Early Eocene to middle Miocene cooling and aridification of East Antarctica

transcript

Article

Volume 14, Number 5

6 May 2013

doi:10.1002/ggge.20106

ISSN: 1525-2027

Early Eocene to middle Miocene coolingand aridification of East Antarctica

S. PasschierDepartment of Earth and Environmental Studies, Montclair State University, 1 Normal Avenue,Montclair, New Jersey, 07043, USA (passchiers@mail.montclair.edu)

S. M. BohatyOcean and Earth Science, National Oceanography Centre, University of Southampton, Southampton,UK

F. Jiménez-EspejoInstituto Andaluz de Ciencias de la Terra, Granada, Spain

Institute of Biogeosciences, Japan Agency for Marine-Earth Science and Technology,Yokosuka, Japan

J. ProssPaleoenvironmental Dynamics Group, Institute of Geosciences, Goethe University Frankfurt,Frankfurt, Germany

U. RöhlMARUM-Center for Marine Environmental Sciences, University of Bremen, Bremen, Germany

T. van de FlierdtDepartment of Earth Science and Engineering, Imperial College London, London, UK

C. EscutiaInstituto Andaluz de Ciencias de la Terra, Granada, Spain

H. BrinkhuisLaboratory of Palaeobotany and Palynology, Department of Earth Sciences, Faculty of Geosciences,Utrecht University, Utrecht, Netherlands

[1] Few high-latitude terrestrial records document the timing and nature of the Cenozoic “Greenhouse” to“Icehouse” transition. Here we exploit the bulk geochemistry of marine siliciclastic sediments from drill coreson Antarctica’s continental margin to extract a unique semiquantitative temperature and precipitation record forEocene to mid-Miocene (~54–13Ma). Alkaline elements are strongly enriched in the detrital mineral fraction infine-grained siliciclastic marine sediments and only occur as trace metals in the biogenic fraction. Hence,terrestrial climofunctions similar to the chemical index of alteration (CIA) can be applied to the alkaline majorelement geochemistry of marine sediments on continental margins in order to reconstruct changes in precipitationand temperature. We validate this approach by comparison with published paleotemperature and precipitationrecords derived from fossil wood, leaves, and pollen and find remarkable agreement, despite uncertainties inthe calibrations of the different proxies. A long-term cooling on the order of ≥8�C is observed between theEarly Eocene Climatic Optimum (~54–52 Ma) and the middle Miocene (~15–13 Ma) with the onset of

transient cooling episodes in the middle Eocene at ~46–45 Ma. High-latitude stratigraphic records currentlyexhibit insufficient temporal resolution to reconstruct continental aridity and inferred ice-sheet developmentduring the middle to late Eocene (~45–37 Ma). However, we find an abrupt aridification of East Antarcticanear the Eocene-Oligocene transition (~34 Ma), which suggests that ice coverage influenced high-latitudeatmospheric circulation patterns through albedo effects from the earliest Oligocene onward.

Components: 7,400 words, 5 figures.

Keywords: Antarctica; Cenozoic; greenhouse; icehouse; chemical index of alteration;Integrated Ocean Drilling Program.

Index Terms: 1039 Geochemistry: Alteration and weathering processes (3617); 1065 Geochemistry: Major andtrace element geochemistry; 1051 Geochemistry: Sedimentary geochemistry; 0726 Cryosphere: Ice sheets; 0790Cryosphere: Weathering (1625, 1886).

Received 8 November 2012; Revised 20 February 2013; Accepted 2 March 2013; Published 6 May 2013.

Passchier, S., S. M. Bohaty, F. Jiménez-Espejo, J. Pross, U. Röhl, T. van de Flierdt, C. Escutia, and H. Brinkhuis(2013), Early Eocene to middle Miocene cooling and aridification of East Antarctica, Geochem. Geophys. Geosyst.,14, 1399–1410, doi:10.1002/ggge.20106.

1. Introduction

[2] Reconstruction of the Cenozoic continentalenvironmental history of Antarctica is importantto understand the role of high-latitude physicaland biogeochemical processes in the global oceanand climate system. Due to extensive ice coverand erosion of sedimentary archives, obtainingterrestrial paleo-environmental records acrossmajor climate transitions has been challenging. Wepresent new bulk geochemical data from IntegratedOcean Drilling Program (IODP) Site U1356 [Escutiaet al., 2011], Ocean Drilling Program (ODP) Site1166 [O’Brien et al., 2001], and McMurdo ErraticD1 [Harwood and Levy, 2000], along with existingdata from the Cape Roberts Project (CRP) drill cores[Barrett, 2007; Passchier and Krissek, 2008] and theCIROS-1 drill core (Figure 1) [Roser and Pyne,1989]. We apply terrestrial climofunctions derivedfrom chemical weathering indices for semiquantita-tive paleoclimatological interpretations, an approachpioneered by Nesbitt and Young [1982].

[3] The average exposed continental crust containsmore than 60% feldspars by volume, with subordinatecontributions from volcanic glass and other labileminerals [Nesbitt and Young, 1982]. Chemicalweathering by meteoric and soil waters containingdissolved carbonic and organic acids transforms alka-line aluminosilicates to clayminerals, thereby deplet-ing minerals in cations, such as Na+, K+, and Ca2+,while conserving Al3+. Numerous recent studies ofefflux from watersheds suggest a strong imprint ofclimate on the chemical weathering in soils [e.g.,

White et al., 1999; Turner et al., 2010; Rasmussenet al., 2011; Eiriksdottir et al., 2013]. In addition,studies of geochemical mass balance and cosmo-genic nuclides show that the degree to which bulksoil profiles become depleted in labile elements isstrongly dependent on the prevailing climate[Sheldon et al., 2002; Riebe et al., 2004; Nordt andDriese, 2010]. Sheldon et al. [2002] demonstratedby analysis of a database of 126 North Americansoils [Marbut, 1935] that different molar oxide ratiosof K, Na, and Ca versus Al in soil weatheringhorizons anticorrelate with mean annual temperature(MAT) and mean annual precipitation (MAP).Sheldon et al. [2002] proposed two climofunctions,which provided paleotemperature and precipitationestimates for paleosols comparable to those acquiredvia other independent proxies.

[4] We assume that the detrital geochemistry of themudrocks in the continental margin successionssemiquantitatively preserves the average paleo-environmental signature of the soil mantle in thesource drainage basins (Figure 1). Physical erosionin continental uplands provides a fine-grainedsediment source from a broad spectrum of sourcerock lithologies. In glacially fed sediment transportsystems, specifically, silt-sized glacial rock flour istransported to the coastal zone as suspended loadvia wind and proglacial meltwater streams. Thedetrital geochemistry of continental margin sedi-ments adjacent to both nonglaciated and glaciatedcatchments primarily records the climatic condi-tions in the vegetated coastal zone, where silicateweathering rates are high. Here the sediment resides

GeochemistryGeophysicsGeosystemsG3G3 PASSCHIER ET AL.: EOCENE-MIOCENE PALEOCLIMATE ANTARCTICA 10.1002/ggge.20106

stabilized in floodplains, immediately prior to thefinal phase of erosion, transport, and offshore depo-sition [Anderson, 2007].

