Marys Medicine

Weakening and strengthening of the indian monsoon during heinrich events and dansgaardoeschger oscillations

Weakening and strengthening of the Indian monsoon during Heinrich events and Dansgaard-Oeschger oscillations • Intensity of Indian monsoon is traced with geochemical and grain size Gaudenz Deplazes1, Andreas Lückge2, Jan-Berend W. Stuut3,4, Jürgen Pätzold3, Holger Kuhlmann3, analyses of sediments Dorothée Husson3, Mara Fant3, and Gerald H. Haug1,5 • Fluvial versus aeolian sediment input to Arabian Sea mimics DO oscillations 1Geological Institute, Department of Earth Sciences, ETH Zürich, Zürich, Switzerland, 2Bundesanstalt für Geowissenschaften • During Heinrich events the Indian monsoon weakened distinctly und Rohstoffe, Hannover, Germany, 3MARUM-Center for Marine Environmental Sciences, Bremen University, Bremen,Germany, 4NIOZ-Royal Netherlands Institute for Sea Research, Texel, The Netherlands, 5DFG-Leibniz Center for Earth SurfaceProcess and Climate Studies, Institute for Geosciences, Potsdam University, Potsdam, Germany Supporting Information:• Readme Abstract The Dansgaard-Oeschger oscillations and Heinrich events described in North Atlantic sediments and Greenland ice are expressed in the climate of the tropics, for example, as documented in Arabian Sea sediments. Given the strength of this teleconnection, we seek to reconstruct its range of environmental impacts. We present geochemical and sedimentological data from core SO130-289KL from the Indus submarine slope spanning the last 80 kyr. Elemental and grain size analyses consistently indicate that Correspondence to: interstadials are characterized by an increased contribution of fluvial suspension from the Indus River. In contrast, stadials are characterized by an increased contribution of aeolian dust from the Arabian Peninsula.
Decadal-scale shifts at climate transitions, such as onsets of interstadials, were coeval with changes inproductivity-related proxies. Heinrich events stand out as especially dry and dusty events, indicating a dramatically weakened Indian summer monsoon, potentially increased winter monsoon circulation, and Deplazes, G., A. Lückge, J.-B. W. Stuut,J. Pätzold, H. Kuhlmann, D. Husson, increased aridity on the Arabian Peninsula. This finding is consistent with other paleoclimate evidence for M. Fant, and G. H. Haug (2014), continental aridity in the northern tropics during these events. Our results strengthen the evidence that Weakening and strengthening of the circum-North Atlantic temperature variations translate to hydrological shifts in the tropics, with major Indian monsoon during Heinrich eventsand Dansgaard-Oeschger oscillations, impacts on regional environmental conditions such as rainfall, river discharge, aeolian dust transport, and Paleoceanography, 29, 99–114, ocean margin anoxia.
Received 17 MAY 2013Accepted 3 JAN 2014 Accepted article online 8 JAN 2014Published online 18 FEB 2014 Paleoproxy and climate-modeling studies have documented correspondences between tropical and high-latitudinal climate variability on various time scales [Behl and Kennett, 1996; Schulz et al., 1998; Peterson et al.,2000; Wang et al., 2001; North Greenland Ice Core Project Members, 2004; Zhang and Delworth, 2005]. Duringthe last glacial period, climate variability is characterized by millennial-scale warmer (interstadial) and colder(stadial) Dansgaard-Oeschger (DO) oscillations [Bond et al., 1993; Dansgaard et al., 1993]. Some of the colderDO oscillations coincide with anomalous occurrences of ice-rafted detritus in North Atlantic sediments, whichare referred to as Heinrich events [Heinrich, 1988; Hemming, 2004]. It is postulated that instabilities of theNorthern Hemisphere glacial ice sheets led to a freshwater input to the North Atlantic during Heinrich events,which caused a strong decrease or even shutdown of the Atlantic meridional overturning circulation (AMOC)[Broecker, 1994; Ganopolski and Rahmstorf, 2001; McManus et al., 2004]. The DO oscillations and Heinrichevents are also expressed in climate records of the Indo-Asian realm [Schulz et al., 1998; Wang et al., 2001;Altabet et al., 2002; Clemens and Prell, 2003; Higginson et al., 2004; Wang et al., 2008; Banakar et al., 2010].
However, forcing and response mechanisms of the Indo-Asian monsoon system in conjunction with thesemillennial-scale oscillations are still poorly understood [Kudrass et al., 2001; Caley et al., 2011].
Sedimentary archives in the northeastern Arabian Sea have been described as sensitive recorders of Indianmonsoonal climate on annual to millennial time scales [Schulz et al., 1998; von Rad et al., 1999; Lückge et al.,2002]. During the past years a controversial discussion evolved whether the mainly productivity-relatedproxies in the Arabian Sea reflect dominantly changes in the Indo-Asian monsoonal intensity or ratherchanges in the AMOC [Pourmand et al., 2004; Schmittner et al., 2007; Böning and Bard, 2009; Ziegler et al., 2010;Caley et al., 2011].
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.

Figure 1. Environmental setting of sediment core SO130-289KL in the northeastern Arabian Sea. (a) SeaWiFS (Sea-viewing Wide Field-of-viewSensor) image of a storm transporting dust from the Arabian Peninsula to the Arabian Sea at 12 March 2000 (SeaWiFS project, NASA/GoddardSpace Flight Center, Orbital Imaging Corporation (ORBIMAGE)). (b) Bathymetric map of the Arabian Sea and the Persian Gulf (geomorphologicdata from General Bathymetric Chart of the Oceans 08). The bathymetric contour of 120 m is highlighted, which is approximating the coastlineduring maximum last glacial lowstand of sea level. The black arrows schematically indicate the wind field during early summer monsoon, whichis favorable to transport dust from the Arabian Peninsula to the indicated core location [Krishnamurti et al., 1980; Sirocko and Sarnthein, 1989].
In this study, we present a multiproxy record of sediment core SO130-289KL from the oxygen minimum zone(OMZ) in the northeastern Arabian Sea (Figure 1) in order to reconstruct monsoon-related climate and oce-anic variability over the last 80 kyr in decadal- to centennial-scale resolution. Sediment total color reflectance(L*) and total organic carbon (TOC) measurements are used to reconstruct marine productivity and organicmatter preservation. Bulk chemical sediment analyses (X-ray fluorescence (XRF)) are applied to infer varia-tions of biogenic and terrigenous input as well as their provenance. Grain size analyses and end-membermodeling of siliciclastic grain size distributions are used to identify transport mechanisms of the sediments.
The combination of high-resolution grain size and elemental analyses sheds new light on the interaction,chronology, and potential causalities between atmospheric and oceanographic driving mechanisms ofsediment formation in the Arabian Sea in decadal- to centennial-scale resolution.
2. Modern Indian Monsoon The modern seasonal cycle over the Arabian Sea is dominated by the Indian monsoon. In boreal summer(June–September) powerful southwesterly winds (Figure 1) are generated by the large pressure gradientbetween the Indian-Tibetan low-pressure cell and a belt of high pressure over the Southern Ocean [Websteret al., 1998]. It is debated if the strong summer monsoon is mainly caused by elevated heating of the Tibetanplateau [Wu et al., 2012] or rather by orographic insulation of warm, moist air over continental India from thecold and dry extratropics [Boos and Kuang, 2010]. In any case, the strong and moist monsoonal southwesterlywinds, which are linked to the Intertropical Convergence Zone (ITCZ), provide intense rainfall to the Indiansubcontinent. This leads, together with snowmelt in the Himalayas, to high runoff rates from the continent[Bookhagen and Burbank, 2010], which results in high terrigenous input to the coastal sediments near theIndus River mouth [Milliman et al., 1984].
The Arabian Sea receives one of the highest amounts of dust input in the world [Rea, 1994]. The highest windspeeds and dust loads around the Arabian Sea are reached in spring and during the summer monsoon period[Ackerman and Cox, 1989; Clemens, 1998; Schulte and Müller, 2001]. In this season dust is mobilized in thenorthern Arabian Peninsula and around the Persian Gulf by northwesterly winds, which result from cycloniccirculation around the Asian low [Middleton, 1986; Sirocko et al., 1993; Clemens, 1998]. These winds arereferred to as Shamal in the Persian Gulf region. The dust transport is peaking during occasional major duststorms [Ackerman and Cox, 1989]. The dust-bearing, dry, and warm winds are lifted up and override the low-level, moist, and cooler southwesterly winds leading to a monsoonal inversion between the two air masses DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
[Findlater, 1969]. The convergence of these winds forms the regional summer ITCZ. The northwesterly windstend to turn east above the monsoon inversion [Sirocko and Sarnthein, 1989], and while lifting up, they dropmost of their load, which is partly picked up by the low-level southwesterly wind resulting in a dust transportto the northeastern Arabian Sea (Figure 1).
The summer monsoon southwesterly winds trigger upwelling of nutrient-rich waters, particularly in thewestern Arabian Sea [Honjo et al., 1999]. Enhanced nutrient supply to surface waters leads to extremely highproductivity in the Arabian Sea during the summer monsoon. Subsequently, high remineralization rates oforganic matter lead to a rapid consumption of oxygen and establish oxygen-depleted conditions in the watercolumn and on the sea floor within the OMZ.
