Record Shows Ancient Temperature Variations Coinciding With Shifts in the Planet’s Biodiversity

Trilobite Fossil Ordovician Strata

A finger points to a small trilobite fossil from the Ordovician strata in Svalbard, Norway. Credit: Adam Jost

Geologists Produce New Timeline of Earth’s Paleozoic Climate Changes

The temperature of a planet is linked with the diversity of life that it can support. MIT geologists have now reconstructed a timeline of the Earth’s temperature during the early Paleozoic era, between 510 and 440 million years ago — a pivotal period when animals became abundant in a previously microbe-dominated world.

In a study appearing today (February 1, 2021) in the Proceedings of the National Academy of Sciences, the researchers chart dips and peaks in the global temperature during the early Paleozoic. They report that these temperature variations coincide with the planet’s changing diversity of life: Warmer climates favored microbial life, whereas cooler temperatures allowed more diverse animals to flourish.

The new record, more detailed than previous timelines of this period, is based on the team’s analysis of carbonate muds — a common type of limestone that forms from carbonate-rich sediments deposited on the seafloor and compacted over hundreds of millions of years.

“Now that we have shown you can use these carbonate muds as climate records, that opens the door to looking back at this whole other part of Earth’s history where there are no fossils, when people don’t really know much about what the climate was,” says lead author Sam Goldberg, a graduate student in MIT’s Department of Earth, Atmospheric, and Planetary Sciences (EAPS).

Goldberg’s co-authors are Kristin Bergmann, the D. Reid Weedon, Jr. Career Development Professor in EAPS, along with Theodore Present of Caltech and Seth Finnegan of the University of California at Berkeley.

Microbial Buildup From the Early Ordovician Strata

In this photo, taken in western Newfoundland, Canada, you can see microbial buildup from the early Ordovician strata. Credit: Kristin Bergmann

Beyond fossils

To estimate Earth’s temperature many millions of years ago, scientists analyze fossils, in particular, remains of ancient shelled organisms that precipitated from seawater and either grew on or sank to the seafloor. When precipitation occurs, the temperature of the surrounding water can change the composition of the shells, altering the relative abundances of two isotopes of oxygen: oxygen-16, and oxygen-18.

“As an example, if carbonate precipitates at 4 degrees Celsius, more oxygen-18 ends up in the mineral, from the same starting composition of water, [compared to] carbonate precipitating at 30 degrees Celsius,” Bergmann explains. “So, the ratio of oxygen-18 to -16 increases as temperature cools.”

In this way, scientists have used ancient carbonate shells to backtrack the temperature of the surrounding seawater — an indicator of the Earth’s overall climate — at the time the shells first precipitated. But this approach has taken scientists only so far, up until the earliest fossils.

“There is about 4 billion years of Earth history where there were no shells, and so shells only give us the last chapter,” Goldberg says.

A clumped isotope signal

The same precipitating reaction in shells also occurs in carbonate mud. But geologists assumed the isotope balance in carbonate muds would be more vulnerable to chemical changes.

“People have often overlooked mud. They thought that if you try to use it as a temperature indicator, you might be looking at not the original ocean temperature in which it formed, but the temperature of a process that occurred later on, when the mud was buried a mile below the surface,” Goldberg says.

To see whether carbonate muds might preserve signatures of their original surrounding temperature, the team used “clumped isotope geochemistry,” a technique used in Bergmann’s lab, which analyzes sediments for clumping, or pairing, of two heavy isotopes: oxygen-18 and carbon-13. The likelihood of these isotopes pairing up in carbonate muds depends on temperature but is unaffected by the ocean chemistry in which the muds form.

Combining this analysis with traditional oxygen isotope measurements provides additional constraints on the conditions experienced by a sample between its original formation and the present. The team reasoned that this analysis could be a good indication of whether carbonate muds remained unchanged in composition since their formation. By extension, this could mean the oxygen-18 to -16 ratio in some muds accurately represents the original temperature at which the rocks formed, enabling their use as a climate record.

Ups and downs

The researchers tested their idea on samples of carbonate muds that they extracted from two sites, one in Svalbard, an archipelago in the Arctic Ocean, and the other in western Newfoundland. Both sites are known for their exposed rocks that date back to the early Paleozoic era.

In 2016 and 2017, teams traveled first to Svalbard, then Newfoundland, to collect samples of carbonate muds from layers of deposited sediment spanning a period of 70 million years, from the mid-Cambrian, when animals began to flourish on Earth, through the Ordovician periods of the Paleozoic era.

When they analyzed the samples for clumped isotopes, they found that many of the rocks had experienced little chemical change since their formation. They used this result to compile the rocks’ oxygen isotope ratios from 10 different early Paleozoic sites to calculate the temperatures at which the rocks formed. The temperatures calculated from most of these sites were similar to previously published lower-resolution fossil temperature records. In the end, they mapped a timeline of temperature during the early Paleozoic and compared this with the fossil record from that period, to show that temperature had a big effect on the diversity of life on the planet.

“We found that when it was warmer at the end of the Cambrian and early Ordovician, there was also a peak in microbial abundance,” Goldberg says. “From there it cooled off going into the middle to late Ordovician, when we see abundant animal fossils, before a substantial ice age ends the Ordovician. Previously people could only observe general trends using fossils. Because we used a material that’s very abundant, we could create a higher-resolution record and could see more clearly defined ups and downs.”

