Geochemistry Geophysics G Volume 8 Geosystemsjzachos/pubs/Roehl_etal_07.pdf · understanding the ultimate fate of this carbon is to develop high-resolution age ... 1997; Svensen et

On the duration of the Paleocene-Eocene thermal maximum(PETM)

Ursula Rohl and Thomas WesterholdCenter for Marine Environmental Sciences (MARUM), Bremen University, Leobener Strasse, D-28359 Bremen,Germany ([email protected])

Timothy J. BralowerDepartment of Geosciences, Pennsylvania State University, University Park, Pennsylvania 16802, USA

James C. ZachosEarth and Planetary Sciences Department, University of California, Santa Cruz, California 95064, USA

[1] The Paleocene-Eocene thermal maximum (PETM) is one of the best known examples of a transientclimate perturbation, associated with a brief, but intense, interval of global warming and a massiveperturbation of the global carbon cycle from injection of isotopically light carbon into the ocean-atmosphere system. One key to quantifying the mass of carbon released, identifying the source(s), andunderstanding the ultimate fate of this carbon is to develop high-resolution age models. Two independentstrategies have been employed, cycle stratigraphy and analysis of extraterrestrial helium (HeET), both ofwhich were first tested on Ocean Drilling Program (ODP) Site 690. These two methods are in agreementfor the onset of the PETM and initial recovery, or the clay layer (‘‘main body’’), but seem to differ in thefinal recovery phase of the event above the clay layer, where the carbonate contents rise and carbon isotopevalues return toward background values. Here we present a state-of-the-art age model for the PETMderived from a new orbital chronology developed with cycle stratigraphic records from sites drilled duringODP Leg 208 (Walvis Ridge, Southeastern Atlantic) integrated with published records from Site 690(Weddell Sea, Southern Ocean, ODP Leg 113). During Leg 208, five Paleocene-Eocene (P-E) boundarysections (Sites 1262 to 1267) were recovered in multiple holes over a depth transect of more than 2200 mat the Walvis Ridge, yielding the first stratigraphically complete P-E deep-sea sequence with moderate torelatively high sedimentation rates (1 to 3 cm/ka, where ‘‘a’’ is years). A detailed chronology wasdeveloped with nondestructive X-ray fluorescence (XRF) core scanning records on the scale of precessioncycles, with a total duration of the PETM now estimated to be �170 ka. The revised cycle stratigraphicrecord confirms original estimates for the duration of the onset and initial recovery but suggests a newduration for the final recovery that is intermediate to the previous estimates by cycle stratigraphy and HeET.

Components: 8385 words, 5 figures, 3 tables.

Keywords: XRF core scanner; sediment chemistry; cyclostratigraphy; rapid climate change; age model; Paleocene-Eocene

thermal maximum.

Index Terms: 1051 Geochemistry: Sedimentary geochemistry; 1616 Global Change: Climate variability (1635, 3305, 3309,

4215, 4513); 4948 Paleoceanography: Paleocene/Eocene thermal maximum; 3036 Marine Geology and Geophysics: Ocean

drilling; 4901 Paleoceanography: Abrupt/rapid climate change (1605).

G3G3GeochemistryGeophysics

Geosystems

Published by AGU and the Geochemical Society

AN ELECTRONIC JOURNAL OF THE EARTH SCIENCES

GeochemistryGeophysics

Geosystems

Article

Volume 8, Number 12

11 December 2007

Q12002, doi:10.1029/2007GC001784

ISSN: 1525-2027

ClickHere

for

FullArticle

Copyright 2007 by the American Geophysical Union 1 of 13

http://dx.doi.org/10.1029/2007GC001784

Received 9 August 2007; Revised 8 October 2007; Accepted 24 October 2007; Published 11 December 2007.

Rohl, U., T. Westerhold, T. J. Bralower, and J. C. Zachos (2007), On the duration of the Paleocene-Eocene thermal maximum

(PETM), Geochem. Geophys. Geosyst., 8, Q12002, doi:10.1029/2007GC001784.

1. Introduction

[2] The Paleocene Eocene thermal maximum(PETM) is one of the most abrupt and transientclimatic events documented in the geologic record[e.g., Zachos et al., 2001, 2005]. This event wasassociated with pronounced warming of the oceansand atmosphere, changes in ocean chemistry, andreorganization of the global carbon cycle [Kennettand Stott, 1991; Koch et al., 1992; Thomas et al.,2002; Zachos et al., 2003, 2005; Tripati andElderfield, 2005; Sluijs et al., 2006]. Warming ofdeep waters and subsequent oxygen deficiencymay have been responsible for extinction of 30–50% of deep-sea benthic foraminiferal species[Thomas and Shackleton, 1996] and planktonicbiota were affected by changes in surface waterhabitats [e.g., Kelly et al., 1996; Bralower et al.,2002; Kelly, 2002; Raffi et al., 2005; Gibbs et al.,2006a, 2006b]; global warming also may have ledto a pulse of speciation or migration among mam-malian groups [e.g., Koch et al., 1992, Bowen etal., 2001; Gingerich, 2003]. The PETM corre-sponds to a significant (�3.5–4.5%) negativecarbon isotope excursion (CIE) recorded in marineand terrestrial sections [e.g., Kennett and Stott,1991; Koch et al., 1992; Bralower et al., 1997;Zachos et al., 2004, 2005; Schouten et al., 2007].The source and triggering mechanism of this eventare still the focus of much debate [e.g., Lourens etal., 2005; Sluijs et al., 2007; Storey et al., 2007].An orbital trigger for the PETM and similar (butless severe) events has been suggested [Lourens etal., 2005], but the specific orbital parameter asso-ciation is still not completely resolved [Westerholdet al., 2007]. Other mechanisms that might explainthe abruptness of the CIE include the input ofmethane into the ocean and atmosphere from thedissociation of methane hydrates in continentalmargin sediments or from the cracking of coalduring rifting of the northern North Atlantic Ocean[Dickens et al., 1995, 1997; Svensen et al., 2004].

