Research Article

Empirical evidence for stability of the 405-kiloyear Jupiter–Venus eccentricity cycle over hundreds of millions of years

Dennis V. Kent, Paul E. Olsen, Cornelia Rasmussen, Christopher Lepre, Roland Mundil, Randall B. Irmis, George E. Gehrels, Dominique Giesler, John W. Geissman, and William G. Parker
PNAS June 12, 2018 115 (24) 6153-6158; first published May 7, 2018; https://doi.org/10.1073/pnas.1800891115

  1. Edited by Lisa Tauxe, University of California, San Diego, La Jolla, CA, and approved April 6, 2018 (received for review January 16, 2018)

See related content:

Significance

Rhythmic climate cycles of various assumed frequencies recorded in sedimentary archives are increasingly used to construct a continuous geologic timescale. However, the age range of valid theoretical orbital solutions is limited to only the past 50 million years. New U–Pb zircon dates from the Chinle Formation tied using magnetostratigraphy to the Newark–Hartford astrochronostratigraphic polarity timescale provide empirical confirmation that the unimodal 405-kiloyear orbital eccentricity cycle reliably paces Earth’s climate back to at least 215 million years ago, well back in the Late Triassic Period.

Abstract

The Newark–Hartford astrochronostratigraphic polarity timescale (APTS) was developed using a theoretically constant 405-kiloyear eccentricity cycle linked to gravitational interactions with Jupiter–Venus as a tuning target and provides a major timing calibration for about 30 million years of Late Triassic and earliest Jurassic time. While the 405-ky cycle is both unimodal and the most metronomic of the major orbital cycles thought to pace Earth’s climate in numerical solutions, there has been little empirical confirmation of that behavior, especially back before the limits of orbital solutions at about 50 million years before present. Moreover, the APTS is anchored only at its younger end by U–Pb zircon dates at 201.6 million years before present and could even be missing a number of 405-ky cycles. To test the validity of the dangling APTS and orbital periodicities, we recovered a diagnostic magnetic polarity sequence in the volcaniclastic-bearing Chinle Formation in a scientific drill core from Petrified Forest National Park (Arizona) that provides an unambiguous correlation to the APTS. New high precision U–Pb detrital zircon dates from the core are indistinguishable from ages predicted by the APTS back to 215 million years before present. The agreement shows that the APTS is continuous and supports a stable 405-kiloyear cycle well beyond theoretical solutions. The validated Newark–Hartford APTS can be used as a robust framework to help differentiate provinciality from global temporal patterns in the ecological rise of early dinosaurs in the Late Triassic, amongst other problems.

The 27-My Newark–Hartford astrochronostratigraphic polarity timescale (APTS) (1) is one of the longest continuous APTS segments presently available and calibrates much of the Late Triassic and earliest Jurassic geologic timescale. It relies on the geomagnetic polarity and cycle stratigraphies of over 6,900 m of coring in the Newark basin (2⇓–4) and an overlapping 2,500-m-thick outcrop section in the nearby Hartford basin (5), both in eastern North America. The composite continental record is paced by 66 McLaughlin lithologic cycles spanning 27 My of the Norian and Rhaetian of the Late Triassic and Hettangian and Sinemurian of the Early Jurassic, reflecting the response of climate to the long astronomical eccentricity variation with a presumed 405-ky period. The 405-ky period cycle is related to the gravitational interaction of Jupiter and Venus (g2–g5 cycle) and is the prominent and most stable term in the approximation of eccentricity of Earth’s orbital variations on geologic timescales despite chaotic behavior of the Solar System (6). The record also encompasses 51 Poisson-distributed geomagnetic polarity intervals (with an additional 15 polarity intervals in the fluvial noncyclic sediments toward the base of the section, which by extrapolation of sediment accumulation rates extend the record an additional ∼6 My into the Carnian), providing a template for global correlation. However, the astronomically paced polarity sequence is anchored at essentially only one level. High-precision U–Pb zircon dates in lavas and intrusions of the Central Atlantic Magmatic Province (CAMP), which are clustered in close spatiotemporal proximity to the Triassic–Jurassic boundary (7), were collapsed to a calibration age of 201.6 Ma for the onset of Chron E23r close to the base of a McLaughlin cycle (Ecc405: k = 498.25, where k is the inferred number of 405-ky eccentricity cycles projected back from the most recent maximum at 0.216 million years ago as k = 1), which immediately underlies the oldest CAMP basalts. In the absence of other directly dated horizons, the APTS relies on the assumption of a continuous section with no substantial hiatus(es) in deposition of these continental sediments, which would potentially result in one or more unrecorded 405-ky cycles. The specific timing of the APTS is also dependent on the untested reliability at this distant age range of the 405-ky cycle.

