Rice and Mosher In Defence of Happy-Slappy Severe Weather Attribution at ATTP
I promised myself I wouldn’t do it, that it wasn’t worth the grief and the nasty taste left in the mouth afterwards, but I did it – I commented over at ATTP on the post on severe weather attribution. I regret it of course, but regret is a two way thing. It turned personal of course, so I’m not going to resist the odd personal attack here, which it’s not in my nature to indulge in, but if you’re going to fling shit, you’ve got to expect some to bounce back off the fan.
John’s also got a post on this subject and I would recommend any person interested in the technical aspects of the two statistical methods involved to read it and comment there. This post is more me just sounding off and trying to explain the ideas I was trying to get across at ATTP and inviting those people who preferred to insult me (even in my absence) rather than engage substantively with my arguments to come here and do the same, where my responses won’t be deleted. Hence I am not anticipating a great number of comments. I expected the usual insults about contrarians and even though I engaged with Steven Mosher politely, it came as no surprise when he resorted to spite in the absence of being able to engage with my points. I didn’t expect this from Tom Fuller though:
Jaime’s argument seems (from the outside) cynical. All of us might reflect on the factors that may have induced his cynicism. The evil that men do, and all that. The behavior of some in the climate community has been bad and that has tainted the views of people like Jamie towards the community entire.
I am not as cynical about motives as is Jaime. But I certainly understand why he might adopt such a stance.
IIRC, it seems that the climate community eventually decided to label the behavior of those few who produced cynicism in those such as Jamie as ‘sub-optimal.’ The blogosphere is replete with the detection of such ‘sub-optimal’ behavior. Perhaps an attribution study on hardened and negative attitudes resulting from such behavior is in order.
What was he thinking? He knows me from this blog; he contributes articles at Cliscep. Arrogantly and condescendingly dismissing my arguments on another blog using the third person (in the wrong gender!) as just ‘hardened cynicism’ born of the bad behaviour of ‘sub optimal’ climate communicators, then suggesting (even sarcastically) that an attribution study be done to determine the extent to which such ‘hardened and negative attitudes’ are due to the failings of such ‘sub optimal’ actors was bound to elicit a response from me. Tom’s not stupid. I don’t think he’s senile. I’ll admit to being cynical, but I won’t be bulldozed into a pigeon-hole using the wrong gender third person by someone who doesn’t even bother to address my argument and should know me rather better than that, as a fellow contributor here.
So that said, what was my argument? I’ll quote it again here:
“A complementary approach is to consider a storyline. For example, given that an event has occurred, how might climate change have influenced this event? If the air was warmer, then we may expect enhanced precipitation. If sea surface temperatures are high, then we may expect a tropical cyclone to be more intense. The focus here tends to be on the thermodynamics (i.e., the energy) and to take the dynamics as given (i.e., the event happened). The storyline approach, on the other hand, is more looking at how anthropogenically-driven climate change might have influenced an event that has actually occurred.”
I’m not sure it’s complementary. It’s also assessing, in a way, the probability of an event which has occurred, occurring, by an informal storyline process of looking only at how thermodynamical changes induced by warming might have made the event more likely and/or more extreme. By ignoring the dynamical circumstances which directly caused the event to occur, your storyline approach dismisses a vital component of attribution analysis, because dynamical influences leading to the occurrence of extreme weather events may be affected by natural and anthropogenic factors and it is only via a thorough examination of past weather events in relation to the specific event in question that can one can disentangle (or at least try to disentangle) these various factors.
“The problem, though, is that although the two approaches are complementary, they’re not actually quite addressing the same issue. The detection and attribution approach is essentially trying to determine how anthropogenic-driven climate change influences the probability of a specific class of event. The storyline approach, on the other hand, is more looking at how anthropogenically-driven climate change might have influenced an event that has actually occurred. There is no real reason why we should prefer one approach over the other; they can both play an important role in aiding our understanding of how anthropogenic influences impact extreme weather events.”
In actual fact, extreme weather attribution studies most often look at specific events which have occurred. That is what people are most interested in. That is the rationale behind most extreme weather attribution studies – to quantitatively determine what role (if any) anthropogenic climate change might have played in the extreme event which unfolded. By examining circulation patterns, past similar weather events and basic thermodynamics, an attribution study provides a tentative assessment of the fraction of attributable risk of that specific event happening in a world warmed by GHGs. It’s not perfect by any means, but it’s preferable to forming a truncated storyline based only upon a knowledge of basic thermodynamic physics combined with the simple observation that a specific event happened, with value added judgement thrown in. I think in such circumstances, one indeed should prefer one approach over the other.
