E., Jweda, J., Vallelonga, P., Kaplan, M.R., 2016. Provenance of dust to Antarctica: a lead isotopic perspective. Geophys. Res. Lett. 43. http://dx.doi.org/10.1002/2016GL068244. Gingele, F.X., De Deckker, P., 2005. Clay mineral, geochemical and Sr–Nd isotopic fingerprinting of sediments in the Murray–Darling fluvial system, southeast Australia. Aust. J. Earth Sci. 52, 965–974. Goldstein, S.L., O’Nions, R.K., Hamilton, P.J., 1984. A Sm–Nd isotopic study of atmospheric dust particulates from major river systems. Earth Planet. Sci. Lett. 70, 221–236. Goudie, A.S., Wells, G.L., 1995. The nature, distribution and formation of pans in arid zones. Earth-Sci. Rev. 38, 1–69. Honda, M., Yabuki, S., Shimizu, H., 2004. Geochemical and isotopic studies of aeolian sediments in China. Sedimentology 51, 211–230.

11

Iriondo, M.H., Krohling, D., 1995. El sistema eólico pampeano. In: Comunicaciones. Museo Provincial de Ciencias Naturales “Florentino Ameghino”, Nueva Serie, vol. 5, pp. 1–68. Jacobsen, S.B., Wasserburg, G.J., 1980. Sm–Nd isotopic evolution of chondrites. Earth Planet. Sci. Lett. 50, 139–155. Johnson, M.S., Meskhidze, N., Solmon, F., Gassó, S., Chuang, P.Y., Gaiero, D.M., Yantosca, R.M., Wu, S., Wang, Y., Carouge, C., 2010. Modeling dust and soluble iron deposition to the South Atlantic Ocean. Atmos. Chem. Phys. 15, 2487–2502. Kanayama, S., Yabuki, S., Zeng, F.J., Liu, M.Z., Shen, Z.B., Liu, L.C., Yanagisawa, F., Abe, O., 2005. Size-dependent geochemical characteristics of Asian dust-Sr and Nd isotope compositions as tracers for source identification. J. Meteorol. Soc. Jpn. 83 (A), 107–120. Lassig, J.L., Cogliati, M.G., Bastanski, M.A., Palese, C., 1999. Wind characteristics in Neuquen, North Patagonia, Argentina. J. Wind Eng. Ind. Aerodyn. 79, 183–199. http://dx.doi.org/10.1016/S0167-6105(98)00110-X. Lamy, F., Gersonde, R., Winckler, G., Esper, O., Jaeschke, A., Kuhn, G., Ullermann, J., Martinez-Garcia, A., Lambert, F., Kilian, R., 2014. Increased dust deposition in the Pacific Southern Ocean during glacial periods. Science 343, 403–407. http:// dx.doi.org/10.1126/science.1245424. Lamy, F., Arz, H.W., Kilian, R., Lange, C.B., Lembke-Jene, L., Wengler, M., Kaiser, J., Urrea, O.B., Hall, I.R., Harada, N., Tiedemann, R., 2015. Glacial reduction and millennial-scale variations in Drake Passage through flow. Proc. Natl. Acad. Sci. USA 112, 13496–13501. http://dx.doi.org/10.1073/pnas.1509203112. Li, F., Ginoux, P., Ramaswamy, V., 2008. Distribution, transport, and deposition of mineral dust in the Southern Ocean and Antarctica: contribution of major sources. J. Geophys. Res. 113, D10207. http://dx.doi.org/10.1029/2007JD009190. Maher, B.A., Prospero, J.M., Mackie, D., Gaiero, D., Hesse, P., Balkanski, Y., 2010. Global connections between aeolian dust, climate and ocean biogeochemistry at the present day and at the last glacial maximum. Earth-Sci. Rev. 99, 61–97. Martínez-García, A., Sigman, D.M., Ren, H., Anderson, R.F., Straub, M., Hodell, D.A., Jaccard, S.L., Eglinton, T.I., Haug, G., 2014. Iron fertilization of the Subantarctic Ocean during the last Ice Age. Science 343 (6177), 1347–1350. http:// dx.doi.org/10.1126/science.1246848. Milana, J.P., 2009. Largest wind ripples on Earth? Geology 37, 343–346. http:// dx.doi.org/10.1130/G25382A.1. Noble, T.L., Piotrowski, A.M., Robinson, L.F., McManus, J.F., Hillenbrand, C.D., Bory, A.J.M., 2012. Greater supply of Patagonian-sourced detritus and transport by the ACC to the Atlantic sector of the Southern Ocean during the last glacial period. Earth Planet. Sci. Lett. 317–318, 374–385. http://dx.doi.org/10.1016/ j.epsl.2011.10.007. Norte, F.A., Ulke, A.G., Simonelli, S.C., Viale, M., 2008. The severe Zonda wind event of 11 July 2006 east of the Andes Cordillera (Argentine): a case study using the BRAMS model. Meteorol. Atmos. Phys. 102 (1), 1–14. Norte, F.A., 2015. Understanding and forecasting Zonda wind (Andean Foehn) in Argentina: a review. Atmos. Clim. Sci. 5, 163–193. Peiretti, R., Dumanski, J., 2014. The transformation of agriculture in Argentina through soil conservation. Int. Soil Water Conserv. Res. 2 (1), 14–20. Prohaska, F.J., 1976. Climates of Central and South America. In: Schwerdtfeger, A. (Ed.), World Survey of Climatology. Elsevier, New York, NY, pp. 13–72. Prospero, J.M., Ginoux, P., Torres, O., Nicholson, S.E., Gill, T.E., 2002. Environmental characterization of global sources of atmospheric soil dust identified with the NIMBUS 7 Total Ozone Mapping Spectrometer (TOMS) absorbing aerosol product. Rev. Geophys. 40 (1), 1002. http://dx.doi.org/10.1029/2000RG000095. Rampsberger, R., Peinemann, N., Stahr, K., 1998. Deposition rates and characteristics of aeolian dust in semi-arid and sub-humid region of the Argentinean Pampa. J. Arid Environ. 39, 467–476. Revel-Rolland, M., De Dekker, P., Delmonte, B., Hesse, P., Magee, J.W., Basile-Doelsch, I., Grousset, F., Bosch, D., 2006. Eastern Australia: a possible source of dust in East Antarctica interglacial ice. Earth Planet. Sci. Lett. 249, 1–13. Simonella, L., Palomeque, E.M.E., Croot, P.L., Stein, A., Kupczewski, M., Rosales, A., Montes, M.L., Colombo, F., García, M.G., Villarosa, G., Gaiero, D.M., 2015. Soluble iron inputs to the Southern Ocean through recent andesitic to rhyolitic volcanic ash eruptions from the Patagonian Andes. Glob. Biogeochem. Cycles 29, 1125–1144. http://dx.doi.org/10.1002/2015GB005177. Skonieczny, C., Bory, A., Bout-Roumazeilles, V., Abouchami, W., Galer, S.J.G., Crosta, X., Stuut, J.B., Meyer, I., Chiapello, I., Podvin, T., Chatenet, B., Diallo, A., Ndiaye, T., 2011. The 7–13 March 2006 major Saharan outbreak: multiproxy characterization of mineral dust deposited on the West African margin. J. Geophys. Res. 116, D18210. http://dx.doi.org/10.1029/2011JD016173. Smith, J., Vance, D., Kemp, R.A., Archer, C., Toms, P., King, M., Zarate, M., 2003. Isotopic constraints on the source of Argentinian loess – with implications for atmospheric circulation and the provenance of Antarctic dust during recent glacial maxima. Earth Planet. Sci. Lett. 212, 181–196. Strecker, M.R., Alonso, R.N., Bookhagen, B., Carrapa, B., Hilley, G.E., Sobel, E.R., Trauth, M.H., 2007. Tectonics and climate of the southern central Andes. Annu. Rev. Earth Planet. Sci. 35, 747–787. Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: Its Composition and Evolution. Blackwell, Oxford. 315 pp. Toggweiler, J.R., Russell, J.L., Carson, S., 2006. Midlatitude westerlies, atmospheric CO2 and climate change during the ice ages. Paleoceanography 21, PA2005. http://dx.doi.org/10.1029/2005PA001154.

