Figures

top

Introduction

The evolution of life on Earth has been punctuated by extreme environmental/climatic events, often leading to mass extinctions (Chen and Benton, 2012

Chen, Z.-Q., Benton, M.J. (2012) The timing and pattern of biotic recovery following the end-Permian mass extinction. Nature Geoscience 5, 375–383.

; Rothman, 2017

Rothman, D.H. (2017) Thresholds of catastrophe in the Earth system. Science Advances 3, e1700906.

). Carbon cycle perturbations, global warming or cooling, and widespread anoxia have all played major roles in shaping the course of biotic macroevolution (Payne et al., 2020

Payne, J.L, Bachan, A, Heim, N.A, Hull, P.M, Knope, M.L. (2020) The evolution of complex life and the stabilization of the Earth system. Interface Focus 10, 20190106.

). However, there is no general model that encompasses the range of scenarios documented by the geological record. For example, in terms of global δ¹³C records, negative (e.g., end Permian) and positive (e.g., end Ordovician) excursions have both been linked to biotic extinctions (Fan et al., 2020

Fan, J.X., Shen, S.Z., Erwin, D.H., Sadler, P.M., Macleod, N., Cheng, Q.M., Hou, X.D., Yang, J., Wang, X.D., Wang, Y., Zhang, H., Chen, X., Li, G.X., Zhang, Y.C., Shi, Y.K., Yuan, D.X., Chen, Q., Zhang, L.N., Li, C., Zhao, Y.Y. (2020) A high-resolution summary of Cambrian to Early Triassic marine invertebrate biodiversity. Science 367, 272–277.

). Thus, while major biotic events were linked to intervals of environmental perturbation and ecologic disturbance, they were not universally driven by one particular environmental process. However, there are no reliable proxies for measuring the extent of carbon cycle instability and associated environmental stress through Earth history.

Here, we develop a carbon cycle turbulence index (CT_index) to measure perturbations in the δ¹³C_carb record (Cramer and Jarvis, 2020

Cramer, B.D., Jarvis, I. (2020) Chapter 11 - Carbon Isotope Stratigraphy. In: Gradstein, F.M., Ogg, J.G., Schmitz, M.D., Ogg, G.M. (Eds.) The Geologic Time Scale 2020. Elsevier, Amsterdam, 309–343. https://doi.org/10.1016/B978-0-12-824360-2.00011-5

) over the last 1,000 Myr. The CT_index emphasises perturbations to the carbon cycle as a metric for ecosystem instability. The rationale for using δ¹³C_carb perturbation is because the δ¹³C_carb record is robust and at high resolution, and short term (0.5–10 Myr) δ¹³C_carb variability over the past 1,000 Myr has been suggested to record ocean-atmosphere system instability (Rothman, 2017

Rothman, D.H. (2017) Thresholds of catastrophe in the Earth system. Science Advances 3, e1700906.

). The CT_index is calculated according to Equation 1:

To measure the degree of carbon cycle turbulence at each point (CT_index (t)), over a pre-determined duration (e.g., τ_span = 0.05 Myr), we compiled all the datum points within the τ_span (centred on the calculating point, t) in array F. In this case, N and μ(t) represent the number and average isotopic value of the array F, respectively, where n is a statistical control to avoid a situation where the sum of the mean deviations is zero. When n = 2, the output of the calculation, CT_index (t), is equal to the standard deviation of the array F (see sensitivity tests for parameters τ_span and n in Supplementary Information). Thus, major positive and negative excursions will both result in a CT_index peak.

Phanerozoic biotic extinction rates (ERs) at the generic level were derived from a palaeobiology database (PaleoDB), and are applied to represent biodiversity changes and ecologic disturbance for comparison with CT_index values (Fig. 1). The fossil record often documents a direct link between enhanced ERs and dramatic reductions in biodiversity, or major ecologic disturbance, during mass extinctions (Alroy et al., 2008

Alroy, J., Aberhan, M., Bottjer, D., Foote, M., Fursich, F.T., Harries, P.J., Hendy, A.J.W., Holland, S.M., Lvany, L.C., Kiessling, W., Kosnik, M.A., Marshall, C.R., McGowan, A.J., Miller, A.I., Olszewski, T.D., Patzkowsky, M.E., Peters, S.E., Villier, L., Wagner, P.J., Bonuso, N., Borkow, P.S., Brenneis, B., Clapham, M.E., Fall, L.M., Ferguson, C.A., Hanson, V.L., Krug, A.Z., Layou, K.M., Leckey, E.H., Nurnberg, S., Powers, C.M., Sessa, J.A., Simpson, C., Tomasovych, A., Visaggi, C.C. (2008) Phanerozoic Trends in the Global Diversity of Marine Invertebrates. Science 321, 97–100.

