The Two Degree Temperature Target is Arbitrary and Untethered
A close look at the origins of the 2C target of the Paris Climate Agreement leads to a surprising conclusion
Pretty much everyone has heard of the 2 degree Celsius temperature target that sits at the center of international climate policy and shapes everything from climate science and assessment to policies and regulations. But very few know where it came from. Even fewer know that the nice round number – TWO – is completely arbitrary, and has become detached from its original scientific justification as scientific understandings have advanced.
If the 2C degree policy target (and its sibling, the 1.5C degree target) are breached in coming years and decades, as many expect, then the aims of international climate policy will have to revisited. This post initiates an on-going discussion here at The Honest Broker of how those policy targets might evolve by documenting the origins of the 2C temperature target and how it is today untethered from its original scientific basis. In this context, reconsidering targets for climate policy is not necessarily a bad thing. In fact, it may help to better focus climate policy on targets and timetables more appropriate for effective action on climate change.
The proximate origin of the 2C temperature target is the 2015 Paris Climate Agreement under the UN Framework Convention on Climate Change (FCCC), which has the goal of
Holding increase in the global average temperature to well below 2°C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5°C above pre-industrial levels, recognizing that this would significantly reduce the risks and impacts of climate change
But why 2 degrees? Why not 3C? Or 2.5C? Or 1.0C? Or any other number?
The reason that the Paris Agreement focuses on 2C can be traced directly to a statement published 20 years earlier in 1995 by the German Advisory Council on Global Change (WBGU) — an expert advisory body to the German government — as a contribution to the first conference of parties to the UNFCCC. The WGBU recognized that addressing climate change meant that “anthropogenic emissions must be reduced to almost zero over the very long term,” foreshadowing today’s commitments to “net-zero” emissions. But how fast would emissions reductions have to occur?
The UNFCCC, which took effect in 1994, articulates a goal to stabilize greenhouse gas concentrations in the atmosphere,
"at a level that would prevent dangerous anthropogenic (human induced) interference with the climate system… [and] such a level should be achieved within a time-frame sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened, and to enable economic development to proceed in a sustainable manner."
The unquantified goal of the UNFCCC prompted a flood of studies and debates over what “dangerous anthropogenic interference” actually meant and how the concept would be converted into targets and timetables that could be used to guide real-world policy implementation. The 1995 WGBU statement provided an answer that was ultimately taken up in international negotiations.
The WGBU focused on quantifying a “tolerable temperature window” for the evolution of future global temperatures that met two goals: the “preservation of creation” and “prevention of excessive costs.” The second goal indicates that the Iron Law of Climate Policy was present at the start. But how to define quantitatively the “tolerable temperature window”?
The WGBU used understandings of global temperature fluctuations (what the IPCC calls “variability”) to establish the bounds of its proposed window — ranging from the low global temperatures of the last ice age to the high temperatures of the last interglacial period.
You can see the description of the “tolerable temperature window” of the WGBU in the image below.
At the time, in 1995, the world had already warmed an estimated ~0.7C from preindustrial levels. So 0.7C plus the additional 1.3C estimated by the WGBU that would exceed the “tolerable temperature window.”
So that is the methodological origin of the 2C increase above pre-industrial levels codified in the Paris Agreement. But it is not the end of the story.
A next task in tracing the origins of the 2C target is to identify where the actual temperature boundaries of “tolerable temperature window” come from – the 10.4C to 16.1C expressed by the WGBU (to which they subtracted/added a buffer of 0.5C, as described in the image above). After all, if the upper end temperature of the last interglacial period was, say, actually lower at 15.5C or higher at 16.6C then those seemingly small differences in temperature would have profound implications for the size of the “tolerable temperature window” and thus also for climate policy.
The temperature ranges for the last ice age and last interglacial periods used by the WGBU come from an obscure 1987 analysis of variations in historical global temperature. The creators of the “tolerable temperature window” explained their methods in more detail in a 1999 paper in Climatic Change. The authors noted that their approach was stylized and “all numerical results should be seen as provisional and that normative settings are not proposed to be accepted now and forever world-wide.” The 2C temperature target was never intended to be viewed as binding on policy, as a scientific threshold or a tipping point. It was always a stylized estimate.
Consider what would happen if instead of relying on an obscure 1987 paper to establish the bounds of the “tolerable temperature window” we instead relied on a somewhat more authoritative source, the 2021 report of the IPCC Working Group 1. The recent IPCC report concluded (with medium confidence) that the last interglacial period “is estimated to have reached 0.5°C–1.5°C higher values” than the preindustrial reference period. This estimate is considerably lower than the temperature of the last interglacial periods estimated in the obscure 1987 chapter used in 1995 by WGBU as the basis for establishing the “tolerable temperature window.”
