Quaternary history of sea ice in the western Arctic Ocean based on foraminifera

Please download to get full document.

View again

of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
Information Report

Concepts & Trends


Views: 23 | Pages: 12

Extension: PDF | Download: 0

Sediment cores from the Northwind Ridge, western Arctic Ocean, including uniquely preserved calca- reous microfossils, provide the first continuous proxy record of sea ice in the Arctic Ocean encompassing more than half of the Quaternary. The cores
  Quaternary history of sea ice in the western Arctic Ocean based onforaminifera Leonid Polyak a , * , Kelly M. Best a , b , Kevin A. Crawford a , b , Edward A. Council c ,Guillaume St-Onge d , e a Byrd Polar Research Center, The Ohio State University, 1090 Carmack Road, Columbus, OH 43210, USA b School of Earth Sciences, The Ohio State University, Columbus, OH 43210, USA c Department of Geological Sciences, Wright State University, Dayton, OH 45435, USA d Institut des sciences de la mer de Rimouski (ISMER), Rimouski, Quebec G5L 3A1, Canada e GEOTOP Research Center, Quebec, Canada a r t i c l e i n f o  Article history: Received 27 July 2012Received in revised form20 December 2012Accepted 22 December 2012Available online 4 February 2013 Keywords: Western Arctic OceanForaminifersQuaternaryMid-Pleistocene TransitionSea ice history a b s t r a c t Sediment cores from the Northwind Ridge, western Arctic Ocean, including uniquely preserved calca-reous microfossils, provide the  󿬁 rst continuous proxy record of sea ice in the Arctic Ocean encompassingmore than half of the Quaternary. The cores were investigated for foraminiferal assemblages along withcoarse grain size and bulk chemical composition. By combination of glacial cycles and unique eventsre 󿬂 ected in the stratigraphy, the age of the foraminiferal record was estimated as ca 1.5 Ma. Foraminiferalabundances, diversity, and composition of benthic assemblages, especially phytodetritus and polarspecies, were used as proxies for sea-ice conditions. Foraminiferal Assemblage Zone 2 in the LowerPleistocene indicates diminished, mostly seasonal sea ice, probably facilitated by enhanced in 󿬂 ow of Paci 󿬁 c waters. A gradual decrease in ice-free season with episodes of abrupt ice expansion is interpretedfor the Mid-Pleistocene Transition, consistent with climatic cooling and ice-sheet growth in the NorthernHemisphere. A principal faunal and sedimentary turnover occurred near the Early e Middle Pleistoceneboundary ca 0.75 Ma, with mostly perennial sea ice indicated by the overlying Assemblage Zone 1. Twosteps of further increase in sea-ice coverage are inferred from foraminiferal assemblage changes in the “ Glacial ”  Pleistocene by ca 0.4 and 0.24 Ma, possibly related to hemispheric (Mid-Brunhes Event) andLaurentide ice sheet growth, respectively. These results suggest that year-round ice in the western Arcticwas a norm for the last several 100 ka, in contrast to rapidly disappearing summer ice today.   2013 Elsevier Ltd. All rights reserved. 1. Introduction As global climate conditions continue to shift toward a warmerplanet, the Arctic Ocean is becoming increasingly vulnerable towarming and its associated effects, due to a set of positive feed-backs regarded as Arctic Ampli 󿬁 cation (Serreze and Barry, 2011).Sea ice is an integral factor that determines the magnitude of thesefeedbacksincludingthealbedoinsummerandinsulationinwinter.Knowledge of paleo-ice conditions is essential for understandingthe trajectory of rapid retreat of sea ice in the Arctic (e.g., Stroeveet al., 2011) and related climatic and hydrographic changes affect-ing the global thermohaline circulation (Rashid et al., 2011).Reconstruction of sea-ice extent in the Arctic is complicated bythe lack of any one unequivocal proxy, and is further complicatedby very low sedimentation rates and biogenic content in the areasof high sea-ice coverage (e.g., Polyak et al., 2010). Not surprisingly,reconstructions of paleo sea ice are being developed for marginal-ice areas such as the Fram Strait (Müller