[3] Identifying potential triggering mechanisms forthe PETM, as well as understanding the relation-ship between forcing and consequences requiresa very precise and high-resolution chronology.For example, quantifying the climate sensitivity

requires robust estimates of the mass of carbonreleased, and hence the rate of the CIE. Untilrecently, however, estimates of the absolute ageof the onset and the duration of the event werepoorly constrained, varying between 54.88 and55.50 Ma, and 100 and 250 ka, respectively [e.g.,Kennett and Stott, 1991; Koch et al., 1992; Aubryet al., 1996; Rohl and Abrams, 2000; Rohl et al.,2000; Farley and Eltgroth, 2003; Giusberti et al.,2007]. By using an astronomically calibrated butfloating timescale, the age of the onset (54.93 to54.98 Ma) and the duration (150 to 220 ka) of theCIE were initially determined at Ocean DrillingProgram (ODP) Site 1051 [Norris and Rohl, 1999]then refined using combined records from Sites690 and 1051 [Rohl et al., 2000]. However, be-cause the onset of the PETM in pelagic sequencesis marked by a pronounced dissolution layer orcondensed interval and the recovery by a litholog-ically uniform carbonate-rich interval, an alterna-tive constant flux age model was developed[Farley and Eltgroth, 2003]. This model is basedon the concentrations of extraterrestrial He (3HeET)and the assumption that the flux of this isotope tothe Earth remained constant during the PETM.Both age models are in agreement for the durationof the main body of the PETM (70–80 ka for the‘‘core’’, the onset, peak, and initial recovery phase(rapid rise in d13C, but low carbonate; here termedphase 1)), but diverge for the final recovery phaseof the CIE (slow rise in d13C, high carbonate; heretermed phase II), with orbital age models produc-ing 140 ka for this interval and He age models30 ka. Identification of cycles in the Ca (or Fe)records in the recovery interval of the Site 690section is complicated due to the high and uniformcarbonate content of the sediments.

[4] A new era in Cenozoic paleoceanography waslaunched with the recovery of Paleogene sedimentsin multisite depth transects during Ocean DrillingProgram Legs 198 (Shatsky Rise, Pacific Ocean[Bralower et al., 2002; Westerhold and Rohl,2006]) and 208 (Walvis Ridge, Southeast AtlanticOcean [Zachos et al., 2004]). These expeditionsyielded the first high-quality, stratigraphically com-plete sedimentary sequences of the early Paleo-gene, recovered in offset, multiple-hole sites. The

GeochemistryGeophysicsGeosystems G3G3

rOhl et al.: paleocene-eocene thermal maximum 10.1029/2007GC001784

2 of 13

lithologic and geochemical records generated fromthese cores exhibit the highly cyclic nature of earlyPaleogene climate, while also demonstrating that theearly Eocene Greenhouse World was punctuated bymultiple transient global warming events, or hyper-thermals [Thomas et al., 2000; Zachos et al., 2004].The occurrence of multiple hyperthermals within thelate Paleocene–early Eocene suggests a repeatedtrigger as their cause. Recently, X-ray fluorescence(XRF) core scanning records fromODPLeg 208 sitesand from ODP Site 1051 spanning a �4.3 millionyear interval of the late Paleocene to early Eocenewere used to establish a longer time series and todevelop a robust and improved chronology of mag-netochrons [Westerhold et al., 2007] which is con-sistent with records from the Bighorn Basin [Wing etal., 2000; Clyde et al., 2007].

[5] One of the obstacles to developing age modelsfor PETM sections is providing a exact definitionof the termination of the CIE on a global scale, e.g.,at Site 690, the location of the termination issomewhat subjective because of the asymptoticshape of the CIE. In addition, the low signal-to-noise ratio of the XRF Ca concentrations in thishigh-carbonate interval has made cycle extractiondifficult and somewhat subjective.

[6] Here we develop a revised chronology for thePETM using high-resolution geochemical datafrom the ODP Leg 208 depth transect in combina-tion with new Barium (Ba) XRF intensity data ofthe expanded section at ODP Site 690 from theWeddell Sea, Southern Ocean (Figure 1). TheBarium (Ba) records, in combination with Fe, Ca,and carbon isotope data from the Leg 208 sites andSite 690, show similar patterns that allow forrefinement of correlation and age calibrations.

These new data provide much better constraintson the durations of each phase of the CIE, partic-ularly the recovery phases (I and II). These recordswill also allow for a more accurate recalibration ofthe He isotope chronology from Site 690 [Farleyand Eltgroth, 2003]. Moreover, we propose that thedefinition of the termination of the CIE be based ona combination of cyclostratigraphic proxies derivedfrom XRF scanner and other methods rather thancarbon isotopes which gradually become uniform,thus making it difficult to define a globally recog-nizable termination point for the recovery.

2. Material and Methods

[7] ODP Leg 208 recovered stratigraphically com-plete and undisturbed PETM sections at four sitesbetween 2.7 and 4.8 km water depth (Sites 1262,1263, 1266, 1267 [Zachos et al., 2005]). At eachsite, the PETM sequence is characterized by anabrupt transition from underlying carbonate richooze to a dark red ‘‘clay layer’’ which then gradesupward into ooze. The most expanded PETMsuccession was recovered at the shallowest site,Site 1263. All Leg 208 sites are condensed duringthe PETM because of severe carbonate dissolution[Zachos et al., 2005] with CaCO3 content <1 wt%in the clay layers, compared to >80 and 90 wt% inthe underlying and overlying oozes. The thicknessof the clay layers increases with water depth, from5 cm at the shallowest Site 1263 (2717 m; paleo-depth �1500 m) to 35 cm in the deepest Site 1262(4755 m; paleodepth �3600 m) [Zachos et al.,2005]. We scanned longer core sections coveringthe late Paleocene-early Eocene interval from ODPSites 1262 (27�11.150S, 1�34.620E) and 1263(28�31.980S, 2�46.770E), as well as short (1 to

Figure 1. Location of ODP Site 690 in the Weddell Sea, Southern Ocean, and ODP Leg 208 Sites 1262 to 1267 onthe Walvis Ridge, South Atlantic. For comparison, ODP Site 1051 located on the Blake Nose in the Western NorthAtlantic, the Bighorn Basin section in Wyoming, USA, and the Forada section in northern Italy are also shown.


rOhl et al.: paleocene-eocene thermal maximum 10.1029/2007GC001784rOhl et al.: paleocene-eocene thermal maximum 10.1029/2007GC001784

3 of 13

2 m) intervals spanning the PETM at Sites 1265(28 �50 .10 0S, 2�38 .35 0E, 3083 m) , 1266(28�32.550S, 2�20.610E, 3798 m), and 1267(28�5.880S, 1�42.660E, 4355 m) and acquiredXRF Fe, Ca, and Ba data. In addition, we collectednew XRF Ba data from Core ODP 690B-19H, butalso XRF Fe, Ba and Ca data for longer coresections (160 to 176 meters below seafloor, mbsf)covering the late Paleocene–early Eocene interval(Leg 113, Maud Rise, 65�9.6290S; 1�12.2960E,2914 m; Figure 1) [Barker et al., 1988]. UpperPaleocene sediments at ODP Site 690 also showoscillations of CaCO3 and clay content asexpressed as white to pale brown banding, similarto the cyclicity in the Leg 208 sections.