). Hence, relationships between CT_index peaks and higher ER values can be employed to investigate coincidence between catastrophic environmental stress and biotic crises through Earth history. Moreover, increasing evidence suggests that we are in the midst of a “sixth mass extinction” (Barnosky et al., 2011

Barnosky, A.D., Matzke, N., Tomiya, S., Wogan, G.O.U., Swartz, B., Quental, T.B., Marshall, C., McGuire, J.L., Lindsey, E., Maguire, K.C., Mersey, B., Ferrer, E.A. (2011) Has the Earth’s sixth mass extinction already arrived? Nature 471, 51–57.

; Ceballos et al., 2015

Ceballos, G., Ehrlich, P.R., Barnosky, A.D., García, A., Pringle, R.M., Palmer, T.M. (2015) Accelerated modern human–induced species losses: Entering the sixth mass extinction. Science Advances 1, e1400253.

), and a pronounced negative excursion in δ¹³CO₂ caused by greenhouse gas emissions has been recorded since the onset of the industrial revolution (Graven et al., 2020

Graven, H., Keeling, R.F., Rogelj, J. (2020) Changes to Carbon Isotopes in Atmospheric CO₂ Over the Industrial Era and Into the Future. Global Biogeochemical Cycles 34, e2019GB006170.

). In order to investigate the current and potential future significance of rising greenhouse gas emissions on biotic ERs, we compare the present day CT_index value to that observed during the Permian/Triassic (PTr) transition, representing the most severe mass extinction of the Phanerozoic.

Full size image

top

Results and Discussion

CT_index vs. environmental perturbations. Our reconstruction of the CT_index through time (Fig. 1; analytical methods and database details are given in the Supplementary Information) highlights that it is both the magnitude and duration of the carbon cycle perturbation that exhibits a major link to the extent of environmental and biotic change. The CT_index offers a unique measure of the intensity of δ¹³C_carb fluctuations, which reflect carbon cycle perturbation and the dynamic balance between the input of carbon and its burial as carbonate and organic matter (Kump and Arthur, 1999

Kump, L.R., Arthur, M.A. (1999) Interpreting carbon-isotope excursions: carbonates and organic matter. Chemical Geology 161, 181–198.

). As such, the CT_index peaks during the late Tonian, through the Cryogenian and Ediacaran glaciations, and into the Cambrian (Fig. 1), broadly correlate with periods of environmental oxygenation (Canfield et al., 2007

Canfield, D.E., Poulton, S.W., Narbonne, G.M. (2007) Late-Neoproterozoic Deep-Ocean Oxygenation and the Rise of Animal Life. Science, New Series 315, 92–95.

; Lyons et al., 2021

Lyons, T.W., Diamond, C.W., Planavsky, N.J., Reinhard, C.T., Li, C. (2021) Oxygenation, Life, and the Planetary System during Earth’s Middle History: An Overview. Astrobiology 21, 906–923.

). The major variability evident in CT_index values (in both magnitude and frequency) reflects well documented instability in the extent of Earth surface oxygenation over this interval (Canfield et al., 2008

Canfield, D.E., Poulton, S.W., Knoll, A.H., Narbonne, G.M., Ross, G., Goldberg, T., Strauss, H. (2008) Ferruginous Conditions Dominated Later Neoproterozoic Deep-Water Chemistry. Science 321, 949–952.

; Sahoo et al., 2016

Sahoo, S.K., Planavsky, N.J., Jiang, G., Kendall, B., Owens, J.D., Wang, X., Shi, X., Anbar, A.D., Lyons, T.W. (2016) Oceanic oxygenation events in the anoxic Ediacaran ocean. Geobiology 14, 457–468.

