Cyclostratigraphy of the Upper Cretaceous Niobrara ... - CiteSeerX

Earth and Planetary Science Letters 269 (2008) 540-553

Contents lists available at ScienceDirect

Earth and Planetary Science Letters j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e p s l

Cyclostratigraphy of the Upper Cretaceous Niobrara Formation, Western Interior, U.S.A.: A Coniacian–Santonian orbital timescale Robert E. Locklair ⁎, Bradley B. Sageman Department of Earth and Planetary Sciences, Northwestern University, 1850 Campus Drive, Evanston, Illinois, 60208, United States

A R T I C L E

I N F O

Article history: Received 25 September 2007 Received in revised form 28 February 2008 Accepted 4 March 2008 Available online 18 March 2008 Editor: M.L. Delaney Keywords: Coniacian Santonian Niobrara formation spectral analysis orbital timescale

A B S T R A C T The Turonian–Campanian Niobrara Formation of the Western Interior basin, U.S.A., is characterized by decimeter-scale rhythmic alternations of chalk and marl beds and decameter-scale oscillations of chalky and marly facies. This study applies recent advances in quantitative cyclostratigraphic analysis to high-resolution geophysical records in order to test for an orbital signal. We studied records of the entire formation (~ 85 m) from two wells in Colorado (40° 17′ N, 104° 38′ W; 40° 14′ N, 104° 41′ W). The study utilized high-resolution time series of borehole resistivity measured from Formation MicroImager tools (FMI). Advanced spectral techniques indicate the presence of eccentricity, obliquity, and precession periodicities throughout the study interval. Deviations of bundling ratios in the lower Fort Hays Member from those predicted by simple eccentricity modulation of precession, previously reported in the literature, are likely due to the influence of obliquity. Deconvolution and frequency tracking of orbital components were used to reconstruct sedimentation rates for the Niobrara Formation and thus develop a high-resolution orbital timescale, which permits calculations of accumulation rates for sedimentary components and the duration of inclusive stages, substages, and, in some cases, biozones. The Niobrara Formation in the studied cores appears to be hiatusfree and therefore provides a continuous record of rhythmic deposition for astronomical timescale development. The relationship between carbonate content and sedimentation rate throughout the Niobrara Formation indicates that the decameter-scale oscillations of chalk and marl are dominantly driven by variations in siliciclastic flux (dilution). The duration, derived from the orbital timescale, between the Turonian–Coniacian and Coniacian–Santonian boundaries is 3.40 ± 0.13 myr, which is generally consistent with radiometric estimates for the Coniacian. The orbitally derived duration of the Santonian Stage is 2.39 ± 0.15 myr, which is very similar to estimates based on curve fitting between radiometric dates. Development of high-resolution orbital chronometers provides excellent time-resolution in intervals that lack radiometric-time control or where control points are widely spaced. © 2008 Elsevier B.V. All rights reserved.

1. Introduction The study of sedimentary cyclicity has long attracted the interest of geoscientists because of its dual promise of enhancing chronostratigraphic resolution and of providing historical evidence for fundamental Earth processes, such as tectonics and climate, interpreted through their roles as drivers of sedimentation. Fischer's (1980) work on rhythmic bedding in the Cretaceous Greenhorn and Niobrara Formations of the Western Interior U.S. stimulated much interest in these relationships. He summarized the pioneering work of Gilbert (1895) and others who recognized that periodic variations in the Earth's orbit might modulate climate and thus produce the rhythmic patterns of sedimentation observed in many sedimentary sequences. It

⁎ Corresponding author. Tel.: +1 847 491 8180; fax: +1847 491 8060. E-mail addresses: [email protected] (R.E. Locklair), [email protected] (B.B. Sageman). 0012-821X/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.epsl.2008.03.021

was Gilbert (1895) who first recognized that orbitally forced stratigraphy might, therefore, contain an embedded high-resolution record of geologic time. This study aims to evaluate the upper Turonian–lower Campanian depositional rhythms of the carbonate-rich Niobrara Formation in the Western Interior basin of North America using recently refined versions of quantitative signal-processing techniques (e.g., Park and Herbert, 1987; Hinnov, 2000). Investigation of Niobrara depositional rhythms is significant because quantification of the orbital signal provides an opportunity to construct a high-resolution orbital timescale through a thick stratigraphic interval, estimated to be at least 5.5 myr in duration (Obradovich, 1993), that includes a conformable record of the Coniacian and Santonian stages. An assessment of durations for the Coniacian and Santonian stages that is independent of the radiometric-time scale can be made using the preserved orbital record. The new timescale also makes possible reconstruction of accumulation rates for the main sedimentary inputs, which allows hypothesized mechanisms for generation of rhythmic strata in the Western Interior (i.e., productivity, dilution,

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

541

redox, scour, and dissolution cycles; Arthur et al., 1984) to be more rigorously tested. Using available biostratigraphic data, the timescale can be exported to other Coniacian–Santonian intervals and thus aid future studies of the poorly understood Oceanic Anoxic Event (OAE) III (Arthur and Schlanger, 1979). The study also demonstrates the utility of high-resolution geophysical logs (FMIs—Formation Micro-resistivity Imaging) for detection of primary depositional rhythms and underscores the utility of these available time series for addressing possible orbital influence in other localities and time intervals.

In recent years, some of the more sophisticated (e.g., MTM; Thomson, 1982) spectral techniques have been employed to identify and track orbital signals within rhythmically bedded facies. These methods allow detailed sedimentation rate histories and high-resolution orbital time scales to be developed (e.g., Olsen and Kent, 1999; Meyers et al., 2001), thus significantly enhancing the value of geochemical data from such facies. With the development of more sophisticated analytical approaches, studies of longer, noisier rhythmic intervals, that are not well constrained geochronologically, have become more feasible.

2. Geologic background

2.2. Turonian–Campanian of the Western Interior Basin

2.1. Milankovitch theory and spectral analytical techniques

As a result of load-induced subsidence and global eustatic rise, the Western Interior region (Fig. 1) was inundated by marine waters episodically from Albian through Maastrichtian time (Kauffman and Caldwell, 1993). Following peak regression of the Greenhorn marine cycle, a late Turonian transgression flooded the seaway and hemipelagic carbonate-rich sediments were deposited from the Texas Gulf Coast to the Western Canadian Sedimentary Basin until early Campanian time (Kauffman and Caldwell, 1993). In the central part of the Western Interior (e.g., Denver basin, Wattenberg Field) these hemipelagic units are termed the Niobrara Formation and they unconformably overlie the middle Turonian Codell Sandstone Member of the Carlile Shale. The dominantly carbonate-poor Campanian Pierre Shale overlies the Niobrara Formation. Formal members of the Niobrara include the Fort Hays Limestone (~ 10 m thick) and the overlying Smoky Hill Chalk (≥80 m thick). Scott and Cobban (1964) subdivided the Smoky Hill Chalk Member into 7 informal chalk and marl submembers (Fig. 2). At a decameter scale, oscillations in predominant lithotype (organic-poor limestone/chalk vs. organic-rich calcareous shale/marl) form the basis for informal Smoky Hill submembers (Fig. 2).

As known to Gilbert (1895), variations in the Earth's orbital parameters alter the amount/distribution of insolation received (Milankovitch cycles) and thus may periodically modulate the Earth's climate system. This climate modulation has the potential to be recorded in climate-sensitive sediments (Hays et al., 1976; Fischer, 1980; Fischer and Bottjer, 1991; DeBoer and Smith, 1994; House, 1995). In an effort to detect such orbital signals, spectral techniques have been applied to ancient strata in various localities (e.g., Anderson, 1984; Herbert and Fischer, 1986; Weedon, 1989; Park et al., 1993; D'Argenio et al., 1997; Olsen and Kent, 1999; Preto et al., 2004; Beckmann et al., 2005), including the Western Interior basin, U.S.A. (Sageman et al., 1997; Meyers et al., 2001). These studies have employed a variety of time series datasets, including bedding thicknesses, oxygen isotope ratios, weight percent calcium carbonate, grayscale pixel data, and drill core magnetic susceptibility, and they have utilized various methodologies, such as bundling ratios, Fourier Analysis with multiple, independent data tapers (MTM), evolutive harmonic analysis (EHA), and amplitude and frequency modulation analysis (AM/FM) (see Hinnov, 2000 and Weedon, 2003 for detailed explanation of these methods). The results indicate time-stratigraphic frequencies that correlate to one or more of Earth's orbital periodicities (roughly: precession = 21 kyr; obliquity = 41 kyr; eccentricity = 100 kyr, 400 kyr, 1000 kyr) and are, therefore, interpreted as Milankovitch cycles.

