ABSTRACT

A thick, apparently continuous section recording eventsof the latest Paleocene thermal maximum in a neritic setting wasdrilled at Bass River State Forest, New Jersey as part of ODP Leg174AX (Miller, Sugarman, Browning, et al. , 1998). Integrated nan-nofossil and magneto- stratigraphy provides a firm chronology sup-plemented by planktonic foraminiferal biostratigraphy. This chrono-logic study indicates that this neritic section rivals the best deep-seasections in providing a complete record of late Paleocene climaticevents. Carbon and oxygen isotopes measured on benthic foraminiferashow a major (4.0 ä in carbon, 2.3 ä in oxygen) negative shiftcorrelative with the global latest Paleocene carbon isotope excursion(CIE). A sharp increase in kaolinite content coincides with the isotopeshift in the Bass River section, analogous to increases found inseveral other records. Carbon and oxygen isotopes remain low andkaolinite content remains high for the remainder of the depositionalsequence above the CIE (32.5 ft, 9.9 m), which we estimate torepresent 300-500 k.y. We interpret these data as indicative of anabrupt shift to a warmer and wetter climate along the North Americanmid-Atlantic coast, in concert with global events associated withthe CIE.

RESUME

Une �paisse s�quence, apparemment continue, carott�e �Bass River State Forest, New Jersey, dans le cadre du Leg ODP174AX (Miller, Sugarman, Browning, et al., 1998), a enregistr� enmilieu n�ritique les �v�nements li�s au Latest Paleocene ThermalMaximum (LPTM). L' interpr�tation stratigraphique est fond�e surl' int�gration de donn�es magn�tostratigraphiques et biostrati-graphiques (nannofossiles calcaires et foraminif�res planctoniques).Nous montrons ainsi que l'enregistrement du LPTM en milieu n�ri-tique dans la coupe de Bass River est de qualit� au moins �gale �celui de coupes oc�aniques (DSDP et ODP). Les isotopes du carboneet de l'oxyg�ne, mesur�s sur les tests de foraminif�res benthiques,montrent un shift majeur (4.0 ä: carbone, 2.3 ä: oxyg�ne) corr�latifde l'excursion isotopique du carbone de la fin du Pal�oc�ne (CIE).Comme dans d'autres coupes, ce shift isotopique dans la coupe deBass River coincide avec une augmentation abrupte de la teneur enkaolinite. Ces deux �v�nements se sont produits au cours d'un inter-valle estim� � <10 k.y. Les valeurs isotopiques de l'oxyg�ne et ducarbone restent faibles, et la teneur en kaolinite reste �lev�e dansliÕintervalle de 32.5 ft (9.9 m) susjacent � lÕexcursion et repr�sentantau moins 300-500 k.y. Nous interpr�tons ces donn�es comme t�moi-gnant d'un changement climatique soudain vers des conditionschaudes et humides le long de la c�te mid-atlantique du continentnord-am�ricain, en liaison avec les �v�nements globaux associ�s

1 Department of Geological Sciences, Rutgers University, Piscataway, NJ 08854, USA2 Institut des Sciences de lÕEvolution, Universit� Montpellier II 34095, Montpellier, cedex 05, France3 Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY 10964, USA4 Department of Geological Sciences, University of Maine at Orono, Orono, ME 04469, USA5 Current address: Department of Geological Sciences, Rutgers University, Piscataway, NJ 08854, USA

1

avec la CIE.

Geological Society of France, Bulletin, in press

An exceptional chronologic, isotopic, and clay mineralogic record of the latest Paleocene thermalmaximum, Bass River, NJ, ODP 174AX

Un enregistrement sédimentaire exceptionnel (chronologie, isotopes et minéralogie) du "LatestPaleocene Thermal Maximum": La coupe de Bass River, New Jersey, ODP Leg 174AX

Benjamin S. Cramer1, Marie-Pierre Aubry2, Kenneth G. Miller1,3, Richard K. Olsson1, James D. Wright4,5,Dennis V. Kent3,5

INTRODUCTION

The latest Paleocene thermal maximum (LPTM) isan apparently unique event in Cenozoic climatic and ocean-ographic history. It is characterized by a large (~4ä) negativecarbon isotope excursion (CIE) lasting < 20 k.y. which hasbeen recorded in both marine and terrestrial records (e.g.,Kennett and Stott 1990, 1991; Koch et al., 1992; Sinha andStott, 1993; Schmitz et al., 1996; Stott et al., 1996). Thecarbon isotope excursion was accompanied by a negativeexcursion (1-3ä) in oxygen isotopes (Kennett and Stott, 1990,1991; Pak and Miller, 1992; Lu and Keller, 1993; Braloweret al. 1995; but see Aubry, 1998 for discussion), a benthicforaminiferal extinction event in deep marine records (e.g.,Beckman, 1960; Tjalsma and Lohmann, 1983; Miller et al.1987a; Thomas, 1989, 1990a,b, 1991; Katz and Miller, 1991;Pak and Miller, 1992; see Thomas, 1998 for further references),and a mammalian turnover in terrestrial records, marking theClarkforkian/Wasatchian boundary in North America and thebase of the Sparnacian ÒstageÓ in Europe (Gingerich, 1989;Hooker, 1991, 1996; Koch et al., 1992; Sinha and Stott,1993; Maas et al. , 1995). More recently it has been shownthat certain short-lived and evolutionarily anomalous speciesof planktonic foraminifera (the so-called Òexcursion taxaÓAcarinina africana, Acarinina sibaiyaensis, and Morozovellaallisonensis, Kelly et al., 1996) have their lowest occurrenceat the level of the CIE. The quantity of isotopically negativecarbon required to account for the large excursion in isotopiccomposition of marine dissolved carbonate and atmosphericcarbon dioxide has so far only been satisfactorily explainedby the dissociation of sediment methane hydrate (Dickens etal. 1995, 1997). According to this hypothesis, long-term warm-ing of oceanic bottom waters throughout the late Paleocene(Shackleton, 1986; Miller et al., 1987b) led to a decreasedsediment thermal gradient which caused an increase in thedepth of the methane hydrate stability field. Dissociation ofpreviously stable methane hydrate could then result in a run-away greenhouse effect.

The LPTM is of interest stratigraphically as a candi-date for the (re)definition of the Paleocene/Eocene (P/E)boundary and from the perspective of climate research as anexample of intense, rapid global warming. Consequently, ithas been the subject of a burgeoning literature detailing the

stratigraphic and paleoenvironmental records of sectionsthroughout the world. This paper presents benthic foramin-iferal carbon and oxygen isotope and clay mineral recordsfrom an apparently complete LPTM section recently drilled(October-November, 1996) at Bass River State Forest, NewJersey (Fig. 1) as part of the New Jersey Coastal Plain DrillingProject, ODP Leg 174AX (Miller, Sugarman, Browning, etal., 1998).

BACKGROUND

ChronostratigraphyThe CIE occurs within planktonic foraminiferal Zone

P5, calcareous nannoplankton Zone NP9, and magnetic polar-ity Chron C24r (Berggren and Aubry, 1998). The P/E boundaryis currently defined as the base of the Ypresian Stage (baseof the Ieper Clay, Belgium) which is correlative with thecalcareous nannoplankton NP10a/b subzonal boundary (Aub-ry, 1998). Nevertheless, many authors have prematurely adopt-ed the use of the CIE and associated stratigraphic markers tocharacterize the P/E boundary. The P/E boundary may be(re)defined at the stratigraphic level of the CIE in recognitionof the variety of chemo- and biostratigraphic markers thatcan serve for long distance correlations in both marine andterrestrial records. In this paper we maintain the current def-inition of the P/E boundary so as to correlate with the NP10a/bsubzonal boundary, placing the CIE in the latest Paleocene.

Climate and EnvironmentThe carbon isotope excursion (CIE) was first de-

scribed by Kennett and Stott (1990, 1991) at ODP Sites 689and 690 on Maud Rise in the Southern Ocean. They notedtwo characteristics of the event: first, both carbon and oxygenisotopes rapidly reached more negative values than have beenrecorded at any other time during the Cenozoic, and second,the surface to deep water oxygen isotope gradient becamereversed during the excursion at the deeper-water Site 690compared to Site 689. They interpreted the isotope record asindicating a brief warm episode reaching to high latitudes.High latitude warming was coincident with formation of warmsaline deep waters at low latitudes, causing a reversal in theoxygen isotope gradient at high latitudes. Although numerousisotope records of the LPTM from various marine sectionshave subsequently been published (e.g., Pak and Miller, 1992;Lu and Keller, 1993; Bralower et al., 1995; Thomas andShackleton, 1996), the record from Site 690 remains the moststratigraphically expanded record of the excursion (see Aubryet al., 1996, 1998 for discussion).

The argument that high latitudes experienced Òsub-tropicalÓ warm conditions during the LPTM has been strength-ened by Robert and Kennett (1992, 1994), who documenteda gradual rise in the proportion of kaolinite in clays depositedthrough the late Paleocene and early Eocene at Sites 689 and690. They interpreted this rise as indicative of increased rainfallduring the late Paleocene (Robert and Kennett, 1992). Thebroad rise is punctuated by a sharp peak in kaolinite/illiteand smectite/illite ratios which is coincident with the CIEand was interpreted as indicating abruptly increased precip-itation induced by an increased continent-to-ocean temper-ature gradient (Robert and Kennett, 1994).

Neritic recordsIn the neritic environment the CIE has been recorded

at Gebel Aweina and Gebel Duwi, Egypt (Schmitz et al.,1996). Due to severe diagenetic infilling of foraminiferaltests, Schmitz et al. (1996) generated benthic foraminiferal

2

records using Lenticulina spp. and Frondicularia spp., taxa

having a thick test which allowed infilling calcite to be re-moved with the use of a needle and sonication. The recordfrom Egypt shows lower δ13 C values than the record fromODP 690 with a less pronounced rebound after the CIE.Schmitz et al. (1996) interpreted this as not supporting lowlatitude formation of bottom waters from this part of theTethys Ocean, because carbon isotope values would then beexpected to decrease toward high latitudes. The oxygen isotoperecord from Egypt does not show a substantial excursion,which led Schmitz et al. (1996) to argue against substantiallow-latitude warming and an enhanced greenhouse effect dur-ing the CIE. The oxygen isotope record at ODP Site 865(equatorial Pacific, Bralower et al., 1995) also suggests thattropical sea surface temperatures remained stable throughthe LPTM. However, despite the great care taken by Schmitzet al. (1996) for sample preparation, we cannot exclude thepossibility that the record from Egypt may reflect an incom-plete section and/or diagenetic alteration of original isotopicvalues.