Ken says:
“Jaime,
But noone is suggesting that it should be a substitute for a study that can consider atmospheric dynamics and thermodynamics. The suggestion is that a storyline approach can complement the more formal D&A approach, not that it replaces it.”
Mosh says:
“huh?
Sorry but you havent addressed the fundamental issue that everyone here who actually does these types of analyses understands.
ya need to do both approaches because they each have a short coming and you want your decision maker to be fully informed of everything you know and everything you dont know.”
As everyone here apparently understands these fundamental issues, please demonstrate that an analysis of extreme weather based upon thermodynamic considerations only, with a large chunk of value judgement thrown into the mix, is complementary [i.e. mutually supplying each other’s lack] to a more rigorous scientific analysis which incorporates thermodynamics, atmospheric dynamics and an examination of past weather events – albeit far from perfectly. What does formal attribution lack which the new, complementary ‘storyline’ attribution brings to the table? Huh? As far as I can see, yes, both methods do have their shortcomings, but the one method does not make up for the shortcomings of the other, and vice versa.
Ken, Izen, have you ever considered the fact that these extreme events must be viewed in the context of past events going back hundreds of years (if possible) and that the prior knowledge incorporated via the Bayesian approach (assumed to be complementary to the frequentist approach) itself relies upon a formal detection and attribution of global warming to anthropogenic factors, which only yields a confident attribution since 1950? Is it not faulty logic and somewhat circular reasoning to claim legitimacy and independence (or complementarity) for a Bayesian approach which inherently relies upon the (time limited) frequentist approach for its prior knowledge?
“Instead, the idea is: take the extreme event as a given constraint and ask if thermodynamic factors are involved in such a way as to worsen it.”
It’s a false dichotomy. You can’t conveniently separate out atmospheric dynamics and thermodynamics and analyse an extreme weather event with reference to just thermodynamics, ignoring any possible impact of the dynamics. You can’t do a magnitude attribution in isolation from dynamics, because the dynamics affects the geography and frequency of the event in question, which in turn can influence its magnitude/severity/impacts. If you don’t know the dynamics, then you don’t know, it’s as simple as that, and consequently you have to pronounce low confidence upon attribution, not dream up a storyline which ignores the uncertainty and focuses instead upon conditional knowledge.
OK Steven, a lot to unpick there, but it’s still not making a lot of sense to me – and the inconsistencies in the argument for the storyline approach are becoming increasingly apparent.
“Jaime you misunderstand the storyline approach. A storyline approach is like an accident investigation, like a forensic science, where the goal is understanding.”
That is a perfectly reasonable motivation. But I don’t think better understanding is the primary motivation. Attribution in the case where a more formal attribution cannot give high confidence appears to be the motive. That is political.
“If an extreme event was mainly caused by purely thermodynamic processes, then the risk-based analysis using a climate model is probably reliable and a strong attribution statement can be made.”
I can’t think of many instances where extreme weather is caused mainly by thermodynamic processes. Can you? Extreme cloud bursts, thunderstorms, intense rainfall over short periods (hours) spring to mind, but not much else. It’s really hard to separate dynamics from weather.
I totally get the false positives vs. the false negatives and I get the suggestion that, where formal attribution fails to come up with a ‘result’, then ‘absence of evidence is not evidence of absence’ and that it might simply be a false negative. But likewise, presence of evidence is not evidence of presence and false positives are a possibility. In my humble opinion, given the current lack of knowledge about the many different factors contributing to extreme weather, the probability of false positives using the unsophisticated and less robust storyline approach is higher than the probability of false negatives using formal attribution, and in any case, it is largely value judgement which determines the relative importance of either possibility. The authors seem to regard the possibility of false positives as a lesser evil than false negatives, that their value judgement is superior to the consensus attribution scientists’ value judgement for the purposes of “actionable information”. Again, political, not scientific.
