Invited research articleTowards determination of the source and magnitude of atmospheric pCO2 change across the early Paleogene hyperthermals
Introduction
Climate sensitivity (CS; the equilibrium temperature increase due to a doubling of CO2) has important implications for policy makers (Knutti and Hegerl, 2008; Rohling et al., 2012; Rogelj et al., 2014; Knutti et al., 2017; Cox et al., 2018). Current understanding of pre-Quaternary climate sensitivity (or Earth-system sensitivity) is based on individual estimates of pCO2 from at least six different proxies (Royer, 2006; Park and Royer, 2011; Martínez-Botí et al., 2015; Anagnostou et al., 2016; Royer, 2016). Taken together, these data reveal Earth-system sensitivity of 1.6 to 9.6 °C during the Cenozoic (Hoffert and Covey, 1992; Hansen et al., 1993; Covey et al., 1996; Bijl et al., 2010; Lunt et al., 2010; Pagani et al., 2010; Royer, 2016), which includes both fast and slow feedbacks. The data used for these estimates are generally based on intervals of Earth history with stable levels of pCO2; the temperature response to a rapid CO2 increase is perhaps fundamentally different from long-term equilibrium Earth-system sensitivity (Zachos et al., 2008; Royer, 2016). Although some workers have studied past intervals of rapid pCO2 and temperature increases as analogs for anthropogenic climate change (e.g., the Paleocene-Eocene Thermal Maximum, PETM; Zachos et al., 2008), existing proxies are generally unable to resolve the shape or magnitude of pCO2 change across these events. The source of these events is also widely debated (Dickens, 2000; Kurtz et al., 2003; Higgins and Schrag, 2006), which leads to large uncertainty when modeling the pCO2 change associated with these events (Panchuk et al., 2008; Zeebe et al., 2009; Cui et al., 2011). For this reason, a new high-resolution pCO2 proxy capable of resolving pCO2 across sub-million year timescales is needed.
The early Paleogene contains at least four such intervals of significant carbon release between 56 and 53.5 Ma, marked by significant negative carbon isotope excursions (CIEs) identified within both marine and terrestrial substrates (Cramer et al., 2003; Nicolo et al., 2007; Abels et al., 2016; Lauretano et al., 2016). High-resolution oxygen isotope measurements on foraminifera preserved within marine sediments suggest global deep sea temperature increases of as much as 11 °C associated with the largest one of these events (Thomas et al., 2002; Zachos et al., 2003; Tripati and Elderfield, 2004; McCarren et al., 2008; Zachos et al., 2008; Dunkley Jones et al., 2013; Hansen et al., 2013; Lauretano et al., 2015). Existing pCO2 proxies, however, generally fail to precisely resolve the pCO2 rise associated with these events (e.g., Gehler et al., 2016), which makes comparison to present-day anthropogenic CO2 release difficult (Zeebe et al., 2016). Furthermore, modeling efforts to simulate pCO2 levels across these events commonly set background pCO2 = ~750 to 1000 ppmv (Panchuk et al., 2008; Zeebe et al., 2009; Cui et al., 2011; Zeebe et al., 2017), higher than any existing proxy-based value in order to simulate reasonable late Paleocene deep ocean temperature; therefore these simulations may result in overestimations of peak pCO2.
In this study, we present a high-resolution pCO2 record across the early Paleogene hyperthermals based on changes in carbon isotope discrimination between the δ13C value of terrestrial organic matter and that of atmospheric CO2 (Δ13C = (δ13CCO2 – δ13Corg)/(1 + δ13Corg/1000) (Schubert and Jahren, 2012). The effect of pCO2 on carbon isotope discrimination has shown potential for reconstructing pCO2 across geologically short (<1 Myr) timescales (Schubert and Jahren, 2015; Breecker, 2017; Hare et al., 2018), including the determination of background and peak levels of pCO2 during the Paleogene hyperthermals (Schubert and Jahren, 2013; Abels et al., 2016; Cui and Schubert, 2017). Here, we demonstrate a new application of this approach towards generating a nearly continuous record of pCO2 change across these events. These results allow for improved understanding of climate sensitivity in response to geologically rapid (i.e., <1 Myr) pCO2 rise, and comparison of pCO2 proxy results with model predictions.