In boreal winter the land-sea pressure gradient is reverse and moderate, dry and cold northeastern windsflow from the Indian-Tibetan high-pressure zone toward the southward shifted ITCZ, which is located atabout 10°S. Northeasterly winds induce convective mixing along the coast, generating a second, smallerproductivity maximum in the northeastern Arabian Sea [Reichart et al., 1998; Kumar et al., 2000]. The wintermonsoon is characterized by overall drier conditions and lower dust input into the Arabian Sea [Sirocko andLange, 1991; Clemens, 1998; Prins and Weltje, 1999]. However, the western Himalayas, also part of the IndusRiver drainage basin, are strongly influenced by the westerlies during the winter. The westerlies bringmoisture from the North Atlantic, Mediterranean, or other inland seas, which leads to high winter snow coverand subsequent snowmelt-runoff during spring [Bookhagen and Burbank, 2010].
3. Materials and Methods 3.1. Core Setting Sediment core SO130-289KL (23°07.34′N, 66°29.84′E, 571 m water depth, Figure 1) was taken at the Sindhcontinental margin within the OMZ that expands today from about 200 to 1200 m water depth [von Rad et al.,1999]. This 20.2 m long piston core was recovered on a levee at the flank of the Indus submarine canyon offthe Indus River mouth. Oceanographic setting and sedimentology of the core setting of SO130-289KL andneighboring cores SO90-136KL and SO90-137KA have been described by, e.g., Schulz et al. [1998], von Radet al. [1999], and von Rad et al. [2002].
The first-order age model for SO130-289KL is based on 20 accelerator mass spectrometry 14C dates and crosscorrelation to radiocarbon dated color records from the same region [Deplazes et al., 2013]. The age modelwas further fine tuned by correlating the high-resolution total color reflectance (L*) records to the ice coreδ18O record of North Greenland Ice Core Project (NGRIP) using the GICC05modelext age scale [NorthGreenland Ice Core Project Members, 2004; Wolff et al., 2010]. Because of this fine tuning it is not possible tointerpret lead or lag relationships between the two records although lead and lag relationships among theproxies measured from the same core are certainly valid. An additional time control point is given by the TobaAsh layer identified in core SO130-289KL at 18.43 m subbottom depth (Figure 2) [Schulz et al., 2002; von Radet al., 2002; Storey et al., 2012].
3.3. Lamination Index A Lamination Index (LI) has been established and is similar to the Bioturbation Index of Behl and Kennett[1996] and von Rad et al. [1999]. A value of 4 characterizes undisturbed sediments with fine laminations(typically < 1 mm to 2 mm thick). A value of 3 represents discontinuous or irregular laminations (typically 1 to4 mm thick). A LI of 2 describes slightly to partly bioturbated sediments. A LI of 1 represents strongly tocompletely bioturbated or homogenous sediments. The LI is not defined on the entire cross-core length buton the 1.7 cm broad image transects of which the total color reflectance (L*) was determined.
3.4. Total Organic Carbon (TOC) Samples for TOC measurements were taken every 2 cm in the interval of 7.51 to 9.79 m core depth (115samples). The sediment samples were dried and TOC was determined with a LECO CS-444 instrument byinfrared detection after combustion at 1400°C. Prior to combustion, TOC samples were acid treated (10% HClat 80°C) to remove inorganic carbon.
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
Figure 2. Comparison of color reflectance (L*) and elemental abundances from the northeastern Arabian Sea (SO130-289KL) with NGRIP δ18O ice core record. (a) NGRIP δ18O record (dark blue) with 20 year resolution over the last 80 kyr [Wolff et al., 2010]. (b) Sediment totalreflectance (L*) [Deplazes et al., 2013] colored corresponding to the Lamination Index (LI). A LI of 4 indicates finely laminated sediments, a LIof 1 indicates homogenous or bioturbated sediments. Elemental concentrations of (c) nickel (green), (d) aluminum (red), (e) calcium (lightblue), and (f) elemental ratio of strontium to calcium Sr/Ca (purple). Greenland interstadial numbers are indicated above, Heinrich events (H)below the L* record. Double-headed arrows indicate stratigraphic tie points. Yellow shaded areas denote relatively warmer climates in theNorthern Hemisphere.
3.5. Conventional XRF Measurements Bulk chemical sediment analyses were performed on samples from almost every 2 cm over the entire 20.2 mlong core (966 samples). This results in an average temporal sampling resolution of 80 years over the entirecore. The analyses were accomplished with X-ray fluorescence (XRF) using Philips PW 2400 and PW 1480wavelength dispersive spectrometers. Forty-two major and trace elements were quantitatively analyzed afterfusion of the samples with lithium metaborate at 1200°C for 20 min (sample/LiBO2 = 1/5). Quality of the re-sults was controlled with certified reference materials (i.e., Community Bureau of Reference, Brussels). Theprecision for major elements was generally better than ±0.5% and better than 5% for trace elements. Threeoutliers in the presented nickel (Ni) record have been removed from the record (3.58, 8.07, and 17.29 m).
3.6. Grain Size Analysis and End-Member Modeling Grain size distributions were studied in 2 cm intervals in the upper half of the core (0.21–9.79 m, 474 samplesparallel to XRF, Figure 3 and Figures S1–S4 in the supporting information). The samples were pretreated withH2O2 (30–35%), HCl (10%), and NaOH (6 g pellets dissolved in 100 mL H2O) to separate the terrigenous DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.

Figure 3. Comparison of color reflectance (L*) with proxies tracing sediment composition, transport, and source over the (top) last 42 kyrand over the time interval of (bottom) 41.6–35.0 kyr before 2000 AD. (a) Sediment total reflectance (green). (b) Median grain size (red)and elemental ratio of zirconium to aluminum Zr/Al (black). (c) Grain size end-member EM1 as proxy for the coarser aeolian dust (maroon).
(d) Elemental ratio of magnesium to aluminum Mg/Al (orange) as provenance indicator of the terrigenous sediment fraction (Figure 5). (e)Total organic carbon (TOC, black). Blue-shaded areas indicate Heinrich events 4 to 1. Greenland interstadial numbers are indicated above,Heinrich events (H) below the L* record. Bølling-Allerød (BA), Younger Dryas (YD), Preboreal (PB), and Holocene (HOL) are assigned.
sediment fraction from organic carbon, calcium carbonate, and biogenic opal, respectively. The presentedgrain size measurements reflect therefore only variations within the siliciclastic fraction. To prevent formationof aggregates Na4P2O7 × 10 H2O was added prior to grain size analysis. Grain size analyses were performed ona laser diffraction particle size analyzer (Beckman Coulter) LS200, resulting in 92 size classes from 0.4to 2000 μm.
An inversion algorithm for end-member (EM) modeling was applied to unmix the polymodal grain size dis-tributions that are composed of sediment subpopulations resulting from different transport mechanisms[Weltje, 1997; Prins and Weltje, 1999; Prins et al., 2000] (Figures S1–S4). The 2-EM and 4-EM models capture thedominantly bimodal grain size distributions best. The 4-EM has a mean coefficient of determination of 89.7%,i.e., the end-member model reproduces on average 89.7% of the input variances [Prins and Weltje, 1999; Stuutet al., 2002]. The 2-EM has coefficient of determination of 64.7%. In spite of the lower coefficient of deter-mination a 2-EM model was selected for the investigated intervals, because of the principle of parsimony(a minimal number of end-members) and a reasonable attribution of end-members to particular mecha-nisms of transport.
4.1. Facies and Chemical Composition of Sediments The sediments of core SO130-289KL are characterized by two sedimentary facies during the last glacial pe-riod. During interstadials the sediments are dark colored and distinctly to indistinctly laminated; duringstadials and Heinrich (equivalent) events they are light colored and in general homogenous or bioturbated.
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
The L* record and Lamination Index (LI) trace these changes with low L* and high LI values during theinterstadials and vice versa during stadials (Figure 2). Heinrich events are characterized by light-coloredsediments having highest L* values in the entire record.
Two similar, alternating facies can be found at the beginning of the current interglacial period (Figures 2 and 3).
The sediments of the Bølling-Allerød (BA) are dark colored and finely to discontinuously laminated, whereas thesediments of the Younger Dryas (YD) are white colored and bioturbated. The sediments from the "Preboreal"(PB, 11.7–10.9 kyr) are unique within the entire record and consist of finely laminated sediments with a reddishcolor. The Holocene sediments are dark colored. The early to mid-Holocene sediments are slightly to partlybioturbated, whereas the middle to late Holocene sediments are discontinuously laminated.
Sediment-color properties have been used in this region as a proxy for TOC (%) [Schulz et al., 1998]. This is inline with the observation that the L* record shows a remarkable similarity to the new TOC measurements inthe interval of 41 to 35.5 kyr before 2000 AD (Figure 3e). A polynomial regression of second order of L*(smoothed over 2 cm) to TOC (%) revealed a coefficient of determination R2 of 0.9. The TOC content isgenerally lower in stadials and Heinrich events than in the interstadials and varies between 0.6% in Heinrichevent 4 and up to 3.6% in the Interstadial 10.