JID:EPSL

12

AID:14402 /SCO

[m5G; v1.213; Prn:24/04/2017; 15:51] P.12 (1-12)

S. Gili et al. / Earth and Planetary Science Letters ••• (••••) •••–•••

Tripaldi, A., Zárate, M.A., Brook, G.A., Li, G.-Q., 2011. Late Quaternary paleoenvironments and paleoclimatic conditions in the distal Andean piedmont, southern Mendoza, Argentina. Quat. Res. 76, 253–263. Vallelonga, P., Gabrielli, P., Balliana, I., Wegner, A., Delmonte, B., Turetta, C., Vanhaecke, F., Rosman, K.J.R., Hong, S., Boutron, C.F., Cescon, P., Barbante, C., 2010. Lead isotopic compositions in the EPICA Dome C ice core and Southern Hemisphere Potential Source Areas. Quat. Sci. Rev. 29, 247–255.

Walter, H.J., Hegner, E., Diekmann, B., Kuhn, G., Rugters Van Der Loeff, M.M., 2000. Provenance and transport of terrigenous sediment in the South Atlantic Ocean and their relations to glacial and interglacial cycles: Nd and Sr isotopic evidence. Geochim. Cosmochim. Acta 64, 3813–3827. Zárate, M.A., Tripaldi, A., 2012. The aeolian system of central Argentina. Aeolian Res. 3, 401–417. http://dx.doi.org/10.1016/j/aeolia.2011.08.002.