“As climate change research becomes increasingly applied, the need for actionable information is growing rapidly. A key aspect of this requirement is the representation of uncertainties. The conventional approach to representing uncertainty in physical aspects of climate change is probabilistic, based on ensembles of climate model simulations. In the face of deep uncertainties, the known limitations of this approach are becoming increasingly apparent. An alternative is thus emerging which may be called a ‘storyline’ approach.”
Kind of gives the game away doesn’t it? It’s not complementary, it’s not a tool for additional understanding, it’s a tool to use as an “alternative” when formal attribution doesn’t give the ‘right’ result for the purposes of actionable information. Better err on the side of caution and pronounce that climate change did play a significant role in extreme weather events A and B, rather than make do with a “low confidence” assessment. Thanks, but no thanks. We should stick with the best that science can offer, even if it does risk missing an attribution here or there (and it’s not averse to returning false positives either). Storylines belong in story books – and there’s still the rather knotty issue that the first chapter was written by formal attribution scientists.
Ken, formal detection and attribution does use all of the available information. The storyline approach suggests focusing only upon some of the available information when other vital information (dynamics) is not available and you just can’t get a good signal on the old wireless attribution detector – which is not useful when you’re trying to form evidence-based policy and convince the more sceptical elements in society that bad weather is principally due to GHG emissions. That’s what I’m objecting to.
The bold gives a fair summary of what I was trying to get across. I thought my last comment to Mosher was polite and to the point, but he then saw fit to then launch a verbal assault upon me, some of which Willard deleted, so I’ve no idea what else he said. If Mosh wants to continue that attack, he’s welcome to do so here. But it does seem that I am instrumental, along with Judith Curry, in making Mosh a new convert to Happy Slappy Storyline Extreme Weather Attribution – I can live with that!
Oh I enjoy Jamie’s “contribution”. Frankly on forst reading Eric’s paper I wasnt a fan much of the storyline approach. Then I read Jamie’s nonsense and Judiths mischaracterization.
So that pushed me to got read the papers cited. Then it hit me.
Oh, I know this approach. Shit I even went and found the old AIAA paper where our team laid out a similar approach to understanding extreme events in war simulations.. basic forensics.
[Playing the ref, and no need to pile on. -W]
One last thing, Ken accused me basically of preferring the formal extreme weather attribution method because I’m not a fan of climate action! This is absurd. He’s spent enough time on this blog to know that I’m not a fan of formal extreme weather attribution either – in fact I would go so far as to say I think it’s largely pseudoscience! But, it’s a bit more robust and scientific than fiddling with storylines constructed via Bayesian priors, so naturally, being a fan of science in general and meteorology in particular, I’m going to defend one approach over the other.
Something to consider is that one of the points made in Winsberg et al. is that a preference for one method over the other involves some kind of value judgement, whether explicit, or not. It seems pretty clear (as evidence by the points you were making) that those who largely oppose climate action prefer the detection and attribution method. Those who are comfortable with climate action also seem comfortable with the storyline suggestion.
Climate Scepticism 
ATTP, as typical for extremists, is projecting. And his gratuitous mis-gendering, in light of his willingness to join in the hang attack on Dr. Crockford, reveals quite a bit of what is-and isn’t- in his character.
Thanks for putting on the hazmat suit and going there do others don’t.
LikeLike
What makes an extreme event extreme? Lots of casualties. The casualty trend is strongly negative. Given that future rise in co2 emissions is strongly linked to rapid economic growth of emerging economies, the casualties will drop even further this century, even if somebody makes up a storyline that tells a different story.
LikeLiked by 1 person
Hunter, thanks, but it wasn’t Ken who misgendered me. In fact he corrected the commenter who did.
LikeLike
You’d think so, Hans, but I’ve found that human casualties are only deserving of empathy when they further the alarmist narrative. When you raise the inconvenient data on economic prosperity and climate-related deaths, watch them make a beeline directly for the poor, unfortunate fowl of the air, fish of the sea, and beasts of the field.
And bees, of course.
LikeLiked by 1 person
Jaime,
I must admit that it’s a bit bizarre to have you complaining about a comment on my blog made by an author of this site. I also asked that you contact me privately if there was any specific comments that you found objectionable. I’m more than happy to correct any untoward personal insults, but I do need to know what they are. I did read the comments, and they all seemed quite benign. If your issue is that people didn’t agree with you, then that happens to be something that I allow on my blog ;-). (also, if you do object to personal attacks, why do you think it’s okay for you to engage in that here?)
