Given that the whole Royal Society Twitter saga relates partly to the possibility of an abrupt methane release, I thought I might post this video that I found on Climate Denial Crock of the Week (I hope Peter Sinclair doesn’t mind me using it). It’s Richard Alley explaining that our current understanding is that although there is a lot of methane on the sea floor, it is very unlikely to be released abruptly. It may well provide a feedback, but a methane “bomb” is unlikely.
This got me thinking a little about these low-probability, high-risk events. Of course, we should study and try to understand these processes, but I can see why they shouldn’t play a big role in the policy debate. We’re unlikely to drastically change our economy because of something very unlikely, even if it does present a high-risk (although how we as a society react to unlikely events does suggest that we don’t always behave in this way). We also have plenty of reason to act without focusing on such possibilities.
There is, however, a role that they can play and that some may not recognise. Richard Alley brings this up at the end of the video, which is why I thought I might mention it. There are various ways in which we can estimate how our climate will evolve under increasing anthropogenic forcings, and we can produce some kind of best estimate for the likely outcome. However, it’s not exact and we know that it could be “better” than this, and it could be “worse”. So, in some sense, there’s a symmetry – an approximately equal chance of these being “better” or “worse” than our best estimate (assuming it’s the median).
However, when it comes to the unexpected outcomes/events, this symmetry – I think – breaks. The chance that something unexpected could make things “better” is extremely unlikely. It would almost need to be fine tuned. It is much more likely that unexpected outcomes will makes things worse. So, although I can see why focusing on these unexpected events may not makes sense from a policy perspective, I do think that we can’t ignore that – if they do occur – they will likely make things worse. They also, probably, depend on the amount and rate of future warming, and so add an extra reason why we should really be seriously considering avoiding the higher emission pathways.
Anyway, that’s my view, for what it’s worth. Watch the video; Richard Alley is an extremely skilled science communicator.
...and Then There's Physics 
I thought I might add a comment about a recent Conversation article by Rich Pancost and Stephan Lewandowsky called Why climate uncertainty is no excuse for doing nothing. It makes the perfectly reasonable argument that being uncertain doesn’t mean that we should wait to be more certain. If anything, it means acting sooner rather than later.
The comments are already dominated by those with whom I am unable to have a sensible discussion (may be my own fault) and some of what is being said makes me think that some don’t understand how risk analysis works. Risk analysis doesn’t involve comparing the probability of something bad happening with the probability of it not happening; it involves comparing the probability of something bad happening with the cost (or risk) associated with reducing the chance of this bad outcome.
There have been some recent papers suggesting a reduced range and best estimate for climate sensitivity. Most of these have issues that suggest that they’re lower limits, rather than being better than other estimates, but that’s not really all that relevant. An increased possibility of a lower climate sensitivity is of course good, but doesn’t change that you still need to compare the probability of a high climate sensitivity with the risk (or cost) associated with reducing the chance of dangerous warming. Also, if the chance of climate sensitivity being high drops from a few percent to 1 percent, that’s not exactly comforting.
Additionally, the main impact of a reduced climate sensitivity is that it simply delays the warming, but only by decades not by centuries. Having more time is clearly good, but since we’ve done little so far, this would not appear to be a particularly good argument for slowling down. At best, it’s an opportunity to breathe a sigh of relief that we might not have left things too late.
I’ll also add that a recent David Roberts article suggests that the assumptions used in IAMs are so poor that we know virtually nothing about the economic impact of climate change or high much it will cost to mitigate, so any kind of risk analysis is extremely difficult.
Richard Alley is brilliant – a pleasure to listen to.
Redundant further evidence: The Two Mile Time Machine. Ideal stocking filler material 🙂
I’m sure we all know the story of the turkey foraging around in the farmyard, whose opinion that human beings are a kind and generous species is reinforced every day with the arrival of the farmer bringing an ever-increasing scoop of grain. Until that is, the day before Christmas Eve (or Thanksgiving, for our American cousins).
For our turkey—due to a lack of information—Christmas is a low probability, high risk event. Which is, unfortunately, similar to the position into which we are putting ourselves with regards to climate change—gambling with our future as we step into the unknown. And yet is it so unknown?