2.3. Lithologic character and bedding of the Niobrara Formation Reviews of Niobrara lithostratigraphy over the past three decades by Kauffman (1969, 1977a,b), Hattin (1982), and Barlow and Kauffman

Fig. 1. Locality map of the Niobrara study area in the Western Interior basin with paleogeographic context for Coniacian–Santonian time (Roberts and Kirschbaum, 1995). The symbol L indicates the location of the Libsack 43-27 and Aristocrat Angus 12-8 (spaced 2 km apart) boreholes (spaced 2 km apart), for which FMI records have been analyzed. Symbol RC indicates the location of Rock Canyon Anticline outcrops and the USGS #1 Portland core, which provide the most constrained biostratigraphic zonation of the Niobrara Formation (Scott and Cobban, 1964; Scott et al., 1986; Cobban, 1993; Kauffman et al., 1993; Walaszczyk and Cobban, 1998, 2000, 2007). Other key outcrop and borehole records useful for determination of stratigraphic relations include Amoco Rebecca Bounds #1(ARB; Bralower and Bergen, 1998), Raton basin (RB; Scott et al., 1986), and Mesa Verde (MV; Leckie et al., 1997).

542

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

Fig. 2. Stratigraphy of the Niobrara Formation in Colorado and correlation of downhole measurements to lithostratigraphy and geochemistry. A) Lithology, resistivity, and calcium carbonate content of the Libsack 43-27 core/borehole are placed in the context of upper Turonian to lower Campanian Niobrara formal/informal lithostratigraphy (LCS—lower chalk and shale member, LM—lower marl member, LC—lower chalk member, MM—middle marl member, MC—middle chalk member, UM—upper marl member, UC—upper chalk member; Scott and Cobban, 1964). More detailed chrono- and biostratigraphy are presented later in a summary figure. B) A higher-resolution correlation of lithology, FMI, and weight percent calcium carbonate through the lower chalk and shale (LCS) interval of the Aristocrat Angus 12-8 core. Lithologies (M—marl, CM—chalky marl, MC—marly chalk, C—chalk) were assigned on the basis of rock color and degree of bioturbation, in addition to carbonate content. Numbers are in reference to chalk beds through this interval. C) Resistivity and calcium carbonate data from the LCS interval of B). D) Weight percent calcium carbonate plotted against the log of resistivity data in B). E) Changes in resistivity and calcium carbonate on a bed-by-bed basis, demonstrating that both carbonate and resistivity increase through the transition from marl to chalk and both decrease from chalk to marl. The point labeled 1 is the difference from chalk 1 in B) to the overlying marl bed and point 2 is the difference from that marl bed to Chalk 2 in B).

(1985) have retained the same formal and informal subdivisions used by Scott and Cobban (1964), shown in Fig. 2. As described in these reviews, the Fort Hays Limestone Member is characterized by thick

(up to 66 cm), micritic to chalky, bioturbated limestone beds interbedded with thin (1–15 cm), weakly calcareous, mostly laminated, organic-rich shales. The overlying Smoky Hill Chalk Member is

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

similar to the Fort Hays Member in that limestone or chalk beds (5– 20 cm) tend to be burrowed and organic-poor whereas marls (10– 50 cm) tend to be more laminated and organic-rich (Dean and Arthur, 1998; Savrda, 1998). Carbonate content is high in the Smoky Hill Chalk due to the ubiquitous coccolith-rich fecal pellets that impart a speckled appearance (Hattin, 1981, 1982), but higher overall terrigenous mud content yields an average Smoky Hill lithology closer to calcareous shale or marl (e.g., 35–75 wt.% CaCO3), as compared to the near-pure chalk or limestone (90–95 wt.% CaCO3) typical of the Fort Hays Member. Informal subdivisions of the Smoky Hill Chalk are based on varying proportions of calcareous shale or marl vs. limestone or chalk. Both scales of cyclicity in the Niobrara are reflected in weight percent carbonate and aluminum data (Dean and Arthur, 1998), reflecting proportionality of predominantly coccoliths and clay minerals, respectively. 2.4. Biostratigraphy Detailed biozonation schemes using ammonite and bivalve index taxa have been developed and refined over the years for Albian through Maastrichtian strata of the Western Interior basin (e.g., Cobban, 1993; Kauffman et al., 1993), but the resolution and geographic range of biozones and index taxa, respectively, show significant variation among different stratigraphic intervals. As a generalization, biozone resolution in the Turonian–Campanian (T-C) study interval is lower than in the other maximum highstand sequence of the underlying Cenomanian– Turonian Greenhorn marine cycle. Some of the most significant contributions to the development of biozonation in the T-C interval include Scott and Cobban (1962, 1964), Kauffman (1967, 1977a,b), Jeletsky (1968), Obradovich and Cobban (1975), Hattin (1982), Kennedy (1984), and Kennedy and Cobban (1991). The Niobrara spans twelve ammonite biozones of the standard Western Interior biozonation scheme (Scott and Cobban, 1962, 1975, 1964; Cobban, 1993; Kauffman et al., 1993), representing late Turonian through early Campanian (T-C) time. The biozonation scheme used in this study is derived from Scott and Cobban (1964), with a modification to the Turonian–Coniacian boundary placement in the Fort Hays Limestone following Walaszczyk and Cobban (1998) and placement of the Santonian–Campanian boundary from nannofossils following Watkins et al. (1993). The diachronous nature of upper and lower Fort Hays Limestone lithofacies boundaries underscores the importance of regional correlations for placement of the Turonian–Coniacian Stage boundary in cores lacking abundant index taxa. Our correlations indicate, as previously noted by others (Scott and Cobban, 1964; Hattin, 1982; Leckie et al., 1997), that informal members of the Smoky Hill Chalk are essentially isochronous from central, northern, and southwestern Colorado.

543

2.6. Cyclostratigraphy Earlier attempts to understand the origins of Niobrara cycles focused on two key factors: cycle timing and depositional mechanisms. Gilbert (1895) first proposed a dilution mechanism related to climate-controlled changes in terrigenous flux, and assumed that Niobrara cycles were the result of orbital precession. Fischer (1980) and Fischer et al. (1985) argued for precessional cycles in the Niobrara Formation based on bundling patterns in both the Fort Hays Limestone and Smoky Hill Chalk, but noted that these patterns displayed stratigraphic deviations from the ideal precessional index (1:5). A combination of dilution and productivity mechanisms with a coordinated redox effect were discussed, but a specific conclusion about control of Niobrara cyclicity was not expressly stated. Hattin (1982) believed that variations in siliciclastic dilution on a relatively constant carbonate flux were responsible for hemipelagic patterns of deposition. Laferriere et al. (1987) investigated precessional bundling patterns in the Fort Hays Member, and in addition to the stratigraphic deviations noted by Fischer (1980) and Fischer et al. (1985), used simple Fourier analysis to study geographic variations in bundling. Bundling variations were hypothesized to result from interference among competing orbital signals (e.g., precessional index and obliquity), as well as possible overprint by tectono-sedimentologic effects. Sageman et al. (1997) and Meyers et al. (2001) later came to similar conclusions about mixed orbital signals and sedimentation effects in the uppermost Cenomanian–lower Turonian Bridge Creek Limestone, and showed how advanced spectral techniques can be used to quantify these processes. Subsequent papers supported the dilution/productivity/redox hypothesis; based on available time scales and bedding patterns, a ±21 kyr precessional driver modulated by the ±100 kyr eccentricity cycle was interpreted for the Fort Hays Limestone (Pratt et al., 1993; Arthur et al., 1985; Arthur and Dean,1991; Dean and Arthur,1998). Similar decimeterscale patterns were noted in the Smoky Hill Chalk, but longer cycles observed in CaCO3 content, on the order of 3–4 m and 30 m (see Fig. 2), were believed to reflect oscillations of several hundred kyr and 2500 kyr, respectively (Dean and Arthur,1998). The shorter of these two cycles was interpreted by Dean and Arthur (1998) to reflect long eccentricity (±404 kyr), whereas the longer cycle was attributed to tectonic or sea level effects despite its similarity in duration to the “grand cycle” of eccentricity (~2000 kyr) noted by Olsen and Kent (1999) in Triassic– Jurassic lake deposits. Advanced spectral techniques have the power to better resolve nested cycle periodicities and are applied to the Niobrara Formation in this study. 3. Materials and methods

2.5. Chronostratigraphy

3.1. Borehole and well logging techniques

Utilizing the Cretaceous time scale of Obradovich (1993), the Niobrara Formation has an estimated duration of about 6.2 ± 0.5 myr based on linear interpolation between Ar–Ar dates from altered volcanic ash beds (bentonites). Obradovich (1993) reports ages for 6 bentonites from biozones correlative to the Niobrara. Based on this time scale, a bulk average sedimentation rate (~ 1.4 cm/kyr) for the Niobrara Formation in the central part of the Denver basin can be established. However, the resolution of dated bentonites is not adequate to reconstruct high-resolution sedimentation rates throughout the entire formation. For instance, the timescale of Obradovich (1993) does not include enough dated bentonites to isolate durations/sedimentation rates of each informal member. The results of a more recent fit of Obradovich's (1993) radiometric ages for the Turonian–Coniacian, Coniacian–Santonian, and Santonian–Campanian boundaries by Ogg et al. (2004) provide an alternative geochronology. These radiometric chronologies will be discussed later in relation to the results from the orbital chronology of the Niobrara Formation.

