Obliquity-dominated glacio-eustatic sea level change in ...forth/people/Hemmo/Abels_2007.pdf · shelf sediments and their outcrop area constitutes the main body of the Rupelian historical

Obliquity-dominated glacio-eustatic sea level change in the earlyOligocene: evidence from the shallow marine siliciclastic Rupelianstratotype (Boom Formation, Belgium)

Hemmo A. Abels,1 Stefaan Van Simaeys,2 Frits J. Hilgen,1 Ellen De Man3 and Noel Vandenberghe2

1Stratigraphy/Paleontology, Department of Earth Sciences, Utrecht University, Budapestlaan 4, 3584 CD, Utrecht, The Netherlands;2Geology, K.U. Leuven, Celestijnenlaan 200E, 3001 Heverlee, Belgium; 3Department of Paleontology, Royal Belgian Institute of Natural

Sciences, Vautierstraat 29, B-1000 Brussels, Belgium

Introduction

The transition from the Eocene green-house world to the Oligocene glaciatedstate is one of the more dramatic shiftsin Cenozoic climate evolution. Duringthe early Oligocene global coolingresulted in the development of a signi-ficant Antarctic cryosphere (e.g. Learet al., 2000) and the onset of markedcyclicity in ice volume. Evidence forhigh-frequency glacial cycles comesfrom benthic d18O records in the openocean that point to a control by the41-kyr obliquity cycle, and 110- and405-kyr eccentricity cycles (Wade andPalike, 2004). However, these recordsreflect both deep water temperaturesand ice volume, besides other possiblefactors such as changes in ocean cir-culation. Independent estimates ofbottom water temperature using Mg/Ca ratios were made in order to deducethe ice volume part in oxygen isotoperecords (Lear et al., 2000). Alternat-ively, shallow marine settings can bestudied to unravel the imprint ofglacio-eustatic sea level variability,avoiding the indirect calibration ofthe ice volume variations (Pekar et al.,2002).

A suitable shallow marine locationto study high-frequency early Oligo-cene glacial cyclicity is the siliciclasticBoom Formation in Belgium. TheBoom Clay is characterized by later-ally persistent lithological alternationsbetween silt and clay. These metre-scale sequences have been ascribed tosea level fluctuations that control theamount of sorting on the sea floor byvarying the wave base (Vandenberghe,1978; Van Echelpoel and Weedon,1990; Vandenberghe et al., 1997,2001). Statistical analysis of lithologyproxies in the outcrop area pointed atan astronomical control on the regularbasic silt – clay sequence, which wasattributed to the 100-kyr eccentricitycycle (Van Echelpoel and Weedon,1990; Vandenberghe et al., 2001). Re-cent drilling of the Lower Oligocenestratotype succession substantiallylengthened the known stratigraphicrecord of the Boom Formation (VanSimaeys et al., 2004) and producedhigh-quality proxy records for litho-logy, providing an excellent oppor-tunity to re-examine the astronomicalorigin of the sedimentary cyclicity andglacio-eustatic sea level fluctuationsduring the early Oligocene.

Geological setting and stratigraphicframework

The studied lower Oligocene BoomFormation was deposited in the south-ern part of the North Sea Basin, a

slowly subsiding epi-continental basinbordered by the Fenno-ScandianShield to the northeast, Central Europeto the south, and the British Isles to thewest (Ziegler, 1990; Fig. 1). The silici-clastic deposits predominantly consistof shallow marine, coastal to outershelf sediments and their outcrop areaconstitutes the main body of theRupelian historical stratotype (Fig. 2;Luterbacher et al., 2004; Van Simaeyset al., 2004). The overall Boom For-mation has a thickness of 150 m in thesubsurface of the Campine blocks, ofwhich only the lower 43 m are exposedin the Rupel area (Fig. 3).Accurate age calibration of the

Rupelian deposits to the GeologicTime Scale (GTS04; Gradstein et al.,2004) remains a moot point because ofthe near absence of age-indicativeplanktonic calcareous microfossils,and uncertainties around paleo-magnetic results (Lagrou et al., 2004).The last occurrence (LO) of nanno-plankton species Reticulofenestraumbilica positions the NP22/NP23boundary just below the base of theBoom Formation (Steurbaut, 1992);this event has an age of � 32.4 Ma inGTS04 (Luterbacher et al., 2004). To-wards the middle of the succession twodistinct dinocyst events nearly coin-cide: the first occurrence (FO) of Sat-urnodinium pansum and the LO ofcommon Enneadocysta pectiniformis(Van Simaeys et al., 2005a), both datedat 29.18 Ma according to Luterbacher