If we instead plug into the “tolerable temperature window” methodology the values reported by the IPCC AR6 for temperatures of the last interglacial period, then the upper threshold for the “tolerable temperature window” is just 15.6C, calculated as:
14.6C preindustrial, plus
0.5C, lower bound of IPCC AR6 temperature range for last interglacial, plus
0.5C, buffer applied in the methodology of the “tolerable temperature window”
Equals 15.6C
This is one full degree lower than the results of the 1995 WGBU statement to the UN FCCC, which ultimately became codified in the 2015 Paris Agreement as the 2C degree temperature target.
The 2C temperature target should actually be a 1C temperature target, under the methodology originally used to create the temperature target adopted by the UNFCCC.
That means that the world is already well outside the “tolerable temperature window,” as global temperatures have already increased by more than 1C since the preindustrial period.
The 2C temperature target serves as an “anchoring device” – in the language of a classic 1998 paper by van der Sluijs and colleagues. They use the concept of climate sensitivity to illustrate how supposedly scientific concepts in climate science become transformed into social and political concepts which require stabilization over time in order to be turned into policy instruments.
Consequently, an anchoring device sometimes requires detachment from its underlying scientific basis in order to serve a political function:
“If the apparent scientific rationale for policy were to be too closely tied to new scientific findings, the basis for policy actions could be undermined, especially within the context of a highly politicized and polarized societal debate on the issue at hand”
If we actually recognized that the 2C target is arbitrary, then it would upset not just an entire epistemic and political community that has been built around it, but it would also open up climate policy to fundamental reconsideration – which may not be a bad thing.
As van der Sluijs and colleagues explain, opening up and questioning an anchor device in climate policy might lead to asking:
“whether the cohering climate concept of greenhouse gas emissions is really the most important, effective and acceptable way of defining international priorities with regard to global environmental change and its close associates – human poverty, inequity, and global consumption and land-use patterns. . . [such a question] would upset the existing commitment by most (though by no means all) policy actors and institutions to the idea of smooth and manageable forms of anthropogenic climate change, corresponding with the idea of its intellectual amenability to the modern epistemic culture of prediction and control”
Temperature targets in climate policy are untethered from their original underlying scientific basis, but nonetheless considerable scientific work is performed to reinforce the significance of 2C. Claims of scientific significance are useful politically because the 2C and 1.5C temperature targets have become reified in policies and regulations, and scientific claims can help to deflect criticism or opposition.
However, we may soon learn that achieving such targets is unrealistic or unattainable. Then the entire architecture of climate policy will need to be revisited. Arguably we are already at that point.
Climate policy needs to be opened up, and new types of targets and timetables are needed to replace the out-of-date and untethered temperature targets. My preferred candidate is to focus policy on achieving zero-fossil fuel energy consumption. But that is a discussion for another time.
Paying subscribers to The Honest Broker receive subscriber-only posts, regular pointers to recommended readings, occasional direct emails with PDFs of my books and paywalled writings and the opportunity to participate in conversations on the site. I am also looking for additional ways to add value to those who see fit to support my work.
There are three subscription models:
1. The annual subscription: $80 annually
2. The standard monthly subscription: $8 monthly - which gives you a bit more flexibility.
3. Founders club: $500 annually, or another amount at your discretion - for those who have the ability or interest to support my work at a higher level.
Thank You. This is very informative. Using computer generated global mean temperature changes as meaningful drivers of climate policy is bad enough but tethering it to some specious determination of a "tolerable temperature window" is just way over the top.
Sanity needs to prevail.
Very helpful commentary on how the official goals are determined.
One concern for me is that they are world-wide averages, making a complex, large set of local and regional climates into a single one.
Another is that the average temperature change does not distinguish between changing daily low temperatures and changing daily high temperatures.
Here in Kansas City there have been 69 record highs from 2000 to 2022. 34 of them have been in the winter months (December to March) and only 11 in the summer months (June to September).
And there have been 70 record lows for KC from 2000 to 2022. Only 12 of these have been in the winter months and 32 have been set in the summer months.
This indicates that changes in the average temperature for this geographically representative area are complex. If the average temperature has gone up by say 0.3 C in this period that could be an entirely beneficial or an entirely dangerous temperature change or somewhere in between.
Likewise, an increase of 2 C over the period 2000 to 2100 could be beneficial or dangerous or in between. The average numbers do not tell us what is correct.
[data at: https://www.weather.gov/eax/kcrecnorm]