If we continue dumping carbon dioxide into the atmosphere at current rates, our coldest years around 2047 will be comparable to our hottest years now, finds a groundbreaking study published in the journal Nature.
“On average, locations worldwide will leave behind the climates that have existed from the middle of the 19th century through the beginning of the 21st century by 2047 if no progress is made in curbing emissions of heat-trapping greenhouse gases, said researchers at the University of Hawaii at Manoa who sought to project the timing of that event for 54,000 locations,” my colleague Lenny Bernstein reports.
The study’s methodology, which pins this climate tipping point to a specific date and location using 39 climate models, is novel and intriguing. Gaining an appreciation for when different parts of the world will step out of the climate of the past and into an entirely new one could help us plan for and adapt to such an eventuality.
But if the climate of the future is to become unrecognizable from the climate of the past, can science really nail down exactly when and where with confidence? Unlikely.
“I simply do not believe any of the models are able to reliably say anything meaningful on the details in degree and timing with such specificity, region by region, let alone city by city, with no mention of uncertainties,” says CWG’s Steve Tracton, who spent decades working on numerical weather prediction models.
David Titley, a professor of meteorology at Penn State, expressed similar reservations.
“We can’t say on one hand that the [current] slowdown in warming air temps is because of random, ‘short-term’ (<30 year) fluctuations (or stochastic processes), then give a prediction (not projection!) to within 12 months of crossing a very specific temperature threshold,” Titley says. “Round to the decade, add a confidence interval, etc….”
The study’s authors provide information about the uncertainty in their predictions, but it’s limited. They provide standard deviations (plus or minus 14 years for the global average tipping point date of 2047 in the business-as-usual greenhouse gas emissions scenario). And for the tipping point in individual cities, there’s a 5-year margin of error, the authors said at a news conference.
But beyond that information on standard deviations and margin of error, the authors don’t really discuss the modeling limitations and the assumptions that have to be made in making such specific predictions.
Fortunately, some science writers stepped in to provide a framework for interpreting the study’s results.
For example, credit Andrew Freedman at Climate Central who writes (bold text is my added emphasis) “Given that there are considerable uncertainties about future greenhouse gas emissions as well as the precise response of the climate system to those emissions, not to mention the uncertainties inherent in computer modeling, the study should not be taken as offering precise predictions.”
Similarly, the New York Times’ Justin Gillis cautions (bold text is my added emphasis): “The research comes with caveats. It is based on climate models, huge computer programs that attempt to reproduce the physics of the climate system and forecast the future response to greenhouse gases. Though they are the best tools available, these models contain acknowledged problems, and no one is sure how accurate they will prove to be at peering many decades ahead.”
Without the kind of qualification provided by Freedman and Gillis, it’s easy to be misled about the clarity of the future.
This new paper provides a valuable new way of very generally thinking about how, when and where the climate may reach a new frontier. As climate models get better and our scientific understanding of climate improves, scientists will be able to sharpen this thinking to some degree. For now, it’s best to say the study provides a reasonable first approximation of climate tipping points. It certainly does not provide a crystal ball.