ABSTRACT

Our results prove that glacio-eustatic sea level oscillations inthe early Oligocene were dominantly obliquity controlled withadditional influence of the �100- and 405-kyr eccentricitycycles. This was derived from spectral analysis of resistivityrecords from an extended downhole section of the Boom Claysuccession in Belgium, that reveals a prevailing obliquitycontrol on the laterally persistent metre-scale alternations ofshallow marine silt- and claystones in the Rupelian historicalstratotype succession. These direct measurements of sea level

variations in a shallow marine setting corroborate that varia-tions with similar frequencies in benthonic oxygen isotoperecords from the open ocean indeed reflect, at least partly, icevolume change. A very tentative astronomical tuning has beenestablished for the Boom Clay succession which awaits futureconfirmation with the addition of more accurate age calibrationpoints.

Terra Nova, 19, 65–73, 2007

Correspondence: Dr Hemmo Anne Abels,

Department of Earth Sciences, Utrecht

University, Budapestlaan 4, Utrecht

3584CD, The Netherlands. Tel.:

00 31 30 253 5125; fax: 00 31 30 253 3486;

e-mail: [email protected]

� 2006 Blackwell Publishing Ltd 65

doi: 10.1111/j.1365-3121.2006.00716.x

et al. (2004). The FO of Distatodiniumbiffii calibrated at 28.43 (Luterbacheret al., 2004) is found higher up in thesuccession (Van Simaeys et al., 2005a).Dinocyst calibration between theNorth Sea Basin and the central Italiansections shows that theRupelian/Chat-tian unconformity in the stratotypearea marks a hiatus covering the inter-val from around 27.5 to 27.0 Maaccording to Van Simaeys et al.(2005b). Sedimentation of Chattiansediments is thought to start at around26.8 Ma (Van Simaeys et al., 2005b).New strontium isotope ages indicatedin Fig. 2 are briefly explained in theMethods section.

Lithology

The Boom Formation is characterizedby a rhythmic alternation of silt and

clay layers. A sinusoidal variation ingrain size is present with the largestproportion of coarse grain sizes in themiddle of the silt bed. The amount oforganic matter, mainly of terrestrialorigin (Vandenberghe, 1978; Vanden-berghe et al., 1997; Laenen, 1998),increases abruptly at the base of thePutte Member (Fig. 2). Individualorganic-rich layers occur after a siltbed in the basal parts of a clay bed.Clay mineral analysis revealed that theillite, chlorite and smectite contentvaries in harmony with grain size,while the kaolinite content shows anopposite distribution. This aims at amore basinal origin of the clay miner-als during the deposition of the claybeds, and a more coastal origin duringthe deposition of the silts (Laenen,1998). The stratigraphic position ofcarbonate-rich layers, that evolved

into septaria horizons, does not showany relationship with grain size, claymineral distribution and organic mat-ter, suggesting a sedimentary ratherthan a diagenetic origin (Vanden-berghe et al., 1997; Fig. 2).

Sedimentological interpretation

The lateral persistence of the individ-ual silt-clay sequences (Vandenbergheet al., 2001) requires a forcing mech-anism that exerts a simultaneous influ-ence over the entire basin at the sametime. Sea level variation influencingthe amount of sorting by varyingthe wave base is the most plausibleprocess that can account for this(Vandenberghe et al., 1997, 1998).The more basinal origin of the clayminerals deposited in the clay bedsduring supposed sea level high-standscorroborates this mechanism. Accord-ingly, the terrestrial organic matter isthen deposited during the transgres-sive part of the sequence (in the basalpart of the clay beds) as a result ofreworking of the vegetation cover ofthe flooded land, while the clay min-eral assemblage indeed points to amore coastal origin during depositionof the silt beds during times of sea levellow-stands. The supposed mechanismof sea level variations triggering thelithological variations fits all observa-tions. A more regional mechanism aschanges in run-off or in sediment loadwould produce local differences in theBoom succession that are not present.The origin of the marly beds howeverremains poorly understood.The Boom Clay succession can thus

be seen as an archive of early Oligo-cene glacio-eustatic sea level fluctua-tions, also because high-frequency sealevel variability is not expected to beof (local) tectonic origin (Miller et al.,2005).

