Lower Cretaceous stage durations combining radiometric data and orbital chronology: Towards a more stable relative time scale?

Earth and Planetary Science Letters 246 (2006) 407 – 417 www.elsevier.com/locate/epsl

Lower Cretaceous stage durations combining radiometric data and orbital chronology: Towards a more stable relative time scale? N. Fiet a,⁎, X. Quidelleur a , O. Parize b , L.G. Bulot c , P.Y. Gillot a a

c

Laboratoire Interactions et Dynamique des Environnements de Surface UMR-CNRS 8148, I.D.E.S., Université Paris Sud Bât. 504, 91405 Orsay Cedex, France b CGES- Sédimentologie, E.N.S.Mines de Paris 35, rue St Honoré, 77300 Fontainebleau, France Centre de Sédimentologie-Paléontologie UMR-CNRS 6019, Université de Provence, 13331 Marseille cedex 04, France Received 20 January 2006; received in revised form 24 March 2006; accepted 8 April 2006 Available online 2 June 2006 Editor: C. P. Jaupart

Abstract We propose an alternative calibration of Lower Cretaceous stage durations constrained by direct absolute dating of each stage combined with orbital chronology. Ten glauconitic horizons sampled in the Vocontian basin (SE, France) from the base of the Lower Hauterivian to Upper Albian, yielded K–Ar ages from 123.3 ± 1.7 Ma to 96.9 ± 1.4 Ma, respectively. The relative duration of each stage has been derived by cyclostratigraphy previously obtained in the south-east France and central and south Italy basins. Using the GL-O standard from the Albian–Cenomanian boundary at 95.3 Ma as the anchor point, a cyclostratigraphic age for each stage boundaries has been extrapolated and thus compared with the K–Ar ages. This shows a very well-defined linear correlation which demonstrates the robustness of the proposed durations of the Lower Cretaceous stages. The estimated durations are 5.3 ± 0.4 my, 5.1 ± 0.3 my, 6.8 ± 0.4 my and 11.6 ± 0.2 my for the Hauterivian, Barremian, Aptian and Albian stages, respectively. It also shows that glauconite minerals are powerful radiochronometric tools, when precisely stratigraphically defined and carefully selected. Moreover, the large discrepancy of the estimated Aptian duration of more than 6 my between the most recent published time scale and this study highlights the problem of the radiometric calibration of the M-0 magnetic chron. Finally, the stage durations and boundary ages proposed here bring strong constraints towards the calibration of the Lower Cretaceous time scale. Such accurate temporal calibration is required before any relationship between major biological crises and magmatic emplacement, for instance, could be further investigated. © 2006 Elsevier B.V. All rights reserved. Keywords: Lower Cretaceous; K–Ar; Geochronology; Cyclostratigraphy; Glaucony; Geologic time scale

1. Introduction The Lower Cretaceous, considered as the period of the warmest climate of the Phanerozoic, undergoes several major crises, such as strong perturbations of the carbon

⁎ Corresponding author. Tel.: +33 1 6915 6757; fax: +33 1 6915 4882. E-mail address: [email protected] (N. Fiet). 0012-821X/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.epsl.2006.04.014

cycle, important accumulation of organic matter, the socalled oceanic anoxic events (OAE), plankton crises in the oceans, and emplacement of large igneous provinces. The accurate temporal calibration of this period is of major relevance to investigate a possible link between these events. Unfortunately, large discrepancies in both absolute ages and relative duration of stages are observed between proposed time scales. This is likely due to: 1) the very low number of radiometric ages available,

408

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417

2) the important variation in the nature and the precision of the stratigraphical support related to each radiometric data, 3) the selection of data and the choice of the anchor points used for the time scale calibration and, 4) the use of fundamentally different approaches, including graphical extrapolation between radiometric points in the Hawaiian seafloor spreading magnetic profile [1], maximum likelihood estimation applied to a radiometric database of high temperature minerals (K-feldspar or biotite) [2,3], and selected radiometric data for both low temperature (glauconite mineral) and high temperature minerals (K-feldspars and biotites) [4,5]. Each of these approaches has limitations, which result on large absolute and/or relative uncertainties for calibration of the Lower Cretaceous Time Scale (LCTS). The Barremian–Aptian boundary (base of the M–0 chron), for example, ranges between 113 Ma [5] and 125 Ma [2], and the duration of the Aptian between 5 and 13 Ma. Clearly, the determination of a precise and more robust chronology at the scale of the sub-stage is necessary to demonstrate any synchronicity, relationships, or interactions between the different geological events of the Lower Cretaceous. Our integrated approach is based on the combination of radiochronological dating of condensed sedimentary horizons, which provides an absolute time frame of each stage, with cyclostratigraphy (Milankovitch cyclicities) performed on continuous sedimentary basin successions and providing a relative measure of time. Here, we determined the duration of the stages and related substages of the Hauterivian–Albian interval from the Tethyan pelagic sedimentary successions using K–Ar analyses on glauconite minerals from the Vocontian basin (SE of France), combined with cyclostratigraphic data supported by a well-constrained biostratigraphic framework (ammonite zones or sub-zones) from both the Umbria–Marche (central Italy) and the Vocontian basins. 2. Geological setting and sampling of glauconitic horizons The Vocontian basin has been investigated in detail for stratigraphy and sedimentology [6,7], and represents a stratigraphic key area relative to the location of international reference sections for the Lower Cretaceous (Fig. 1). Glauconite minerals have been collected throughout this interval in condensed sedimentary deposits, which are stratigraphically well-constrained by ammonite fauna (Table 1). Except two of them (for which precise stratigraphic positions were available, see Table 1), most of them are located at the base of biozones. The ten selected glauconitic-rich horizons follow rigorous criteria for K–Ar dating (e.g. [8]). The