[5] Silicate chemical weathering in uplands andperiglacial environments is kinetically limited dueto the high erosion rates and abundant supply ofsilt-sized feldspars with a large specific surface area[Anderson, 2005]. The rate of leaching of cations isinitially rapid but diminishes exponentially within afew thousand years as mineral surfaces age [Egli et al.,2001; Anderson, 2005]. In contrast to transport-limitedweathering where changes in the fluid saturation statecomplicate the climate-weathering relationship, kinet-ically limited weathering is regulated, spatially andtemporally, by temperature, precipitation (runoff),and vegetation [West et al., 2005; Anderson, 2005,2007; Eiriksdottir et al., 2013].

[6] Building upon the above considerations, we applyterrestrial climofunctions based on the North Ameri-can soils database [Marbut, 1935] to siliciclastic sedi-ments on the Antarctic continental margin. Weacknowledge that uncertainties may arise from re-gional differences between North America and Ant-arctica but argue that the application of terrestrialclimofunctions is justified by the fact that both EastAntarctica and North America represent petrologicallydiverse high-latitude cratonic regions periodically cov-ered by ice sheets. Moreover, in combination withpublished vegetation data (Table S1 in the supporting

information)1 from several Antarctic sites [Pross et al.,2012; Francis et al., 2007, 2009 and referencestherein], this approach provides a unique opportunityto reconstruct Antarctic continental paleotemperatureand precipitation dynamics from the peak of the earlyEocene “Greenhouse” to the middle Miocene“Icehouse” (~54–13 Ma).

2. Methods and Materials

[7] The CRP [Barrett, 2007] and ODP Site 1166[O’Brien et al., 2001] cores were recovered fromthe Antarctic continental shelf in the Ross Sea andPrydz Bay, respectively, whereas IODP Site U1356was drilled on the Wilkes Land continental rise(Figure 1) [Escutia et al., 2011]. As documentingAntarctic continental conditions was a primaryobjective for these expeditions, the drill sites werepositioned into seaward prograding sediment wedgesthat received material eroded from East Antarcticsources [Cooper et al., 1991]. The sample ofMcMurdo Erratic D1 is part of a collection of erraticsdocumenting the otherwise poorly sampled middle tolate Eocene Antarctic glacial history [Harwood andLevy, 2000].

[8] The major element chemistry of bulk mudrocksamples from the Antarctic drill cores was deter-mined using XRF and inductively coupled plasmaatomic emission spectrometry (ICP-AES) instru-mentation. We also incorporated data from stratarecovered in the CIROS-1 drill hole in McMurdoSound [Roser and Pyne, 1989] into our compila-tion, but a lack of carbonate data precluded applica-tion of the precipitation climofunction. Age modelsfor the drill cores were based on Tauxe et al. [2012]for IODP Hole U1356A, Lavelle [1998], McIntosh[1998], and Florindo et al. [2005] for the CRP drillcores, Florindo et al. [2003] for ODP Hole 1166A,and on Harwood et al. [1989], Rieck [1989], andWilson et al. [1998] for the CIROS-1 drill core.The middle Eocene age assessment for McMurdoErratic D1 is derived from Levy and Harwood[2000]. All ages were converted to Gradstein etal. [2004] (Tables S2 and S4, supporting informa-tion). It should be noted however that the nannofossildatums used to calibrate the magnetostratigraphy forCIROS-1 [Wilson et al., 1998] are controversial[Watkins, 2007]. Published bulk chemical data arealso available for the middle to upper Eocene ofthe La Meseta Formation on the northern AntarcticPeninsula [Dingle et al., 1998]. These sediments,

CIROS-1CRP 1,2,3

10000-1000-2000 2000-3000-4000-5000

120°E

30°E 0°60°S 30°W

120°W

150°W 180°E 150°E

Figure 1. Location of ODP Site 1166, IODP SiteU1356, CRP 1, 2, 3, and CIROS-1 drill holes projectedonto reconstructed topography for Antarctica at theEocene-Oligocene boundary [Wilson et al., 2012].

1All supporting information may be found in the online version ofthis article.

however, contain more than 40% sand [Dingle et al.,1998]; hence, interpretations of CIA-based weather-ing ratios derived from whole-rock geochemistryare not possible.

[9] Most of the new ICP-AES data for IODP SiteU1356 were collected on board the JOIDESResolution. Additional Miocene samples from IODPSite U1356 and all samples from ODP Site 1166were analyzed in the ICP laboratory at MontclairState University. The sample preparation on boardthe JOIDES Resolution followed identical protocolsto those applied in the ICP laboratory at MontclairState University [Murray et al., 2000]. In brief, theprocedures involve fusion of 0.1 g of sample mixedwith 0.4 g LiBO4 flux in a furnace and dissolutionof the pellets in 7% HNO3, followed by filtrationand a dilution prior to analysis. Elemental concentra-tions were calculated using 10 or more U.S. Geolog-ical Survey standards. The analytical uncertaintiesfor oxide weight percentages in repeat runs andmonitor standards are typically 1–5%.

[10] The bulk major element geochemistry ofmarine mudrocks reflects provenance, weatheringhistory, and post-depositional alteration or diagenesisand deposition of salt [McLennan et al., 1993]. Insediments with limited diagenetic alteration and saltdeposition, the relative abundances of labilemineral-bound alkali and alkaline Earth elementsare controlled by paleoclimatic conditions [Nesbittet al., 1996] which affect the intensity of chemicalweathering in the source areas [Riebe et al., 2004].The margin architecture [Cooper et al., 1991]and high sedimentation rates on the order of 10–100m/Myr [Florindo et al., 2003, 2005; Tauxe et al.,2012] indicate that the dominant sediment flux isderived from nearby Antarctic continental sources.In addition to directly routed terrigenous sediments,the marine sediments investigated here containsubordinate proportions of wind-blown dust;marine biogenic components, such as calcareousforaminifera and nannofossils; siliceous microfos-sils; and salt. In sediments with a large terrigenousfraction, corrections can be made for the presenceof carbonate [Nesbitt and Young, 1982]. Further-more, Al, K, and Na occur only in trace metalconcentrations in the biogenic fraction [Breweret al., 2008]. Hence, in our analysis and interpreta-tion of the bulk chemical data, we considered thefollowing:

1. The elements of interest to the climofunctions aremajor elements in most rock-forming mineralsand occur in trace metal concentration in other(biogenic) components of marine sediments

[Brewer et al., 2008]. As a result, the weightpercentages of these elements in averageupper crust [Wedepohl, 1995] are between 11and 34 times higher than the weight percentagesin a purified diatomite. For the samples in thisstudy, we estimated from smear slides that theterrigenous fraction contributes between 30%and 100% to the overall sediment composition.Following the above cited proportions of alkalineelements between biogenic and detrital phases,this means that even in the most diatom-richsamples, 79–94% of the alkaline earth elementsin the bulk sediment will be contributed fromthe detrital fraction.

