The Late Devonian extinction event Towards the end of the Devonian, at the boundary between the Frasnian and Famennian ages about million years ago , a great extinction turned the living world upside-down. This time, it was not a single, abrupt event, but instead a series of extinctions spread out over a relatively short period of two to four million years.
At that time, the colonization of land by terrestrial plants, invertebrates and the first tetrapods was still in its early days. Marine species, on the other hand, were very hard hit. Large underwater reef ecosystems, constructed mainly by rugose and tabulate corals and by stromatoporoids, dramatically vanished. They would not return until million years later in the Triassic Period, built that time by scleractinian corals and calcisponges.
Suggested causal mechanisms for the extinction include one or more extraterrestrial impacts 7 , 8 , large-scale volcanism 9 , 10 , either climate cooling 11 or warming 12 , and enhanced continental weathering, through mountain building 13 or development of terrestrial forests 5 , Increase continental weathering in particular would have resulted in eutrophication, planktonic algal blooming and subsequent widespread marine anoxia.
Better constraints on the rate and timing of environmental change before, during and after the extinction could resolve the diverging hypotheses. Indeed, numerous different chronologies have been constructed for the F—F extinction event synthesis in Fig.
Recently, strong efforts have been made to improve the Devonian time scale through cyclostratigraphy 20 , 21 , For the F—F interval, De Vleeschouwer et al.
Subsequently, De Vleeschouwer et al. Pas et al. This cyclostratigraphic approach resulted in a F—F boundary age of The F—F age is based on a cyclostratigraphic extrapolation from ash beds Recently, Percival et al. Using this age, extrapolating the De Vleeschouwer et al. Frasnian—Famennian time scales and their uncertainties proposed in recent studies. This uncertainty may be under evaluated, since it is based on an extrapolation from an ash beds The present study brings together the latest developments in the fields of astrochronology and radio-isotopic dating to conduct a cyclostratigraphic analysis of a multi-proxy data set for the Steinbruch Schmidt section.
It subsequently integrates the high-precision radio-isotopic date 25 to tie the floating chronology to an absolute time frame. These results are then compared with previous chronologies. This approach leads to a unique high precision chronology for the F—F interval, anchored in absolute time and at unprecedented resolution.
The Steinbruch Schmidt section was deposited at tropical palaeolatitudes Fig. This section offers outstanding conditions for a stratigraphic framework and timescale that can be globally correlated.
It contains well-developed Lower and Upper Kellwasser black-shale intervals and associated positive carbon-isotope excursions 3 , 27 , a high-precision radio-isotopic date 19 , 25 , as well as an excellent biostratigraphic framework For this study, the 5. As mentioned above, there is an ash bed in the Steinbruch Schmidt section, intercalated between the two Kellwasser horizons. Geological Setting and lithological column of the Steinbruch Schmidt section and of other records used in this paper.
The Frasnian—Famennian F—F boundary is marked by a dotted red line, the ash bed dated by 25 by dotted orange line, and the lower and upper Kellwasser LWK, UKW black shale intervals are underlined by grey areas. The X-ray beam was focused by a poly-capillary lens on a spot with a diameter of 25 mm Mo—Ka. An integration time of 60 s per point was chosen, satisfying the conditions for reaching the Time of Stable Accuracy and Time of Stable Reproducibility and providing the ideal compromise between high analytical accuracy and precision and high sample throughput 29 for analytical details.
All time-series analyses are performed on the R platform All proxy series are interpolated to 0. First, the function eTimeOpt within the Astrochron package 32 is used to evaluate sedimentation rates and their stratigraphic evolution 33 , The eTimeOpt method provides a quantitative assessment of a sequence of possible sedimentation rates, adopting a sliding window approach.
Within each window, sedimentation rates that result in precession amplitude modulation patterns and frequency ratios in accord with astronomical theory are singled out.
