Climate 201: What On Earth is a Climate Change Scenario?

Author’s note: In 2020 I started publishing a series of podcast episodes under the title of “Climate 201”. The idea was to introduce topics in climate change, climate science, and climate policy — which I’ve studied for some years as a student- and explore them — and their implications — in much more detail than the simplistic framings you often see on the news. Ideally in the process I wanted to answer any questions that my audience had about climate change. By the end, the interested layperson should be up on some of the jargon and nuances of the discussion surrounding climate change.

You can find all of the episodes in the Climate 201 series here.

What on Earth is a climate scenario/RCP?

Hello, and welcome to Climate 201.

One of the things we discuss most often when it comes to climate change is uncertainty. The earth is an incredibly complex system, and human activities influence it in a number of different ways, which can themselves interact in a complex way with natural processes and cycles that occur in the ocean and atmosphere. Consequently, when we kick the system into a place we’ve never seen before — despite our best attempts to understand and model it — there is inevitably going to be some uncertainty about the results.

This is reflected in the range of estimates that we have for something like the climate sensitivity, the amount that the climate will warm if we double CO2. According to the IPCC, this is somewhere between 1.5C and 4.5C, with a best guess of around 3C. We’ll discuss this more in another episode.

However, the largest measure of uncertainty in what will happen to the climate over the next century is about what humans do. It’s not yet clear which path we will take in terms of our emissions. So if you want to attempt to model what will happen to the climate in the future, you have to contend with this uncertainty.

Some early climate modellers didn’t worry about this too much and just tried to answer more specific questions such as “What happens to the climate if we double CO2?” or “What happens to climate if we increase our CO2 emissions by 1% a year?” [The average % increase of CO2 over the last 250 years or so has been around 3% a year, just for reference — in the last couple of years this has been around 1% a year, but in 2017 it was 2.7%, so it’s very variable and dependent on a number of different factors we’ll talk about in a future episode.]

More recently, though, there have been a lot of attempts made to try and compare the potential outcome across different ranges of potential human actions. In other words, how is the climate likely to change in the next hundred years, depending on what we do? How much depends on the decisions we take? And this entails the creation of climate scenarios.

Now obviously the creation of climate scenarios is an extremely difficult thing — because, in a very real sense, you are effectively attempting to predict the future. There are a huge number of factors that contribute to our greenhouse gas emissions in the future. Does the world economy continue to grow, or does it saturate at some point? Does the world population continue to grow? How do developing nations grow — do they end up burning fossil fuels at the same rate as Americans or Europeans, or do they grow with alternative technologies? How does technology change — do we find, develop, and use low-carbon solutions in place of fossil fuels? We’re talking about a century here; perhaps we find some new use for fossil fuels, or some new process that’s in very high demand that produces greenhouse gases. Does the availability of fossil fuels get worse as they begin to run out, which starts to price them out of competition with the alternatives? How do our agricultural practices change? How about transport? And so on, and so on.






Particularly in the 1990s and the early 2000s, when a lot of climate scenarios were first developed, things could plausibly evolve in a number of different ways. Renewables were still very expensive, although research was ongoing; it was clear that globalisation was going to continue, leading to rapid economic growth in the “developing world”, which might mean more consumption of fossil fuels; there was no globally agreed-upon target for reducing greenhouse gas emissions or setting any kind of global temperature target. You could plausibly argue that the world might start taking climate seriously and start cutting their emissions… or that they would plough on with reckless abandon to dig up and burn virtually every fossil fuel that was available.

When it comes to climate models, there are obviously dozens that everyone is using. If you don’t all model the same thing, then you’re clearly not going to be able to make comparisons between the results of the models. So for this reason, starting in the early 90s, climate scientists developed a set of “representative scenarios” which might indicate, broadly, predictions for how things might develop in the future. Some have lower emissions, some have higher emissions, and so on. This way, everyone can run the same scenarios, and they can then compare the modelling results in like-for-like scenarios.