History shows us that life on earth has been moulded not by the periods of relative stability, but by rapid-change events of which the best known is probably that which wiped out the large dinosaurs. Now we’re possibly manufacturing our own event: ‘boldly going’ on a wing and a prayer, for no reason other than it’s too easy to continue with a short-term agenda rather than plan for the long haul.
As for the turkey, uncertainty is not our friend.
I think that it’s misleading to first discuss uncertainties and unlikely effects of climate change, because no choice is only about climate change.
The more correct framing (IMO) compares real alternative choices. We might, e.g., consider the potential differences between one realistic scenario of the BAU type with another realistic scenario, where a rising path of rather heavy carbon taxation is introduced (this path would probably include a reduction in other taxes).
The second comparison may turn out to be even more difficult to do, but it that’s the case then that’s direct evidence for not knowing, what’s wise.
Pekka,
I’m not sure I’m quite getting what you mean here. If one’s talking about climate change, then considering uncertainties seems obviously relevant. Of course, when considering what to do, one then has to consider the impact of any possible decision on other factors. That’s – as I understand it – is the whole risk analysis issue.
Uncertainties of estimates of the strength of climate change and its consequences are relevant, but it’s not possible to simply decide, what kind of climate we are going to have. What’s possible is to decide on policies or more directly on specific actions. Those decisions are going the have some influence on climate change and some influence on other things. The influence on climate change is uncertain also due to other reasons than the uncertainties in the climate science. The other effects of the decision are also uncertain.
What we must compare are the positive and negative influences of that particular decision (as compared to another possible decision). Even, if the climate change is judged a very serious threat, we must consider also, how much dent the decision can make in that threat.
There are good reasons to think that the worst credible consequences of climate change are more severe than the worst credible consequences of the policies proposed to mitigate climate change. Therefore risk aversion moves the balance of arguments towards stronger action at least for many of the proposed policies, but it’s quite likely that some of the proposals would have very little effect on climate, and therefore provide little benefits to justify the costs.
I have said it many times, and I say it again: Arguing that doing something or doing a lot is meaningless until it’s said, what that something or a lot refers to specifically. Only when the proposal is explicit, it’s possible to argue properly, whether the proposal has merit.
Indirect decisions like deciding on carbon tax are problematic as long as there’s little knowledge on, where that kind of incentive leads to. That makes deciding on the proper level of tax difficult. Even so policies based on gradually increasing carbon taxes may be the best alternative available.
Pekka,
I don’t think anything you’ve said is very different to what I’ve said. I agree that we have to consider a very complex set of issues. What are the implications of the various possible policy options and how will they influence future climate change (and other things)? There is indeed a huge amount of uncertainty, probably more so in terms of the implications of the various policy options than in climate change itself.
Forgot all about paleo, Anders, and in particular the fact that, even as we see unexpectedly rapid large-scale changes in both polar regions (WAIS passed its tipping point?… whatever), models remain unable to manage the transition from current to a Pliocene-like climate? But make that transition we will, even with just the level of CO2 we have. Sure, a crash emissions reduction program would probably put us on a different path, but I would suggest we’re unlikely to take that path so long as people like you keep saying “there is indeed a huge amount of uncertainty.” In the only important sense there isn’t.
Recall Ken Caldeira’s statement of the issue:
Pekka’s “gradually” amounts to an agreement to mug rather a lot of little old ladies.
Steve,
I thought you wouldn’t like this post 🙂
Did I actually say that? I didn’t think I did. I don’t think we can ignore that there is uncertainty, but – as Richard Alley says – the “uncertainty is mostly on the bad side”. My point was more that focusing on individual event that are low-probability, high-risk would seem counterproductive. Pointing out, however, that such events are possible and that they will almost certainly be detrimental might be a better way to approach this issue.
Roberts notes that the paper he discusses comes to much the same conclusion as Ken:
So not so complex after all.
Steve,
Yes, I tend to agree. If you are doing some kind of risk analysis in which you know there is a risk associated with one set of actions but have no real idea of the risks associated with doing something different (i.e., mitigating) then it would seem to make sense to prioritise minimising the known risks.
In a sense that is why I have no issue with promoting a proper risk analysis, because one would hope that any sensible analysis would conclude that following an emission pathway that has a non-negligible chance of resulting in an existential threat, is a path we should not be following. I would much rather see people arguing why such a risk is worth taking (I obviously don’t think it is) than pretending it doesn’t exist.