Fig. 1. Location map of sampled sections (Lambert coordinates). 1: Escragnolles (955.5–3169.95); 2: La Palud (924.2–3172.75); 3: Jas de Coeur (889.48–3206.28); 4: Gargas (843.35–3181.85); 5: Les Ferres (982.60–3182.50); 6: Rougon (925.8–3174.3); 7: La Colle (934.1–3179.1). Insert: Location of area in France.

sections are located far from thrust belts or main faults zones. Burial depth did not exceed 1600 m [6,9]. The sedimentary deposits are not affected by pedogenetic alteration and are constituted of moderately welded marls and, in one case, of stylolite-free limestones, which argues for the lack of diagenetic effects on the glauconite minerals. These clay minerals are authigenic to rarely perigenic (no apparent reworking). The grains have been selected by hand picking following washing, sieving, magnetic separation, and ultrasonic treatment with 10% acetic acid. X-ray diffractometry shows well-crystallized closed minerals in the analyzed size fraction (125– 300 μm). The selected glauconitic minerals are evolved to highly evolved (K values between 5.4% and 6.9%). Moreover this selection rules out the possible contribution of the inherited substratum of glauconitization, which could yield significant excess radiogenic argon (e.g. [10]). 3. K–Ar radiometric dating of glauconite minerals Table 1 lists the K–Ar ages obtained using the Cassignol–Gillot technique [11]. They have been duplicated in most cases to better than 0.2%, which attests to the homogeneity of the glauconite minerals separation. Note that all uncertainties mentioned in this paper are given at the 1σ level.

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417

409

Table 1 Stratigraphic location of sampled glauconitic horizons Sample

Ammonite zone and/or sub-zone

Sections (Fig. 1)

K%

40

Ar⁎ (%)

40

Ar⁎ (1014 atom/g)

K–Ar age (Ma)

Un. (Ma)

Escra 03 Escra 02

Dispar Dispar (Blanchetti)

1a 1b

Escra 01

Inflatum (Cristatum)

1c

Palud 02

Base Loricatus

2a

Palud 01

Dentatus (Lyelli)

2b

JdC

Jacobi

3

Cav-B

Upper Tobleri (eq. Melchioris)

4

LF 04

Subnodosocostatum

5

Rougon

Emerici (eq. Hugii)

6

COL

Radiatus (Buxtorfi)

7

5.838 6.126 6.111 5.419 5.575 5.419 5.419 5.864 5.916 5.910 6.614 6.293 5.754 5.477 5.427 5.405 6.822 6.822 6.885 6.885 6.681 6.564

82.8 92.8 92.5 92.3 94.1 92.1 92.4 94.1 90.9 94.4 91.7 93.3 92.9 93.3 88.4 88.3 94.4 94.0 97.1 97.3 95.5 95.7

6.0633 6.4847 6.4791 5.9484 6.0940 5.9383 5.9322 6.5648 6.6274 6.5933 7.4672 7.0838 6.5780 6.2736 6.3515 6.3392 8.2053 8.1602 8.7648 8.7612 8.8967 8.7615

96.8 98.6 98.8 102.2 101.7 102.0 101.9 104.1 104.2 103.8 105.0 104.7 106.3 106.5 108.7 108.9 111.6 111.0 117.9 117.9 123.2 123.5

1.4 1.4 1.4 1.4 1.4 1.4 1.4 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.6 1.6 1.7 1.7 1.7 1.8

Mean K–Ar age (Ma)

Glauconitic age (Ma)

96.8 ± 1.4 98.7 ± 1.4

96.9 ± 1.4 98.8 ± 1.4

101.9 ± 1.4

102.0 ± 1.4

104.0 ± 1.5

104.1 ± 1.5

104.8 ± 1.5

105.1 ± 1.5

106.4 ± 1.5

106.5 ± 1.5

108.8 ± 1.5

108.9 ± 1.5

111.3 ± 1.6

111.7 ± 1.6

117.9 ± 1.7

118.3 ± 1.7

123.3 ± 1.7

123.6 ± 1.7

Escra 02 is located at the base of the very glauconitic marls of the Escragnolles section considered as Vraconian (base of the Dispar biozone) [9]. CavB is sampled in a condensed very fossiliferous glauconitic horizon of the famous Gargas section [9], visible in temporary excavations. The ammonite assemblages indicate the upper part of the Tobleri zone near the top of the Gargasian. In the absence of a precise chronometer for this portion of time, the K–Ar age will be considered close to the extrapolated cyclostratigraphic age at the Gargasian–Clansayesian boundary. Palud samples have been stratigraphically revisited from the ammonite data published by [9]. Palud 01 belongs to the Dentatus/Lyelli sub-zone and Palud 02 most probably to the base of Loricatus zone (and not Lautus zone). Note that all glauconitic horizons, excepted LF 04, which is polyzonal (Furcata at the base to Subnodosocostatum at the top), are buried by sediments belonging to the same ammonite biozone. Potassium content (K%) was determined by atomic flame emission. Radiogenic argon (40Ar⁎) determination was performed using the Cassignol–Gillot technique [11]. Age uncertainties (Un.), which include the GL-O standard uncertainty (1%), are quoted at the 1σ level. Decay constants of [48] were used. Note that in order to limit the effect of adsorbed water on small aliquot weight (typically 50 mg here), K and Ar determinations have been performed closely in time for each age determination. Glauconitic ages account for time of closure of glauconite minerals (see text).