The Libsack 43-27 and Aristocrat Angus 12-8 boreholes are located in the Wattenberg field on the west side of the Denver basin (northcentral Colorado; Fig. 1). The holes are spaced roughly 2 km apart and were drilled in 2003 by EnCana Oil and Gas (U.S.A.). High-resolution petrophysical logs and core splits were provided by EnCana for our study. Both boreholes penetrate the entire Niobrara Formation, which was cored with excellent recovery in both wells. To the best of our knowledge, these boreholes are the only complete cored records of the entire Niobrara Formation. Well-log correlations within the field and visual examination of the cores indicate that both of the wells contain conformable sequences of the Niobrara with no fault offset. The well-log suite includes Formation Micro-resistivity Images (FMIs), output from a micro-resistivity tool that resolves bedding features at centimeter scale (Fig. 2). The FMI sample spacing is 3 cm and 5 mm for the Libsack and Aristocrat Angus boreholes, respectively. The Niobrara is 84.5 m thick in both locations. Based on lithologic correlation and bentonite stratigraphy, offset between log depths and core depths

544

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

was readily determined and corrected for. Both cores exhibit typical Niobrara bedding rhythms as shown in Fig. 2A and B. Carbonate content was measured by coulometric titration (Engleman et al., 1985). The FMIs show patterns typical of Niobrara lithologic changes, at both the decameter (~ 10 m) and decimeter scale (~ 20 cm), with higher resistivity values corresponding to carbonate-rich intervals (or beds) and vice versa (Fig. 2). At the decimeter scale, higher-resolution analyses of weight percent calcium carbonate show correspondence with both resistivity changes and alternation of chalk and marl lithologies (Fig. 2B). A positive correlation exists between weight percent calcium carbonate and resistivity, although resistivity measurements increase dramatically in samples with carbonate content above 75%. Correlation coefficients improve slightly when the resistivity results are adjusted to a logarithmic scale against carbonate. Despite the lack of a linear relationship between resistivity and carbonate, the resistivity trends and positive relationship with both lithology and carbonate content indicate that the relative variance is related to, and records, chalk-marl oscillations of the primary depositional system. No relationship was observed between resistivity and total organic carbon content. 3.2. Spectral techniques The FMI data from the Libsack core are evenly spaced and have a Nyquist frequency of 0.17 cycles/cm. These FMI data are recorded from 9 sensors (same vertical height) on the logging tool. We used the average of these nine measurements to develop a time series. Based on Obradovich's (1993) interpolation of Ar–Ar dates to create a Cretaceous time scale for the Western Interior and the resulting average bulk sedimentation rate estimates for the Niobrara Formation, the Nyquist frequency is sufficient to resolve any Milankovitch periodicities present. The Niobrara time series was constructed from the FMI dataset, as were individual segments representing each informal member of the Smoky Hill Chalk, as well as the Fort Hays Member. Each segment was analyzed separately in order to reduce the likelihood of noise and distortion due to the possibility of large sedimentation rate changes at member boundaries and/or faciesrelated amplitude changes in the FMI. The segmented series were used as an aid in interpretation of results from spectral analyses of the entire interval. Although the Niobrara contains intervals of conspicuous rhythmic character, detection, discrimination, and quantification of potential orbital periodicities in the time series are challenging for three main reasons. First, time control in the Niobrara is somewhat poorly resolved. Published Turonian–Campanian bentonite dates are not from the Niobrara Formation itself and must be imported at biozone resolution from outside the study area (Montana and the Gulf Coast region). Because of this the direct conversion of spatial frequencies to temporal frequencies using two prominent, dated bentonites cannot be conducted as was done for the underlying Bridge Creek Limestone (Sageman et al., 1997). A second challenge involves discrimination of potentially competing orbital signals. Laferriere et al. (1987) noted variations in bundling ratios through time and space in the Fort Hays Limestone Member, an interval generally regarded to reflect eccentricity modulation of precession (Fischer, 1980). Deviations from expected bundling patterns may reflect obliquity influence. The final challenge relates to the length of the time series. The Niobrara alternates, at a decameter scale, between predominantly chalk and marl facies. Changes in rhythmic expression, and sedimentation rates, are associated with these facies changes and they can distort the time series. Our approach to resolving the Niobrara depositional rhythms employs a combination of spectral techniques. First, the multi-taper method (MTM; Thomson, 1982) is performed on the entire time series to establish statistically significant (F-test; Thomson, 1990) strati-

graphic frequencies and make preliminary interpretations of spectral peaks and average/modal sedimentation rate. The time series is then segmented into 8 intervals corresponding to the eight formal/informal Niobrara members and spectral analysis is performed on each segment. These shorter time series are generally less noisy and easier to interpret, due to greater sedimentation rate stability. The interpretations of the segments are useful for documenting a potential range of sedimentation rates through the entire formation. A third technique, metronomic AM/FM analysis (Herbert, 1994), is used to test the preliminary MTM frequency interpretations. Given the frequency ranges for eccentricity, obliquity, and precession determined by MTM, the time series is bandpassed (using a Gaussian filter in Analyseries; Paillard et al., 1996) for each suspected Milankovitch component. The filtered series are analyzed metronomically (by cycle number) which diminishes the effects of sedimentation rate changes (Herbert, 1994). Amplitude and frequency modulations of the metronomic series are subsequently analyzed by MTM (following the methods of Hinnov, 2000) to test for predicted modulations associated with each orbital parameter. This is essentially a quantitative treatment of bundling patterns where competing signals have been removed. Confirmed spectral peaks are then tracked stratigraphically by a moving window MTM (Evolutive Harmonic Analysis; Olsen and Kent, 1999; Meyers et al., 2001). Frequency changes of spectral peaks through the entire stratigraphic interval are compared to frequency results from the segmented MTMs to ensure consistency. The tracking profiles are also used to create a high-resolution, temporal series by assigning a constant periodicity to a single signal, which may exhibit variability in stratigraphic frequency. The new time series, corrected for sedimentation rate changes, is tuned to this particular periodicity so that the output is entirely in the time domain and, therefore, readily interpretable. These Fourier techniques were applied to the entire dataset, which also includes roughly 5 m of the lowermost Pierre Shale. These strata have greater carbonate content than typical Pierre Shale and are correlative to the uppermost Niobrara in Kansas (Hattin, 1982). All results reported herein are from analyses using a time-bandwidth product of two (3 data tapers; see Thomson, 1982). 4. Niobrara depositional frequencies 4.1. Multi-taper method Frequency analysis (MTM) of the entire Niobrara Formation yields noisy results, although statistically significant peaks (or clusters) are revealed; note particularly the robust power response at ~0.75 cycles/m (Fig. 3). Based on the Ar–Ar geochronology of Obradovich (1993), this spatial frequency is consistent with an eccentricity period (specifically a 95 kyr period with sedimentation rate of 1.4 cm/kyr). All sedimentation rate values discussed herein represent effective sedimentation rates (uncorrected for compaction). Dean and Arthur (1998) calculated a similar sedimentation rate (1.23 cm/kyr) for the Smoky Hill Chalk from the Coquina Oil, Berthoud State No. 4 well near Fort Collins, Colorado, based on thickness and the Obradovich (1993) time scale. Assuming this average sedimentation rate for the Niobrara Formation and the predicted Cretaceous orbital periodicities (Berger and Loutre, 1994; Hinnov, 2000; Laskar et al., 2004), the suspected stratigraphic frequency ranges for eccentricity, obliquity, and precession are indicated in Fig. 3. Peaks outside of the highlighted range may be noise, unrelated to orbital perturbations, or may represent spatial shifts in orbitally induced stratigraphic rhythms due to sedimentation rate changes (either faster or slower than the average of 1.4 cm/kyr). MTM analysis of individual members demonstrates that stratigraphic frequencies do indeed vary within the formation; however, results from some intervals appear just as noisy as the results from analysis of the entire sequence. This suggests the possibility of sedimentation