2. Kryc et al. [2003] demonstrated that scavenging ofAl is negligible for sediment with >3–5 wt % ter-rigenous matter, which applies to the cores studiedhere, where the terrigenous fraction is estimated tobe between 30% and 100% and the remaining frac-tion primarily consists of biogenic silica [Escutiaet al., 2011]. Given the low density of biogenicsilica, the lower value of 30% terrigenous fractionis a conservative estimate and it likely correspondsto >30% weight percent terrigenous fraction.

3. Carbonate content was routinely measured forCRP, ODP, and IODP samples [Dietrich andKlosa, 1998; Dietrich et al., 2000, 2001;Shipboard Scientific Party, 2001; Escutia et al.,2011]. Most samples yielded<2 wt % carbonate,and corrections for CaO present in calcium car-bonate were applied [Nesbitt and Young, 1982].A few samples, including McMurdo Erratic sam-ple D1, contained >10 wt % carbonate and wereeliminated from the calculations of CIA-K to avoidthe introduction of errors related to extensive car-bonate corrections. Samples with abundant car-bonate cement in the CIROS-1 drill hole [Bridleand Robinson, 1989] were also eliminated. No car-bonate correction was applied to the samples fromSite 1166, because the CaO concentrations were<1 wt % and carbonate corrections yielded nega-tive values. Because no carbonate wt % data areavailable for the CIROS-1 drill hole, no CIA-Kand MAP calculations are presented.

4. The contribution of Na from salts is negligible insediments with a large siliciclastic fraction.Krissek and Kyle [2000] showed that Na valuesand major element ratios for washed andunwashed samples taken less than a meter apartin the CRP cores were indistinguishable.

5. Dissolution of feldspar and authigenic clay for-mation, due to the geothermal heat flux affectingporewater temperatures, is generally negligiblefor burial depths shallower than approximately1000–1500 m. However, in the case of cores

in the Ross Sea, active magmatism during thepast 25 Myr [Martin et al., 2010] may haveincreased porewater temperature conditions atshallower depth. Therefore, core sections werecarefully chosen and sections with documentedevidence for aluminosilicate diagenesis, such asthe base of CRP-3 below 340 m below seafloor[Ehrmann et al., 2005], were eliminated fromthe analysis [cf. Passchier and Krissek, 2008].

6. Samples from the CIROS-1 drill hole wereselected based on particle size (more than60% mud), and samples with Al2O3/TiO2 ratios<18 were eliminated to avoid a volcanic over-print [cf. Passchier and Krissek, 2008].

[11] Following the approach of Sheldon et al. [2002],applying their climate data set and the soil chemistrydatabase of Marbut [1935], we constructed slightlymodified climofunctions. The soils of the Marbut[1935] database encompass a wide range of topo-graphic settings, relief, and source rock lithologiesand are covered by a variety of vegetation types [Shel-don et al., 2002]. Therefore, climofunctions developedfrom this database potentially findwidespread applica-tion on continents with similar physiographic attri-butes as North America. Using the climate data fromSheldon et al. [2002] and the major element geochem-istry data for soil B horizons in the Marbut [1935]database, we created linear least squares regressionequations forMAP andMAT (Figure 2). The equationfor MAT is identical to the climofunction used inSheldon et al. [2002]. For MAP, we eliminated datapoints for pedons with calcrete horizons (with carbon-ate nodules) as our goal was to address the depletion ofCaO in the silicate fraction.

[12] The climofunctions we applied to the bulkmajor element geochemistry are

MAT ¼ 18:5 S½ � þ 17:3 (1)

in which the S index is represented by the molarratio of Na2O and K2O to Al2O3 [Sheldon et al.,

2002]. The applied climofunction for MAP wasslightly modified using only the data points forpedons without calcrete horizons, to emphasizethe silicate mineral-bound components:

MAP ¼ 143:75e0:0232 CIA�K½ � (2)

in which CIA (chemical index of alteration)[Nesbitt and Young, 1982], represents 100 timesthe molar ratio of Al2O3 to Na2O,CaO*, K2O andAl2O3, and CIA-K, the same equation withoutK2O [Maynard, 1992]. CaO* designates CaO indetrital silicate minerals only. Although the regres-sion coefficients are small for the MAT data, thecorrelations for both sets of variables are significantwith p < 0.01. Using the regression equations toestimate MAT and MAP from S and CIA-K,respectively, we calculate the standard error of theprediction on the y axis (MAT and MAP) usingthe STEYX function in Excel [2007]. The standarderrors were �3.6�C and �182 mm for MAT andMAP, respectively.

[13] The relatively large standard errors for MAT andMAP originate from physiographic effects other thantemperature and precipitation, such as topography,variable mineral-specific weathering rates due tosource rock variability, vegetation, and soil pH. Re-cent field studies, however, confirm the strong cli-matic imprint on the chemical depletion of soils,despite uncertainties in the effects of these other vari-ables [e.g., Turner et al., 2010; Ferrier et al., 2010;Nordt and Driese, 2010, Rasmussen et al., 2011].We assume that there is negligible chemical alter-ation of detrital sediment components duringsuspended load transport of mud from coastal sourceto continental margin sink [e.g., Nesbitt et al., 1996].

3. Results

[14] On a CaO* + Na2O � Al2O3 � K2O (CN-A-K)ternary diagram (Figure 3), the samples from Sites

y= 143.75e0.0232x

R²=0.7102

y= -18.516x+17.298R²= 0.3711

0 20 40 60 80 100 120 0 0.2 0.4 0.6 0.8

Figure 2. Least squares regression plots for (left) MAT and (right) MAP using data from soil B horizons[Sheldon et al., 2002].

1166 and U1356 and the CRP drill cores plot inapproximate alignment with the CN-A axis, whichis consistent with differentiation of detrital materialsvia chemical weathering from an intermediatecrystalline source rock or a mixture of source rocksof average upper crustal composition [Nesbitt andYoung, 1984, 1989]. The alignment of plotted datapoints with the CN-A axis (Figure 3) also illustratesthat differences in source rock composition didnot significantly impact the alkaline elementgeochemistry. If variability in the abundance ofCa- and Na-bearing plagioclase and K-feldsparwas governed by source rock changes, thesample compositions would align parallel to theCN-K axis.

[15] Pronounced changes in alkaline elementgeochemistry are noted between the Eocene“Greenhouse” and Miocene “Icehouse” sediments(Figure 4). Values of CIA-K for the Antarctic drillholes decrease from an average of ~86 during the

10 15 20 25 30 35 40 45 50 5530

0010 15 20 25 30 35 40 45 50 55

Miocene Oligocene Eocene

Age (Ma)

CRPU1356

1166McMurdoErra c D1

CIROS-1

CRP U13561166

AVG upper crust

Figure 4. (a) Mean annual temperature (MAT) and (b) mean annual precipitation (MAP) calculated from the detritalgeochemistry in Antarctic continental margin drillcores using climofunctions based on theMarbut [1935] soil chemistrydatabase. Values for average (AVG) upper crust were calculated from Wedepohl [1995]. Gray shadings designateconfidence intervals calculated from the standard errors. The smoothing trendline is derived using the LOWESS(Locally Weighted Scatterplot Smoothing) algorithm [Cleveland, 1979].