[8] Relatively fast nondestructive core loggingmethods enable continuous measurements at muchfiner scales (down to millimeter scale) than arepractical for discrete sampling methods [Rohl andAbrams, 2000]. The X-ray fluorescence (XRF)core scanner acquires bulk-sediment chemical datafrom split core surfaces [Richter et al., 2006;Tjallingii et al., 2007]. Although elemental inten-sities are dominantly proportional to concentration,they are also influenced by the energy level of theX-ray source, the count time, and the physicalproperties of the sediment [Rohl and Abrams,2000]. Ba data were collected every 1 cm down-core over a 1 cm2 area with a generator setting of50 kV and a sampling time of 30 seconds directlyat the split core surface of the archive half withXRF Core Scanner II at the University of Bremen.The split core surface was covered with a 4 mmthin SPEXCerti Prep Ultralene1 foil to avoidcontamination of the XRF measurement unit anddesiccation of the sediment.

3. Results and Discussion

[9] All records (Fe, Ca, Ba) show distinctive andcorrelative elemental patterns in the clay layer ofthe basal CIE, despite the condensed nature of thisinterval. The Fe cyclicity at Site 690 disappears inthe upper half of the event where sediments arecharacterized by uniformly high (>85%) CaCO3

contents [Rohl et al., 2000], whereas at Site 1263the Fe records still show distinct variations(Figure 2). The Ba intensity data at Site 690 almostperfectly correlate to the Ba concentration andaccumulation record of Bains et al. [2000], whointerpret their data as biogenic barium indicative ofexport paleoproductivity (Figure 2). Dickens et al.[2003], on the other hand, noted that changes in theBa and barite fluxes, if global, must be largely a

consequence of changes in the supply of Ba to theocean, either from rivers or methane hydrates.Moreover, some of the total Ba in sediments islikely recycled during early diagenesis: adsorbedon mineral surfaces, co-precipitated with Fe-Mnoxyhydroxides or precipitated as barite [Paytanand Kastner, 1996].

3.1. Interpretation of New High-Resolution XRF Scanning Ba RecordsFrom the South Atlantic and SouthernOceans

[10] The Leg 208 data along with the new Site 690data both complement and improve upon theoriginal Site 690 records of Rohl et al. [2000]which was compromised by several factors. First,the records were generated with a first generationXRF Core Scanner [Rohl and Abrams, 2000],which had a restricted elemental range between Kand Sr. As a consequence, clear identification ofcycles in the Ca and Fe records in the phase II ofthe recovery interval was complicated due to thevery high carbonate content of the sediments andthe low signal-to-noise ratios [Rohl et al., 2000;Kelly et al., 2005]. The addition of Ba yields arecord with a much higher signal-to-noise ratiothan Ca and Fe. Because the total Ba analyzed,potentially consists of both marine sourced Ba inbarite (productivity), as well as Ba derived fromterrigenous sources, it is most useful in intervals inwhich the carbonate content is either exceptionallyhigh (very low Fe intensities) or low (extremelylow Ca intensities). Second, only a single core wasoriginally analyzed at Site 690 and thus it was notpossible to undertake any further detailed timeseries analysis for deciphering longer-term trendslike modulation of precession cycles by orbitaleccentricity. The long continuous cores from Leg208, though of lower resolution, are continuousover millions of years and thus provide the neces-sary framework for more precise identification ofcycle periodicity.

[11] In general, it appears that sedimentary Bacontent is not significantly affected by changes inredox conditions as minor changes in sulfate con-centration are not an issue [Torres et al., 1996;Dickens et al., 2003]. Thus the XRF Ba dataprovide a means of fine-tuning the high-resolutioncorrelation between individual holes of each site(auxiliary material1 Figure S2 and Tables S1 andS2) which were originally defined using mainly

1Auxiliary materials are available in the HTML. doi:10.1029/2007GC001784.



4 of 13

magnetic susceptibility data [Zachos et al., 2004].The new site-to-site correlation is straightforwardas the Ba intensity curves show a high signal-to-noise ratio. It is particularly effective in the highcarbonate recovery interval where the Ca and Fecycles are somewhat less prominent. The high-resolution Ba data provide a more refined inter-hole correlation than can be achieved with bulkcarbon isotope data alone, especially in the asymp-totic part of the recovery interval (auxiliary mate-rial Figure S2 and Table S2).

[12] Though Ba concentrations should primarilyreflect the flux of Ba to the seafloor which tendsto be dominated by barite, extreme changes inredox conditions may also affect concentrations.The spike in Ba at the base of the CIE, for example,might reflect a geochemical front that lead todiagenetic barite precipitation [e.g., Dickens etal., 2003] rather than original biogenic Ba concen-tration as reducing conditions set in during the

PETM. An investigation of both excess and bio-genic barium data obtained from discrete samplesfrom these sites supports this interpretation [Chunet al., 2006]. Regardless, the pattern can be used toassess relative completeness of individual sections.For example, the lack of this spike at the base ofthe clay layer at Site 1265 (auxiliary materialFigure S1), confirms the presence of an unconfor-mity that truncates the CIE at this site [Zachos etal., 2005].

3.2. On the Duration of the PETM

[13] We used the high-resolution Ba based site tosite correlations to assess the tempo of sedimentarycycles through the PETM interval (Figure 2). Thisincluded an analysis of longer-term trends of cyclicpatterns, e.g., modulation of the precession cyclesby eccentricity within Chron C24r (see Westerholdet al. [2007] for more details) (Figure 3). Sites 690and 1051 were previously correlated using preces-

Figure 2. Compilation of Fe (red, counts per second, cps; total counts, area), Ca (blue, counts per second, cps; totalcounts, area), Ba XRF data (green, total counts), Ba accumulation rates [from Bains et al., 2000], and bulk carbondata for Sites 690 [from Bains et al., 1999] and 1263 [from Zachos et al., 2005] across the PETM interval. Numbersindicate precession cycle assignments and letters indicate horizons as identified by Zachos et al. [2005]. Please notethat precession cycles 6 to 8 can be identified in the Ba curves, whereas the Fe shows less clear and the Ca recordsexhibits no clear cyclicity at all for this interval (recovery phase of the PETM).



5 of 13

sion cycles in Fe- and Ca-intensity data along withhigh-resolution stable isotope records [Rohl et al.,2000]. For Site 690 � 11 precessional cycles wereidentified between the base of the CIE and thepoint where d13C reaches post excursion values; 4cycles lie within the ‘‘core’’ of the d13C excursionfrom the initial decrease to the level where valuesbegin to rebound, and another 7 cycles lie withinthe recovery phase. The duration of the entire CIEwas thus estimated at �220 ka [Rohl et al., 2000].