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

545

Fig. 3. Frequency results from the FMI dataset for the entire Niobrara Formation and lower 5 m of the Pierre Shale with a sample resolution of 3 cm. A) Results are from multi-taper method (MTM) harmonic analyses using three 2-pi data tapers. Power spectra (thin, grey line) and F-test (thick, black line) for statistical significance of frequencies are shown. The horizontal line indicates 95% significance. Bracketed regions indicate possible frequency range for eccentricity (E), obliquity (O), and precession (P) bands corresponding to radiometric-based estimates of bulk sedimentation rate (1.4 cm/kyr) of the Niobrara Formation in Wattenberg Field. B) Spectral results from metronomic assignment of upper envelopes for amplitude modulations of bandpassed frequencies of the suspected precession range for the Niobrara in the Libsack 43-27 dataset. Results are from analyses using three 2-pi data tapers. F-test spectra are shown with dark lines. Labeled peaks are in kyr. C) FM precession spectra with F-test results presented with dark lines. Labeled peaks are in kyr.

rate changes within some of the informal members. Another observation from the MTM results is that the marl members (generally thicker) are noisier and, therefore, more difficult to interpret due to lower carbonate content and decreased signal amplitude. The frequency analysis of marls also produces peaks with greater F-test values than chalks or limestones, likely due to a greater number of cycles from a thicker interval. Higher frequency F-test peaks, around 6 cycles/m, may represent semi-precession with a corresponding period of ~11.5 kyr or represent precession for intervals of decreased sedimentation rate (~ 1 cm/kyr). 4.2. Bandpassed frequency analysis Using the long-term average value of 1.4 cm/kyr as a reference, stratigraphic frequency distributions observed in the MTM analyses of Niobrara members suggest that actual sedimentation rates may vary from roughly 0.9 to 2.3 cm/kyr through the study interval. Based on these ranges of frequency variability in the eccentricity, obliquity, and precession signals, the time series was bandpassed for each orbital component. The bandpassed ranges are 0.5–1.0, 1.2–2.5, and 2.5–

6.0 cycles/m for eccentricity, obliquity, and precession, respectively. These ranges do not include the entire frequency spectrum for each orbital parameter, as concluded from member-level MTMs because filtering the entire range of each component would lead to slight overlap between orbital signals and complicate further analyses. The chosen frequency ranges are justifiable in that they include the entire frequency range for each component through the vast majority of the stratigraphic series (this likely does not hold for portions of the middle marl). Lower frequency bandpassing was also conducted for evaluation of potential long eccentricity (404 kyr) and “grand” eccentricity (~ 2400 kyr) cycles. The stratigraphic series of bandpassed frequencies for eccentricity, precession, and obliquity show typical bundling patterns throughout the sequence, even in the middle marl. The variability of amplitude and frequency from the precession-filtered series was quantified (using AM/FM analysis) and the cycle count for eccentricity, obliquity, and precession were also tallied (71, 176, and 345, respectively). Both AM and FM metronomic series for precession were analyzed with MTM. The results for the precession spectrum (Fig. 3B–C) quantitatively indicate significant bundling ratios in both the AM and FM

546

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

series. The most relevant precession series ratios indicate the presence of 400, 200, 125, 114, 85, 57, and 47 kyr periods, near predicted combination tones of Cretaceous precessional periods (18.6, 18.7, 21.9, 22.6, and 23.1 kyr). These periods were obtained by multiplying the ratios by an average precession period of 21.7 kyr; metronomic modulation tone periods are summarized in Hinnov (2000). A longer period modulation is also present with a period of ~ 1700 kyr. The bandpass results for long eccentricity and precession are plotted with the original time series for comparison and to ensure the validity of the bandpass (Fig. 4). The extracted long period signal (~ 1700 kyr) is also presented in Fig. 4A and this signal can also be compared to the original time series.

4.3. Evolutive harmonic analysis In order to quantify sedimentation rate changes in the Niobrara, Evolutive Harmonic Analysis (EHA) was performed on the time series. Sedimentation rate changes are expressed as stratigraphic shifts of significant frequencies through the series. EHA was performed on the entire series and was also performed for each member segment, since the amplitudes of chalk and marl intervals vary (see Fig. 2). The eccentricity range was analyzed with a 12-meter window and an incremental step of 10 cm. The same step was used for the obliquity range, but a 10-meter window was sufficient to track frequencies in this range. The precession range was explored and tracked in the

Fig. 4. Bandpassed frequencies for suspected orbital signals for the Niobrara Formation from the Libsack 43-27 FMI data series. A) Filtered frequencies for long period Niobrara rhythms at the scale of informal members. The number of 400 kyr cycles within each member-scale oscillation is tallied and the corresponding long period is calculated based on the average number of 400 kyr cycles contained within the oscillations. B) Filtered frequencies for 20 kyr cycles (grey) and 400 kyr cycles (black) with cycle counts for the 400 kyr oscillations. The 400 kyr cycles are correlated to the original times series (FMI) by dashed lines for peaks and solid lines for troughs.

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

Aristocrat Angus core, but the confidence of this result is lower due to difficulties related to the number of precessional signals present and the magnitude of spatial frequency shifts associated with sedimentation rate changes. In short, it is very difficult to identify and track a single precession component throughout the entire series. Identification and tracking of statistically significant frequencies on EHA diagrams is not always straightforward. Although Meyers et al. (2001) were able to make confident interpretations for a 12-m thick time series from the underlying Bridge Creek Limestone Member of the Greenhorn Formation, longer intervals with significant changes in facies and sedimentation rates, like the Niobrara, are more difficult to evaluate. The specific problems encountered in our study include: 1) changes in amplitude/sensitivity among facies; 2) stratigraphic changes in the statistical significance of frequencies; and 3) the effects of rapid sedimentation rate changes that cause smearing at stratigraphic horizons within evolutive spectrograms. These limitations were overcome by utilizing all available cyclostratigraphic results as aids to frequency identification and tracking on EHA diagrams. For example, the results for bandpassed frequency analysis were used as rough estimates for frequency changes through the series. Because the duration between peaks (or troughs) of a known Milankovitch component should be constant, variability in depth between peaks can be used to assess changes in frequency (in cycles/meter) through the series. This was performed for the filtered 400, 100, and 40 kyr periods. The results are represented by the thin, blue curves overlaying the EHA diagrams in Fig. 5. The MTM results of individual members were also called upon to constrain frequency tracking, as

547

were segmented EHA results. The interpretation of one Milankovitch component was aided by evaluation of another. For instance, confident tracking of the 95 kyr period in the lower 20 m of Fig. 5A could only be made by observation of stratigraphically correlative frequency changes in the obliquity range (Fig. 5B–C). Finally, we conducted several tracking exercises for both the eccentricity and obliquity frequency ranges. These tracking models were performed to assess different interpretations of frequency shifts in the EHA diagrams, specifically targeting the areas of lowest confidence. The results of each tracking model were then assigned a given period (either 95 or 40 kyr) and the spatial frequencies were converted to temporal frequencies by accounting for sedimentation rates. The tuned series for each model was analyzed by MTM and comparisons were made to determine which models produced dubious results. Our discrimination was based on the amplitude and statistical significance of frequencies independent of the frequency that was tracked. The resulting frequency tracking is designated by the bold lines in Fig. 5 and the 95 kyr tuned spectra are shown in Fig. 6. With a probability threshold set at 0.8 for frequency tracking, the spectra yield statistically significant peaks at 696, 370, 236, 140, 120, 95, 71, 57, 50, 44, 38, 31, 25.7, 23.5, 21.3, 19.6, 18.3, and 16.9 kyr periods. Intervals with probabilities less than 0.8 were linearly interpolated; this could introduce a minor amount of error. Because the 12-meter moving window EHA technique results in the absence of data at the bottom and top of the series, there are no results for the lower 6 m of the Fort Hays Limestone or the uppermost 1 m of the Smoky Hill Chalk (~ 5 m of Pierre Shale was used in the time series). Tracking of these

Fig. 5. Results of evolutive harmonic analyses (EHA) through hypothesized eccentricity and obliquity frequency ranges of the Niobrara Formation. Abbreviations for informal members are the same as in Fig. 2A. Eccentricity and obliquity results are from a 12 and 10 m moving window, respectively. Identification and tracking of orbital frequencies are based on the synthesis of MTM, EHA, and bandpassed signal analyses. Bandpass overlays (thin, blue lines) are used as a guide for signal tracking and are computed by assigning a temporal period (i.e., 100 and 40 kyr) to a cycle of measured thickness to produce a frequency. A) Probability of frequencies in the eccentricity ranges with a bandpassed frequency overlay (blue) and tracked 95 kyr component (black). B) Probabilities associated with obliquity range. C) Amplitude in the obliquity range.