What I actually said was It seems pretty clear (as evidence by the points you were making) that those who largely oppose climate action prefer the detection and attribution method. Those who are comfortable with climate action also seem comfortable with the storyline suggestion. I think this is a perfectly defensible comment. It does seem pretty clear that there is an association between those who’d typically be dismissive of the need for climate action and objecting to the storyline approach. Similarly, those who seem comfortable with climate action don’t seem to object to also considering the stroyline approach. Also, I don’t really see why this is such an objectionable thing to say. It seems pretty self-evident, even if it doesn’t apply in all cases.
LikeLike
OK, now you’re just annoying me Ken. When the first commenter in effect suggested that I was just a troll making a ‘nuisance’ of myself on your blog, I responded and Willard deleted my response, but not the original insult. I then decided that I had had enough of trying to comment at your blog under those circumstances and said so, which Willard also deleted. I left it there, only to check back and find that Mosh (predictably) had personally attacked me and Tom Fuller had left his absurd and insulting remarks. There would have been little point in me responding because my responses would probably have been deleted. So I didn’t, choosing instead to let off steam here. That’s my privilege. If you want to run your blog the way you do, allowing polite commenters to be harassed by other commenters and denying them the chance to respond, then that’s your choice, but don’t expect me not to remark upon it.
“If your issue is that people didn’t agree with you, then that happens to be something that I allow on my blog ;-). (also, if you do object to personal attacks, why do you think it’s okay for you to engage in that here?)”
I don’t as a matter of course attack people personally on this blog. That’s not something I do or condone. My issue was not that people disagreed with me, only that most failed to engage substantively with the points I raised, preferring instead to start sniping. Fine. If that’s the way people want to behave. It reflects more on their lack of an argument than on me. My objection was that my response to that sniping was deleted, so therefore I stopped commenting.
LikeLiked by 4 people
Ken,
“Similarly, those who seem comfortable with climate action don’t seem to object to also considering the stroyline approach.”
Well you need to tell that to the D&A community who, almost without exception, have considered the ‘stroyline [sic] approach’ and yet decided against it. That is the complaint being made by Winsberg et al.
What I find bizarre is the sight of a climate science sceptic having to point out what the climate science consensus is on this subject, and then defend it, only to receive so much opposition from a set of climate science consensus groupies who argue against it! What was it you keep telling us about denialism?
LikeLiked by 3 people
Jaime,
Okay, I’ve gone through the comments again and found one that might have been implying that you were a troll and have moderated that. Mosher’s comment was moderated last night. I’m not moderating Tom’s, partly because I can’t really see the problem and partly because I really can’t bring myself to moderate a comment on my blog by a cliscep author because of a complaint by a cliscep regular.
John,
Except, one of the reasons for preferring the D&A approach appears to be because it reduces the risk of reputational harm if someone were to claim a positive result that subsequentally turns out to be false. This is a perfectly valid concern and I can see why many scientists might have this preference. However, it doesn’t change that the formal D&A approach then runs the risk of suggesting false negatives. Hence – in my view – there is merit in considering an alternative approach that may fill in some of the holes. Of course, I’m not suggesting that the storyline approach should/would always suggest a link between climate change and extreme events. I do, however, think there may well be cases when there is a link, and that it’s possible to infer such a link based on our understanding of the underlying physics, and that may not be evident from a D&A approach. Hence, a storyline approach may provide more relevant information. Father’s day, family has arrived, so I’ve rushed and may not have explained that as well as I could have.
LikeLike
No worries Ken. I too will be busy tonight so I will not be rushing to respond. However, what I will do is replicate your comment at my post and respond to it there. This makes more sense to me because I want to respond within the context provided by my previous statements on this subject.
LikeLiked by 1 person
Story telling – precursor to science. ‘ … and so the Sky God sent down his fire to tell us of his anger.’
[Edited by RD to show gif]
LikeLiked by 3 people
¿Do weather models include CO2 concentration or green house gas forcing?
LikeLike
“The bold gives a fair summary of what I was trying to get across. I thought my last comment to Mosher was polite and to the point, but he then saw fit to then launch a verbal assault upon me, some of which Willard deleted, so I’ve no idea what else he said. ”
1. If you had no idea of what I said, why do you think it was a verbal assault?
2. Willard warned me about PLAYING THE REF and PILING ON by repeating a criticism
others had made.