Section snippets
Methods
Schubert and Jahren (2015) showed how pCO2 can be quantified based on a relative change in carbon isotope discrimination between some time t and a reference time (t = 0), designated as Δ(Δ13C):wherewhich can be expanded as:
By rearranging Eq. (1), one can therefore quantify pCO2(t) using the following equation:
Results
Consistent with previous studies comparing the size of the marine and terrestrial CIEs (Bowen et al., 2004; McInerney and Wing, 2011; Abels et al., 2012, 2016), we observed smaller magnitude changes in δ13C value inferred from the marine record (Fig. 1A) compared with the terrestrial record (Fig. 1B) across each event. After correcting for changes in δ13CCO2, we observed an increase in carbon isotope discrimination (Δ13C) across each of the four CIEs (Fig. 1C), consistent with elevated levels
Discussion
Available proxy data suggest relatively low pCO2 (348 + 112/−76 ppmv) during the early Paleogene (i.e., similar to 20th and 21st century levels) while model simulations generally set background pCO2 during this interval ~2 to 3 × higher than these proxies suggest (Cui et al., 2011; Panchuk et al., 2008; Zeebe et al., 2009, 2017). The lowest pCO2 estimates (<300 ppmv) are primarily based on paleosol carbonate (Cerling, 1992; Sinha and Stott, 1994; Royer et al., 2001) and the revised stomatal
Conclusions
These data represent the first high-resolution pCO2 record across the early Paleogene hyperthermals and show that pCO2 tracks temperature changes across the entirety of this interval. Background pCO2 was likely ~1.5× higher than that determined using other proxies; we assert that these new estimates are more in line with global surface temperatures that are >10 °C warmer than today (Hansen et al., 2013). Revision of the background pCO2 used in model simulations of these events will serve to
Acknowledgement
This research was supported by NSF EAR award #1603051. Y.C. thanks D. Royer for helpful discussions and an Obering postdoctoral fellowship from the Department of Earth Sciences at Dartmouth College.
References (76)
- et al.
Improving the Ginkgo CO2 barometer: implications for the early Cenozoic atmosphere
Earth Planet. Sci. Lett.
(2016) Atmospheric pCO2 control on speleothem stable carbon isotope compositions
Earth Planet. Sci. Lett.
(2017)- et al.
Quantifying uncertainty of past pCO2 determined from changes in C3 plant carbon isotope fractionation
Geochim. Cosmochim. Acta
(2016) - et al.
Atmospheric pCO2 reconstructed across five early Eocene global warming events
Earth Planet. Sci. Lett.
(2017) Climate model and proxy data constraints on ocean warming across the Paleocene-Eocene Thermal Maximum
Earth Sci. Rev.
(2013)- et al.
Beyond methane: towards a theory for the Paleocene–Eocene Thermal Maximum
Earth Planet. Sci. Lett.
(2006) - et al.
A high-resolution benthic stable-isotope record for the South Atlantic: implications for orbital-scale changes in Late Paleocene–Early Eocene climate and carbon cycling
Earth Planet. Sci. Lett.
(2014) Stomatal density and stomatal index as indicators of paleoatmospheric CO2 concentration
Rev. Palaeobot. Palynol.
(2001)CO2-forced climate thresholds during the Phanerozoic
Geochim. Cosmochim. Acta
(2006)- et al.
Phanerozoic atmospheric CO2 change: evaluating geochemical and paleobiological approaches
Earth Sci. Rev.
(2001)
Incorporating the effects of photorespiration into terrestrial paleoclimate reconstruction
Earth Sci. Rev.
New atmospheric pCO2 estimates from palesols during the late Paleocene/early Eocene global warming interval
Glob. Planet. Chang.
Terrestrial carbon isotope excursions and biotic change during Palaeogene hyperthermals
Nat. Geosci.
Environmental impact and magnitude of paleosol carbonate carbon isotope excursions marking five early Eocene hyperthermals in the Bighorn Basin, Wyoming
Clim. Past
Changing atmospheric CO2 concentration was the primary driver of early Cenozoic climate
Nature
Atmospheric lifetime of fossil fuel carbon dioxide
Annu. Rev. Earth Planet. Sci.