Glauconite minerals are formed from a Fe-rich precursor neo-formed in various porous substrates such as carbonate skeleton, faecal pellets or terrigeneous particles. This precursor evolves into a glauconitic mineral by successive re-crystallizations, K2O incorporation and concomitant dissolution of the initial substratum. The entire evolution process occurs at the sediment–sea water interface. A high rate of detrital influx and/or a deposition of a non-permeable horizon of sediments, and/or an abrupt fall of sea level leading to emersion, will inhibit the glauconitization process. The closure age, or apparent age, thus corresponds to the end of the glauconitization process after the deposition of the glauconitized sediment. This apparent age, as given by K–Ar dating, can be distinguished from the glauconitic age, which corresponds to the beginning of the glauconitization process when sediment is being deposited. The duration of this process, calculated from the K content, is based from different studies [10] made in recent

environments (Holocene–Pleistocene), and in the lower Eocene (Uppermost Ypresian–Lowermost Lutetian). Observations of glauconite minerals have been done in present-day sea-bottom sediments from Congolese continental shelf, Aegean Sea (Würmian glauconitic clays), off Vancouver Island. The timing of glauconitization have been investigated using the differences between the age of the sediment deposits and the age of closure of glauconites. The estimation gives duration less than 20 ka for a slightly evolved glauconite (4.2% K) [10]. In the lower Eocene, at Mont Cassel (North of France), 60 m of glauconitic deposits in an offshore environment, containing evolved to highly evolved glauconites have been investigated in a stratigraphic interval of about 1 Ma of duration. Two sequences of glauconites, respectively of 20 and 40 m thick, have been observed. From the base to the top, these sequences show a linear trend from highly evolved (6.6% K) to evolved (5% K) glauconite minerals, interpreted as recording the variation of the sedimentation

410

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417

rate from condensed to low. Consequently, the durations of the highly evolved glauconites formation occurring in the two condensed levels at the base of the sequences must be lower than 500 ka. If we consider the average sedimentation rate of the whole sequence, it is on the order of 400 ka. These different studies argue for the duration of the glauconitization processes ranging from 10–20 ka for slightly evolved glauconites (4% K) to 300–400 ka for highly evolved glauconites (6.5% K) [8,10]. Therefore, we have used the K content of the glauconite to quantify the timing of this process, from 0.1 Ma for the less evolved (5.4% K) to 0.4 Ma for highly evolved glauconitic minerals (6.9% K). The apparent ages obtained by K–Ar and the glauconitic ages, which account for the duration of the glauconitization process as explained above, are listed in Table 1. Note that this correction is relatively minor and, in all cases, the two ages remain undistinguishable within uncertainties. Smith et al. [12] suggested that the period of glauconite mineral formation could take place over periods between 0 and 8 Ma, based on 15 single-grain 40Ar/39Ar dating of this mineral. But the 39Ar recoil, which is a consequence of the neutron irradiation necessary for dating by this technique, has probably not been properly accounted for in that study. This is illustrated by the age distribution of GL-O glauconite single grains they obtained (from 95.8 to 88.3 Ma) with several peaks centered over 95.5, 94.5, 92.5, 91.0 and 88.0 Ma. With such distribution, it appears impossible to obtain the 95 Ma K–Ar age, which is routinely measured for the GL-O standard [8,13] using 100 mg aliquots. Consequently, we can only assume that no reliable duration of the genesis of glauconite minerals can be derived from the Smith et al. [12] study. The K–Ar data, optimized to account for the delay of the glauconization process as a function of their K content [10], yield ages from 123.6 ± 1.7 Ma for the base of Hauterivian to 96.9 ± 1.4 Ma for the Upper Albian (Vraconian) (Table 1). 4. Astronomical calibration of stage durations 4.1. Methodology The relative duration of each stage, and in many cases substages within the Lower Cretaceous, has been estimated from cyclostratigraphic analysis of rhythmic sedimentation in different basins. This approach investigates the cyclicities expressed in the sedimentary successions at different time frequencies in order to reveal their origin [14]. The recognition of the orbital parameters of the Earth (precession of the equinoxes, obliquity

and eccentricity of the Earth orbit), also called the Milankovitch cycles, as the preponderant forcing parameters provides an excellent opportunity to precisely calibrate the duration of each geological event or period. Frequencies have been shown to remain constant for the last 200 Ma [15]. Two main approaches are commonly used. The first approach is based on the analysis of the fluctuations of geochemical or physical properties of the lithology (carbonate content, magnetic susceptibility, color) by spectral analysis. The wavelength ratios of the main periodicities deduced from the periodograms are compared with time/period ratios of the Milankovitch cycles in order to identify a possible link with the Earth orbital parameters. In sedimentary successions comprising important fluctuations in the sedimentation rates, a tuning of the sedimentary record from a main extracted frequency can be carried out with the aim of correcting the induced possible deviations through time [16,17]. The second approach is based on the analysis of the lithological stacking pattern, mainly from the thickness and hardness of strata and/or from the color variations. It uses the recognition of different orders of cyclicities and their logic of fitment, compared to the Earth orbital parameters pattern, in a similar fashion than Russian dolls. This approach allows the conversion of the cyclic lithological record into time cycles in the Milankovitch band frequencies [18,19]. 4.2. The Hauterivian, Barremian and Albian stage durations The Hauterivian, Barremian and Albian stages have been calibrated from Tethyan ammonite zonations on the pelagic succession of the Vocontian basin [20,21] and of central Italy [17–19,22], respectively. The stage and sub-stage boundaries are supported by magnetostratigraphy coupled with ammonite Tethyan zonation for the Barremian [23], and planktonic foraminifera, nannofossils and dinoflagellate cysts zonations for the Albian [24,25]. For this latter interval, the micropaleontological zonations of central Italy have been translated to the Tethyan ammonite zonations of the Vocontian basin from the first occurrence (FO) of the micropaleontological taxa in the two basins, assuming negligible diachronism of these events within the same paleogeographic province [20,26]. For these different periods, the timing has been obtained from both spectral analysis and “Russian dolls” fitment methodologies ([18,19,21,27]; Fiet, unpublished data for Hauterivian). The durations obtained from independent studies with different approaches match well. For example, the Albian stage is

