Mass extinctions aren’t fun times. There’s a reason (usually more than one, actually) species disappear in droves. That makes untangling these reasons a challenge. The geological crime scene investigation is tough given that clues can be elusive after millions of years, and the events are complex.
The extinction that wiped out (most of) the dinosaurs, for example, saw both a massive asteroid impact and long-lived volcanic eruptions that covered most of what is now India in lava flows.
The record of that warming in the geologic record isn’t very good, though. The problem has been to find a suitable climate record in rocks that were deposited fast enough to show relatively short time periods in detail. To obtain that sort of record, a team led by the University of Missouri’s Kenneth MacLeod scratched through rocks in Tunisia for crushed up pieces of fossil fish bits.
The researchers tick both boxes—good climate record and lots detail—by turning to Tunisian rocks with fish bones. The rocks there that date to the time of the end-Cretaceous mass extinction have been well studied, with the ages and characteristics of various layers nailed down pretty precisely. At the time of the extinction event 66 million years ago, the sediments in these rocks were being deposited in a shallow ocean in the tropics. Many fossils of critters made of calcium carbonate were altered over time, making them unreliable sources of information. But these fish bits are made of more resilient calcium phosphate.
It’s the oxygen atom in that calcium phosphate that holds the secrets—just as it does in the H2O of glacial ice or the calcium carbonate of other critters. The ratio of heavier oxygen-18 to lighter oxygen-16 isotopes changes with temperature. So as you go down an ice core, or down through a layer of rock, shifts in this isotope ratio mark warming or cooling.
In this case, the researchers made measurements across 9 meters of rock, spanning about 350,000 years of Earth history. At 40 points in time, they collected dozens of tiny fish fossil pieces to analyze for oxygen isotopes. The results show a change at the time of Chicxulub asteroid impact that lasted 100,000 years.
The oxygen isotope ratio shifted by an amount that represents water that was about 5°C warmer than it had been. After about 100,000 years of that, temperatures return to roughly where they were before the impact—although the samples are a little too sparse here to say more than that.
You’d expect the tropical ocean to warm less than the global average, so this represents more than 5°C global warming. (For comparison, unconstrained greenhouse gas emissions could bring human-caused global warming to about 5°C by the year 2100.) This lines up with some estimates of the expected warming. A 2002 study that estimated the change in CO2 from plant fossils, for example, put the average global warming in the ballpark of 7.5°C. The most interesting bit here is seeing evidence of how long that warmer climate lasted—100,000 years also fits in with climate model simulations of this time. So while the world changed in the blink of an eye, the misery took its time in leaving.
All the estimates of the drivers of short-term cooling and long-term warming all have pretty large error bars still. It’s data like this that helps reduce some of that uncertainty—one tiny little piece of crushed fish fossil at a time.