Beware the Werewolves, Part Two

After a longer delay than expected, it's time to bring you the second part in my explanation of why you may need to reconsider your current views on the dangers posed by werewolves.

If you have no idea what I'm talking about and that sounds crazy, you should probably look at my earlier post on this topic. Or not. You could always just follow along and try to pick up on context clues to catch up if you'd prefer.

For a quick refresher, the last post I wrote on this topic mostly focused on how inappropriate an analogy used in a post by Victor Venema was. He had said:

Imagine you are driving on a curvy forest road and it gets more foggy. Do you slow down or do you keep your foot on the pedal? More fog means more uncertainty, means less predictability, means that you see the deer in your headlights later. Climate change mitigations sceptics like talking about uncertainty. They seem to see this as a reason to keep the foot op the pedal.

While this is madness, psychology suggests that this is an effective political strategy.

I won't rehash what's wrong with that analogy, but I'd like to reemphasize one last time just how disgustingly inappropriate I find it to be. Venema explains in his post "uncertainty can have multiple meanings" yet as I said before:

Venema knew there were many potential meanings for “uncertainty” and intentionally chose the one which allowed him to paint the people he is criticizing as insane when he could have chosen any number of other meanings, including ones which would have painted them as perfectly sane and rational.

But let's go beyond the analogy and examine Venema's actual arguments. In his analogy, the uncertainty introduced for the driver by the fog inherently increases risk. If you looked at the chances of the driver hitting a deer, adding fog to the mix could only make that more likely. He presents that as an, "Ah-hah!" moment, as a demonstration skeptics are irrational and that uncertainty means we should be more inclined to combat global warming.

Aside from that not being a remotely appropriate analogy for the global warming situation, there's a simple counterpoint Venema fails to even consider: What if the driver was already driving slower than necessary? People don't always drive the fastest speed they can safely drive. If fog changed the safe driving speed on a road from 55 to 45, that wouldn't mean a person who was already driving 45 would need to slow down. When you start thinking like that, you can find a ton of problems with Venema's core argument:

The damages of climate change rise with its magnitude (let's call this "temperature increase" for simplicity). I will argue in the next section that these damages rise faster than linear. If the relationship were linear, twice as much temperature increase would mean twice as much damages. Super-linear means that damages rise faster than that. Let us for this post assume that the damages are proportional to the square of the temperature increase. Any other super-linear relationship would show the same effect: that more uncertainty means higher risks.

In this case, if there were no uncertainty and the temperature in 2100 will increase by 4 degrees Celsius. For comparison, the temperature increase in 2100 is projected to be between 3 and 5.5°C for the worst scenario considered by the IPCC; RCP8.5. With 4 degrees warming the damages would be 16*D (4^2*D) dollar or 16*H human lives.

In the case with uncertainty, the temperature in 2100 would still increase by 4 degrees on average, but it could also be 3°C or 5°C. The damages for 4 degrees are still 16*D dollar. At 3 degrees the damage would be 9*D and at 5 degrees 25*D dollar, which is on average 17*D. The total damages will thus be higher than the 16*D dollar we had for the case without uncertainty.

I don't want to get bogged down with the details of how he has grossly over-simplified the global warming issue in his attempts to portray the people he disagrees with as irrational though. Well, actually I do, but I feel like it'd put people to sleep. So instead, let's look at how this argument compares to Venema's analogy.

For this argument, Venema assumes increased uncertainty about global warming must inherently be evenly distributed on both sides of some central estimate. That is, uncertainty both increases and decreases the potential dangers. In his analogy, however, uncertainty can only increase potential dangers. Venema doesn't explain this discrepancy. Let's consider it ourselves.

For his baseline examination, Venema used the value of 16D (gotten by taking 4 degrees squared). When he added uncertainty, he got these potential values:

9, 16, 25

Corresponding to the supposed damage caused by 3, 4 and 5 degrees of warming. The average of this is 17.3, greater than the 16 gotten without uncertainty (I'm going to stop including the variable as its superfluous). Now I want to stress there are a ton of things I take issue with for this relationship, but for the sake of this discussion, let's ignore that and assume the exponential relationship is fine. What would happen if we increased uncertainty further? Following in Venema's footsteps, we'd get:

4, 9, 16, 25, 36

Which gives an average of 18, even larger than before. But wait, the fog only added uncertainty in the direction that made things worse. Couldn't that mean we should really use:

9, 16, 25, 36

Which gives an average of 21.5, meaning things are even worse? That would be in line with Venema's analogy. That doesn't actually justify it, but then, Venema doesn't justify any of his choices, so... maybe? Or maybe we should have only added uncertainty that would have favored the "skeptic" position?