So obviously one thing that should be clear is that the actual future — what we’ll actually experience — is not going to correspond exactly to any one climate scenario. They’re not perfect predictions — instead, they just hope to span the space of all possible futures. And, by the way, as should be obvious by now, you can’t let the perfect be the enemy of the… something… when it comes to this type of research. The same is true of models, which are obviously imperfect and none of which will actually correspond exactly to reality. If we waited until we had perfect computer models and perfect knowledge about how emissions would develop in the future before we were willing to say anything about how climate change was going to turn out… we wouldn’t need to run the simulations… because we could just look outside.

And the other thing that should be clear is, given how influential the scenarios developed by the Intergovernmental Panel on Climate Change can be in terms of the climate-changed worlds that we model, and the conclusions that we draw from that… that yes, there is a lot of politics surrounding what should go into these scenarios, how they should be used, and how they should be referred to, and what they really represent. But we’ll come onto that later.

The earlier sets of scenarios aren’t used as much any more by recent modelling, so I’ll just briefly mention that they were called SA90, IS92, and SRES. The SRES scenarios came from the “Special Report on Emissions Scenarios.” What made these scenarios slightly different was that, while previous scenarios had just specified where energy was coming from, or dealt with predictions for population, energy mix, and fossil fuel use, the SRES tried to map things out in terms of “storylines” for how humanity might change into the future that were a bit more coherent.

For example, one version of the future might involve rapid economic growth, which is what governments prioritise at all costs, population increasing to 9 billion, and everyone gradually “converging” towards a similar lifestyle in terms of their energy demands as globalisation continues apace. These were the SRES A1 scenarios — and they were further divided based on how the world gets its economic growth — whether the world focused on fossil fuels, a more balanced mix of technologies, or new technologies that reduced greenhouse gas emissions, for example.

Another version of the future might involve economic growth but also a transition away from energy-intensive industry and towards a more service-based economy — as we’ve seen in the UK, which has helped to reduce emissions although unintentionally. Perhaps population grows more slowly, and globalisation becomes more about strong climate policies, greater emphasis on energy efficiency and ecological solutions — a world that prioritises the environment more. These would be the B1 scenarios.

And each set of scenarios also tries to take into account possible geopolitical differences — maybe globalisation stops or is reversed, and the world is more divided between some groups that act together to try and reduce environmental harms, and others that don’t. These contrasting scenarios are the A2 and the B2 scenarios, respectively.

Once these scenarios for how the world might change in the future have been developed, the storylines are converted into what they might actually mean for emissions and concentrations of greenhouse gases, which is then what you feed into the models to determine how the climate changes.

The SRES scenarios have since been superseded by two sets of scenarios, which are still in use today and which I’m going to spend most of my time talking about.

The first and arguably most important, certainly the most influential, set of scenarios are the RCPs. RCP stands for Representative Concentration Pathway. The concentrations here are concentrations of greenhouse gases.

So all that the RCPs really are is a set of numbers — the concentrations of each greenhouse gas in the atmosphere projected to be there in each year, from 2005 to 2100. These pathways are chosen to give a certain amount of net radiative forcing by the end of the century — remember that the radiative forcing, measured in Watts per square metre, is a measure of how much additional heating is occuring due to the presence of greenhouse gases. Currently we estimate that the net radiative forcing due to humans is around 2.29 watts per square metre, for reference — that’s from the IPCC’s report in 2014, so the figure is a little out of date now. That’s including the positive effects from greenhouse gases, and the negative effects from aerosols.

So in terms of the RCPs, then, the numbers they’re given represent the radiative forcing due to humans in the year 2100. For this, we have the most important four:

RCP2.6, RCP4.5, RCP6, and RCP8.5.

RCP2.6 is obviously the most ambitious of these scenarios, where the net contribution from humans to radiative forcing is about the same in 2100 as it is today — we’ll talk a little bit more about what it means later on.


RCP4.5 is a little less optimistic, RCP6 less so, and RCP8.5 is really the doomsday scenario where we would have to burn colossal amounts of fossil fuels, right the way up until the end of the century.