The team is now looking to analyze older muds, dating back before the appearance of animals, to gauge the Earth’s temperature changes prior to 540 million years ago.

“To go back beyond 540 million years ago, we have to grapple with carbonate muds, because they are really one of the few records we have to constrain climate in the distant past,” Bergmann says.

Reference: “A high-resolution record of early Paleozoic climate” by Samuel L. Goldberg, Theodore M. Present, Seth Finnegan, and Kristin D. Bergmann, 1 February 2021, Proceedings of the National Academy of Sciences.
DOI: 10.1073/pnas.2013083118

This research was supported, in part, by NASA and the David and Lucile Packard Foundation.

3 Comments on "Record Shows Ancient Temperature Variations Coinciding With Shifts in the Planet’s Biodiversity"

  1. “… this COULD mean the oxygen-18 to -16 ratio in SOME muds accurately represents the original temperature at which the rocks formed, …”

    It is well-known that plants discriminate between C12 and C13 in CO2, preferring the lighter isotope. What assurance do we have that there isn’t a similar biological mechanism that selects for one oxygen isotope over the other in calcifiers? If temperatures are calibrated against carbonates in the shells of calcifiers, might not carbonate muds need a different calibration curve if the carbonates in the muds are chemical precipitates rather than the product of biological activity?

    A good experiment would have compared the carbonate oxygen-isotope ratios in identifiable fossil shells against the bulk isotope ratios in the mud. I don’t see anything in the above synopsis to indicate that they did that.

    Where are the expert reviewers when you need them?

    • Torbjörn Larsson | February 3, 2021 at 5:10 pm | Reply

      A press release is not “a synopsis” but information of a paper and a lure for other scientists or interested public to take a look, often elaborated by the scientists with context not in the paper.

      Science peer review isn’t made on the press releases obviously – the scientists and universities where applicable are responsible for that – but on the paper that you didn’t read.

      I haven’t either, it is paywalled, but I note that the press release claims that “The likelihood of these isotopes pairing up in carbonate muds depends on temperature but is unaffected by the ocean chemistry in which the muds form.” Mind that the mud carbonates are precipitated from the ocean chemistry and not from fossils – they could correlate in trends, but they may not be identical in calibration.

      And they checked the trend with available fossil record trend, before suggesting that this can be extended to where there are no fossils.

      From the abstract:

      “Despite resetting of the clumped-isotope thermometer at both sites, our samples indicate relatively little change to their bulk δ18O due to low fluid exchange. Consequently, both sequences preserve temporal trends in δ18O. Motivated by this result, we compile a global suite of bulk rock δ18O data, stacking overlapping regional records to minimize diagenetic influences on overall trends. We find good agreement of bulk rock δ18O with brachiopod and conodont δ18O trends through time. Given evidence that the δ18O value of seawater has not evolved substantially through the Phanerozoic, we interpret this record as primarily reflecting changes in tropical, nearshore seawater temperatures and only moderately modified by diagenesis.”

      Else it would have been a good question.

      I can’t get to the paper metrics – and it is fairly young anyway – but it looks to me like it was a direct submission and reviewed as well as edited and commented for significance by Mark Thiemens:

      “Edited by Mark Thiemens, University of California San Diego, La Jolla, CA, and approved January 11, 2021 (received for review July 6, 2020)”

      “Mark Howard Thiemens (born January 6, 1950 in St. Louis, Missouri) is the Distinguished Professor and Chancellors Associates Chair in the Department of Chemistry and Biochemistry at the University of California San Diego.[1] He is best known for the discovery of a new physical chemical phenomena termed the mass independent isotope effect.[2]

      His studies have crossed a broad range of topics including basic physical and quantum chemistry, solar system origin, tracking the origin and evolution of life on early earth; stratospheric chemistry, climate change and greenhouse gas identification, Mars atmospheric chemistry, past and future and isotope geochemistry.”

      [ ]

      I don’t think you could find a better peer reviewer!

      • Torbjörn

        You remarked, “Science peer review isn’t made on the press releases obviously …” Fair enough, but the press releases should be accurate enough to be able to determine whether there are any serious problems, and whether interested persons would find it worth their while to peek behind the paywall.

        Yes, they made the point, and supported it, that diagenesis generally has negligible effect on the δ18O-16 ratios. However, the claim is made, “We find good agreement of bulk rock δ18O with brachiopod and conodont δ18O trends through time.” “Good agreement” isn’t defined. It isn’t specified whether that means that the slopes of the trends were identical, but there were different y-offsets for the lines, or whether the slopes of the lines were just similar with an intersection, or if some other combination of slope and intercept best describes the relationship. Probably the original article had graphs, but I’m not about to pay a lot of money to look at every “press release” I find interesting. In any event, the authors and the science editor could have been more precise about what was meant by “good agreement with trends.” Thus, it doesn’t really answer my original question about whether the mud ‘thermometer’ requires a different calibration curve or if they will be content to use the biogenic calibration curve, with expanded error bars.

        Would the best possible peer reviewer approve a publication that presented ambiguous terms and leave important questions unanswered?

Leave a comment

Email address is optional. If provided, your email will not be published or shared.