et al., 2009, 2012; Bonnet et al., 2010; Spielhagen et al., 2011), but not for the central parts of  the Arctic Ocean.This paper aims to reconstruct Quaternary sea-ice conditions inthe western Arctic based on a foraminiferal record from theNorthwindRidge,westernArcticOcean(Fig.1),wheremodernsea-ice retreat is especially pronounced (Stroeve et al., 2011). Unlikemost sediment cores from the Arctic Ocean, where calcareous ma-terialispreservedonlyintheLateand,partially,MiddlePleistocene(e.g., Jakobsson et al., 2001; Spielhagen et al., 2004; Polyak et al., 2009; Stein et al., 2010), the record under study has abundant cal- careousmicrofossilsgoingbackintotheEarlyPleistocene,estimatedca 1.5 Ma. Pronounced changes in benthic foraminiferal assem-blages,alongwithlithostratigraphicproxies,allowforthe 󿬁 rst-timecharacterization of the Quaternary history of sea ice in the westernArctic Ocean, the most ice-covered oceanic region of the world. *  Corresponding author. Fax:  þ 1 614 2924697. E-mail address:  polyak.1@osu.edu (L. Polyak). Contents lists available at SciVerse ScienceDirect Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev 0277-3791/$  e  see front matter    2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.quascirev.2012.12.018 Quaternary Science Reviews 79 (2013) 145 e 156  2. Paleoceanographic context A cyclic character of lithological, geochemical, and paleobio-logical proxy changes has been identi 󿬁 ed from recent detailedstudies of sediment cores from various parts of the Arctic Ocean(e.g., Jakobsson et al., 2000; Spielhagen et al., 2004; O ’ Regan et al.,2008; Adler et al., 2009; Polyak et al., 2009; Stein et al., 2010). Although not all proxies are completely deciphered, the generalinterpretation recognizes that these changes have been caused byalternation of full-glacial, deglacial (iceberg dominated), andinterglacial/major interstadial environments of the Late to MiddleQuaternary. Glacial e interglacial contrasts are most pronounced inthe hydrographically more isolated western Arctic Ocean due to itsremoteness from the deep-water connection with the Atlantic andthe gyre-type surface circulation (Beaufort Gyre) (Polyak and Jakobsson, 2011). Early Pleistocene to pre-Quaternary history of the Arctic Ocean has been partially investigated only in the ACEXdeep borehole (Fig.1) representing the Transpolar Drift-dominatedenvironments and suffering from poor preservation of paleonto-logical remains (Backman et al., 2006; O ’ Regan et al., 2008).Despite a generally robust identi 󿬁 cation of glacial e interglacialcyclicity, changes in associated sea-ice conditions in the ArcticOcean remain mostly unassessed. A notable exception is an ele-vated occurrence of subpolar planktonic foraminifers at somestratigraphic intervals including the estimated Last Interglacial(Nørgaard-Pedersen et al., 2007; Adleret al., 2009). Although more understanding is needed for the distribution of these foraminifers(local blooms vs. long-distance transport by currents, selectivepreservation, etc) and for the accurate age of these events, theyclearly show the potential of paleobiological proxies for repre-senting sea-ice conditions. In this study we capitalize on thestratigraphically longest calcareous foraminiferal record recoveredthusfarfromtheArcticOceantogaininsightsintosea-icehistoryinthewesternArctic.Weespeciallyfocusonbenthicforaminifersthathave been thus far underutilized as paleo-sea ice proxies due topreservationandcountingsizeissues(e.g.,Croninetal.,2008;Scott et al., 2008) and perennial sea ice predominance in recent Arctichistory,whichcomplicatesidenti 󿬁 cationofproxiesrelatedtolowerice extent. 