The developing APTS has been successfully used for global correlations in marine and nonmarine facies (e.g., refs. 8 and 9; see ref. 1). Nevertheless, there have been persistent suggestions made largely on the basis of nonmarine biostratigraphic correlations that several million years of Rhaetian (latest Triassic) time are missing in a cryptic unconformity that supposedly occurs just above Chron E23r in the Newark Supergroup basins (e.g., refs. 10 and 11). If true, this would have consequences of comparable magnitude in the timing of events based on anchoring the astrochronology below the alleged cryptic unconformity to the U–Pb-dated CAMP lavas above it. Therefore, a test is needed of the conjoined assumptions of stratigraphic continuity and the 405-ky periodicity for the long eccentricity cycle that are implicit in the construction of the Newark–Hartford APTS.

The Late Triassic-age Chinle Formation (Fm.) of the American Southwest (Fig. 1A) consists of fluvial and minor lacustrine facies interfingered with paleosols and, importantly for the task at hand, has numerous sandstone horizons with volcaniclastic detritus containing detrital zircons amenable for U–Pb dating (12). The Black Forest Bed (BFB) within the Petrified Forest Member (Mb.) of the Chinle Fm. at Chinde Point in the northern sector of Petrified Forest National Park (PFNP) was the first unit to be successfully U–Pb zircon dated in the PFNP section (13). Recent high-precision U–Pb zircon dating of the BFB (12) made it an attractive target for calibration of the APTS. Sampling of outcrop sections demonstrated the feasibility of obtaining a magnetostratigraphy (14) even though revised long-distance lithostratigraphic correlations of the Sonsela Mb. of the Chinle Fm. (15) indicate that there may be a large gap in the composite magnetostratigraphic section. A main scientific objective of the inaugural drilling effort of the Colorado Plateau Coring Project (16) at PFNP (core CPCP-PFNP-13-1A; henceforth PFNP-1A) (Fig. 1B) was to obtain a magnetostratigraphic sequence directly supported by high-precision U–Pb detrital zircon dates for the Chinle Fm. with unequivocal superposition and a diagnostic polarity signature for correlation to the APTS. Here, we report on paleomagnetic and geochronologic results from the upper ∼280 m of the 519-m-long core recovering a section of the Chinle Fm. down from the lower Owl Rock Mb. to the Mesa Redondo Mb., the Moenkopi Fm. and the uppermost Coconino Sandstone (Fig. 1B).

Fig. 1.
Fig. 1.

Geographic location and stratigraphy of scientific drill core CPCP-PFNP-13-1A in Petrified Forest National Park (Arizona). (A) Nested maps showing location of core PFNP-1A in Petrified Forest National Park (PFNP) on the Colorado Plateau in western North America. (B) Stratigraphic log of continuously cored section of the Bidahochi, Chinle, Moenkopi, and Coconino formations recovered in core PFNP-1A is shown with lithostratigraphic unit names (38), and the predominant color of the sediments from a digitally smoothed, enhanced color profile based on spectrophotometer data provided by LacCore (courtesy of Anders Noren, University of Minnesota, Minneapolis). Core length (in meters core depth) was converted to stratigraphic thickness (in meters stratigraphic depth) assuming flat-lying strata with core drilled 30° from vertical, and hence stratigraphic thickness is 86.6% of core length. ABQ, Albuquerque; AZ, Arizona; CA, California; CO, Colorado; FLG, Flagstaff; GJT, Grand Junction; LV, Las Vegas; NM, New Mexico; NV, Nevada; PFNP, Petrified Forest National Park; PFNP-A, Private or State Trust land; PFNP-CPCP13-1A, core site at Chinde Point.