Upper Paleocene-lower Eocene sections were con-tinuously cored at Clayton and Island Beach, New Jersey(Fig. 1). Gibson et al. (1993) noted several significant changesof paleoenvironmental indicators in the Clayton section at alevel stratigraphically correlative with the CIE which hassince been shown to contain a ~8 ä negative excursion inbulk carbonate carbon and oxygen isotopes (Thomas et al.,1997). In the Clayton section the lithology changes from aclayey, glauconitic sand (40+% sand) to a non-glauconiticclay (~1% sand) in the interval containing the CIE. The samelevel marks the beginning of a rise in kaolinite content in thesediment clay fraction which extends ~9 m (~30 ft) abovethe CIE. Kaolinite makes up 4-5% of the clay fraction belowthe CIE and rises to ~50% by 3 m (10 ft) above the CIE.Gibson et al. (1993) interpreted these and other events asindicative of increased precipitation and higher temperaturesaccompanied by low-oxygen bottom-water conditions and anincrease in water depth along the New Jersey coast. Pak etal. (1997) generated an isotope record through the upperPaleocene-lower Eocene at Island Beach and correlated anegative excursion in carbon and oxygen isotopes near theP/E boundary with the CIE; however, the exact stratigraphiccorrelation of the excursion was uncertain because of an un-conformity at this level.

BASS RIVER BOREHOLE

The Bass River core was drilled by Rutgers Universityand the New Jersey Geological Survey during October-November, 1996, and is the first site drilled as part of ODPLeg 174AX, complementing shelf drilling by Leg 174A (Aus-tin, Christie-Blick, Malone, et al., 1998). The site was thefourth hole drilled as part of the New Jersey Coastal PlainDrilling Project, following Leg 150X drilling at Island Beach,Cape May, and Atlantic City (Miller et al., 1994a, b, 1996)(Fig. 1). A full description of the Bass River core and onsiteoperations is published in Miller, Sugarman, Browning, etal. (1998). Paleomagnetic (both AF demagnetization and Kir-schvink analysis), clay mineral, stable isotope, and sedimentdata generated for this study are available from the WorldData Center-A for Paleoclimatology, 325 Broadway, Boulder,CO, USA; http://www.ngdc.noaa.gov/paleo/paleo.html;email: paleo@ngdc.noaa.gov.

LithostratigraphyThe lithostratigraphy of the upper Paleocene-

lowermost Eocene interval in New Jersey has been somewhatuncertain. Olsson and Wise (1987a, b) noted the presence of

a Ògray clay, silt, and sand unitÓ in the subsurface separatingthe upper Paleocene Vincentown Formation from the lowerEocene Manasquan Formation. Gibson et al. (1993) correlatedthis unit with a section in the Clayton core which they assignedto the Manasquan Formation. They place the Vincentown/Ma-nasquan contact at the lithologic change where the CIE hasbeen subsequently shown to occur (Thomas et al., 1997). Wenote that this unit at Clayton is not typical Manasquan lithology(glauconitic sandy green clay) and that the interval below theCIE is not typical Vincentown lithology (quartz sand to glau-conite sand). Liu et al. (1997) correlated an upper Paleoceneunit of interbedded gray clay and sandy silt in the IslandBeach core with the unnamed unit of Olsson and Wise (1987a,b). Owens et al. (1997) assigned this unit to the VincentownFormation. We adopt the terminology of Olsson and Wise(1987a, b), assigning gray clays, silts, and fine sands of theuppermost Paleocene in New Jersey to an unnamed unit litho-logically distinct from the Vincentown Formation below andthe Manasquan Formation above.

Miller, Sugarman, Browning, et al. (1998) locatetwo depositional sequences within the upper Paleocene Vin-centown Formation, in which they include the unnamed unitof Olsson and Wise (1987a, b): 1138.6-1240.8 ft (347.05-378.20 m) and 1243.9-1256.5 ft (379.14-382.98 m) (Fig. 2).Surfaces were observed in the core at 1240.8 and 1243.9 ft(378.20 and 379.14 m), either of which could be the boundarybetween these two sequences. Sediment becomes glauconiticand sandy below ~1236 ft (~376.7 m), just above the sequenceboundary, and we assign this interval and the glauconiticmuds between 1241.5 and 1248.5 ft (378.41 and 380.54 m)to the Vincentown Formation. Above the basal glauconiticsandy silts the uppermost Paleocene sequence (1138.6- 1236ft, 356.92-376.73 m) is a very slightly sandy, slightly glauc-onitic silty gray clay containing evidence of bioturbation andvery few sedimentary structures which becomes a slightlysilty greenish-gray clay above ~1171 ft (~356.9 m). We cor-relate these clays and silty clays with the unnamed unit ofOlsson and Wise (1987a, b) and Liu et al. (1993, IslandBeach borehole) and the upper Vincentown-Manasquan For-mations of Gibson et al. (1993, Clayton borehole) (Fig. 2). Adistinct unconformity at 1138.6 ft (347.05 m) separates theuppermost Paleocene unnamed unit from the lower EoceneManasquan Formation. A very thin (4.0 ft, 1.22 m) sequenceoccurs in the lowermost Manasquan with an upper sequenceboundary marked by a surface at 1134.6 ft (345.83 m).

Core samplingSamples for planktonic foraminiferal, isotope, and

clay mineral analysis, taken at a sampling interval of 1-5 ft.(30-150 cm), were dried in a 40¡C oven and weighed dry.Samples were disaggregated in sodium metaphosphate at roomtemperature and washed through a 63 µm screen. The fine(<63 µm) fraction was saved for use in preparing samplesfor clay analysis. The sand (>63 µm) fraction was dried in a40¡C oven and weighed to determine sediment weight percentsand. Splits were separated from the sand fraction for isotopeanalysis. Samples for calcareous nannofossil studies weretaken at a sampling interval of ~1 ft (~30 cm), with increasedsample density in the vicinity of distinct lithologic surfaces.

Ten to fifteen cm3 quarter-round sections of corewere taken for magnetostratigraphic analysis. Samples weretaken at 5-10 ft (150-300 cm) intervals through the sectionwith additional samples at 1-2 ft (30-60 cm) intervals tobetter resolve magnetozone boundaries.

Biostratigraphy

3

Calcareous nannoplankton biostratigraphy (Fig. 2)

is discussed in Miller, Sugarman, Browning, et al. (1998)and Aubry et al. (manuscript in prep.). Preservation of calcar-eous nannofossils is excellent throughout the upper Paleocene-lower Eocene. The interval between 1242.5 and 1244 ft (378.71and 379.13 m) belongs to Zone NP6 while the interval between1211.4 and 1240.0 ft (370.36 and 377.95 m) belongs to ZoneNP8. Zone NP7 is missing, suggesting that the surface notedat 1240.8 ft (378.20 m) is an unconformity. The base ofZone NP9 is placed between samples at 1210.6 and 1211.4 ft(368.99 and 369.24 m) based on the lowest occurrence (LO)of Discoaster multiradiatus at 1210.6 ft (368.99 m). Basedon the highest occurrence (HO) of Heliolithus megastypus at1171.1 ft (356.95 m) and the LO of Rhomboaster calcitrapa,Discoaster araneus, and D. anartios at 1170.9 ft (356.89 m),Aubry et al. (manuscript in prep.) subdivide Zone NP9 intoSubzones NP9a and NP9b. Zone NP9 is truncated by theunconformity at 1138.6 ft (347.05 m). The juxtaposition ofthe LOs of Tribrachiatus digitalis and T. bramlettei at 1138.4ft (346.98 m) indicates that the sediment immediately abovethe unconformity belongs to Subzone NP10b (subzonalscheme of Aubry, 1996). Subzone NP10b is restricted to avery thin interval, with the LO of T. contortus at 1138.1 ft(346.89 m) indicating Subzone NP10d. Subzone NP10c ismissing, suggesting that there is a hiatus which may correspondto a subtle surface in the core noted at 1138.25 ft (346.94 m).The HO of T. contortus at 1136.2 ft (346.31 m) determinesthe top of Subzone NP10d. The NP10/NP11 contact is uncon-formable.

Planktonic foraminiferal biostratigraphy (Fig. 2) isdiscussed in Miller, Sugarman, Browning, et al. (1998). Plank-tonic foraminiferal zonal boundaries are uncertain due to poorpreservation of foraminifera between samples at 1225.3 and1170.0 ft (373.47 and 356.62 m) and inconsistencies in theoccurrence of datums between 1230 and 1245.4 ft (374.90and 379.60 m). Planktonic foraminiferal Zone P4 is identifiedin samples from 1225.3 to 1247.5 ft (373.47 to 380.24 m).The base of Subzone P4c is identified at 1230.0 ft (374.90m) by the LO of Acarinina soldadoensis and the top of SubzoneP4a is identified at 1247.5 ft (380.24 m) by the HO of Acarininasubsphaerica. The interval from 1234.3 to 1245.4 ft (376.22to 379.60 m) is assigned to Subzone P4b, although the LO ofM. aequa, generally found in association with the LO of A.soldadoensis, at 1245.4 ft (379.60 m) suggests that this intervalmay actually belong to Subzone P4c. In either case, thisconflicts with identification of calcareous nannofossil ZoneNP6 below 1242.5 ft (378.71 m). The HO of G. pseudome-nardii at 1225.3 ft (373.47 m) marks the top of Zone P4.Poor preservation of planktonic foraminifera between samplesat 1225.3 and 1170.0 ft (373.47 and 356.62 m) preventsconfident assignment of this interval to Zone P5. The intervalfrom 1145 to 1170 ft (349.00-356.62 m) is confidently assignedto Zone P5 based on the presence of Morozovella aequa, M.gracilis, M. subbotinae, M. velascoensis, and Acarinina sol-dadoensis. In addition, three species (Acarinina africana, A.sibaiyaensis, and Morozovella allisonensis) that are associatedwith the LPTM at ODP Site 865 in the tropical Pacific (Kellyet al., 1996) occur in the interval beginning at 1171 ft (356.9m). Acarinina africana and A. sibaiyaensis are found through-out the interval from 1171 ft to 1145 ft (356.9 to 349.0 m)whereas M. allisonensis is confined to a thin interval from1171 to 1165 ft (356.9 to 355.1 m). Sediments between 1110and 1130 ft (338.33 and 344.42 m) are assigned to Zone P6bbased on the LO of Morozovella formosa formosa at 1130 ft(344.42 m). A sample at 1137 ft (346.56 m) is assigned tolowermost Eocene Zone P6a, based on the absence of Moro-zovella velascoensis and M. formosa formosa and the co-occurrence of Pseudohastigerina wilcoxensis, Morozovella

formosa gracilis, and Morozovella aequa.

MagnetostratigraphyNatural remanent magnetism (NRM) was measured

on a 2G DC SQUID 3-axis cryogenic magnetometer atLamont-Doherty Earth Observatory. Step demagnetization ofNRM at 5 mT increments to 45 mT was performed using alarge-volume alternating field (AF) demagnetization coil. Themost stable paleomagnetic direction was calculated from lineardemagnetization trends using the method of least-squares anal-ysis (Kirschvink, 1980). In most cases a series of 3-4 demag-netization steps between 20 and 45 mT were used includingthe origin; demagnetization patterns indicate that any over-prints are removed by around 20 mT, allowing the character-istic magnetization to be isolated.

