4

From a 2016 The Atlantic article, Human Extinction Isn't That Unlikely:

UN-approved climate models estimate that the risk of six to ten degrees Celsius of warming ... “On a more plausible emissions scenario ... a 10-percent risk,”

The quote is from

Sebastian Farquhar, the director of the Global Priorities Project

Is this a consensus estimate?

Martin Schröder
  • 548
  • 1
  • 6
  • 16
FrozenKiwi
  • 167
  • 5
  • 1
    So yes, I guess I'm just looking for supporting evidence and would accept either that GPP is reputable, or confirmation from an alternative source. – FrozenKiwi Sep 10 '20 at 12:00
  • 3
    The Global Priorities Project is related to the Centre for Effective Altruism, both looking at what actions (collective or individual) may produce the greatest long term benefits. Established by academic philosophers at the University of Oxford, it places substantial value on the long-term survival of humanity, and so has a particular concern about catastrophe risks, more so than other similar ideas such as the [Copenhagen Consensus](https://en.wikipedia.org/wiki/Copenhagen_Consensus) – Henry Sep 10 '20 at 12:18
  • I have removed the troublesome part, which biased the call for evidence in only one direction. As Skeptics, we should be open to the evidence, no matter which way it points. – Oddthinking Sep 10 '20 at 18:28
  • I note that the claim is NOT "there is a 10% chance". The claim is "if we don't tamp down our emissions, and instead follow a more plausible path, there is a 10% chance". – Oddthinking Sep 10 '20 at 18:35
  • How the concept of probability/chance is calculated or what does it mean, here and in similar discussions? I never understood sentences like there is 1 over 10000 chance that the rocket launch goes wrong, and stuff like that. It seems quite a pictorial use of concepts like probability. I know it is not related to this question, but can someone point me to the right direction? I am not a climate change skeptics, this is a general recurrent doubt of mine. – Alchimista Sep 11 '20 at 17:22
  • @Oddthinking thanks for the edit, and the follow-on comment. I hadn't picked up on that qualification, which is a pretty big point and because it changes the odds from 10% to 10% * the odds that we continue emitting (which I think is pretty low, in spite of current evidence) – FrozenKiwi Sep 12 '20 at 11:29
  • @Alchimista I think people use phrases like that because it tends to be more comprehensible than a 0.01% chance. Same thing, but people aren't very good at visualizing low-probability events. It doesn't mean that if there have been 99999 successful launches that you should cancel your ticket on the next one though :-) – FrozenKiwi Sep 12 '20 at 11:34
  • @Alchimista here is a good article on the problems with communicating probability and uncertainty. key line: "We tend to think that “1 in 5” is more likely to happen than “20%.” https://hbr.org/2016/11/why-its-so-hard-for-us-to-visualize-uncertainty – FrozenKiwi Sep 12 '20 at 11:39
  • @FrozenKiwi thank you. But it is not really my point. My doubt is really on the sense however wrote down. It seems to me a stretch of the concept of probability. This is also on mechanical failure etc. Can we really state a probability that something goes wrong while boarding a plane? Even out of statistics our fly has nothing to do with all the numbers collected. I think I should pose a Q and express myself in a more precise way. – Alchimista Sep 13 '20 at 13:56

1 Answers1

11

A 10% chance of 6-10°C by the end of the century is not in the findings of the IPCC AR5 (2013) report. This seems relevant given the reference to "UN approved climate models" in the claim.

The summary in the IPCC AR5 report is (pg. 1031):

Global mean temperatures will continue to rise over the 21st century if greenhouse gas (GHG) emissions continue unabated. Under the assumptions of the concentration-driven RCPs, global mean surface temperatures for 2081–2100, relative to 1986–2005 will likely be in the 5 to 95% range of the CMIP5 models; 0.3°C to 1.7°C (RCP2.6), 1.1°C to 2.6°C (RCP4.5), 1.4°C to 3.1°C (RCP6.0), 2.6°C to 4.8°C (RCP8.5). Global temperatures averaged over the period 2081–2100 are projected to likely exceed 1.5°C above 1850-1900 for RCP4.5, RCP6.0 and RCP8.5 (high confidence), are likely to exceed 2°C above 1850-1900 for RCP6.0 and RCP8.5 (high confidence) and are more likely than not to exceed 2°C for RCP4.5 (medium confidence). Temperature change above 2°C under RCP2.6 is unlikely (medium confidence). Warming above 4°C by 2081–2100 is unlikely in all RCPs (high confidence) except for RCP8.5, where it is about as likely as not (medium confidence). {12.4.1, Tables 12.2, 12.3, Figures 12.5, 12.8

Note that increases in the 6-10°C range only seem at all plausible under "RCP8.5" which is Representative Concentration Pathway 8.5. The RCPs correspond to different assumptions about how much carbon people will emit -- in RCP8.5 carbon emissions continue to increase over the 21st century. This is the worst case considered by the IPCC.

Note that the article in the OP is dated April 29, 2016, well before there were indications of higher sensitivity reported in 2020 using the models being developed to support AR6 -- so the claim by Mr. Farquhar couldn't have been based on these more recent, and still tentative, results.

(Aside, note that Mr. Farquhar may assume that emissions will be even higher than what the IPCC considers the worst case, which could explain the discrepancy).

Martin Schröder
  • 548
  • 1
  • 6
  • 16
Dave
  • 2,433
  • 19
  • 23
  • Good answer. One factor that could explain part of the discrepancy could be differing baselines. By 1986–2005 (the reference period in your quote), temperatures had already risen by ~0.5 degrees relative to the (notoriously vague) 'pre-industrial levels' referred to eg in the Paris agreement – waltzfordebs Sep 10 '20 at 18:55
  • Good answer, and probably the best we can do with this dataset, given there doesn't really seem to be any way to assign a likelyhood to each given pathway. Thanks! – FrozenKiwi Sep 12 '20 at 11:42