The L* record is also mimicked by the concentration of the trace element nickel (Ni). Ni is delivered to thesediment mainly in association with organic matter [van der Weijden et al., 2006] and underlines that the L*record traces the organic matter content over the entire record.
In the time period of 81 to 68 kyr before 2000 AD (mainly end of Marine Isotope Stage (MIS) 5, Lisiecki andRaymo [2005]) element concentrations of terrigenous, siliciclastic origin (e.g., Al, Fe, K, Si, Ti, and Zr) reproducethe DO oscillations traced by the L* and Ni records only partly and show a reduced amplitude in theirvariability (Figures 2 and S5). In contrast, during most of the last glacial period ( 68–14 kyr, mainly MIS 4–2)the terrigenous element concentrations mimic the DO oscillations traced by the L* and Ni records. From 13 to2 kyr before 2000 AD elements of terrigenous origin show again a different, temporarily opposite patterncompared to the L* and Ni records. The CaCO3 content, traced by calcium (Ca) concentrations, shows apattern that is inverse to the siliciclastic element concentrations. In summary, the sediment consists mainly ofCaCO3, a siliciclastic fraction and organic carbon; biogenic opal is rare (< 1%) [von Rad et al., 1999].
Elements of siliciclastic origin were additionally plotted normalized to Al to investigate changes in thechemical composition related to transport mechanisms or provenance of the terrigenous sediments that arenot depending on dilution by other sedimentary constituents [Lückge et al., 2001]. Terrigenous element ratiossuch as Ti/Al, Mg/Al, and Zr/Al are neither depending on redox nor on productivity changes and show similar,inverse millennial- to centennial-scale variability patterns as the terrigenous element concentrations(Figures 3, S5, and S6). The element ratio of Sr/Ca follows in general the pattern of Ca (%) (Figure 2).
4.2. Grain Size Analysis The median grain size record of the terrigenous fraction over the last 41 kyr shows in general higher valuesduring the end of the last glacial period and lower values during the current interglacial period (Figure 3). Theabsolute values vary between 4 and 24 μm. On millennial-scale, the median grain size record shows apattern concomitant with the DO oscillations as defined by the L* record with low values during interstadials,higher values during stadials, and highest values during Heinrich events. At prominent climate transitions,such as at the onset of Interstadial 8, the median grain size shows abrupt changes, which are coeval withchanges in L*, in element concentrations and in element to Al ratios (Figure 3). The "two-end-member (EM)model" is in accordance with the median grain size record. The end-members EM1 and EM2 have modal grainsizes of 22.5 μm and 4.7 μm (Figure S3). The element ratio of Zr/Al traces the general pattern in the me-dian grain size and 2-EM model in great detail (Figure 3b).
5.1. Paleoproductivity and OMZ Intensity Total color reflectance (L*) [Deplazes et al., 2013] and Ni (%) show a similar millennial- to centennial-scalepattern similar to paleoclimate records from Greenland [North Greenland Ice Core Project Members, 2004; Wolffet al., 2010] (Figure 2), North Atlantic [Bond et al., 1993], and from the Indo-Asian realm [e.g., Wang et al., 2001].
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
The millennial-scale variability characterized by DO oscillations and intercalated Heinrich events has beenrecognized at several locations documented by different proxies in the Arabian Sea [Schulz et al., 1998; Altabetet al., 2002; Clemens and Prell, 2003; Ivanochko et al., 2005]. However, at least three hypotheses were proposedwhich describe the imprint of millennial- to orbital-scale climate oscillations in Arabian Sea sediments.
The first hypothesis assigns the dominant role to the Indian southwest summer monsoon [Schulz et al., 1998;Schulte and Müller, 2001; Altabet et al., 2002; Clemens and Prell, 2003; Caley et al., 2011]. A strong southwestsummer monsoon leads to coastal and open-ocean upwelling of nutrient-rich subsurface waters mostly inthe western Arabian Sea causing high productivity in large parts of the Arabian Sea. High surface waterproductivity leads to high export production and high oxygen demand driven by remineralization of organicmatter and ultimately to anoxic conditions hampering bioturbation and potentially enhancing preservationof organic matter. Accordingly, Clemens and Prell [2003] used mainly proxies that are potentially productivityrelated (δ15N, opal mass accumulation rate (MAR), percent Globigerina bulloides, excess Ba MAR, andlithogenic grain size) to define the Arabian Sea Summer Monsoon stack (SM stack).
In the second hypothesis, the productivity and intensity of the OMZ in the northern Arabian Sea are not onlydetermined by the summer monsoon but also by an interplay of summer and winter monsoon elements.
During stadials and Heinrich events enhanced northeastern monsoonal winds lead to winter mixing andintermediate water formation down to at least 600 m water depth resulting in weaker OMZ conditions[Reichart et al., 1998; Reichart et al., 2004; Klöcker and Henrich, 2006]. In contrast, interstadials are characterizedby productivity maxima and shallow winter mixing.
The third hypothesis describes productivity and OMZ intensity changes as a function of changes in theAMOC. The intensity of the OMZ might be influenced by changes in the formation of Subantarctic ModeWater and Antarctic Intermediate Water (SAMW-AAIW) [Schulte et al., 1999; Pichevin et al., 2007; Böning andBard, 2009]. Stable isotope studies on planktonic and benthic foraminifera off Somalia suggest enhancedintermediate water supply to the northern Indian Ocean when the AMOC was reduced [Jung et al., 2009].
Intensified northeastward injection of oxygen-rich SAMW-AAIW would lead to increased ocean ventilationand decreased intensity of the OMZ during stadials and Heinrich events compared to interstadials. However,climate-model simulations of oxygen and nutrient concentrations during the last glacial period suggest thatchanges in consumption are more important than ventilation in the Arabian Sea [Schmittner et al., 2007].
According to these model simulations, reduced AMOC conditions lead to a decreased nutrient delivery to theupper Indo-Pacific Ocean, reducing productivity and causing a reduced OMZ intensity as a consequence ofreduced oxygen consumption. The model, however, does not reproduce the strong OMZ conditions in theArabian Sea.
In our study the DO oscillations and the current interglacial period are traced in high resolution by the L*record and Ni (%) that show a high correlation with organic matter contents. Organic petrographic analysesindicate that the organic matter is predominantly of marine origin [Lückge et al., 1999, 2012]. The observedsignal cannot be the result of dilution by carbonate or siliciclastic contents as they show a different pattern,for example, during the deglaciation. The L* record could in principle reflect OMZ intensity and thereforerepresent a preservation rather than a productivity signal. The Sr/Ca record (Figure 2) that has beeninterpreted as an aragonite dissolution record [Reichart et al., 1998; Klöcker and Henrich, 2006; Böning andBard, 2009] and therefore as an OMZ intensity proxy follows, in general, the L* oscillations. However, surfacesediment studies from the Oman and Pakistan margins suggest that bottom water oxygen concentration isnot solely driving the amount of buried organic matter [Cowie et al., 2009]. Instead, a high settling flux oforganic matter (i.e., high export productivity in the surface water) [Pedersen et al., 1992] seems to be needed.
The basic question whether the different proxies reflect mainly primary productivity or bottom water oxygenconcentration or a combination of both (i.e., oxygen exposure time) [Sinninghe Damsté et al., 2002] has been amatter of debate since several decades [Tyson, 2005]. It is difficult to separate these two mechanisms becausethey are intrinsically correlated. This means that increased productivity can directly influence the oxygenconcentration of the bottom waters. The correspondence of our L* record with Arabian Sea SM stack [Clemensand Prell, 2003] (Figure S6) and results of previous studies linking the observed Northern Hemispheremillennial-scale pattern to productivity proxies in the Arabian Sea [Reichart et al., 1998; von Rad et al., 1999;Sinninghe Damsté et al., 2002; Pourmand et al., 2004] suggest, however, that monsoonal driven productivityplays a major but not an exclusive role in determining the L* and organic matter record.
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
5.2. Terrestrial Siliciclastic Signature and Depositional Regime Climate variations in monsoon winds and precipitation influence not only the deposition of biogenic(as traced by L*) but also of lithogenic sedimentary components. Therefore, geochemical composition andproperties of the siliciclastic fraction that are independent of productivity or OMZ intensity but dependenton fluvial or aeolian input were analyzed. Since the L* measurements and the geochemical analyses wereperformed in the same core and in high resolution the records can be compared on a time scale of onaverage 80 years and in case of the onset of Interstadial 8 even in a resolution of 40–60 years.
Element concentrations of siliciclastic, terrigenous origin like Al (%) closely follow the DO pattern over mostof the presented glacial record (Figures 2 and S5). Warmer interstadials correlate with higher siliciclasticconcentrations than colder stadials. Element ratios like Ti/Al, Mg/Al, and Zr/Al, which are not influenced byproductivity or redox conditions, follow a similar pattern like Ti- and Mg-element concentrations (Figures S5and S6). This indicates that the millennial-scale pattern in the siliciclastic fraction is not just a result of dilutionby biogenic carbonate input or preservation but reflects a change in input, source, or transport conditions.