CRP1166U1356

Figure 3. Ternary plots of CaO* + Na2O � Al2O3 �K2O (CN-A-K) system [Nesbitt and Young, 1984, 1989].The distribution of the data points is consistent with lossof alkalinity through chemical weathering.

Early Eocene Climatic Optimum to ~63 duringthe middle Miocene “Icehouse,” whereas values forthe S index increase from~0.24 to ~0.60. The calculatedterrestrial MAT ranges from ≲4 to 14�C and MAPsrange from ~400 to 1200 mm (Figure 4). Themaximum values are below the upper limits ofMAT > 18�C or MAP > 1600 mm that have beensuggested for complete leaching of labile mineralsfrom soils [Sheldon et al., 2002; Rasmussen et al.,2011]. Corresponding to an absence of significantfeldspar weathering, lower limits can be establishedfrom values for average upper crust (Figure 4)[Wedepohl, 1995], which yield MAT < 7.7�C andMAP < 460 mm in our calibrations. In portions ofthe Oligocene and Miocene record, CIA-K andS index values plot below average upper crustalcompositions (Figure 4), and therefore, the calcu-lated MAPs and MATs should be regarded asmaximum values. Because of a lack of carbonatedata for the CIROS-1 drill hole and very highcarbonate values for D1 [Passchier, 2000], we wereunable to calculate CaO*, and hence CIA-K, forthese records.

[16] An abrupt decrease in humidity from1000–1250 mm to 500–800 mm is apparent at theEocene-Oligocene transition (Figure 4b). In theOligocene, regional differences emerge betweensites with higher MAP values at Site U1356 offWilkes Land (~750 mm) than at the CRP sites(~600 mm) in the Ross Sea (Figure 4b). The MATsaverage 14�C for the early and middle Eocene withsome variability in the late Eocene and steadilydecline through the Oligocene and Miocene(Figure 4a). The data density in the mid to lateEocene portion of the record is low due to poorrepresentation of this stratigraphic interval. Therecords from the CRP sites, CIROS-1, and IODPSite U1356 partially overlap in time and also inpaleotemperature trends (Figure 4a). Overall, along-term cooling on the order of ≥8�C isinferred over a ~41 Myr time interval betweenthe Early Eocene Climatic Optimum and themiddle Miocene.

4. Discussion

4.1. Sediment Recycling

[17] Thiry [2000] points out that inheritance ofancient kaolinite complicates the paleoclimatologicalinterpretation of clay minerals in marine sediments.However, the methodology we use here is consider-ably different as we address the bulk geochemistry

of mudrocks, which include a significant fine tocoarse silt fraction. The methodology based on bulksediment weathering ratios, pioneered by Nesbittand Young [1982], considers the degree to whichthe feldspars in the mudrocks are altered. The centralidea is that survival of fresh feldspar is related to theprevailing climate, with more silt-sized feldsparconverted to clay minerals under warm and humidconditions, and survival of abundant feldspar duringcold and dry conditions. As glaciers produce glacialrock flour through erosion of crystalline cratonicrocks, a steady supply of silt is maintained, whichdepending on the paleoclimatic conditions, will beaffected by different degrees of chemical weatheringin the coastal zone immediately before transport anddeposition in the marine environment.

[18] Although erosion of weathered sedimentaryplatform rocks can affect the composition of glacialsediments directly overlying these platform rocks,the effects are small in offshore sediment cores, asdemonstrated by the CIA values of Quaternarydiamicts from Antarctic drill cores, which are 47–56[Passchier and Krissek, 2008; this study, andunpublished data], compared to 46 for average uppercrust [Wedepohl, 1995]. Kaolinite reworking hencemay result in a maximum of approximately 20%decrease in alkalinity of the bulk sediment, translat-ing into a less than 20% increase in temperatureand humidity.

4.2. Comparison With Vegetation-BasedClimate Reconstructions

[19] The high MAP of >1000 mm calculated forthe Eocene at Sites U1356 and 1166 is in closeagreement with the MAP for subtropical andtemperate rain forest biomes on East Antarctica asreconstructed from pollen records in the same cores[Pross et al., 2012], and with plant macrofossilassemblages in Eocene strata near the AntarcticPeninsula and on subantarctic islands [Franciset al., 2007, 2009; and references therein]. A decreasein humidity is apparent from the early Oligoceneonward. An estimated MAP of 500–800 mm forthe late Oligocene and early Miocene is higher thanthe modern MAP of 150–400 mm for an Antarcticcontinent without vegetation. MAP levels <460mm, however, are below the applicable limit of theclimofunction (Figure 4b).

[20] In agreement with our results, temperaturesderived from plant macrofossils on the AntarcticPeninsula, on subantarctic islands, and in the RossSea are relatively warm in the early and middleEocene,withMATof 7–15�C (Table S1 andFigure 5).

Leaf fossils and pollen fromOligocene-Miocene drillcores in the Ross Sea suggest the presence of a low-diversity Nothofagus-dominated forest and shrub-land [Askin and Raine, 2000], typical of modernconditions along the southern tip of South Americawith MAT of 5–8�C [Prebble et al., 2006; Franciset al., 2009], giving way to an impoverishedshrub tundra and mossy tundra vegetation in themid-Miocene [Lewis et al., 2008].

[21] A discrepancy between vegetation-derivedtemperatures and the MAT derived from the soilclimofunction, however, is apparent in the lateEocene at 37–35 Ma (Figure 5). The major elementdata for the upper Eocene are from the CIROS-1drill hole (Figure 4a) for which the age model iscontroversial, whereas the vegetation data are fromsubantarctic islands (Francis et al., 2009) and Hole1166A in Prydz Bay [Macphail and Truswell,2004; Truswell and Macphail, 2009]. Wilson et al.[1998] argue for a late Eocene cooling episodeand onset of glaciation. Watkins [2007], however,based on a detailed reanalysis of the nannofossilbiostratigraphy prefers a significantly youngerlatest Eocene to early Oligocene age for the lowerportion of the CIROS-1 drill hole. Unless the agemodel issues are resolved, the terrestrial paleoclimaterecord for the late Eocene remains controversial.

4.3. Climatic Implications

[22] Despite a lack of evidence for significant icegrowth and aridification, the temperature recordsderived from both vegetation reconstructions and

weathering indices may indicate transient Antarcticcooling episodes beginning in the mid-Eocene~46–45 Ma, a time of global cooling, alpine glacia-tion, and sea ice formation in the Arctic [Stickleyet al., 2009]. Although evidence for Eocene glacia-tion in Antarctica and/or elsewhere in the SouthernHemisphere has been invoked by studies of surfacetextures of sand-sized quartz grains in high-latitudedrill cores [Margolis and Kennett, 1970; Strand et al.,2003], our MAP record for the Antarctic coast impliesthat the ice cover, if present, was likely restricted tosmall-scale mountain glaciers and ice caps duringthe early to mid-Eocene (~55–45 Ma). The strati-graphic coverage and age control of the middle to lateEocene records in Antarctic drill cores, however, arepoor and preclude any definitive conclusions.