Results

Following the procedures described in Materials and Methods and SI Appendix, characteristic remanent magnetization (ChRM) directions were successfully isolated in 132 of the 174 oriented subsamples extracted from ∼24–278 meters core depth (mcd) [∼21–248 meters stratigraphic depth (msd)] encompassing the lower Owl Rock Mb., the entire Petrified Forest Mb. including the BFB, and down into the middle part of the Sonsela Mb. The ChRM direction for each sample was converted to a virtual geomagnetic pole (VGP) whose latitude with respect to the 210-Ma reference (north) paleopole (17), referred to as rVGP latitude, is used to designate polarity. A plot of rVGP latitudes with depth in core PFNP-1A shows a series of magnetic polarity zones that are designated from the top down as PF1r (partim), PF2n, PF2r, PF3n, PF3r (with a thin normal polarity interval, PF3r.1n), PF4n, PF4r, and PF5n (partim) (Fig. 2). The paleomagnetic signature of sediments of the Owl Rock Mb. from immediately below the erosional unconformity with the late Cenozoic Bidahochi Fm. (not sampled) at 20.6 msd to around 35 msd is erratic, which we attribute to complications during coring and recovery of this shallow and poorly indurated material. The coherence of the polarity data improves sufficiently down core to establish that the lowermost Owl Rock Mb. is in a reverse polarity magnetozone (PF1r) that extends down to a normal magnetozone PF2n (41.1–48.0 msd) straddling the contact with the Petrified Forest Mb. at 41.7 msd. The underlying reverse polarity magnetozone PF2r (48.0–64.5 msd) is sparsely documented because the pale-colored cross-bedded sands and silts of the BFB (∼55.5–66.9 msd) that dominate this interval are poor paleomagnetic recorders, with only 3 out of 17 samples providing acceptable results. Finer-grained red lithologies are more common in the underlying part of the Petrified Forest Mb. and record substantial evidence of a thick (64.5–125.4 msd) normal polarity magnetozone (PF3n), a modestly thick (125.4–153.9 msd) reverse magnetozone (PF3r) that includes a thin (144.5–149.5 msd) normal polarity interval in its lower part, and a normal magnetozone (PF4n; 153.9–187.3 msd) that extends below the contact with the Sonsela Mb. (160.3 msd). The generally coarser and less red lithologies of the Sonsela Mb. proved more difficult for obtaining a coherent paleomagnetic signature. The available data delineate reverse polarity magnetozone PF4r from 187.3 to 215.7 msd and normal polarity magnetozone PF5n from 215.7 msd to the lowest sample with acceptable data in this suite at 240.8 msd. There are also three putative (one-sample) polarity magnetozones of uncertain origin (e.g., inverted samples?) that are not interpreted further.

Fig. 2.
Fig. 2.

Magnetostratigraphy of core PFNP-1A (for rock units and depth scale see legend to Fig. 1). rVGP latitudes are for sample ChRM directions converted to virtual geomagnetic poles whose rotated latitude with respect to the 210-Ma mean reference pole for North America (17) are plotted versus core (mcd) and stratigraphic (msd) depth for the upper Chinle Fm. Positive (northerly) and negative (southerly) rVGP latitudes correspond to normal and reverse polarity, delineated in polarity column by filled and open bars, respectively. Solid circles are accepted sample data with MAD values of 16° or less; crosses are rejected sample data (SI Appendix, Table S3). Proposed correlation indicates that the sampled section extends from around 209–215 Ma of the Newark–Hartford APTS chronology (1), which is based on a U–Pb-dated anchor at 201.6 Ma for 405-ky eccentricity cycles (k, peaks numbered from most recent at 0.216 Ma as k = 1 for Ecc405:k). Some global events and trends in the Late Triassic that can be integrated with the age-validated APTS include the large-bodied sauropodomorph-rich La Esquina assemblage [e.g., Coloradisaurus (34)] in the upper part of the Los Colorados Fm. in the Ishigualasto Basin in Argentina (25), and the Revueltian assemblage of the Chinle Fm. with rare small-bodied theropod dinosaurs [e.g., Chindesaurus from the Petrified Forest Mb. in the PFNP, Arizona, and the Hayden Quarry at Ghost Ranch, New Mexico (24)] that is only about a million years younger and completely lacks sauropodomorphs. The 65-km-diameter Manicouagan impact crater in Quebec, Canada, has melt rocks dated by U–Pb zircon geochronology to 215.5 Ma (35) that are characterized by normal polarity (36). The atmospheric carbon dioxide concentration, pCO2 in parts per million (ppm) is also shown as a red curve with data points as circles based on the paleosol proxy (37).

