November 2, 2021

We are “quite literally” in the “last chance saloon,” says Prince Charles. Humanity is “about 5-1 down at half-time,” says Boris Johnson. People will curse this generation of politicians worse than they did that which appeased Hitler, says Justin Welby, the Archbishop of Canterbury, because they are allowing “a genocide on an infinitely greater scale”.

Are things quite that bad? Well, if you read the headlines, yes: we’re on course for disaster and we don’t seem to be turning. “Business as usual” will lead to a catastrophic collapse of Himalayan glaciers; and devastating heatwaves in the southern United States, according to the New York Times. What’s more, the same newspaper said last week ahead of COP26, “If we continue with business as usual, by the end of the century, it will be too hot to go outside during heat waves in the Middle East and South Asia.”

But all these stories share one thing: they are based on the IPCC’s RCP 8.5 scenario. It’s not an exciting name, but 8.5 is often described as “business as usual”. Hence the headlines.

RCP 8.5 is not business as usual, though; it’s an unlikely worst case. This means a large fraction of the public debate on climate change mitigation is driven by an increasingly implausible scenario, which was unlikely when it was proposed, and is even less so now. The more we focus on this scenario, though, the more pessimistic — and the more hopeless — the situation will seem.

Back in 2007, after the fourth IPCC report was published, climate scientists wanted to create new emissions scenarios, imagining how they might change in the coming decades and what impact they would have on the climate. But while there was a plan to make in-depth scenarios ahead of the fifth report in 2013, scientists would need to start running models for that report as soon as 2010. They needed something that could be used in the meantime.

So four “representative concentration pathways”, RCPs, were developed: RCP 2.6, 4.5, 6.0 and 8.51. They were, roughly speaking, a highly optimistic scenario regarding emissions, two middling scenarios, and a highly pessimistic one. RCP 8.5 was there as a realistic worst case. Even at the time, it was viewed as unlikely. But two unfortunate things confused that picture.

First, of the four RCPs, only 8.5 imagined a world with no climate policy. And second, its authors described it as “a high-emission business as usual scenario” – meaning that it was at the high end of emissions for business as usual, but which was taken to mean in some quarters that high emissions were business as usual.

So RCP 8.5 became synonymous with “business as usual”. There are dozens of studies published every year describing it as such: this one says that “Business-as-usual will lead to super and ultra-extreme heatwaves in the Middle East and North Africa”. It is a mainstream part of climate science.

“The problem isn’t that people study these scenarios,” says Zeke Hausfather, a climate scientist at the Breakthrough Institute. “It’s that they frame them as the most likely outcome in the absence of policy. That was probably never true, but I’m sure it’s not true now.”

Most obviously, we don’t live in a world without climate policies. The Paris protocols commit signatories to goals of keeping warming below 2°C and aiming for below 1.5°C. So “business as usual” is a global effort to reduce emissions.

Second, and most important, the world has changed its energy mix far faster than RCP 8.5 expected. Hausfather points out that “current solar prices are below what the models think they’ll be in 2050”, and coal use has dropped in recent years. In order for emissions to reach what RCP 8.5 imagined, we’d need to increase coal use per capita by about 700% from today’s levels, spectacularly reversing the recent decline.

And this means that the really devastating levels of warming envisioned by RCP 8.5 are even less likely: we’re probably looking at a situation of 3°C warming above pre-industrial levels, compared with the 4°C, 5°C or even 6°C that RCP 8.5 predicts. As Hausfather said in a Nature piece last year, 3°C is still “a catastrophic outcome”, but it’s a heck of a lot better than it used to be.

It’s worth noting that while the emissions that get the world to RCP 8.5 are highly unlikely, the outcomes that 8.5 predicts are still plausible. One of the uncertainties in the climate is the feedbacks. For instance, as the world gets hotter, ice melts. Ice reflects sunshine, while dark earth absorbs it. So the less ice there is, the faster the world warms, reducing ice. That’s a positive feedback system: it tends to accelerate changes.

On the flip side, the more CO2 there is in the atmosphere, the faster plants grow, and the more they suck from the atmosphere. That’s a negative feedback system, and it tends to moderate changes.

The exact magnitude of the many possible feedback systems in the atmosphere is uncertain. As the Siberian permafrost melts, will it release gigatonnes of CO2? It is possible that even under more optimistic emissions scenarios, the feedback systems will push more carbon into the atmosphere, so we end up with the same CO2 concentrations as we would have done under RCP 8.5. And there’s another layer of uncertainty, which is “climate sensitivity”. It’s not precisely known how much warming a given increase in CO2 concentrations will cause; perhaps even if the amount of carbon in the atmosphere is relatively low, the warming will be greater (although that now looks less likely).

So really bad outcomes are possible even without RCP 8.5-level emissions. We’d have to get “incredibly unlucky” with feedbacks for the world to end up warming that much, says Hausfather, but we can’t eliminate the possibility.