[23] The onset of continental-scale ice growth onAntarctica at ~34 Ma is relatively well establishedfrom deep-sea proxy records, glacigenic sediments,and ice sheet modeling [Hambrey et al., 1991;Coxall et al., 2005; DeConto et al., 2008]. The pro-nounced aridification we observe near the Eocene-Oligocene boundary (Figure 4) is in agreement withthe growth of an ice sheet and its influence on at-mospheric circulation patterns: The size of the icesheet determines the strength of the polar high-pressure cell through an albedo feedback andpromotes the onset of strong northward directedcold and dry katabatic winds. Following thisreasoning, our record supports an absence of largecontinental ice sheets in Antarctica in early to mid-Eocene time [DeConto et al., 2008], as conditionsremained relatively humid. Although we do not have

Miocene Oligocene Eocene

Age (Ma)

10 15 20 25 30 35 40 45 50 55

T (vegeta on)

Figure 5. Comparison of MAT based on the soil climofunction and T (vegetation), which is MAT derived frompublished vegetation reconstructions [Askin and Raine, 2000; Birkenmajer and Zastawniak, 1989; Francis and Poole,2002; Francis et al., 2007, 2009; Lewis et al., 2008; Macphail and Truswell, 2004; Poole et al., 2005; Prebble et al.,2006; Pross et al., 2012; Raine, 1998; Truswell and Macphail, 2009; Warny et al., 2009]. Error bars, where reported,indicate range based on standard errors for T (vegetation). The gray lines indicate the confidence intervals for theMAT based on the soil climofunction (see Figure 4a).

a continuous record across the Eocene-Oligoceneboundary, the ~3�C drop in temperature observedin the running average of our data set (Figure 4b) iscomparable to a 3–4�C drop in sea surface tempera-tures of the nearest high latitude ODP Site 277[Liu et al., 2009].

[24] Midlatitude terrestrial paleoclimate recordsshow similar aridification and cooling trends acrossthe Eocene-Oligocene transition and of the samemagnitude although there is considerable geographicvariability. Paleoprecipitation records from thewest-central United States are geographically diverseand records from central Oregon show significantaridification, whereas the effects are smaller inthe dryer areas of the continental interior of theUnited States due to dryer conditions in the Eocene[Sheldon and Retallack, 2004]. In contrast, terrestrialpaleotemperature records show a significant drop inthe midcontinent area of the United States [Zanazziet al., 2007], whereas a gradual temperature changeis observed in the more maritime regions of centralOregon [Sheldon and Retallack, 2004], Argentina[Kohn et al., 2004], and the Isle of Wight [Sheldonet al., 2009].

[25] In Antarctica, regional differences in humidityobserved between the Oligocene records of theWilkes Land (U1356) and the Ross Sea (CRP)coasts (Figure 4) are in agreement with the resultsfrom a regional climate model for a 2�CO2

scenario [DeConto et al., 2007]. Themodel, however,predicts year-round subfreezing coastal temperaturesfor a fully glaciated continent, in contrast to temper-atures reconstructed from vegetation [Thorn andDeConto, 2006] and our geochemical records.The results of the proxies may be biased towardinterglacial conditions when plant biomass produc-tion reached a maximum, and because the soilgeochemical climofunctions are calibrated to peakinterglacial MATs. Due to Antarctica’s strong sea-sonality, there is possibly also a bias toward thesummer season. During portions of the Oligoceneand Miocene, however, S index values are below av-erage upper crustal compositions and the calculatedMATs should be regarded as maximum temperatures(MAT < 7.7�C). Hence, model results for a fullyglaciated continent are consistent with our proxyevidence for the earliest Oligocene and portions ofthe early to mid-Miocene.

5. Conclusions

[26] Geochemical signatures of marine sediments onthe Antarctic continental margin together with

vegetation reconstructions carry unique semiquanti-tative evidence of continental paleoclimatic condi-tions. The strong partitioning of alkaline elementsin the detrital fraction of marine sediments allows ap-plication of soil climofunctions, which gauge alka-line element depletion in relation to mean annualtemperature and precipitation in the source area.There is a strong agreement between chemical andvegetation proxies for Eocene-Miocene times in thisstudy, indicating ≥8�C cooling from the early Eoceneto middle Miocene. Although our data compare wellwith results from other proxy records and modeling,indicating the onset of continental-scale ice growth atthe Eocene-Oligocene transition, further drilling onthe Antarctic continental margin will be necessaryto reconstruct continental aridity and inferred ice devel-opment during the mid to late Eocene (~45–37 Ma).

Acknowledgments

[27] Nathan Sheldon kindly provided the data for soil Bhorizons of Marbut [1935] and the climatological informationfor the pedons. Samples were provided by the Integrated OceanDrilling Program. Expedition 318 Scientific Participants arethanked for shipboard collaboration in the data collection andpost-cruise discussions. A list of participants and affiliationscan be found here: http://iodp.tamu.edu/scienceops/precruise/wilkesland/participants.html. Shipboard data collection alsobenefited from the technical assistance of Chieh Peng, DavidHoupt, and Bob Olivas. SP acknowledges funding forshipboard participation and post-cruise research from theConsortium for Ocean Leadership (IUSSP410-T318A72),SMB and TvdF from NERC UK IODP, FJJE from ResearchGroup 0179 (Junta de Andalucía) and Spanish MEC projectCTM2011-24079, and JP and UR from the DeutscheForschungsgemeinschaft (DFG). Four anonymous reviewersprovided suggestions for improvement to a previous versionof the manuscript.

References

Anderson, S. P. (2005), Glaciers show direct linkage betweenerosion rate and chemical weathering fluxes, Geomorphology,67, 147–167, doi:10.1016/j.geomorph.2004.07.010.

Anderson, S. P. (2007), Biogeochemistry of glacial landscapesystems, Ann. Rev. Earth Planet. Sci., 35, 375–399,doi:10.1146/annurev.earth.35.031306.140033.

Askin, R. A., and J. I. Raine (2000), Oligocene and earlyMiocene terrestrial palynology of the Cape Roberts drill holeCRP-2/2A, Victoria Land Basin, Antarctica, Terra Antart., 7,493–501.

Barrett, P. J. (2007), Cenozoic climate and sea level historyfrom glacimarine strata off the Victoria Land coast, CapeRoberts Project, Antarctica, in Glacial Processes andProducts, edited by M. J. Hambrey, P. Christoffersen, N. F.Glasser, and B. Hubbart, International Association of Sedimen-tologists, Special Publication, 39, 259–287, doi:10.1002/9781444304435.ch15.