; He et al., 2019

He, T.C., Zhu, M.Y., Mills, B.J.W., Wynn, P.M., Zhuravlev, A.Y., Tostevin, R., Pogge Von Strandmann, P.A.E., Yang, A.H., Poulton, S.W., Shields, G.A. (2019) Possible links between extreme oxygen perturbations and the Cambrian radiation of animals. Nature Geoscience 12, 468–474.

; Li et al., 2020

Li, Z., Cao, M., Loyd, S.J., Algeo, T.J., Zhao, H., Wang, X., Zhao, L., Chen, Z.-Q. (2020) Transient and stepwise ocean oxygenation during the late Ediacaran Shuram Excursion: Insights from carbonate δ238U of northwestern Mexico. Precambrian Research 344, 105741.

).

As the Earth progressed to near modern levels of oxygenation in the Palaeozoic (Krause et al., 2018

Krause, A.J., Mills, B.J.W., Zhang, S., Planavsky, N.J., Lenton, T.M., Poulton, S.W. (2018) Stepwise oxygenation of the Paleozoic atmosphere. Nature Communications 9, 4081.

) and life continued to evolve and diversify (Lenton et al., 2016

Lenton, T.M., Dahl, T.W., Daines, S.J., Mills, B.J.W., Ozaki, K., Saltzman, M.R., Porada, P. (2016) Earliest land plants created modern levels of atmospheric oxygen. PNAS 113, 9704–9709.

), the ensuing environmental stability generally resulted in lower peaks in CT_index values, with the exception of the major peak across the PTr boundary (Fig. 1). However, despite the general decrease, the CT_index trend is of sufficient resolution (see Fig. S-4) to distinguish nearly all of the major individual carbon cycle perturbations during the Phanerozoic (Fig. 1). Thus, although there are many potential controls on the δ¹³C_carb record (Lenton et al., 2018

Lenton, T.M., Daines, S.J. (2018) The effects of marine eukaryote evolution on phosphorus, carbon and oxygen cycling across the Proterozoic–Phanerozoic transition. Emerging Topics in Life Sciences 2, 267–278.

), the CT_index provides an integrated and robust assessment of the overall extent of environmental perturbation.

Deciphering extinction rates and environmental perturbations. The CT_index trends correlate with biodiversity fluctuations, ERs and ecologic disturbance through the Neoproterozoic and Phanerozoic (Fig. 1). Several important intervals of biodiversification, including the initial evolution of the Ediacaran biota, the Cambrian explosion, the Great Ordovician Biodiversification, the Carboniferous-Permian, and the mid-Triassic and Jurassic-Cretaceous radiations (Sepkoski, 1981

Sepkoski, J.J. (1981) A factor analytic description of the Phanerozoic marine fossil record. Paleobiology 7, 36–53.

; Alroy et al., 2008

Alroy, J., Aberhan, M., Bottjer, D., Foote, M., Fursich, F.T., Harries, P.J., Hendy, A.J.W., Holland, S.M., Lvany, L.C., Kiessling, W., Kosnik, M.A., Marshall, C.R., McGowan, A.J., Miller, A.I., Olszewski, T.D., Patzkowsky, M.E., Peters, S.E., Villier, L., Wagner, P.J., Bonuso, N., Borkow, P.S., Brenneis, B., Clapham, M.E., Fall, L.M., Ferguson, C.A., Hanson, V.L., Krug, A.Z., Layou, K.M., Leckey, E.H., Nurnberg, S., Powers, C.M., Sessa, J.A., Simpson, C., Tomasovych, A., Visaggi, C.C. (2008) Phanerozoic Trends in the Global Diversity of Marine Invertebrates. Science 321, 97–100.

; Erwin et al., 2011

Erwin, D.H., Laflamme, M., Tweedt, S.M., Sperling, E.A., Pisani, D., Peterson, K.J. (2011) The Cambrian Conundrum: Early Divergence and Later Ecological Success in the Early History of Animals. Science 334, 1091–1097.