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417

calculated at 11.6 ± 0.2 Ma using the stratonomic approach [19] and confirmed at 11.8 ± 0.4 Ma using spectral analyses on the same sedimentary succession of the central Italy basin [17]. The cyclostratigraphic durations of 5.3 ± 0.4 Ma, 5.1 ± 0.3 Ma and 11.6 ± 0.2 Ma for the Hauterivian, Barremian and Albian stages, respectively, are well determined and all agree with the last LCTS of the International Commission of Stratigraphy (ICS) [28]. 4.3. The Aptian stage The duration of the Aptian stage remains challenging for the LCTS because of its strong temporal instability in the literature. In the last LCTS of the ICS, the duration of the Aptian stage is estimated at 13.0 ± 2.0 Ma [28], while it was only at 5 Ma in some previous LCTS (e.g. [5]). For the last decade, several cyclostratigraphic studies have been carried out independently in different sedimentary depositional environments and different basins within the Tethyan domain. A cyclostratigraphic analysis of shallow-water carbonate deposits in South Italy led to about 7.2 Ma for the Aptian stage duration [29]. Another study investigated the cyclic magnetic susceptibility variations of hemipelagic carbonate sediments deposited with a high sedimentation rate in the Vocontian basin (SE of France) on a 22 m thick representative interval [30,31]. We extrapolate the results on the whole Aptian succession and obtain duration of about 6.9 Ma. This value is in total agreement with the recently derived duration of 6.7 ± 0.2 Ma (Fiet et al., in preparation) from a more detailed spectral analysis study of color and lithological variations of the same sedimentary deposits, which took into account the variations of sedimentation rates and changes in the sedimentary alternations. A third cyclostratigraphic study related to the famous “Scisti a fucoidi” Formation, a pelagic carbonate succession with a low sedimentation rate located in central Italy, constrained the Aptian duration at 6.4 ± 0.2 Ma [22]. The different cyclostratigraphic approaches of the Aptian interval show a great homogeneity of the estimated durations (less than 14% of variability), despite different depositional environments, sedimentation rates and basins have been considered. These concordant results demonstrate that local sedimentological biases, such as condensation intervals or small hiatuses, have been properly accounted for in the above studies. However, the small variability in the Aptian durations is probably due to the stratigraphic uncertainties on the stage boundaries characterization within these different environments. The base of the Aptian stage is defined by

411

the ICS at the base of the M–0 magnetic chron. This boundary is used for the cyclostratigraphy applied in central Italy sections [18,22]. In South Italy and in the Vocontian basin, the M–0 magnetic chron is not recorded because of diagenetic alteration of the magnetic minerals. The boundary is thus located, using the ammonite standard zonation, at the base of the Tuarkyricus zone [32]. However, the exact correspondence between these two scales remains controversial. Moreover, the top of the Aptian is historically defined using ammonite zonations, but with an endemic taxon as reference marker only observed in Germany within the boreal province, but not within the Tethyan domain [33–35]. Because of this controversy, we follow the proposition of the last LCTS [28] and place the boundary at the lowest occurrence of Praediscosphaera columnata calcareous nannofossil, synchronous with the last occurrence of Ticinella bejaouaensis planktonic foraminifer. Thus, we cannot rule out a small diachronism of these biostratigraphic events between the different basins. Taking into account these considerations, an average Aptian duration of 6.8 ± 0.4 Ma estimated from cyclostratigraphy can be deduced. Note that it is in strong disagreement with the value of 13 Ma proposed within the last LCTS of the ICS [28]. 5. Discussion 5.1. Stage and sub-stage durations: cyclostratigraphy vs. geochronology The occurrence of both stratigraphically well-dated glauconitic horizons and several cyclostratigraphic studies for the Hauterivian–Albian interval provides a great opportunity to compare the durations derived from the orbital chronology and from the radiometric geochronology. It should lead to more stable durations for the stages and most of the related sub-stages. The comparisons have been carried out on the stratigraphic intervals containing glauconitic horizons closed to an ammonite zone or sub-zone boundary for an optimization of the extrapolated durations from radiometric data. The selected intervals are the Hauterivian stage and the Upper Aptian, Middle Albian, Upper and Uppermost Albian substages (Fig. 2). The estimated durations based on the two independent approaches are very similar (Fig. 2) in spite of possible uncertainties due to the nature of the glauconitic horizons (condensation levels), their stratigraphic location slightly shifted compared to the biostratigraphic boundaries, and the position of the biostratigraphic boundaries for the cyclostratigraphic estimations.