Brewer, T. S., M. J. Leng, A.W.Mackay, A. L. Lamb, J. J. Tyler,and N. G. Marsh (2008), Unravelling contamination signalsin biogenic silica oxygen isotope composition: The role ofmajor and trace element geochemistry, J. Quat. Sci., 23,321–330.

Birkenmajer, K., and E. Zastawniak (1989), Late Cretaceous-early Tertiary floras of King George Island, inWest Antarctica:Their Stratigraphic Distribution and Palaeoclimatic Signifi-cance, Geological Society, London, Special Publications, 47,227–240, doi:10.1144/GSL.SP.1989.047.01.17.

Bridle, I. M., and P. H. Robinson (1989), Diagenesis, inAntarctic Cenozoic History From the CIROS-1 Drill Hole,McMurdo Sound, edited by P. Barrett, DSIR Bulletin, 245,pp. 201–207, DSIR Publishing, Wellington.

Cleveland, W. S. (1979), Robust locally weighted fitting andsmoothing scatterplots, J. Am. Stat. Assoc., 74, 829–836.

Cooper, A. K., P. J. Barrett, K. Hinz, V. Traube, G. Leitchenkov,and H. M. J. Stagg (1991), Cenozoic prograding sequences ofthe Antarctic continental margin: A record of glacio-eustaticand tectonic events, Mar. Geol., 102, 175–213.

Coxall, H. K., P. A. Wilson, H. Pälike, C. H. Lear, andJ. Backman (2005), Rapid stepwise onset of Antarctic glaci-ation and deeper calcite compensation in the Pacific Ocean,Nature, 433, 53–57, doi:10.1038/nature03135.

DeConto, R., D. Pollard, and D. Harwood (2007), Sea icefeedback and Cenozoic evolution of Antarctic climate andice sheets, Paleoceanography, 22, PA3214, doi:10.1029/2006PA001350.

DeConto, R. M., D. Pollard, P. Wilson, H. Pälike, C. Lear, andM. Pagani (2008), Thresholds for Cenozoic bipolar glaciation,Nature, 455, 653–656. doi:10.1038/nature07337.

Dietrich, H. G., and D. Klosa (1998), Carbonate contents inCRP-1 samples—Initial results, Terra Antart., 5, 645–646.

Dietrich, H. G., D. Klosa, and C. Wittich (2000), CarbonateContents in CRP-2/2A, Victoria Land, Antarctica, TerraAntart., vol. 7, pp. 389–391.

Dietrich, H. G., D. Klosa, and C. Wittich (2001), CarbonateContent in CRP-3 Drill Core, Victoria Land Basin, Antarctica,Terra Antart., vol. 8, pp. 299–302 .

Dingle, R. V., A. Marenssi, and M. Lavelle (1998), Highlatitude Eocene climate deterioration: Evidence from thenorthern Antarctic Peninsula, J. South Am. Earth Sci., 11(6),571–579, doi:10.1016/S0895-9811(98)00035-2.

Egli, M., P. Fitze, and A. Mirabella (2001), Weathering andevolution of soils formed on granitic, glacial deposits:Results from chronosequences of Swiss alpine environments,Catena, 45, 19–47. doi:10.1016/S0341-8162(01)00138-2.

Ehrmann, W., M. Setti, and L. Marinoni (2005), Clay mineralsin Cenozoic sediments off Cape Roberts (McMurdo Sound,Antarctica) reveal palaeoclimatic history, Palaeogeogr.Palaeoclimatol. Palaeoecol., 229, 187–211, doi:10.1016/j.palaeo.2005.06.022.

Eiriksdottir, E. S., S. R. Gislason, and E. H. Oelkers (2013), Doestemperature or runoff control the feedback between chemicaldenudation and climate? Insights from NE Iceland, Geochim.Cosmochim. Acta, 107, 65–81. doi:10.1016/j.gca.2012.12.034.

Escutia, C., H. Brinkhuis, A. Klaus, and the Expedition 318Scientists (2011), Proc. IODP, 318: Tokyo (Integrated OceanDrilling Program Management International, Inc.), doi:10.2204/iodp.proc.318.2011.

Excel (2007), Microsoft Excel. Microsoft, 2007, ComputerSoftware, Redmond, WA.

Ferrier, K. L., J. W. Kirchner, C. S. Riebe, and R. C. Finkel(2010), Mineral-specific chemical weathering rates over millen-nial timescales: Measurements at Rio Icacos, Puerto Rico,Chem. Geol., 277, 101–114. doi:10.1016/j.chemgeo.2010.07.013.

Florindo, F., S. M. Bohaty, P. S. Erwin, C. Richter,A. P. Roberts, P. A. Whalen, and J. M. Whitehead (2003),Magnetobiostratigraphic chronology and palaeoenvironmentalhistory of Cenozoic sequences from ODP sites 1165 and 1166,Prydz Bay, Antarctica, Palaeogeogr. Palaeoclimatol.Palaeocol., 198, 69–100, doi:10.1016/S0031-0182(03)00395-X.

Florindo, F., G. S. Wilson, A. P. Roberts, L. Sagnotti, andK. L. Verosub (2005), Magnetostratigraphic chronology of alate Eocene to early Miocene glacimarine succession from theVictoria Land Basin, Ross Sea, Antarctica, Global Planet.Change, 45, 207–236. doi:10.1016/j.gloplacha.2004.09.009.

Francis, J. E., and I. Poole (2002), Cretaceous and earlyTertiary climates of Antarctica: Evidence from fossil wood,Palaeogeogr. Palaeoclimatol. Palaeoecol., 182, 47–64.doi:10.1016/S0031-0182(01)00452-7.

Francis, J. E., et al. (2009), From greenhouse to icehouse—TheEocene/Oligocene in Antarctica, Develop. Earth Environ.Sci., 8, 309–368, doi:10.1016/S1571-9197(08)00008-6.

Francis, J., A. Ashworth, D. Cantrill, J. Crame, J. Howe,R. Stephens, A. Tosolini, and V. Thorn (2007), 100 millionyears of Antarctic climate evolution: evidence from fossilplants, in Antarctica: A Keystone in a Changing World,edited by A. K. Cooper, P. J. Barrett, H. Stagg, B. Storey,E. Stump, and W. Wise, pp.19–27, The National AcademiesPress, Washington D.C., doi:10.3133/of2007-1047.kp03.

Gradstein, F. M., et al. (2004), A Geologic Time Scale 2004,Cambridge University Press, Cambridge, UK.

Hambrey, M. J., W. U. Ehrmann, and B. Larsen (1991),Cenozoic glacial record of the Prydz Bay continental shelf,East Antarctica, in J. Barron, B. Larsen, et al., Proc. ODP,Sci. Results, 119, College Station TX (Ocean DrillingProgram), pp. 77–132, doi:10.2973/odp.proc.sr.119.200.1991.

Harwood, D. M., and R. H. Levy (2000), Introduction andoverview, in Paleobiology and Paleoenvironments of Eo-cene Fossiliferous Erratics, 76, pp. 1–18, edited by J. D.Stilwell, and R. M. Feldmann, McMurdo Sound, EastAntarctica, Antarctic Research Series, doi:10.1029/AR076p0001.