Paleomagnetic data were divided into three qualitycategories based on the demagnetization pattern (Fig. 2). 1)Good demagnetization patterns with linear demagnetizationtrends (MAD angles <5¡) are typical of samples from mostof the clay-rich uppermost Paleocene sequence. 2) Noisydemagnetization trajectories which nonetheless allow at leastthe polarity of the characteristic component to be reasonablydetermined (MAD angles from 5¡ to 15¡) are typical of samplesfrom sandier portions of the section. 3) Demagnetization pat-terns showing hardly any linearity (MAD angles >15¡) oftenappear to be associated with intervals surrounding a reversalin magnetic polarity.

Based on clearly reversed samples at 1247.2 and1253.1 ft (380.15 and 381.24 m) the interval from 1247.2 to1253.1 ft (380.15 to 381.24 m) is tentatively interpreted as areversed polarity interval (Fig. 2). Integration with biostratig-raphy suggests that this interval represents Chron C26r. Sam-ples between 1243.7 and 1246.0 ft (380.15 and 379.78 m)may be overprinted and are not interpreted. All samples from1193.00-1241.30 ft (363.63-378.35 m) show normal polarityand this interval is confidently interpreted as representingChron C25n, although we cannot completely rule out over-printing. We note that this conflicts with assignment of theinterval from 1234.3 to 1241.3 ft (376.22 to 378.35 m) toplanktonic foraminiferal Zone P4b. Samples between 1137.00and 1191.60 ft (346.56 and 363.20 m) show reversed polarityexcept for a sample at 1174.30 ft (357.93 m) which showsevidence of overprinting. The reversed polarity interval from1137.0 to 1191.6 ft (346.56 to 363.20 m) is confidently inter-preted as representing Chron C24r. Samples above 1135.00ft (345.19 and 345.95 m) show inconsistent polarity and aregenerally very weakly magnetized. We confidently interpretsamples at 1125.0 and 1135.0 ft (342.90 and 345.95 m) asnormal polarity and one sample at 1129.9 ft (344.39 m) asreversed polarity and tentatively correlate this interval withChron C24n with possibly either C24n.1r or C24n.2r repre-sented at 1129.9 ft (344.39 m).

Isotope analysisSamples of Cibicidoides spp. (mostly C. succedens,

C. alleni and C. cf. vulgaris), Anomalinoides acuta, Gyroid-inoides octocamerata and Stensioina (Gavelinella) beccari-iformis were picked for carbon and oxygen isotope analysis.Sample spacing was 5-10 ft (150-300 cm) below 1190 ft(362.7 m) and above 1160 ft (353.6 m) and 1-2 ft (30-60 cm)between 1160 and 1190 ft (353.6 and 362.7 m). Above 1173ft (357.53 m) preservation is excellent; below 1173 ft (357.53m) preservation varies from poor to very poor (Fig. 3). Verypoorly preserved specimens of C. succedens lacking an aper-ture are virtually indistinguishable from poorly preserved Nut-talides truempyi. Although every effort was made to run

4

only specimens which could be confidently identified, low

abundances required that some specimens were run whichlacked an aperture. It is not certain that these samples werestrictly Cibicidoides spp. and this may account for the vari-ability in the isotope record (especially oxygen) below 1173ft (357 m). Published values for offsets of Nuttalides fromCibicidoides are 0.0ä δ13 C and -0.15ä δ18 O (Shackletonet al., 1984) and -0.12ä δ13 C and -0.23ä δ18 O (Pak andMiller, 1992), so any mixture of genera should have a minimaleffect on the isotope records.

Most samples of Cibicidoides spp., A. acuta , and G.octocamerata were sonicated for <2 sec in distilled water.Stensioina beccariiformis samples and some Cibicidoides spp.samples below 1173 ft (357 m; just below the base of theCIE) were not sonicated due to the fragility of the specimens.Isotope analysis was performed at the University of Maine,Orono. The samples were reacted in phosphoric acid (H3PO4)at 90¡C in an Autocarb peripheral attached to a MicromassPrism II mass spectrometer. The analytical precision of theNBS19 and NBS20 standards analyzed during the sampleruns was 0.06 and 0.05ä for δ18 O and δ13 C values, respec-tively.

The foraminiferal isotopic record (Fig. 4) is a com-posite of spliced, overlapping records using assemblages ofCibicidoides spp., A. acuta, G. octocamerata, and S. beccari-iformis. Values for Cibicidoides spp. are ~-1ä in oxygenand ~1ä in carbon below 1173 ft (357.53 m). A majornegative excursion in both carbon and oxygen isotopes occursbeginning at 1173.0 ft (357.53 m) with the lowest values(-3.18 δ13 C, -3.69 δ18 O) occurring at 1171.1 ft (356.95 m).Between 1173.0-1171.1 ft (357.53-356.95 m) the magnitudeof the excursion in Cibicidoides spp. is -4.0ä in δ13 C and-2.3ä in δ18 O. Between 1169 and 1146 ft (356.31 and 349.3m) values recover gradually from -3 to -1ä in carbon andfrom -3 to -2ä in oxygen.

Splicing of the isotopic record is necessary becauseof an abrupt change in the benthic assemblage between 1173.0and 1171.1 ft (357.53 and 356.95 m) (Cramer, 1998 andunpublished). No specimens of S. beccariiformis are presentabove 1171.1 ft (356.95 m) and no specimens of G. octocam-erata or A. acuta are present below 1171.8 ft (357.16 m).There is a correlative change in Cibicidoides species assem-blage. Cibicidoides spp. samples below 1173.0 and above1165.0 ft (below 357.53, above 355.09 m) are generally amixture of C. alleni and C. succedens while those between1171.8 and 1166.0 ft (357.16-355.40 m) are generally of C.cf. vulgaris. We note in this context that the relatively highvalues in both carbon and oxygen isotopes (-0.85 ä and-2.18 ä, respectively) in one Cibicidoides spp. sample at1171.1 ft (356.95 m) was obtained using a monospecificsample of C. succedens. Cibicidoides succedens are very rarein the interval from 1171.1 to 1165.0 ft (356.95 to 355.09 m)and it is likely that all C. succedens specimens found in thisinterval were bioturbated upward from below the level of theexcursion and therefore show relatively high isotopic values.

Clay mineral analysisClay analysis was run on the ~<2 µm fraction at a

sampling interval of 1-4 ft (30-120 cm) from 1136.2-1219.4ft (346.31-371.67 m). The <63 µm sieved fraction was placedin 1 L settling columns filled with distilled water. The <2µm size fraction was obtained by allowing the suspension tosettle for 3.5 hours at room temperature and pipetting off theupper 4 cm of liquid. The clay fraction was treated with 5%acetic acid to remove carbonate and allowed to dry to apaste. Smear slides were prepared from the paste and thesurface of the slide was smoothed with a spatula. This methodhas been found to achieve well oriented samples which yield

diffraction patterns nearly identical to those obtained by suc-tion onto a ceramic tile (Theisen and Harward, 1962; Gibbs,1965). The slides were analyzed dry and after glycol solvationat the New Jersey Geological Survey on a Diano 2100e X-raydiffractometer (XRD) using a Cu target. Counts per second(CPS) were measured at 0.04 ¡2Θ increments from 5 - 33¡2Θ.

The clay assemblage consists primarily of kaoliniteand a mixed layer illite/smectite (>80% illite), possibly witha minor component of illite. Illite and the mixed-layer il-lite/smectite are treated as a single illite/smectite component.It is possible that there is a glauconite component present,which shares some refraction angles with illite. A glauconite001 refraction should occur at the same angle as illite 001;however, no significant glauconite refraction peak shouldoccur at the same angle as illite 002 (Moore and Reynolds,1989). Therefore, any significant variation through the sectionin the amount of glauconite clay present should cause variationin the ratio of the intensities of the illite 001 and 002 peaks(which are expressed as illite-smectite 001/004* and 002/003peaks in this section). No discernible variation is observed inthis ratio, so we assume that glauconite does not contributesignificantly to variation in clay mineral abundances.

Quantitative analysis was performed on the glycolsaturated traces using procedures described in Moore andReynolds (1989). All ratios of kaolinite to illite/smectite peaks(kaolinite 001 and 002 and illite/smectite 001/004*, 002/003and 003/005) show the same variation through the section,indicating that the results of the quantitative analysis arerobust. For details of clay mineral analysis procedures seeCramer (1998).

At 1214.5 ft (370.18 m), kaolinite is essentially absentfrom the clay assemblage (Fig. 5). Kaolinite content graduallyrises to ~34% at 1158.0 ft (352.96 m) and declines to <2% atthe upper unconformity (1138.7 ft, 347.08 m). Superimposedon the gradual rise is a sharp spike between 1171.8 ft and1171.1 ft (357.16-356.95 m) with values of 32-36% kaolinite(Figs. 5, 6). This spike indicates a pulse of kaolinite coincidentwith the CIE which increased the kaolinite content of thesediment being deposited at this site by >10% over the first-order variation.

Coarse fractionThe percentage of sand in the sediment was calculated

based on the weight of samples before and after washingthrough a 63 µm sieve. Relative proportions of componentsof the sand fraction were determined by visual examinationunder a light microscope. The total coarse fraction (sand-sized,>63 µm) below 1173.0 ft (357.53 m) varies between 5-20%(Fig. 5). From 1171.8 ft (357.16 m) to the upper unconformity(1138.7 ft, 347.08 m), the total sand fraction is only ~1%,occasionally increasing to 3%. The one exception is the sampleat 1171.1 ft (356.95 m) which contains ~10.3% sand-sizedmaterial, nearly all of which is (predominantly planktonic)foraminifera (Fig. 6). From 1171.1 ft (356.95 m) to the topof the section, foraminifera make up 80-100% of the sandfraction, while below 1175.2 ft (358.2) the sand fraction isdominantly glauconite and quartz. Glauconite grains in thisinterval are well sorted and of the same size and shape asquartz grains, indicating that the glauconite and quartz weretransported by the same mechanism. Glauconite is thereforetreated as a terrigenous component derived from older exposedmarine deposits rather than an authigenic component producedduring the late Paleocene.

5

DISCUSSION

StratigraphyIntegration of paleomagnetic (Cramer, 1998; this

study), calcareous nannofossil (Aubry in Miller, Sugarman,Browning, et al., 1998 and Aubry et al., manuscript in prep.),and planktonic foraminiferal (Olsson, in Miller, Sugarman,Browning, et al., 1998) zonations provide a reliable chronologyfor the uppermost Paleocene sequence at Bass River (Figs. 2,7). In constructing an age model, we use the Chron C25n/C24rreversal, the LO of D. multiradiatus (marking the base ofnannofossil Zone NP9), and the CIE, which essentially coin-cides with the LO of R. calcitrapa (base of nannofossil SubzoneNP9b) (Table 1, Fig. 7). Our rationale in choosing thesethree datums is twofold: 1) to pick datums which are mostlikely to be reliable in the Bass River borehole and which donot directly contradict other stratigraphic datums; and 2) topick datums which will allow correlation of the Bass Riverrecord with the global record of the LPTM. The reversal inmagnetic polarity between 1191.6 and 1193.0 ft (363.20 mand 363.63 m) is very clearly delineated (Fig. 2) and itsoccurrence within Zone NP9 indicates that this is the ChronC25n/C24r reversal. The LO of D. multiradiatus is clearlydelineated in an interval of good nannofossil preservation.The CIE itself (1171.1 ft, 356.95 m) is also used as a chrono-logic datum, taking the age of 55.52 Ma determined by Aubryet al. (1996) based on the ODP Site 690 stratigraphy relativeto the Berggren et al. (1995) timescale. The fact that the CIEin the Bass River section falls on a sedimentation rate lineextrapolated from the LO of D. multiradiatus and the ChronC25n/C24r reversal (Fig. 7) lends credence to the age calcu-lated by Aubry et al. (1996) and to the use of the CIE as astratigraphic datum.