These are by far the most widely modelled climate scenarios. For example, when I search for RCP8.5 in “Google Scholar”, I find almost 18,000 scientific papers have been published which use this scenario. And it’s important to understand that quite often, when you read headlines in newspaper about, say — “scientists predict that droughts will be 3x as common by the end of the century”, or “scientists predict that hurricanes will be 50% wetter by the end of the century due to climate change” — more often than not, the underlying study will end up using one of these scenarios. If you don’t know which scenario you’re using, or how likely it is, then it’s quite misleading to treat all predictions as equal — because, of course, all these model results are assuming that humanity behaves a certain way in the future to generate these greenhouse gas concentrations.

One thing you’ll notice about the RCPs is that they are essentially concentration pathways chosen to give you these numbers for radiative forcing — which are essentially chosen to span the range of feasible outcomes fairly well. But we only specify the greenhouse gas concentrations in the atmosphere, and not how we get to those concentrations. So you have to back-track from these concentrations a little bit to try to figure out what kind of world you’re going to be in for each of these scenarios, to figure out what greenhouse gas emissions will lead to which concentration pathways.

And converting from emissions of greenhouse gases to the actual concentrations of those gases that are in the atmosphere is not always straightforward, because of the Earth’s carbon cycle. For example, half of the carbon we emit is absorbed by the oceans. But in the future, the ocean will be less effective at absorbing CO2 as it starts to saturate and warms up. So, depending on how you model those ocean processes, it might be that emitting a tonne of CO2 in the future will lead to a higher concentration of CO2 in the atmosphere, because the ocean is absorbing a smaller fraction of the CO2. Some models include processes whereby carbon dioxide can be added to the atmosphere from the world as it warms — for example, due to the destruction of forests which act as reservoirs of carbon, or from the melting of permafrost in the Arctic which releases methane. The magnitude and nature of these “carbon-cycle feedbacks” is still much debated about, but obviously they will impact how your emissions of greenhouse gases from the Earth turn into concentrations in the atmosphere — for example, if you emit CO2 which warms the planet and causes it to emit more CO2 through these processes, the overall concentration would be higher than from your emissions alone.

Nevertheless, we can still work out some things about the worlds that are implied by the different RCPs. We’ll focus mostly here on the best-case and worst-case scenarios here, RCP2.6 and RCP8.5, with a little on RCP4.5. RCP6, for whatever reason, is hardly used at all.

So let’s describe what typically might happen to emissions in these scenarios. In RCP 2.6, CO2 emissions peak in 2020 and sharply decline after that, halving globally by 2040 and getting very close to zero by 2070 before becoming net negative by the end of the century. In other words, by the end of the century, not only have global emissions fallen to zero, but we are sucking CO2 back out of the atmosphere — which we’ll discuss more in our episode on negative emissions. Typically, when this is run in most models, this leads to CO2 concentrations that peak at around 450 parts per million in the atmosphere — the CO2 concentration is currently around 416ppm and rises by 2–3ppm a year. In most models, temperatures stay below 2C throughout this scenario, which means it’s the only one of the scenarios that is compatible with the Paris Agreement, and the most ambitious mitigation scenario that’s considered.

RCP4.5 is a more pessimistic, albeit perhaps more realistic scenario. In this scenario, emissions gradually increase until the middle of the century, peaking at around 2040–50 slightly higher than they are today. They then gradually fall, before hitting a constant “floor” of around 5 GtC a year, around half of today’s value, in 2080. I suppose this is meant to represent emissions that are hard to get rid of e.g. due to industry, agriculture, and transport as opposed to power generation. By the end of the century, CO2 concentrations are around 520ppm and continuing to rise slowly. The net result, in most models, is that the planet heats by 2.5–2.7C by the end of the century — so substantially above the agreed-upon 2C of warming from Paris.

RCP8.5 is the scenario where we burn fossil-fuels like crazy. Emissions rise very rapidly throughout most of the century, eventually stabilising at around three times today’s levels. The result is a CO2 concentration that rises to a staggering 950ppm by 2100, and temperatures that skyrocket, perhaps to between 4–6C above where they were preindustrially. To call this a doomsday scenario is almost an understatement.