Harwood, D. M., P. J. Barrett, A. R. Edwards, H. J. Rieck, andP. N. Webb (1989), Biostratigraphy and chronology, in AntarcticCenozoic History From the CIROS-1 Drill Hole, McMurdoSound, 245, 231239, edited by P. Barrett, DSIR bulletin, DSIRPublishing, Wellington.

Kohn, M. J., J. A. Josef, R. Madden, R. F. Kay, G. Vucetich,and A. A. Carlini (2004), Climate stability across theEocene-Oligocene transition, southern Argentina, Geology,32, 621–624, doi:10.1130/G20442.1.

Krissek, L. A., and P. R. Kyle (2000), Geochemical indicatorsof weathering, Cenozoic palaeoclimates, and provenancefrom fine-grained sediments in CRP-2/2A, Victoria LandBasin, Antarctica, Terra Antart., 7, 589–597.

Kryc, K. A., R. W. Murray, and D. W. Murray (2003),Al-to-oxide and Ti-to-organic linkages in biogenic sediment:Relationships to paleo-export production and bulk Al/Ti, EarthPlanet. Sci. Lett., 211(1-2), 125–141. doi:10.1016/S0012-821X(03)00136-5.

Lavelle, M. (1998), Strontium-isotope stratigraphy of theCRP-1drill hole, Ross Sea, Antarctica, Terra Antart., 5(3),691–696.

Levy, R. H., and D. M. Harwood (2000), Tertiary marinepalynomorphs from the McMurdo Sound erratics, Antarctica,in Paleobiology and Paleoenvironments of Eocene Fossilif-erous Erratics, 76, pp. 183–242, edited by J. D. Stilwell,and R. M. Feldmann, McMurdo Sound, East Antarctica, Ant-arctic Research Series.

Lewis, A. R., et al. (2008), Mid-Miocene cooling and theextinction of tundra in continental Antarctica, PNAS, 105(31),10676–10680, doi:10.1073/iti3108105.

Liu, Z., M. Pagani, D. Zinniker, R. DeConto, M. Huber,H. Brinkhuis, S. R. Shah, R. M. Leckie, and A. Pearson(2009), Global cooling during the Eocene-Oligocene climatetransition, Science, 323(5918), 1187–1190, doi:10.1126/science.1166368.

Macphail, M. K., and E. M. Truswell (2004), Palynology ofSite 1166, Prydz Bay, East Antarctica, In Cooper,A. K., O’Brien, P. E., and Richter, C. (Eds.), Proc. ODP,Sci. Results, 188: College Station, TX (Ocean DrillingProgram), 1–38. doi:10.2973/odp.proc.sr.188.013.2004.

Marbut, C. F. (1935), Soils of the United States, in U.S.Deptartment of Agriculture Atlas of American agriculture,Pt. III. Advance Sheets, No. 8., 98 pp.

Margolis, S. V., and J. P. Kennett (1970), Antarctic glaciationsduring the Tertiary recorded in sub-Antarctic deep-sea cores,Science, 170, 1085–1087.

Martin, A. P., A. F. Cooper, and W. J. Dunlap (2010), Geochro-nology of MountMorning, Antarctica: Two-phase evolution ofa long-lived trachyte-basanite-phonolite eruptive center, Bull.Volcanol., 72, 357–371, doi:10.1007/s00445-009-0319-1.

Maynard, J. B. (1992), Chemistry of modern soils as a guide tointerpreting Precambrian paleosols, J. Geol., 100, 279–289.

McIntosh, W. C. (1998), 40Ar/39Ar Geochronology of volcanicclasts and pumice in CRP-1 core, Cape Roberts, Antarctica,Terra Antart., 5(3), 683–690.

McLennan, S. M., S. Hemming, D. K. McDaniel, andG. N. Hanson (1993), Geochemical approaches to sedimenta-tion, provenance and tectonics, in Processes Controlling theComposition of Clastic Sediments, edited by Johnsson M. J.and A. Basu, Geol. Soc. Am. Spec. Paper, 284, 21–40.

Murray, R. W., D. J. Miller, and K. A. Kryc (2000), Analysis ofmajor and trace elements in rocks, sediments, and interstitialwaters by inductively coupled plasma–atomic emission spec-trometry (ICP-AES). ODP Tech. Note, 29 (Online). Availablefrom World Wide Web: http://www-odp.tamu.edu/publica-tions/tnotes/tn29/INDEX.HTM.

Nesbitt, H. W., and G. M. Young (1982), Early Proterozoicclimates and plate motions inferred frommajor element chem-istry of lutites, Nature, 299, 715–717, doi:10.1038/299715a0.

Nesbitt, H. W., and G. M. Young (1984), Prediction of someweathering trends of plutonic and volcanic rocks based onthermodynamic and kinetic considerations, Geochim.Cosmochim. Acta, 48, 1523–1534, doi:10.1016/0016-7037(84)90408-3.

Nesbitt, H. W., and G. M. Young (1989), Formation and dia-genesis of weathering profiles, J. Geol., 97(2), 129–147.

Nesbitt, H. W., G. M. Young, S. M. McLennan, and R. R. Keays(1996), Effects of chemical weathering and sorting on thepetrogenesis of siliciclastic sediments, with implications forprovenance studies, J. Geol., 104(5), 525–542.

Nordt, L., and S. Driese (2010), A modern soil characterizationapproach to reconstructing physical and chemical propertiesof paleo-Vertisol, Am. J. Sci., 310, 37–64, doi:10.2475/01.2010.02.

O’Brien, P. E., et al. (2001), Proc. ODP, Init. Repts., 188:College Station, TX (Ocean Drilling Program), doi:10.2973/odp.proc.ir.188.2001.

Passchier, S. (2000), Geochemistry, mineralogy and grain-sizeof the Sirius Group and related glacial deposits in Antarctica:Implications for paleoclimate and ice sheet drainage, PhDthesis, The Ohio State University.

Passchier, S., and L. A. Krissek (2008), Oligocene–MioceneAntarctic continental weathering record and paleoclimatic

implications, Cape Roberts Drilling Project, Ross Sea,Antarctica, Palaeogeogr. Palaeoclimatol. Palaeoecol., 260,30–40. doi:10.1016/j.palaeo.2007.08.012.

Poole, I., D. Cantrill, and T. Utescher (2005), A multi-proxyapproach to determine Antarctic terrestrial palaeoclimate dur-ing the Late Cretaceous and Early Tertiary, Palaeogeogr.Palaeoclimatol. Palaeoecol., 222, 95–121, doi:10.1016/j.palaeo.2005.03.011.

Pross, J., et al. (2012), Persistent near-tropical warmth on theAntarctic continent during the early Eocene epoch, Nature,488, 73–77. doi:10.1038/nature11300.