3. there was no verbal assualt,
I didnt verbally assault you. none of my comments were edited or deleted Nice try though.
like you said
‘I gave up commenting here yesterday, because it turned personal and my responses were deleted (but not the insults) ”
So what it is .. on one hand you say the insults were there ( there are no insults)
On the other hand, you think willard deleted something ( he didnt) and assume that it was an assualt.
I played the ref and piled on, basically repeated a criticism made by others.
Now to what you say here :
“That is a perfectly reasonable motivation. But I don’t think better understanding is the primary motivation. Attribution in the case where a more formal attribution cannot give high confidence appears to be the motive. That is political.”
The simple fact is the storyline approach is not what you think it is. This comment above is typical of
your approach. You think you know the motivations. Lets put it this way. When you cant even characterize the approach properly, when you can’t even understand that the approaches are complementary, what makes you think you can divine motivations?. And furthermore, it doesnt matter
what the motivations are. A good decision maker can always disregard or weigh the evidence based on
their evaluation of the motivations. After all everyone who does D&A using the risk approach, knows that they risk false negatives. They are NOT free of motivations.
If you use a D&A approach you will have false negatives. guaranteed.
If you use a storyline approach, you will have false positives. guaranteed.
Just from my history doing decision support, my approach was to do both methods and clearly explain the pitfalls of both approaches. If decision makers asked for my opinion on which was correct, I informed them that I was not a decision maker.. above my paygrade.
LikeLike
Jaime,
Glad to hear that Ken didn’t also misgender you.
LikeLiked by 1 person
So Hans Erren asks a good question:
What makes an extreme event in the real world?
For the climate consensus, basically everything is extreme.
LikeLiked by 3 people
Steven, fair enough if you didn’t make any other comment. I assumed (wrongly) that Willard deleted something because ‘playing the ref’ is what he left in place in one of my comments that he deleted.
Still, saying,
“Oh I enjoy Jamie’s “contribution”. Frankly on forst reading Eric’s paper I wasnt a fan much of the storyline approach. Then I read Jamie’s nonsense and Judiths mischaracterization.”
in response to somebody else complaining that I was a nuisance troll is hardly positive engagement or a fair response to my last comment to you. So verbal assault is probably a bit strong; generally insulting and dismissive is more like it. But I wasn’t objecting to that. I’m no snowflake. You’ve said worse about me. It doesn’t bother me. What bothers me is Willard deleting my replies, so I exited the thread.
” . . . what makes you think you can divine motivations?” I didn’t divine them. I looked at the quote you supplied and I read them from that, in addition to using my own value judgement which I freely admitted.
It appears to me that there is a motivation for wanting to introduce a less than scientifically robust ‘storyline’ approach to attributing extreme weather; else, why do it? I have argued that the motivation is primarily political, policy orientated and you provided me the quote from the paper which strongly suggests this is indeed the case.
Now your argument is that you think I am unable to divine motivations, partly because I don’t even understand the method being advocated, but from what I said to you I understood enough apparently to concede your point that, in the case of formal attribution, ‘absence of evidence is not evidence of absence’. Correct. It’s not. That’s why attribution scientists don’t say: “climate change played no part in this event”, they say “we have low confidence that climate change played a part in this event”.
But it seems this isn’t good enough for your authors. They want something more. Hence they construct storylines where climate change might have played a part. but they’re as uncertain as the analyses carried out by scientists which are unable to demonstrate a robust attribution. Are they going to admit to this every time they use their method and say “it’s perfectly possible climate change played a significant role here, but we can’t say for sure, because we could be wrong, because we’re only assessing the thermodynamics in artificial isolation from the atmospheric dynamics, so it might be a false positive”?
No, they’re going to say something like “Although we can’t attribute this specific event to climate change, this is exactly what we would expect in a warming world”. Then they’re going to insist that this ‘expert’ assessment be included as part of the formal attribution process, along with the more formal attribution by scientists. That’s then part of the ‘actionable information’ which they apparently seek.
LikeLiked by 1 person
Beth, Wilson tells a great story. Maybe we should get Wilson to do all future extreme weather attributions! He knows a lot about storms at sea. I cried too when Wilson drifted away.
LikeLiked by 1 person