Global Warming Is Worse Than Most Models Predict, Data-driven New Study Warns

Projections on climate change by 2100 have been too optimistic and emission curbig targets are too low, warn Carnegie Institute scientists

Ice melting in Greenland
Ice melting in Greenland John McConnico, AP

Global warming will be even worse than we presently think, scientists warn in a new paper published in Nature on Wednesday. How much worse? In all scenarios of greenhouse-gas emissions, the estimates of temperature increase by the Intergovernmental Panel on Climate Change are about 15 percent too low.

This result is based on a new model, based on picking from dozens of previously published climate models that successfully simulated the recent past, plus actual observations. The key concept is that the models best at simulating the past would likely be best at projecting future warming too.

The conclusion of Patrick Brown and Ken Caldeira of the Carnegie Institute for Science that, if anything, science has been too optimistic, applies whether we stop emitting greenhouse gases today or continue on our present trajectory: projections have been too optimistic across the board.

If greenhouse-gas emissions continue to rise throughout the century (the “business-as-usual” scenario – also known as “the worst case scenario”), existing climate models predict the global mean temperature will rise by 3.2 degrees to 5.9 degrees Celsius (5.8 degrees to 10.6 degrees Fahrenheit) above pre-industrial levels by the year 2100.

“Our projection basically eliminates the low end of the range, but retains the upper end of the range and moves the central estimate up 15 percent,” Brown explains.

So much for halting temperature rise at 2 degrees Celsius, which was already considered to be a disastrous level of global warming. The 2-degree hopeful limit was set in the 2016 Paris Agreement climate accord that the United States has since abandoned.

Their ultimate conclusion is that greenhouse-gas emissions need to be reduced more than had been thought if we hope to stabilize the global mean temperature, Brown says. And make no mistake, emissions are still increasing globally.

Climate modles that simulate the current climate the best tend to project more global warming
Patrick Brown and Ken Caldeira, Nature

Reducing the uncertainty

All climate models predict average global warming because of human greenhouse-gas emissions. The question is by how much, and the spread is not small. Why should we believe Brown and Caldeira rather than some more benign model?

Climate modeling is an imprecise science, partly because of the vast number of parameters involved, plus the unknowns, which include the behavior of feedbacks in the climate system, notably cloud feedback.

“The wide range comes from there being no consensus on how to best model many key aspects of the climate system – in particular how clouds might change with warming,” Brown explained to Haaretz.

“[Other] models do not explicitly incorporate modern observations. They are based on physics and simulate a world that is internally consistent, but may not match observations of the recent past very well,” he added.

Their model is likely to prove more reliable because it factors in actual observation, not just theory, they claim. Technically, Brown and Caldeira took various climate models and constrained them with actual observational data regarding the amount of light and heat energy that the Earth exchanges with space.

Based on observation, they predict that the cooling effect from clouds will be diminished in the future, and so the world will get warmer than is presently expected.

To sum up, the observationally informed warming projection for the steepest emissions scenario in 2100 is about 15 percent warmer than the IPCC says. Moreover, Brown and Caldeira estimate their work diminishes the uncertainty of previous projections by about a third.

In April 2016, the European Geosciences Union reported on the likely consequences of a 1.5 degree Celsius increase in global mean temperature, compared with an increase of 2 degrees based on existing climate models. That extra increase, from 1.5 to 2 degrees, means heat waves would last around a third longer and storms would be about a third more intense. Even in the arid Middle East, which is expected to become drier and drier, the storms that do arrive are expected to be stronger.

It bears adding that there are tons of scenarios the Carnegie scientists cannot humanly factor in – one being the actual future emission of greenhouse gases.

Another is the possibility of rapid, nonlinear melting of the Greenland and Antarctic ice sheets. That could happen, they point out, but this imponderable (and unthinkable) cannot be represented in any of the models they studied. Ergo, it has an effective probability of zero, even though a 2016 paper predicts that very thing even in the lesser case of a 2-degree Celsius increase in temperature. And if that happens, all bets are off.