The changes in the Ti/Al, Mg/Al, and Zr/Al records are as abrupt and concomitant with changes in the L*record (Figure 3).
In contrast, signals derived from the terrigenous fraction younger than 14–13 kyr before 2000 AD showanother—partly opposite—pattern compared to the L* record. Sedimentary studies have shown that off thelarge Indus River mouth most of the terrigenous sediments are transported from the delta to the deep-seafan via large submarine canyon channel systems bypassing the continental slope [von Rad and Tahir, 1997].
The modern Indus submarine canyon starts about 3.5 km off the coast in about 20 m water depth anddeepens seaward. At the shelf break region (135 m) the Indus submarine canyon has a maximum depth ofabout 1030 m [Giosan et al., 2006; Inam et al., 2007]. Therefore, the continental margin outside the Indussubmarine canyon gets mostly covered by hemipelagic sediments with only few turbidites [von Rad andTahir, 1997]. During sea level lowstands, like toward the end of the last glacial period, most of the fluvialsediments were deposited in channel-levee complexes of the Indus submarine fan, frequently in the form ofturbidites [Prins and Postma, 2000]. During deglacial sea level rise, the deposition of turbidites decreasedstrongly on the Indus submarine fan leading to increased sedimentation in near-shore zones [Prins andPostma, 2000; Prins et al., 2000] and possibly on the continental slope because of mobilization and reworkingof sedimentary material [von Rad et al., 1999]. This shift in depositional regime and redirection of sedimenttransport could explain the observed high terrigenous portion from the Bølling-Allerød to the onset of theHolocene associated with a maximum siliciclastic deposition in the reddish clays of the Preboreal (Figure 3).
This time interval coincides with meltwater pulse MWP-1a and partly with MWP-1b representing periods ofmaximum eustatic sea level rise [Stanford et al., 2011] (Figure 4). During the Holocene the terrigenous signalreflects nearly the opposite trend expected for summer monsoon conditions [Fleitmann et al., 2003; Guptaet al., 2003], which might reflect enhanced sediment deposition within the delta and a change in depositionalregime on the continental shelf [Limmer et al., 2012]. Today, sediments are mainly deposited within theinnermost shelf and at the delta, whereas the outer shelf (> 90 m) is characterized by a lack of deposition[von Rad and Tahir, 1997]. The most recent evolution of sediment transport by the Indus River is additionallyinfluenced by extensive damming in the catchment area [Milliman et al., 1984; Clift et al., 2008; Lückgeet al., 2012].
The onset of significant changes in the depositional regime and sediment provenance seems to haveoccurred at 14 to 13 kyr before 2000 AD corresponding to a "threshold" sea level stand of 70–80 m(Figure 4). It seems that the L* record and the terrigenous signature follow tightly the DO oscillations and theSM stack (Figure S6) when the sea level was lower than 70–80 m below the modern sea level (Figure 4). Byanalogy with present-day seasonal variations, warmer and more humid interstadials are expected to becharacterized by elevated snowmelt and precipitation leading to high erosion and river sediment loads. Thisagrees with the observed enhanced contribution of siliciclastic material during the interstadials. The terrig-enous sediment pattern is opposite to the SM stack or shows a considerably reduced millennial-scalevariability when the sea level was above 70–80 m during the time periods of 80–68 and < 14 kyr. Weinterpret that the siliciclastic oscillations reflect monsoon variability during most of the last glacial period[North Greenland Ice Core Project Members, 2004]. Before and after this time period they are influenced by acomplex interplay of changes in provenance, depositional regime, and monsoon intensity.
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
Figure 4. Comparison of Arabian Sea proxies and sea level. (a) Sediment total reflectance record (L*, green). Greenland interstadialnumbers are indicated above, Heinrich events (H) below the L* record. MIS indicates Marine Isotope Stage [Lisiecki and Raymo,2005]. (b) Highest-probability sea level history (black) based on six key deglacial sea level records over the last 22 kyr [Stanford et al.,2011]. The black and dashed lines show the 99% and 95% confidence intervals, respectively. Sea level reconstruction from northernRed Sea for 82–13 kyr before 2000 AD based on benthic δ18O considering two temperature corrections: in dark blue the alkenone seasurface temperatures as a maximum assumption for deep water temperature changes (benthic Tmax) and in purple constant tem-peratures as a minimum approach (benthic Tmin) [Arz et al., 2007]. (c) Aluminum (red). Grey-shaded areas indicate time intervals inwhich the sea level was higher than 75 m. The sedimentation model indicates changes in the depositional regime depending onthe sea level.
5.3. Fluvial Versus Aeolian Signature The siliciclastic input into the Arabian Sea can be of fluvial or aeolian origin. As the core location is near theIndus River delta, most of the terrigenous fraction is expected to be of fluvial origin [von Rad et al., 1999].
Grain size analysis and end-member modeling are valuable tools to distinguish between sediments of aeolianand of fluvial origin in the Arabian Sea [Prins and Weltje, 1999]. At the continental slope off the Indus Riverdelta, the fluvial fraction is characterized by finer grain sizes, because the coarser size fraction is either de-posited in the delta or funneled within the submarine canyon [Giosan et al., 2006; Inam et al., 2007]. In con-trast, the aeolian fraction has relatively coarser grain size distributions [Prins and Weltje, 1999; Prins andPostma, 2000]. Based on this basic premise and previous studies [Prins and Weltje, 1999; Prins and Postma,2000] the coarser EM1 is assigned as aeolian dust and the finer EM2 as fluvial mud (Figures 3 and S3). Theabsolute values of the EM have to be interpreted with caution and give only qualitative estimations of theaeolian versus fluvial contributions.
The grain size analyses over the last 41 kyr generally show increased aeolian contribution (EM1) during theend of the last glacial and enhanced fluvial contribution (EM2) during the interglacial period (Figure 3). This isin line with previous results from continental, marine, and ice core studies that show elevated dust loadsduring the last glacial period [Sirocko and Lange, 1991; Pourmand et al., 2004; Ruth et al., 2007; Roberts et al.,2011; Sun et al., 2012].
On a millennial-scale, grain size analyses show a distinct increase of coarser aeolian dust (EM1) duringHeinrich events 4 to 1 (Figure 3). In contrast, interstadials are characterized by fluvial mud (EM2) andcoarser-grained aeolian dust (EM1) is nearly absent. This trend is clearly expressed during Heinrich event4 and the subsequent Interstadial 8 (Figure 3). The measured median grain size record can be extendedover the entire record by using Zr/Al as a grain size proxy. Zirconium is commonly associated withthe coarser-grained heavy minerals [von Rad et al., 1999; Dypvik and Harris, 2001], and Zr/Al is tracingthe measured median grain size record in great detail (Figure 3). The Zr/Al ratio suggests that Heinrichevents and stadials are consistently characterized by coarser grain sizes throughout the entirerecord (Figure S6).
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
The grain size analyses strongly support the interpretation that the interstadials were more humid andwarmer leading to enhanced deposition of fluvial mud (EM2), whereas the stadials and especially theHeinrich events were drier and colder leading to enhanced deposition of aeolian dust (EM1). The relativecontribution of the coarser aeolian dust (EM1) abruptly decreases from a high percentage to 0% at thetransition from Heinrich event 4 to Interstadial 8 (Figure 3). This change in dust contribution appears to haveoccurred within less than 60 years. It is important to note that this change happens exactly at the time whenthe L* record and TOC (%) are recording the abrupt onset of Interstadial 8. This is indicating that there is nolead or lag relation between abrupt atmospheric changes traced by grain size analysis and marine produc-tivity traced by L* record and TOC (%) on a decadal-scale. It is noteworthy that these transitions seem tooccur in less than 100 years. This is 200–300 years faster than modeled oxygen concentration changes dueto changes in AMOC intensity [Schmittner et al., 2007]. However, studies of the modern oceanic circulationsuggest that, for example, changes in the formation of AAIW can be communicated within decades to theArabian Sea [Fine et al., 2008]. The observed synchronicity between grain size, L*, and other compositionalproxies suggests a dominant role of the monsoons for the oceanographic conditions and sediment formationin the Arabian Sea. Though it is likely that the oceanographic changes in the Arabian Sea were, especially on alonger time scale, also influenced by oscillations in SAMW-AAIW formation [Schmittner et al., 2007; Junget al., 2009].
The inferred higher export productivity during interstadials is classically explained with increased nutrientavailability by upwelling or changes in the oceanic current system, which are related to the southwestsummer monsoon [e.g., Schulz et al., 1998]. The observed multiproxy pattern could alternatively or addi-tionally indicate increased nutrient injection via the Indus River provoking enhanced marine productivity inthe region that is influenced by the Indus River plume and subsequent sediment deposition. The importanceof fluvial nutrient delivery on coastal productivity at the study site has been suggested for recent times[Lückge et al., 2012] and for other locations on various time scales [Peterson et al., 2000; Dagg et al., 2004]. Theimportance of riverine nutrient delivery during glacial times might have been underestimated in the north-eastern Arabian Sea.
