A record of the Earth’s temperature covering half a billion years

Global temperature records go back less than two centuries. But that doesn't mean we have no idea what the world was doing before we started building thermometers. There are various things—tree rings, isotope ratios, and more—that register temperatures in the past. Using these temperature proxies, we've managed to reconstruct thousands of years of our planet's climate.

But going back further is difficult. Fewer proxies get preserved over longer times, and samples get rarer. By the time we go back past a million years, it's difficult to find enough proxies from around the globe and the same time period to reconstruct a global temperature. There are a few exceptions, like the Paleocene-Eocene Thermal Maximum (PETM), a burst of sudden warming about 55 million years ago, but few events that old are nearly as well understood.

Now, researchers have used a combination of proxy records and climate models to reconstruct the Earth's climate for the last half-billion years, providing a global record of temperatures stretching all the way back to near the Cambrian explosion of complex life. The record shows that, with one apparent exception, carbon dioxide and global temperatures have been tightly linked. Which is somewhat surprising, given the other changes the Earth has experienced over this time.

Past climates

The work done here by an international team involves a combination of proxy data and climate models. While there are a number of land-based proxies, they tend to come with very large uncertainties. So, the researchers focused on a single type of proxy: the ratio of oxygen isotopes found in the shells of sea organisms. There are some questions regarding the accuracy of these, as using them requires that the ratio of these isotopes in the oceans has remained constant over time.

To compensate for that, the researchers used two methods of converting these proxies into temperatures. One method assumed that oxygen isotope ratios in seawater have remained constant; the second method used a slow, constant change over the time period covered.

Climate models provide a way of converting these proxies, which typically come from a single geographic location, to a global temperature. By using details like the continental configuration and carbon dioxide levels, the models can estimate which reasonable global temperatures are consistent with the proxy data, meaning a specific temperature at a specific location on the globe. The researchers used an ensemble of climate models so that the results weren't dependent on any particular implementation of atmospheric physics.

The results, which the researchers call PhanDA, estimate global temperatures over the last 485 million years, going back to the end of the Cambrian, the period that saw the diversification of the major groups of present-day animal life.

So, what does PhanDA look like? One key feature is that it overlaps with the Cenozoic, which started with the mass extinction that ended all non-avian dinosaur lineages. We've got a better history of the Cenozoic climates, so these provide an important test of whether PhanDA's temperatures match those obtained independently. The consistency between them is an important validation of the new work.

Overall, the researchers find that the global mean temperature has likely varied from a low of about 11° C, seen in the recent glacial periods, up to a high of 36° C, seen about 90 million years ago, though similar extremes were seen during the PETM. Other major climate events, such as the warming produced in the wake of the eruptions that formed the Siberian Traps, showed up in the record. There are both long periods of warming trends (such as one that covered most of the Mesozoic) alternating with cooling (which has dominated the present Cenozoic). The researchers suggest these are driven by the assembly and breakup of supercontinents.

More of this period was spent in warm greenhouse climates (41 percent of the period) than in icehouse climates (31 percent). The researchers found that most of the difference between these climates occur in the polar regions. Changes do occur in the tropics, but they're considerably smaller in magnitude. So, during an icehouse period, the difference between equatorial regions and high latitudes is on the order of 30° to 50° C. By contrast, during hothouse periods, the equator-to-pole difference tended to be on the order of 15° to 25° C.

Heating the globe

One thing that is clear from comparing this record with carbon dioxide is that there's a close correlation between the two. There are some exceptions, but the two tend to move in parallel throughout this entire period. The big exception is in the Cretaceous (a period dominated by dinosaurs), which saw a hothouse climate develop, while carbon dioxide levels appeared to remain flat. We've known about this discrepancy for a while but don't have a good explanation for it; the new research doesn't really change that situation.

The strong correlation between carbon dioxide and temperatures is also somewhat surprising because we know there's been an additional significant climate influence over this period: the Sun. As the Sun has been warming up as it ages, the amount of energy reaching the Earth from the Sun should have increased by 4.2 percent over the period covered by this study. So, there should be an additional warming trend superimposed over the fluctuations in carbon dioxide. But the study doesn't see it.

The researchers suggest a couple of other factors that might offset this trend. For example, the fraction of the planet covered by oceans has trended downward during this same period, although there have been significant fluctuations over time. Since the oceans absorb more sunlight, that may partly offset the Sun heating up. Similarly, the ecosystems present early in the period covered by this study might have produced more methane, a potent greenhouse gas. So, in contrast to the confusion about the Cretaceous hothouse, we seem to have a lot of potential explanations for this oddity.

Two findings in this study may have implications for our future climate. The first is a measure of the long-term sensitivity of the climate to a doubling of carbon dioxide. They come up with a value of about 8° C of warming for the doubling. And that's... a lot. It's important to remember that this is a separate measure from the IPCC's estimation of the climate's sensitivity to CO₂ (which is 2–5° C). Here, any slow-acting feedbacks have millions of years to influence the climate. And this work's estimate also includes the unknown influences described above (other greenhouse gases, solar changes, etc.).

Beyond all the information produced by this work, the most important thing is that the researchers made it easy to build on. All the data and software have been made available to the research community, and the software has been structured to make it easy to incorporate additional proxy data into the analysis. So, as researchers get additional estimates of past temperatures from anywhere around the globe, they can be plugged into this work to potentially refine the picture painted by this one paper.

Science, 2024. DOI: 10.1126/science.adk3705  (About DOIs).