412

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417

Fig. 2. Comparisons of the durations of selected stages or sub-stages, estimated by orbital chronology and extrapolated from K/Ar radiochronology on glauconite minerals. For the latter, the glauconitic horizons used for the calculations are indicated between brackets.

From these results, we can extend this comparative approach to the whole of the Hauterivian–Albian period, even in the intervals without glauconitic horizons. 5.2. Calibration of the stage and substage ages Calibration of the stage and substage durations, from K–Ar radiometric ages and cyclostratigraphic data within the whole of the interval investigated here, requires a common radiometric anchor point. The GLO international standard [8] has been chosen because: 1) it is a pure, highly evolved glauconite similar to those analyzed here; 2) the preparation and the analysis of the glauconite minerals have been made in the same

laboratory conditions as the studied samples; 3) it is one of the standards used in the Orsay UPS-IPGP K– Ar laboratory to validate the volumetric calibration of the argon line; and 4) its stratigraphic position is very well defined. In order to validate the age of 95.0 Ma [8] for the GL-O standard, and hence the recommended value of 6.679 × 1014 atom/g of 40Ar⁎ that we use in the Orsay UPS-IPGP K–Ar laboratory for argon signal calibration, we have dated the widely used standard MMhb-1 [36] using the same protocol. From nine independent argon determinations (Fig. 3), we obtained a mean age of 525.0 ± 2.1 Ma (K = 1.560%) for MMhb-1 relative to GL-O at 95.0 Ma [8]. Within uncertainties, this agrees with recent

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417

Fig. 3. Replicated K–Ar ages of MMhb-1 relative to the recommended value for GL-O [13]. Note that in order to account for the recognized inhomogeneity of MMhb-1 standard for lower sizes, aliquots of about 60 mg (from split 6-58-1/4) have been analyzed here.

determinations of 523.1± 2.6 or 525.1± 2.3 Ma, depending on the primary standard used [37], and 523.2 ± 0.9 Ma [38], and hence validates the age of GL-O. Additional constraint is provided by the good agreement, within uncertainties, between K–Ar dating of the Matuyama– Brunhes geomagnetic reversal at 786 ± 8 ka [39], relative to the recommended value for GL-O [8], and the independent Astronomical Polarity Time Scale (APTS) age of 779 ± 2 ka [40]. The GL-O standard is precisely located within the Carcitanensis ammonite sub-zone in the Mantelii zone at the base of the Cenomanian [13]. The end of the glauconitization process of GL-O on carbonated substrates took place during the formation of a hard ground at the top of the sub-zone. The age of GL-O is thus coincident with the top of this sub-zone [13]. The duration of the H. carcitanensis sub-zone may be calculated assuming an equal duration of each ammonite zone and sub-zones within each zone. Considering a duration of 4.45 Ma for the Cenomanian, as determined by cyclostratigraphy using precession signals recorded in Western Europe basins [41], and the ammonite zonation of the Anglo-Paris basin as the stratigraphic framework (7 zones and 3 sub-zones in the Mantelii zone [3,13]), this sub-zone spans about 0.3 Ma. The maximum duration between the formation of GL-O and the Albian/Cenomanian boundary should thus be 0.3 Ma, which is also in agreement with the time of

413

formation of GL-O estimated from its K content (6.55%). Consequently, the age of the Albian–Cenomanian boundary can be determined by combining the radiometric age of GL-O with the time of formation of the glauconite minerals, and is proposed as our anchor point at 95.3 ± 1.1 Ma for this study. Note that this age is in good agreement with K–Ar and Rb/Sr glauconite ages obtained around the Albian–Cenomanian boundary at 96.0 ± 1.9 Ma on drill cores in the Vocontian basin [42]. The similarity of K–Ar and Rb/Sr ages independently obtained supports their reliability if they are carefully selected, and indicates “no further pervasive recrystallization affecting the glauconite grains, either by heat flux or fluid flow after deposition of the sediments” [42]. The cyclostratigraphic ages of each sampled glauconitic horizon have been calculated using the GL-O anchor point and the orbital durations defined above. Note that for horizon 2b located within the middle Albian sub-stage, biozones have been considered of equal duration. The comparison of the two databases shows a very well-defined linear correlation (Fig. 4), indicating a strong coherence in the estimations of the stage and related substages durations proposed in this paper. 5.3. The Aptian stage, a controversial duration: implication for the age of the M–0 magnetic chron The linear correlation obtained in Fig. 4 highlights the controversy regarding the duration of the Aptian stage and its implication for the LCTS. Indeed the most

Fig. 4. Correlation diagram between K–Ar and extrapolated cyclostratigraphic ages. R: Pearson's correlation coefficient.

414

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417

recent LCTS [28] suggests a duration of 13.0 ± 2.0 Ma, while we obtained 6.8 ± 0.4 Ma (Fig. 5). The LCTS duration of Gradstein et al. [28] is evaluated based only on two anchorage radiometric points. The first corresponds to an 40Ar/39Ar age from a bentonite located in the Nutfieldensis ammonite zone (Upper Aptian) in Germany [1]. The second is derived from an 40Ar/39Ar age obtained on a weathered basalt at the top of the MIT guyot in the Western Pacific [43], which records a reverse-polarity magnetic zone attributed to the M–0 magnetic chron, and is thus located at the Barremian–Aptian boundary [44].