5.4. Provenance of Aeolian Dust The Arabian Sea is surrounded by deserts in India, Pakistan, Iran, Iraq, and the Arabian Peninsula that are allpotential sources of aeolian dust [Sirocko and Lange, 1991; Pourmand et al., 2004]. Mineralogical and chemicalstudies of sediment samples from the Arabian Sea indicate that magnesium can be used as a potentialprovenance indicator of the Arabian Peninsula over the last 24 kyr [Sirocko et al., 2000]. The distribution ofMg-rich minerals in the sediments shows a persistent pattern with a maximum close to the Arabian Peninsulaand decreasing values in the more distal part of the Arabian Sea [Sirocko and Lange, 1991; Sirocko et al., 2000].
This has been interpreted as deposition of dust plumes that are rich in Mg-bearing minerals. Palygorskite(Mg-rich clay) that can be found in sabkhas and Mesozoic rocks of the Arabian Peninsula or dolomite that isformed today in coastal sabkhas around the Persian Gulf are potential source-typical minerals [Sirocko andLange, 1991; Sirocko et al., 1993]. In contrast to Mg, Al reveals low concentrations in the sediments off theArabian Peninsula [Sirocko et al., 2000]. Accordingly, Prins et al. [2000] proposed that the excess of Mg to Al, i.e.,Mg/Al and Ti/Al values of sediment samples, can be used to distinguish an aeolian "Arabian source" from afluvial "Indus River source." In this study it was also shown that Mg/Al seems to not reflect grain size at the Indussubmarine fan and the Oman continental slope.
The Mg/Al record of SO130-289KL shows a millennial-scale pattern similar to the aeolian end-member EM1derived from the grain size analyses (Figures 3 and S6). A cross plot of Mg/Al versus Ti/Al indicates thatsediments deposited during Heinrich events rather have an Arabian source signature, whereas interstadialsrather have an Indus River source signature (Figure 5).
The Mg/Al record in general increases toward the end of the last glacial period, which suggests an increasedrelevance of the Arabian source. This long-term trend could partly be related to the lowering of the eustaticsea level leading to an exposure of the Persian Gulf, which represents a potential source of Mg-rich minerals[Sirocko et al., 1993]. At the end of the last glacial period the sea level was 120 m lower than today, so thatthe Persian Gulf was largely exposed (Figure 1b). However, the observed millennial-scale oscillations of Mg/Alare likely not reflecting changes in sea level fluctuations but rather changes in monsoonal climate variability, DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
Figure 5. Provenance of terrigenous sediments deposited during major interstadials (reddish) and Heinrich events (bluish) over the timeinterval of 62–35 kyr. (a) Cross plot of elemental ratios of magnesium to aluminum Mg/Al and titanium to aluminum Ti/Al to trace thesource of the terrigenous sediment from the Arabian Peninsula (aeolian) versus Indus River as shown in Prins et al. [2000]. (b) Sediment totalreflectance (L*) record with indicated samples used to characterize the major interstadials and Heinrich events in Figure 5a.
because the sea level seems not to fluctuate as abruptly as the above described proxies [Arz et al., 2007;Siddall et al., 2008].
It was previously postulated that in times of weakened summer monsoon the winter monsoon strengthened[Reichart et al., 1998; Sun et al., 2012] and that this could have led to increased aeolian input from thenortheastern deserts like the Thar Desert [Prins and Weltje, 1999]. Based on the presented data an additionalinput cannot be excluded but the Mg/Al record suggests that such an input was not important enough tooverprint, for example, the high Mg/Al values associated with Heinrich events, which are typical for anArabian Peninsula provenance. Additionally, studies in the Thar Desert showed that during the last glacialperiod there was no major dune formation and that dust transport seems to depend mainly on the strengthof the southwest summer monsoon [Kar et al., 2001; Glennie et al., 2002].
5.5. Climate Implication for the Arabian-Indian Realm The grain size and geochemical analyses are consistently indicating that the interstadials during the lastglacial period were characterized by an increased contribution of fluvial mud from the Indus River to theArabian Sea (Figures 3 and S6). The Indus River discharge depends on rainfall and snowmelt [Bookhagen andBurbank, 2010]. This trend is therefore suggesting that warmer and more humid climate conditions in asso-ciation with an intensified summer monsoon in the Indus River catchment area led to increased snowmeltand precipitation resulting in increased erosion and river sediment loads. This trend could be additionallystrengthened by a weakened Indian winter monsoon and an increased westerly (winter) circulation, leadingto increased snowfall and subsequent snowmelt in the western Himalayas. In contrast, the colder climatestates are characterized by a relative increase of aeolian dust from the Arabian Peninsula.
Dust mobilization and transport from the Arabian Peninsula to the Arabian Sea have been interpreted in termsof continental aridity, wind strength, wind directions, and provenance [Clemens, 1998; Prins and Weltje, 1999; DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.

Figure 6. Co-occurrence of Heinrich events and increased aridity or decreased precipitation intensity reconstructed at different sites in thenorthern (sub) tropics. (top) The core locations of (bottom) the proxy records. Green coloring shows the distribution of average rainfall in Julyover the period of 1979 to 1995 [Janowiak and Xie, 1999]. (a) Bulk carbonate δ18O record (dark blue) at IODP Site U1308 tracing Heinrichevents in the North Atlantic [Hodell et al., 2008a]. (b) Elemental ratio of zirconium to aluminum Zr/Al (maroon) from the northeastern ArabianSea (this study) reflecting grain size, source aridity, and wind strength. (c) Elemental ratio of iron to potassium Fe/K (red) at core GeoB9508 offSenegal tracing Sahel aridity [Mulitza et al., 2008]. (d) Magnetic susceptibility (MS, light green) at site PI-6 in the Lake Péten Itzá reflectingwet-dry cycles [Hodell et al., 2008b]. (e) Stalagmites δ18O records from Hulu Cave (PD, black; MSD, purple; MSL, orange) [Wang et al., 2001]and Sanbao Cave (SB26, light blue; SB10, red; SB22, dark green) [Wang et al., 2008] reflecting the intensity of the East Asian monsoon. Forcomparison, the Hulu δ18O record is plotted 1.6‰ more negative to account for the higher Hulu values than Sanbao Cave [Wang et al., 2008].
Pourmand et al., 2004]. Increased wind strength leads to an increase of the transported terrigenous grain sizes,whereas the total amount of transported dust seems to reflect mainly source aridity on the Arabian Peninsula[Clemens and Prell, 1990; Rea, 1994; Clemens, 1998]. The observed increase in aeolian contribution and grain sizein core SO130-289KL could therefore reflect increased wind strength of the northwesterly winds inMesopotamia and around the Persian Gulf and/or decreased summer monsoon-ITCZ intensity over the ArabianPeninsula, i.e., increased dust source aridity. This is in line with continental climate reconstructions, whichpropose a strengthening of the Shamal [Glennie et al., 2002] and increased aridity on the Arabian Peninsuladuring colder climates [Preusser et al., 2002; Rosenberg et al., 2011]. The ITCZ-summer monsoon system mightnot only have decreased in intensity but might also have shifted southeastward during colder climate states,triggered by the increased strength of the northwesterly winds and a decreased strength of the southwesterlymonsoonal winds. Dune reconstructions from the northeastern Arabian Peninsula (Wahiba Sands) stand incontrast to a major shift of the ITCZ near the ground during the last glacial period [Preusser et al., 2002].
However, at higher tropospheric levels the strengthened northwesterly winds might have penetrated farther tothe east, leading to higher dust entrainment to the study site.
5.6. Heinrich Events Related Droughts in the Tropical Northern Hemisphere Heinrich events in the northeastern Arabian Sea stand out as especially prominent peaks of aeolian dustcontribution, which suggest a strong reduction of precipitation in the Indus River catchment area and DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
increased aridity on the Arabian Peninsula (Figure 3). For comparison, we analyzed a set of key paleoclimaterecords that is located at the northern rim of the modern boreal summer ITCZ position and therefore reactmost sensitively on a southward migration, contraction, and/or intensity decrease of the ITCZ monsoonal rainbelt (Figure 6).
Mulitza et al. [2008] used the element ratio of Fe/K to reconstruct relative contributions of atmospheric dustand fluvial suspension in marine sediments close to the mouth of the Senegal River (Figure 6c). They foundabrupt onsets of arid conditions in West African Sahel coinciding with Heinrich events. Hodell et al. [2008b]analyzed magnetic susceptibility in Lake Péten Itzá, Guatemala, which traces alternations between clay andgypsum deposition reflecting wet-dry cycles (Figure 6d). The most arid conditions were observed duringHeinrich events, which is in line with reconstructions from the Cariaco Basin [Peterson et al., 2000; Deplazeset al., 2013]. Wang et al. [2001, 2008] found evidence for reduced East Asian monsoon intensity coincidingwith Heinrich events by analyzing δ18O in stalagmites from Hulu and Sanbao Caves (Figure 6e).
Heinrich events are characterized by increased freshwater input and cold temperatures in the North Atlantic[Hodell et al., 2008a] (Figure 6a) and a reduction of the AMOC [McManus et al., 2004]. Climate simulationsindicate an ITCZ southward shift and a decrease in ITCZ summer monsoon intensity after a major reduction ofthe AMOC [Zhang and Delworth, 2005; Krebs and Timmermann, 2007]. Stager et al. [2011] have postulated thatHeinrich event 1 coincided with a catastrophic drought in the tropical Afro-Asian region and Schefuss et al.