[14] The large number of precession cycles in theupper recovery interval at Site 690 (and impliedlow sedimentation rate), however, seemed incon-sistent with the shift to high carbonate contentwhich would indicate an acceleration of accumu-lation rates. This suspicion was reinforced by theHe isotope chronology developed for Site 690which suggested that the upper carbonate richrecovery interval was relatively short comparedto the underlying dissolution and lower recovery

intervals yielding a total duration for the CIE ofabout 120 ka [Farley and Eltgroth, 2003]. Theoverestimation of cycles was clearly a consequenceof the low signal-to-noise ratios in the Ca records.However, the He record also has a high degree ofuncertainty over this interval, in part, because ofthe relatively low number of samples [Farley andEltgroth, 2003], and also because of possible errorsin sedimentation rates based on magnetochrondurations [Westerhold et al., 2007]. As a conse-quence, the He model likely underestimates theduration of the carbonate rich layer.

[15] The PETM section from Site 690 remains oneof the most expanded and intensively studiedreference sections for the CIE. Yet, location ofthe termination of the recovery phase of the PETMis somewhat subjective because of the asymptoticshape of the CIE. A conservative placement of thetop of the recovery at 167.10 mbsf [Rohl et al.,2000] (Figure 2) lies at the top of ODP-113-690B-

Figure 3. Compilation of Fe (red, total counts), Ba intensity (green, total counts), and bulk carbon data (ViennaPeedee belemnite, vPDB) for the late Paleocene–early Eocene interval of Sites 690 [from Cramer et al., 2003] and1262 and 1263 [from Zachos et al., 2005]. 1, top Fasciculithus spp. [Westerhold et al., 2007]; 2, decreaseFasciculithus spp. [Aubry et al., 1996].



6 of 13

19H [Kelly, 2002], which is overlain by a coringgap (Figures 2 and 3) rendering this site less thanideal for defining the termination of the event.Fortunately, a continuous section is available atSite 1263 where multiple holes were drilled acrossthe boundary. Thus, using new Ba based cycleconstraints on Sites 690 and 1263, we reassessedthe duration of the entire PETM. The terminationof the CIE at Site 1263 and the ‘‘reference section’’at Site 690 were defined (Tables 1 and 2), byidentifying an inflection point in the bulk d13Ccurve. Although the inflection is minor, it appearsto lie at a similar level from a cyclostratigraphicpoint of view [Zachos et al., 2005], in the middleof a precession cycle. The Ba record from Site1263 and for Site 690 show 5 cycles within theclay layer of the CIE and 3.5 cycles in the recoveryinterval phase II (Figure 2), which suggests that the7 cycles in the Ca record from the recovery intervalphase II at Site 690 as described by Rohl et al.[2000] are probably half precession cycles. Weassigned the precession cycle to the modern meanprecession period of 21 ka (average of the 19 and23 ka precessional bands [Herbert et al., 1995;Westerhold et al., 2007]), resulting in a newastronomically calibrated estimate of �170 ka forthe duration of the PETM.

[16] Our estimated 170-ka duration for the PETMcan be further evaluated by considering it withinthe context of longer-term cyclostratigraphic trendsof the upper Paleocene–lower Eocene sediments atSites 690, 1262 and 1263. Detailed time seriesanalysis of the records spanning magnetochronsC25n and C24r were undertaken by Westerhold etal. [2007] indicating that Chrons C24r and C25ncontain 148 and 23 precession cycles, respectively.The eccentricity cycles (405, 100 ka) were alsoextracted by filtering. This new Leg 208 cyclo-stratigraphic framework provided the context forassessing our proposed cyclostratigraphy for thePETM clay layer. Starting at the PETM (Figure 3)

we identify the same number of precession cyclesin Site 690 and Leg 208 sites with respect to theeccentricity related modulation pattern in the Feand Ba precession cycles. For example, precessioncycles 11 to 15 in both Site 1262 and 690 can becorrelated to a short eccentricity cycle (‘‘3’’ [seeWesterhold et al., 2007, Figure 4]) above thePETM. The congruent modulation pattern of Leg208 and Site 690 precession related sedimentarycycles lead to consistent stratigraphic featureswhich are global in nature. In particular precessioncycles 16 and 17 (Figure 3) located in a pro-nounced eccentricity cycle minimum are prominentfeatures which should also be present in othersections around the globe. The consistency of ourcyclostratigraphy is corroborated by biostratigraphicfeatures. For example, the position of the abruptdecrease in abundance of Fasciculithus spp. at Site690 [Aubry et al., 1996] (cycle 17 in Figure 3) andat Site 1262 [Westerhold et al., 2007] (top of cycle19 in Figure 3), respectively, is roughly correlativebetween the sites.

3.3. Sediment Accumulation Rates

[17] The new estimate for the total duration of thePETM (170 ka) is shorter than that derived fromthe original orbital model and longer than thatderived from the He age model (about 90–140 ka) [Farley and Eltgroth, 2003]. How does thisaffect accumulation rates (auxiliary materialFigure S3), particularly that of carbonate throughthe recovery interval when rates should have accel-erated [Dickens et al., 1997; Farley and Eltgroth,2003; Kelly et al., 2005; Zachos et al., 2005]? To

Table 1. Carbon Isotope and Ba Data Tie Points FromODP Site 690 Used for Correlation to the Leg 208 P-EBoundary Sections

Tie Points 690, mbsf 1263, rmcd

Fe intensity tie point cycle 8 167.20 332.79Fa 169.09 334.14Ca 170.01 334.84Onset CIE 170.64 335.27Fe intensity tie point cycle -1 170.99 335.54Fe intensity tie point cycle -2 171.40 336.04

aLetters after Bains et al. [1999].

Table 2. Carbon Isotope Tie Points From ODP Site690 and Assigned Ages Used for Correlation and Datingthe ODP Leg 208 PETM Sections

TiePointsa

690,mbsf

690 Age,±ka

(Rohl00)b

690 Age,±ka

(F&E03)b

690 Age,±ka

(This Study)

H 166.13 230.00 183.00 196.87G 167.12 218.00 118.60 153.50F 169.05 108.00 100.15 94.23E 169.39 88.00 93.60 81.17D 169.56 76.00 90.20 71.25C 170.02 47.20 67.00 42.38B 170.33 28.40 37.48 21.90A 170.63 1.00 1.00 0.75PEB 170.64 0.00 0.00 0.00A- 171.24 �34.00 �62.00 �30.80B- 172.81 �125.00 �102.53

aLetters after Bains et al. [1999].

bTime (ka) at ODP Site 690 relative to the P-E boundary set to 55Ma.

Ages are from Rohl et al. [2000] and Farley and Eltgroth [2003].



7 of 13

estimate accumulation rates, we used the orbitallytuned, but floating age model from ODP Leg 208[Westerhold et al., 2007], which provides tuned agesfrom the termination of Chron 25n to the onset of theCIE (1.113 Ma), and for the total duration of ChronC24r (3.118 Ma). For the duration of the PETM/CIE, we simply adopt our 170 ka estimate. Sedimentaccumulation rates are estimated for the uppermostPaleocene and lower Eocene at Site 1263 and Site690 cores using the depths, age model, and dura-tions as described above (Table 3 and Figure 4). Forboth sites, the sedimentation rates are close to 2 cm/ka before the onset of the CIE, drop dramatically(�1.0 cm/ka) during the body (clay layer) of theCIE/PETM, and then increase in the recovery inter-val (3 cm/ka). Although these estimates possessminor uncertainties it is clear that sedimentation

rates did increase during phase II of the recovery,but not tenfold as derived from He isotopes.