Methods

Statistical analysis was performed onthe AO90 and MSFL resistivity re-cords of the Dessel-1 borehole (for

Fig. 1 (a) Paleogeographical reconstruction of the North Sea Basin during earlyOligocene times. (b) Locality map of the Dessel borehole site, the Swenden quarry inRumst and outcrop area of the Boom Formation.

Fig. 2 The FMI, AO90 and MSFL resistivity records of the Dessel-1 borehole covering the Rupelian stage. Important bio-stratigraphical events are shown with their inferred age. Successive highs (lows) in the AO90 record are labelled with an odd (even)number. In black the name of septaria horizons that were recognized in the borehole data and in grey other septaria horizonsknown from the Boom Clay. The latter were pinpointed in the Dessel stratigraphy by bed-to-bed correlation with other boreholeand outcrop data of the same interval (see also Vandenberghe et al., 2001). Ages are from Luterbacher et al. (2004); age 1, 3, 4 and5). Strontium isotope dates are explained in Methods section (ages 2).

Early Oligocene sea level variability • H. A. Abels et al. Terra Nova, Vol 19, No. 1, 65–73

.............................................................................................................................................................

66 � 2006 Blackwell Publishing Ltd

Terra Nova, Vol 19, No. 1, 65–73 H. A. Abels et al. • Early Oligocene sea level variability

.............................................................................................................................................................


location see Fig. 1). This borehole wasselected for its higher resolutioncompared to other investigated bore-holes and the presence of a FullboreFormation Micro-Imager (FMI)log (Fig. 2). The Micro-SphericallyFocused Log (MSFL) has a depth ofinvestigation/analysis from between2.5 to 15 cm, while the high-resolutiondeep laterolog measurement (AO90)has a depth of investigation of 300–400 cm.The resistivity records are an accu-

rate proxy for the silt-clay alternationsin the Boom Formation, exhibitinghigher values in the silt beds (Van-denberghe et al., 2001). Carbonatecontents in the clay did not influencethe periodicity analysis. As a minorbackground component, calcite is pre-sent only in the short interval betweenthe septaria levels S10 and S30. In theother parts of the section only septariahorizons are present of which thestratigraphic positions are well known.Of these only S180 (Fig. 2) exhibits asignal in the used resistivity records.The presence of some carbonate willpossibly only slightly increase thenoise in the spectral analysis.

The two resistivity logs show adistinct offset for the coarse-grainedsediments in the lower Belsele-Waas

Member and the upper sandier unit,which is the reflection of higher per-meability in the coarser sediments.For the statistical analysis the intervalbetween 212 and 282 m in the Dessel-1 borehole has therefore been selectedbecause above and below distortionby increasing amounts of sand oc-curs. The lack of conspicuous inter-vals that could point to erosionallevels and the extreme lateral conti-nuity indicate that the successionwithin the selected interval is continu-ous, at least until the level of the basicsilt-clay sequences.The statistical analysis was carried

out with the Analyseries programversion 1.1.1 of Paillard et al. (1996)with compromise settings on the num-ber of lags and percentage of theseries. Subsequently, the Blackman-Tuckey power spectral analysis wasapplied using a Barlett window.Age assessment by means of

strontium isotope stratigraphy wasperformed by measuring the 87Sr/86Srratio of benthic foraminifera fromnine horizons and evaluating it tothe secular variation curve for Oli-gocene marine waters using the pro-cedure of McArthur et al. (2001; agesderived from Look-Up Table,Version 4).

Fig. 3 Photograph of a Boom Clay outcrop section in the Swenden quarry in Rumst,Belgium, which is located 45 km to the WSW of Dessel (Fig. 1). Light (dark) horizonsrepresent silt (clay) beds. White lines indicate successive upper limits of a clay-siltcouplet. The white numbers represent successive clay-silt alternations as numbered inthe Dessel borehole (Fig. 2). The black cycle numbers indicate the numbering ofVandenberghe (1978). Septaria horizons recognized in the quarry are indicated, aswell as the so-called double band (DB) and red layer (R). The Terhagen to Puttemember transition is based on the rapid increase in preserved organic matter and canbe recognized over large distances (Vandenberghe et al., 2001).