In using the Chron C25n/C24r reversal, the LO ofD. multiradiatus, and the CIE as stratigraphic datums werecognize that these are not entirely consistent with planktonicforaminiferal stratigraphy and magnetostratigraphic interpre-tation in the interval from ~1220 ft (371.86 m) to the uncon-formity at 1240.8 ft (378.20 m) (Fig. 7). The low position ofthe HO of G. pseudomenardii and the high position of theLO of M. subbotinae are easily explained by poor preservationof planktonic foraminifera between 1170 and 1225, as notedabove. We note that, although an age for the HO of A. mckannaiis provided in BKSA95, this datum is generally less reliablethan the others provided. The LO of A. soldadoensis is es-sentially consistent with a sedimentation rate line extrapolatedfrom the Chron C25n/C24r reversal and the LO of D. multi-radiatus. However, we hesitate to interpret this LO as repre-senting the FAD of the species due to the LO of M. aequa at1245.4 ft (379.60 m), which should occur in association withthe LO of A. soldadoensis. In addition, magnetostratigraphicinterpretation of Chron C25n extending downward at least tothe unconformity at 1240.8 ft (378.20 m) is more consistentwith the LO of M. aequa at 1245.4 ft (379.60 m) than withthe LO of A. soldadoensis at 1230.0 ft (374.90 m). However,we cannot rule out the possibility that the magnetic polarityrecord may be overprinted in this interval. These conflictsmay be resolved in future more detailed planktonic foramin-iferal and magnetostratigraphic studies.

Our age model indicates a sedimentation rate of ~19.3m/m.y. (Fig. 7). The sedimentation rates above the CIE andbelow the LO of D. multiradiatus are poorly constrained.However, for convenience and in order to estimate rates ofenvironmental changes we apply a uniform sedimentationrate of 19.3 m/m.y. throughout the sequence, noting that inreality this rate may vary significantly. This sedimentationrate indicates an age at the upper sequence boundary (1138.7

ft, 347.08 m) of 55.0 Ma which we note is a minimum agesince the NP9/NP10 biochronal boundary has an age estimateof 55.0 Ma (BKSA95). We estimate a maximum age for thissequence boundary (1138.7 ft, 347.08 m) of 55.3 Ma basedon assuming a reasonable sedimentation rate (<50 m/m.y.)above the CIE. Extrapolation from the LO of D. multiradiatusto the lower sequence boundary (1240.8 ft, 378.20 m) indicatesan age of 56.6 Ma. We estimate an age for this sequenceboundary (1240.8 ft, 378.20 m) of 56.3-56.6 Ma, with theminimum age estimated assuming that the magnetostrati-graphic interpretation is correct. Our age model and theseconstraints on the ages of the two sequence boundaries allowscorrelation of the Bass River record of the LPTM with theglobal record with a high degree of precision.

We recognize, though, that the accuracy of theseages depends on a chronology derived from a compositesection for Chron C24r derived using DSDP Site 550 (lateChron C24r) and ODP Site 690 (early Chron C24r) (Aubryet al., 1996). As recognized by Aubry et al. (1996) and Berg-gren and Aubry (1998), a numerical chronology will be firmlyestablished only when a continuous section allowing accurateestimation of sedimentation rates through Chron C24r is re-covered. Moreover, there is currently no means to determinethe completeness of sections at or close to the NP9a/b subzonalboundary. The Bass River borehole provides a record of Sub-zone NP9b and the LPTM, which are preserved in very fewother sections (Aubry, 1998). Although this record is consistentwith the Site 550/690 composite section it does not contain acomplete section through Chron C24r and we are currentlyunable to determine if a gap occurs within Subzone NP9a.Yet the correlation that we observe between the isotopic shifts,the kaolinite enrichment, the occurrence of the planktonicforaminifera excursion taxa and that of the LPTM calcareousnannoplankton taxa (Aubry, unpublished data) as seen inseveral deep sea records, indicates that the Bass River sectionyields an essentially complete record of the LPTM. Althoughthe relative chronology is firm, we emphasize that the agesand durations calculated based on our age model must betreated with caution until an independent chronology is de-rived.

Significance of the Bass River sectionThe Bass River section provides an apparently com-

plete record of events surrounding the LPTM extending fromnear the beginning of magnetic polarity Chron C25n to nearthe end of Biochron NP9, an estimated 800-1100 k.y. priorto and 200-500 k.y. after the CIE (Fig. 5). There are apparentlyno hiatuses in this record, although, as in any sedimentarysection, we cannot rule out the possibility of small stratigraphicgaps (<<100 k.y.). The Bass River section is therefore one ofthe most stratigraphically complete records of the LPTM.Good to excellent preservation of foraminifera throughoutthe section has allowed the generation of a carbon and oxygenisotopic record through the LPTM providing a record of cli-mate changes in a neritic setting. The location of Bass River

6

in a neritic setting also allows examination of changes in the

Table 1. Age model for the uppermost Paleocene section,Bass River.

Datums Depth interval (ft) Mean depth Agelower upper (ft) (m) (Ma)

CIE 1171.5 1170.4 1170.95 356.906 55.52C25n/C24r 1193.0 1191.6 1192.30 363.413 55.904

NP8/NP9 1210.6 1211.4 1211.00 369.113 56.2

terrestrial climate through such proxies as clay mineralogy.In addition, identification of three planktonic foraminiferaltaxa previously identified at ODP Site 865 (Kelly et al.,1996) confirms the global distribution of these taxa whichevidently evolved in response to environmental changes dur-ing the LPTM. Studies at Bass River have also allowed thedocumentation of a faunal turnover in calcareous nanno-plankton at the LPTM (Aubry, unpublished data).

Implications for New Jersey lithostratigraphyThe Bass River section provides an opportunity to

reevaluate previous records of the upper Paleocene-lowerEocene from Clayton and Island Beach, NJ. As in the Claytonsection, the CIE at Bass River marks the transition betweenthe slightly sandy, glauconitic lower and clay-rich upper por-tions of the uppermost Paleocene unnamed unit (compareFigs. 2, 5 in this study with Fig. 3 of Gibson et al., 1993).Based on a lithologic and stratigraphic comparison betweenIsland Beach, Bass River, and Clayton (Cramer, 1998), it isclear that the section at Island Beach does not contain theupper, sand-barren portion of this unit (compare Fig. 2 ofLiu et al., 1997). Pak et al. (1997) tentatively correlated anisotope excursion in the lowermost Manasquan sediment atIsland Beach with the CIE. Aubry et al. (manuscript in prep.)have demonstrated that the unconformity at Island Beach atthe level of the excursion noted by Pak et al. (1997) separatesnannofossil Subzone NP9a from Subzone NP10a with only~1 ft (0.3 m) representing Subzone NP9b just below theunconformity. This implies that the isotopic record from IslandBeach is a juxtaposition of the very initial portion of the CIEwith lighter lower Eocene isotopic values.

Isotope shiftThe shift in isotopic values at Bass River is large in

comparison with other records. In general shape, the carbonisotope record is similar to records from Gebel Aweina andGebel Duwi, Egypt (Schmitz et al. 1996) (Fig. 8), showingthe persistence of low δ13 C values for >200-400 k.y. afterthe shift; in contrast, deep-sea records show a nearly fullrecovery to pre-excursion values within <100 k.y. (e.g. Site690 (Fig. 8): Kennett and Stott 1990, 1991; Site 865: Braloweret al. 1995). For this reason we characterize the isotopicrecord at Bass River as containing a ÒshiftÓ in isotopic valuesrather than an ÒexcursionÓ which implies both a shift and arapid recovery. This shift is correlative with the latest Pale-ocene CIE and occurred on similarly short timescales (<10k.y., Fig. 6). It is likely that the persistence of low δ13 Creflects differences in cycling of carbon between shelf envi-ronments and open-ocean environments.

The oxygen isotope record at Bass River also showsa persistence of low values above the excursion, similar tothe pattern in carbon isotopes. Application of the paleotem-perature equation assuming no ice volume (δw=-1.2ä, Shack-leton and Kennett, 1975) implies an increase in bottom watertemperature on the shelf from ~14¡C pre-excursion to ~20¡Cpost-excursion with a peak temperature of 25¡C at the CIE.This is most likely a maximum estimate, since increasedprecipitation could result in a significant decrease in δ18 O. Ifthe ~-1.8 ä shift in δ18 O values between 1173.9 ft (357.81)and 1170.4 ft (356.74 m) was due entirely to freshwaterinput this would imply a decrease in salinity of ~8 ä (assumingmodern Delaware River, PA δ18 O of ~-8 ä, Fairbanks, 1982)to ~12 ä (assuming modern Roanoke River, NC δ18 O of -5ä, Fairbanks, 1982). Realistically, the oxygen isotopic recordfrom Bass River must reflect some combination of temperatureand freshwater input. Warming (and increased precipitation)occurred in less than 20 k.y. with a slight rebound to cooler

temperatures within ~10 k.y. immediately following the CIEand gradual cooling continuing for at least the subsequent300-500 k.y. (Figs. 5, 6).

In contrast, neritic isotope records from Egypt showno significant excursion in oxygen (Fig. 8; Schmitz et al.1996). Although the oxygen isotopic record from Egypt mayreflect diagenetic alteration, one other published dataset sup-ports the idea that tropical sea-surface temperatures may, infact, have been stable through the LPTM (ODP Site 865,equatorial Pacific, Bralower et al., 1995). The record at BassRiver demonstrates that warming did occur in shallow watersat relatively low latitudes. We speculate that equatorial recordsdo not show warming during the LPTM due to enhancedpoleward transport of heat through the production of warmintermediate/deep waters in equatorial regions.

This speculation is not consistent with Schmitz etal. (1996), who rejected the hypothesis of production of warmdeep water in the Egypt-Israel portion of the Tethys duringthe LPTM. Their conclusion was based on the observationthat δ13 C at Gebel Aweina and Gebel Duwi, Egypt, wassignificantly more negative during the LPTM than δ13 C atODP Site 690 (Fig. 8). In the deep sea, bottom water δ13 Cbecomes more negative as it travels away from the sourcedue to the breakdown of organic matter as it settles from thesea-surface and in the surface sediment. However, in neriticsections carbon isotopic values are potentially strongly influ-enced by terrigenous input having extremely negative δ13 Crelative to the marine environment. This is a likely explanationfor the relatively low carbon isotopic values in the recordsfrom Egypt and Bass River.