The strong Aptian duration discrepancy observed (Fig. 5) between our result and Gradstein et al. [28] time scale could not rely on stratigraphic boundaries assignment, since they are identical. On the other hand, we think the M–0 magnetic chron duration and age are not suitably well-constrained because the quality of the material used for radiometric dating and the stratigraphic attribution of the reverse-polarity magnetic zone recorded at the top of the MIT guyot remain perfectible. Unfortunately, pacific seamount basalts and magmatic intrusions are indirectly calibrated to the biostratigraphic data (rarely by ammonites) of the sedimentary cover.

Fig. 5. Comparisons between the estimated durations of the Hauterivian–Albian stages from this study, and the proposed durations of the last LCTS for the same time interval.

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417

Moreover, the calibration is sometimes only derived from magnetostratigraphy, which, in a context of generally poor core recovery, is of low reliability. To this regard, we think that the calibration of the geologic time scale should not rely on a set of often weathered (sampled underwater) basaltic (low K, high Ca) whole rock 40 Ar/39Ar ages [43–45]. In addition, varying spreading rates, as well as late magmatic intrusions can significantly limit the use of oceanic basalt for LCTS calibration (e.g. [46]). From the above remarks, it ap-pears that the M–0 chron remains to be calibrated with high quality radiometric ages from volcanic products stratigraphically well-constrained and distributed around this magnetic event before it can be used for the LCTS calibration. 5.4. The glauconite mineral as a relative geochronometer The estimation of the stage and related substage durations for the whole Hauterivian–Albian interval investigated here, by K–Ar dating on glauconite minerals and cyclostratigraphy, are indistinguishable. Clearly, this indicates that any underestimation of the radiometric ages based on glauconite minerals is identical for all horizons analyzed here, and would only lead to a constant shift of all ages. A prior study on several Cretaceous condensed horizons throughout the Vocontian basin displayed rejuvenation of the glauconite measured ages, interpreted as being due to tectonic activity since the Upper Cretaceous [47]. Based on the new results presented here, the rejuvenated ages are most probably due to other influences, including analyses of only slightly evolved glauconite, as well as the nature of the glauconite-bearing rocks (not indicated by the authors) and their related differential diagenesis (oxidizing vs. reduced environmental conditions at the water–sediment interface, and acidic vs. basic fluid flows in the sediment during the formation of glauconite minerals).

415

Hauterivian, Barremian, Aptian and Albian stages, respectively. The coherence of the two data sets strongly argues that glauconite minerals are powerful radiochronometers, when precisely stratigraphically defined and carefully selected only as evolved or highly evolved. Moreover, a delay of closure of the glauconite minerals of several Ma, up to 5 Ma as sometimes suggested, seems clearly unrealistic here. Glauconite minerals allow direct dating of the sedimentary sequences and are common in sedimentary basins. Our study confirms their potential for use as absolute radiochronometer. However, high-quality K–Ar or 40Ar/39Ar step heating ages from well-positioned bentonites within each stage of the Lower Cretaceous are still needed to compare absolute dating from highly evolved glauconite and high-temperature minerals. The calibration of the Hauterivian–Albian interval developed in this paper highlights the problem of the estimated Aptian duration in the most recent published LCTS [28] and, consequently, the radiometric age of the M–0 magnetic chron. The difference of more than 6 Ma in the estimated duration is probably caused by the too scarce radiometric anchorage points available to better constrain the M–0 chron through time, and possibly combined with dating of altered rocks. Finally, the excellent correlation between cyclostratigraphy and K– Ar radiometric dating of glauconites constraints the stage duration within the Lower Cretaceous, from the Hauterivian to the Albian. Acknowledgments D. Besson, G. Friès, J.-L. Latil, and M. Pagel are thanked for stimulating discussions about biostratigraphy, diagenesis and field assistance. The original version of the manuscript was greatly improved following careful reading by Y. Gallet and J. Holt. Reviews by two anonymous referees helped to improve the overall clarity of the text. This is LGMT contribution No. 60.

6. Conclusions References The present study shows that K–Ar dating of well separated highly evolved glauconite from numerous stage boundaries within the Lower Cretaceous is (1) self-coherent and (2) in very good agreement with the astronomical calibration. The calibration of the Hauterivian–Albian stage and related substages durations from independent radiochronologic and cyclostratigraphic approaches are strongly correlated. Thus, we propose durations of 5.3 ± 0.4 Ma, 5.1 ± 0.3 Ma, 6.8 ± 0.4 Ma, and 11.6 ± 0.2 Ma for the

[1] J.D. Obradovich, A Cretaceous time scale, in: W.G.E. Caldwell (Ed.), Evolution of the Western Interior Basin, Geol. Assoc. Can. Spec. Pap., vol. 39, 1993, pp. 379–396. [2] F.M. Gradstein, F.P. Agterberg, J.G. Ogg, J. Hardenbol, P. Van Veen, J. Thierry, Z. Huang, A Mesozoic time scale, J. Geophys. Res. 99 (1994) 24,051–24,074. [3] F.M. Gradstein, F.P. Agterberg, J.G. Ogg, J. Hardenbol, P. Van Veen, J. Thierry, Z. Huang, A Triassic, Jurassic and Cretaceous time scale, SEPM Spec. Pub. 54 (1995) 95–126. [4] W.J. Kennedy, G.S. Odin, in: G.S. Odin (Ed.), The Jurassic and Cretaceous Time Scale in 1981, Numerical Dating in