[2011] have reported an ITCZ southward shift over southern Africa. Our study suggests together with a set ofkey paleoclimate records that the continental northern tropics around the globe were indeed impacted byincreased aridity, but not only during Heinrich event 1 but also during Heinrich events 6 to 2. This indicatesthat the hypothesized effect of Heinrich events on the tropical hydroclimate occurred under a variety ofbackground conditions (greenhouse gas concentrations, ice sheet extent, and insolation) during the lastglacial period.
Elemental and grain size analyses indicate independently a high sensitivity of the studied site to monsoonalclimate variability during the last glacial period. Our decadal- to centennial-scale Arabian Sea data suggest 1. The sedimentary depositional system at the continental slope off Pakistan seems to change with a "threshold" sea level of 70–80 m. Fluvial suspended load is mainly transported along the Indus submarinecanyon to the deep sea when sea level is lower than this threshold ( 68–14 kyr). During this time interval,inorganic properties of the hemipelagic sediments closely trace millennial-scale climate variability. The de-position center of sedimentation shifts from the deep sea toward the shelf and delta when the sea level ishigher than 70–80 m, which leads to reduced climate sensitivity of inorganic properties at the study site.
2. The fluvial versus aeolian sedimentary contribution varies during the last glacial period in concordance with the Dansgaard-Oeschger oscillations. Interstadials are characterized by a fluvial signature indicatingwarmer and more humid glacial conditions leading to enhanced runoff and transport of suspended loadby the Indus River to the ocean. Stadials reveal an increased aeolian contribution that at least partlyoriginates from the Arabian Peninsula. This leads to the conclusion that colder climates were charac-terized by decreased ITCZ-Indian summer monsoon intensity and by increased northwesterly windstrength and aridity over the Arabian Peninsula. Heinrich events 6 to 1 stand out as prominent, especiallydry and dusty events.
3. The abrupt and coeval change of inorganic, organic, oceanic, and atmospheric proxies at major climate transitions, e.g., the beginning of Interstadial 8, suggests a dominant role of the Indian summer monsoonon the oceanographic conditions and on subsequent sediment deposition in the Arabian Sea duringMIS 3. Productivity-related proxies of the Arabian Sea (L*, TOC) can therefore be interpreted in terms ofmonsoonal climate variability.
4. The close relationship between Indus River sediment and marine organic matter contributions during the last glacial period indicates that regional productivity was at least partially fed by fluvial nutrient injection.
5. Finally, this study together with other key paleoclimate records suggests increased continental aridity in the northern tropics during Heinrich events 6 to 1. Circum-North Atlantic temperature variations translateto hydrological shifts in the tropics, with major impacts on regional environmental conditions in themonsoonal world.
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
We are thankful to Frank Korte for Ackerman, S. A., and S. K. Cox (1989), Surface weather observations of atmospheric dust over the southwest summer monsoon region, conducting the XRF analyses. We ac- Meteorol. Atmos. Phys., 41(1), 19–34.
knowledge Wiebke Keil, Ursula Röhl, Altabet, M. A., M. J. Higginson, and D. W. Murray (2002), The effect of millennial-scale changes in Arabian Sea denitrification on atmospheric Georg Scheeder, and Thomas CO2, Nature, 415(6868), 159–162.
Westerhold for technical assistance. We Arz, H. W., F. Lamy, A. Ganopolski, N. Nowaczyk, and J. Pätzold (2007), Dominant Northern Hemisphere climate control over millennial-scale also thank A. Nele Meckler and Alfredo glacial sea-level variability, Quat. Sci. Rev., 26(3-4), 312–321.
Martínez-Garcia for valuable discus- Banakar, V. K., B. S. Mahesh, G. Burr, and A. R. Chodankar (2010), Climatology of the Eastern Arabian Sea during the last glacial cycle sions. We are thankful to S. Clemens for reconstructed from paired measurement of foraminiferal δ18O and Mg/Ca, Quat. Res., 73(3), 535–540.
his helpful suggestions to improve the Behl, R. J., and J. P. Kennett (1996), Brief interstadial events in the Santa Barbara basin, NE Pacific, during the past 60 kyr, Nature, 379(6562), manuscript. Mara Fant and Dorothée Husson were supported through a Bond, G., W. Broecker, S. Johnsen, J. McManus, L. Labeyrie, J. Jouzel, and G. Bonani (1993), Correlations between climate records from North MARUM Summer Student Fellowship.
Atlantic sediments and Greenland ice, Nature, 365(6442), 143–147.
We acknowledge the Bundesministerium Böning, P., and E. Bard (2009), Millennial/centennial-scale thermocline ventilation changes in the Indian Ocean as reflected by aragonite für Bildung und Forschung (BMBF, Bonn) preservation and geochemical variations in Arabian Sea sediments, Geochim. Cosmochim. Acta, 73(22), 6771–6788.
for funding the SO130 cruise (project Bookhagen, B., and D. W. Burbank (2010), Toward a complete Himalayan hydrological budget: Spatiotemporal distribution of snowmelt and rainfall and their impact on river discharge, J. Geophys. Res., 115, F03019, Boos, W. R., and Z. Kuang (2010), Dominant control of the South Asian monsoon by orographic insulation versus plateau heating, Nature, Broecker, W. S. (1994), Massive iceberg discharges as triggers for global climate change, Nature, 372(6505), 421–424.
Caley, T., B. Malaize, S. Zaragosi, L. Rossignol, J. Bourget, F. Eynaud, P. Martinez, J. Giraudeau, K. Charlier, and N. Ellouz-Zimmermann (2011), New Arabian Sea records help decipher orbital timing of Indo-Asian monsoon, Earth Planet. Sci. Rev., 308(3–4), 433–444.
Clemens, S. C. (1998), Dust response to seasonal atmospheric forcing: Proxy evaluation and calibration, Paleoceanography, 13(5), 471–490.
Clemens, S. C., and W. L. Prell (1990), Late Pleistocene variability of Arabian Sea summer monsoon winds and continental aridity: Eolian records from the lithogenic component of deep-sea sediments, Paleoceanography, 5(2), 109–145.
Clemens, S. C., and W. L. Prell (2003), A 350,000 year summer-monsoon multi-proxy stack from the Owen ridge, Northern Arabian sea, Mar.
Geol., 201(1–3), 35–51.
Clift, P. D., et al. (2008), Holocene erosion of the Lesser Himalaya triggered by intensified summer monsoon, Geology, 36(1), 79–82.
Cowie, G. L., S. Mowbray, M. Lewis, H. Matheson, and R. McKenzie (2009), Carbon and nitrogen elemental and stable isotopic compositions of surficial sediments from the Pakistan margin of the Arabian Sea, Deep Sea Res., Part II, 56(6–7), 271–282.
Dagg, M., R. Benner, S. Lohrenz, and D. Lawrence (2004), Transformation of dissolved and particulate materials on continental shelves influenced by large rivers: Plume processes, Cont. Shelf Res., 24(7–8), 833–858.
Dansgaard, W., et al. (1993), Evidence for general instability of past climate from a 250-kyr ice-core record, Nature, 364(6434), 218–220.
Deplazes, G., et al. (2013), Links between tropical rainfall and North Atlantic climate during the last glacial period, Nat. Geosci., 6(2), 213–217.
Dypvik, H., and N. B. Harris (2001), Geochemical facies analysis of fine-grained siliciclastics using Th/U, Zr/Rb and (Zr plus Rb)/Sr ratios, Chem.
Geol., 181(1–4), 131–146.
Findlater, J. (1969), A major low-level air current near the Indian Ocean during the northern summer, Q. J. R. Meteorol. Soc., 95(404), 362–380.
Fine, R. A., W. M. Smethie, J. L. Bullister, M. Rhein, D. H. Min, M. J. Warner, A. Poisson, and R. F. Weiss (2008), Decadal ventilation and mixing of Indian Ocean waters, Deep Sea Res., Part I, 55(1), 20–37.
Fleitmann, D., S. J. Burns, M. Mudelsee, U. Neff, J. Kramers, A. Mangini, and A. Matter (2003), Holocene forcing of the Indian monsoon recorded in a stalagmite from Southern Oman, Science, 300(5626), 1737–1739.
Ganopolski, A., and S. Rahmstorf (2001), Rapid changes of glacial climate simulated in a coupled climate model, Nature, 409(6817), 153–158.
Giosan, L., S. Constantinescu, P. D. Clift, A. R. Tabrez, M. Danish, and A. Inam (2006), Recent morphodynamics of the Indus delta shore and shelf, Cont. Shelf Res., 26(14), 1668–1684.
Glennie, K. W., A. K. Singhvi, N. Lancaster, and J. T. Teller (2002), Quaternary climatic changes over Southern Arabia and the Thar Desert, India, in Tectonic and Climatic Evolution of the Arabian Sea Region, edited by P. D. Clift et al., pp. 301–316, Geological Society, London.
Gupta, A. K., D. M. Anderson, and J. T. Overpeck (2003), Abrupt changes in the Asian southwest monsoon during the Holocene and their links to the North Atlantic Ocean, Nature, 421(6921), 354–357.