Clay mineralogyThe clay mineralogy of marine sediment has fre-

quently been used as an indicator of terrestrial climate (e.g.,Biscaye, 1965; Griffin et al. , 1968; Robert and Chamley,1987; Chamley, 1989; Robert and Chamley, 1991, 1992).The three components typically emphasized are kaolinite,illite and smectite. Kaolinite and smectite are both productsof intense chemical weathering on land. Smectite forms inpoorly drained soils in humid regions while kaolinite developsin well-drained soils receiving high precipitation. Illite is themajor constituent of immature soils which have undergonelittle chemical weathering and is characteristic of cold orarid climates or areas of steep relief which interfere with soilformation and prevent extensive chemical weathering. In ma-rine sediments the relative proportion of these minerals gen-erally reflects their abundance in nearby terrestrial environ-ments, although the absolute abundance of each can be affectedby the processes which transport them to the marine environ-ment (see Chamley, 1989, for full discussion).

It is possible that the observed increase in kaolinitecontent across the LPTM in the Bass River section is due todilution by another clay component below ~1175 ft (358.1m) which is greatly reduced above ~1170 ft (356.6 m). It isunlikely that this component is illite, since illite is ubiquitousin the marine environment and its abundance generally re-sponds to dilution by other clays such as kaolinite or smectite(Biscaye, 1965). There could be a glauconite component con-tributing to the observed intensity of illite-smectite peaks.The disappearance of glauconite sand at the level of the CIEis consistent with a reduction in clay-size glauconite. However,as discussed above, we find no evidence in the XRD tracesto support significant presence of clay-sized glauconite inthe section. We therefore conclude that the observed increasein kaolinite content reflects an actual increase in kaolinitesedimentation at the Bass River site during the latest Paleocene.

7

In the Bass River section there is a very abrupt

increase in kaolinite deposition relative to illite/smectite oc-curring in <20 k.y. coincident with the CIE (Fig. 6). Thiry(1998) has noted that it may take as long as one to severalmillion years for the clay minerals in soil profiles to equilibratewith climatic conditions. Therefore, the increase in kaolinitein the Bass River section is much too rapid to reflect productionof kaolinite in the terrestrial environment and must reflecterosion of previously deposited clays. The likely source ofkaolinite in the Bass River section is erosion of exposedUpper Cretaceous deposits in the New Jersey Coastal Plain.In the Cenomanian Raritan Formation, deposited subaeriallyin a flooding plain and mangrove swamp environment, ka-olinite content can reach >90% and is generally ~60% of theclay assemblage (Owens and Sohl, 1969). Kaolinite remainsa significant component of subaerial and marine depositsthroughout the remainder of the Cretaceous. It is likely thaterosion of these Upper Cretaceous deposits was ongoing inthe late Paleocene and that variations in kaolinite content inthe Bass River section must be due either to changes in theproximity of the Bass River site to the kaolinite source and/orto sorting mechanisms operating during transport and dep-osition of clay minerals.

Changes in proximity to source can be effected eitherby movement of a point source (river mouth) or by waterdepth changes affecting the distance from shore. Variationdue to movement of a point source would cause very localizedpatterns of clay sedimentation. The similarity in shape andtiming of the increase in kaolinite at Bass River (this study)and Clayton (Gibson et al., 1993; Thomas et al., 1997) impliesthat the variations in clay sedimentation were not localized.Further investigation is needed to confirm that variations inclay sedimentation were ubiquitous throughout the New Jerseycoastal plain, but the similarity of records at Bass River andClayton is inconsistent with a point source for kaolinite.

It is also unlikely that the variation in clay mineralsis due to water depth variations. In general, kaolinite contentof marine sediments decreases with increasing distance fromshore while smectite and mixed-layer clays increase (e.g.Biscaye, 1965; Chamley, 1989). The abrupt increase in ka-olinite is therefore most consistent with a decrease in waterdepth and distance from shore. The elimination of the terrige-nous sand component at the level of the CIE (1171.1 ft,356.95 m) together with benthic foraminiferal faunal evidence(Cramer, 1998) indicates an increase in water depth. There isa hint that kaolinite sedimentation covaries with sand in theinterval below ~1180 ft (359.7 m), suggesting that kaolinitecontent may in fact be responding to minor changes in waterdepth in this interval, but the high kaolinite content abovethe CIE is clearly unrelated to water depth (Figure 5).

The rapid increase in kaolinite relative to illite/smec-tite coincident with the CIE is most likely due to environmentalchanges differentially affecting transportation of clay miner-als. The δ18 O record indicates an increase in temperatureand a decrease in salinity due to increased precipitation atthe CIE, persisting for the remainder of the section (1138.6-1171.8 ft, 347.05-357.17 m). An increase in precipitation isconsistent with 1) an increase in terrigenous components suchas kaolinite and/or 2) differential transport of kaolinite relativeto illite/smectite into the neritic environment1. Whatever thesorting mechanism, we believe that the precise coincidenceof the kaolinite peak with the carbon and oxygen isotopic

1It has been suggested that kaolinite flocculates readily in relativelylow salinity water and is deposited in estuarine environments whilesmectite and mixed layer clays are preferentially transported intothe marine environment (Chamley, 1987).

shifts (both in New Jersey and high latitude ODP Site 690,

Robert and Kennett, 1992, 1994) and the lack of consistencybetween variations in kaolinite content and other sedimento-logic parameters indicates a climatic control on kaolinite dep-osition during and after the LPTM.

SUMMARY

Isotopic records from the upper Paleocene sectionat Bass River section show major negative shifts in δ13 C andδ18 O correlative with the late Paleocene CIE recognized inmany other marine and terrestrial sections. Integratedmagneto- and nannofossil biostratigraphy demonstrates thatthis record is complete from near the base of Chron C25n tomid-Chron C24r and mid-Zone NP8 to upper Zone NP9b.Therefore, the Bass River section preserves a record of theLPTM in a neritic setting which is apparently as complete asthe deep sea record from ODP Site 690. The magnitude ofthe negative shift in carbon (4.0ä) and oxygen (2.3ä) iso-topes is comparable to the record from ODP Site 690. Thenegative shift in δ18 O indicates that rapid warming (and wet-tening) at relatively low latitudes (Bass River) as well as athigh latitudes (ODP Site 690) occurred during the LPTM.

The isotope excursion is accompanied by a sharpincrease in kaolinite content. The onset of the isotope excursionand the spike in kaolinite content are synchronous and occurwithin < 20 k.y. The increase in kaolinite correlative withthe LPTM observed at ODP Site 690 (Robert and Kennett,1994), Clayton, NJ (Gibson et al., 1993), and in this study atBass River, NJ appears to be a climate signal related toincreased temperature and/or precipitation. The mechanismfor producing this climatic response has not yet been deter-mined.

The climate system in New Jersey apparently didnot return to pre-LPTM conditions during the remainder ofthe Paleocene Epoch. Carbon and oxygen isotopes do notreturn to pre-excursion values after the excursion in the BassRiver section, a pattern which is similar to the neritic recordsfrom Egypt (Schmitz et al., 1996). Persistence of low δ13 Cafter the CIE may be a shallow-water phenomenon; however,a full explanation awaits further study and records from otherneritic sites. Continued high kaolinite content and low oxygenisotope values after the excursion suggest that the climate inthe neritic environment, at least in New Jersey, remainedwarmer and wetter for at least 300-500 k.y. after the excursion.

ACKNOWLEDGMENTS

Funding for Bass River was provided by the New JerseyGeological Survey for direct drilling expenses and the NationalScience Foundation (Earth Sciences Division, Continental DynamicsProgram and Ocean Science Division, Ocean Drilling Program) forscience support. We thank J.V. Browning for helpful discussion ofNew Jersey lithostratigraphy and P. Sugarman, J. Dooley, and theNew Jersey Geological Survey for providing access to the XRD.This study has benefited from suggestions by M. Thiry, W.A. Berg-gren, G.M. Ashley, R.M. Sherrell, and Y. Rosenthal and criticalreviews by R.D. Norris and M. Steinberg. This material is partiallybased upon work supported under a National Science FoundationGraduate Fellowship. This is ISEM contribution 99-024.

REFERENCES CITED

AUBRYÊM.-P. (1996). Ñ Towards an upper Paleocene - lower Eocenehigh resolution stratigraphy. In: AUBRYÊM.-P. & BENJAMINI , C.,Eds., Paleocene/Eocene Boundary Events in Space and Time Ñ

8

Israel Journal of Earth Sciences, 44, 239-253.

AUBRYÊM.-P. (1998). Ñ Stratigraphic (dis)continuity and temporalresolution of geological events in the upper Paleocene-lowerEocene deep sea record. In: AUBRYÊM.-P., LUCASÊS., & BERG-GRENÊW.A., Eds., Late Paleocene-Early Eocene Climatic and Bi-otic Events in the Marine and Terrestrial Records. Ñ ColumbiaUniversity Press, New York, 37-66.

AUBRYÊM.-P., BERGGRENÊW.A., STOTTÊL., & SINHAÊA. (1996). ÑThe upper Paleocene-lower Eocene stratigraphic record and thePaleocene-Eocene boundary carbon isotope excursion: Implica-tions for geochronology. In: KNOXÊR.W.O'B., CORFIELDÊR.M., &DUNAYÊR.E., Eds., Correlation of the Early Paleogene in North-west Europe, Special Publication 101Ñ Geological Society, Lon-don, 353Ð380.

AUSTINÊJ., CHRISTIE-BLICKÊN., MALONEÊM., et al. (1998). Ñ Pro-ceedings of the Ocean Drilling Program, Initial Reports, 174A.Ñ College Station, TX, Ocean Drilling Program.

BECKMANÊJ.-P. (1960). Ñ Distribution of benthonic foraminifera atthe Cretaceous-Tertiary boundary of Trinidad, West Indies. ÑInternational Geological Congress, Report of the 21st Session,Norden, Part V: The Cretaceous-Tertiary Boundary, 57Ð69.

BERGGRENÊW.A., KENTÊD.V., SWISHERÊC.C., & AUBRYÊM.-P.(1995). Ñ A revised Cenozoic geochronology and chronostratig-raphy. In: BERGGRENÊW.A., KENTÊD.V., & HARDENBOLÊJ., Eds.,Geochronology, Time Scales and Global Stratigraphic Correla-tion, Special Publication 54Ñ Society for Sedimentary Geology,129Ð212.

BERGGRENÊW. A. & AUBRYÊM.-P. (1998) Ñ The Paleocene/EoceneEpoch/Series boundary: Chronostratigraphic framework and esti-mated geochronology. In AUBRY, M.-P., LUCAS, S. AND BERGGREN,W. A., Eds., Late Paleocene-Early Eocene Climatic and BioticEvents in the Marine and Terrestrial Records Ñ Columbia Uni-versity Press, New York, 18-36.

BISCAYEÊP.E. (1965). Ñ Mineralogy and sedimentation of recentdeep-sea clay in the Atlantic Ocean and adjacent seas and oceans.Ñ Geological Society of America, Bulletin, 76, 803Ð832.

BRALOWERÊT.J. , ZACHOSÊJ.C., TH O M A S ÊE., PARROWÊM.,PAULLÊC.K., KELLYÊD.C., PREMOLI-SILVAÊI., SLITERÊW.V., & LO-HMANNÊK.C. (1995). Ñ Late Paleocene to Eocene paleoceanogra-phy of the equatorial Pacific Ocean: stable isotopes recorded atOcean Drilling Program Site 865, Allison Guyot. Ñ Paleocean-ography, 10, 965Ð971.