; Darroch et al., 2018

Darroch, S.A.F., Smith, E.F., Laflamme, M., Erwin, D.H. (2018) Ediacaran Extinction and Cambrian Explosion. Trends in Ecology & Evolution 33, 653–663.

; Fan et al., 2020

Fan, J.X., Shen, S.Z., Erwin, D.H., Sadler, P.M., Macleod, N., Cheng, Q.M., Hou, X.D., Yang, J., Wang, X.D., Wang, Y., Zhang, H., Chen, X., Li, G.X., Zhang, Y.C., Shi, Y.K., Yuan, D.X., Chen, Q., Zhang, L.N., Li, C., Zhao, Y.Y. (2020) A high-resolution summary of Cambrian to Early Triassic marine invertebrate biodiversity. Science 367, 272–277.

), all coincide with low CT_index values. By contrast, pronounced CT_index peaks correspond to higher ERs, as evident during mass extinctions (Fig. 1b).

However, since the datum sampling densities of δ¹³C_carb database and PaleoDB are clearly different, it is difficult to directly calculate the regression relationship between the CT_index and ERs. To address this, we begin by identifying significant global perturbation events during the Phanerozoic (Fig. S-4, Table S-2). These 41 events comprise the major δ¹³C_carb perturbations (Rothman, 2017

Rothman, D.H. (2017) Thresholds of catastrophe in the Earth system. Science Advances 3, e1700906.

), mass extinctions (Bambach, 2006

Bambach, R.K. (2006) Phanerozoic Biodiversity Mass Extinctions. Annual Review of Earth and Planetary Sciences 34, 127–155.

), and biodiversifications (Fan et al., 2020

Fan, J.X., Shen, S.Z., Erwin, D.H., Sadler, P.M., Macleod, N., Cheng, Q.M., Hou, X.D., Yang, J., Wang, X.D., Wang, Y., Zhang, H., Chen, X., Li, G.X., Zhang, Y.C., Shi, Y.K., Yuan, D.X., Chen, Q., Zhang, L.N., Li, C., Zhao, Y.Y. (2020) A high-resolution summary of Cambrian to Early Triassic marine invertebrate biodiversity. Science 367, 272–277.

). While the CT_index trend reflects the degree of environmental perturbation on the timescale of individual events (Fig. S-3), consideration needs to be given to the method for calculating ER ratios. Kocsis et al. (2019)

Kocsis, A.T., Reddin, C.J., Alroy, J., Kiessling, W. (2019) The r package divDyn for quantifying diversity dynamics using fossil sampling data. Methods in Ecology and Evolution 10, 735–743.

explored 12 different metrics to calculate ER values (Fig. 1b), which also consider the potential errors caused by the PaleoDB itself. For each stage, we use the maximum, minimum and ‘sqs2f3’ methods (Kocsis et al., 2019

Kocsis, A.T., Reddin, C.J., Alroy, J., Kiessling, W. (2019) The r package divDyn for quantifying diversity dynamics using fossil sampling data. Methods in Ecology and Evolution 10, 735–743.

) to establish a broad range for the calculation of ER values (Fig. 2, Fig. S-5). Then, each major CT_index peak is considered responsible for the corresponding ER perturbation.

Full size image

Cross plots show that the CT_index and ER values for 31 of these events are broadly positively correlated (red points; Fig. 2), with R² = 0.835 and p < 0.001. The remaining 10 events (blue points; Fig. 2) comprise two types of outlier: (1) events with high ER and low CT_index values, occurring from 541 to 480 Ma (from the Cambrian to the first stage of the Ordovician), and (2) events with high CT_index and low ER values, occurring in Silurian. Given that the dataset for these latter two intervals is smaller than that of other periods of the Phanerozoic (Fig. S-4a), these outliers may result from the incompleteness of the PaleoDB. However, a fundamental difference in ecosystem stability for these periods, relative to the rest of the Phanerozoic, cannot be ruled out. Nevertheless, the CT_index suggests that environmental turbulence was intense during the early Cambrian (but not the middle to late Cambrian) and through the entire Silurian. While the high ER and low CT_index signal for 541–480 Ma may represent an unstable ecosystem, the high CT_index and low ER signal for the Silurian reflects a more stable ecosystem that did not suffer a major extinction.