So you can see that the major climate scenarios span a staggering range of different possibilities. We have a world where rapid climate action is coordinated globally — emissions never rise again, peaking around last year, and breaking the trend of pretty much continuous increases which has held since the 1700s. In this world, we slash our emissions, eventually to the extent that we can suck billions of tonnes of carbon dioxide out of the atmosphere every year by the end of the century. And when this is done, we keep to the 2C Paris target, just about. This is RCP2.6.

On the other end of the spectrum is a world that burns a truly phenomenal and heroic amount of coal and other fossil fuels, ignoring skyrocketing temperatures as they do so, and the net result is a climate hell-scape that pushes large parts of the world into being essentially uninhabitable by the end of the century. This is the RCP8.5 world. And then there are a whole bunch of scenarios in between.

=> obviously controversial, because scenarios determine what we simulate with finite computing time, influences predictions that people make.

=> Ultimately it amounts to a particular set of lines on a graph. The question of how you interpret them is huge, though.

=> in particular, interesting; what would climate experts, economists, scientists etc. draw as the “most likely” scenario? I’d bet that it’s none of these.

Now it’s easy to be incredibly critical of these scenarios — there are flaws with each of them, and I am sure that no-one who made them would argue that any of them are a “best guess” for what will happen. They’re supposed to span a range of possible futures. But this is quite a subtle interpretation that is often lost on these things. When a climate scientist sees that some study is making some conclusions, the first thing we’ll do is look to the scenario they modelled and make our judgements based on this. But the vast majority of people likely have no idea about these scenarios (and why would they, because they’re obscured in this jargon?)

One of the major controversies that has shown up in recent times surrounds the RCP8.5 scenario, and how it should be interpreted. I have my own views on this, but I’ll lay out some of the different points for you here.

Some people have referred to RCP8.5 as a “business as usual” scenario, and it’s sometimes referred to as “business as usual.” But what does business as usual mean? For some, that implies “if current trends continue”. In other words, the “business as usual” scenario is what happens if we don’t include additional climate policy. For some, it carries the implication that this is what we’d do if we had no climate policies in place at all — if we had no knowledge of climate change whatsoever and no one had ever done anything that we had to mitigate it.

The COVID-19 pandemic is perhaps a good example. A worst-case scenario might assume that people continue meeting up in the same way they always did — even as the bodies are piling up. Whereas in actual fact, that worst-case scenario is unlikely to be realised, because people will to some extent — see what’s going on and change their behaviour.

In my opinion, if RCP8.5 is anything, it’s really a worst-possible-case scenario. Perhaps you could argue back in 2000 or 2005 when this was first developed, you could imagine that something like RCP8.5 could actually happen; if the world was run by fossil-fuel interests who wanted to do nothing more than exploit the world’s fossil fuel reserves to their maximum possible extent to spur economic growth up until the end of the century. In that sense, you could make a case for it then as a “worst-possible-case” scenario, if nothing whatsoever was done.

But the reality is that a lot of things have been done. As I never tire of telling you, renewable electricity is now the cheapest form of electricity generation across 85% of the world’s population. Carbon regulations have been slow to come, and the fossil fuel industry is still being subsidised, but some positive change has arrived as economies have increasingly chosen to move away from coal to natural gas and renewables. Coal is an expensive, dirty option with a poor reputation — and as countries get wealthier, they often enforce stricter air pollution standards; the air pollution alone from activities like burning coal is part of what currently contributes to one in six deaths, and while I would argue that our world pays nowhere near enough attention to environmental and sustainability issues, we’re not quite as awful as the humans who are in charge of the RCP8.5 world — perhaps largely because it’s not even profitable to burn all that coal any more.

I’m not the only person to make this criticism — and so RCP8.5 has to be viewed as a worst-case scenario if anything. Given the progress we have made on climate, and the progress we’ve made on making renewables cheaper in the last few decades, for something like RCP8.5 to happen nowadays would probably require a shift away from climate policies and towards fossil fuels… which might happen, but it feels increasingly against the tide of history.

Others feel even more strongly about this — Google “RCP8.5 is bollocks” and you’ll see a lot of people in climate discussing this.

