Opening the future
29 Mar 2011
By 2100, science will boom, and with widespread birth control — and a resurgence of puritan morals — the population will be kept at a sustainable level, about two billion. “Industrial enterprises that formerly befouled the world with smoke, refuse and cinder heaps” will be “cleaner in their habits than a well-trained cat.” So H. G. Wells imagined in his 1933 novel, The Shape of Things to Come.
The future imagined by Wells is strikingly similar in many of its details to one evoked much more recently by researchers aiming for their own sense of the shape of things to come. As part of their work assessing the scientific understanding of climate change for the United Nations in 2000, a small team of economists, climatologists and other researchers created a set of scenarios, each of which represented a possible future world. One termed ‘B1’ represented a world with “a high level of environmental and social consciousness combined with a globally coherent approach to a more sustainable development.”
Wells’s goals and motivations, too, have a lot in common with those of scenario researchers today — so much that he has been called ‘the founder of futures studies’, the field that pioneered the scientific use of scenarios1. Through his fiction, Wells made a serious effort to understand the world’s driving forces and trends, with the hope that it could help society make a choice about the best path to follow. Now, with a major revamping of how the climate research community builds and uses scenarios, scientists are coming that bit closer to realizing this ambition.
The change in scenario development now taking place is like switching from Windows, a closed-source operating system, to an open-source system such as Linux. The new process will make it much easier for researchers, policymakers and activists to take socio-economic scenarios created under the aegis of the Intergovernmental Panel on Climate Change (IPCC), and add in details or tweak the assumptions — just as open-source software allows a whole community of programmers to tweak and add on to programs. Instead of shopping for a particular future that looks appealing, this process could encourage people to figure out how to create the kind of future they want — and how to avoid the futures they fear.
The earlier process of scenario development, which took years, started with a handful of qualitative ‘storylines’, and then simulated effects on the planet from each of these scenarios. However, the new process is starting instead with four distinct possibilities for greenhouse-gas emissions, and leaving it more open as to how the world might follow one or another path. According to Brian O’Neill of the National Center for Atmospheric Research in Boulder, Colorado, who develops models to explore climate scenarios, this will open up important new avenues of research. “It will be more flexible and allow more types of future to be considered,” including assessment of various mitigation and adaptation options. Critics of past attempts at scenario development say that this is a step in the right direction to a more open evaluation of policy options. But the risk is that the variety of scenarios under development will become overwhelming; and how so many scenarios can best be used to inform us about the future is a question still under debate.
The last time the IPCC released a set of scenarios was in 2000, after a drawn-out process of development that began in the mid-1990s.2 These were known as Special Report on Emissions Scenarios (SRES). Now these scenarios “are essentially 15 years old, from the day of conception,” says Nebojsa Nakicenovic, an economist at the International Institute for Applied Systems Analysis in Laxenburg, Austria, who led the project. “They have limited shelf-life,” he says. Also, the initial SRES scenarios “did not include any climate policies, beyond what was agreed at that time, which wasn’t much,” because that was the limit of the IPCC’s mandate at the time, says Nakicenovic. The new process, however, will make these issues more obvious, and also make it easier to do scenarios research that is relevant to policymakers.
The process of creating the SRES scenarios started with a set of storylines, each giving a sketch of a possible future. These qualitative descriptions highlight trends and values that would characterize a specific path that the world could follow, such as being mainly geared towards high economic growth — exemplified by the high-emissions A1 scenario — or dedicated to preserving the environment, as the low-emissions B1 scenario. By tweaking the assumptions behind each storyline — such as how much people care about sustainability, equality or material wealth — scenario developers could factor in faster or slower population growth, for example. They could also attempt to capture qualities that are harder to quantify, for example better or worse governance.
To turn these narratives into something quantitative, four teams of researchers each used their own integrated assessment model — a computer simulation that combines a simple version of an economic model with a simple version of a climate model. Changing the settings in an integrated model so that it represents a specific storyline, and then running the simulation, gives rise to a specific ‘socio-economic scenario’ for the future, suggesting how economics, populations and technologies may change. The B1 scenario, for example, is meant to describe a world deeply concerned with the environment and sustainable development. It could be called a ‘peak’ world, in which fossil-fuel use, population and emissions all reach a peak around mid-century and then fall sharply in the latter half of the century. By 2100, even without any specific policies to cut emissions, continued development of clean technology would cause emissions to drop to less than half of today’s level. However, the outcome could be starkly different. Wells’s various science fiction novels highlight the range of possible futures — from the gentle, eco-friendly world of The Shape of Things to Come (similar to the B1 scenario), to a planet ravaged by high-tech conflict and nuclear warfare in his 1914 The World Set Free.
