Maybe drawing a line in the sand isn’t the best approach to climate change, especially when the sand is shifting.
A paper published in the journal “Nature Geoscience” in September theorized that the Earth might be warming a little less than climate models have predicted, by 0.3 degrees Celsius (0.54 degrees Fahrenheit).
That small number drew a lot of attention, because almost all climate scientists agree that it takes a substantial amount of emissions to warm the planet by even 0.1 degrees Celsius. If the lower estimate is correct, it would give the world more wiggle room to meet emission and warming targets.
Responses were predictable. Climate skeptics said, “All the climate models are wrong!”
“The Daily Telegraph” newspaper in Britain actually used that as a headline, “Climate models are ‘wrong.’”
Climate alarmists considered the new estimate too revisionary, and couldn’t imagine actually increasing the world’s carbon budget.
Some climate scientists noted that the new level was still at the low end of a range predicted by a group of climate models, essentially saying, “This is within the possibility of how things might go – if things go that way.”
And the general public was left thinking:
“What’s going on with these climate models?”
So it’s time to ask why there’s such a variation in climate models, and what climate scientists are doing to improve their models.
Butterfly Effect
It turns out that one of the most important factors climate scientists are now trying to understand are clouds.
In fact, how clouds work and what’s going on in the deep ocean are probably the two biggest puzzles climate scientists are trying to solve.
“Every global climate model does clouds the wrong way. Some do it worse than others, but they do it wrong,” said Joel Norris, professor of climate and atmospheric sciences at the Scripps Institution of Oceanography in La Jolla, Calif., part of the University of California, San Diego.
Clouds cool the planet because they reflect sunlight back into space. They also warm the planet because they trap heat. Different kinds of clouds have different effects. And it’s hard to predict where clouds will generate in a long-range scenario.
Clouds give climate scientists the willies.
“One of the things I’ve been looking at is how clouds have been changing over years and decades,” Norris said.
That’s not a simple lab experiment, he explained. It requires monitoring equipment and other assets, like airplanes to fly into clouds. Clouds might have changed little, maybe 1 percent, but “that 1 percent actually matters a lot for how much the Earth warms,” Norris observed.
Recent observations have shown more cloud coverage shifting toward the poles and cloud tops stretching higher into the atmosphere. Norris said we have better evidence for those changes, “and if they continue as we expect, they will also matter.”
“One of the challenges (in climate science) is that things that matter happen on a small scale, and how do you represent that in a simplified way?” he said.
In current climate work, “clouds are a big part, but aerosols and how they interact with clouds are also getting a lot of study,” Norris said.
“Another big part is trying to understand how the atmospheric circulation will change and how that will affect rainfall patterns where people live,” he said, adding that there are also questions about ocean circulation, especially in the deep ocean.
“We know small things vary. And that matters,” said Andrew Gettelman, senior scientist at the National Center for Atmospheric Research in Boulder, Colo. Gettelman wrote the book “Demystifying Climate Models” with co-author Richard Rood.
Climate Model Shortcomings
“It’s hard to know if our climate models are correct. We don’t really know the current state of the atmosphere,” he said. “We can’t describe things happening in the atmosphere at the level of detail we need.”
Climate models begin with the same approach as models used for short-term weather forecasting. In weather prediction, getting the large-scale motions of the atmosphere right is the critical step, Norris said. Climate prediction presents a more complex challenge on a time-scale of decades.
So why trust climate models at all?
“The advantage is, we know what the energy balance is. We have these large-scale constraints for climate modeling, which we don’t have for weather modeling,” Gettelman noted. “Those large-scale constraints help us because we can add up all the elements in the system.”
“My ultimate philosophy of climate modeling is, you constrain every aspect of the model to be realistic. The foundation is the fundamental laws of physics. We don’t violate any of them,” he said.
Norris, who isn’t a climate modeler himself, thinks there are limits to the usefulness of climate models.
“Climate models have been popular, but I don’t know in the big scope of things how much we’ve learned from them,” he explained.
“The way I look at them is, they’re a useful tool,” Norris said. “Climate models are better for analyzing processes, and not for getting a quantitatively precise answer.”
Predicting and Shaping the Future
It’s time for some definitions and background:
A “climate model” is a set of equations designed to approximate the workings of the Earth’s climate system in a simplified way. Still, climate models can be complex, sometimes requiring a supercomputer to process.
An “ensemble” is a selected group of climate models. No single climate model is likely to mirror the Earth’s climate system exactly, so scientists use an ensemble to produce a range of possible outcomes.
A “forcing” is a factor that affects climate. Positive forcing makes the planet warmer; negative forcing makes it cooler.
The troposphere is the lowest region of the atmosphere, extending up about 10 kilometers from the Earth’s surface.
The Paris Climate Accord was negotiated in 2015 by 195 nations plus the European Union. A goal of the accord is “holding the increase in the global average temperature to well below 2 degrees C above pre-industrial levels, and to pursue efforts to limit the temperature increase to 1.5 degrees C.”
If you want to run a climate model for future decades, you have to make assumptions about processes and inputs. You’ll need to make projections. How many volcanoes will erupt in the next 30 years, and how much ash will enter the atmosphere? How will cloud formations emerge, and where?
While it is possible the climate models themselves are badly flawed, it’s more likely that the assumptions have been far enough off to cause a miscalculation of projected warming – not by degrees or even a single degree, but by a fraction of a degree.
And that’s exactly what the authors said in the “Nature Geoscience” paper, published online as “Emission budgets and pathways consistent with limiting warming to 1.5 degrees C.” The paper was written by Richard Millar of the Environmental Change Institute at the University of Oxford, with nine co-authors.
The new, 0.3 degree Celsius-lower estimate of warming seemed to take everyone by surprise. It probably shouldn’t have. Dozens of papers have appeared in the scientific literature that address the climate models’ overestimation of warming.
In June, the very same journal included a paper titled “Causes of differences in model and satellite tropospheric warming rates,” by Benjamin Santer of the Lawrence Livermore National Laboratory and 15 co-authors, including climatologist Michael Mann.
“In the early 21st century, satellite-derived tropospheric warming trends were generally smaller than trends estimated from a large multi-model ensemble,” the authors noted.
“We conclude that model overestimation of tropospheric warming in the early 21st century is partly due to systematic deficiencies in some of the post-2000 external forcings used in the model simulations,” they wrote.
Climate scientists will be debating the latest findings for years. The bottom line is there’s a belief out there right now that the planet has warmed a little less than scientists had expected.
And that’s a good thing.