Is global warming's 'hiatus' really just a statistical error?

In a changing climate, it's easy to understand the impulse to seek out good news. But some scientists may have been too hopeful.

Many climate scientists have observed that, since 1998 or so, the rise of global temperatures had stopped, or at least slowed down, in what has been variously called climate change's "pause" or "hiatus."

But, there was no global warming "hiatus," say Stanford scientists.

The pause was actually a statistical error, according to a new study published Thursday in the journal Climatic Change.

“The alleged or purported hiatus in the warming of the global climate system does not have a sound scientific basis,” says study lead author Bala Rajaratnam, an assistant professor of statistics and Earth system science at Stanford University.

Dr. Rajaratnam and his team examined all the data published relating to a global warming pause. “Using the language of statistics and the discipline of mathematics, we tested to see if these claims could stand up to statistical scrutiny,” he says.

It did not. The researchers found no statistical evidence that the rate of global warming stopped or slowed down between 1998 and 2013, as previously thought.

Climate change skeptics had hailed the supposed hiatus as a sign that global warming had ended, or that the warming was just part of natural variances. 

Climate change researchers had been baffled by the hiatus, as it contradicted models predicting continued global warming.

But Rajaratnam says his work restores confidence in those climate projections. The temperatures that initially caught scientists’ attention as a potential climate change pause are actually within expected ranges for variability, he says. 

The existing statistical methods weren't appropriate to handle the data related to the purported hiatus, which was one of the problems with previous studies, says Rajaratnam. The Stanford researchers had to create new methods to thoroughly examine the statistical question.

One reason Rajaratnam and his team had to reconsider statistical methods was that the classical techniques needed more data points than the 15 year period provided. 

Furthermore, the team had to consider the relationship of one temperature data point to the ones that were taken leading up to that instant. For example, one hot day might be part of a heat wave, and those influences must be considered. The researchers maintained those links in their models.

“What is so different about this paper from other papers? We actually have environmental scientists and trained statisticians on board,” Rajaratnam says. By combining the fields, the researchers could better consider the nuances of the question. “That is really critical.”

Other scientists have questioned the warming “hiatus” recently as well. 

Researchers wondered where the heat could be disappearing to, as humans were still producing global warming-causing greenhouse gases. In 2014 they found that oceans might be heating up. While air temperatures weren't rising, scientists reasoned, the warming may have continued deep in the ocean.

The data itself may have been wrong too. The National Oceanic and Atmospheric Administration (NOAA) reported in June that the recording tools may have generated inaccurate data. When NOAA scientists accounted for that error, they found that global warming did not slow down between 1998 and 2013.

The Stanford researchers used both NOAA’s adjusted data and the original data in their statistical analysis.

“To say that the warming has slowed down or has stalled, it is not supported scientifically at all,” says Rajaratnam. 

Even if there had been a hiatus, it might be over now. The Met Office, the UK government agency that studies global weather, found that temperatures in 2014 and 2015 broke records, and that the warmer temperatures are likely to continue.

Rajaratnam hopes researchers use his work as a tool to better examine climate trends. 

“We have to be a little more responsible when making such statements,” Rajaratnam says about the claim of a global warming pause. Before making such a contentious claim, he says, researchers need to make sure “it’s really backed up by rigorous statistical and scientific analysis.”

Rajaratnam is confident in his team’s results.

“I challenge anyone to try and see if they can invalidate them,” he says.