CHAMLEYÊH. (1989). Ñ Clay Sedimentology. Ñ Springer-Verlag,New York, 623 p.

CRAMERÊB.S. (1998). Ñ The late Paleocene Thermal Maximum,Bass River, NJ: Neritic response to a runaway greenhouse event(Masters Thesis). Ñ Rutgers University, New Brunswick, NJ,132 p.

DICKENSÊG.R., O'NEILÊJ.R., REAÊD.K., & OWENÊR.M. (1995). ÑDissociation of oceanic methane hydrate as a cause of the carbonisotope excursion at the end of the Paleocene. Ñ Paleoceanogra-phy, 10, 965Ð971.

DICKENSÊG.R., CASTILLOÊM.M., & WALKERÊJ.G.C. (1997). Ñ Ablast of gas from the latest Paleocene: Simulating first-order effectsof massive dissociation of oceanic methane hydrate. Ñ Geology,25, 259Ð262.

FAIRBANKS ÊR.G. (1982). Ñ The origin of continental shelf andslope water in the New York Bight and Gulf of Maine: Evidencefrom H2

18 O/H216 O ratio measurements. Ñ Journal of Geophys-

ical Research, 87, 5796-5808.

GIBBSÊR.J. (1965). Ñ Error due to segregation in quantitative claymineral X-ray diffraction mounting techniques. Ñ The AmericanMineralogist, 50, 741Ð751.

GIBSONÊT.G., BYBELLÊL.M., & OWENSÊJ.P. (1993). Ñ Latest Pale-ocene lithologic and biotic events in neritic deposits of southwest-ern New Jersey. Ñ Paleoceanography, 8, 495Ð514.

GINGERICHÊP.D. (1989). Ñ New earliest Wasatchian mammalianfauna from the Eocene of northwestern Wyoming: compositionand diversity. Ñ Papers on Paleontology, Museum of Paleontolo-gy, University of Michigan, 28, 321-326.

GRAHAMÊD.W., CORLISS ÊB.H., BENDERÊM.L., & KEIGWINÊL.D.(1981). Ñ Carbon and oxygen isotopic disequilibria of Recentbenthic foraminifera. Ñ Marine Micropaleontology, 6, 483Ð497.

GRIFFINÊJ.J., WINDOMÊH., & GOLDBERGÊE.D. (1968). Ñ The distri-bution of clay minerals in the world ocean. Ñ Deep-Sea Research,15, 433Ð460.

HOOKERÊJ.J. (1991). Ñ The sequence of mammals in the Thanetianand Ypresian of the London and Belgium Basins: Location of thePalaeocene-Eocene boundary. Ñ Newsletters on Stratigraphy,25, 75Ð90.

HOOKERÊJ.J. (1996). Ñ Mammalian biostratigraphy across thePaleocene-Eocene boundary in the Paris, London and Belgianbasins. In: KNOXÊR.W.O'B., CORFIELDÊR.M., & DUNAYÊR.E., Eds.,Correlation of the Early Paleogene in Northwest Europe, SpecialPublication 101Ñ Geological Society, London, 205Ð218.

KATZÊM.E. & MILLERÊK.G. (1991). Ñ Early Paleogene benthicforaminiferal assemblage and stable isotope composition in theSouthern Ocean, ODP Leg 114. In: CIESIELSKI, P.F., KRISTOF-FERSEN, Y., et al., Proceedings of the Ocean Drilling Program,Scientific Results, 114. Ñ College Station, TX, Ocean DrillingProgram, 481Ð512.

KELLYÊD.C., BRALOWERÊT.J., ZACHOSÊJ.C., PREMOLI-SILVAÊI., & TH-OMASÊE. (1996). Ñ Rapid diversification of planktonic foramin-ifera in the tropical Pacific (ODP Site 865) during the late Paleocenethermal maximum. Ñ Geology, 24, 425Ð426.

KENNETTÊJ.P. & STOTTÊL.D. (1990). Ñ Proteus and proto-Oceanus:Ancestral Paleogene oceans as revealed from Antarctic stableisotopic results. In: BARKER, P.F., KENNETT, J.P., et al., Proceed-ings of the Ocean Drilling Program, Scientific Results, 113. ÑCollege Station, TX, Ocean Drilling Program, 865Ð880.

KENNETTÊJ.P. & STOTTÊL.D. (1991). Ñ Abrupt deep-sea warming,palaeoceanographic changes and benthic extinctions at the end ofthe Paleocene. Ñ Nature, 353, 225Ð229.

KIRSCHVINKÊJ.L. (1980). Ñ The least-squares line and plane andthe analysis of palaeomagnetic data. Ñ Geophysical Journal ofthe Royal Astronomical Society, 62, 699Ð718.

KOCHÊP.L., ZACHOSÊJ.C., & GINGERICHÊP.D. (1992). Ñ Correlationbetween isotope records in marine and continental carbon reser-voirs near the Palaeocene/Eocene boundary. Ñ Nature, 358,319Ð322.

LIUÊC., BROWNINGÊJ.V., MILLERÊK.G., & OLSSONÊR.K. (1997). ÑPaleocene benthic foraminiferal biofacies and sequence stratigra-phy, Island Beach borehole, New Jersey. In: MILLERÊK.G. & SNY-DERÊS.W., Eds, Proceedings of the Ocean Drilling Program, Sci-entific Results, 150X. Ñ College Station, TX, Ocean DrillingProgram, 267Ð275.

LUÊG. & KELLERÊG. (1993). Ñ Climatic and oceanographic eventsacross the Paleocene-Eocene transition in the Antarctic IndianOcean: Inference from planktic foraminifera. Ñ Marine Micropa-leontology, 21, 101Ð142.

MAAS ÊM.C., ANTHONYÊM.R.L., GINGERICHÊP.D., GUNNELLÊG.F., &KRAUSEÊD.W. (1995). Ñ Mammalian generic diversity and turn-over in the late Paleocene and early Eocene of the Bighorn andCrazy Mountains Basins, Wyoming and Montana (USA). ÑPalaeogeography, Palaeoclimatology, Palaeoecology, 115,181Ð207.

MILLERÊK.G., JANECEKÊT.R., KATZÊM.E., & KEILÊD.J. (1987a). ÑAbyssal circulation and benthic foraminiferal changes near the

9

Paleocene/Eocene boundary. Ñ Paleoceanography, 2, 741Ð761.

MILLERÊK.G., FAIRBANKSÊR.G., & MOUNTAINÊG.S. (1987b). Ñ Ter-tiary oxygen isotope synthesis, sea level history, and continentalmargin erosion. Ñ Paleoceanography, 2, 1Ð19.

MILLERÊK.G., BROWNINGÊJ.V., LIUÊC., SUGARMANÊP., KENTÊD.V.,VANÊFOSSENÊM., QUEENÊD., GOSS ÊM., GWYNNÊD., MULLIKINÊL.,FEIGENSONÊM.D., AUBRYÊM.-P., & BURCKLEÊL.D. (1994a). Ñ At-lantic City Site Report. In: MILLERÊK.G., et al., Proceedings ofthe Ocean Drilling Program, Initial Reports, 150X. Ñ CollegeStation, TX, Ocean Drilling Program, 35Ð58.

MILLERÊK.G., SUGARMANÊP., VANÊFOSSENÊM., LI UÊC., BROWN-INGÊJ.V., QUEENÊD., AUBRYÊM.-P., BURCKLEÊL.D., GOSSÊM., &BUKRYÊD. (1994b). Ñ Island Beach Site Report. In: MILLERÊK.G.,SUGARMANÊP., et al., Proceedings of the Ocean Drilling Program,Initial Reports, 150X. Ñ College Station, TX, Ocean DrillingProgram, 5Ð33.

MILLERÊK.G., SUGARMANÊP.J., BROWNINGÊJ.V., OLSSONÊR.K., PE-KARÊS.F., RE I L L YÊT.J . , CRAMERÊB.S., AU B R Y ÊM.-P. ,LAWRENCEÊR.P., CURRANÊJ., STEWARTÊM., METZGERÊJ.M., UPTE-GROVEÊJ., BUKRYÊD., BURCKLEÊL.D., WRIGHTÊJ.D., FEIGEN-SON ÊM.D., BRENNERÊG.J., & DALTONÊR.F. (1998). Ñ Bass RiverSite Report. In: MILLERÊK.G., SUGARMANÊP.J., BROWNINGÊJ.V.,et al., Proceedings of the Ocean Drilling Program, Initial Reports,174AX. Ñ College Station, TX, Ocean Drilling Program, 5-43.

MOOREÊD.M. & REYNOLDSÊJ. (1989). Ñ X-Ray Diffraction and theIdentification and Analysis of Clay Minerals. Ñ Oxford UniversityPress, New York, 332 p.

OLSSONÊR.K. & WISEÊS.W. (1987a). Ñ Upper Maestrichtian tomiddle Eocene stratigraphy of the New Jersey slope and CoastalPlain. In: VAN HINTE, J.E., WISE, S.W., JR., et al., Initial Reportsof the Deep Sea Drilling Project, 92. Ñ Washington, DC, U.S.Government Printing Office, 1343Ð1365.

OLSSONÊR.K. & WISEÊS.W. (1987b). Ñ Upper Paleocene to middleEocene depositional sequences and hiatuses in the New JerseyAtlantic Margin. In: ROSS ÊC.A. & HAMANÊD., Eds., Timing andDepositional History of Eustatic Sequences: Constraints on Seis-mic Stratigraphy, Special Publication 24Ñ Cushman Foundationfor Foraminiferal Research, 99Ð112.

O'NEILÊJ.R., CLAYTONÊR.N., & MAYEDAÊT.K. (1969). Ñ Oxygenisotope fractionation in divalent metal carbonates. Ñ Journal ofChemical Physics, 51, 5547Ð5558.

OWENSÊJ.P. & SOHLÊN.F. (1969). Ñ Shelf and deltaic paleoenviron-ments in the Cretaceous-Tertiary formations of the New JerseyCoastal Plain. In SUBITZKY, S., Ed., Geology of Selected Areas inNew Jersey and Eastern Pennsylvania and Guidebook of Excur-sions. Ñ Rutgers University Press, New Brunswick, NJ, 235-278.

OWENSÊJ.P., MILLERÊK.G., & SUGARMANÊP.J. (1997). Ñ Lithos-tratigraphy and paleoenvironments of the Island Beach borehole,New Jersey Coastal Plain Drilling Project. In: MILLERÊK.G. &SNYDERÊS.W., Eds, Proceedings of the Ocean Drilling Program,Scientific Results, 150X. Ñ College Station, TX, Ocean DrillingProgram, 305Ð315.

PAKÊD.K. & MILLERÊK.G. (1992). Ñ Paleocene to Eocene benthicforaminiferal isotopes and assemblages: Implications for deepwa-ter circulation. Ñ Paleoceanography, 7, 405Ð422.

