RCP 8.5 is officially bollox
Back in 2019, I created a Twitter hashtag, #RCP85isBollox, to highlight the danger of building the case for climate action on a wildly implausible scenario. This month RCP 8.5 was retired.
This is a long post for those who remember #RCP85isBollox.
That was the hashtag I invented back in 2018 as part of my campaign to caution the clean energy community that RCP 8.5 - the extreme scenario heavily promoted by the IPCC, climate scientists, journalists and activists, and accepted on trust by political, business and financial leaders - was wildly implausible.
Last week, the journal Geoscientific Model Development published a paper by Professor Detlef van Vuuren and others on the scenarios that will be used for the next iteration of the IPPC’s influential Assesement Reports, AR7, due out between 2028 and 2029. In it, they finally admit what anyone paying attention has known for many years: that RCP 8.5 (and its successor scenario, SSP5-8.5) is "implausible".
The authors are recommending a High scenario that will sit somewhere around RCP 6.7 - still too high to be truly plausible, but a huge improvement. Most importantly, what the abandonment of RCP 8.5 means is that AR7 will need to be scrubbed of findings that rely on the use of RCP 8.5 as explicit or de facto Business as Usual (BAU). That is a colossal undertaking. There are 105,000 papers in Google Scholar that mention RCP 8.5 (including an incredible 2,290 published so far this year) and many of them - perhaps the majority - use the scenario inappropriately.
My only critique of the new scenario paper is the wording: that RCP8.5 should be retired because it is "no longer" plausible. This is a fig-leaf, allowing those who have clung to RCP 8.5 for a decade longer than they should to claim that it was only recent developments in clean energy that rendered RCP 8.5 implausible. That’s nonsense, RCP 8.5 was implausible the moment it was developed.
RCP 8.5 was always implausible
The first line of the 2011 paper that presented RCP 8.5, also with Professor Van Vuuren as lead author, explicitly states that scenarios used in climate research are to be “plausible descriptions of how the future may evolve” [my emphasis].
To reach the carbon concentration designated in RCP8.5 would require global coal use to grow by around 7x from 2015 to 2100. To put that in perspective coal’s growth between 2015 and 2025 has been less than 10%.
How could the world have given such prominence to a scenario that is so wildly out of sync with real developments in the energy industry?
The roots of RCP 8.5 date back to the time when China’s coal use was surging by more than 10% per year. At that point it was perhaps an interesting academic question to ask what might happen if that growth continued for a century - although even then, that outcome could not have passed any test of plausibility.
In 2006, China set a national target in its 11th Five‑Year Plan to cut energy use per unit of GDP by 20% by 2010. That spelled an almost immediate end to China’s precipitous growth in coal-fired power (I remember the celebrations) and should have spelled the death of RCP 8.5 before it was even born. From that point on, no one familiar with the assumptions behind RCP 8.5 could in good faith pretend it was plausible, still less that it was BAU.
By 2017, there was no justification to continue using RCP 8.5 for any purpose. Justin Ritchie and Hadi Dowlatabadi published a puplished a paper showing there weren’t enough recoverable coal reserves on the planet to follow RCP8.5 even if you wanted to. They concluded that “RCP8.5 no longer offers a trajectory of 21st-century climate change with physically relevant information for continued emphasis in scientific studies or policy assessments” but were ignored by mainstream climate researchers.
I’m embarrassed to say it didn’t cross my mind until around 2017 (over a decade after founding New Energy Finance) that it might be worth checking the climate modellers’ homework against the scenarios being produced by my own team and colleagues in the energy industry. What I found did my head in: it was clear that RCP 8.5 was already wildly implausible.
I became very concerned that by hanging its hat on research underpinned by RCP 8.5, the climate community was creating an way for those intent on defending the status quo in the energy system to brush off the need for change - and I determined to warn people way from the precipice.
