Blue skies, red-hot temps in Cretaceous

NASA-JSC-ES&IA

A reduction in cloud cover may have led to the "supergreenhouse" warming of the Cretaceous.

The Cretaceous (145 million to 65 million years ago) was one long heat wave. Average annual temperatures at the equator topped 38 degrees Celsius and even polar temperatures hovered near a temperate 10 degrees Celsius. Exactly how the planet got so warm is something that scientists have struggled to explain. Now, researchers say they may have the answer: A slight reduction in cloud cover that allowed more sunlight to reach Earth could have been the necessary nudge needed to push the planet into a "supergreenhouse" world.

Explaining the climate of the Cretaceous has been difficult because "supergreenhouse climates are warmer than what the models want them to be," says Lee Kump, a geologist at Pennsylvania State University in University Park. High carbon dioxide levels have been blamed, but the geological record indicates that carbon dioxide levels were no more than four times greater than modern levels during the Cretaceous — and at that level, climate models don't predict such high temperatures. So, either estimates of ancient carbon dioxide levels are wrong, Kump says, or something else contributed to the warming.

Kump and climate modeler David Pollard of Penn State decided to consider a factor that they say is often overlooked by paleoclimatologists: clouds. Clouds help keep Earth cool by reflecting sunlight back into space. But if something could have altered the planet's cloud cover during the Cretaceous, they thought, then perhaps it would explain the warming.

Based on the nature of how clouds form, the pair had reason to think changes in cloud patterns were likely at this time. Tiny particles or aerosols in the atmosphere called cloud condensation nuclei (CCN) serve as platforms where water vapor can condense into cloud droplets to form clouds. When CCNs are abundant, more cloud droplets form, making clouds denser and brighter, Kump says. When CCNs are scarce, fewer cloud droplets form, making clouds thinner. Droplets also tend to get larger, he adds, which ultimately causes fewer clouds because larger droplets force clouds to rain.

Today, CCNs are plentiful mainly due to the outpouring of pollution emitted into the atmosphere. But in the past, the main source of CCNs was likely biological. Algae in the oceans, for example, produce dimethylsulfide, which is transformed via photochemical reactions into acids that act as CCNs, Kump says. But when carbon dioxide levels began to rise during the Cretaceous — probably due to increased volcanic activity — higher ocean temperatures probably reduced the algae's productivity. That's because marine microorganisms get their nutrients from ocean upwellings, Kump says, which are diminished as temperatures rise.

Using the assumption that there was a decrease in biological productivity that would have led to fewer CCNs, Kump and Pollard modeled the Cretaceous climate and successfully recreated the period's supergreenhouse environment. "We got less bright clouds and fewer of them," Kump says. "It was a surprise, but both factors amplify warming." Cloud cover was reduced from 64 to 55 percent, leading to a 6 percent increase in the amount of sunlight that reached Earth, the researchers reported today in Science. The small change was enough to trigger a positive feedback system of warming that was necessary to reach supergreenhouse temperatures.

"It's an important feedback mechanism that we have been leaving out of our paleoclimate simulations," says Jeff Kiehl of the National Center for Atmospheric Research in Boulder, Colo. "The nice thing about their argument is that it's pretty firmly grounded in both observation and theory." But Karen Bice of the Woods Hole Oceanographic Institution in Woods Hole, Mass., questions whether biological productivity was actually limited during the Cretaceous. "The marine community didn't appear to be terribly stressed in response to warmer temperatures," she says.

Kump says that the next step is to investigate the geological record to look for evidence of changes in biological productivity and to conduct "more comprehensive modeling." But Kiehl already plans to investigate this feedback in a climate simulation of the Permian-Triassic Boundary some 251 million years ago. "We who model and try to understand different geological time periods and their climates need to consider the role of this cloud effect for paleoclimate simulations," he says.

Erin Wayman

Back to top