548

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

Fig. 6. Spectra tuned to frequencies of the 95 kyr period. A) 95 kyr tuned spectra for the Niobrara Formation. F-test spectra are shown with thick, black lines. The horizontal line represents 95% confidence level. Labeled peaks are in kyr (E = eccentricity; O = obliquity; P = precession). B) Results of MTM analysis for the eccentricity tuned upper envelope of the amplitude modulation series. Power spectra of this tuned series show less noise than the untuned modulation series and validates the tuning method.

intervals was performed on short segments with a small window size (1.5 m) and stitched to the rest of the sequence to create a more complete composite. The precession range was bandpassed from the eccentricity tuned time series and MTM was performed on the amplitude modulation series (Grippo et al., 2004). This analysis was conducted as an additional test of the tuning procedure (Fig. 6B). The results indicate modulation tones of 1667, 439, 244, 167, 130, 103, and 93 kyr. All of the results from the Libsack FMI reported above are consistent with analyses of the FMI from the neighboring Aristocrat Angus borehole. The sedimentation rate curve derived from the eccentricity tuned series is compared to sedimentation rates estimated from interpretations of tracked obliquity frequencies and the eccentricity bandpass (Fig. 7; all procedures are tallied in Table 1). The sedimentation rate histories from the various methods are in general agreement, but differ slightly. The range of effective sedimentation rate falls between 0.75 and 2.35 cm/kyr, which is not an unexpected result given the geochronological constraints. The sedimentation rate history derived from the eccentricity bandpass is of higher resolution than the EHAbased sedimentation rate reconstructions because a long window size (10–12 m) was needed to confidently track signals via EHA and thus the result is significantly time averaged. In general, sedimentation rate was greater during deposition of marls, at least at the decameter scale, as indicated by all three sedimentation rate approaches. Sediment

accumulation was fastest (and more variable) in the middle marl member, which likely explains the relative methodological difficulties associated with that unit. 5. Durations of the Niobrara Formation, Coniacian Stage, and Santonian Stage The durations of each informal member and the cumulative duration of the formation, based on application of the suite of cyclostratigraphic methods described above, are listed in Table 1. The EHA-based durations are calculated by conversion of sedimentation rate with thickness to time. Summation of time is tallied from the thickness of each member. The durations of the Coniacian and Santonian Stages are derived by the same method, although the stratigraphic intervals are biostratigraphically chosen. The results for durations of informal Niobrara members are generally consistent among the various methods/models for the lower 40 m of the formation. Results show more variance in durations for the middle marl, middle chalk, and upper marl units. Our calculations for duration of the Niobrara Formation using different cyclostratigraphic methods range from 6.1 to 6.7 myr, which is within the range of error (~ 6.2 ± 0.5 myr) for the Ar–Ar interpolation method of Obradovich (1993). The biostratigraphic control of dated bentonites also allows for comparison of radiometric and orbital durations of shorter intervals in the Niobrara (Fig. 8). For instance,

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

549

Fig. 7. Sedimentation rate history of the Niobrara Formation in the context of informal member lithology and the original FMI time series. The sedimentation rate history on the left was determined by bandpass frequency analysis of short eccentricity (100 kyr). The horizontal dashed lines correlate intervals of increased sedimentation rate to more conductive (carbonate-poor) intervals. The two sedimentation rates on the right (bold tracks obliquity and thin tracks eccentricity) were calculated by evolutive analysis (Fig. 5) and are more time averaged than the bandpassed analysis of sedimentation rate history.

radiometric data bracket the duration of the lower marl member at 1.42 ± 0.99 myr. The orbital duration of this interval is 0.85 to 1.12 myr (Table 1). This indicates that determination of time between dated bentonites with the Ar–Ar method is better suited for long intervals

(e.g., the entire Niobrara Formation) and that shorter intervals, like the lower marl, are difficult to resolve using radioisotopic constraints. The spectral analytical results indicate that construction of an orbital timescale provides better constraint at high resolution. This

Table 1 Lithostratigraphic and stage boundaries in the Niobrara Fm. from the Libsack core Durations (kyr) Libsack 43-27 Libsack Unit

m Top

Bottom Thickness

FH

2149.0

2157.6

8.6

LCS LM LC MM

2139.3 2127.0 2117.1 2103.8

2149.0 2139.3 2127.0 2117.1

9.7 12.2 9.9 13.3

MC UM UC

2094.0 2081.2 2072.9

2103.8 2094.0 2081.2

9.8 12.8 8.3

Obliquity tracked continuous

Eccentricity tracked continuous

Stage boundaries (m)

Obliquity MTM Eccentricity 400 kyr tracked bandpass segments bandpass cycle counts cycle counts segmented

2156 Turonian– Coniacian

716

738

700

Incomplete Incomplete Incomplete

745 906 759 886

727 1036 834 1066

747 909 808 888

784 917 749 885

804 903 782 795

779 874 717 734

701 899 684

698 1115 680

746 870 751

645 937 575

632 972 675

738 5947 6685

700 5845 6545

~ 725 5701 6426

~ 725 5440 6165

~ 725 5383 6108

3568 2512 6080

3342 2608 5950

3411 2375 5786

3345 2215 5560

3226 2205 5431

2112.5 Coniacian– Santonian

Total FH Mbr Total SH Mbr Total Niobrara

8.6 75.9 84.5

722 851 ~ 2078.5 Santonian– 754 Campanian 716 5624 6340

Coniacian Santonian Both stages

43.5 34.0 77.5

3300 2391 5691

Average stage durations were calculated based on results from various methods.

Angus 12-8 Precession tracked segments 3583 2390 5973

Average S.D. 3396 2385 5781

134 146 236

550

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

Fig. 8. Lithostratigraphy, chronostratigraphy, and biostratigraphy of the Niobrara Formation. A) Lithology of the Libsack 43-27 core with upper Turonian to lower Campanian Niobrara formal/informal lithostratigraphy, biostratigraphic boundaries (Walaszczyk and Cobban, 2007), imported bentonite stratigraphy (Obradovich, 1993), and a cumulative orbital timescale with 400 kyr cycle counts compared to occurrences of age diagnostic taxa for central Colorado (Scott and Cobban, 1964; Bralower and Bergen, 1998) and Kansas (Hattin, 1982; Watkins et al., 1993). B) Age dates for North American bentonites shown in A) (Obradovich, 1993) with re-calibrated estimates from Ogg et al. (2004). C) Ammonite zonation (Cobban, 1993; Leckie et al., 1997) of the Western Interior. Numbered biozones are plotted to the right in A) for a compilation of imported biostratigraphy to the Libsack core, following lithostratigraphic correlations. The synthesis of Niobrara biostratigraphic zonation is much like the original rendition of Scott and Cobban (1964) with respect to stage boundaries, but with differences to the substage placements.

improvement to geochronologic resolution facilitates calculations of accumulation rates for sedimentary components (e.g., organic carbon and carbonate), which has particular significance for investigations of the enigmatic OAE III interval. The accumulation rates will also help to evaluate debated paleoceanographic mechanisms responsible for limestone–shale rhythms in cyclostratigraphically unique facies of the Fort Hays Limestone Member of the Niobrara Formation. The duration of the Coniacian Stage ranges from 3.26 to 3.50 myr, based on the Niobrara orbital record (Table 1). This result is much longer than the 2.4 myr duration reported by Obradovich (1993) and similar to those reported by Palmer and Geissman (1999) and Remane (2000) (Table 2). A more recent compilation (Ogg et al., 2004) places the Coniacian at 3.5 ± 0.3 myr; the error range overlaps our estimate from the orbital record. A Coniacian duration of ~ 3.2 myr is also supported by the presence of eight suspected ~ 400 kyr period oscillations in chalks of the Anglo-Paris basin (Grant et al., 1999). The Santonian Stage duration ranges from 2.24 to 2.53 myr as determined from the orbital timescale. These results are shorter than the Ar–Ar based duration of 2.8 myr (Obradovich, 1993). Our results for the