One piece of valid criticism arose in a paper by Jianlang Wang et al. In 2017, which was “The implications of fossil fuel supply constraints on climate change projections: a supply-side analysis.” Because one criticism that is valid of the RCPs is that they don’t take into account calculated limits on the possible supply of fossil fuels, instead just considering how demand might change.

Very long-term listeners will remember from our TEOTWAWKI episode, Peak Oil, that there is a lot of debate surrounding how many fossil fuels are available and whether they will ever “run out” in a conventional sense. New reserves are being discovered all the time, as are new methods of extraction, like fracking for natural gas which was hardly done a few years ago, or the extraction of oil from shale rather than wells. It might be the case, as the BP economist Christoph Ruhl chillingly argued, that “anything can be turned into oil, if there is demand”… and “oil will never run out because there will always be a price at which the last drop can clear the market.” But this is really just semantic trickery. At some point, supply constraints do kick in, if only by making new exploration and extraction of fossil fuels more expensive than the alternatives.

This paper, by Wang et al, tried to assess those supply constraints, and their argument is essentially that fossil fuel production will peak and prevent a scenario like RCP8.5 from ever happening.

Glen Peters and Zeke Hausfather discussed this controversy in an op-ed in Nature called “Emissions — the business as usual scenario is misleading.”

They said:

“RCP8.5 was intended to explore an unlikely high-risk future. But it has been widely used by some experts, policymakers and the media as something else entirely: as a likely ‘business as usual’ outcome. A sizeable portion of the literature on climate impacts refers to RCP8.5 as business as usual, implying that it is probable in the absence of stringent climate mitigation. The media then often amplifies this message, sometimes without communicating the nuances. This results in further confusion regarding probable emissions outcomes, because many climate researchers are not familiar with the details of these scenarios in the energy-modelling literature.”

I’ll just interrupt here should point out that this is not necessarily something that patronises people, to say that they don’t understand. I have friends who are literally PhD candidates in Atmospheric Physics, far smarter than I am, who are very active and engaged in the climate debate, and who — when I asked which was the most likely climate scenario — said the worst-case scenario, RCP8.5. So this is not something that comes from ignorance necessarily, you have to be quite deep in the weeds to realise this.

Zeke and Glen continue:

“Happily — and that’s a word we climatologists rarely get to use — the world imagined in RCP8.5 is one that, in our view, becomes increasingly implausible with every passing year5. Emission pathways to get to RCP8.5 generally require an unprecedented fivefold increase in coal use by the end of the century, an amount larger than some estimates of recoverable coal reserves6. It is thought that global coal use peaked in 2013, and although increases are still possible, many energy forecasts expect it to flatline over the next few decades7. Furthermore, the falling cost of clean energy sources is a trend that is unlikely to reverse, even in the absence of new climate policies7.

Assessment of current policies suggests that the world is on course for around 3 °C of warming above pre-industrial levels by the end of the century — still a catastrophic outcome, but a long way from 5 °C7,8. We cannot settle for 3 °C; nor should we dismiss progress.”

That fivefold increase in coal use, in particular, would represent a pretty incredible decision to turn away from cheap alternatives and start to burn coal like there’s no tomorrow. And, according to some researchers, there may not even be sufficient coal supply to burn that much.

However, there is a slight wrinkle to point out here. One of the things that we mentioned was that the RCPs only include concentrations of greenhouse gases and they don’t specify how those gases might end up in the atmosphere. It’s possible that, if climate sensitivity is higher than we thought it was, or if those carbon-cycle feedbacks — where a heating planet emits its own carbon, remember — could mean that a lower-emissions trajectory still leads to the same eventual amount of greenhouse gases ending up in the atmosphere. So some will argue that it’s still not beyond the realms of possibility that something like RCP8.5 could happen, particularly if these climate feedbacks turn out to be a dominant effect towards the end of the century. Nevertheless, I think most people who’ve given this serious thought would say that a world with little climate action probably looks more like RCP6 or RCP4.5 than it does RCP8.5.