Fig. 4 Blackman-Tuckey (BT) power spectra of the (a) MSFL and (b) AO90 resistivityrecords between 212 and 282 m in the Dessel borehole, and (c) of the tuned AO90record in the time domain. Grey shades indicate distinct spectral peaks. (d) Powerspectrum of a sinusoid tuned to the same interval of the obliquity target curve by usingthe identical tie-points as for the AO90 tuning. 90% confidence intervals (c.i.) andbandwidth (b.w.) are indicated with vertical and horizontal lines respectively.


.............................................................................................................................................................


Results

Statistical analysis

The spectral analysis performed onthe selected interval of the AO90 andMSFL depth records reveals spectralpower in both resistivity records atperiods around 1.13 and 1.45 m(Fig. 4a,b). Bandpass filtering of theseperiods shows that they are related tothe basic silt-clay sequences (Fig. 5).In addition, a 3.65-m peak is presentin both resistivity records and a 14-mpeak in the power spectrum of AO90.Filtering these peaks shows that the3.65 m peak reflects bundles of two or

three basic sequences and the 14 mpeak in AO90 reflects intervals ofmore and less silty beds (Fig. 5).

Astronomical forcing

Spectral analysis of depth series re-sults in power spectra that may bedistorted by diagenetic processes andbioturbation and by changes in sedi-mentation rate related to long-termtrends and amplification of the astro-nomical forcing (Fischer et al., 1991;Herbert, 1994; Van der Laan et al.,2005). Power spectra of depth seriesthat contain distinct peaks revealimportant information about cyclic

forcing mechanisms, even if age con-trol is limited, by comparing spectralpeaks with astronomical frequencyratios. In case of the Boom Claysuccession, bandpass filtering of thedistinct spectral peaks reveals thatthese three major periodicities arepresent throughout the studied inter-val (Fig. 5). Therefore a relativelyconstant sedimentation rate can beexpected, while the separation of forexample the 1.45 m peak into a 1.13,1.37 and 1.54-m peak may be causedby small variations in sedimentationrate. The consistent regular cyclicity ina long interval is an argument forastronomical forcing of the sedimen-

Fig. 5 AO90 and MSFL resistivity records and their high- and low-frequency bandpass filters. Filtered intervals relate to distinctspectral peaks in the spectral analysis of the depth proxy series. To the left small-scale cycle numbers are indicated. Higheramplitudes in filters indicate periods of dominance of that period.


.............................................................................................................................................................


tary alternations (Van Echelpoel andWeedon, 1990). Moreover, there is aclose resemblance of the ratio of1 : 2.8 : 9.7 between the distinct spec-tral peak periods and the ratio of1 : 2.65 : 9.9 for the astronomicalfrequencies of obliquity (41-kyr) andshort- (95- and 123-kyr) and long-term (405-kyr) eccentricity, whichprovides the final convincing argu-ment for astronomical forcing of theBoom Clay lithological cyclicity.

Obliquity control

Obliquity control (41-kyr) of theapproximately 50 silt-clay sequencesin the studied part of the Boom Claysuccession (Fig. 5) results in a timespan of 2050 kyr for the studied inter-val, which is in agreement with theknown age constraints (Fig. 2). If pre-cession would be designated to thebasic sequence, only 1050 kyr wouldbe present in the interval. This durationfits within the limits of the age con-straints although it would be short andhiatuses that bound the successionremarkably long. However, the ratioof the spectral peaks does not at allresemble astronomical forcing (Fig. 4).Designation of eccentricity would ac-count for a duration of 5 Myr, which isimpossible for the studied interval con-sidering the available age constraints(Fig. 2). The lithological cyclicity isthus dominantly controlled by obli-quity forcing on sea-level fluctuationswith additional imprint of the �100-and 405-kyr eccentricity cycles. Ourinterpretation of the astronomical for-cing in the Boom Clay differs fromprevious studies that attributed thebasic sequence to short-term eccentri-city instead of obliquity (Van Echel-poel and Weedon, 1990).