The eTimeOpt function requires the user to provide target eccentricity and precession periodicities to be tested. The precession periodicities Therefore, the kyr and kyr periodicities have remained unchanged throughout geological history. Numerical calculations by Laskar et al. In this study, eccentricity periodicities of The eTimeOpt chronology is then compared with a time scale obtained though the age modelling protocol designed by 6.
A tentative relative age was assigned to each of those tie-points, abiding to the kyr astrochronologic framework for Western Canada Fig. Those authors then applied a Monte Carlo procedure that distorts time-differences between consecutive tie-points, i. Their goal was to reach the best expression of the Milankovitch cycles, under the presumption that the chronology with the clearest orbital signature advocated to be the most accurate detailed description in 6.
The new results from the Steinbruch Schmidt section are incorporated into this correlation framework and assigned relative age tie-points with 0 kyr corresponding to the F—F boundary identical to those in 6. Finally, the Wavelet Transform evolutive spectra CWT 38 , obtained through the biwavelet package 39 is applied. Continuous wavelet is an evolutive spectral techniques that reveals the evolution of periodicities for the studied sections and detects changes in cycle thickness and, hence, sediment accumulation rate.
Magnetic susceptibility MS is viewed as a proxy for detrital input e. Ti evolution includes very strong peaks. By using the logarithmic signal, modest base-level variations are enhanced while positive peak values are reduced.
The Steinbruch Schmidt section is dominated by carbonate lithologies, with the two Kellwasser dark marls or shales. Within these argillaceous intervals, MS increases. Although the LKW dark marly layer ranges between 0. The UKW marly dark layer ranges between 3. LogTi values display strong variations, with a slight increase in the average value during the marly Kellwasser intervals. The application of eTimeOpt produces a first assessment of a potential Milankovitch imprint in these different proxy records.
As an input parameter, eTimeOpt requires an estimate of a range of plausible sedimentation rates for the studied interval. As outlined in the introduction, LKW-to-F—F-boundary duration estimates range from kyr 6 to more than three times that timespan The corresponding stratigraphic interval at the Steinbruch Schmidt section consists of about cm of section.
Hence, plausible sedimentation rates range between 0. Continuous Wavelet Transform of the Magnetic susceptibility signal, together with the different results from eTimeOpt red and green lines and ORTA black dotted line on the magnetic susceptibility record.
However, eTimeOpt for amplitude modulation of precession cannot consider the full sedimentation rate interval as the proxy-series resolution is too low to resolve precession cycles for sedimentation rates lower than 0. Running eTimeOpt again on all proxies, investigating the amplitude modulation of short eccentricity by long eccentricity circumvents this problem. These settings lead to eTimeOpt results for the full sedimentation rate range.
Despite the fundamentally different eTimeOpt settings, the sedimentation rates remain similar ranging between 0. The eTimeOpt results appear thus consistent, for both amplitude modulation of precession by short eccentricity or short eccentricity by long eccentricity. The decrease in the amount of tie points from 7 to 5 compared to 6 gives more freedom for distortions.
The kyr that separate the extreme tie points dictate a mean sedimentation rate of 0. The ORTA transforms the signal from the distance to the time domain between the different tie-points using a Monte Carlo approach, retaining the chronology that has the best expression of the expected Milankovitch frequencies between all different proxies. The algorithm yields an optimized result with a total duration of 1, ka, corresponding to a mean sedimentation rate of 0.
This result fall close to those obtained through eTimeOpt. Finally, the results are examined using a continuous wavelet transform Fig. These periods fit respectively with short and long eccentricity, with sedimentation rates around 0. Magnetic susceptibility correlation of Steinbruch Schmidt section with other F—F records H, CG-1, Section C, Sinsin, and Fuhe, see 6 and visual distinct magnetic susceptibility features for correlations red lines and arrows.