Heinrich, H. (1988), Origin and consequences of cyclic ice rafting in the Northeast Atlantic Ocean during the past 130,000 years, Quat. Res., 29(2), 142–152.
Hemming, S. R. (2004), Heinrich events: Massive Late Pleistocene detritus layers of the North Atlantic and their global climate imprint, Rev.
Geophys., 42, RG1005, Higginson, M. J., M. A. Altabet, D. W. Murray, R. W. Murray, and T. D. Herbert (2004), Geochemical evidence for abrupt changes in relative strength of the Arabian monsoons during a stadial/interstadial climate transition, Geochim. Cosmochim. Acta, 68(19), 3807–3826.
Hodell, D. A., J. E. T. Channell, J. H. Curtis, O. E. Romero, and U. Röhl (2008a), Onset of "Hudson Strait" Heinrich events in the eastern North Atlantic at the end of the middle Pleistocene transition (similar to 640 ka)?, Paleoceanography, 23, PA4218, Hodell, D. A., et al. (2008b), An 85-ka record of climate change in lowland Central America, Quat. Sci. Rev., 27(11-12), 1152–1165.
Honjo, S., J. Dymond, W. Prell, and V. Ittekkot (1999), Monsoon-controlled export fluxes to the interior of the Arabian Sea, Deep Sea Res., Part II, 46, 1859–1902.
Inam, A., P. D. Clift, L. Giosan, A. R. Tabrez, M. Tahir, M. M. Rabbani, and M. Danish (2007), The geographic, geological and oceanographic setting of the Indus River, in Large Rivers: Geomorphology and Management, edited by A. Gupta, pp. 333–345, John Wiley & Sons, Ltd., New York.
Ivanochko, T. S., R. S. Ganeshram, G. J. A. Brummer, G. Ganssen, S. J. A. Jung, S. G. Moreton, and D. Kroon (2005), Variations in tropical con- vection as an amplifier of global climate change at the millennial scale, Earth Planet. Sci. Rev., 235(1-2), 302–314.
Janowiak, J. E., and P. P. Xie (1999), CAMS-OPI: A global satellite-rain gauge merged product for real-time precipitation monitoring appli- cations, J. Clim., 12(11), 3335–3342.
Jung, S. J. A., D. Kroon, G. Ganssen, F. Peeters, and R. Ganeshram (2009), Enhanced Arabian Sea intermediate water flow during glacial North Atlantic cold phases, Earth Planet. Sci. Rev., 280(1-4), 220–228.
Kar, A., A. K. Singhvi, S. N. Rajaguru, N. Juyal, J. V. Thomas, D. Banerjee, and R. P. Dhir (2001), Reconstruction of the late Quaternary envi- ronment of the lower Luni plains, Thar Desert, India, J. Quat. Sci., 16(1), 61–68.
Klöcker, R., and R. Henrich (2006), Recent and Late Quaternary pteropod preservation on the Pakistan shelf and continental slope, Mar. Geol., DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
Krebs, U., and A. Timmermann (2007), Tropical air-sea interactions accelerate the recovery of the Atlantic Meridional Overturning Circulation after a major shutdown, J. Clim., 20(19), 4940–4956.
Krishnamurti, T. N., Y. Ramanathan, R. Pasch, and P. Greiman (1980), Quick look Summer MONEX atlasRep., Florida State University, Kudrass, H. R., A. Hofmann, H. Doose, K. Emeis, and H. Erlenkeuser (2001), Modulation and amplification of climatic changes in the Northern Hemisphere by the Indian summer monsoon during the past 80 k.y, Geology, 29(1), 63–66.
Kumar, S. P., M. Madhupratap, M. D. Kumar, M. Gauns, P. M. Muraleedharan, V. V. S. S. Sarma, and S. N. De Souza (2000), Physical control of primary productivity on a seasonal scale in central and eastern Arabian Sea, Proc. Indian Acad. Sci., Earth Planet. Sci., 109(4), 433–441.
Limmer, D. R., P. Böning, L. Giosan, C. Ponton, C. M. Köhler, M. J. Cooper, A. R. Tabrez, and P. D. Clift (2012), Geochemical record of Holocene to Recent sedimentation on the Western Indus continental shelf, Arabian Sea, Geochem. Geophys. Geosys., 13, Q01008, doi: Lisiecki, L. E., and M. E. Raymo (2005), A Pliocene-Pleistocene stack of 57 globally distributed benthic δ18O records, Paleoceanography, 20, Lückge, A., M. Ercegovac, H. Strauss, and R. Littke (1999), Early diagenetic alteration of organic matter by sulfate reduction in Quaternary sediments from the northeastern Arabian Sea, Mar. Geol., 158(1–4), 1–13.
Lückge, A., H. Doose-Rolinski, A. A. Khan, H. Schulz, and U. von Rad (2001), Monsoonal variability in the northeastern Arabian Sea during the past 5000 years: Geochemical evidence from laminated sediments, Palaeogeogr. Palaeoclimatol. Palaeoecol., 167(3–4), 273–286.
Lückge, A., L. Reinhardt, H. Andruleit, H. Doose-Rolinski, U. von Rad, H. Schulz, and U. Treppke (2002), Formation of varve-like laminae off Pakistan: Decoding 5 years of sedimentation, in The Tectonic and Climatic Evolution of the Arabian Sea Region, edited by P. D. Clift et al.,pp. 421–431, Geological Society, London.
Lückge, A., G. Deplazes, H. Schulz, G. Scheeder, A. Suckow, S. Kasten, and G. H. Haug (2012), Impact of Indus River discharge on productivity and preservation of organic carbon in the Arabian Sea over the twentieth century, Geology, 40(5), 399–402.
McManus, J. F., R. Francois, J. M. Gherardi, L. D. Keigwin, and S. Brown-Leger (2004), Collapse and rapid resumption of Atlantic meridional circulation linked to deglacial climate changes, Nature, 428(6985), 834–837.
Middleton, N. J. (1986), Dust storms in the Middle East, J. Arid Environ., 10(2), 83–96.
Milliman, J. D., G. S. Quraishee, and M. A. A. Beg (1984), Sediment discharge from the Indus River to the Ocean: Past, present, future, in Marine Geology and Oceanography of Arabian Sea and Coastal Pakistan, edited by P. Haq and J. D. Milliman, pp. 65–70, van Nostrand Reinhold,New York.
Mulitza, S., M. Prange, J. B. Stuut, M. Zabel, T. von Dobeneck, A. C. Itambi, J. Nizou, M. Schulz, and G. Wefer (2008), Sahel megadroughts triggered by glacial slowdowns of Atlantic meridional overturning, Paleoceanography, 23, PA4206, North Greenland Ice Core Project Members (2004), High-resolution record of Northern Hemisphere climate extending into the last inter- glacial period, Nature, 431(7005), 147–151.
Pedersen, T. F., G. B. Shimmield, and N. B. Price (1992), Lack of enhanced preservation of organic matter in sediments under the oxygen minimum on the Oman Margin, Geochim. Cosmochim. Acta, 56(1), 545–551.
Peterson, L. C., G. H. Haug, K. A. Hughen, and U. Röhl (2000), Rapid changes in the hydrologic cycle of the tropical Atlantic during the last glacial, Science, 290(5498), 1947–1951.
Pichevin, L., E. Bard, P. Martinez, and I. Billy (2007), Evidence of ventilation changes in the Arabian Sea during the late Quaternary: Implication for denitrification and nitrous oxide emission, Global Biogeochem. Cycles, 21, GB4008, Pourmand, A., F. Marcantonio, and H. Schulz (2004), Variations in productivity and eolian fluxes in the northeastern Arabian Sea during the past 110 ka, Earth Planet. Sci. Rev., 221(1–4), 39–54.
Preusser, F., D. Radies, and A. Matter (2002), A 160,000-year record of dune development and atmospheric circulation in southern Arabia, Science, 296(5575), 2018–2020.
Prins, M. A., and G. Postma (2000), Effects of climate, sea level, and tectonics unraveled for last deglaciation turbidite records of the Arabian Sea, Geology, 28(4), 375–378.
Prins, M. A., and G. J. Weltje (1999), End-member modeling of siliciclastic grain-size distributions: The Late Quaternary record of eolian and fluvial sediment supply to the Arabian Sea and its paleoclimatic significance, in Numerical Experiments in Stratigraphy: Recent Advances inStratigraphic and Sedimentologic Computer Simulations, edited by J. W. Harbaugh et al., pp. 91–111, Society for Sedimentary Geology,Tulsa, USA.
Prins, M. A., G. Postma, J. Cleveringa, A. Cramp, and N. H. Kenyon (2000), Controls on terrigenous sediment supply to the Arabian Sea during the late Quaternary: The Indus Fan, Mar. Geol., 169(3–4), 327–349.
Rea, D. K. (1994), The paleoclimate record provided by eolian deposition in the deep sea: the geologic history of wind, Rev. Geophys., 32(2), Reichart, G. J., L. J. Lourens, and W. J. Zachariasse (1998), Temporal variability in the northern Arabian Sea Oxygen Minimum Zone (OMZ) during the last 225,000 years, Paleoceanography, 13(6), 607–621.