Discussion

Early Oligocene glacial cyclicity

Our results confirm that the high-frequency variations in the benthic

oxygen isotope record in the equatorialPacific (Wade and Palike, 2004) indeedpartially reflect ice volume changes ofthe incipient Antarctic ice sheet andthat these variations are controlledby obliquity and �100- and 405-kyreccentricity orbital cycles. Unfortu-nately, no quantitative estimates ofabsolute sea level variations can bederived from the Boom Clay deposits.The lower part of the Boom For-

mation is characterized by a moredominant obliquity signal, while to-wards the upper part the imprint ofeccentricity becomes more prominent(Fig. 5). In Fig. 6 a detail of the upperpart is shown to visually depict theinterplay between obliquity and �100-and 405-kyr eccentricity in the EarlyOligocene eustatic sea-level variation.Gradual increases in grain size (bluearrows in Fig. 6) end rather abruptlywith sudden decreases (red arrows).This strongly suggests a gradualincrease in ice volume and hence a

slow sea level lowering followed byrapid deglaciation and sea level rise,which show remarkable similaritieswith the late Pleistocene glacialhistory.At around 240 m (Fig. 6) twice as

many silt-clay alternations are presentas expected from the normal obliquityforced cyclicity, as can be seen in the1.30–1.70 m filter. We tentativelyinterpret these alternations as beingrelated to precession forcing, althoughspectral analysis in the depth domaindoes not reveal a distinct peak thatcan be related to the precession band(Fig. 4a,b). However, Van Echelpoeland Weedon (1990) did find a cyclewith half the length of the basic silt-clay sequence in the Boom Clay thatprobably reflects this kind of preces-sion related alternations. The presenceof precession related cycles is to beexpected in this particular intervalbecause it coincides with a 400-kyreccentricity maximum.

Fig. 6 Detail of the Dessel stratigraphy showing the AO90 resistivity record, its 21-point moving average, and the 1.30–1.70, 3.00–4.65 and 13.3–18.0 m filters of therecord. Blue (dashed red) arrows indicate gradual increasing (decreasing) grain size. Apattern of gradual increasing ice volume and fast deglaciations can be recognized, dueto the interplay of obliquity with �100- and 405-kyr eccentricity. To the left small-scale cycle numbers are shown.

Fig. 7 Tuning of the AO90 and MSFL (dashed record and filters) resistivity record of the Dessel borehole to the Laskar et al.(2004) obliquity target curve (grey lines) using available age datums (superscript refers to identical legends in Fig. 2). Dashedarrows show intervals where the astronomical target curves suggest periods of forcing dominance of obliquity or obliquity andeccentricity. Ages of the top of the lower hiatus and base of the upper hiatus were calculated by counting small-scale cycles thathowever may not continue to be obliquity cycles in those intervals. For the top interval, the cycle pattern of nearby well-logs is alsoused. Asterisks indicate used age tie-points for the time-series analysis. Indicated Oi-events are from Wade and Palike (2004).


.............................................................................................................................................................



.............................................................................................................................................................


Astronomical tuning

Ideally the Boom Clay successionwould be tuned to astronomical targetcurves, providing astronomical agecontrol that ties the Rupelian histor-ical stratotype to the geological timescale. For a tuning, characteristicpatterns in the target curves and proxyrecords and at least one highly reliableage calibration point have to be pre-sent. However, the age calibrationpoints for the Rupelian successionlack precision. The present age dataserve as a starting point for a verytentative tuning that correlates theprolonged sea level low-stand frombed numbers 43 to 61 to the 405-kyreccentricity and 1.2-Myr obliquityminima at 30.0 Ma (Fig. 7), followingthe results of other authors regardingthe phase-relation of sea level low-stands to 41-kyr and 1.2-Myr obli-quity minima and �100- and 405-kyreccentricity minima (Wade and Palike,2004; Abels et al., 2005) For thetuning subsequently every next siltbed (sea level low-stand) was correla-ted with the next obliquity minimum.The whole tuning may have to be

shifted one, or maybe even two, 405-kyr eccentricity cycles older or youngerif more precise age calibration pointsbecome available, even though avail-able age tie-points fit well with thisprovisional tuning. Note that the tun-ing reveals a consistent relation ofcoarser intervals (sea level low-stands)with �100- and 405-kyr eccentricityminima, although the used tuningprocedure only correlated individualsmall-scale cycles with obliquity. TheAO90 depth series is transformed to atime series by using a limited number ofage tie-points (* in Fig. 7). The powerspectrum of this time series showspeaks that correspond to 41.7, 100and 417 kyr periods (Fig. 4c), whilesome power is also present at preces-sion frequencies. In addition, we tuneda simple sinusoid to the same interval ofthe obliquity curve using the sametie-points to check the amount ofspectral power we introduced to theAO90 depth series by tuning it to anastronomical target curve and so con-structing the time series. When thepower spectra of the two time seriesare compared (Fig. 4c,d) the�100- and405-kyr peaks in the Boom Clay spec-trum stand out very clearly, againindicating the presence of these period-

icities in the early Oligocene sea levelarchive of the Boom succession.