[18] We also compared our new orbital chronologywith one of the more expanded marine sections, theForada section in northern Italy which was depos-ited in a hemipelagic, near-continental setting. Inthis section, the Paleocene succession of limestone-marl couplets is sharply interrupted by a �3.30-m-thick unit of clays and marls representing the ‘‘claylayer’’ of the PETM [Giusberti et al., 2007]. TheCIE main excursion at Forada is about 3.4 m longand therefore is almost 3 times more expanded thanthe corresponding interval at ODP Site 690(Figure 5). Several parameters (hematite, carbon-ate, d13C, radiolarians) oscillate in a cyclical fash-ion and are interpreted to represent precession

Table 3. PETM Age Model for Sites 690 and 1262–1267a

Cycle

Depthmbsf

Site 690

DepthmcdSite1262

Depthrmcd

Site 1263

DepthmcdSite1265

DepthmcdSite1266

DepthrmcdSite1267

RelativeAge

(pre = 21.0 ka)to Cycle 1

RelativeAge

(pre = 21.0 ka)to ELMO(= 1000)

RelativeAge

(pre = 21.0 ka)to Onset CIE

20 160.42 135.88 326.56 308.24 - 227.12 399 2428 40219 160.98 136.09 327.01 308.66 - 227.35 378 2449 38118 161.57 136.31 327.53 309.19 - 227.62 357 2470 36017 - 136.56 328.08 no data - 227.89 336 2491 33916 162.96 136.81 328.52 309.91 - 228.14 315 2512 31815 163.54 137.04 329.11 310.44 - 228.37 294 2533 29714 164.32 137.27 329.58 310.81 302.05 228.70 273 2554 27613 164.87 137.50 330.13 311.29 302.50 228.95 252 2575 25512 165.29 137.75 330.78 311.76 302.95 229.25 231 2596 23411 - 138.12 no data 312.31 303.52 229.57 210 2617 21310 166.24 138.38 no data 313.01 304.10 229.90 189 2638 1929 - 138.73 332.12 313.64 304.70 230.30 168 2659 1718 167.20 139.03 332.79 314.21 305.12 230.64 147 2680 1507 - 139.32 333.39 314.66 305.52 230.93 126 2701 1296 - 139.53 333.88 315.23 305.93 231.22 105 2722 1085 169.29 139.71b 334.29c 315.45b 306.21b 231.40b 84 2743 874 169.65 139.82b 334.57c 315.59b 306.40b 231.52b 63 2764 663 169.98 139.94b 334.82c 315.72b 306.56b 231.63b 42 2785 452 170.30 140.03b 335.04c 315.80b 306.67b 231.69b 21 2806 241 170.60 335.24c 315.88b 306.75b 231.76b 0 2827 3Onset CIE 170.64 140.12 335.27 315.91 306.78 231.79 0�1 170.99 140.23 335.64 316.31 306.92 - �21 2848 �18�2 171.40 140.32 - 316.74 307.20 232.05 �42 2869 �39�3 171.88 140.53 - end of splice - 232.27 �63 2890 �60�4 172.40 - - - 232.45 �84 2911 �81�5 172.80 140.85 336.86 308.10 232.63 �105 2932 �102�6 173.20 141.04 337.11 308.43 232.85 �126 2953 �123�7 173.49 141.27 337.42 308.80 233.10 �147 2974 �144�8 no data 141.52 337.82 309.13 233.35 �168 2995 �165�9 no data 141.74 338.20 309.55 233.65 �189 3016 �186�10 no data 141.95 338.47 309.87 233.88 �210 3037 �207

aNotes: mcd, meters composite depth rmcd, revised meters composite depth. Relative ages are in ka.

bCorrelation of Site 1263 to Site 690.

cCorrelation to Site 1263 by Ba data.



8 of 13

cycles. The ‘‘core’’ of the excursion interval spans5 complete cycles (�105 ka). The 1.4-m thickrecovery interval in the Forada section containssix distinct limestone-marl couplets, and is inter-preted to represent six precessional cycles with aduration of �126 ka [Giusberti et al., 2007],providing a total CIE duration of �231 ± 22 ka,5–10% longer than the estimate of Rohl et al.[2000].

[19] For the core of the PETM, only 4 cycles wereidentified at Site 690 and 1263, while 5 wereidentified at Forada over the same time interval.This difference is just an artifact of the way inwhich the cycles are defined, using Fe (or clay) forthe ODP records (Fe data reflecting the terrigenouscomponent) (Figures 2, 3, and 4) whereas for theForada section the cycle count is based on thecarbonate record (Figure 5). The hematite record atForada [Giusberti et al., 2007] also exhibits 4cycles that vary roughly inversely with carbonateand would be the equivalent (as same number) ofthe 4 Fe cycles identified at the ODP sites.

[20] With regard to the recovery interval, Giusbertiet al. [2007] also noted that the seven precessioncycles from the corresponding recovery interval(phases I and II) at Site 690 [Rohl et al., 2000]exceeds that at Forada by one cycle. They corre-lated cycle by cycle using the numbering schemefor Site 690. Unfortunately, the sample resolutionfor the Forada record is not high enough, especiallyin the recovery interval, to precisely correlatedetails in records for this part of the CIE. However,with our new data we were able to revise thecorrelation of Forada and the deep-sea sites. Onepossible additional correlation point is the LO/decrease of Fasciculithus spp. which is associatedwith precession cycle 16. The pronounced mini-mum in the eccentricity modulation of precessioncycles 16 and 17 aligns closely to the thicklimestone bed (marker bed, cycle 16) at Forada(Figure 5) consistent with the proposed detailedcorrelation within and around the CIE between thesites. If we use this event as a tie point, then thenumber of precessional cycles at Forada closelymatches that of the deep-sea sections.

Figure 4. Sedimentation rates (top) for Site 1263 and (bottom) for Site 690 plotted versus Fe (red, total counts),Ba (green, total counts) and d13C (black, Vienna Peedee belemnite, vPDB) data. For Site 690 the sedimentation ratesfrom Rohl et al. [2000] and Farley and Eltgroth [2003] are also shown for comparison.