Given the overall lithostratigraphic, biostratigraphic, and geochronologic framework of the upper Chinle Fm (12, 18), the most plausible correlation of the local magnetozones (PF1r to PF5n, youngest to oldest) equates them to chrons E17r to E14n (younger to older) of the APTS, spanning ∼209.5–215.5 Ma (Fig. 2).

The BFB is a distinctive white tuffaceous crevasse-splay sandstone (13, 19) that can be more than 10 m thick. It occurs within the upper part of the mostly red Petrified Forest Mb. and can be readily traced in outcrops throughout the northern part of the PFNP, including Chinde Point. A ∼50-m-thick normal polarity interval equivalent to magnetozone PF3n was also found just below the BFB in outcrop (14) but was miscorrelated to Chron E15n, although the thin normal within the underlying reverse polarity interval (PF3r) that secures its correlation to Chron E15r is well represented in the outcrop section. Paleomagnetic results were not obtained higher in the outcrop section that might have allowed identification of the critical PF2r magnetozone that straddles the BFB. Detrital zircons in an early study yielded U–Pb dates that indicated a likely, yet poorly defined, depositional age of 209 ± 5 Ma for the BFB (13). In a subsequent study, 16 prismatic zircons from a sample about 1.5 m above the base of the BFB at Chinde Point were analyzed by chemical abrasion–thermal ionization mass spectrometry (CA-TIMS); five distinguishably younger zircons gave a weighted mean U–Pb date of 209.93 ± 0.07 Ma (2σ) (12).

Here, we report a U–Pb CA-TIMS detrital zircon age of 210.08 ± 0.22 Ma for a coherent cluster of 6 out of 14 zircons analyzed from the BFB in core PFNP-1A (SI Appendix). The two U–Pb CA-TIMS zircon dates for the BFB differ by an insignificant 0.15 My. Sample 52Q2 (210.08 ± 0.22 Ma) in the BFB is from within magnetozone PF2r, which is correlated to Chron E16r whose APTS age range of 209.95–210.25 Ma was based on counting 20.62–21.35 long eccentricity (405 ky) cycles back from the high-precision U–Pb zircon date at 201.6 Ma in the Newark section (1, 7). The U–Pb CA-TIMS zircon date for the BFB agrees precisely with the predicted APTS age for Chron E16r. This agreement immediately discounts assertions that even a small fraction of a 405-ky cycle let alone millions of years are missing below the CAMP lavas in the Newark section.

The overall stratigraphic coherence of the magnetic correlations with the BFB and other U–Pb zircon dates can be gauged in a plot of depth versus age (Fig. 3). The magnetic correlations to the APTS (coefficient of determination, R2 = 0.9845) suggest a relatively uniform net sediment accumulation rate of about 34.8 msd/My from the upper Sonsela, through the Petrified Forest, and up into the lower Owl Rock members. A tight linear regression of composite heights of lithostratigraphic members from outcrop (12, 20) versus their stratigraphic depths in PFNP-1A (SI Appendix, Fig. S5 and Table S1) allows the U–Pb-dated levels in outcrop to be compared with those reported here from core PFNP-1A. The dates for the BFB from outcrop (sample BFBo) and core (sample 52Q2) are indistinguishable, as described above, supporting the use of the base of the distinctive BFB as a regional time horizon. Sample 158Q2 along with samples GPU and P57-C confirm that the included magnetozone PF4n indeed correlates to Chron E15n, whereas sample 182Q1 along with sample GPU confirm that included magnetozone PF4r most probably correlates to Chron E14r. However, sample GPL has a reported U–Pb CA-TIMS zircon date of 218.02 ± 0.28 Ma (12) that is much too old for its projected level at 207.8 msd in PFNP-1A, which would place it within magnetozone PF4r whose correlative Chron E14r ranges back to only 214.92 Ma (Fig. 3). Further discussion of GPL and other dated samples from outcrop lower in the Chinle Fm. are deferred pending acquisition of additional U–Pb CA-TIMS zircon dates directly from the core.

Fig. 3.
Fig. 3.