3. Materials and methods Piston cores 92AR-P39 and 93AR-P23 (hereafter referred to asP39 and P23) were collected on the 1992 and 1993 U.S. GeologicalSurvey P1 cruises from the Northwind Ridge extending from theChukchi Sea margin to the interior of the western Arctic Ocean(Fig.1; Table 1). Both core sites are bathed by the Upper Polar Deep Water,withshallowerP23sitebeingclosetothelowerboundaryof the Arctic Atlantic Water (Rudels, 2009). The summer sea-icemargin was located south of the ridge in climatological data, butshifted to its northern edge in recent years (e.g., Stroeve et al.,2011), which makes the Northwind Ridge an area of choice forstudying the history of sea ice in the western Arctic.Two samples from core P23 investigated by Mullen and McNeil(1995) contained calcareous benthic foraminifers that had simi-larity to pre-Quaternary (Pliocene to Late Miocene) fauna from theBeaufort e MackenzieBasinofArcticCanada.However,stratigraphicand paleoceanographic context for this core remained unin-vestigated. To test the earlier assessment of foraminiferal assem-blages and utilize them for evaluating paleo-environments, wehave performed a detailed study of core P23 along with P39 col-lectedfarthersouthontheNorthwindRidge(Fig.1).Duetoitsmoresouthernlocation,P39representsoveralllowerlong-termicecoverresulting in higher sedimentation rates and, thus, a higher-resolution record (Polyak et al., 2009; Crawford, 2010). Difference inwaterdepths mayalso contribute tovarying sedimentation ratesas P23 site at the ridge top is more likely affected by currents.Upon collection, cores have been stored in a refrigerated facility(USGSMenloPark)and,otherthanpartiallydryingout,remainedina very good condition. Due to a negligible amount of labile organicmatter and strong oxic conditions on the central Arctic Ocean  󿬂 oor(e.g., Stein and Macdonald, 2004), carbonaceous material in thesesedimentsisminimallyaffectedbydiageneticdissolutioncausedbypost-collection oxidation (unlike sediments from the continentalmargins).Coreshavebeensampledonseveraloccasionsforvariousanalyses including discrete and u-channel samples. Earlier litho-and magnetostratigraphic results on a series of cores from thewesternArcticOceanincludingP23andP39havebeendescribedinPolyaketal.(2009).TodetailearlierbulkchemicalcompositionXRFmeasurements spaced at 2 e 6 cm, u-channels from core P23 wereanalyzed with the 0.5 cm resolution on the Itrax XRF scanner atthe INRS-ETE (Quebec City). Foraminiferal counts along withcoarse ( > 63  m m) grain content measurements were done at theByrd Polar Research Center, with samples mostly taken at 2-cmintervals from subsections cut along the core length. Total plank-tonicandcalcareousbenthicforaminifers > 150 m mwerecountedinalmostallsamplescollected(Fig.2;Suppl.1),withsparserintervals in the unfossiliferous lower part of P39. Detailed counts of benthicforaminifers in 63 e 125 and > 125  m m size fractions were primarily Fig. 1.  Index map with location of cores P23 and P39 (red circles) and other sitesdiscussed in the paper (yellow circles). Arrows show Transpolar Drift and BeaufortGyre circulation. Pink line  e  climatological late-20th century summer ice extent (15%concentration), white line  e  2007 summer ice extent (historical minimum). Yellowboxes  e  sites of earlier paleo-sea-ice studies based on benthic foraminifers. NR, MR,and LR   e  Northwind, Mendeleev, and Lomonosov ridges, respectively. FS and BS  e Fram and Bering straits. BM  e  Beaufort e Mackenzie area. GIN seas  e  Greenland e Ice-land e Norwegian seas.  Table 1 Sediment core information.Core ## Latitude N Longitude W Waterdepth (m)Corelength(cm)Corediameter(cm)P1-92AR-P39 75  50.7 0 156  01.9 0 1470 687 8P1-93AR-P23 76  57.3 0 155  03.9 0 951 572 8 L. Polyak et al. / Quaternary Science Reviews 79 (2013) 145 e 156  146  done in core P23, with P39 added for an enhanced characterizationof the uppermost stratigraphy that is strongly compressed andpartially missing in P23 (Suppl. 