So what are the dangers here? Well, I think it should be acknowledged that there is a temptation to try and scare people who don’t take climate change seriously into taking action by emphasising these worst-case scenarios. But if you take this too far, it tips over into climate doomerism, which I’ll talk more about I think in a future episode.

In essence, it’s a behavioural-psychological debate. If you tell people good news about the progress we’re making on climate, but that there’s still a long way to go before we can be sure that we have really tackled the problem, does this make them more likely to make the efforts that are required? Or does it make people complacent? Similarly, if you emphasise the worst-possible case scenario, does the fear of the potential horrors that — realistically, still just about plausibly — might be visited upon us motivate people to take positive actions to head off that disaster? Or does it lead to a type of paralysis — an assumption that we’re doomed, and that nothing we do can possibly matter, or an assumption that it would cost too much or prove too unpopular to actually act on the problem, and we may as well try to ignore the upsetting future — or, maybe worse still, defend ourselves against it while letting the rest of the world go to hell? If you still call RCP8.5 business as usual, in spite of the progress that has been made, then doesn’t it imply the efforts we have made have been useless so far? When in fact, people since the 1980s who have been motivated by trying to fix climate change have made the worst-case scenario less likely.

Yet popular books like David Wallace-Wells’ “The Uninhabitable Earth” arguably focus too much on this worst-case scenario. A lot of the examples of the worst-case climate impacts that Wallace-Wells draws on are from papers that model RCP8.5, the worst-case climate scenario.

I won’t go into too much detail on this book, or the article that it began as, because it’s been done brilliantly by a team of scientists on the website “Climate Feedback” which allows scientists a forum to comment on the claims in popular media articles. Go and google Climate Feedback on “The Uninhabitable Earth” if you want to see more on this.

Naturally I have a pretty nuanced opinion on this. I definitely think there are doomers out there who are way too keen to emphasise how terrible things might be and exaggerate the danger for their own political ends, and I think that’s counterproductive. I also think that there are people out there who are too keen to say that the climate problem is basically solved when actually, it still requires massive efforts from us to get anything close to the Paris target.

We can all agree, surely, that the best way to make decisions is when we have the best possible information. If all of your attempts to adapt to climate change are based on a worst-case scenario that’s hugely worse than anything that is likely to happen, you’re obviously not going to be preparing correctly. We live, unfortunately, in the real world, where people need to be able to assess different priorities. Frankly, I think climate change is going to be bad enough, and a difficult enough problem to solve already, without dreaming up doomsday scenarios that are, hopefully, getting increasingly unlikely. We need to understand what’s really likely to be ahead of us, and have a realistic assessment of how much progress we’ve made.


People will always have their own motivations to stake out these extreme positions. Extreme positions are simple, eye-grabbing, attention-catching, and easier to fit in a tweet. If I was doing podcasts about how climate change was going to kill everyone and we were all totally doomed, or podcasts about how it was all an exaggerated hoax, I bet I’d get more listeners than my attempts to describe the nuances of the ongoing debates in the climate community, the uncertainties and the shades of grey. People don’t want to hear that stuff — and so a lot of people stake out these extreme positions. And, as always, the reality is usually somewhere between the caricatured extreme positions.

With all that said, is RCP2.6 any more realistic than RCP8.5? The fact that we have four main climate scenarios — and in only one of them, the most optimistic, do we actually achieve our “climate target” of 2C, which was once considered “unacceptably dangerous warming”, should tell you something about how bad the situation is.

The implications of RCP8.5 — that we’ll quintuple our use of coal over the next century, for example — might seem ridiculous. But then again, so might RCP2.6, and the idea that we will put a brake on the runaway train of increasing emissions and that they will stop increasing immediately — despite having increased pretty much every single year for the last three centuries. So might the idea that we can totally transform how we do agriculture, energy production, industry and transportation in the next few, short decades — and then end up spending billions of dollars a year just to clean up our historic mess with negative emissions — yet this is what is required for something like RCP2.6 to happen.