Each socio-economic storyline can lead to a slew of potential climate impacts. But the climate component of integrated assessment models is very crude. So in SRES scenarios, for any given storyline the value of the emissions generated by the integrated model was fed into a state-of-the-art climate model, which can generate more realistic results for the temperature rise, rainfall shifts and other changes. This whole process — from generating storylines to climate impacts — was grindingly slow. The SRES report got started in the mid-1990s, but the storylines weren’t published soon enough to give researchers time to use them as the basis for research that was included in the IPCC’s third assessment report in 2001. Even by the next assessment report, published in 2007, many of the studies cited were still based on scenarios published in the 1990s.
The new approach “is a reaction to a feeling that there’s not enough time to go through this long laborious process,” says economist James Edmonds of the University of Maryland in College Park, who has been involved in its development. Rather than the IPCC coordinating and overseeing the process, the organization has now taken a back seat to allow the scientific community to develop the scenarios independently. The response from scientists working on this for the IPCC’s fifth assessment report — due to be published in 2014 — has been to turn the old approach on its head. Instead of starting with the qualitative storylines, they are starting out with greenhouse-gas emissions trajectories. These are known in IPCC terminology as representative concentration pathways (RCPs) and span a wide range of plausible future-emissions pathways (see Box 1 below). Climate scientists are now running these alternate emissions trajectories in their models and are expected to release results this summer on how severe the impacts may be from each RCP3.
While the climate models are grinding away to uncover the likely impacts of a particular course of action, another process has been running in parallel — trying to understand the actions that would be necessary for us to follow any of the pathways, such as prioritizing the environment over economic growth, or favouring global cooperation over regional alliances. Such decisions, however, are political rather than scientific. By starting the scenario-development process with emissions pathways, rather than with storylines, and making the socio-economic scenarios open source, so to speak, it “avoids a bunch of thorny issues,” says Edmonds, such as how specific climate policies would affect emissions4.
With the new process, the emissions — the RCPs — are fixed. So instead of asking what would be the likely impacts of a certain group of assumptions, the question now is: what decisions and developments would take us down one path or another? “The socio-economic scenarios are now in the forefront,” O’Neill says, and that transparency could ultimately aid decision-making. “If we don’t put our assumptions about the future on the table, then we have the same situation we had in the run-up to the financial crisis, and that is, we blindly follow the assumptions about the future that are built into our disciplines and models,” says Angela Wilkinson of Oxford University, a physicist and futures scholar, who has developed scenarios for Royal Dutch Shell and more recently a set of financial scenarios. One of the reasons cited for the severity of the financial crash is that many derivatives traders trusted commonly used risk-analysis software that, it turned out, severely underestimated their risks5.
However, besides misunderstanding the risks of derivatives, much deeper assumptions went unquestioned, such as that “growth can go on forever,” Wilkinson says. To help the financial sector reveal and test its assumptions, Wilkinson and colleagues developed two different scenarios they call ‘growth’ and ‘health’. The first aims for high economic growth, whereas the second scenario, which aims for stability and health of the whole financial system, represents a major shift from current priorities6.
Aside from forcing scenario developers to take a closer look at their underlying assumptions, the new approach has the benefit of making it easier for decision-makers to explore more diverse policy options. After all, there are a myriad ways for the world to reach any given climate stabilization goal. “There are not just two ways to get to 550 parts per million [of carbon dioxide] — but two thousand or two million ways,” says climatologist Mike Hulme of the University of East Anglia in the UK. The danger in exploring a huge number of possibilities, says O’Neill, is “that people will think that we don’t know anything about anything — and that’s not the case.” Even if there are multiple ways to get to any certain goal, research could reveal patterns and trends among various scenarios — something O’Neill has been studying.
In a study published last year, O’Neill and colleagues used an integrated assessment model to analyse how various population-related parameters would influence future emissions. Previous studies have simulated changes in population size, but O’Neill and colleagues found that factors other than sheer numbers can have a sizeable influence on emissions. They found that an increase in the average age of the population in the coming decades could drive emissions down by 20% by 2100, for example, whereas including realistic trends on urbanization could push emissions up 25% by the end of the century7. Further studies could vary the assumptions and “explore, in a quasi-quantitative way, just how big of a difference it makes,” Hulme says.
