Reconciling the science, economics and policy of climate sensitivity: would a lower value buy us more time?

Roz Pidcock

A recent high-profile study suggested earth’s sensitivity to greenhouse gases in the coming decades could be lower than previously thought – a premise some have argued means we can slow efforts to bring down emissions.

The authors have just published a new paper on how climate sensitivity can inform policy. We spoke to them about it and they told us even if estimates were to come down a bit, current policies to tackle climate change still aren’t tough enough.

Climate sensitivity

A lively discussion is taking place online and in the media about how sensitive the earth is to increasing carbon dioxide – something scientists call climate sensitivity.

The recent paper by Alexander Otto and colleagues, published in Nature Geoscience, focussed on earth’s Transient Climate Response (TCR). That’s the amount of surface warming we can expect from a doubling of atmospheric carbon dioxide above pre-industrial levels.

An alternative option is equilibrium climate sensitivity (ECS), which includes all known processes that influence the climate, including the slow uptake of heat by the oceans. We wrote more about the difference between the two measures here.

“Call of motivation”

Based on temperature data between 1970 to 2009, the paper suggested a best estimate for TCR of 1.4 degrees – which is about 20 per cent lower than the average from the new suite of climate models used by the Intergovernmental Panel on Climate Change (IPCC).

Climate skeptic Tory peer Matt Ridley took this result to mean climate change is likely to be “slow and harmless”. But one of the paper’s co-authors, professor Myles Allen, quickly dismissed Ridley’s interpretation.

Lead author Otto tells Carbon Brief that a lower value of TCR is no reason to slow the pace of emissions reduction – quite the opposite. He says:

“[I]t is not about slowing or pausing our efforts at all, because the change we are talking … this is more a call of motivation for getting things done than a call for laying back.”

Otto, Allen and colleagues have just published a new paper in Environmental Research Letters explaining the role climate sensitivity plays in informing climate policy. We spoke to them about it to see how they came to these conclusions.

Social cost of carbon

Scientists predict destructive consequences of climate change like more frequent wildfires, extended periods of drought and a higher risk of strong storms could incur heavy costs in the future.

One way to represent this is by estimating the cost of future damages caused by one additional unit of carbon if it were to be released into the atmosphere today – known as the social cost of carbon.

Measured in dollars per tonne of carbon dioxide, this essentially allows a tonne-by-tonne comparison with a particular policy approach, to work out its cost-effectiveness.

The main point the new paper makes is that TCR is better for working out the social cost of carbon than ECS. That’s because ECS is far more complex to calculate, which means it comes with a large range of uncertainty.

Reducing uncertainty

On the other hand, the uncertainty range for TCR is smaller because it can be better constrained by past surface temperature measurements, according to the authors.

The authors show that when they used TCR calculated between 1970 and 2009 to work out the social cost of carbon instead of ECS over the same period, the uncertainty in the result reduced by 30 per cent.

The caveat to their conclusion that TCR is more policy-relevant than ECS, however, is that it only holds true if the damage caused by climate change is assumed to increase gradually with temperature. By contrast, Otto tells us:

“If we expect climate change to be like a threshold, beyond which damages are very high â?¦ then the [equilibrium] climate sensitivity is the more interesting parameter (because it tells us whether or not we will be crossing this assumed threshold)”.

Tougher policy

Given the authors’ conclusion that TCR is the most appropriate measure to inform climate policy, a big question is what difference the new suggestions of a lower TCR might have in practice.

Specifically, if TCR is 20 per cent lower than current climate models predict, how does this impact the social cost of carbon?

Climate change policy researcher Dr Chris Hope has done these calculations using the extremely comprehensive PAGE09 model, an updated version of the one used to underpin the 2006 Stern Review on the economics of climate change.

Hope’s working show that using a TCR value of 1.4 degrees instead of 1.8 degrees, the social cost of carbon goes down from about $100 to $80 per tonne.

But Hope says since the UK’s carbon floor price is currently £16 per tonne, and that only the most energy intensive firms pay it, this means:

“[W]e can indeed say that the new estimates of the TCR show that our climate policy is misguided, but not in the direction that Matt Ridley believes; the new TCR estimates support a much tougher policy than the one we currently have in place.”

We asked Otto if a lower TCR provides support for slowing or pausing current efforts to cut emissions. He told us:

“No, not at all â?¦ Because even with 30 per cent lower values of TCR â?¦ we would still look at significantly higher temperatures than the two degree target at the end of the century, when continuing our current emission trends.”

A complex business

These calculations are extremely complicated. For a start, the whole uncertainty range of TCR has to be input into the model – not just the best estimate. Then the average value is taken from many thousands of model runs, each of which evolves in a slightly different way.

And there are other uncertainties besides climate sensitivity. The precise value of social cost of carbon depends heavily on the cost attached to climate damages. But uncertainties aside, it appears the scientists making the new estimates don’t think lowering climate sensitivity is a reason to sit back and relax when it comes to current climate policy.

🗂️ back to the index