Depth versus age plot for core PFNP-1A based on correlation of magnetostratigraphy with the Newark–Hartford APTS (1). Stratigraphic units, graduated depths, and color log of PFNP-1A core as in Figs. 1 and 2. Red crosses are magnetozone boundaries in PFNP-1A correlated to the APTS (SI Appendix, Table S2); solid red line is a linear regression for base magnetozone PF1r to base magnetozone PF4r to their correlative chron ages. Blue circles are U–Pb CA-TIMS detrital zircon dates from PFNP-1A reported here (SI Appendix, Table S4); light blue squares are published U–Pb CA-TIMS detrital zircon dates from outcrop correlated here to PFNP-1A using a regression on member boundaries mapped in an outcrop composite section (12, 20) versus those in PFNP-1A (SI Appendix, Fig. S5). Linear regression on U–Pb CA-TIMS dates (blue line) based on sample data from core PFNP-1A (excluding sample 177Q1), and by filled circles (see SI Appendix, Fig. S6 for regressions including U–Pb CA-TIMS zircon dates on outcrop samples). U–Pb zircon date for Manicouagan crater impact melt rocks (35), which are characterized by normal polarity (36), is shown for reference. Inset Shows a paleocontinental reconstruction of Pangea positioned according to a 220-Ma mean global paleopole (17) with some key continental localities indicated by filled circles connected by arrows to their relative positions at 200 Ma by open circles. Map adapted with permission from ref. 25.

The U–Pb CA-TIMS zircon dates for samples 52Q2, 158Q2, and 182Q1 from the core are highly consistent with the depth–age relationship for samples BFBo, GPU, KWI, and P57-C from outcrop projected to PFNP-1A as described above (SI Appendix, Fig. S6). However, given the added stratigraphic uncertainties of projecting the dates from surface exposures to PFNP-1A, we prefer to use the dates obtained from the core (samples 52Q2, 158Q2, and 182Q1) for a more detailed analysis. This set of sample data provides a regression for Y (meters stratigraphic depth in PFNP-1A) with respect to X (age in Ma) (Y = 34.34 * X − 7,158; R2 = 0.9994) that is in remarkably close agreement with a regression of predicted depth versus ages from the APTS via magnetostratigraphy (Y = 34.84 * X − 7,258 m; R2 = 0.9845) and yield highly consistent estimates for sediment accumulation rate of 34.3–34.8 m/My (Fig. 3).

Discussion

The very close agreement between the ages predicted by the magnetic polarity correlations and the U–Pb zircon geochronology directly confirms the timing of the APTS back to about 215 Ma, at around the Chron E14r/14n boundary; this immediately refutes suggestions that significant time is missing in the Newark–Hartford sequence used to build the APTS. The congruence between the Colorado Plateau–zircon-calibrated polarity timescale and the Newark–Hartford APTS can be directly attributed to the reliability of the 405-ky eccentricity variation recorded in the Newark–Hartford sequence, in terms of its correct identification and counting of its expression as the long eccentricity climate cycle (21) as well as the periodicity being close to 405 ky as calculated from modern celestial mechanics (6). In fact, the age difference (8.50 My) estimated from slightly less than 21 long eccentricity cycles between the U–Pb-dated tie points at 210.08 Ma for the BFB within Chron E16r (Ecc405:518.87–519.60, centered at 519.24), and 201.6 Ma just below first CAMP basalt (Ecc405:498.25) is indistinguishable from the difference in the U–Pb dates (8.48 My, with an estimated uncertainty of ±0.3 My from propagation of the individual 2σ errors), which would imply a cycle duration of close to 404 ky, or within an insignificant 1 ky or 0.25% of the hypothesized 405-ky period. The periodicity of the Jupiter–Venus-long eccentricity cycle indeed must be very close to 405 ky even at this remote age range and we see no justification given the present observational uncertainties to assume a different value.

The 405-ky oscillation is the largest and most stable term in the periodic approximation of Earth’s eccentricity, but theoretical solutions formally constrain orbital motions to only about the past 50 million years, with different solutions yielding deviations of up to about ±405 ky over 250 My (6). Nonetheless, this apparent stability, however plausible, is entirely theoretical. Independent U–Pb CA-TIMS zircon dating is just beginning to confirm the accuracy of these solutions in the Cenozoic (22). Our work shows that the 405-ky eccentricity cycle can be used with confidence as a framework of geologic timescales back to at least 215 Ma in the Late Triassic, which also places an empirical constraint on the dynamics of the Solar System (6) over a time span more than four times longer than orbital solutions allow.

