Rising temperatures are producing some very predictable effects: gradually worsening droughts, a steady rise in sea levels, and so on. But we also risk crossing thresholds in which aspects of the climate suddenly shift to new behaviors. That process seems to have happened in part of the Arctic Ocean, and there are indications that the main circulating current in the Atlantic Ocean may be approaching a shutdown.
A lot of effort has gone into modeling the economic cost of climate change in general, but we haven’t figured out what crossing a tipping point might do to the world economy. This week, three researchers attempted to explore the subject, integrating estimates of the cost of tipping points with combined economics/climate models. The results suggest that we may be under-valuing the current cost of our carbon emissions and accepting a much higher level of financial risk than we might think.
What percent do you tip?
Climate tipping points are easy to understand on a conceptual level. At some poorly defined point in the future’s warming, some natural systems will shift to a different type of behavior. That behavior will make it unlikely that the system will return to its initial state.
For an example, we can turn to the permafrost found in many sub-Arctic regions. Warming will inevitably eliminate the “perma” from the frost, thawing the soil and potentially releasing much of the carbon stored there. Because that carbon will wind up as carbon dioxide in the atmosphere, it will contribute to warming that makes permafrost re-forming unlikely.
That sort of tipping point will have dramatic implications for the areas of the planet where permafrost is found. But it will also enhance future warming, which will have widespread impacts across the globe. So while the economic costs of climate change may slowly scale with warming up to the tipping point, they’re likely to experience a sudden jump if we blow past that point.
Quantifying the economic costs of climate change is typically done using the “social cost of carbon,” which puts a dollar value on each tonne of carbon we emit. Its price naturally rises as we continue to emit more carbon because the accumulation of carbon in the atmosphere makes it more difficult to avoid higher economic costs.
Calculating the social cost of carbon is challenging, even when warming produces effects that increase in severity along a relatively smooth curve. Making these calculations is considerably more difficult when the effects jump after a tipping point. Still, a team of researchers was able to find attempts to estimate the costs of eight of the tipping points. The researchers took these estimates and integrated them into a climate/economic model to see how the numbers would affect the social cost of carbon.
Melting ice, thawing carbon
Three of the tipping points the researchers considered—the thawing of permafrost, the release of methane from ocean clathrates, and a drying out of the Amazon—act by putting more carbon into the atmosphere. Two others involve the disintegration of the ice sheets in either Greenland or the West Antarctic, which would accelerate sea-level change. Three others—changes in the reflectivity of the Earth due to the loss of ice and snow, a shutdown of the Atlantic Ocean circulation, and increased variability of the Indian monsoon—have more subtle impacts.
The above isn’t a full list of the climate tipping points we’ve identified; it’s simply the list of effects for which we have a decent estimate of economic impacts. So we can assume that the total impact of tipping points will be higher than the figures generated here.
The researchers started with an estimate that put the social cost of carbon at $52/tonne and used their model to figure out how each of these tipping points changed that price. Two of them—destabilizing the Atlantic circulation and altering Earth’s reflectivity—actually decreased the social cost of carbon, although by less than two percent each. The former limits the warming in the North Atlantic, and the latter produces more warming near the poles, both of which are economically favorable.
Everything else, however, showed a net negative. These results ranged from relatively minor impacts due to ice sheet stabilization (two and three percent increases for the two ice sheets) to rather dramatic effects from destabilizing carbon stores. The release of permafrost carbon boosted the social cost of carbon by over eight percent, and the release of methane from ocean clathrate deposits added a 13 percent hit.
If we go past all the tipping points, the social cost of carbon will go up by roughly 25 percent.
Risk tolerance
There are many uncertainties in all these estimates, and the researchers explored them using 10,000 separate runs of their model. The runs made it clear that while the uncertainty is large, it’s very unlikely to bail us out, since it’s not evenly distributed on either side of the 25 percent estimate. There is almost no probability that the total impact of these tipping points would lower the social cost of carbon. At the same time, there’s a notable chance that the tipping points could more than double the social cost of carbon.
The uncertainty is not evenly distributed in time, either. There’s much more uncertainty about the impact of crossing tipping points by mid-century, in which case the effects might occur against the backdrop of a climate that is otherwise not dramatically different from our present one. In contrast, the uncertainty of the impact is far lower by the end of the century, when so much else is going wrong that the social cost of carbon is likely to be very high in any case.
Overall, this work is an early effort based on a limited number of studies that looked at individual tipping points. We’ll want to see further efforts to limit the uncertainty of these studies, and the climate effects should be integrated with the updated models and estimates in upcoming IPCC climate reports.
At the same time, the study clearly indicates that the riskiest tipping points are the ones that pump a lot of additional carbon dioxide into the atmosphere. That idea on its own should provide a strong incentive to monitor these effects to detect early warning signs, and we should study them further to make sure we know the right warning signs.
PNAS, 2021. DOI: 10.1073/pnas.2103081118 (About DOIs).