Next, in order to explore the potential thresholds of the mass extinctions, we use a logistic curve (sigmoid curve) to fit the data (Fig. 2b), which is a widely used approach for deep time extinction studies (Roopnarine, 2006

Roopnarine, P.D. (2006) Extinction cascades and catastrophe in ancient food webs. Paleobiology 32, 1–19.

). Logistic curves represent the battle between ecosystem stability and environmental instability. Ecosystems are generally unaffected at low levels of volatility, but when the degree of volatility increases to a certain threshold, the biotic ER rapidly increases (e.g., the PTr and TJ mass extinctions).

In addition to distinguishing these well studied events, the logistic curve also provides new insight into broader trends through time. Here we note that a previous study of major δ¹³C_carb events suggests that the thresholds of catastrophe in the Earth system have not changed systematically over time (Rothman, 2017

Rothman, D.H. (2017) Thresholds of catastrophe in the Earth system. Science Advances 3, e1700906.

). Building on this, our data implicitly suggest that, although life has evolved into more complex forms, ecosystem tolerance to environmental turbulence has not significantly improved through time (discounting the Cambrian and Silurian; Fig. S-5c). This observation naturally leads to consideration of whether the CT_index may be used to evaluate the possible nature of Earth’s “sixth mass extinction” in the modern and future environment.

Contrasting the CT_index during the PTr and from 1700 to 2100 A.D. Current high extinction rates of some taxonomic groups suggest that the Earth may be entering a “sixth mass extinction”, which may have a causal relationship with increasing greenhouse gases in the atmosphere (Barnosky et al., 2011

Barnosky, A.D., Matzke, N., Tomiya, S., Wogan, G.O.U., Swartz, B., Quental, T.B., Marshall, C., McGuire, J.L., Lindsey, E., Maguire, K.C., Mersey, B., Ferrer, E.A. (2011) Has the Earth’s sixth mass extinction already arrived? Nature 471, 51–57.

; Ceballos et al., 2015

Ceballos, G., Ehrlich, P.R., Barnosky, A.D., García, A., Pringle, R.M., Palmer, T.M. (2015) Accelerated modern human–induced species losses: Entering the sixth mass extinction. Science Advances 1, e1400253.

). However, there is still a widely recognised “last kilometre gap” between the fossil record and the modern bio-record and their extinction rates, for the following reasons: (1) most modern data focus on terrestrial vertebrates (e.g., mammals and birds), whereas the fossil record dominantly reflects marine invertebrates (e.g., brachiopods, gastropods and bivalves) (Kocsis et al., 2019

Kocsis, A.T., Reddin, C.J., Alroy, J., Kiessling, W. (2019) The r package divDyn for quantifying diversity dynamics using fossil sampling data. Methods in Ecology and Evolution 10, 735–743.

), and (2) there are different time spans when considering a particular extinction. Specifically, the “Big Five” Phanerozoic mass extinctions are only constrained across a broad time resolution of millions of years, so a comparison with modern events occurring on a timescale of a hundred to a thousand years is not straightforward, and (3) there is uncertainty in the near future. For example, calculation of ERs for modern species is hampered since it remains unknown whether “critically endangered species” will actually go extinct (Barnosky et al., 2011

Barnosky, A.D., Matzke, N., Tomiya, S., Wogan, G.O.U., Swartz, B., Quental, T.B., Marshall, C., McGuire, J.L., Lindsey, E., Maguire, K.C., Mersey, B., Ferrer, E.A. (2011) Has the Earth’s sixth mass extinction already arrived? Nature 471, 51–57.

). (4) Other unknowns, including continental configuration, pre-existing climate state, and the extant biota at the time of CT_index perturbation might also affect how the ER responds.