And I think there is a growing consensus now that RCP8.5 should be regarded as an absolute-worst-possible case climate change scenario — even though it’s quite often referred to as “business-as-usual”, while a lot of what has to happen for RCP2.6 to happen is also quite dramatically optimistic. So I urge caution when you read some disastrous headline about the impacts of climate change: if it’s based on RCP8.5, just remember that is quite an unlikely trajectory for us to take in the 21st century.

As I say, climate change is definitely going to be bad enough — and worth doing everything we can to fight — without overstating how dangerous the situation is, or indeed understating how much progress we have made. The actions humanity has taken, in terms of finding cheaper alternatives to fossil fuels and pursuing climate goals, has made a worst-case scenario like RCP8.5 — where the effects of runaway climate change would bury our civilization — much, much less likely. It wasn’t always clear that this would be the case, and a civilization that just burned all of the available fossil fuels to generate power may well have ended up on an RCP8.5-like trajectory. But we have made progress. I think that’s something to celebrate, while acknowledging that avoiding the worst possible scenario we could dream up hardly means we’re finished.

Another criticism that is often made of the widespread use of these scenarios is that the way they are defined inevitably implies certain things and certain limits. For example, you might find it interesting that — although the Paris Agreement sets an aspirational target to limit global warming to 1.5C, and the IPCC had an entire report on the impact of limiting global warming to 1.5C, none of the scenarios we’ve discussed would actual limit temperatures to below 1.5C — in most models, RCP2.6 just keeps to below 2C. Although it’s true to a certain extent that the RCPs just tell us what concentrations will be, in practice RCP2.6 does involve negative emissions at the end of the century — it assumes that we will be able to suck billions of tonnes of CO2 out of the atmosphere.

So you can see again that, even though it’s a line on a graph, there are inevitably some political and technological assumptions that get implied by the scenario, one way or another, based on what the people who created the scenario considered feasible. It’s more or less assumed that it would take until 2040 to halve emissions, and 2070 to get them to net zero, and that negative emissions would be required thereafter. Essentially, the original scenario modellers assumed that it wouldn’t be possible to cut carbon emissions faster than in RCP2.6, which would need negative emissions later on.

This was dealt with by Grubler et al, who considered a new “Low Energy Demand” scenario, RCP1.9, which is now being modelled. In this scenario, CO2 emissions are very rapidly cut in the near-term — falling by about half every decade — with what the authors envision as a combination of energy efficiency, renewables, and demand-side changes in people’s behaviour, all of which result in a world with a much lower demand for energy. This, in turn then allows net emissions to fall close to zero by 2050. Now, this scenario may be unrealistic — but by allowing ourselves to imagine and model it, we can get a better understanding of what might happen if it were to come true.

This is by no means a comprehensive criticism of the IPCC scenarios or the use of scenarios in climate modelling, but one additional point I’d raise is to say that there has also been a good deal of modelling — for example, by the HAPPI project — which determines what happens to the climate in terms of temperature targets, 1.5C, 2C, and 3C. We deal in temperature targets, even though the scenarios don’t, so it makes sense to have this stuff in there to fill in some of the gaps.

So I end the whole thing with appealing to you to remember what climate scenarios are: these are scenarios. They are potential futures, none of which is particularly likely to actually be realised, but which together show you how climate change could transpire across some cases that are best-imaginable, and some doomsday worst-case scenarios. They should always, always be taken with a grain of salt. They are not supposed to be predictions, much less prescriptions, for how climate change will unfold; instead, they just allow climate modellers to model a range of different future scenarios that they can all agree on, to try and find the differences that arise due to different levels of human action — from RCP2.6 where we do quite a bit to mitigate climate change, and RCP8.5 where we do quite a lot to exacerbate it from where we are today. But these caveats certainly don’t mean the scenarios are immune from criticism, and indeed the debate around how appropriate each one is will continue to rage.

They are not perfect, but they certainly have their uses. There is always going to be a role for exploring the worst-case scenario for what could happen with climate change — if only so we know what our actions are avoiding and helping to avert. But if we are too pessimistic, or too unrealistic, we risk focusing on the wrong things and creating the wrong incentives for people. So next time you see a study about climate change reported in the media, or a paper on climate change, try and see what happens in these different scenarios, and understand their limitations.