Opening up the scenarios, and making it more straightforward for many researchers to tweak and add on to them — as with open-source software — will probably create a huge number of storylines, some differing subtly in their assumptions and details and some widely different. “You might see a proliferation of scenarios,” Wilkinson says. “It’s very complex and clumsy — but it’s probably the right process.” Hulme agrees. “I think the proliferation of these socio-economic scenarios will be unavoidable — and I think that’s a good thing,” he says. “It will help draw attention to where different assumptions and value judgments are being made.”
However, some are sceptical about how well the process will work. Granger Morgan, a policy expert at Carnegie Mellon University in Pittsburgh, Pennsylvania, is bothered by the huge number of possible scenarios that will emerge. “I keep asking people, ‘To what end?’,” Morgan says. Proponents say that exploring scenarios may help identify which policy levers to pull — but such answers are only “partially satisfactory,” Morgan says, as “you can tell an almost infinite number of stories about the future.”
To help tame the vast number of possibilities and to coordinate research efforts, the IPCC authors will endorse a small set of storylines, which many researchers plan to use as the starting point for their socio-economic scenarios. Researchers involved say these storylines will be published in draft form by this summer. They’re still hashing out how detailed each storyline will be, and what kind of information it will include, with possibilities including more detailed demographics and world oil prices.
Back in 1902, H. G. Wells gave a speech to London’s Royal Institution in which he proclaimed, “the time is drawing near when it will be possible to suggest a systematic exploration of the future.” He thought that humanity would soon identify laws of society and politics that would be as solid a science as geology. That still seems far out of reach, but researchers working on scenarios today share his overarching goal of understanding how our actions and assumptions today could change the shape of things to come.
Box 1: Representative concentration pathways.
It took more than two years to hash out the first step in the IPCC’s new approach to scenarios: choosing a handful of emissions trajectories, known as representative carbon pathways (RCPs). The RCPs then became the basis for new runs of the latest climate models. In parallel, the RCPs are also intended to be focal points for research on socio-economic scenarios, to see what kinds of developments might take the world along each pathway.
The IPCC authors chose four emissions trajectories to focus on, and have labelled them based on how much heating they would produce at the end of the century — 8.5, 6, 4.5 and 2.6 watts per square metre (W/m2). They chose four trajectories to avoid the common misconception that the middle scenario is the most likely. The figure shows the annual carbon dioxide emissions (in billions of tons of carbon) out to 2100 for each of the RCPs. Data from RCP database (http://go.nature.com/Rmyxyt).
The range covered by the RCPs is wider than that contemplated in previous IPCC reports, partly reflecting a general shift in outlook to one where possible future emissions trajectories look more extreme than they did a decade ago. But the RCPs also include a scenario that some think is incredibly optimistic. The 2.6 W m−2 scenario has greenhouse-gas emissions dropping to zero by about 2070, and then continuing to fall, so that the world’s emissions become negative — actually pulling greenhouse gases out of the air and locking them away — for decades. This pushes the bounds of what is plausible with mitigation, some experts say.
On the high end, the 8.5 W m−2 case, carbon dioxide levels soar above an enormous 1,300 parts per million by the century’s end — and are still rising fast. Some resource experts — such as Jean Laherrère, a petroleum geologist retired from French oil company Total — say this is unrealistic, because people won’t be able to produce enough oil, coal and gas to produce that much carbon dioxide. Nebojsa Nakicenovic agrees, but for different reasons: “the high end is impossible,” he says, because the impacts would be so severe that it’s inconceivable that the world would not take some kind of action. (Return to main text.)
1. Wagar, W. W. World Future Soc. Bull. 17, 25–29 (1983). ↑
2. Nakicenovic, N. & Swart, R. (eds) Special Report on Emissions Scenarios (Cambridge Univ. Press,2000). ↑
3. Moss, R. H. et al. Nature 463, 747–756 (2010). ↑
4. Pielke, R. Jr, Wigley, T. & Green, C. Nature 452, 531–532 (2008). ↑
5. Lessig, L. Why the banks all fell down. Newsweek (18 October 2008); available viahttp://go.nature.com/oUyJgf. ↑
6. Wilkinson, A. et al. Beyond the Financial Crisis (Univ. Oxford, 2009); available athttp://www.sbs.oxford.edu/financial-scenarios. ↑
7. O’Neill, B. C. et al. Proc. Natl Acad. Sci. USA 107, 17521–17526 (2010). ↑
Text © 2011 Nature Publishing Group, a division of Macmillan Publishers Limited. All Rights Reserved. Photo © ITV Global / The Kobal Collection.