Thanks for the thoughtful comment! I’ve actually recently become worried about similar issues and am planning a post on the question whether climate risk is decreasing. There’s a lot in here so let me decompose.
Overall climate risk looks something like
(1) Emissions >
(2) Warming as a function of climate sensitivity and tipping points >
(3) Climate impacts at the level of warming >
(4) Vulnerability to direct climate impacts >
(5) Vulnerability to indirect climate impacts (e.g. dealing with climate-induced migration)
Going through the chain:
Expected warming has gone down a lot Expectations of (1) have strongly moved down.
Climate sensitivity has remained roughly stable, narrowing somewhat (for this and (1) see here), I think recent work on tipping points suggests that they are less severe than previously thought (see here, and discussion of this and a paper from last year here), overall this does not suggest a stronger climate response than previously assumed (2), if one had to make a directional update it would probably be a downward one.
So overall we should expect less warming and, crucially, we should expect a disproportionate probability decrease in tail warming scenarios (because emissions observed essentially rule out RCP 8.5).
Insofar as mechanisms 3-5 are independent of our knowledge of them, I think it is safe to say “it looks like climate risk has significantly decreased”. Of course, we could be badly surprised.
How expected climate risk might not have decreased What got me worried was Figure 4 from the SMP of the recent IPCC Synthesis Report suggesting higher risks at lower levels of warming (for all observed reasons for concern, the likely occurrence has moved downward between AR5 and A56 (2014 to 2022)):
I haven’t yet had time to do the math on this, but it could be that, as you suggest, expected climate risk is not decreasing (or even increasing) because, at the same time as we observe much lower expected warming we discover that impacts occur at lower temperatures.
Assessing this requires a lot more work, also because a lot of the knowledge here is very politicized, but it could be that I was wrong here and that risks have not decreased.
More correctly I should have said something like “It looks like climate risk has decreased a lot based on emissions and expected warming, but of course this holds impacts constant and this could be wrong.”
Verbalising the distinctions between 1 − 5 was something I was struggling with, so thanks for putting it so concisely and comprehensively. I agree with all the points you have made and the clarification at the end, which is what I was trying to say in a jumbled up way.
My impression on tipping point sensitivity was based on specific events happening significantly ahead of projections from modelling. I will have a read through the linked paper suggesting tipping points aren’t as bad as thought and comment on your linked post from March if necessary, but otherwise will update based on that.
I also agree that while the expected temperature trajectory is moderating (as well as the risk of higher trajectories), we may be underestimating the “political climate sensitivity” which is a function of the risks you provided in Figure 4 rather than warming, and which appear to be getting worse. I also don’t think great power conflict is significantly exacerbated by these indirect effects until much higher warming but Israel / Arab world and Pakistan / India are a couple of conflicts I think could be worsened and would still be of global concern, despite not being between Great Powers.
I agree with you that the biggest uncertainty right now is on “how does warming and its impact translate into societal consequences?” and I would be keen for anyone reducing the uncertainty there.
As I also discussed in the podcast, I think the biggest indirect longtermist risk from climate likely stems from a situation where non-great-power conflicts made more likely through climate become more catastrophic (e.g. through more distributed bio WMDs).
Thanks for the thoughtful comment! I’ve actually recently become worried about similar issues and am planning a post on the question whether climate risk is decreasing. There’s a lot in here so let me decompose.
Overall climate risk looks something like
(1) Emissions >
(2) Warming as a function of climate sensitivity and tipping points >
(3) Climate impacts at the level of warming >
(4) Vulnerability to direct climate impacts >
(5) Vulnerability to indirect climate impacts (e.g. dealing with climate-induced migration)
Going through the chain:
Expected warming has gone down a lot
Expectations of (1) have strongly moved down.
Climate sensitivity has remained roughly stable, narrowing somewhat (for this and (1) see here), I think recent work on tipping points suggests that they are less severe than previously thought (see here, and discussion of this and a paper from last year here), overall this does not suggest a stronger climate response than previously assumed (2), if one had to make a directional update it would probably be a downward one.
So overall we should expect less warming and, crucially, we should expect a disproportionate probability decrease in tail warming scenarios (because emissions observed essentially rule out RCP 8.5).
Insofar as mechanisms 3-5 are independent of our knowledge of them, I think it is safe to say “it looks like climate risk has significantly decreased”. Of course, we could be badly surprised.
How expected climate risk might not have decreased
What got me worried was Figure 4 from the SMP of the recent IPCC Synthesis Report suggesting higher risks at lower levels of warming (for all observed reasons for concern, the likely occurrence has moved downward between AR5 and A56 (2014 to 2022)):
I haven’t yet had time to do the math on this, but it could be that, as you suggest, expected climate risk is not decreasing (or even increasing) because, at the same time as we observe much lower expected warming we discover that impacts occur at lower temperatures.
Assessing this requires a lot more work, also because a lot of the knowledge here is very politicized, but it could be that I was wrong here and that risks have not decreased.
More correctly I should have said something like “It looks like climate risk has decreased a lot based on emissions and expected warming, but of course this holds impacts constant and this could be wrong.”
Verbalising the distinctions between 1 − 5 was something I was struggling with, so thanks for putting it so concisely and comprehensively. I agree with all the points you have made and the clarification at the end, which is what I was trying to say in a jumbled up way.
My impression on tipping point sensitivity was based on specific events happening significantly ahead of projections from modelling. I will have a read through the linked paper suggesting tipping points aren’t as bad as thought and comment on your linked post from March if necessary, but otherwise will update based on that.
I also agree that while the expected temperature trajectory is moderating (as well as the risk of higher trajectories), we may be underestimating the “political climate sensitivity” which is a function of the risks you provided in Figure 4 rather than warming, and which appear to be getting worse. I also don’t think great power conflict is significantly exacerbated by these indirect effects until much higher warming but Israel / Arab world and Pakistan / India are a couple of conflicts I think could be worsened and would still be of global concern, despite not being between Great Powers.
I haven’t done a deep dive on it but my reading has leant towards political instability being very sensitive to increases in risks, risks which are plausible at temperatures expected in the next 30 years. That being said, the frequent example of the Syria drought-conflict could be the wrong narrative in favour of unsustainable agricultural policies. So I think I need to investigate more.
Thanks for the discussion!
Thank you!
I agree with you that the biggest uncertainty right now is on “how does warming and its impact translate into societal consequences?” and I would be keen for anyone reducing the uncertainty there.
As I also discussed in the podcast, I think the biggest indirect longtermist risk from climate likely stems from a situation where non-great-power conflicts made more likely through climate become more catastrophic (e.g. through more distributed bio WMDs).