Santonian are similar to more recent studies (e.g., Ogg et al., 2004), which cite values of 2.3 ± 0.1 myr. The combined duration of the Coniacian and Santonian is 5.55 to 6.02 myr (Tables 1 and 2). Our estimates for the durations of the Coniacian and Santonian may contain some error associated with the correlation of biostratigraphic zones to our time series. However, such error is thought to be minimal in the context of gross sedimentation rates and thicknesses of the Niobrara members. A recent study of geochemical oscillations in Coniacian–Campanian claystones of the tropical Atlantic yielded spectral results indicating the presence of sixty-five 22 kyr cycles and a duration of 1.46 myr (Beckmann et al., 2005). This duration for the Coniacian, Santonian, and early Campanian seems anomalously short in comparison to both the orbital time scale-based durations, presented here, as well as those based on radiometric techniques. One possible explanation for this discrepancy involves the presence of hiatus in the tropical Atlantic. Another possibility is that the tropical Atlantic orbital signal has been misinterpreted. If the 65 cycles in this record had a periodicity of ~ 100 kyr (eccentricity), then the duration of the study

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

551

interval would be roughly 6.5 myr, which is consistent with our results and those based on radiometry.

diachronous (this unit thins to only a couple meters in western Kansas; Amoco Rebecca Bounds #1 core). The uppermost bed of the lower chalk and shale member may be diachronous, even at distances of 25 km, and therefore, the Niobrara Formation may be entirely conformable in the Pueblo, Colorado area. The cyclostratigraphic analyses of the current study do not support an unconformity at this stratigraphic level in the northern Colorado area. In addition, correlation of bentonite markers along the Front Range of Colorado also suggests conformity through the lower chalk and shale (LCS) and lower marl section at Pueblo, Colorado and in the USGS #1 Portland core. Advanced spectral techniques can be applied to a wide range of time series. This study demonstrates the applicability of these techniques to high-resolution petrophysical records. Not all FMIs (or time series in general) record orbitally influenced depositional rhythms, but rhythmic intervals with compositional oscillations in relatively conductive and resistive sediments, and with constrained porosities, may be suitable for such analyses. The advantages of this approach include continuous, high-resolution time series, availability of datasets, and sufficient sensitivity to primary depositional signal. Given the number of these records already available, analyses of FMI datasets through rhythmically bedded strata are relatively cost effective in comparison to geochemical time series and require much less time to prepare than other continuous datasets, such as (pixel) grayscale records. Further, the FMI is more likely to have better continuity than even the most well-recovered core intervals because it is a measurement of the borehole wall.

6. Discussion

7. Conclusions

Spectral analytical results of decimeter-scale rhythmic oscillations of chalk and marl deposition in the Niobrara Formation indicate the presence of orbital frequencies related to eccentricity, obliquity, and precession. The results also indicate variability in sensitivity to Milankovitch-scale depositional control between decameter-thick chalk and marl facies. These facies changes make tracking of individual orbital frequencies difficult. Interpretation of the complete time series is significantly aided by segmentation of the time series and bandpass filtering techniques. Tracking of frequencies related to a single orbital periodicity allows for the reconstruction of the sedimentation rate history and generation of an orbital timescale (shown in Fig. 8). Comparison of sedimentation rates for chalk and marl facies suggests a dilution-driven depositional mechanism (increased accumulation rate of siliciclastics during marl deposition) at the decameter scale. In light of noisier spectral results for marl intervals and the correlation of Santonian sandy facies to the west and south, siliciclastic dilution may be influencing the signal strength and sensitivity to orbital forcing. Although the temporal scale of change between chalk and marl submembers is roughly 1.7 myr, interpretation of this period as a long modulation of precession is not consistent with the predicted 2.3 myr cycle. Nevertheless, similar periods have been noted from other long records of orbital cyclicity in the Mesozoic (Olsen and Kent, 1999; Grippo et al., 2004) and therefore, the decameter rhythms of the Niobrara may be orbitally forced. An alternative explanation for these longer Niobrara ‘rhythms’ invokes changes in accommodation and/or sediment supply, presumably related to tectonic processes (as mentioned by Dean and Arthur, 1998). Walaszczyk and Cobban (2007) interpreted a stratigraphic gap in the uppermost lower chalk and shale (LCS) informal member of the Smoky Hill Chalk based on the co-occurrence of Volviceramus involutus and Cremnoceramus deformis in the uppermost limestone bed of this informal unit near Pueblo, Colorado. This apparent gap includes the “Inoceramus gibbosus” zone of uppermost lower Coniacian in northern Europe. As noted by Walaszczyk and Cobban (2007), these taxa were not found in the same section (Scott and Cobban, 1964) and the localities are roughly 25 km apart. This is a significant point because the base of the lower chalk and shale informal member is diachronous (Hattin, 1982) and the top of this informal member is also likely

The Niobrara Formation exhibits depositional rhythmicity from the upper Turonian to lower Campanian. The interval is free of hiatus based on our lithologic observations and time-frequency analysis and records orbital forcing of sedimentation related to eccentricity, obliquity, and precession. Tracking of orbital signals through the entire sequence permits generation of a sedimentation rate history and a high-resolution orbital timescale that facilitates comparison of stage durations to estimates from radiometry. The sedimentation rate profile for the Niobrara indicates faster rates of deposition during decameter-thick marl intervals and slower rates during decameter-thick chalk intervals, thereby suggesting oscillations in siliciclastic flux as a controlling mechanism for these rhythms. The orbital-based durations for the Coniacian and Santonian are consistent with results from radiometric dating, but are at significantly higher resolution. With respect to long time series exhibiting facies changes, segmentation of the time series aids frequency tracking techniques for calculation of sedimentation rate histories. Segmented series also facilitate the use of smaller moving window sizes in EHA analysis, due to the shortened series length, permitting sedimentation rate calculations that are less time averaged. Bandpassing of frequencies for suspected orbital periods provides an additional aid to signal tracking techniques in EHA analysis and provides a higher-resolution sedimentation rate history than EHA-based reconstructions in the Niobrara Formation. Tracking of frequencies related to eccentricity, obliquity, and precession yield similar results in terms of both sedimentation rate histories and durations. By tuning a long time series to the temporally stable periodicity of eccentricity, a calibrated record of Turonian–Campanian obliquity and precession periodicities is generated through an interval with high cycle counts. These periodicities, especially when compared to results from other long records in the Cretaceous (e.g., Grippo et al., 2004), may help document and constrain long-term changes in the Earth's precession rate (Hinnov, 2000). Generation of an orbital timescale in the Niobrara Formation creates a temporal framework that can be exported to other biostratigraphically constrained localities in the Western Interior basin and elsewhere. The orbital timescale also facilitates calculation of accumulation rates for geochemical components, such as organic carbon, that will aid in the

Table 2 Ages and durations of the Coniacian–Santonian Radiometry

Orbital

Obradovich Palmer and Remane (2000) (1993) Geissman (1999) Boundary ages (Ma) Turonian– 88.7 Coniacian Coniacian– 86.3 Santonian Santonian– 83.5 Campanian Error (my) 0.5 Durations (my) Coniacian 2.4 Santonian 2.8 Coniacian and 5.2 Santonian Uncertainty (my) Coniacian Santonian

Niobrara Ogg et al. Beckmann (2004) et al. (2005) Fm. this study Equatorial Atlantic

