SACRAMENTO, Calif. — Climate change scientists don’t like to use the term “prediction.” Rather, they’re making “projections” about the future of the planet as sea levels rise, wildfires sweep the West and hurricanes become more ferocious.
There’s a good reason for that.
In a world awash in misinformation — about medicine, politics and climate, and pretty much everything else — part of a scientist’s job now involves teaching the public about how science works. Convincing the public to have faith in science means making precise, measured projects about the future.
They’ve got to overcome the big question: Can you really make accurate projections about what the planet will look like in 50 years, a century from now?
Climate scientists think they can, based on the past five decades of climate science that has proven accurate. Futurists, such as Jamais Cascio, a distinguished fellow for the Institute for the Future, a nonprofit foresight group based in Silicon Valley, study present trends and available data to lay out plausible outcomes for the future.
Today, a lot of Cascio’s work is centered around climate change, helping people prepare for the future and make informed decisions for a warming world.
“Everything in the world,” Cascio said, “every future outcome will have to be examined through the lens of climate.”
In the future, climate change may only get worse. But how much worse will it get?
Scientists have relied on climate models for over 50 years. To people who aren’t scientists, it’s challenging to understand the calculations that go into these projections. So, what exactly is a climate model?
Meteorologists can make weather predictions for the next hour, or even week, based on weather data and forecast models that use humidity, temperature, air pressure, wind speed, among other current atmospheric, land and oceanic conditions. But with climate, a specific region’s weather averaged over decades, is a little more challenging to project and understand.
An extension of weather forecasting, climate models factor in even more atmospheric, land and oceanic conditions to make longer-term forecasts. Using mathematical equations and thousands of data points, the models create representations of physical conditions on earth and simulations of the current climate.
Climate models predict how average conditions will change in a region over the coming decades as well as how the climate appeared before humans recorded it.
Researchers can then understand how these changing conditions could impact the planet, which is useful especially for understanding climate change, said Zeke Hausfather, a climate scientist and director of climate and energy at the Breakthrough Institute, an environmental research center based in the Bay Area.
“Perhaps the most important (purpose) is to try to suggest the types of changes that might occur as the world continues to emit CO2 and other greenhouse gases,” Hausfather said.
The first climate model, developed over 50 years ago in the early days of climate science, helped scientists gauge how the ocean and atmosphere interacted with each other to influence the climate. The model predicted how temperature changes and shifts in ocean and atmospheric currents could lead to climate change.
Today, these models are much more complicated and run on some of the world’s most powerful supercomputers. A decade ago, most models broke up the world into 250-kilometer segments, but now the models are 100 square kilometers. More regional patterns emerge when simulations are at a finer scale.
“People aren’t drawing a picture of temperature and carbon dioxide and drawing a line through it and then extrapolating that into the future,” said Gavin A. Schmidt, a senior climate adviser at NASA.
Through these advancements in technology, these models are becoming even more useful to scientists in understanding the climate of the past, present and future.
“Fortunately, they don’t do such a terrible job,” Schmidt said.
All of this works toward convincing the public and businesses to take action.
A majority of Americans already notice the effects of climate change around them, according to a Pew Research Center survey from 2020. But individuals, businesses and politics must “adapt to a radically and dangerously changing climate,” Cascio said.
On the individual level, people must consider the climate in all of their monumental decisions: whether to have children; which car to buy; how to invest; when and where to buy a house. Governments are tasked with climate decisions that impact the future of entire nations, such as whether to invest in alternative energy or write policy curbing emissions.
Are climate models useful?
Instead of thinking about climate models as whether or not they are right, Schmidt said climate models should be considered as to whether they provide useful forecasts.
“Do they tell us things? Do they get things right more than you would have done without them?” Schmidt said.
Usually, the answer is yes, and what these models inform scientists is crucial for their understanding of the future climate.
Hausfather knows this better than anyone, as he led a study published in the journal Geophysical Research Letters analyzing the accuracy of early climate models. Some of the findings were included in the latest report from the United Nations’ Intergovernmental Panel on Climate Change published in August.