4, 9, 16, 25

Hey look at that, now the average is down to 13.5, below any of the other values we've seen so far. Arbitrarily choosing how we want to apply uncertainty to our calculations really pays off. We can make our data say whatever we want!

And no, it's not just the analogy I'm using to argue this problem exists in Venema's post. He uses this very same sort of approach himself:

"Judith Curry thinks that we should take even more uncertainty into account: 'I think we can bound [equilibrium climate sensitivity] between 1 and 6°C at a likely level, I don’t think we can justify narrowing this further. ... [T]here is a 33% probability that that actual [climate] sensitivity could be higher or lower than my bounds. To bound at a 90% level, I would say the bounds need to be 0-10°C.' "

If the climate sensitivity were zero, the damages in 2100 would be zero. Estimating the temperature increase for a climate sensitivity of 10°C is more challenging. If we would still follow the carbon-is-life scenario mitigation skeptics prefer (RCP8.5), we would get a temperature increase of around 13°C in 2100**. It seems more likely that civilization will collapse before, but 13°C would give climate change damages of 132*D, which equals 169*D. The average damages for Curry's limiting case are thus 85*D, a lot more than the 16*D for the case were we are certain. If the uncertainty monster were this big, that would make the risk of climate change a lot higher.

Now, I'm sure Venema would say he was just using the values Judith Curry listed so he wasn't actually arbitrarily choosing how to adjust the uncertainty distribution. The kernel of truth in that claim only serves to highlight the more serious problem with Venema's entire approach. While Curry did in fact give the range of 0-10, she never suggested all the values in-between are equally likely.

Think about it. Venema is simply averaging the supposed damage for each temperature value together. That would only make sense if each temperature value was equally likely. If some values are actually more likely than others, his approach will give completely bogus results. While his approach considers each value equally likely, effectively drawing a flat line straight across for them, Curry's remark which he quotes stemmed in part from the common belief at the time the likelihood of the values might be something more like:

slide11

Which is a figure Curry had posted on her site a number of times as it is from the IPCC AR4, and it reflected a common view of what climate sensitivities might be. As it shows, a person saying the range of potential climate sensitivities is 0-10 in no way suggests they think each value is equally likely.

This goes back to the werewolf issue. Philosophically speaking, we cannot know with absolute certainty there are no werewolves. As such, I would estimate the number of werewolves as being somewhere between 0 and a non-zero number. I would put an extremely low probability on their being any werewolves, but it would be positive.

With Victor Venema's sort of butchering, he'd likely portray that as me saying it is just as likely there are werewolves as there not being any. Or maybe he'd say because I believe the range is 0 to seven plus billion, we should average all numbers between 0 and seven billion to come up with the number of werewolves I believe exist. Or maybe he'd come up with some other ridiculous mischaracterization of my innocuous remarks to paint me as some sort of delusional maniac who needs to be committed. Who knows? Any of that would make as much sense as what he did here.

And there's still more. The figure I show above displays what are known as probability density functions (PDFs). That's basically the name for showing the various possibilities with their corresponding probabilities. Judith Curry has said this about them:

In my opinion, we have a strong profession obligation NOT to simply the uncertainty by portraying it as a pdf, when the situation is characterized by substantial uncertainty that is not statistical in nature.

There's a bit of poor editing there, and I haven't included any of the details or explanations she gave (you can go to the post to see them), but the point is clear enough. Curry doesn't think graphs like the one I displayed can fully characterize the uncertainty in the results.

So first, Venema would have been unable to use his approach if he had looked at the range Curry gave in anything resembling a sensible manner rather than simply assuming she felt all values for climate sensitivity from 0 to 10 are equally plausible. Then, if he had looked at what Curry had actually said about her beliefs, he'd have found there is actually no easy or direct way to put them into numbers like he wanted. But instead, he just... came up with a ridiculous misrepresentation of what Curry believes.