Finally, a number of global events in the Late Triassic can now be integrated more confidently into the framework of the APTS (Fig. 2). In particular, the vexing problem of Late Triassic biotic provinciality and the delay in rise to ecological dominance of tropical dinosaurs until after the end-Triassic mass extinction (23) can be explored without relying on biostratigraphy for time control, which has in the past resulted in circular reasoning. Dinosaurs had evolved by the early Late Triassic (Carnian), and by the Norian, relatively large, herbivorous, sauropodomorph dinosaurs were abundant at mid-to-high latitudes, but apparently absent from low latitudes (23, 24). Based on correlation of paleomagnetic polarity sequences to the confirmed APTS, we can now more securely place the diverse La Esquina assemblage, with its abundant large herbivorous sauropodomorph dinosaurs from the upper third of the Los Colorados Formation of Argentina (paleolatitude ∼40° S; Fig. 3, Inset) in Chron E15n at about 213 Ma (25). This assemblage is thus contemporaneous with the diverse assemblages of the early part of the Revueltian biozone (18) of the western United States (paleolatitude, ∼10°N), in which dinosaurs are rare and small-bodied, and sauropodomorphs (and all herbivorous dinosaurs) are absent (26). Like in the western United States, strata in chrons E15n-E16n in the Newark and Fundy basins (27) with paleolatitudes of ∼15°N produce diverse tetrapod track assemblages but lack putative sauropodomorph footprints, and dinosaur track-makers are rare and small-bodied. The Fleming Fjord Fm. bone and track assemblages from Greenland (paleolatitude, ∼43°N) have both abundant large-bodied sauropodomorph skeletons and footprints (28) and correlate to chrons possibly only as old as E14n (∼216 Ma) (29), similar in age to the La Esquina assemblage in the Los Colorados Fm. as well to the Revueltian assemblages in the Chinle and contemporaneous assemblages in the Newark and Fundy basins. The absence of large-bodied herbivorous dinosaurs in the tropics of Pangea, the strong biotic provinciality (30, 31), and the 30-My delay in the rise of dinosaurian ecological dominance in the tropics (23), is thus not an artifact of biostratigraphic miscorrelation as construed by some (32) but a real feature of that world in which it can now be quantified both in time and space.

Materials and Methods

A total of 174 oriented subsamples were extracted from ∼24–278 mcd (∼21–248 msd) from the finest-grained most red-colored sediment facies of the lower Owl Rock Mb., the entire Petrified Forest Mb. including the BFB, and down into the middle part of the Sonsela Mb. Progressive thermal demagnetization was used to isolate the characteristic component of the natural remanent magnetization of each specimen that tended to be carried predominantly by hematite (SI Appendix, Fig. S1). A least-squares linear fit using principal-component analysis over the three to seven (typically five) demagnetization steps in the range of 300–600 °C and anchored to the origin (SI Appendix, Fig. S2) was used to assess the characteristic remanent magnetization (ChRM) in each sample. Results were deemed acceptable if the maximum angular deviation (MAD) (33) was 16° or less; 132 of the 174 samples (76%) satisfied this criterion (average MAD of 8°) and fell into two populations (SI Appendix, Fig. S3), shallow northerly and shallow southerly groupings that are interpreted to represent normal and reverse polarities of the Late Triassic geomagnetic field that provide a magnetostratigraphy (SI Appendix, Tables S2 and S3).