The birth of #RCP85isBollox
As I talked to people in the clean energy sector, it became clear that almost none had even heard of RCP 8.5, let alone taken the time to see how it mapped against the energy scenarios being produced by the IEA, EIA, BloombergNEF and others with real expertise in energy system modelling. They simply took it as an article of faith that the world was headed for catastrophic climate impacts at the global level in the coming few decades.
When I started to explain my findings and concerns on social media, at first no one paid any attention. It was a dissonant narrative and it was shrugged off.
It was only when I created the hashtag #RCP85isBollox that I attracted the attention of the climate police - a self-appointed group of noisy social media pundits, more intent on maintaining message conformity within the climate movement than in intellectual and academic integrity.
The level of vitriol and trolling that I started to receive was off the charts. I had articles written about how wrong I was, how politically motivated and toxic I was. I could list the worst offenders here, but they know who they are and they should be thoroughly ashamed of their behaviour.
Sadly, critics included quite a few well-regarded scientists, who weighed in to defend the scenario on which so much of their work hung. It took thick skin to stick to my guns, which I did even when that meant having to dive deep into the scenario literature to rebut attempts by Dr Richard Betts to blind the lay audience with science. I ended up blocked by Dr Katharine Hayhoe after asking her about the U.S. National Climate Assessment, which derived the financial cost of climate action by comparing costs under RCP 8.5 (a scenario with a 2100 global population of 12.3 billion) with costs under RCP 4.5 (a scenario with a 2100 global population of 8.7 billion).
2019 to 2023
I became so sure of my ground that in December 2019, in a lookahead piece as we entered the new decade, I predicted that "At some point during the next 10 years, it will become clear that RCP 8.5, an utterly terrifying, high-end scenario when it was first mooted in 2007, can no longer be considered plausible, much less base case." I even predicted, in private messages, that the epiphany would not happen in time for IPCC AR6.
There was a turning point in January 2020, when climate scientists Zeke Hausfather and Glen Peters published a comment in Nature that has now been cited over 1,440 times, entitled “Emissions – the ‘business as usual’ story is misleading”, urging people to “Stop using the worst-case scenario for climate warming as the most likely outcome — more-realistic baselines make for better policy.” Peters later came on Cleaning Up and admitted that he and Zeke wrote the piece because of the provocation provided by #RCP85isBollox.
In March 2020 I went on Chris Nelder’s wonderful Energy Transition Show podcast (if you don’t subscribe to it, you really should) and debated the assumptions behind RPC 8.5 with one of the people behind it, Dr. Nico Bauer, a climate modeller with the Potsdam Institute for Climate. I finally understood why the scenarios produced by climate models rely on insane amounts of coal: they combine the assumption that population and the economy continue to grow, with the assumption that oil and gas run out. The only thing the models were allowed to use to power transportation was coal-to-liquids. Incredibly, Bauer defend this decision, despite the fact that EVs were taking off and coal-to-liquids was flat-lining, and that we now knew the volume of coal required simply didn’t exist.
In July 2021, I had the great planetary systems scientist Johan Rockström on Cleaning Up, and pushed him on RCP 8.5. What he said was “I share with you the critique on RCP 8.5. I’ve been criticizing that for a very long time. I find it very unsatisfying that it assumes fossil fuel burning at levels that will never happen… I never use RCP8.5 for anything.”
Professor Jim Skea, at the time vice-chair and now chair of the Intergovernmental Panel on Climate Change for AR7 came on Cleaning Up in March 2021. What he said of RCP 8.5 was “You need to think all of the policies that have been put in place over the last three years are reversed. You need to think that all of the wind turbines that have been installed are dismantled and we don’t replace our photovoltaic cells.” In other words, it was entirely implausible.
Nevertheless, the misinformation continued, even from those claiming the mantle of scientists. Dishonourable mention must go to Christopher Schwalm et al, who produced a paper in 2020 in defence of RCP 8.5, claiming it was the best fit for observed emissions. It was a statistical trick, combining slower energy-related emissions with a one-off correction to land-based emissions to show past emissions consistent with RCP 8.5, and then to slyly suggest that this proved RCP 8.5 would continue to be the best fit until 2050, despite soaring emissions growth over that period in RCP 8.5 and slowing growth in the real world.