416

[5] [6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417 Stratigraphy, vol. 1, John Wiley and Sons, New York, 1982, pp. 557–592. G.S. Odin, Geological time scale, C. R. Acad. Sci., Ser. II 318 (1994) 59–71. G. Friès, Dynamique du bassin subalpin méridional de l'Aptien au Cénomanien, Mémoire des Sciences de la Terre Ecole des Mines de Paris, Université Paris VI, 1986. G. Fries, O. Parize, Anatomy of ancient passive margin slope systems: Aptian gravity-driven deposition on the Vocontian palaeomargin, western Alps, south-east France, Sedimentology 50 (6) (2003) 1231–1270. G.S. Odin, et al., Interlaboratory standards for dating purposes, in: G.S. Odin (Ed.), Numerical Dating in Stratigraphy, vol. 1, John Wiley and Sons, Chichester, 1982, pp. 123–150. P. Cotillon, Le Crétacé inférieur de l'arc subalpin de Castellane entre l'Asse et le Var. Startigraphie et sédimentologie, 520 pp., B. R.G.M., Paris, 1971. G.S. Odin, P.D. Fullagar, Geological significance of the glaucony facies, in: G.S. Odin (Ed.), Green Marine Clays, Elsevier, Amsterdam, 1988. C. Cassignol, P.-Y. Gillot, Range and effectiveness of unspiked potassium–argon dating: experimental groundwork and applications, in: G.S. Odin (Ed.), Numerical Dating in Stratigraphy, John Wiley, New York, 1982, pp. 159–179. P.E. Smith, N.M. Evensen, D. York, G.S. Odin, Single-grain 40 Ar–39Ar ages of glauconies: implications for the geologic time scale and global sea level variations, Science 279 (1998) 1517–1519. G.S. Odin, J.C. Hunziker, in: G.S. Odin (Ed.), Radiometric Dating of the Albian–Cenomanian Boundary, Numerical Dating in Stratigraphy, vol. 1, John Wiley and Sons, Chichester, 1982, p. 630. A.G. Fischer, B. D'Argenio, I. Premoli Silva, H. Weissert, V. Ferreri, Cyclostratigraphic approach to Earth's history: an introduction, SEPM Spec. Pub. 81 (2004) 5–17. A. Berger, M.F. Loutre, J. Laskar, Stability of the astronomical frequencies over the Earth's history for paleoclimate studies, Science 255 (1992) 560–566. H. Mayer, E. Appel, Milankovitch cyclicity and rock-magnetic signatures of palaeoclimatic change in the Early Cretaceous Biancone formation of the southern Alps, Italy, Cretac. Res. 20 (1999) 189–214. A. Grippo, A.G. Fischer, L.A. Hinnov, T.M. Herbert, I. Premoli Silva, Cyclostratigraphy and chronology of the Albian stage (Piobbico core, Italy), SEPM Spec. Pub. 81 (2004) 57–83. N. Fiet, G. Gorin, Lithological expression of Milankovitch cyclicity in carbonate-dominated, pelagic, Barremian deposits in central Italy, Cret. Res. 21 (2000) 457–467. N. Fiet, B. Beaudoin, O. Parize, Lithological analysis of Milankovitch cyclicity in pelagic Albian deposits of central Italy: implications for the duration of the stage and substages, Cret. Res. 22 (2001) 265–275. L.G. Bulot, J.P. Thieuloy, Les biohorizons du Valanginien du Sud-Est de la France: un outil fondamental pour les corrélations au sein de la Téthys occidentale, Lower Cretaceous Cephalopod Biostratigraphy of the Western Tethys, Geol. Alpine, vol. 20, Mem H.S, Grenoble, 1994, pp. 15–41. Z. Huang, J.G. Ogg, F.M. Gradstein, A quantitative study of Lower Cretaceous cyclic sequences from the Atlantic Ocean and the Vocontian basin (SE France), Paleoceanography 8 (1993) 275–291. N. Fiet, Temporal calibration of the Aptian stage and related substages by a cyclostratigraphic approach applied to the pelagic

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34] [35]

[36]

[37]

[38]

