I’ve written about bad analogies before. One doesn’t get to see one every day, so they are to be cherished. There’s a doozy in an article about effective communication on climate extremes. Like all the best bad analogies, it involves something everyday and relatable. In this case it’s basketball, shoes* and slam dunks. It was cooked up to explain why it’s confusing to talk about the effect of climate change on extremes – heatwaves, more specifically – in the “standard way”. The standard way being to say things like “climate change made the recent record heat 10 times more likely”.
In all fairness, the way that climate scientists have been taught to talk about extreme events can be confusing and is very easy to get wrong. Advice is conflicting and changes faster than some people can easily deal with. The advice in the article doesn’t really clear things up though and, in the service of clarity, it offers one of the worst analogies I’ve seen in well… days probably.
Go and read it.
The post is contrasting different ways of doing event attribution with the specific example of a heatwave in India and Pakistan. The author seems to prefer what is often referred to as the “storyline” approach in which one estimates how much warmer a heatwave was because of climate change (in the example, 1degC). This is contrasted with the PEA approach (PEA standing for Probabilistic Event Attribution) which tells us how much more likely a particular class of events has become due to climate change (in the example, 30 times more likely).
One objection he makes to the PEA approach is that in order to make it work, you need to define a class of events, based on the observed event. For example, where the maximum temperature equals or exceeds what was observed. You don’t have to do it that way, of course; you can choose other definitions, but the objection is that it splits the world into “event” and “non-event”. This, Patrick interprets as defining out of existence any heatwave that doesn’t meet the criteria. This is silly of course. The event being defined is not a generic definition of a “heatwave”, it’s a criteria specifically chosen for this particular event, which in turn was chosen because temperatures were exceptionally high even for a heatwave. This is woolly thinking at best, sleight of hand at worst.
A second objection is a misunderstanding made on behalf of people who are, presumably, believed by the author to be more easily confused than he is. These hypothetical readers would if told that “Because of climate change, the probability of an event such as that in 2022 has increased by a factor of about 30” understand that “the weather patterns that cause heatwaves are occurring 30 times more frequently”. It’s the same woolliness around whether we’re talking heatwaves in general or this specific heatwave. Once we sort that out, then actually that is, more or less, the correct interpretation. Without climate change, the range of weather patterns that would drive temperatures that high, would have been rather small. With climate change – that is, in the real world – the range of weather patterns reaching that temperature is much broader. In fact, you’re 30 times more likely to encounter one.
All of which leads to the analogy:
He asks us to consider a basketball player who can jump 28 inches one time in a hundred. The player buys new shoes which make every jump one inch higher, but they don’t jump any more often than before (a crucial, albeit bizarre detail). Now, there’s a one in ten chance they’ll jump 28 inches. Under the PEA formulation, we’d say that the chance of the player jumping 28 inches had increased by a factor of 10. Patrick takes exception to this, saying that it’s misleading because it makes it seem like the player jumps ten times more often and that the best way to quantify the impact is to say that each jump is an inch higher (the storyline approach). Furthermore, he argues, a jump is no longer a jump in the PEA formulation if it doesn’t attain 28 inches. (Patrick does allow that if 28 inches happened to be a “meaningful threshold” like the air needed to dunk the ball then it would be fine and relevant).
The analogy is strange, because I don’t think anyone would make the mistake that Patrick is supposing they would. If someone told me that the magic shoes they’d just given me would increase my chances of jumping a particular height by a factor of 10, my first assumption would not be that they would turn me into some sort of hyperactive kangaroo. The analogy is also sort of weird when it comes to matching up elements in the analogy with the thing it’s supposed to represent. If you straighten all that out then the analogy is not quite the slam dunk it’s supposed to be.
The aside about “meaningful” thresholds is interesting because current thinking favours choosing event-defining thresholds that are most closely connected to the impacts of the event. For example, if excess deaths were highest during the warmest three days of a longer heatwave, then the event might be defined in terms of the average maximum temperature for those three days. On the other hand, if deaths were associated with an extended period of high temperatures, which never got phenomenally high, then the average weekly temperature might be used to define the event.