89

89

89.3

85.8

85.8

85.8

83.5

83.5

83.5

1

0.5

0.7–1.0

3.2 2.3 5.5

3.2 2.3 5.5

3.5 2.3 5.8

b 1.46

3.4 2.39 5.79

3.2–3.8 2.2–2.4

552

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553

evaluation of processes during the comparatively understudied Oceanic Anoxic Event III (Arthur and Schlanger, 1979). Acknowledgements EnCana Oil and Gas, Inc., (U.S.A.) provided the materials for this study. Financial support was provided by grant 42588-AC8 of the Petroleum Research Fund of the American Chemical Society (B.B.S). The Geological Society of America provided financial assistance to R.E. L. E. Gustason is thanked for all his efforts related to the logistics of this project and for numerous discussions of Western Interior stratigraphy. S. Meyers is thanked for his help with the spectral methods employed in this study. J. Ogg provided helpful suggestions on an earlier version of the manuscript. R.M. Leckie and one anonymous reviewer provided helpful comments. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.epsl.2008.03.021. References Anderson, R.Y., 1984. Orbital forcing of evaporite sedimentation. In: Berger, A.L., Imbrie, J., Hays, J., Kukla, G., Saltzman, B. (Eds.), Milankovitch and Climate. Reidel Publ., Dodrecht, pp. 147–162. Arthur, M.A., Dean, W.E., 1991. An holistic geochemical approach to cyclomania: examples from Cretaceous pelagic limestone sequences. In: Einsele, G., Ricken, W., Seilacher, A. (Eds.), Cycles and Events in Stratigraphy. Springer-Verlag, Berlin, pp. 126–166. Arthur, M.A., Schlanger, S.O., 1979. Cretaceous “Oceanic Anoxic Events” as causal factors in development of reef-reservoired giant oil fields. AAPG Bull. 63, 870–885. Arthur, M.A., Dean, W.E., Bottjer, D., Scholle, P.A., 1984. Rhythmic bedding in MesozoicCenozoic pelagic carbonate sequences; the primary and diagenetic origin of Milankovitch-like cycles. In: Berger, A., Imbrie, J., Hays, J.D., Kukla, G., Saltzman, B. (Eds.), Milankovitch and Climate, pt. 1. Reidel Pub. Co., Amsterdam, pp. 191–222. Arthur, M.A., Dean, W.E., Pollastro, R., Scholle, P.A., Claypool, G.E., 1985. A comparative geochemical study of two transgressive pelagic limestone units, Cretaceous Western Interior basin, U.S. In: Pratt, L.M., Kauffman, E.G., Zelt, F.B. (Eds.), FineGrained Deposits and Biofacies of the Cretaceous Western Interior Seaway: Evidence of Cyclic Sedimentary Processes. SEPM Field Trip Guidebook, vol. 4, pp. 16–27. Barlow, L.K., Kauffman, E.G., 1985. Depositional cycles in the Niobrara Formation, Colorado Front Range. In: Pratt, L.M., Kauffman, E.G., Zelt, F.B. (Eds.), Fine-Grained Deposits and Biofacies of the Cretaceous Western Interior Seaway: Evidence of Cyclic Sedimentary Processes. SEPM Field Trip Guidebook, vol. 4, pp. 199–208. Beckmann, B., Flögel, S., Hofmann, P., Schulz, M., Wagner, T., 2005. Orbital forcing of Cretaceous river discharge in tropical Africa and ocean response. Nature 437, 241–244. Berger, A., Loutre, M.F., 1994. Astronomical forcing through geological time. In: de Boer, P.L., Smith, D.G. (Eds.), Orbital Forcing and Cyclic Sequences, International Association of Sedimentologists. Special Publication No. 19, pp. 15–24. Bralower, T.J., Bergen, J.A., 1998. Cenomanian–Santonian calcareous nannofossil biostratigraphy of a transect of cores drilled across the Western interior seaway. In: Dean, W.E., Arthur, M.A. (Eds.), Stratigraphy and Paleoenvironments of the Cretaceous Western Interior Seaway, U.S.A: SEPM. Concepts Sedimentol. Paleontol., vol. 6, pp. 59–77. Cobban, W.A., 1993. Diversity and distribution of Late Cretaceous ammonites, Western Interior, U.S. In: Caldwell, W.G.E., Kauffman, E.G. (Eds.), Evolution of the Western Interior Basin. Spec. Pap.-Geol. Assoc. Can., vol. 39, pp. 435–452. D'Argenio, B., Amodio, S., Ferreri, F., Pelosi, N., 1997. Hierarchy of high-frequency orbital cycles in Cretaceous carbonate platform strata. Sediment. Geol. 113, 169–193. Dean, W.E., Arthur, M.A., 1998. Geochemical expression of cyclicity in Cretaceous pelagic limestone sequences: Niobrara Formation, Western Interior Seaway. In: Dean, W.E., Arthur, M.A. (Eds.), Stratigraphy and Paleoenvironments of the Cretaceous Western Interior Seaway, U.S.A: SEPM. Concepts Sedimentol. Paleontol., vol. 6, pp. 227–255. De Boer, P.L., Smith, D.G., 1994. Orbital forcing and cyclic sequences. Spec. Publ. Int. Assoc. Sedimentol. 19, 1–14. Engleman, E.E., Jackson, L.L., Norton, D.R., Fischer, A.G., 1985. Determination of carbonate carbon in geological materials by coulometric titration. Chem. Geol. 53, 125–128. Fischer, A.G., 1980. Gilbert-bedding rhythms and geochronology. In: Yochelson, E.I. (Ed.), The Scientific Ideas of G.K. Gilbert. Spec. Pap.-Geol. Soc. Am., vol. 183, pp. 93–104. Fischer, A.G., Bottjer, D.J., 1991. Orbital forcing and sedimentary sequences. J. Sediment. Petrol. 61, 1063–1069. Fischer, A.G., Herbert, T., Premoli-Silva, I., 1985. Carbonate bedding cycles in Cretaceous pelagic and hemipelagic sediments. In: Pratt, L.M., Kauffman, E.G., Zelt, F.B. (Eds.), Fine-Grained Deposits and Biofacies of the Cretaceous Western Interior Seaway: Evidence of Cyclic Sedimentary Processes. SEPM Field Trip Guidebook, vol. 4, pp. 1–10.

Gilbert, G.K., 1895. Sedimentary measurement of geologic time. Geology 3, 121–127. Grant, S.F., Coe, A.L., Armstrong, H.A., 1999. Sequence stratigraphy of the Coniacian succession of the Anglo-Paris Basin. Geol. Mag. 136, 17–38. Grippo, A., Fischer, A.G., Hinnov, L.A., Herbert, T.M., Premoli Silva, I., 2004. Cyclostratigraphy and chronology of the Aptian (Piobbico Core, Italy). In: D'Argenio, B., Fischer, A.G., Premoli Silva, I., Weissert, H., Ferreri, V. (Eds.), Cyclostratigraphy: Approaches and Case Histories. SEPM Spec. Publ., vol. 81, pp. 57–82. Tulsa, OK. Hattin, D.E., 1981. Petrology of Smoky Hill Member, Niobrara Chalk (Upper Cretaceous), in type area, western Kansas. AAPG Bull. 65, 831–849. Hattin, D.E., 1982. Stratigraphy and depositional environment of Smoky Hill Chalk Member, Niobrara Chalk (Upper Cretaceous) of the type area, western Kansas. Bulletin-Kansas State Geological Survey, vol. 225. 108 pp. Hays, J.D., Imbrie, J., Shackleton, N.J., 1976. Variations in the Earth's orbit: pacemaker of the ice ages. Science 194, 1121–1132. Herbert, T.D., 1994. Reading orbital signals distorted by sedimentation: models and examples, in: DeBoer, P.L., Smith, D.G., (Eds.), Orbital forcing and Cyclic Sequences. Spec. Publ. Int. Assoc. Sedimentol. 19, 483–508. Herbert, T.D., Fischer, A.G., 1986. Milankovitch climatic origin of mid-Cretaceous black shale rhythms in central Italy. Nature 321, 739–743. Hinnov, L.A., 2000. New perspectives on orbitally forced stratigraphy. Annu. Rev. Earth Planet. Sci. 28, 419–475. House, M.R., 1995. Orbital forcing timescales: an introduction. In: House, M.R., Gale, A.S. (Eds.), Orbital Forcing Timescales and Cyclostratigraphy. Geol. Soc. Spec. Publ., vol. 85, pp. 1–18. Jeletzky, J.A., 1968. Macrofossil zones of the marine Cretaceous of the Western Interior of Canada and their correlation with the zones and stages of Europe and the Western Interior of the United States. Geol. Surv. Can., Paper, vol. 62–72. 66 pp. Kauffman, E.G., 1967. Coloradan macroinvertebrate assemblages, central Western Interior, United States. Paleoenvironments of the Cretaceous Seaway — A symposium. Publ. Colorado School of Mines, pp. 67–143. Kauffman, E.G., 1969. Cretaceous marine cycles of the Western Interior. Mt. Geol. 6, 227–245. Kauffman, E.G., 1975. Dispersal and biostratigraphic potential of Cretaceous benthonic Bivalvia in the Western Interior. Cretaceous Systems in the Western Interior of North America. Spec. Pap.-Geol. Assoc. Can., vol. 13, pp. 163–194. Kauffman, E.G., 1977a. Geological and biological overview: Western Interior Cretaceous Basin. In: Kauffman, E.G. (Ed.), Cretaceous facies, faunas and paleoenvironments across the Western Interior Basin. Mt. Geol., vol. 14, pp. 75–99. Kauffman, E.G., 1977b. Upper Cretaceous cyclothems, biotas, and environments, Rock Canyon Anticline, Pueblo, Colorado. In: Kauffman, E.G. (Ed.), Cretaceous Facies, Faunas, and Paleoenvironments Across the Western Interior Basin. Mt. Geol., vol. 14, pp. 129–152. Kauffman, E.G., Caldwell, W.G.E., 1993. The Western Interior Basin in space and time. In: Caldwell, W.E., Kauffman, E.G. (Eds.), Evolution of the Western Interior Basin. Spec. Pap.-Geol. Assoc. Can., vol. 39, pp. 1–30. Kauffman, E.G., Sageman, B.B., Elder, W.P., Kirkland, J.I., Villamil, T., 1993. Cretaceous molluscan biostratigraphy and biogeography, Western Interior Basin, North America. In: Caldwell, W.G.E., Kauffman, E.G. (Eds.), Evolution of the Western Interior Basin. Spec. Pap.-Geol. Assoc. Can., vol. 39, pp. 397–434. Kennedy, W.J., 1984. Ammonite faunas and the “standard zones” of the Cenomanian to Maastrichtian stages in their type areas, with some proposals for the definition of these stages boundaries by ammonites. Bull. Geol. Soc. Den. 33, 147–161. Kennedy, W.J., Cobban, W.A., 1991. Stratigraphy and interregional correlation of the Cenomanian–Turonian transition in the Western Interior of the United States near Pueblo, Colorado, a potential boundary stratotype for the base of the Turonian stage. Newsl. Stratigr. 24, 1–33. Laferriere, A.P., Hattin, D.E., Archer, A.W., 1987. Effects of climate, tectonics and sea-level changes on rhythmic bedding patterns in the Niobrara Formation (Upper Cretaceous), U.S. Western Interior. Geology 15, 233–236. Laskar, J., Robutel, P., Joutel, F., Gastineau, M., Correia, A.C.M., Levrard, B., 2004. A longterm numerical solution for the insolation quantities of the Earth. Astron. Astrophys. 428, 261–285. Leckie, R.M., Kirkland, J.I., Elder, W.P., 1997. Stratigraphic framework and correlation of a principal reference section of the Mancos Shale (Upper Cretaceous), Mesa Verde, Colorado: New Mexico Geological Society Guidebook. 48th Field Conference, Mesozoic Geology and Paleontology of the Four Corners Region, pp. 163–216. Meyers, S.R., Sageman, B.B., Hinnov, L.A., 2001. Integrated quantitative stratigraphy of Cenomanian–Turonian Bridge Creek Limestone Member using evolutive harmonic analysis and stratigraphic modeling. J. Sediment. Res. 71, 628–644. Obradovich, J., 1993. A Cretaceous time scale, in: Caldwell, W.G.E., Kauffman, E.G., (Eds.), Evolution of the Western Interior Basin. Spec. Pap.-Geol. Assoc. Can. 39, 379–396. Obradovich, J.D., Cobban, W.A., 1975. A time-scale for the Late Cretaceous of the Western Interior of North America. In: Caldwell, W.G.E. (Ed.), The Cretaceous System in the Western Interior of North America. Spec. Pap.-Geol. Assoc. Can., vol. 13, pp. 31–54. Ogg, J.G., Agterberg, F.P., Gradstein, F.M., 2004. The Cretaceous Period. In: Gradstein, F.M., Ogg, J.G., Smith, A.G. (Eds.), A Geologic Time Scale. Cambridge University Press, New York, pp. 344–383. Olsen, P.E., Kent, D., 1999. Long-period Milankovitch cycles from the Late Triassic and Early Jurassic of eastern North America and their implications for the calibration of the early Mesozoic time-scale and the long-term behaviour of the planets, in: Shackleton, N.J., McCave, I.N., Weedon, G.P., (Eds.), Astronomical (Milankovitch) calibration of the geological time-scale, Philosophical Transactions — Royal Society. Math., Phys. Eng Sci. 357, 1761–1786. Paillard, D., Labeyrie, L., Yiou, P., 1996. MacIntosh program performs time series analysis. EOS Trans., AGU 77, 379.