Peters and Hausfather remained two of the few climate scientists prepared to criticise RCP 8.5 publicly. Peters tweeted statements like: “Could global coal use per person increase five fold after being flat for 50+ years)? If your answer is bollocks, then you think RCP 8.5 is bollocks!” pointing out “SSP5-8.5 is basically a pathway with a World War II mobilisation to burn as much coal & gas as possible” (drawing a sanctimonious complaint about his choice of words from Bauer, but no effective rebuttal).
Dishonourable mention must also go to Joe Romm who, after I had linked to a post by Roger Pielke Jr, sent me an unsolicited email calling Pielke “a very bad researcher and terrible human being” and his work “crap”. You can judge Pielke’s credibility for yourself - I had him as a guest on Cleaning Up in June 2022. I have always found he takes more effort than his critics to stick to the orthodoxy of the IPCC. Indeed, if you really want to do a deep dive into how climate science went off the rails and prioritised public advocacy over robust science, there is no better starting point than Pielke Jr and Justin Ritchie’s 2021 review of the development and use of RCP 8.5 “Distorting the view of our climate future: The misuse and abuse of climate pathways and scenarios”, or their non-paywalled commentary article “How Climate Scenarios Lost Touch With Reality”.
In 2023 Hausfather blew Schwalm et al out of the water in a piece simply entitled Emissions are no longer following the worst case scenario, which contained the image at the top of this piece. At that point I felt I was pretty much done with RCP 8.5, and not coincidentally, also pretty much done with Twitter.
Caveat
I want to be absolutely clear. Nothing about #RCP85isBollox should ever have been taken to imply that did not accept the need for urgent climate action. Judge me by my efforts to promote such action.
Let’s be quite clear. Climate change is real. It is caused by humans. It is already having catastrophic impacts. These impacts will worsen. There are potential tipping points that could be existential. We have to act urgently if we are to avoid the worst impacts and the risk of tipping points.
But let’s also be honest. While we already see catastrophic local impacts, the potential catastrophic global impacts of climate change - many metres of sea level rise, the Amazon turning into tundra, all the things that kick off tipping points - will only play out over many decades, centuries and millennia. The urgency to act comes from the few decades we have to change course. I’m a systems dynamics guy, and what worries me is the lags in the system: when you build fossil assets, they emit CO2 for decades; those trap heat in the atmosphere for for centuries; that heat stays in the oceans for millennia. Once globally significant impacts do emerge, so clearly that no one can deny anthropogenic climate change is responsible, it will be impossible to turn the clock back.
We have to act consistently over many decades - that’s how long it will take to fundamentally reengineer our energy system, transportation and infrastructure to eliminate carbon emissions, while at the same time pulling more people out of poverty and improving human welfare. This means we need a broad coalition for action. If we start with 51% political support, then inevitably when times are tough, the mandate for continued action won’t be there. That is why it is so important that the rationale for climate action is built on the most robust science possible, to appeal across political divides.
Before you take to the comments to point about how temperatures are running ahead of the models, please stop. You would be talking about climate sensitivity not emissions or concentration scenarios. The C in RCP stands for concentration. The RCPs are pathways of concentrations. If we are seeing temperatures running ahead of those predicted based on CO2 concentrations, or impacts running ahead of those predicted based on temperatures, then that is what we must research. We can’t just pretend that we will be in world of 1,100 ppm of CO2 by 2100, when the current trajectory takes us to 540 ppm. That’s not science.
Similarly with tipping points. Yes, they are real, yes, urgent action is needed to avert them, but no, we won’t be seeing their impacts this century. And no, they are not well characterised in the science - partly because the step function used to explore climate tipping points was almost always RCP 8.5. If you want to know how long a kettle takes to boil over at setting 5, you can’t time it on setting 10 and confidently state it will take twice as long - these things are non-linear. Tipping point research needs to start with plausible scenarios - that word again.