To address these issues, we note that if the core hypothesis of the CT_index stands, then the sixth mass extinction should be accompanied by a distinct CT_index peak. We thus reconstruct CT_index values from 1700 to 2100 A.D., representing the time period prior to, and after, the industrial revolution, and we contrast this with the PTr transition at ∼252 Ma, representing the most severe mass extinction of the Phanerozoic. We used δ¹³CO₂ curves from ice core to reconstruct the modern CT_index trend due to its complete and robust record over the last several centuries, in addition to the availability of modelled near future values (Rubino et al., 2013

Rubino, M., Etheridge, D.M., Trudinger, C.M., Allison, C.E., Battle, M.O., Langenfelds, R.L., Steele, L.P., Curran, M., Bender, M., White, J.W.C., Jenk, T.M., Blunier, T., Francey, R.T. (2013) A revised 1000 year atmospheric δ13C‐CO₂ record from Law Dome and South Pole, Antarctica. Journal of Geophysical Research: Atmospheres 118, 8482–8499.

; Graven et al., 2020

Graven, H., Keeling, R.F., Rogelj, J. (2020) Changes to Carbon Isotopes in Atmospheric CO₂ Over the Industrial Era and Into the Future. Global Biogeochemical Cycles 34, e2019GB006170.

). However, we note that other carbon hosts, including surface ocean dissolved inorganic carbon, coral samples, and CH₄ in Antarctica ice core also sensitively record significant CT_index peaks (Fig. S-7). This approach is reinforced by an observed strong similarity in the CT_index output when calculated using the δ¹³CO₂ and δ¹³C coral records (Fig. S-7). To further enhance the reliability of our comparison, we focus on a very short interval during the PTr transition, between Bed 24e to Bed 25 in the Meishan GSSP section, which documents the extinction level itself (Burgess et al., 2014

Burgess, S.D., Bowring, S., Shen, S. (2014) High-precision timeline for Earth’s most severe extinction. PNAS 111, 3316–3321.

). This interval has 11 original δ¹³C_carb datum points across an estimated interval of ∼3,000 years. Hence, we calculate the CT_index for both the PTr and the industrial revolution period using the same parameters (n = 1; τ_span = 100 years), and the datasets were pre-normalised to a datum density of one point per year.

The relative size of the CT_index peaks calculated under the same conditions for the two time intervals is of practical comparative significance (Fig. 3). The CT_index peak during the PTr ranges from 0.47 to 0.78, whilst the CT_index peak of the industrial revolution period ranges from 0.61 to 1.61, with the latter reflecting the predicted range for the δ¹³CO₂ trend into the near future. If fossil fuel emissions peak at the present level and are at a low level through the next century, it has been suggested (Graven et al., 2020

Graven, H., Keeling, R.F., Rogelj, J. (2020) Changes to Carbon Isotopes in Atmospheric CO₂ Over the Industrial Era and Into the Future. Global Biogeochemical Cycles 34, e2019GB006170.

) that δ¹³CO₂ may recover from −8.8 ‰ to 7.0 ‰, which leads to a moderate CT_index peak (line SSP1-1.9; Fig. 3). However, in other scenarios without such effective limits on fossil fuel emissions, δ¹³CO₂ will likely continue to decrease through the next century (line SSP8.5), causing a dramatic peak in the CT_index that may be even greater than the PTr CT_index peak (Fig. 3). While the current extinction of species is driven by biodiversity loss that is only partly associated with fossil fuel driven climate change, the empirical relationship between ecosystem stability and biotic ERs implies that, if fossil fuel emissions are not significantly decreased, Earth’s “sixth mass extinction” has the potential to be severe.

Full size image

top

Acknowledgements

The contributors to the PaleoDB database and Geologic Time Scale are thanked for their provision of taxonomic and carbon isotope data for the database. We are grateful to Gavin Foster and two anonymous reviewers for their comments, which greatly improved this manuscript. This study was supported by five NSFC grants (41821001, 41930322, 41977264, 42073002, 92055212), and by 111 Project of China (BP0820004). ZL is funded by the Fundamental Research Funds for National Universities, China University of Geosciences (Wuhan).

Editor: Gavin Foster

top

References

top

Supplementary Information

The Supplementary Information includes:

• The Geochemical and Paleobiology Databases
• Analytical Approaches and Statistical Analysis
• Datum Density vs. Bias
• Comparison of Carbon Cycle Perturbations and CT_index Values between the Permian-Triassic Mass Extinction and the Industrial Revolution Period of the Anthropocene
• Tables S-1 and S-2
• Figures S-1 to S-8
• Supplementary Information References

Download the Supplementary Information (PDF).