PAKÊD.K., MILLERÊK.G., & BROWNINGÊJ.V. (1997). Ñ Global sig-nificance of an isotopic record from the New Jersey Coastal Plain:Linkage between the shelf and deep sea in the late Paleocene toearly Eocene. In: MILLERÊK.G. & SNYDERÊS.W., Eds, Proceedingsof the Ocean Drilling Program, Scientific Results, 150X. Ñ Col-lege Station, TX, Ocean Drilling Program, 305Ð315.

ROBERTÊC. & CHAMLEYÊH. (1987). Ñ Cenozoic evolution of conti-nental humidity and paleoenvironment, deduced from the kaolinitecontent of oceanic sediments. Ñ Palaeogeography, Palaeoclima-tology, Palaeoecology, 60, 171Ð187.

ROBERTÊC. & CHAMLEYÊH. (1991). Ñ Development of early Eocenewarm climates, as inferred from clay mineral variations in oceanicsediments. Ñ Global and Planetary Change, 89, 315Ð331.

ROBERTÊC. & CHAMLEYÊH. (1992). Ñ Late Eocene-early Oligoceneevolution of climate and marine circulation: Deep-sea clay mineralevidence. In: KENNETTÊJ.P. & WARNKEÊD.A., Eds., The AntarcticPaleoenvironment: A Perspective on Global Change. Ñ AmericanGeophysical Union, Washington, D.C., 97Ð117.

ROBERTÊC. & KENNETTÊJ.P. (1992). Ñ Paleocene and Eocene ka-olinite distribution in the South Atlantic and Southern Ocean:Antarctic climatic and paleoceanographic implications. Ñ MarineGeology, 103, 99Ð110.

ROBERTÊC. & KENNETTÊJ.P. (1994). Ñ Antarctic subtropical humidepisode at the Paleocene-Eocene boundary: Clay-mineral evi-dence. Ñ Geology, 22, 211Ð214.

SCHMITZÊB., SPEIJER ÊR.P., & AUBRYÊM.-P. (1996). Ñ Latest Pale-ocene benthic extinction event on the southern Tethyan shelf(Egypt): Foraminiferal stable isotopic (δ13 C, δ18 O) records. ÑGeology, 24, 347Ð350.

SHACKLETONÊN.J. (1974). Ñ Attainment of isotopic equilibrium be-tween ocean water and the benthonic foraminifera genus Uvigerina:Isotopic changes in the ocean during the last glacial. Ñ ColloqueInternational C. N. R. S., 219, 203Ð210.

SHACKLETONÊN.J. (1986). Ñ Paleogene stable isotope events. ÑPalaeogeography, Palaeoclimatology, Palaeoecology, 57,91Ð102.

SHACKLETONÊN.J. & KENNETTÊJ.P. (1975). Ñ Paleotemperature his-tory of the Cenozoic and the initiation of Antarctic glaciation;oxygen and carbon isotope analyses in DSDP sites 277, 279, and281. In: KENNETTÊJ.P., HOUTZÊR.E., et al., Initial Reports of theDeep Sea Drilling Project, 29. Ñ Washington, D.C., U.S. Gov-ernment Printing Office, 743Ð755.

SHACKLETONÊN.J., HALLÊM.A., & BOERSMAÊA. (1984). Ñ Oxygenand carbon isotope data from Leg 74 foraminifers. In:MOOREÊT.C.ÊJR., RABINOWITZÊP.D., et al., Initial Reports of theDeep Sea Drilling Project, 74. Ñ Washington, D.C., U.S. Gov-ernment Printing Office, 599Ð612.

SINHAÊA. & ST O T TÊL.D. (1993). Ñ Recognition of the Pale-ocene/Eocene boundary carbon isotope excursion in the ParisBasin, France (Abstract). In: Symposium on the Correlation ofthe Early Paleogene in Northwestern Europe, 1-2 December 1993.Ñ Geological Society, London.

SPIESS ÊV. (1990). Ñ Cenozoic magnetostratigraphy of Leg 113drill sites, Maude Rise, Weddell Sea, Antarctica. In: BARKERÊP.F.& KENNETTÊJ.P., Eds, Proceedings of the Ocean Drilling Program,Scientific Results, 113. Ñ College Station, TX, Ocean DrillingProgram, 261Ð318.

STOTT ÊL.D., SINHAÊA., THIRYÊM., AUBRYÊM.-P., & BERGGRENÊW.A.(1996). Ñ Global δ13 C changes across the Paleocene-Eoceneboundary: Criteria for terrestrial-marine correlations. In:KNOXÊR.W.O'B., CORFIELDÊR.M., & DUNAYÊR.E., Eds., Correla-tion of the Early Paleogene in Northwest Europe, Special Publi-cation 101Ñ Geological Society, London, 381Ð399.

THIESENÊA.A. & HARWARDÊM.E. (1962). Ñ A paste method forpreparation of slides for clay mineral identification by X-ray dif-fraction. Ñ Soil Science Society Proceedings, 26, 90Ð91.

THIRYÊM. (1998). Ñ Do marine clays record paleoclimates? ÑProceedings of the 2nd Mediterranean Clay Meeting, 1, 78-97.

THOMASÊE. (1989). Ñ Development of Cenozoic deep-sea benthicforaminiferal faunas in Antarctic waters. In: CRAMEÊJ.A., Eds.,Origins and Evolution of Antarctic Biota, Special Publication

10

47Ñ Geological Society, London, 283Ð296.

THOMASÊE. (1990a). Ñ Late Cretaceous through Neogene deep-seabenthic foraminifers, Maude Rise, Weddell Sea, Antarctica. In:BARKERÊP.F. & KENNETTÊJ.P., Eds, Proceedings of the OceanDrilling Program, Scientific Results, 113. Ñ College Station,TX, Ocean Drilling Program, 849Ð864.

THOMASÊE. (1990b). Ñ Late Cretaceous-early Eocene mass extinc-tions in the deep sea. In: SHARPTONÊV.L. & WARDÊP.D., Eds.,Global Catastrophes in Earth History, Special Publication 247ÑGeological Society of America, Boulder, CO, 481Ð495.

THOMASÊE. (1991). Ñ The latest Paleocene mass extinction of deep-sea benthic foraminifera; results of global climate change (Ab-stract). In Geological Society of America, Annual meeting. ÑAbstracts with Program, 23, 141.

THOMASÊE. (1998). Ñ Biogeography of the late Paleocene benthicforaminiferal extinction. In: AUBRYÊM.-P., LUCASÊS., & BERG-GRENÊW.A., Eds., Late Paleocene-Early Eocene Climatic and Bi-otic Events in the Marine and Terrestrial Records. Ñ ColumbiaUniversity Press, New York, 214-243.

THOMASÊE. & SHACKLETONÊN.J. (1996). Ñ The Paleocene-Eocenebenthic foraminiferal extinction and stable isotope anomalies. In:KNOXÊR.W.O'B., CORFIELDÊR.M., & DUNAYÊR.E., Eds., Correla-tion of the Early Paleogene in Northwest Europe, Special Publi-cation 101Ñ Geological Society, London, 401Ð441.

THOMASÊE., GIBSON ÊT., BYBELLÊL., & ZACHOSÊJ.C. (1997). Ñ Thelate Paleocene thermal maximum in New Jersey (Abstract). InAmerican Geophysical Union, Annual Fall Meeting. Ñ EOS,Transactions of the American Geophysical Union, 78, F363.

TJALSMA ÊR.C. & LOHMANNÊG.P. (1983). Ñ Paleocene-Eocenebathyal and abyssal benthic foraminifera from the Atlantic Ocean.Ñ Micropaleontology, Special Publication 4, 90 p.

72°W7375 747677

38

39

40

41°N

200

m

1000 m

2000 m

3000 m

904

1073

906

10721071

905

Island Beach

Cape May

Atlantic City

Bass River

Ocean City '99

903

Ancora

902

Clayton

Freehold

Allaire

Belleplain

ACGS#4

12

3

45

67

7B

DelmarvaPeninsula

New Jersey

Cenoz

oic ou

tcrop

Cretac

eous

outcr

op

pre-

Cretac

eous

NEW JERSEY TRANSECT

Ew9009 MCS

other MCS

ODP Leg 150, 150X

ODP Leg 174A, 174AX

USGS BERG boreholes

Future platform/semi-submersible drilling

ODP Site 690

NewJersey

Egypt

LATE PALEOCENE

Figure 1. Site map, showing location of all onshore and offshore New Jersey boreholes. Inset shows apaleogeographic reconstruction for the late Paleocene (after Thomas, 1998) showing the location of NewJersey, ODP Site 690, and Egypt sections.

Figure 1. Carte du New Jersey, donnant la position géographique des sites carottés en mer et à terre. Encadré:reconstitution de la paléogéographieau Paléocène récent (d'après Thomas, 1998) montrant la position du NewJersey, du Site ODP 690 et des coupes egyptiennes.

.

?

-60 0 60

R N

1120

1130

1140

1150

1160

1170

1180

1190

1200

1210

1220

1230

1240

1250

1260

g g g

gg

g

sg

g

g

g

g

g

g

g g g

g

s

s

s ss

g

ggg

g ggg g

345

350

355

360

365

370

375

380

PF CN

C24r

C25n

P6b

?

P5

P5?

P4c

P4a

P1cP3

NP11

NP9b

NP9a

NP8

NP6

NP10d

NP5

?

Clay

Clayey silt

Silty clay

Sand

Muddy sand/sandy mud

Shell bed

g Glauconite

Porcellanite/porcellanitic zone

s Sideritic nodule

Lithology symbols

MAD <5

NRMN

W,UP

1160.5

NRM

NW,UP

1205.5

MAD 5-15

NRM

N

W,UP

1188.0

NRM

N

W,UP

1133.9

MAD >15

1191.6

NRM

N

W,UP

N

W,UP

NRM

1174.3

C24n

C26r?

nodata

Depth(ft)

Inclination

Chr

ons

Biozones

Lith

olog

y

Depth(m)

nodataC24n

C24n.?r

Vin

cent

own

Fm

.M

anas

quan

Fm

.H

orne

rsto

wn

Fm

.no

t rec

ogni

zed

unna

med

uni

tH

orne

rsto

wn

Fm

.V

ince

ntow

nF

m.

Man

asqu

an F

m.

Pal

eoce

neE

ocen

eP

ALE

OC

EN

E/E

OC

EN

EB

OU

ND

AR

Y IN

TE

RV

AL

Ser

ies

Formation

uppe

rmos

t Pal

eoce

ne s

eque

nce

1* 2*

P4b?

NP

10b

g g

g g

g g g

3*

Vin

cent

own

Fm

.H

orne

rsto

wn

Fm

.M

anas

quan

Fm

.

Seq

uenc

es11

43.7

/114

0.8

1156

.5g

g11

38.6

1134

.6

1138

.25

P6a

Figure 2. Lithologic column showing paleomagnetic data with interpreted chronal boundaries and

planktonic foraminiferal (PF, Olsson in Miller, Sugarman, Browning, et al., 1998) and calcareous

nannoplankton (CN, Aubry in Miller, Sugarman, Browning, et al., 1998 and Aubry et al., manuscript in

prep.) zonations. Examples of demagnetization patterns are provided, showing the typical quality of data.