And lets end with the concluding paragraph, which I give here in full:
“This [meaning the storyline approach] is the most effective way to communicate the science, and it should be favored over alternatives that do more to motivate political action than they do to convey our best understanding of the situation.“
Our “best understanding” of the situation is one that includes the widest range of analyses. That includes two salient pieces of information in the case of the south Asian heatwave. First, an event at least as bad in the specified way as the observed event was 30 times more likely because of climate change. Second, the particular event was 1 degree warmer because of climate change. It’s not EITHER/OR that’s most useful here, it’s AND, which is exactly what the world weather attribution team actually reported in this case.
Neither of these approaches comes close to describing the full set of things that could be said or to describing the situation in toto and its relationship to climate change. For a start, there are caveats about each method that are relevant to the discussion. To convey our best understanding of the situation, we need to convey exactly how good that understanding is as well as a lot of other things: what observations are available, what were the impacts, how do impacts vary with the meteorological parameters, how good are the models, are there other considerations, etc. etc.
-fin-
* I once bought a pair of shoes** that promised that I would walk 15% faster with them on. I don’t think this swayed my decision to purchase them, but it’s hard to tell. They did have leather soles though, which meant that on a surprisingly wide range of surfaces – anything polished, smooth or slightly damp – my feet would, briefly and without warning, be moving significantly faster than the rest of me usually in an unanticipated direction. None of this is particularly germane to the notional theme of my article, except the passing reference to shoes and the weird marketing hook.
** My history of shoe buying has not been a happy one. At the insistence of my wife, I once went into a shoe shop where the shoes were beguilingly displayed under brass fitted lamps on hardwood and velvet pedestals. There was a library-like hush within, through which an assistant insinuated himself. I was, I must admit, momentarily lost, wondering how you chose shoes when you cannot see the prices, or helpful signs saying “50% off selected brands”, and in my next moment of awareness I was sitting in a comfortable chair and contemplating my existing shoes. They had been placed on their own small stool, my feet for the moment still mostly inside them. The best word to describe my shoes was not, in fact, shoes. They were walking boots that I had used regularly for fifteen years. What was left of the sole had come away from the upper in places and the various prominences with which my feet are generously endowed had worn through the leather revealing that I was undeniably wearing bright pink socks. The padded ankle sleeve had split open and wispy clouds of synthetic lambswool spewed from the opening like something poisonous growing on damp wood.
The assistant contemplated my existing footwear. For a moment, I saw through his eyes and was transported to a place of horror and sorrow.
“Has sir ever had a good shoe?” he asked. The word good had a peculiarly suggestive emphasis that made me want to kick holes in the oak panelling. However, I am a coward, so I just said “No”.
I emerged from the shop several hours later, with my new shoes – my good shoes – in a bag. I had vague recollections of my feet being measured by callipers and other steel contraptions (does sir know his feet are rather different sizes?) and refusing to purchase a shoe tree, or a selection of soft brushes and polishes. However, the shock of paying was such that my mind shied away from the memories and they were thus confused, fugitive, untrustworthy things.
The reason for the purchase was that I was imminently to attend the final review meeting for a project and my go-15%-faster shoes had finally given up the ghost leaving me with the boots and a pair of trainers that had a lived-in air. I put the good shoes in my luggage and set off for the Netherlands.
The hotel was a twenty minute walk from the institute where the meeting was being held. This being the Netherlands, the road was flat and the pavements broad and it being a nice day, we decided to walk. My troubles started as I came down the stairs into the reception. As I went down each step, my foot lifted slightly in the shoe and the leather rubbed against my heel with a peculiar cold heat.
“It’s fine”, I thought, “New shoes. I’ll walk it off”.
I made it through the hotel car park without a problem and I was feeling reassured, but then a bit of rough ground between the car park and road rubbed not just my heels, but also the tops and sides of my feet. Something started gently to ooze.
When walking with ones colleagues it is considered polite to make conversation. I find this difficult at the best of times, but with both shoes now rubbing the whole circumferences of both my feet I was reduced to a pre-linguistic state, with the whole of my attention, and whatever processing power my brain had dedicated to computing exactly how much of each foot remained and how to keep the rest.
One advantage I found, of a good shoe was that the leather was relatively impermeable. I could feel things squelching inside, but externally, the beautifully burnished leather showed not a sign.
To minimise the pain, I tried varying my gait. For a short while, I found that a longer stride minimised the rubbing along the edges of the foot. However, it increased attrition at the heel so I alternated this with shorter shuffling foot steps which spared the heel at the expense of the sides. After a while, I discovered that I could take pressure off both the heel and sides temporarily, by sliding into each step and jamming my toes together in the front of the shoe. These considerations consumed me utterly.