Samples for U–Pb zircon dating were also taken at the Rutgers Core Repository. Each sample was a 10- to 20-cm-thick quarter-round cut using a bandsaw with a carborundum-tipped blade, targeting sandstones that were poorly sorted and contain abundant accessory minerals indicative of volcanic detritus. The samples were crushed, and mineral concentrates were then purified using standard mineral separation techniques, including sieves, magnetic separation, and density separation at the LaserChron Center at the University of Arizona, mounted on epoxy disks and polished to about one-half of the crystal width followed by laser ablation-inductively coupled plasma mass spectrometry (LA-ICPMS) analyses, providing a survey of the range of zircon ages present. At the Berkeley Geochronology Center, 20–30 crystals displaying the youngest LA-ICPMS age from each sample were extracted from the mount and prepared for zircon U–Pb chemical abrasion, thermal ionization mass spectroscopy (CA-TIMS). Before TIMS analysis, all zircons were pretreated using thermal annealing at 850 °C for 48 h, followed by chemical abrasion with concentrated HF in pressurized dissolution capsules at 220 °C for 8–12 h. The analytical data of four samples from the Chinle Fm. subjected to U–Pb zircon CA-TIMS analyses are presented in SI Appendix, Fig. S4 and Tables S4 and S5.

Acknowledgments

We thank the National Park Service, particularly superintendent Brad Traver, for permission to core in the park and for logistical support during site selection and drilling. On-site and laboratory core processing, scanning, and archiving were carried out by LacCore, particularly Anders Noren, Kristina Brady, and Ryan O’Grady; on-site core-handling volunteers Justin Clifton, Bob Graves, Ed Lamb, Max Schnurrenberger, and Brian Switek are thanked for their around-the-clock efforts, and drilling manager Doug Schnurrenberger for overseeing a superb coring project. We sincerely thank the two journal reviewers for insightful comments that allowed us to improve the data analyses. This project was funded by National Science Foundation (NSF) Collaborative Grants EAR 0958976 (to P.E.O. and J.W.G.), 0958723 (to R.M.), 0958915 (to R.B.I.), 0959107 (to G.E.G.), and 0958859 (to D.V.K.), and by Deutsche Forschungsgemeinschaft for International Continental Scientific Drilling Program support. Additional support was provided by NSF Grant EAR-1338583 (to G.E.G.) to the Arizona LaserChron Center; P.E.O. acknowledges support from the Lamont-Climate Center, R.M. acknowledges support of the Ann and Gordon Getty Foundation, and D.V.K. is grateful to the Lamont–Doherty Incentive Account for support of the Paleomagnetics Laboratory. Curatorial facilities for the work halves of the CPCP cores are provided by the Rutgers Core Repository. Any opinions, findings, or conclusions of this study represent the views of the authors and not those of the US Federal Government. This is a contribution to IGCP-632, and is Petrified Forest Paleontological Contribution 54, and Lamont–Doherty Earth Observatory Contribution 8208.

Footnotes

  • ↵1To whom correspondence should be addressed. Email: dvk{at}rutgers.edu.

  • Author contributions: D.V.K. and P.E.O. designed research; D.V.K., P.E.O., C.R., C.L., R.M., R.B.I., J.W.G., and W.G.P. performed research; R.B.I. and J.W.G. assisted in recovery and description of core; R.B.I. and W.G.P. provided regional geologic context; W.G.P. assisted in providing access to the Petrified Forest National Park; D.V.K., C.R., R.M., G.E.G., and D.G. analyzed data; and D.V.K., P.E.O., and R.M. wrote the paper.

  • The authors declare no conflict of interest.

  • This article is a PNAS Direct Submission.

  • See Commentary on page 6104.

  • This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1800891115/-/DCSupplemental.

Published under the PNAS license.

Back to top
Article Alerts
Email Article
Citation Tools
Request Permissions
Empirical evidence for stability of the 405-kiloyear Jupiter–Venus eccentricity cycle over hundreds of millions of years
Dennis V. Kent, Paul E. Olsen, Cornelia Rasmussen, Christopher Lepre, Roland Mundil, Randall B. Irmis, George E. Gehrels, Dominique Giesler, John W. Geissman, William G. Parker
Proceedings of the National Academy of Sciences Jun 2018, 115 (24) 6153-6158; DOI: 10.1073/pnas.1800891115
Digg logo Reddit logo Twitter logo Facebook logo Google logo Mendeley logo
Proceedings of the National Academy of Sciences: 115 (24)
Table of Contents

Submit

Article Classifications

Jump to section

You May Also be Interested in

Birds nestling on tree branches
Parent–offspring conflict in songbird fledging
Some songbird parents might improve their own fitness by manipulating their offspring into leaving the nest early, at the cost of fledgling survival, a study finds.
Image credit: Gil Eckrich (photographer).

Similar Articles