Reichart, G. J., H. Brinkhuis, F. Huiskamp, and W. J. Zachariasse (2004), Hyperstratification following glacial overturning events in the northern Arabian Sea, Paleoceanography, 19, PA2013, Roberts, A. P., E. J. Rohling, K. M. Grant, J. C. Larrasoana, and Q. S. Liu (2011), Atmospheric dust variability from Arabia and China over the last 500,000 years, Quat. Sci. Rev., 30(25–26), 3537–3541.
Rosenberg, T. M., F. Preusser, D. Fleitmann, A. Schwalb, K. Penkman, T. W. Schmid, M. A. Al-Shanti, K. Kadi, and A. Matter (2011), Humid pe- riods in southern Arabia: Windows of opportunity for modern human dispersal, Geology, 39(12), 1115–1118.
Ruth, U., M. Bigler, R. Röthlisberger, M. L. Siggaard-Andersen, S. Kipfstuhl, K. Goto-Azuma, M. E. Hansson, S. J. Johnsen, H. Y. Lu, and J. P. Steffensen (2007), Ice core evidence for a very tight link between North Atlantic and east Asian glacial climate, Geophys. Res. Lett., 34,L03706, doi: Schefuss, E., H. Kuhlmann, G. Mollenhauer, M. Prange, and J. Pätzold (2011), Forcing of wet phases in southeast Africa over the past 17,000 years, Nature, 480(7378), 509–512.
Schmittner, A., E. D. Galbraith, S. W. Hostetler, T. F. Pedersen, and R. Zhang (2007), Large fluctuations of dissolved oxygen in the Indian and Pacific oceans during Dansgaard-Oeschger oscillations caused by variations of North Atlantic Deep Water subduction, Paleoceanography,22, PA3207, Schulte, S., and P. J. Müller (2001), Variations of sea surface temperature and primary productivity during Heinrich and Dansgaard-Oeschger events in the northeastern Arabian Sea, Geo-Mar. Lett., 21(3), 168–175.
Schulte, S., F. Rostek, E. Bard, J. Rullkötter, and O. Marchal (1999), Variations of oxygen-minimum and primary productivity recorded in sediments of the Arabian Sea, Earth Planet. Sci. Rev., 173(3), 205–221.
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.
Schulz, H., U. von Rad, and H. Erlenkeuser (1998), Correlation between Arabian Sea and Greenland climate oscillations of the past 110,000 years, Nature, 393(6680), 54–57.
Schulz, H., K. C. Emeis, H. Erlenkeuser, U. von Rad, and C. Rolf (2002), The Toba volcanic event and interstadial/stadial climates at the marine isotopic stage 5 to 4 transition in the northern Indian Ocean, Quat. Res., 57(1), 22–31.
Siddall, M., E. J. Rohling, W. G. Thompson, and C. Waelbroeck (2008), Marine Isotope Stage 3 sea level fluctuations: Data synthesis and new outlook, Rev. Geophys., 46, RG4003, .
Sinninghe Damsté, J. S., W. I. C. Rijpstra, and G. J. Reichart (2002), The influence of oxic degradation on the sedimentary biomarker record II.
Evidence from Arabian Sea sediments, Geochim. Cosmochim. Acta, 66(15), 2737–2754.
Sirocko, F., and H. Lange (1991), Clay-mineral accumulation rates in the Arabian Sea during late Quaternary, Mar. Geol., 97(1–2), 105–119.
Sirocko, F., and M. Sarnthein (1989), Wind-borne deposits in the northwestern Indian Ocean: Record of Holocene sediments versus modern satellite data, in Paleoclimatology and Paleometeorology: Modern and Past Patterns of Global Atmospheric Transport, edited by M. Leinenand M. Sarnthein, pp. 401–433, Kluwer Academic Publishers, Dordrecht.
Sirocko, F., M. Sarnthein, H. Erlenkeuser, H. Lange, M. Arnold, and J. C. Duplessy (1993), Century-scale events in monsoonal climate over the past 24,000 years, Nature, 364(6435), 322–324.
Sirocko, F., D. Garbe-Schönberg, and C. Devey (2000), Processes controlling trace element geochemistry of Arabian Sea sediments during the last 25,000 years, Global Planet. Change, 26(1–3), 217–303.
Stager, J. C., D. B. Ryves, B. M. Chase, and F. S. R. Pausata (2011), Catastrophic drought in the Afro-Asian monsoon region during Heinrich event 1, Science, 331(6022), 1299–1302.
Stanford, J. D., R. Hemingway, E. J. Rohling, P. G. Challenor, M. Medina-Elizalde, and A. J. Lester (2011), Sea-level probability for the last de- glaciation: A statistical analysis of far-field records, Global Planet. Change, 79(3-4), 193–203.
Storey, M., R. G. Roberts, and M. Saidin (2012), Astronomically calibrated Ar/ Ar age for the Toba supereruption and global synchronization of late Quaternary records, Proc. Natl. Acad. Sci. USA, 109(46), 18,684–18,688.
Stuut, J.-B. W., M. A. Prins, R. R. Schneider, G. J. Weltje, J. H. F. Jansen, and G. Postma (2002), A 300-kyr record of aridity and wind strength in southwestern Africa: Inferences from grain-size distributions of sediments on Walvis Ridge, SE Atlantic, Mar. Geol., 180(1–4), 221–233.
Sun, Y. B., S. C. Clemens, C. Morrill, X. P. Lin, X. L. Wang, and Z. S. An (2012), Influence of Atlantic meridional overturning circulation on the East Asian winter monsoon, Nat. Geosci., 5(1), 46–49.
Tyson, R. V. (2005), The "productivity versus preservation" controversy: Cause, flaws, and resolution, in The Deposition of Organic-Carbon-Rich Sediments: Models, Mechanisms and Consequences, edited by N. B. Harris, pp. 17–33, Society of Sedimentary Geology, Tulsa, USA.
van der Weijden, C. H., G. J. Reichart, and B. J. H. van Os (2006), Sedimentary trace element records over the last 200 kyr from within and below the northern Arabian Sea oxygen minimum zone, Mar. Geol., 231(1–4), 69–88.
von Rad, U., and M. Tahir (1997), Late Quaternary sedimentation on the outer Indus shelf and slope (Pakistan): Evidence from high-resolution seismic data and coring, Mar. Geol., 138(3–4), 193–236.
von Rad, U., H. Schulz, V. Riech, M. den Dulk, U. Berner, and F. Sirocko (1999), Multiple monsoon-controlled breakdown of oxygen-minimum conditions during the past 30,000 years documented in laminated sediments off Pakistan, Palaeogeogr. Palaeoclimatol. Palaeoecol.,152(1–2), 129–161.
von Rad, U., K.-P. Burgath, M. Pervaz, and H. Schulz (2002), Discovery of the Toba Ash (c. 70 ka) in a high-resolution core recovering millennial monsoonal variability off Pakistan, in The Tectonic and Climatic Evolution of the Arabian Sea Region, edited by P. D. Clift et al., pp. 445–461,Geological Society, London.
Wang, Y. J., H. Cheng, R. L. Edwards, Z. S. An, J. Y. Wu, C. C. Shen, and J. A. Dorale (2001), A high-resolution absolute-dated Late Pleistocene monsoon record from Hulu Cave, China, Science, 294(5550), 2345–2348.
Wang, Y. J., H. Cheng, R. L. Edwards, X. G. Kong, X. H. Shao, S. T. Chen, J. Y. Wu, X. Y. Jiang, X. F. Wang, and Z. S. An (2008), Millennial- and orbital-scale changes in the East Asian monsoon over the past 224,000 years, Nature, 451(7182), 1090–1093.
Webster, P. J., V. O. Magana, T. N. Palmer, J. Shukla, R. A. Tomas, M. Yanai, and T. Yasunari (1998), Monsoons: Processes, predictability, and the prospects for prediction, J. Geophys. Res., 103(C7), 14,451–14,510.
Weltje, G. J. (1997), End-member modeling of compositional data: Numerical-statistical algorithms for solving the explicit mixing problem, Math. Geol., 29(4), 503–549.
Wolff, E. W., J. Chappellaz, T. Blunier, S. O. Rasmussen, and A. Svensson (2010), Millennial-scale variability during the last glacial: The ice core record, Quat. Sci. Rev., 29, 2828–2838.
Wu, G., Y. Liu, B. He, Q. Bao, A. Duan, and F. F. Jin (2012), Thermal controls on the Asian summer monsoon, Sci. Rep., 2(404), Zhang, R., and T. L. Delworth (2005), Simulated tropical response to a substantial weakening of the Atlantic thermohaline circulation, J. Clim., Ziegler, M., L. J. Lourens, E. Tuenter, F. Hilgen, G. J. Reichart, and N. Weber (2010), Precession phasing offset between Indian summer mon- soon and Arabian Sea productivity linked to changes in Atlantic overturning circulation, Paleoceanography, 25, PA3213, .
DEPLAZES ET AL.
2014. American Geophysical Union. All Rights Reserved.

Source: http://www.stuut.tv/Deplazes_et_al_2014.pdf