The denouement
Sadly, with the turn in the political environment around 2024, everything I warned the climate community about came to pass.
There was always going to be a moment of popular pushback against climate action, at the moment when climate measures started to affect people’s personal choices in terms of transport and heating solutions. It was very unfortunate that moment came just as there was a huge surge of energy price inflation, largely driven by the bounce-back from Covid and then the Russian invasion of Ukraine, but not helped by policy-makers taking their eye off energy costs in their zeal for an unachievable 1.5C and an net zero 2050.
The reliance of climate science on the implausible RCP 8.5 presented opponents of climate action with an enormous and easy attack surface, just as I had warned. In the end it came to the attention of President Trump, who called out RCP 8.5 by name in his directive on Restoring Gold Standard Science, saying: “agencies have used Representative Concentration Pathway (RCP) scenario 8.5 to assess the potential effects of climate change in a “higher” warming scenario. RCP 8.5 is a worst-case scenario based on highly unlikely assumptions like end-of-century coal use exceeding estimates of recoverable coal reserves.”
What next
The tragedy of the #RCP85isBollox debacle is that the climate community spent a two decades telling people we were heading for 4°C to 7°C of warming, and quantifying the benefits of action by comparing RCP 8.5 with RCP 4.5 (looking at you, editors and authors of supposedly authoritative U.S. National Climate Assessments). The reality is that we are already doing better than RCP 4.5. and are heading for 2.5°C to 3°C of warming*, but still facing a very dark future of climate disruption. The need for urgent action to bend the curve down towards 2°C and below.
We need to learn the lessons. It is not just that it is intellectually flawed to exaggerate science. It’s morally wrong to do so in order stir up fear in pursuit of public policy objectives, no matter how desirable those objectives are. Climate scientists should be held to the same standards as medical or pharmaceutical researchers. It is also pointless: the current political atmosphere is surely proof that the majority of the general public simply cannot be frightened into climate action by stories of future impacts, no matter how luridly described.
We face a huge communications challenge. The public is blissfully unaware that most of the media coverage of future climate impacts is based a scenario that is now officially implausible. We have to explain to people that everything they thought they knew about climate impacts is probably wrong - but that nonetheless it is vital and urgent to act on climate change.
That is all for tomorrow. Today is a day to celebrate: it’s a good day for science. It's official: #RCP85isDead!
* These are the central ranges cited by major institutions like the IEA, UN, Bloomberg, ClimateAction Tracker, etc. It is important to note that these are only central bands, i.e. there are also uncertainty ranges above and below them.
Well, that was a long article with lots of assertions and no support, no footnotes, proving you were right all along.
However, I am living in the drought at the top of the Colorado River Basin, which has stopped providing the amount of water it is legally expected to supply, and that is likely to harm millions of people.
The Colorado River is a local collapse, so it is not important to your picture of Climate Change.
There are already people displaced by climate every year, all over the world, and especially by heat and lack of water. What percentage recover, financially and otherwise?
Errors should be corrected, yes. But if you think errors in climate science make a difference in what people do, as opposed to all the lies about climate science, I think you are not paying attention. It's kind of sweet to believe that the science matters to the deniers, at this point.
My husband just got a fire call for a fire in Grass Valley, Wayne County Utah, and left. That's not far from the 2025 Monroe Fire. Chances are it's a wildfire, a grass fire in that location..
It would be a nice peice of data science to figure out when the damage from climate change and hydrocarbon extraction will be greater than the economic benefits of not changing sooner.
Kudos on exposing RCP8.5 before it was popular to do so.
“The reality is that we are already doing better than RCP 4.5. and are heading for 2.5°C to 3°C of warming, but still facing a very dark future of climate disruption.” Genuinely interested in what catastrophic impacts you expect with this level of warming, given that all severe impacts are found with RCP8.5?