And it gets worse. While Venema quotes Curry as saying she believes the plausible range of climate sensitivity values is 0-10, the reality is she has openly discussed how that view has changed over time. Here is a post she wrote about a year ago with quite a bit of discussion of climate sensitivity. It says:

On a 2010 thread (can’t find it now), I stated that I thought the ‘very likely’ range for equilibrium climate sensitivity was something like 0.5-10C (I can’t exactly recall the lower bound). My rationale for this was the AR4 ECS figure (shown above). I had no reason at that time to ‘reject’ any of the values shown in that figure.
...
Where does this leave us for now, in terms of bounding ECS?

lower bound: I would use Nic Lewis’ values for the lower bound: 1.05 (for the 5-95% range) and 1.2 (for the 17-83% range)
upper bound: we need to look at climate models, for which there is no meaningful pdf, we can only look at discrete simulations. The highest model simulations seem to be slightly less than 5C.

It was trivially easy for me to find this post. I just typed "Judith Curry climate sensitivity" into Google and found it on the first page of my results. Even if it didn't show up quickly for other people, there is still no excuse for Venema relying on a comment Curry made five years ago as a demonstration of what her views are today. She has made no secret of her views on issues like this. She regularly talks about them on her blog. Choosing to just assume she holds the same beliefs now as she did half a decade ago is...

And yet, it still gets worse. Victor Venema wrote his post to tell skeptics uncertainty makes the case for action against global warming stronger, specifically referring to Judith Curry and her “fans” in the title of his post:

Fans of Judith Curry: the uncertainty monster is not your friend

While quoting her and discussing her views in that post. A person reading all that would think this is some new idea to Judith Curry and her “fans,” or at least one they’ve ignored. Nobody would ever guess Curry has written about this argument multiple times over years, even so far back as five years ago.

Venema put the time and effort into writing a blog post to tell these people all about how they didn’t know what they were talking about, completely failing to note they had been talking about the very argument he was making for half a decade. And naturally, because of this, he fails to address any of the responses that have been given to his argument over the years.

I don't really have a wrap-up today. All the things I'd be inclined to say are too harsh or just too snarky to merit the space. I honestly don't know how people could do a worse job advocating for a cause than people like Victor Venema do. I am only half kidding when I say I think I could probably do a better job convincing people to take action to combat the potential rise of lycanthropes than global warming advocates are doing to convince people to combat rising temperatures.

4 comments

  1. We probably have to agree to disagree on whether the analogy fits. I think it does, it provides an everyday situation where clearly uncertainty increases the risk, which is what happens in the cases more mathematically discussed below it. But it is normally easy to disagree with an analogy because it is never the real thing, it is an analogy.

    My post was not about Pascal' wager, catastrophes or thick tailed distributions. Just about the influence of uncertainty, even in a very moderate case.

    The problem of comparing the base case: "4, 9, 16, 25, 36" with: "9, 16, 25, 36" and "4, 9, 16, 25" is that the latter two cases also have a different mean. Feel free to argue that my analogy also does that, that was not intended and my actual argument was about cases where the mean stays the same only the uncertainty changes.

    You may be interested in: “Macroeconomic policy and the optimal destruction of vampires” (1982) by Dennis Snower
    Although human beings have endured the recurring ravages of vampires for centuries, scarcely any attempts have been made to analyze the macroeconomic implications of this problem and to devise socially optimal policy responses. Over the past few centuries, a number of prominent investigators… have suggested that all vampires should be destroyed… [We show that] such a policy would not be socially optimal. via The Stand-Up Economist.

  2. Victor Venema, while you may continue to feel your analogy was appropriate, I'm not sure what point commenting here just to say so serves. This post does discuss and criticize your analogy, but it also discusses far more than your analogy. It shows the argument you used is only true if a number of unstated assumptions are made, a point you have helped confirm here by explaining your argument relies on an assumption you never stated in your post or showed to be true for global warming. In fact, this post goes so far as to show many climate scientists believe the unstated assumption you relied upon and now express here not to be true.

    Given your comment here implicitly acknowledges the central problem with your argument highlighted in this post, I just don't get what you think commenting on your analogy might accomplish. Thanks for the tip about the paper though. It looks interesting.

  3. And there is even more ......

    "Imagine you are driving on a curvy forest road and it gets more foggy. Do you slow down or do you keep your foot on the pedal? More fog means more uncertainty, means less predictability, means that you see the deer in your headlights later. Climate change mitigations sceptics like talking about uncertainty. They seem to see this as a reason to keep the foot op the pedal."

    Indeed, climate change mitigations sceptics won't mitigate that's to say they won't do anything to let the fog disappear or to prevent the fog. They adapt that's to say, when the fog is growing thicker, while talking about all kinds of uncertainty, they, if necessary, slow down. Victor and friends are busy building mills blowing the fog away or try to invent methods of preventing fog

Leave a Reply

Your email address will not be published. Required fields are marked *