9 of 13

[21] Given the massive global carbonate dissolu-tion at the onset of the PETM, could we haveunderestimated the number of precession cycles inthe base of the clay layer in the pelagic site? This ispossible, though it seems unlikely given that theextent of carbonate erosion probably differed be-tween Maud Rise and Walvis Ridge, and that nopure clay layer was produced at Site 690 [Zeebeand Zachos, 2007]. However, Site 690 does showevidence for a significant decrease in carbonatecontent. As such, it might be possible that oneprecession cycle was removed in the middle ofthe onset of the PETM between cycle �1 and 1(Figure 2) and that the equivalent of 10 ka ismissing. Our study and the cycle identification atSite 690 and correlation to the Leg 208 sites result inan estimate of 5 precession cycles. The Foradasection seems to exhibit 6 cycles for the sameinterval. However, we are confident that the newcycle-based age model is robust as the new Leg 208records show individual precession cycles of therecovery interval that are perfectly modulated by

shorter eccentricity cycles. Therefore we can ex-clude the possibility that the number of peaks orprecession cycles has been underestimated or over-estimated. At this point a few words about modula-tion are warranted (auxiliarymaterial Figure S4): thelong term record shows that the PETM is locatedbetween a 405-ka maximum and 405-ka minimum[Westerhold et al., 2007, Figure 6]. This constraintdemonstrates that the duration of the clay layer ofthe PETM cannot be longer than the equivalent of 7precession cycles, which would already be themaximum estimate. If we assume that the clay layeris longer then the equivalent of 7 precession cyclesthan the 405-ka filter will move out of phase. TheLeg 208 cyclostratigraphy clearly demonstrates thatthe clay layer represents a minimum of 5 and amaximum of 7 precession cycles (auxiliary materialFigure S4). In the future, we recommend thatsediment models simulate the effect of cyclicalsediment flux to determine if the expression of theprecession cycle signal can be obscured by a com-bination of dissolution and bioturbation.

Figure 5. Latest high-resolution correlation of Site 690 Fe (total counts) and d13C data (Vienna Peedee belemnite,vPDB) to the Forada section (CaCO3 and d13C data) in northern Italy (modified from Giusberti et al. [2007]). Lettersindicate horizons as identified by Zachos et al. [2005] for Site 690. PFDI, Planktonic Foraminiferal Dissolution [fromLuciani et al., 2007]; BEE, Benthic Extinction Event.



10 of 13

3.4. Timing of the CIE Relative toOrbital Phases

[22] The final point concerns the relationship of thePETM to orbital phases. The question of whetherthe CIE and other hyperthermals was triggered bythe 100 or 405 ka eccentricity cycles has beenexplored in several papers [Cramer et al., 2003;Lourens et al., 2005] using a variety of pelagicsediment cycle records including those fromWalvisRidge [Lourens et al., 2005]. A recent reevaluationof the longer cyclic records from Walvis Ridgeindicates that the CIE did not occur during a timeof maximal variability in insolation (405-kamaximum) as suggested previously, but with adecreasing segment of a 405-ka eccentricity cycle[Westerhold et al., 2007, Figure 6b]. Nevertheless,there seems to be a relation to a 100-ka eccentricitycycle maximum [Westerhold et al., 2007, Figure 6c]:the amplitude modulation of the short eccentricitycycle clearly visible in Figure 3 (with maxima of the100-ka cycles at position of precession cycles �10,�5, 2, 8, 13, 20) indicates that the onset of the CIEoccurred halfway (after precession cycle �2 and�1) within the short eccentricity cycle comprisingprecession cycles �2 through 3, which represents aperiod of relatively high-amplitude climate variationand could be a mechanism for forcing climatechange on 10-ka timescales.

4. Conclusions and Implications

[23] High resolution Fe and Ba intensity recordswere generated for the P-E boundary intervals atODP Site 690 on Maud Rise and Sites 1262 and1263 on Walvis Ridge. The P-E boundary sectionat Site 690 is the one of the most expanded in thedeep sea, while the sections at Sites 1262 and 1263lie within the longest, most stratigraphically con-tinuous and astronomically tuned sequence ofmarine sediments spanning the upper Paleoceneand lower Eocene. The new data highlight signif-icant features and striking changes in response tothis brief episode of extreme warmth. The cycle-to-cycle correlation between ODP Sites 1262 and1263 in the context of the astronomically calibratedlonger-term record [Westerhold et al., 2007] indi-cates that the cycles have durations in the preces-sional frequency band.

[24] The lower portion of the PETM whichincludes the dissolution phase and lower recoveryinterval contains 5 precession cycles, while theupper recovery interval contains 3.5 cycles. Thetotal duration of the PETM is now estimated to be

�170 ka, roughly mid-way between previous esti-mates based on cycle stratigraphy and He isotopes.One key implication is that the carbonate richupper recovery layer represents a time of enhancedaccumulation rates consistent with the notion of asupersaturated ocean [Farley and Eltgroth, 2003;Kelly et al., 2005].

Acknowledgments

[25] We thank two reviewers for their comments. This re-

search used samples and/or data provided by the Ocean Drilling

Program (ODP). ODP is sponsored by the U.S. National

Science Foundation (NSF) and participating countries under

management of Joint Oceanographic Institutions (JOI), Inc.

Funding for this research was provided by the Deutsche

Forschungsgemeinschaft (DFG) to U. Rohl and T. Westerhold

and by NSF grants EAR-0120727 and 0628486 and ACS-PRF

grant 42705-AC8 to T. J. Bralower and to J. C. Zachos. We

thank Gar Esmay (ECR) for assistance with shipping Site 690

cores to Bremen, Alex Wulbers (BCR) for logistical assistance,

and Vera Lukies (Bremen) for support in the XRF Core

Scanners Lab. The complete data set presented in this paper

is available online in the WDC-MARE PANGAEA database at

http://www.pangaea.de (doi:10.1594/PANGAEA.667443).

References

Aubry, M.-P., W. A. Berggren, L. Stott, and A. Sinha (1996),The upper Paleocene–lower Eocene stratigraphic record andthe Paleocene-Eocene boundary carbon isotope excursion:Implications for geochronology, in Correlation of the EarlyPaleogene in Northwest Europe, edited by R. W. Knox,R. M. Corfield, and R. E. Dunay, Geol. Soc. Spec. Publ.,101, 353–380.

Bains, S., R. M. Corfield, and R. D. Norris (1999), Mechan-isms of climate warming at the end of the Paleocene,Science, 285, 724–727.

Bains, S., R. D. Norris, R. M. Corfield, and K. L. Faul (2000),Termination of global warmth at the Paleocene/Eoceneboundary through productivity feedbacks, Nature, 407,171–174.

Barker, P. F., et al. (1988), Proceedings of the Ocean DrillingProgram, Initial Reports, vol. 113, doi:10.2973/odp.proc.ir.113.1988, Ocean Drill. Program, College Station, Tex.

Bowen, G. J., P. K. Koch, P. D. Gingerich, R. D. Norris,S. Bains, and R. M. Corfield (2001), Refined isotopestratigraphy across the continental Paleocene-Eocene bound-ary on Polecat Bench in the Northern Bighorn Basin, inPaleocene-Eocene Stratigraphy and Biotic Change in theBighorn and Clarks Fork Basins, Wyoming, Univ. Mich.Pap. Paleontol., vol. 33, edited by P. D. Gingerich, pp.73–88, Ann Arbor.