You said it in your immediate prior comment, Anders.
Re the ESS methane, I’m not sure I entirely buy Richard’s apparent conclusion that a major clathrate release is a low-probability event. The usual reason given is the lack of such a release during the Eemian, but a clathrate release would likely be triggered by warming currents rather than high-latitude insolation (with a corresponding relative lack of insolation in the tropics), so I take that explanation with a large grain of salt. That said, as I’ve mentioned before I’m more concerned with the prospect of a rapid loss via burning of the yedoma.
So, the tipping points we have include, in no particular order: The clathrates, the yedoma, Arctic sea ice loss, ice sheet loss, sea level rise, shifts in precipitation, increasing severe weather driven by Ferrel cell instability, ocean acidification, ocean anoxia, tropical rain forest conversion to savanna, boreal forest loss through burning and insect predation, assorted other atmospheric and oceanic circulation changes such as a major slowing of the AMOC, loss of habitability of much of the tropics, spread of tropical diseases into densely populated northers regions, and an abrupt warming pulse from removing aerosols. I’m sure I’m forgetting a few.
While some of these are dead certainties and others are not (yet), the obvious point is that a proper risk consideration (per the above, no analysis is possible, sorry) has to include all of them., otherwise we’re fooling ourselves.
Steve,
That was my point though. Focusing on one would seem counterproductive. Considering the possibility of such events should, however, be part of the risk analysis.
I say consideration because an analysis worthy of the term seems quite impossible. Can you point to one, or even an outline of how to go about it? Repeating Rosen and Guenther: “(H)umanity would be wise to mitigate climate change as quickly as possible without being constrained by existing economic systems and institutions, or risk making the world uninhabitable.” So we don’t need to analyze the risk in the sense you seem to mean, we just need to credibly point to it. I thinks that’s been done.
Relevant news in the Guardian yesterday: “Amazon deforestation picking up pace, satellite data reveals: Data indicates 190% rise in land clearance in August and September compared with same period last year“
Steve,
Sure, I broadly agree. I wasn’t really meaning it in the way you think I meant it. I was meaning comparing the risks associated with climate change and the risks associated with acting. Given that the latter are either small (according to the IPCC) or unknown (according to some recent papers discussed by David Roberts) it would seem obvious that we should act to minimise the risks associated with climate change. Of course, the problem is that even though it seems obvious to you and me, it isn’t to others and convincing them otherwise seems to be much harder than I would have expected.
That’s a mind-blowing non-sequitur. Why should unexpected events necessarily be “bad”? You have not justified this assertion.
I would argue that adaptation allows us to find what you call the median automatically. We adapt to the “bad” effects of global warming,while pocketing the “good” effects. This is precisely what humans are good at. There will be good effects and bad effects to temperature increases. Anyone arguing the opposite is implicitly stating that the climate of 1850 was optimal – perfect. Any deviation, at least in the warm direction, being undesirable. This is unlikely.
Of course there is some temperature increase above which it is all bad. Is 4C that limit? I don’t see why it should be. I think we can pocket the benefits of warming and adapt to the costs over the next 100yrs. The good news is, if we manage to maintain a healthy world market, we will do this automatically! This is why economic growth and keeping markets free and properly functioning is our most important challenge. It is also the only way we can truly contront the unknown challenges of the future. By being wealthy and knowledgeable enough. The future cannot be planned. We need wealth and knowledge.
You do realise that there are some things to which we cannot adapt. You’re simply arguing that we should happily continue modifying our climate into something that we – as a species – have never before experienced and that somehow being wealthy will save us. Personally, you views seem absurd. Others may disagree.
I’ll respond more to this
Well, partly simply because of the range of parameter space that’s available. There’s much more that would be bad than would be good. In more detail, what might we expect? A sudden methane release. That would almost certainly be bad. A much more rapid than expected reduction in ice. That too would increase the warming rate – bad. Are there any things that could be good? I guess suddenly getting an increase in the type of clouds that increases albedo and reduces our overall warming. That might be good, but is entirely inconsistent with paleoclimate. Of course, you should bear in mind that I mean bad for us (the human race) not the planet specifically. The planet’s fine.
And here’s another nasty surprise. I suppose the biological ones could be the worst.