After twenty minutes of this, I was starting to moan slightly with each footstep. My imagination painted my feet as bloody, septic stumps – something that would disgust a zombie. I could feel fluids oozing and squelching between my toes. The only relief I could get by now was walking flat footed, lifting each foot directly upwards and placing it down vertically too. Needless to say, this made forward progress difficult and was hard to do without attracting concerned attention particularly from someone you are nominally having a conversation with.
When we arrived at the leafy campus, all I could think about was tearing off the shoes and wondering how I could discreetly amputate my feet at the ankle. Consequently, I made quite the impression at security when I couldn’t reliably remember my own name, or which country I had come from. I was fantasising about putting my feet into an ice-cold vat of water or, possibly, liquid nitrogen. I asked for directions to the bathroom in a voice only two octaves higher than my usual talking voice and hobbled off with the kind of just-learning-to-walk locomotion that is cute in a 5-minute old calf but earns an adult human several square metres of personal space.
In the echoing coolness of the bathroom, I crashed into an empty stall and slammed the door shut. With the weight off my feet I had some respite. The leather, to its credit, remained unblemished. I tugged at the laces, opened them to their widest extent and slipped the shoe off.
I’m not sure what noise I made; I just know that it was loud. I was suddenly aware that all sound on the other side of the door had ceased and there was a kind of listening silence. I tried to remove the second shoe quietly, but stifling the sound somehow made it worse. With the shoes off, I thought my troubles were over and was determined to survey the extent of the damage.
So I peeled off the first sock.
If the first and second noises I had made were disquieting, they had, at least, a note of relief in them. Before the sock was half off, I realised that the material had fused together with my skin such that the removal of the former led directly to the removal of the latter. The involuntary noises that accompanied this minor act of flensing hadn’t the slightest note of relief in them.
I stopped. On the far side of the door I could hear the sound of someone washing their hands as quickly as possible, torn between the needs of hygiene and some atavistic necessity to flee from whatever horror was unfolding in the end cubicle.
Somehow, I got both socks off. In the cold light of the toilet cubicle, the damage seemed comically slight. Blisters had formed and then detached themselves from my feet. The revealed under-skin was bright red. Overall the damage was reassuringly minor. Able to think coherently for the first time since I had left the hotel, I realised I had to make it through the whole day and then walk back to the hotel in the evening.
First things first though, I had to get the shoes back on and re-join the everyday world. I swaddled my feet in a heavy wrapping of toilet paper. It was deliciously soft. I then stretched the socks over the top and pushed the whole lot into the shoes. I found that I could walk again, after a fashion. I carefully shuffled my way through the whole day, and spent as much of it sitting as possible.
For the walk home, I decided to go barefoot. This was fine for the first five minutes and I was saved from having to explain by the fact if you do something sufficiently eccentric, no one asks questions (leastwise not to your face). After five minutes, the bottoms of my feet started to burn. After ten, the urgent messages being sent by the nerve endings on the undersides of my feet were outshouting the signals from the damage already inflicted by the good shoes.
Walking on grass helped. The coolness was delicious. Unfortunately, the grass strips were discontinuous, separated by areas of gravel, cobblestones, asphalt and paving. Each of these hard surfaces had it’s own particular flavour of pain, with which I became well acquainted. Gravel could feel disarmingly soft until one hit a larger than average stone. Cobbles made my feet flex and stretch in strange ways. The asphalt’s uneven surface was rough and sharp and strewn with loose grit that rolled underfoot. Paving stones had no give in them at all. They just beat at the bottom of my feet mercilessly.
When we got back to the hotel, there were a few social niceties to dispense with – arranging dinner plans, and making small talk. I did this through gritted teeth. When I could finally get away, I left damp, slightly bloody footprints on the polished floor of the reception, staircase and corridor.
I spent much of the rest of the week wearing the old boots. Footwear has never felt so blessedly comfortable. I carried the good shoes around and, if called upon to look smart, I would slip them on. The shoes looked good, but in them I had the distracted air of a man who knows the mob has his wife. I don’t remember much else about the review meeting. I’m told it went well.
Pingback: A marine heatwave by any other name | Diagram Monkey