successions of the Umbria–Marche basin (Central Italy), Bull. Soc. Geol. France 171 (2000) 103–113. F. Cecca, G. Pallini, Latest Hauterivian–Barremian ammonite biostratigraphy in the Umbria–Marche Apennines (central Italy), in: L.-G. Bulot, M. Argot, H. Arnaud (Eds.), Lower Cretaceous Cephalopod Biostratigraphy of the Western Tethys, Geol. Alpine, vol. 20, Mem H.S, Grenoble, 1994, pp. 205–218. E. Erba, Aptian–Albian calcareous nannofossil biostratigraphy of the Scisti a fucoidi cored at Piobbico (central Italy), Riv. Ital. Paleontol. Stratigr. 94 (2) (1988) 249–284. N. Fiet, E. Masure, Albian dinocystes from the Umbria–Marche basin (Central Italy): new biozonation for the Tethyan realm, Cret. Res. 22 (2001) 63–77. P.J. Hoedemaeker, P.F. Rawson, Report on the 5th International Workshop of the Lower Cretaceous Cephalopod Team (Vienna, 5 September 2000), Cret. Res. 21 (2000) 857–860. T.D. Herbert, Paleomagnetic calibration of Milankovitch cyclicity in Lower Cretaceous sediments, Earth Planet. Sci. Lett. 112 (1992) 15–28. F.M. Gradstein, J.G. Ogg, A.G. Smith, W. Bleeker, L.J. Lourens, A new geologic time scale with special reference to Precambrian and Neogene, Episodes 27 (2004) 83–100. B. D'Argenio, V. Ferreri, H. Weissert, S. Amodio, F.P. Buonocunto, L. Wissler, A multidisciplinary approach to global correlation and geochronology. The Cretaceous shallow-water carbonates of Southern Apennines, Italy, SEPM Spec. Pub. 81 (2004) 103–123. P. Kößler, J.O. Herrle, E. Appel, J. Erbacher, C. Hemleben, Magnetic records of climatic cycles from Mid-Cretaceous hemipelagic sediments of the Vocontian basin, SE France, Cret. Res. 22 (2001) 321–331. J.O. Herrle, Paleoceanographic and Paleoclimatic Implications on Mid-Cretaceous Black Shale Formation in the Vocontian Basin and the Atlantic: Evidence from calcareous Nannofossils and Stable Isotopes, Institut and Museum Univ. Tübingen, Germany, 2002 113 pp. G. Delanoy, Biostratigraphie des faunes d'ammonites à la limite Barrémien–Aptien dans la region d'Angles–Barrême–Castellane, Ann. Mus. Hist. Nat. Nice 12 (1997) 270. R. Casey, The age of the Argiles à Bucaillella of Normandy, the systematic position of the Cretaceous ammonite genera Bucaillella and Archoplites, and the delimitation of the Aptian–Albian boundary, Cret. Res. 20 (1999) 609–628. H.G. Owen, The base of the Albian stage; comments on recent proposals, Cret. Res. 23 (2002) 1–13. W.J. Kennedy, A.S. Gale, P.R. Bown, M. Caron, R.J. Davey, D. Gröcke, D.S. Wray, Integrated stratigraphy across the Aptian– Albian boundary in the Marnes Bleues, at the Col de PréGuittard, Arnayon (Drôme), and at Tartonne (Alpes de Haute Provence), France: a candidate Global Boundary Stratotype Section and Boundary Point for the base of the Albian stage, Cret. Res. 21 (2000) 591–720. S.D. Samson, E.C.J. Alexander, Calibration of the interlaboratory 40Ar/39Ar dating standard, MMhb-1, Chem. Geol. 66 (1987) 27–34. P.R. Renne, C.C. Swisher, A.L. Deino, D.B. Karner, T.L. Owens, D.J. DePaolo, Intercalibration of standards, absolute ages and uncertainties in 40Ar/39Ar dating, Chem. Geol. 145 (1998) 117–152. T.L. Spell, I. McDougall, Characterization and calibration of 40 Ar/39Ar dating standards, Chem. Geol. 198 (2003) 189–211.

N. Fiet, et al. / Earth and Planetary Science Letters 246 (2006) 407–417 [39] X. Quidelleur, J. Carlut, V. Soler, J.-P. Valet, P.-Y. Gillot, The age and duration of the Matuyama–Brunhes transition from new K– Ar data from La Palma (Canary Islands) and revisited 40Ar/39Ar ages, Earth Planet. Sci. Lett. 208 (2003) 149–163. [40] L. Tauxe, T. Herbert, N.J. Shackleton, Y.S. Kok, Astronomical calibration of the Matuyama–Brunhes boundary: consequences for magnetic remanence acquisition in marine carbonates and the Asian loess sequences, Earth Planet. Sci. Lett. 140 (1996) 133–146. [41] A.S. Gale, Cyclostratigraphy and correlation of the Cenomanian stage in Western Europe, Orbital Forcing Timescales and Cyclostratigraphy, Geological Society Special Publication, vol. 85, 1995, pp. 177–197. [42] D. Rousset, S. Leclerc, N. Clauer, J. Lancelot, M. Cathelineau, J.-F. Aranyossy, Age and origin of Albian glauconites and associated clay minerals inferred from a detailed geochemical analysis, J. Sediment. Res. 74 (5) (2004) 631–642. [43] M.S. Pringle, R.A. Duncan, Radiometric ages of basement lavas recovered at Loen, Wodejebato, MIT, and Takuyo–Daisan

[44]

[45]

[46]

[47]

[48]

417

guyots, northwestern Pacific ocean, Proceeding O.D.P., Scientific Results 144 (1995) 547–557. J.E.T. Channell, E. Erba, M. Nakanishi, K. Tamaki, Late Jurassic–Early Cretaceous time scales and oceanic magnetic anomaly block models, SEPM Spec. Pub. 54 (1995) 51–63. E.L. Winterer, J.H. Natland, R.J. Van Waasbergen, R.A. Duncan, M.K. Mc Nutt, C.J. Wolfe, I. Premoli Silva, W.W. Sager, W.V. Sliter, The Mesozoic Pacific: Geology, Tectonics, and Volcanism, Geological Monography American Geophysical Union, New York, 1993 307–334 pp. A.A.P. Koppers, H. Staudigel, R.A. Duncan, High-resolution 40Ar/ 39 Ar dating of the oldest oceanic basement basalts in the western Pacific basin, Geochem. Geophys. Geosyst. 4 (2003) 11. M. Conard, H. Kreuzer, G.S. Odin, Potassium–Argon dating of tectonized glauconites, in: G.S. Odin (Ed.), Numerical Dating in Stratigraphy, Wiley, Chichester, 1982, pp. 321–332. R.H. Steiger, E. Jäger, Subcommission on geochronology: convention on the use of decay constants in geo and cosmochronology, Earth Planet. Sci. Lett. 36 (1977) 359–362.