The blue line shows globally averaged surface air temperature since the last ice age, 24,000 years ago, created by assimilating paleoclimate records with a computer model of the climate system. Time is stretched for the past 1,000 years to visualize recent changes. Warming begins at the end of the last ice age, roughly 18,000 years ago, then temperatures stabilize. While previous studies showed a slight cooling over the past 10,000 years, the new analysis shows a slight warming trend. The curve steepens in recent decades with the accumulation of atmospheric greenhouse gases.Osman et al./Nature
A new effort to reconstruct Earth’s climate since the last ice age, about 24,000 years ago, highlights the main drivers of climate change, and how far out of bounds human activity has pushed the climate system.
The University of Arizona-led study uses a technique for reconstructing past temperatures developed by co-authors at the University of Washington. The study, published Nov. 10 in Nature, has three main findings:
- It verifies that the main drivers of climate change since the last ice age are rising greenhouse gas concentrations and the retreat of the ice sheets
- It suggests a general warming trend over the last 10,000 years – settling a decade-long debate in the paleoclimatology community about whether this period trended warmer or cooler
- The magnitude and rate of warming over the last 150 years far surpasses the magnitude and rate of changes at any other time over the last 24,000 years
“This reconstruction suggests that current temperatures are unprecedented in 24,000 years, and also suggests that the speed of human-caused global warming is faster than anything we’ve seen in that same time,” said senior author Jessica Tierney, an associate professor at the University of Arizona.
The UW team developed the method that allowed researchers to use computers to make sense of the paleoclimate data in marine sediments, allowing for more regional detail and precision in the temperature history.
“The fact that we’re today so far out of bounds of what we might consider normal is cause for alarm and should be surprising to everybody,” said lead study author Matthew Osman, a postdoctoral researcher at the University of Arizona.
An online search of “global temperature change since the last ice age” would produce a graph of global temperature change over time that was created eight years ago.
“Our team’s reconstruction improves on that curve by adding a spatial dimension,” Tierney said.
Different methods exist for reconstructing past temperatures. The team combined two independent datasets – temperature data from marine sediments and computer simulations of climate – to create a more complete picture of the past.
“Paleoclimate records provide the only record we have of these past climates, but these records are imperfect and they have gaps in space and time. Climate models provide simulations based on the laws of physics, but lack the observational record,” said co-author Gregory Hakim, a UW professor of atmospheric sciences. “Combining models and paleoclimate proxies – using the technique we developed for the Last Millennium Reanalysis – provides the best spatially complete estimate of the actual past climate, constrained by physics.”
The researchers looked at the chemical signatures of marine sediments to get information about past temperatures. Because temperature changes over time can affect the chemistry of a long-dead animal’s shell, paleoclimatologists can use those measurements to estimate temperature in an area. It’s not a perfect thermometer, but it’s a starting point.
Computer-simulated climate models, on the other hand, provide temperature information based on scientists’ best understanding of the physics of the climate system, which also isn’t perfect.
The team decided to combine the methods to harness the strengths of each. This is called data assimilation and is also commonly used in weather forecasting.
“To forecast the weather, meteorologists start with a model that reflects current weather, then add in observations such as temperature, pressure, humidity, wind direction and so on to create an updated forecast,” Tierney said.
The team applied this same idea to past climate.
“We found remarkable agreement between the output from our assimilation method and the values in independent paleoclimate records,” Hakim said. “This independent validation of our estimates for the past climate variability is an indication of how much the results can be trusted.”
The other co-authors are Robert Tardif at the UW; Jiang Zhu at the National Center for Atmospheric Research; Jonathan King at the University of Arizona; and Christopher Poulsen at the University of Michigan. The research was funded by the National Science Foundation, the Heising-Simons Foundation, and the National Center for Atmospheric Research.