Hausfather, along with co-author Schmidt, compared 17 model projections of global average temperature developed between 1970 and 2007 with actual changes in global temperature observed through the end of 2017.
Hausfather and his colleagues found promising news: Most of the models have been quite accurate. More specifically, 10 of the model projections show results consistent with observations. Of the remaining seven model projections, four projected more warming than observed while three projected less warming than observed.
But Hausfather and his colleagues realized this wasn’t telling the whole story. After accounting for differences between modeled and actual changes in atmospheric carbon dioxide and other factors driving the climate, it turns out 14 of 17 model projections were “effectively identical” to warming observed in the real world.
“That was strong evidence that these models are effectively right,” Hausfather said. “They’re doing a very good job of predicting global temperatures.”
The accuracy was particularly impressive in the earliest climate models, Hausfather said, especially given the limited observational evidence of warming at the time.
But not all of the early models were error-free. One of the first climate models, created in 1971 by climate scientists Rasool and Schneider, projected that the world would cool due to the cooling effect of atmospheric aerosols.
“(The researchers) thought that the cooling effect of these aerosols from burning fossil fuels that would reflect sunlight back to space would be much stronger than the warming effects of the greenhouse gas,” Hausfather said.
While the 1970s were still in the early days of climate research, most of the scientific literature of the time was still pointing toward a warming future as much more likely. Yet, Rasool and Schneider’s model still spurred a slew of news stories about a potential ice age. Even today, the model “still gets trotted out every now and then by folks trying to discredit climate science today,” Hausfather said.
Now the model is proven to be wrong. It’s a consensus among climate scientists that the planet is not cooling — instead it’s warming at an alarming rate.
Even today, despite the promise of climate models shown by Hausfather’s study, these models still have their limitations, especially with regard to the uncertainty of future emissions. Climate scientists are physicists — not economists or political scientists, and it’s challenging to understand how policy will shape emissions standards.
“We don’t have a crystal ball that can predict the future human behavior in terms of how much our emissions will change,” Hausfather said. “We can just predict how the climate will respond to the emissions.”
Issues of accuracy in climate models also still arise when models are pushed outside of their specific parameters. To combat this, climate models focus their projections on physical conditions seen in the natural world, instead of statistical probability, Schmidt said.
Researchers have more confidence in the predictability of physics than statistics, because physics doesn’t change into the future. Researchers can have confidence that they can use these models outside of the time period where they have observational data, such as looking at climate during the last ice age, Schmidt said.
“How things get expressed might be different but the basic physics … the underlying processes don’t really change,” Schmidt said.
Hausfather said there’s still a lot of work still to improve climate models, but they are consistently getting better over time. Simulations of the Earth become sharper as more physical processes are added and computer power grows.
Why make projections for the future?
While climate scientists focus on physics to make forecasts for the future climate, Cascio and other futurists place scientific data in a larger context, making foresight based on climate change, new technological developments, as well as political and social movements. Futurism is “essentially anticipatory history,” Cascio said.
“The idea is to take the science and embed it into a historian’s understanding of how the world works to try to get a sense of what are the possible outcomes that we see going forward,” Cascio said.
But, just like with climate models, uncertainty is inherent to the nature of projections. Futurists do not want to over-promise, but they provide a forecast of what could happen and reasons why it could happen, Cascio said.
Most of Cascio’s work with climate change projects a grim future. In his perspective, an “absolutely radical” and “transformative” climate plan is necessary to make the necessary change. Plans that are “sensible and acceptable (are) almost definitely not enough.”
“I really want to be wrong about all of this stuff,” Cascio said, “because there are no futures that are not really depressing for the next generation.”
Despite the despair projected by many climate scientists and futurists, there’s still hope. If global emissions can be brought down to zero, Hausfather said the best climate model estimates illustrate that the world will stop warming.
“It’s not too late to act,” Hausfather said. “The world is not locked into a particular amount of warming.”
Cascio still tries to consider himself a long-term optimist for the future, because the changes necessary to mitigate climate change will also lead to a much more “transparent and equitable” world, he said.
“If we can make it through the second half of this century, there’s a very good chance that what we’ll end up with is a really wonderful world,” Cascio said.