Demagnetization plots are plotted on an orthogonal axis, with the horizontal axis representing ÒNorthÓ and

the vertical axis representing ÒUpÓ for open circles and ÒWestÓ for closed circles. Cores were not oriented

in the horizontal plane, so declination data (closed circles) are inconsistent between samples. Three

formational nomenclatures are shown: 1* after assignment of Gibson et al. (1993) at Clayton; 2* after

assignment of Owens et al. (1997) at Island Beach; 3* this study, after Olsson and Wise (1987a, b) and

after assignment of Liu et al. (1997) at Island Beach.

Figure 2. Corr�lations lithologiques, magneto- et biostratigraphiques dans la coupe de Bass River.

biozonations a partir des foraminif�res planctoniques (PF): Olsson in Miller,Sugarman, Browning, et al.,

1998; a partir des nannofossiles calcaires (CN): Aubry in Miller, Sugarman, Browning, et al., 1998 and

Aubry et al., soumis. Deux exemples de d�magn�tisation progressive typique sont donn�s pour rendre

compte de la qualit� des donn�es. Les valeurs de demagn�tisation sont report�es sur un diagramme, dont

l'axe horizontal represente le Nord et l'axe vertical repr�sente "up" pour les cercles ouverts et "ouest" pour

les cercles ferm�s. Les carottes n'�tant pas orient�es par rapport � l'horizontale, les valeurs de la

d�clinaison (cercles ferm�s) sont tr�s variables. Trois interpr�tations lithostratigraphiques sont donnes: 1*

d'apr�s Gibson et al. (1993) base sur Clayton; 2* d'apr�s Owens et al. (1997) base sur Island Beach; 3*

cette �tude, d'apr�s Olsson et Wise (1987a, b) et d'apr�s Liu et al. (1997) base sur Island Beach.

Figure 3. Scanning electron micrographs showing preservation of benthic foraminifera. 1. C. alleni(1180.0 ft, 359.66 m); 2. C. alleni (1163.0 ft, 354.48 m); 3. C. succedens (1161.0 ft, 353.87 m); 4. C.succedens (1183.0 ft, 360.58 m); 5. S. beccariiformis (1183.0 ft, 360.58 m); 6. G. octocamerata (1165.0 ft,355.09 m); 7. A. acuta (1170.4 ft, 356.74 m); 8. close-up of broken A. acuta test (1165.0 ft, 355.09 m).Note the difference in preservation of specimens from above and below 1171.0 ft (356.92 m).

Figure 3. Microphotographes (MEB) montrant l'�tat de conservation des tests des foraminif�resbenthiques. 1. C. alleni (1180,0 ft, 359,66 m); 2. C. alleni (1163,0 ft, 354,48 m); 3. C. succedens (1161,0ft, 353,87 m); 4. C. succedens (1183,0 ft, 360,58 m); 5. S. beccariiformis (1183,0 ft, 360,58 m); 6. G.octocamerata (1165,0 ft, 355,09 m); 7. A. acuta (1170,4 ft, 356,74 m); 8. d�tail d'un test cass� de A. acuta(1165,0 ft, 355,09 m). Notez la diff�rence d' �tat de conservation des tests au dessus et en dessous de1171,0 ft (356,92 m).

δ13C (PDB) δ18O (PDB)

T (°C)

Depth(ft)

Depth(m)

Gyroidinoides octocamerata

Stensioina beccariformis

Anomalinoides acuta

Cibicidoides/Cibicides spp.

-4 -3 -2 -1 0 1 2

1140

1150

1160

1170

1180

1190

1200

1210

122015 20 25

-4-3-2-1

350

355

360

365

370

Figure 4. Carbon and Oxygen isotope records from Bass River. Temperature is calculated for Cibicidoidesspp. based on the equation T=16.9-4.38(dc-dw)+0.10(dc-dw)2 (OÕNeil et al., 1969; Shackleton, 1974)assuming an ice free world with dw=-1.2ä (Shackleton and Kennett, 1975) and that Cibicidoides spp.secretes calcite which is depleted by 0.64ä relative to d18O equilibrium (Graham et al., 1981).

Figure 4. Isotopes du carbone et de l'oxyg�ne dans la coupe de Bass River. La temp�rature d�termin�e surla base de Cibicidoides spp. est �tablie � partir de l'�quation T=16.9-4.38(dc-dw)+0.10(dc-dw)2 (O'Neil etal., 1969; Shackleton, 1974) en admettant qu'il n'y avait de glace, avec dw=-1.2 ä (Shackleton etKennett,1975) et que Cibicidoides spp. secr�te une calcite appauvrie de 0.64 ä par rapport � l'�quilibre dud18O (Graham et al., 1981).

.

g g g

gg

g

sg

g

g

g

g

g

g

gg g g

g

s

s

g

ggg g

Depth(ft)

CNZone

NP9b

NP9a

NP8

NP6

NP10d

C24r

C25n

C26n

350

360

370

380

NP

10b

1140

1150

1160

1170

1180

1190

1200

1210

1220

1230

1240

1250

Lith

olog

y

-4-3-2-1

15 20 25

-3 -1 1

13C (PDB) 18O (PDB) Age(k.y. from CIE)

Cibicidoides spp.

T (°C)

% Kaolinite

% Sand

25 75

400

200

0

-200

-400

-600

-800

-1000

Depth(m)

Chron

gg

Clay mineralogy

10 30

Figure 5. Environmental indicators through the upper Paleocene sequence at Bass River. Temperature wascalculated as in Figure 4. Age was calculated based on a sedimentation rate of 19.3 m/m.y. See Figure 2 forlithology symbols.

Figure 5. Indicateurs pal�oenvironnementaux dans la s�quence dÕ�ge pal�oc�ne sup�rieur de Bass River.Temp�rature calcul�e comme pour la Figure 4. Ages �tablis sur la base d'un taux de s�dimentation de 19.3m/m.y. Voir la figure 2 concernant la signification des symboles lithologiques.

-3 -1 1 -3-2-10 10

15 20 25T (°C)

13C (PDB) 18O (PDB)

Cibicidoides spp.

Age(k.y. from CIE)

10 30

Clay mineralogy

% Kaolinite

Depth(ft)

1166

1167

1168

1169

1170

1171

1172

1173

1174

1175

1176

Sand %

-60

-40

-20

0

20

40

60

OtherForamsMicaQuartz

Glauconite

Figure 6. Environmental indicators across the latest Paleocene thermal maximum at Bass River. See Figure5 for explanation.

Figure 6. Indicateurs pal�oenvironnementaux dans la sequence dÕ�ge pal�oc�ne sup�rieur de Bass River.Voir Figure 5 pour explication.

.

g g g

gg

g

sg

g

g

g

g

g

g

g

g

g g g

g

s

s

g

ggg g

gNP

11N

P10

b

NP10d

P4a

P5

P6b

P4b

?P

4c

C24

rC

25n

1140

1160

1180

1200

1220

1240

345

350

355

360

365

370

375

380D

epth

(m

)

Age (Ma)

Dep

th (

ft)

NP

9bN

P9a

NP

8N

P6

C24

n

LO A. soldadoensis

LO D. multiradiatusNP8/NP9

HO A. mckannai

C25n/C24r

5556

HO G. pseudomenardii

LO M. subbotinae

1138

.612

43.7

1240

.8

LO T. digitalisLO T. bramlettei

19.3

m/m

.y.

57

P5

NP10b

P6a

NP8 NP9a NP10a

P4b P4c

C25n C24rC25r

NP9b

P4a

HO A. subsphaericaLO M. aequa

1134

.6

Rec

over

y

Lith

olog

y

Mag

neto

stra

tigra

phy

Bio

stra

tigra

phy

PF CN

Global Timescale (BKSA95)

CIE

Surface in core

Foraminiferadatum

MagnetostratigraphicdatumNannoplanktondatum

CIE

g g

NP9a/NP9bLO R. calcitrapa

P6a

Figure 7. Age model for the upper Paleocene sequence at Bass River. Error bars on plotted points reflectsampling interval. Sedimentation rate of 19.3 m/m.y. was calculated based on the C25n/C24r Chronboundary and the NP8/NP9 Biochron boundary using ages from BKSA95. See text for discussion.Lithology symbols are as in Figure 2.

Figure 7. Interpr�tation par rapport au temps de la s�quence pal�oc�ne de Bass River. Les barres d'erreurrefl�tent les incertitudes li�es au pas d'�chantillonnage. Un taux de s�dimentation de 19.3 m/m.y. est estim�sur la base de l'inversion magn�tique (Chron C24n/C24r) et de la limite biochronale NP9/NP10. Voir letexte pour discussion. symboles lithologiques comme pour la figure 2.

170

160

180

70

60

75

65

80

δ13C (PDB) δ18O (PDB) Age

(k.y

. fro

m C

IE)

Bass

Riv

er (f

t)69

0 (m

bsf)

Geb

el A

wei

na (m

)

1150

1160

1170

1180

1190

1200

1210

-4-3-2-101-4 -3 -2 -1 0 1 2

-700

-600

-500

-400

-300

-200

-100

0

100

200

300

400

C24

rC

25n

NP

9bN

P9a

NP

8

Mag

neto

chro

nsC

N b

ioch

rons

Gebel Aweina,Egypt

Bass River,NJ

ODP Site 690,Maud Rise

Figure 8. Comparison of isotope records from Bass River (this study), ODP Site 690 (Kennett and Stott,1990, 1991), and Gebel Aweina (Schmitz et al., 1996). Bass River Cibicidoides spp. analyses are plottedas circles, ODP Site 690 Nuttalides truempyi analyses (Kennett and Stott, 1990, 1991) are plotted asdiamonds, and Gebel Aweina Lenticulina spp. analyses (Schmitz et al., 1996) are plotted as triangles. Therecords for ODP Site 690 and Gebel Aweina are plotted relative to Bass River using tie-points at the CIEand the base of Chron C24r for Site 690 (185.47 mbsf, Spiess, 1990) and the base of nannoplankton ZoneNP9 for Gebel Aweina (58.6 m, Schmitz et al., 1996). Age model is based on a sedimentation rate of 19.3m/m.y. at Bass River.

Figure 8. Comparison des courbes isotopiques dans les coupes de Bass River (cette �tude), di Site ODP690 (Kennett and Stott, 1990, 1991), et du Gebel Aweina (Schmitz et al., 1996). Cerckles: analyses surCibicidoides spp., Bass River; losanges: analyses sur Nuttalides truempyi, ODP Site 690 (Kennett andStott, 1990, 1991); triangles: analyses sur Lenticulina spp., Gebel Aweina (Schmitz et al., 1996). lescourbes du Site ODP 690 et du Gebel Aweina sont �tablies par rapport a la courbe de Bass River, enutilisant les points de rep�res que constitutent le CIE et la base du chron C24r au Site 690 (185.47 mbsf,Spiess,1990), et la base de la zone NP9 dans la coupe du Gebel Aweina (58.6 m, Schmitz et al., 1996).Ages �tablis sur la base d'un taux de s�dimentation de 19.3 m/m.y.