2 e 3). Detailed counts in P23 wereperformed at mostly 2-cm intervals in the upper 35 cm, at 4 cmbetween 35 and 90 cm, at 8 cm between 90 and 180 cm, andat w 15 cm in the rest of the core; in P39 similar counts were doneintheuppermost200cmatevery 4cm.Thissamplingstrategywasaimed at characterizing the Early to Middle Pleistocene transitionand the  “ Glacial ”  Pleistocene above in most detail, as discussedbelow (Fig. 3).Foraminiferal abundances were calculated per gram of drysedimentexcludinglargepebbles.Usingsamplevolumesinsteadof weightswouldhaveintroducedmorebiasassedimenthaspartiallydried out during storage. All samples had a similar volume of  w 6 e 7 cm 3 . Identi 󿬁 cation of benthic foraminifers in small size fractions(atleast > 63 m m)isessentialforacomprehensivecharacterizationof the assemblages, especially in polar areas (e.g., Lagoe, 1977; Scott and Vilks,1991), while > 125  m m counts are helpful for a compari-son with many other studies. Taxonomic identi 󿬁 cations werecomparedtomultiplesourceswithafocusonArcticstudies(Green,1960;Lagoe,1977;ScottandVilks,1991;Wollenburg,1992;Ishman and Foley, 1996; McNeil, 1997; Scott et al., 2008) (Suppl. 4). Fisher a  and Shannon e Wiener  H(S)  diversity indices (Figs. 3 and 4) were calculated using the PAST (PAleontology STatistics) program(Hammer et al., 2001). A preliminary study of ostracodes has alsobeenperformed by T. Cronin inpilot samples from P23. 4. Foraminiferal proxy approach Benthic foraminifers were divided into three proxy groupsbased on their ecological preferences. One comprises taxa that arecommonly considered to be Arctic or bipolar endemics (e.g., Green,1960; Herman,1973; Lagoe,1977) predominated in our material by Stetsonia horvathi  and  Bolivina arctica , and also including  Valvur-ineria arctica ,  Epistominella arctica ,  Buliminella elegantissima hen-soni ,  Eponides tumidulus horvathi , and a few accessorial species.Someof thesetaxawerealsoreportedfromSubantarctic(Corneliusand Gooday, 2004) and Quaternary deposits in the North Atlantic Fig. 2.  Correlation of P23 and P39: content of sand and coarser grains ( > 63  m m), XRF Mn and Ca (bulk sediment; Itrax data for P23, hand-held scanner data for P39), abundances of planktonic (PF) and calcareous benthic (BCF) foraminifers per gram. Blue and pink correlation lines  e  major inclination drop and the top of predominantly brown sediment,respectively; diamond e B. aculeata  peak; asterisk e warm-water planktonic peak; triangle e prominent IRD peak. Interglacial Marine Isotope Stages to MIS21 (most apparent) areshown to the right. L. Polyak et al. / Quaternary Science Reviews 79 (2013) 145 e 156   147  (Pawlowski, 1991; Collins et al., 1996) co-occurring with either modern presence of sea ice or its extension to lower latitudesduring glacial periods. One notable exception is the occurrence of  Stetsonia arctica  ( S. horvathi  of most authors) in the deposits of theBengal Fan (Scott and Leger, 1990), which indicates that the ulti-matecontrolonthisspeciesis probablynotseaice,butlowcontentof freshfood in organic matter inputs to sea 󿬂 oor. Nevertheless, seaice is one of the major factors in modulating this type of benthicenvironment(e.g.,SteinandMacdonald,2004),andtheconsistencyin co-occurrence of sea-ice dominated settings with this and rela-ted species suggests their strong relation to ice-covered waters.Theother foraminiferalgroupthatisakeyforreconstructingiceconditions comprises the so-called phytodetritus species, notably Epistominella exigua  and  Eponides weddellensis . Distribution of these species peaks in the frontal oceanic zones, where conditionsare suitable for pulsed, seasonal production and export of freshorganic matter (phytodetritus) utilized by fast-response benthicmeiofauna (e.g., Gooday, 1988, 1993; Smart et al., 1994; Thomas et al., 1995). The Marginal Ice Zone (MIZ) in the Arctic is oneexample of such a frontal zone, as demonstrated by direct obser-vations including long-term deep-sea stations (e.g., Falk-Petersenet al., 2000; Hoste et al., 2007). Foraminiferal composition from Fig. 3.  Major characteristics of foraminiferal distribution in P23 (polar and phytodetritus species content, total abundances in  > 63 and  > 150 mm size fractions, Fisher  a  diversityindex, planktonic/benthic ratio) along with sand ( > 63  m m) content. Ranges of extinct species are shown on top. Vertical lines show inferred events of sea-ice expansion; yellow linemarks the turnover between foraminiferal Assemblage Zones 2 and 1. Interglacial Marine Isotope Stages to MIS21 are shown above the sand curve. >63 m 02468100 1 2 3H(S)        F       i     s       h     e     r  AZ1 AZ2 >125 m 02468100 1 2 3H(S)        F       i     s       h     e     r  AZ1 AZ2 Fig. 4.  