Transient Middle Eocene atmospheric CO2 and temperature variations
Science
A humid climate state during the Palaeocene/Eocene Thermal Maximum
Nature
Use of carbon isotopes in paleosols as an indicator of the P(CO2) of the paleoatmosphere
Glob. Biogeochem. Cycles
Early Eocene carbon isotope excursions: evidence from the terrestrial coal seam in the Fushun Basin, Northeast China
Geophys. Res. Lett.
Paleoclimate data constraints on climate sensitivity: the paleocalibration method
Clim. Chang.
Emergent constraint on equilibrium climate sensitivity from global temperature variability
Nature
Orbital climate forcing of δ13C excursions in the late Paleocene–early Eocene (chrons C24n–C25n)
Paleoceanography
Slow release of fossil carbon during the Palaeocene-Eocene Thermal Maximum
Nat. Geosci.
Past extreme warming events linked to massive carbon release from thawing permafrost
Nature
Methane oxidation during the late Palaeocene thermal maximum
Bull. Soc. Geol. Fr.
A Palaeogene perspective on climate sensitivity and methane hydrate instability
Phil. Trans. R. Soc. A Math. Phys. Eng. Sci.
Future climate forcing potentially without precedent in the last 420 million years
Nat. Commun.
Temperature and atmospheric CO2 concentration estimates through the PETM using triple oxygen isotope analysis of mammalian bioapatite
Proc. Natl. Acad. Sci.
Solar interior structure and luminosity variations
Sol. Phys.
Very large release of mostly volcanic carbon during the Palaeocene–Eocene Thermal Maximum
Nature
How sensitive is the worlds climate
Res. Explor.
Atmospheric CO2 effect on stable carbon isotope composition of terrestrial fossil archives
Nat. Commun.
Deriving global climate sensitivity from palaeoclimate reconstructions
Nature
Climate change 2013: the physical science basis
Sensitivity of the Eocene climate to CO2 and orbital variability
Clim. Past
Cited by (21)
Response to palaeoclimate by Early Cretaceous terrestrial organic-rich shales in the Yin'e Basin: Evidence from sporopollen, n-alkanes and their compound carbon isotopes
2024, Palaeogeography, Palaeoclimatology, PalaeoecologyQuantitative reconstruction of atmospheric pCO<inf>2</inf> sources during Eocene hyperthermal events based on data from the Fushun Basin, Northeast China
2022, Palaeogeography, Palaeoclimatology, PalaeoecologyCitation Excerpt :The Eocene strata in the Fushun Basin in Northeast China, including the Guchengzi, Jijuntun and Xiloutian formations, consist of thick organic-rich fine-grained deposits, which contain records of various hyperthermal events during the Eocene. In this study, we adopt the optimised methods of Cui et al. (Cui and Schubert, 2018; Cui et al., 2020) to calculate the Eocene pCO2 by δ13CTOC and δ13C27–31 and compare the differences in the results between the two proxies. The relationship between vegetation and the CO2 concentration was identified and quantified by biomarker compounds and organic macerals.
Differences in carbon isotope discrimination between angiosperm and gymnosperm woody plants, and their geological significance
2021, Geochimica et Cosmochimica ActaCarbon cycle variability in tropical Atlantic across two Early Eocene hyperthermals
2021, Geoscience FrontiersCitation Excerpt :We used a carbon cycle Grid Enabled Integrated Earth system model (cGENIE; Ridgwell, 2007; Cui et al., 2011; Vervoort et al., 2019) to calculate the 13C-depleted carbon release rate, amount and related paleoclimate response to the carbon cycle variability using the carbon isotopes of calcium carbonate from Site 1258 and the aforementioned age model. The initial atmospheric pCO2 was set at 834 ppm (~3 × preindustrial atmospheric level or PAL), within the range of proxy pCO2 estimate at ~54 Ma (Foster et al., 2017; Cui and Schubert, 2018), but smaller than the proxy pCO2 estimate of 1400 ppm at ~53 Ma by Anagnostou et al. (2016). This value was chosen in keeping with previous modeling study by Cui et al. (2011) (834 ppm), Gutjahr et al. (2017) (866 ppm), and Zeebe et al. (2009) (1000 ppm).