Steve,
I’m obviously not using personal anecdote to refute proper research, but just to offer some reassurance on the amphibian population of the Picos de Europa: I had the priviledge of travelling there in May this year, and I’ve never seen so many newts in my entire life. They were bloody everywhere!
vtg, from what I can gather from the paper it hasn’t spread throughout the park yet. Overall it sounds like early days for this. But amphibian populations can crash quickly, as with the red-legged frog here in California.
Steve Bloom wrote:
What does that mean? One of the conclusions drawn is that of Pentti Linkola, perhaps the best known environmentalist in Finland. The Wikipedia article on him tells quotes his writings:
I would guess that few of us would be ready to go that far. We all put the limit at some point. We may do it fully intuitively, but that’s not satisfactory to me. I would insist that the decision must be based on some kind of comparison of risks and benefits of every significant decision even when fully quantitative cost-benefit analyses are not possible.
Thus every major decision related to mitigation must weigh the benefits of the reduction of warming that the particular decision is likely to produce to the costs of all kind that the decision has. Both the benefits and the costs must be considered taking into account all relevant risks and uncertainties. From the side of the benefits the reduction in the worst case risks may be the most essential factor or at least one of the important factors. From the uncertainties that weigh against the measure one of the most important is likely the effectiveness of the measure: Does it really lead to the originally estimated worldwide reduction in the CO2 emissions?
It should also be asked, whether some alternatives that may be much more cost-efficient get less funding, when an ineffective policy has been chosen only because it’s intuitively attractive. (I do think that some of the renewable energy schemes implemented in Europe suffer severely from that, when money and human resources are put in mass deployment of immature technologies rather than highly accelerated research of wider set of alternatives.)
Wow, Pekka, that’s quite an excursion, a bit too weighted in favor of the status quo IMO.
Sadly, conservation and efficiency don’t seem to be intuitively attractive. Waste has powerful advocates.
Also, Godwin’s Law.
Wow, Pentti Linkola’s views are remarkably extreme. Abhorrent.
Yeah, it was a big mistake for Rosen and Guenther to quote him. IIRC the Unabomber has similar views. Maybe Pekka can reference him next, for a little variety, or possibly put up a suitable billboard.
Steve,
I don’t give up insisting that each decision must be justified on its own merits. Generalities are not sufficient. Being sure that future is gloomy, if it cannot be changed, does not make such decisions right, that cannot be justified as helping in avoiding the gloomy future more than they cause damage underway.
Wishful thinking is only wishful thinking, whether it comes from those, who propose wait and see, or from those, who propose doing something drastic without good reasons to believe that that will help.
Well, drastic. I seem to recall that by no later than 20 years ago there was enough evidence to do the easy stuff. We didn’t do it and still haven’t done most of it. So your reaction seems a little bizarre to me.
Yes, we’ll have to hard stuff too, and the longer we wait the harder it will get.
But “propose doing something drastic without good reasons to believe that that will help”? Srsly? You don’t think a nearly 3 ppm increase in CO2 in the last year is drastic? There’s lots of drastic happening, and very little of it is mitigation.
Steve,
The point is valid also for minor decisions, but the extent of analysis required depends on the potential effects of the decisions.
I emphasized the extreme views of Linkola and drastic action only to make it clear to (almost) everyone that the proposal of doing everything possible (or as quickly as possible) cannot be taken literally, unless you agree with views as extreme as those of Linkola. The only alternative to that is to compare benefits to costs (by costs I mean all negative consequences including opportunity costs from stopping some alternatives).
Benefits can be compared to costs in many ways from semiquantitative to fully quantitative, purely qualitative arguments do not work. Risks and uncertainties must be taken into account in the comparison.
There are actions that have almost certainly a positive net effect, but I don’t believe that they bring us far.
vtg, by complete happenstance I just came across this article (not an academic paper, but by a trained observer) discussing a sharp drop in amphibian populations almost literally in my back yard. While I had never gone to the park in the spring to see the newt migration (similar to what you saw in Spain, I expect), I had assumed it was continuing as before. The tree frog decline is also a surprise. Bullfrog encroachment might explain things, but they’re not here (yet, anyway). Decades ago this area also had significant red-legged frog populations, but they’re long gone. So now I’m wondering how the local toads and salamanders are doing.
“Don’t it always seem to go
That you don’t know what you’ve got
’til it’s gone”