Conclusions

Spectral analysis of borehole resistivityrecords indicates that shallow marinelithological sequences in the early Oli-gocene Boom Clay succession reflectglacio-eustatic sea-level variations thatare primarily driven by the 41-kyr obli-quity cycle. In addition, a secondaryimprint of�100- and 405-kyr eccentri-city has been identified. These cyclesreflect changes in Antarctic ice volumeand demonstrate that high-frequencyvariations recognized in open oceanbenthic d18O records indeed partlyreflect ice volume. Furthermore, itcould be shown that glaciations tendedto grow slowly, and ended ratherabruptly. A provisional astronomicaltuning is provided using present agecalibration points and correlation ofsubsequent sea level low-stands withobliquity minima. This tuning clearlyshows the presence of �100- and405-kyr eccentricity cyclicity in theglacio-eustatic sea level archive ofthe Boom Clay, and the presence ofsome precession-related variations.

Acknowledgements

Henk Brinkhuis is thanked for criticalreading of an earlier version of the manu-script and for helpful discussions. NIRASis greatly acknowledged for providing theresistivity data of different boreholes. H.A.acknowledges the financial support of theDutch Science Foundation (NWO) andE.D.M. the support of the Belgian SciencePolicy (Grant WI/36/C03). Strontium iso-tope analyses were performed at the RoyalHolloway University (UK), under super-vision of M. Thirlwall and S. Duggen. Twoanonymous reviewers are thanked for theirconstructive comments.

References

Abels, H.A., Hilgen, F.J., Krijgsman, W.,Kruk, R.W., Raffi, I., Turco, E. andZachariasse, W.J., 2005. Long-periodorbital control on middle Mioceneglobal cooling: integrated stratigraphyand astronomical tuning of the BlueClay Formation on Malta. Paleoceano-graphy, 20, PA4012.

Fischer, A.G., Herbert, T.D., Napoleone,G., Premoli Silva, I. and Ripepe, M.(1991). Albian pelagic rhythms (Piobib-ico core). J. Sediment. Petrol., 61, 1164–1172.

Gradstein, F.M., Ogg, J.G., Smith, A.G.,Bleeker, W. and Lourens, L.J., 2004. Anew Geologic Time Scale, with specialreference to Precambrian and Neogene.Episodes, 27, 83–100.

Herbert, T.D., 1994. Reading orbital cyclesdistorted by sedimentation: models andexamples. Spec. Publ. Int. Assoc. Sedi-ment., 19, 483–507.

Laenen, B., 1998. The geochemical signa-ture of relative sea-level cycles recog-nised in the Boom Clay. AardkundigeMededelingen, 9, 61–82.

Lagrou, D., Vandenberghe, N., VanSimaeys, S. and Hus, J., 2004. Magnet-ostratigraphy and rock magnetism of theBoom Clay (Rupelian stratotype) inBelgium. Neth. J. Geosci., 83, 209–226.

Laskar, J., Robutel, P., Joutel, F., Gasti-neau, M., Correia, A.C.M. and Levrard,B., 2004. A long-term numerical solutionfor the insolation quantities of the Earth.Astron. Astrophys., 428, 261–285,doi: 10.1051/0004–6361:20041335.

Lear, C.H., Elderfield, H. and Wilson,P.A., 2000. Cenozoic deep-sea tempera-tures and global ice volumes fromMg/Ca in benthic foraminiferal calcite.Science, 287, 269–272.

Luterbacher, H.P., Ali, J.R., Brinkhuis, H.,Gradstein, F.M., Hooker, J.J., Monechi,S., Ogg, J.G., Powell, J., Rohl, U.,Sanfilippo, A. and Schmitz, B., 2004. ThePaleogene period. In: A GeologicalTimescale 2004 (F.M. Gradstein, J. Oggand A. Smith, eds), pp. 384–408. Cam-bridge University Press, Cambridge, UK.

McArthur, J.M., Howarth, R.J. andBailey, T.R., 2001. Strontium isotopestratigraphy: LOWESS version 3: bestfit to the marine Sr-isotope curve for0–509 Ma and accompanying Look-upTable for deriving numerical age.J. Geol., 109, 155–170.