R.E. Locklair, B.B. Sageman / Earth and Planetary Science Letters 269 (2008) 540-553 Palmer, A.R., Geissman, J., 1999. 1999 Geologic Time Scale. Geological Society of America. CTS004. Park, J., Herbert, T.D., 1987. Hunting for paleoclimatic periodicities in a geologic time series with an uncertain time scale. J. Geophysi. Res. 92, 14,027–14,040. Park, J., D'Hondt, S.L., King, J.W., Gibson, C., 1993. Late Cretaceous precessional cycles in double time: a warm-Earth Milankovitch response. Science 261, 1431–1434. Pratt, L.M., Arthur, M.A., Dean, W.E., Scholle, P.A., 1993. Paleoceanographic cycles and events during the Late Cretaceous in the Western Interior Seaway of North America. In: Caldwell, W.G.E., Kauffman, E.G. (Eds.), Evolution of the Western Interior Basin. Spec. Pap.-Geol. Assoc. Can., vol. 39, pp. 333–354. Preto, N., Hinnov, L.A., De Zanche, V., Mietto, P., Hardie, L.A., 2004. The Milankovitch interpretation of the Latemar platform cycles (dolomites, Italy): implications for geochronology, biostratigraphy, and Middle Triassic carbonate accumulation. In: D'Argenio, B., Fischer, A.G., Premoli-Silva, I., Weissert, H., Ferreri, V. (Eds.), Cyclostratigraphy: Approaches and Case Histories. SEPM Spec. Publ., vol. 81, pp. 167–182. Tulsa, OK. Remane, J., 2000. Explanatory Note to the International Stratigraphic Chart. International Union of Geological Sciences, Trondheim, Norway. 17 pp. Roberts, L.N.R., Kirschbaum, M.A., 1995. Paleogeography of the Late Cretaceous of the Western Interior of Middle North America — Coal Distribution and Sediment Accumulation. U. S. Geol. Surv., Prof. Pap., vol. 1561. 115 pp. Sageman, B.B., Rich, J., Arthur, M.A., Birchfield, G.E., Dean, W.E., 1997. Evidence For Milankovitch periodicities in Cenomanian–Turonian lithologic and geochemical cycles, Western Interior U.S. J. Sediment. Res. 67, 285–301. Savrda, C.E., 1998. Ichnocoenoses in the Niobrara Formation: implications for benthic oxygenation histories. In: Dean, W.E., Arthur, M.A. (Eds.), Stratigraphy and Paleoenvironments of the Cretaceous Western Interior Seaway, U.S.A. SEPM, Concepts Sedimentol. Paleontol., vol. 6, pp. 137–151. Scott, G.R., Cobban, W.A., 1962. Clioscaphites saxitonianus (McLearn), a discrete ammonite zone in the Niobrara Formation at Pueblo, Colorado; Article 90. U.S. Geological Survey Professional Paper, Report: P 0450-C, p. C85.

553

Scott, G.R., Cobban, W.A., 1964. Stratigraphy of the Niobrara Formation at Pueblo, Colorado. U.S. Geological Survey Professional Paper, Report: P 0454-L, pp. L1–L30. Scott, G.R., Cobban, W.A., Merewether, E.A., 1986. Stratigraphy of the Upper Cretaceous Niobrara Formation in the Raton basin, New Mexico. New Mexico Bureau of Mines and Mineral Resources, Bulletin, vol. 115. 34 pp. Thomson, D.J., 1982. Spectrum estimation and harmonic analysis. IEEE Proceedings 70, 1055–1096. Thomson, D.J., 1990. Quadratic-inverse spectrum estimates: applications to paleoclimatology. Philos. Trans. -R. Soc. Lond. A 332, 539–597. Walaszczyk, I., Cobban, W.A., 1998. The Turonian–Coniacian boundary in the United States Western Interior. Acta Geol. Pol. 48, 495–507. Walaszczyk, I., Cobban, W.A., 2000. Inoceramid Faunas and Biostratigraphy of the Upper Turonian–Lower Coniacian of the Western Interior of the United States. Spec. Pap. Palaeontol., vol. 64. The Palaeontological Association of London. 117 pp. Walaszczyk, I., Cobban, W.A., 2007. Inoceramid fauna and biostratigraphy of the upper Middle Coniacian–lower Middle Santonian of the Pueblo Section (SE Colorado, US Western Interior). Cretac. Res. 28, 132–142. Watkins, D.K., Bralower, T.J., Covington, J.M., Fisher, C.G., 1993. Biostratigraphy and paleoecology of the Upper Cretaceous calcareous nannofossils in the Western Interior basin, North America. In: Caldwell, W.G.E., Kauffman, E.G. (Eds.), Evolution of the Western Interior Basin. Spec. Pap.-Geol. Assoc. Can., vol. 39, pp. 521–537. Weedon, G.P., 1989. The detection and illustration of regular sedimentary cycles using Walsh power spectra and filtering, with examples from the Lias of Switzerland. J. Geol. Soc. Lond. 146, 133–144. Weedon, G.P., 2003. Time-series analysis and Cyclostratigraphy; Examining Stratigraphic Records of Environmental Cycles. Cambridge University Press, Cambridge. 274 pp.