As a result, there’s an ongoing row among academics about whether RCP 8.5 (and its equally pessimistic successor2) should be used at all. Roger Pielke Jr, a professor of public policy and environmental studies at the University of Colorado, Boulder, argues that it essentially has no place: that it’s based on an implausible scenario, even if you could end up somewhere similar because of uncertainties elsewhere. “Two wrongs don’t make a right,” he says.

But Richard Betts, a climate scientist at Exeter University and the Met Office, and one of the authors of the UK government’s climate change risk assessment, argues differently. We need to be aware of the risks of these 4°C worlds that we could – plausibly – still find ourselves in, he says. And there’s been lots of research done on worlds like that, mostly using RCP 8.5.

Gavin Schmidt, the director of NASA’s Goddard Institute, adds that using extreme scenarios allows modellers to tease out causality in a way that more narrow ones don’t. “It allows you to see non-linearity, allows you to look for thresholds,” he says. “You get a stronger signal to noise ratio.” The actual emission scenarios are less likely, he agrees, but studying them helps us understand the climate system better.

The trouble is that if a journalist sees two graphs in a study, one of a plausible RCP 4.5 world and one of a much more dramatic RCP 8.5 world, they’ll probably want to run the RCP 8.5 graph, “because it’s more impressive,” says Hausfather. “And they’ll say scientists predict this much sea level rise, or this many heat deaths. And it’s not accurate because you’d need to add ‘but only if we burn all our coal or get very unlucky’.”

There’s also a risk that RCP 8.5 will crowd out research into other areas. For instance, Our World in Data’s recent efforts to understand human impacts on biodiversity have been hampered because almost all the research into coral reef collapse has been carried out using RCP 8.5. Predictably enough, all the coral dies in that scenario. But what would coral reefs look like under 2°C of warming, or 3°C? We don’t really know.

The RCP 8.5 debate is heated. Pielke Jr and Schmidt in particular have a longstanding mutual animosity, and Schmidt feels Pielke Jr has “spent 20 years trying to elbow out scientists from the centre ground” on climate change. So it’s hard for a journalist to get involved without taking sides.

But it’s important, because the more we associate the worst-case scenario with business as usual, the more pessimistic the public debate will be. “It might demotivate people,” says Hausfather. “It’s much easier to see the Paris goals [of no more than 1.5°C warming] as achievable if you know we’re on course for a 3°C world, rather than a 5°C one.”

I disagree with Pielke Jr: RCP 8.5 has its place, because as Betts says, there’s a whole range of research into unlikely but plausible scenarios using it, and if you’re doing risk assessments, you need to look at outcomes that probably won’t happen but which would be disastrous if they did, so you can try to avoid them. At the moment the papers looking at those scenarios all use 8.5.

But on the other hand, if the IPCC puts out RCPs without explicitly saying which are the most likely, then policymakers and journalists will take whichever scenario most suits their needs, whether that’s pretending there’s no problem, or magnifying the problem for the sake of a headline. This isn’t the fault of climate modellers. But no one involved in the IPCC is explicitly saying “RCP 8.5 is pretty unlikely,” and that fateful phrase “business as usual” is still attached to it.

Hausfather has a solution for this: attach explicit percentage likelihoods to the different scenarios – say that RCP 4.5 is 45% probable, or RCP 8.5 is 5% probable. They’d be necessarily subjective but at least it would show that no one thinks they’re all equally likely, or that RCP 8.5 is the course we’re already on. Betts agrees.

There’s a risk that climate sceptics will leap on ideas like this, and say that climate scientists have systematically overstated the risk of climate change; Betts says that this already happens. But climate change is an unhedgeable risk, and we need to pay attention to even unlikely outcomes if they’re very bad, in the same way that you wouldn’t play Russian roulette even if it’s only a 17% chance of death. Besides, even the less dramatic 3°C worlds involve dangerous levels of sea level rise, heat waves, and millions of unnecessary deaths a year. You’d think that’d be worrying enough.

Every week, people see headlines, based on RCP 8.5 scenarios, that say things like three billion people could face “near-unliveable conditions” by 2070. Understandably, people are rattled; and some are even saying they’re not having children. People are more scared than they need to be.

Climate change is extremely bad. But we’re not in the “last chance saloon”, if that means that we face some inevitable catastrophe, or 5-1 down and facing ruinous defeat; and the politicians at COP26 are not guilty of facilitating a worse genocide than Hitler’s. Archbishop Welby apologised for that. But it’s an understandable mistake if he’s always being told that “business as usual” means a march to doom.

  1.  The numbers refer to the amount of “radiative forcing”: that is, the change to the Earth’s energy balance, in watts per metre squared. So under RCP 6.0, each square metre of the earth’s surface would absorb six more watts of energy than it would lose to space, relative to pre-industrial times.
  2.  In the latest IPCC report, the RCPs have been replaced with something else, called SSPs. Only one of the new SSPs, SSP5, is compatible with RCP 8.5. Research using both is combined into a scenario called SSP5-8.5. I’m going to refer to RCPs throughout, for simplicity.