Plots of faunal diversity indices (Fisher  a  vs.  H(S) ) in > 63  m m and > 125  m m size fractions in P23. Data points from Assemblage Zones 1 and 2 (Fig. 3) are shown by trianglesand circles, respectively. Diversity patterns differ between the assemblage zones, especially in the  > 63  m m size fraction, as emphasized by regression lines (solid for AZ1 andpunctured for AZ2). L. Polyak et al. / Quaternary Science Reviews 79 (2013) 145 e 156  148  the Arctic MIZ has not been investigated in detail due to its stronglocalization and widespread dissolution of carbonaceous remainsin bottom waters under seasonal ice cover (Steinsund and Hald,1994; Husum and Hald, 2012). Nevertheless, phytodetritus spe- cies in modern/sur 󿬁 cial Arctic sediments has been reported onlyfrom areas adjacent to the MIZ (Polyak, 1990; Hald and Steinsund, 1992; Wollenburg and Mackensen, 1998). A more complete pic- ture is likely more complex as some of the polar species may becapable of feeding on phytodetritus (Cornelius and Gooday, 2004)andmayoccupy transitional habitatsbetweenice-coveredand MIZenvironments, e.g.,  Epistominella arctica  (Pawlowski, 1991;Wollenburg and Mackensen,1998; Cornelius and Gooday, 2004). Some cosmopolitan species that are common in the ArcticOcean (e.g.,  Cibicidoides wuellerstor   󿬁 , Oridorsalis tener, Cassidulinateretis  ( neoteretis )) may also have speci 󿬁 c adaptations to habitatsunder the ice cover, but a wide distribution of these species com-plicates their use for reconstructing ice conditions. In particular, C. teretis  is common throughout the Arctic Oceanwith an apparentaf  󿬁 nity to Atlantic-derived waters at depths of   w 200 e 900 m(Lagoe, 1979; Polyak, 1990; Ishman and Foley, 1996), and is there- fore commonly used as an indicator of this water mass in paleorecords (e.g., Jennings and Weiner, 1996; Lubinski et al., 2001). However, the actual controls on the Arctic habitats of this speciesmay be related to food supply and need more in-depth inves-tigation (Wollenburg and Mackensen,1998). 5. Results 5.1. Lithology and general stratigraphy Both cores consist of interlaminating grayish and brown bandswith layers of coarse (sand or coarser) detritus (to w 145 cm in P23and 422 cm in P39) and predominantly brown, bioturbated, overall 󿬁 ner-grainedsedimentbelow(especiallyconsistentfrom170cminP23) (Fig. 2; Crawford, 2010). The brown coloration in Arctic Ocean sediments is largely controlled by the content of manganese oxy-hydroxides ( Jakobsson et al., 2000; Polyak et al., 2004; März et al., 2011), and is accordingly well approximated by the distribution of Mn (Fig. 2). Multiple studies suggest that high-Mn sedimentaryunitsrepresentinterglacialor majorinterstadialintervals, probablyinrelationtohigherinputsfromthemarginsduetohighersealeveland lower sea-ice extent (e.g., Jakobsson et al., 2000; O ’ Regan et al.,2008; Adler et al., 2009; Löwemark et al., 2012). Co-occurrence with bioturbation further con 󿬁 rms more productive conditions,which require less extensive ice cover (Löwemark et al., 2012). Incontrast, grayish units, commonly with high content of coarsegrains (Ice Rafted Debris, IRD) are indicative of glacial/deglacialenvironments. Some IRD peaks have characteristically high com-position of detrital carbonates approximated by Ca content (Fig. 2;Clark et al., 1980; Polyak et al., 2009; Stein et al., 2010). These car- bonates, mostly composed of dolomites, are associated with theCanadian Arctic provenance and thus indicate pulses of icebergdischarge from the Laurentide ice sheet. In cores with elevatedsedimentation rates, IRD layers are distinguished from  󿬁 ne-grained, gray sediments representing maximal glaciations (e.g.,Adler et al., 2009; Polyak et al., 2009), but in a compressed record such as core P23 these lithologies merge together as generalizedIRD-rich glacial intervals.DowncoredistributionofMn,Ca,andcoarsegrainpeaks,almostentirely composed of IRD in