Bralower, T. J., D. J. Thomas, J. C. Zachos, M. M. Hirschmann,U. Rohl, H. Sigurdsson, E. Thomas, and D. L. Whitney(1997), High-resolution records of the late Paleocene thermalmaximum and circum-Caribbean volcanism: Is there a causallink?, Geology, 25, 963–966.

Bralower, T. J., et al. (2002), Proceedings of the Ocean DrillingProgram, Initial Reports, vol. 198, Ocean Drill. Program,College Station, Tex. (Available at http://www-odp.tamu.edu/publications/198_IR/198ir.htm).



11 of 13

Chun, C. O. J., M. L. Delaney, and J. C. Zachos (2006),Geochemical response of the marine environment acrossthe Paleocene-Eocene Thermal Maximum, Walvis RidgeTransect (ODP Sites 1262, 1263, 1266) (abstract), paperpresented at Climate & Biota of the Early Paleogene 2006,Prov. Govt. of Biscay, Bilbao, Spain.

Clyde, W. A., J. A. Finarelli, S. L. Wing, D. Schankler, andA. Chew (2007), Basin-wide magnetostratigraphic frame-work for the Bighorn Basin, Wyoming, Geol. Soc. Am. Bull.,119(7), 848–859, doi:10.1130/B26104.1.

Cramer, B. S., J. D. Wright, D. V. Kent, and M.-P. Aubry(2003), Orbital climate forcing of d13C excursions in the latePaleocene–early Eocene (chrons C24n–C25n), Paleoceano-graphy, 18(4), 1097, doi:10.1029/2003PA000909.

Dickens, G. R., J. R. O’Neill, D. K. Rea, and R. M. Owen(1995), Dissociation of oceanic methane as a cause of thecarbon isotope excursion at the end of the Paleocene, Paleo-ceanography, 10, 965–971.

Dickens, G. R., M. M. Castillo, and J. G. C. Walker (1997),A blast of gas in the latest Paleocene: Simulating first-ordereffects of massive dissociation of oceanic methane hydrate,Geology, 25, 259–262.

Dickens, G. R., T. Fewless, E. Thomas, and T. J. Bralower(2003), Excess barite accumulation during the Paleocene-Eocene thermal maximum: Massive input of dissolvedbarium from seafloor gas hydrate reservoirs, in Causes andConsequences of Globally Warm Climates in the EarlyPaleogene, edited by S. L. Wing et al., Spec. Pap. Geol.Soc. Am., 369, 11–23.

Farley, K. A., and S. F. Eltgroth (2003), An alternative agemodel for the Paleocene-Eocene thermal maximum usingextraterrestrial (super 3) He, Earth Planet. Sci. Lett., 208,135–148.

Gibbs, S. J., P. R. Bown, J. A. Sessa, T. J. Bralower, and P. A.Wilson (2006a), Nannoplankton extinction and originationacross the Paleocene-Eocene Thermal Maximum, Science,314, 1770–1773, doi:10.1126/science.1133902.

Gibbs, S. J., T. J. Bralower, P. R. Bown, J. C. Zachos, and L. M.Bybell (2006b), Shelf and open-ocean calcareous phyto-plankton assemblages across the Paleocene-Eocene ThermalMaximum: Implications for global productivity gradients,Geology, 34, 233–236, doi:10.1130/G22381.1.

Gingerich, P. D. (2003), Mammalian responses to climatechange at the Paleocene-Eocene boundary: Polecat Benchrecord in the northern Bighorn Basin, Wyoming, in Causesand Consequences of Globally Warm Climates in the EarlyPaleogene, edited by S. L. Wing et al., Spec. Pap. Geol. Soc.Am., 369, 463–478.

Giusberti, L., D. Rio, C. Agnini, J. Backman, E. Fornaciari, F.Tateo, and M. Oddone (2007), Mode and tempo of the Pa-leocene-Eocene thermal maximum in an expanded sectionfrom the Venetian pre-Alps, Geol. Soc. Am. Bull., 119,391–412, doi:10.1130/B25994.1.

Herbert, T. D., S. L. D’Hondt, I. Premoli-Silva, E. Erba, and A.G. Fischer (1995), Orbital chronology of Cretaceous-EarlyPalaeocene marine sediments, in Geochronology, TimeScales and Global Stratigraphic Correlations: A UnifiedTemporal Framework for a Historical Geology, edited byW. A. Berggren, D. V. Kent, and J. Hardenbol, Spec. Publ.SEPM Soc. Sediment. Geol., 54, 81–94.

Kelly, D. C. (2002), Response of Antarctic (ODP Site 690)planktonic foraminifera to the Paleocene–Eocene thermalmaximum: Faunal evidence for ocean/climate change, Pa-leoceanography, 17(4), 1071, doi:10.1029/2002PA000761.

Kelly, D. C., T. J. Bralower, J. C. Zachos, I. Premoli Silva, andE. Thomas (1996), Rapid diversification of planktonic for-

aminifera in the tropical Pacific (ODP Site 865) during thelate Paleocene thermal maximum, Geology, 24, 423–426.

Kelly, D. C., J. C. Zachos, T. J. Bralower, and S. A. Schellen-berg (2005), Enhanced terrestrial weathering/runoff and sur-face ocean carbonate production during the recovery stagesof the Paleocene-Eocene thermal maximum, Paleoceanogra-phy, 20, PA4023, doi:10.1029/2005PA001163.

Kennett, J. P., and L. D. Stott (1991), Abrupt deep-sea warm-ing, palaeoceanographic changes and benthic extinctions atthe end of the Palaeocene, Nature, 353, 225–229.

Koch, P. L., J. C. Zachos, and P. D. Gingerich (1992), Correla-tion between isotope records near the Palaeocene/Eoceneboundary, Nature, 358, 319–322.

Lourens, L. J., A. Sluijs, D. Kroon, J. C. Zachos, E. Thomas,U. Rohl, J. Bowles, and I. Raffi (2005), Astronomical pacingof late Palaeocene to early Eocene global warming events,Nature, 435, 1083–1087.

Luciani, V., L. Giusberti, C. Agnini, J. Backman, E. Fornaciari,and D. Rio (2007), The Paleocene-Eocene Thermal Maxi-mum as recorded by Tethyan planktonic foraminifera in theForada section (northern Italy), Mar. Micropaleontol., 64,189–214, doi:10.1016/j.marmicro.2007.05.001.

Norris, R. D., and U. Rohl (1999), Carbon cycling and chron-ology of climate warming during the Palaeocene/Eocenetransition, Nature, 401, 775–778.