These tie points have to respect the biostratigraphic constraints underlined by grey area and carbon isotope correlations Fig. The black numbers are ages assigned to each tie-point, according to the existing astrochronologic framework of Section C and Kowala 23 , Famennian, h. Carbon isotope correlation of Steinbruch Schmidt section with other F—F records H, CG-1, Section C, Sinsin, Fuhe, and Kowala, see 6 obtained by visually correlating distinct features in carbon isotope geochemistry blue ties and dotted arrows.
These tie points are accepted only if they respect the biostratigraphic constraints underlined by grey area and magnetic susceptibility correlations Fig. Abbreviations as in Fig. Two fully independent spectral techniques eTimeOpt from 33 , and the ORTA from 6 applied on the different proxies from the Steinbruch Schmidt section, lead to similar results for the duration of the whole section.
The narrow range of results is also illustrated by the average difference between different duration estimates and the median estimate of only 37 kyr Table 1.
The subsequent discussion is primarily based on the results obtained by ORTA. Results are reported for window mid-points, which implies that the results for the extremities of the record need to be extrapolated. Nevertheless, all results are very close as the difference between the various results obtained through the different techniques and those from ORTA on magnetic susceptibility never amounts to more than 83 kyr Table 1.
The Steinbruch Schmidt bentonite layer at bed 36, located between the LKW and the UKW horizons has been recently re-dated by 25 , who reported a weighted mean age of Combining this age with the present cyclostratigraphy provides an integrated time scale for the Frasnian—Famennian boundary interval Fig. The new F—F boundary age is younger than most previous estimates see Fig. However, it is important to note that this young age is primarily inherited from the younger U—Pb age 25 , rather than resulting from issues in the cyclostratigraphic framework, which conforms with previous astrochronologic studies 6.
The Lower and Upper Kellwasser black argillaceous horizons represent respectively between 80 to 96 kyr and to kyr of time equivalent to approximately one short eccentricity cycle, Fig. The interval between the onset of the Lower Kellwasser carbon-isotope excursion Integrated anchored chronology for the Frasnian—Famennian boundary, obtained through the combination of ORTA and the 25 radio-isotopic date orange dotted line.
The grey bands mark the extension of the lower and upper black shally Kellwasser intervals. The light blue arrow marks the onset of positive carbon excursion associated with the Lower Kellwasser The red dotted line marks the Frasnian—Famennian F—F boundary Uncertainty assessment on cyclostratigraphic results is a challenging endeavour An uncertainty assessment can be proposed by considering the results obtained through different cyclostratigraphy techniques and proxies, associated with the uncertainty on the ash bed.
Considering the techniques applied on different proxies generates seven different cyclostratigraphic duration estimates for the studied stratigraphic interval Table 1. Percival et al. Consideration of those uncertainties gives a Frasnian—Famennian boundary age of The newly constructed integrated and anchored time scale obtained for the Steinbruch Schmidt section also allows for the evaluation of the impact of astronomical forcing on the Kellwasser environmental perturbations.
A connection between orbital forcing and organic carbon accumulation has been frequently postulated 46 , 47 , However, two contrasting hypotheses exist regarding the exact nature of this link synthesis in 49 : one in which organic matter accumulation is favoured during eccentricity maxima and one during eccentricity minima.
The eccentricity maxima hypothesis involves a context of strong seasonal contrast between dry and wet seasons, allowing strong fluvial discharge, productivity blooms and organic-matter accumulation.
As extinctions were mostly of tropical groups climate change may have been involved, and there is geological evidence for cooling of the global climate at the end-Frasnian event and near the end of the Devonian Period.
Cooling may have been caused by a drop in the amount of carbon dioxide in the atmosphere. Carbon dioxide is a greenhouse gas that helps warm the planet, so if levels fall, cooling will follow. In the Late Devonian, large trees evolved and formed the first forests. As plant life expanded, they used up more carbon dioxide in photosynthesis.
When dead plant material decays, carbon dioxide is returned to the atmosphere, but some plant material e. This buried plant material removes carbon permanently from the atmosphere and often forms coal.
0コメント