the upper unit, provides a robustlithostratigraphic correlation of cores P23 and P39, further rein-forced by the position of unique litho-, bio-, and magnetostrati-graphic events (Fig. 2). Consistent with higher sedimentation ratesat a more southern site, core P39 has a more expanded but,accordingly, shorter stratigraphy. A combination of the two coresallows for an identi 󿬁 cation of more details in the upper part of therecord as well as an evaluation of the longer stratigraphy. Evenmore detail for the uppermost stratigraphy can be obtainedthrough correlationwith the Northwind Ridge cores such as 92AR-P25 (P25 in Fig. 1) (Crawford, 2010; Yurco et al., 2010) and closely located 88AR-P3 and P5 (Poore et al., 1994), collected yet farthersouth.Calcareous foraminiferal abundance patterns correlate well be-tween the Northwind Ridge cores, exempli 󿬁 ed by P23 and P39, inthe upper part of the record (above pink correlation line in Fig. 2).However, further down-core P39 is practically barren of any cal-careous skeletal remains, similar to almost all known Arctic Oceancores (e.g., Cronin et al., 2008); whereas, core P23 contains wellpreserved benthic and planktonic foraminifers and ostracodes inthe lower part of the record. Regardless of the causes for thispreservation,itprovidesarareopportunitytogaininsightsintotheearlier Quaternary Arctic oceanic environments. Similar micro-faunal distribution has been found in HOTRAX ’ 05 core HLY0503-03JPC (Darby et al., 2009) raised close to P23 from 590-m waterdepth. This core has not yet been investigated in detail, but a pre-liminary evaluation indicates the consistency of P23 bio-stratigraphy for the top of the Northwind Ridge. 5.2. Age model Constraining the age of Arctic Ocean sediments has been, andcontinues to be dif  󿬁 cult. Various chronostratigraphic methods thatworkwellinotheroceansencounterproblemsintheArcticbecauseof very low sedimentation rates, low biological production, andstrong oxygenation of bottom sediments that causes diageneticalteration of biogenic material and even magnetic carriers(Channell and Xuan, 2009; Xuan and Channell, 2010; Xuan et al., 2012). However, the explicitly cyclic nature of Quaternary ArcticOcean sediments allows for a comparison with the global glacial e interglacial cyclicity, which produced meaningful results on sev-eral cores with relatively high resolution ( Jakobsson et al., 2000;Adler et al., 2009; Stein et al., 2010) and has been veri 󿬁 ed againsta longer-term stratigraphy in the ACEX record (O ’ Regan et al.,2008). We have applied this approach to the Northwind RidgecoresbycountingIRD-richgrayishlayersandMn-richbrownlayersas glacial and interglacial intervals, respectively. Resultant stratig-raphy is veri 󿬁 ed by several unique events (Fig. 2):(1) Peak occurrence of benthic foraminifer  Bulimina aculeata (Ishman et al., 1996) at the level estimated as MIS5a in therevised stratigraphy ( Jakobsson et al., 2001; Nørgaard- Pedersen et al., 2007; Adler et al., 2009); (2) Prominent drop in paleomagnetic inclination that consistentlyoccurs in Arctic Ocean cores at the level estimated as MIS7( Jakobsson et al., 2000; Spielhagen et al., 2004; O ’ Regan et al.,2008; Polyak et al., 2009); (3) Peak of planktonic foraminifers almost entirely composed of a temperate-water species  Turborotalita egelida  (possiblyasubspeciesof  T.quinqueloba )atastratigraphiclevelestimatedas MIS11, which makes perfect sense for such an anomalouslywarm-water fauna (more discussion in Cronin et al., 2013);(4) Prominent IRD peak with the  󿬁 rst noticeable increase indetrital carbonates signifying the onset of massive icebergdischarges from the Laurentide ice sheet (Polyak et al., 2009;Steinetal.,2010;PolyakandJakobsson,2011).Thelikelytiming of this event is MIS16, which was a pivotal point in thedevelopmentof100-kacyclicity(e.g.,Clarketal.,2006)andthebeginningof Laurentide iceberg pulses (Heinrich events) in theNorth Atlantic (Hodell et al., 2008). L. Polyak et al. / Quaternary Science Reviews 79 (2013) 145 e 156   149
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!