Paytan, A., and M. Kastner (1996), Benthic Ba fluxes in thecentral equatorial Pacific: Implications for the oceanic Bacycle, Earth Planet. Sci. Lett., 142, 439–450.

Raffi, I., J. Backman, and H. Palike (2005), Changes in calcar-eous nannofossil assemblages across the Paleocene/Eocenetransition from the paleo-equatorial Pacific Ocean, Palaeo-geogr. Palaeoclimatol. Palaeoecol., 226, 93–126.

Richter, T. O., S. Van der Gast, R. Koster, A. Vaars, R. Gieles,H. C. De Stigter, H. De Haas, and T. C. E. Van Weering(2006), The Avaatech XRF Core Scanner: Technical descrip-tion and applications to NE Atlantic sediments, in New Tech-niques in Sediments Core Analysis, edited by R. G.Rothwell, Geol. Soc. Spec. Publ., 267, 39–51.

Rohl, U., and L. J. Abrams (2000), High-resolution, downholeand non-destructive core measurements from Sites 999 and1001 in the Caribbean Sea: Application to the late Paleocenethermal maximum, Proc. Ocean Drill. Program Sci. Results,165, 191–203.

Rohl, U., T. J. Bralower, R. D. Norris, and G. Wefer (2000),New chronology for the late Paleocene thermal maximumand its environmental implications, Geology, 28, 927–930.

Schouten, S.,M.Woltering, I. C. Rijpstra, A. Sluijs, H. Brinkhuis,and J. S. Sinninghe Damste (2007), The Paleocene–Eocenecarbon isotope excursion in higher plant organic matter: Differ-ential fractionation of angiosperms and conifers in the Arctic,Earth Planet. Sci. Lett., 258, 581–592.

Sluijs, A., et al. (2006), Subtropical Arctic Ocean temperaturesduring the Palaeocene/Eocene thermal maximum, Nature,441, 610–613.

Sluijs, A., G. B. Bowen, H. Brinkhuis, L. J. Lourens, andE. Thomas (2007), The Palaeocene-Eocene thermal maxi-mum super greenhouse: Biotic and geochemical signatures,age models and mechanisms of global change, in Deep TimePerspectives on Climate Change: Marrying the Signal FromComputer Models and Biological Proxies, edited byM. Williams et al., Micropaleontol. Soc. Spec. Publ., Geol.Soc., London, in press.

Storey, M., R. A. Duncan, and C. C. Swisher III (2007),Paleocene-Eocene Thermal Maximum and the opening ofthe northeast Atlantic, Science, 316(587), doi:10.1126/science.1135274.



12 of 13

Svensen, H., S. Planke, A. Malthe-Sørensen, B. Jamtveit,R. Myklebust, T. R. Eidem, and S. S. Rey (2004), Releaseof methane from a volcanic basin as a mechanism for initialEocene global warming, Nature, 429, 542–545.

Thomas, D. J., J. C. Zachos, T. J. Bralower, E. Thomas, andS. Bohaty (2002), Warming the fuel for the fire: Evidencefor the thermal dissociation of methane hydrate during thePaleocene-Eocene thermal maximum, Geology, 30, 1067–1070.

Thomas, E., and N. J. Shackleton (1996), The latest Paleocenebenthic foraminiferal extinction and stable isotope anoma-lies, in Correlation of the Early Paleogene in NorthwestEurope, edited by R. W. Knox et al., Geol. Soc. Spec. Publ.,101, 401–441.

Thomas, E., J. C. Zachos, and T. J. Bralower (2000), Deep-seaenvironments on a warm Earth: Latest Paleocene-earlyEocene, in Warm Climates in Earth History, edited byB. Huber, K. MacLeod, and S. Wing, pp. 132–160, Cam-bridge Univ. Press, New York.

Tjallingii, R., U. Rohl, M. Kolling, and T. Bickert (2007),Influence of the water content on X-ray fluorescence core-scanning measurements in soft marine sediments, Geochem.Geophys. Geosyst., 8, Q02004, doi:10.1029/2006GC001393.

Torres, M. E., H. J. Brumsack, G. Bohrmann, and K.-C. Emeis(1996), Barite fronts in continental margin sediments: A newlook at barium remobilization in the zone of sulfate reductionand formation of heavy barites in diagenetic fronts, Chem.Geol., 127, 125–139.

Tripati, A., and H. Elderfield (2005), Deep-sea temperatureand circulation changes at the Paleocene-Eocene ThermalMaximum, Science, 308, 1894–1898.

Westerhold, T., and U. Rohl (2006), Data report: Revised com-posite depth records for Shatsky Rise Sites 1209, 1210, and

1211 [online], Proc. Ocean Drill. Program Sci. Results, 198.(Available at http://www-odp.tamu.edu/publications/198_SR/122/122.htm)

Westerhold, T., U. Rohl, J. Laskar, I. Raffi, J. Bowles, L. J.Lourens, and J. C. Zachos (2007), On the duration of mag-netochrons C24r and C25n and the timing of early Eoceneglobal warming events: Implications from the Ocean DrillingProgram Leg 208 Walvis Ridge depth transect, Paleoceano-graphy, 22, PA2201, doi:10.1029/2006PA001322.

Wing, S. L., H. Bao, and P. L. Koch (2000), An early Eocenecool period? Evidence for continental cooling during thewarmest part of the Cenozoic, in Warm Climates in EarthHistory, edited by B. T. Huber, K. G. MacLeod, and S. L.Wing, pp. 197–237, Cambridge Univ. Press, Cambridge,U. K.

Zachos, J. C., M. Pagani, L. Sloan, E. Thomas, and K. Billups(2001), Trends, rhythms, and aberrations in global climate65 Ma to Present, Science, 292, 686–693.

Zachos, J. C., M. W. Wara, S. Bohaty, M. L. Delaney, M. R.Petrizzo, A. Brill, T. J. Bralower, and I. Premoli-Silva(2003), A transient rise in tropical sea surface temperatureduring the Paleocene-Eocene thermal maximum, Science,302, 1551–1554.

Zachos, J. C., et al. (2004), Proceedings of the Ocean DrillingProgram, Initial Reports [CD-ROM], vol. 208, Ocean Drill.Program, College Station, Tex.

Zachos, J. C., et al. (2005), Rapid acidification of the oceanduring the Paleocene-Eocene Thermal Maximum, Science,308, 1611–1615.

Zeebe, R. E., and J. C. Zachos (2007), Reversed deep-seacarbonate ion basin gradient during Paleocene-Eocene ther-mal maximum, Paleoceanography, 22 , PA3201,doi:10.1029/2006PA001395.



13 of 13

Documents

Geochemistry Geophysics G Volume 8 Geosystemsjzachos/pubs/Roehl_etal_07.pdf · understanding the ultimate fate of this carbon is to develop high-resolution age ... 1997; Svensen et