Over the last 10 years, global temperatures rose about 25 percent more slowly than scientists had previously predicted. A study published in the journal Science reports that that's because the climate models factored in increases in carbon dioxide and other greenhouse gases but didn’t count on a decrease in water vapor — the most abundant greenhouse gas — in the middle part of the atmosphere.
Researchers say the results are nevertheless consistent with what’s been known for years about climate change: that the planet has gotten warmer and most of that warming is anthropogenic.
Decadal average temperatures in the 1990s and 2000s stayed on trend — the last decade was the warmest decade on record — but there’s been a flattening of the trend from the warm years, says National Oceanic and Atmospheric Administration scientist Robert Portmann, one of the study’s authors, and this study helps explain why.
“What we expect to see is a trend in temperatures with variability on top of that,” he says. “The debate is: how much of this is due to variability? I believe that the last 10 years is due to variability.”
The study actually reveals a variability in the climate system that the models had failed to capture: the variability of water in the lower stratosphere.
“That variability is the new thing here,” Portmann says. “It was a bit of a surprise to us, but the mechanisms involved were all outlined before. This is a forcing that models, and most people, have missed.”
Andrew Gettleman, project scientist at the National Center for Atmospheric Research, agrees. “The study results are consistent with what we understand about how the climate is forced,” he says. “It confirms we understand how the system works.”
What remains unknown is why the amount of water vapor declined. It seems to have increased through the 1980s and 1990s, though the data record, from NOAA balloons and satellites, is not perfect, making conclusions difficult. More recently, the data has improved.
Satellite records of stratospheric water show a distinct drop starting in 2000, Portmann says. It’s possible, he says, that global warming itself is causing the amount of water vapor to decrease, meaning the phenomenon could be classified as feedback rather than variability. “But then it’s peculiar that water vapor increased in the 1980s and 1990s, and there was an overall warming trend then. So it could be feedback, but it’s not straightforward.” While he thinks it’s more likely to be related to variability in the system, both explanations remain possible, and the paper makes that clear, he says.
As an aside, Portmann stresses that this study applies only to water vapor in the stratosphere — the second major layer of the Earth’s atmosphere, directly above the troposphere. “It’s important to keep in mind that tropospheric water vapor does go up when the temperature goes up, generally, but that is separate,” Portmann says. “That’s a positive, strong climate feedback.”
Stratospheric water, meanwhile, comes mainly from evaporation over the tropical regions of the planet; the moisture passes through progressively colder air as it rises through the troposphere and upwards. The colder the air, the less moisture it can hold. “The cold region there sets the boundaries, and temperatures there have gotten colder,” Gettelman says. “That’s most likely related to modes of climate, not a result of something that’s been done to it — more like an El Nino. I expect we might see the water vapor come back this year — but that’s not necessarily a prediction.”
It’s well known that water vapor follows temperature, he says: colder means more condensation, then clouds form, taking water out of the atmosphere. “There’s a strong correlation between temperature and water vapor, but we don’t fully understand what drives temperatures in those regions.”
That question and how scientists simulate this variability in their climate models are two areas researchers will be turning to next.
“It’s important that processes affecting stratospheric water vapor are better understood; it’s important for decadal scale predictions, which are becoming an important topic,” Portmann says. “A forcing such as the one we identified are important for that.”
Decadal predictions are indeed “one of the next big things,” Gettelman says, not only for the obvious predictive value of seeing 10 years ahead, but also for testing climate models. The trouble, he says, is figuring in the low frequency variabilities, especially with the ocean. “You have to get the ocean right; not knowing the state of the ocean is a big problem in decadal prediction. If you get a pile of warm water in the ocean current, it can get transmitted around the planet and then pop up and affect the atmosphere years later.” Right now, the atmosphere and the ocean surface are far better monitored than the deep ocean.
This study, meanwhile, fills in another piece of the puzzle, and it confirms that water vapor is the primary greenhouse gas. “It’s an amplifier of what we do to the system,” Gettelman says. “Heat up the planet, and water comes out of the ocean.”