Miller, K.G., Kominz, M.A., Browning,M.A., Wright, J.D., Mountain, G.S.,Katz, M.E., Sugarman, P.J., Cramer,B.S., Christie-Blick, N. and Pekar, S.F.,2005. The Phanerozoic record of globalsea-level change.Science, 310, 1293–1298.

Paillard, D., Labeyrie, L. and Yiou, P.,1996. Macintosh program performs time-series analysis.EOS Trans. AGU, 77, 379.

Pekar, S.F., Christie-Bick, N., Kominz,M.A. andMiller, K.G., 2002. Calibrationbetween eustatic estimates from back-stripping and oxygen isotopic records forthe Oligocene. Geology, 30, 903–906.

Steurbaut, E., 1992. Integrated strati-graphic analysis of Lower Rupeliandeposits (Oligocene) in the BelgianBasin. Ann. Soc. Geol. Belg., 115, 287–306.

Van der Laan, E., Gaboardi, S., Hilgen,F.J. and Lourens, L.J., 2005. Regionalclimate and glacial control on high-resolution oxygen isotope records fromAin el Beida (latest Miocene, northwest


.............................................................................................................................................................


Morocco): a cyclostratigraphic analysisin the depth and time domain. Paleo-ceanography, 20, 1, doi: 10.1029/2003PA000995.

Van Echelpoel, E. and Weedon, G.P.,1990. Milankovitch cyclicity and theBoom Clay Formation: an Oligocenesiliciclastic shelf sequence in Belgium.Geol. Mag., 127, 599–604.

Van Simaeys, S., De Man, E., Van-denberghe, N., Brinkhuis, H. andSteurbaut, E., 2004. Stratigraphic andpalaeoenvironmental analysis of theRupelain–Chattian transition in the typeregion: evidence from dinoflagellatecysts, foraminifera, and calcareousnannofossils. Palaeogeogr. Palaeoclima-tol. Palaeoecol., 208, 31–58.

Van Simaeys, S., Munsterman, D. andBrinkhuis, H., 2005a. Oligocene dino-flagellate cyst biostratigraphy of thesouthern North Sea Basin. Rev. Palaeo-bot. Palynol., 134, 105–128.

Van Simaeys, S., Brinkhuis, H., Pross, J.,Williams, G.L. and Zachos, J.C., 2005b.Arctic dinoflagellate migrations markthe strongest Oligocene glaciations.Geology, 33, 709–712.

Vandenberghe, N., 1978. Sedimentology ofthe Boom Clay Formation, Belgium.Bull. Belg. Ver. Geol., 102, 5–77.

Vandenberghe, N., Laenen, B., VanEchelpoel, E. and Lagrou, D., 1997.Cyclostratigraphy and climaticeustacy. Example of the Rupelianstratotype. C R Acad. Sci. II A, 325,305–315.

Vandenberghe, N., Laga, P., Steurbaut, E.,Hardenbol, J. and Vail, P., 1998.Tertiary sequence stratigraphy at thesouthern border of the North Sea basinin Belgium. In: Mesozoic and CenozoicSequence Stratigraphy of EuropeanBasins (P. De Graciansky, J. Hardenbol,T. Jacquin and P. Vail, eds), SEPMSpec. Publ., 60, 119–154.

Vandenberghe, N., Hager H., Van denBosch M., Verstraelen A., Leroi S.,Steurbaut E., Prufert J., and Laga P.,2001. Stratigraphic correlation by calib-rated well logs in the Rupel Group bet-ween North Belgium, the Lower Rhinearea in Germany and Southern Limburg,and the Achterhoek in the Netherlands.In: Contributions to the Paleogene andNeogene Stratigraphy of the North SeaBasin (N. Vandenberghe, ed.), Aardkun-dige Mededelingen 11, pp. 69–84. LeuvenUniversity Press, Leuven, Belgium.

Wade, B.S. and Palike, H., 2004. Oligoceneclimate dynamics. Paleoceanography, 19,PA4019, doi: 10.1029/2004PA001042.

Ziegler, P.A., 1990. Geological Atlas ofWestern and Central Europe. ShellInternationale Petroleum MaatschappijB.V., Den Haag, 239 pp.

Received 13 April 2006; revised versionaccepted 2 October 2006


.............................................................................................................................................................


Documents

Obliquity-dominated glacio-eustatic sea level change in ...forth/people/Hemmo/Abels_2007.pdf · shelf sediments and their outcrop area constitutes the main body of the Rupelian historical