Prebble, J. G., J. I. Raine, P. J. Barrett, and M. J. Hannah(2006), Vegetation and climate from two Oligoceneglacioeustatic sedimentary cycles (31 and 24 Ma) cored bythe Cape Roberts Project, Victoria Land Basin, Antarctica,Palaeogeogr. Palaeoclimatol. Palaeoecol., 231, 41–57,doi:10.1016/j.palaeo.2005.07.025.

Raine, J. I. (1998), Terrestrial palynomorphs from CapeRoberts Project drill hole CRP-1, Ross Sea, Antarctica. TerraAntart., 5, 539–548.

Rasmussen, C., S. L. Brantley, D. Richter, A. Blum, J. Dixon,and A. White (2011), Strong climate and tectonic control onplagioclase weathering in granitic terrain, Earth Planet. Sci.Lett., 301(3-4), 521–530, doi:10.1016/j.epsl.2010.11.037.

Riebe, C. S., J. W. Kirchner, and R. C. Finkel (2004),Erosional and climatic effects on long-term chemicalweathering rates in granitic landscapes spanning diverse cli-mate regimes, Earth Planet. Sci. Lett., 224, 547–562,doi:10.1016/j.epsl.2004.05.019.

Rieck, H. J. (1989), Paleomagnetic stratigraphy, in AntarcticCenozoic History From the CIROS-1 Drill Hole, McMurdoSound, 245, 153–158, edited by P. Barrett, DSIR bulletin,DSIR Publishing, Wellington.

Roser, B. P., and A. R. Pyne (1989), Wholerock geochemistry,in Antarctic Cenozoic History From the CIROS-1 Drill Hole,McMurdo Sound, 245, 175–184, edited by P. Barrett, DSIRbulletin, DSIR Publishing, Wellington.

Sheldon, N. D., G. J. Retallack, and S. Tanaka (2002),Geochemical climofunction from North American soils andapplication to paleosols across the Eocene-Oligocene boundaryin Oregon, J. Geol., 110, 687–696.

Sheldon, N. D., and G. J. Retallack (2004), Regionalpaleoprecipitation records from the late Eocene and Oligoceneof North America, J. Geol., 112(4), 487–494.

Sheldon, N. D., R. L. Mitchell, M. E. Collinson, and J. J. Hooker(2009), Eocene-Oligocene transition paleoclimatic andpaleoenvironmental record from the Isle of Wight (UK), Geol.Soc. Am. Spec. Pap., 452, 249–259, doi:10.1130/2009.2452(16).

Shipboard Scientific Party (2001), Site 1166, In O’Brien,P. E., Cooper, A. K., Richter, C., et al., Proc. ODP, Init. Repts.,188, College Station, TX (Ocean Drilling Program), 1–110.doi:10.2973/odp.proc.ir.188.104.2001

Stickley, C. E., K. E. St John, N. Koç, R.W. Jordan, S. Passchier,R. B. Pearce, and L. E. Kearns (2009), Evidence for middleEocene Arctic sea ice from diatoms and ice-rafted debris,Nature, 460, 376–379, doi:10.1038/nature08163.

Strand, K., S. Passchier, and J. Näsi (2003), Implications ofquartz grain microtextures for onset Eocene/Oligoceneglaciation in Prydz Bay, ODP Site 1166, Antarctica,Palaeogeogr. Palaeoclimatol. Palaeoecol., 198, 101–111,doi:10.1016/S0031-0182(03)00396-1.

Tauxe, L., et al. (2012), Chronostratigraphic framework for theIODP Expedition 318 cores from the Wilkes Land Margin:Constraints for paleoceanographic reconstruction, Paleoceano-graphy, 27, PA2214, 19, doi:10.1029/2012PA002308.

Thiry, M. (2000), Palaeoclimatic interpretation of clay minerals inmarine deposits: An outlook from the continental origin, Earth-Sci. Rev., 49, 201–221, doi:10.1016/S0012-8252(99)00054-9.

Thorn, V. C., and R. DeConto (2006), Antarctic climate at theEocene/Oligocene boundary—Climate model sensitivity to highlatitude vegetation type and comparisons with the palaeobotanicalrecord, Palaeogeogr. Palaeoclimatol. Palaeoecol., 231(1),134–157, doi:10.1016/j.palaeo.2005.07.032.

Truswell, E. M., and M. K. Macphail (2009), Polar forests on theedge of extinction: What does the fossil spore and pollen evi-dence from East Antarctica say?, Aust. Syst. Bot., 22, 57–106,doi:10.1071/SB08046.

Turner, B. F., A. F. White, and S. L. Brantley (2010), Effects oftemperature on silicate weathering: Solute fluxes and chemicalweathering in a temperate rain forest watershed, JamiesonCreek, British Columbia, Chem. Geol., 269, 62–78, doi:10.1016/j.chemgeo.2009.09.005.

Warny, S., R. A. Askin, M. J. Hannah, B. A. R. Mohr,J. I. Raine, D. M. Harwood, F. Florindo, and the SMSScience Team (2009), Palynomorphs from a sediment corereveal a sudden remarkably warm Antarctica during the middleMiocene, Geology, 37(10), 955–958, doi:10.1130/G30139A.1.

Watkins, D. K. (2007), Quantitative analysis of the calcareousnannofossil assemblages from CIROS-1, Victoria LandBasin, Antarctica, J. Nannoplankton Res., 29(2), 130–137.

Wedepohl, K. H. (1995), The composition of the continentalcrust, Geochim. Cosmochim. Acta, 59, 1217–1232, doi:10.1016/0016-7037(95)00038-2.

West, A. J., A. Galy, and M. Bickle (2005), Tectonic andclimatic controls on silicate weathering, Earth Planet. Sci.Lett., 235(1–2), 211–228, doi:10.1016/j.epsl.2005.03.020.

White, A. F., A. E. Blum, T. D. Bullen, D. V. Vivit, M. Schulz,and J. Fitzpatrick (1999), The effect of temperature on experi-mental and natural chemical weathering rates of granitoidrocks, Geochim. Cosmochim. Acta, 63(19/20), 3277–3291,doi:10.1016/S0016-7037(99)00250-1.

Wilson, D. S., S. S. Jamieson, P. J. Barrett, G. Leitchenkov,K. Gohl, and R. D. Larter (2012), Antarctic topography at theEocene–Oligocene boundary, Palaeogeogr. Palaeoclimatol.Palaeoecol., 335-336, 24–34, doi:10.1016/j.palaeo.2011.05.028.

Wilson, G. S., A. P. Roberts, K. Verosub, F. Florindo, andL. Sagnotti (1998), Magnetobiostratigraphic chronology of theEocene-Oligocene transition in the CIROS-1 core, Victoria Landmargin, Antarctica: Implications for Antarctica glacial history,Geol. Soc. Am. Bull., 110, 35–47, doi:10.1130/0016-7606(1998)1.

Zanazzi, A., M. J. Kohn, B. J. MacFadden, and D. O. Terry(2007), Large temperature drop across the Eocene-Oligocenetransition in central North America, Nature, 445, 639–642,doi:10.1038/nature05551.

Early Eocene to middle Miocene cooling and aridification of East Antarctica

Documents