"It is the unhappy fate of the scientist today that he must play the role of Cassandra in the body politic, sending his fellow men to bed with nightmares in the hope to be heard in time."

- Arthur von Hippel, in "The Molecular Designing of Materials" (h/t @upbeatprof)

Friday, September 8, 2017

Neptune's Revenge

Graphic is excerpted from an excellent visualization at Axios.com, shows all Atlantic hurricane intensities from 2003-2015. Category five storms are highlighted and named. It illustrates the peculiar feast/famine pattern of hurricanes that makes statistical inference difficult.

After a worst case storm hits Houston another seems to be aiming for Miami. A worst case storm for Miami could be very bad indeed. I've seen a trillion dollars in possible damages bandied around - that would set the whole country back. (*) It almost seems as if a malign force is aiming hurricanes so as to do the most possible damage. It's weird and maybe karmic, as if Neptune were hurling his trident against his newly discovered enemies. It's the stuff of nightmares.

But is it really the "climate change wake-up call" that some people are claiming?

I think a lot of people have been far too quick and adamant about the climate change connection to Harvey. The main impact feature of Harvey was its slowness to advance, and its remaining parked near enough to shore to keep drawing on Gulf moisture. While impressive, Harvey was not a record-setting storm before it landed, and the factors that made it memorable don't seem all that directly related to climate.

Of course, Harvey is an instance of the changed climate; there's no disputing it was *affected* by the changed climate. Everything is, nice days as well as bad ones.

It's natural to ask whether a given extreme event was more representative of the changed climate than a comparable event would be in an unchanged climate. But that doesn't mean that science can provide a good answer.

Harvey, basically, was a Category 3 storm at landfall, Category 4 at peak, and it met an atmosphere where there were no strong steering winds just after it landed. I don't see any reason this is impossible under an unchanged climate. Is it more likely under the changed climate? Maybe. Perhaps it's slightly more likely than not.  But I don't think there is an especially strong case that it is definitely so.

Beyond that, I object to the formulation "Hurricane Harvey was not caused by climate change but was made worse by it". So what do some scientists mean when they say that so unequivocally?

The context is a rather uncertain expectation that tropical storms will remain about as common as in the undisturbed climate or perhaps slightly more rare, but that the more vigourous among them will become stronger than in the past. There is weak observational evidence for these claims, but the theory and modeling is clear about a raised "speed limit".

When the public asks about attribution of specific events, the old habit of the scientific community of saying "you cannot attribute individual events to a changed climate" has been replaced by a weird sort of fractional attribution. "Rainfall in strong events is expected to go up X per cent, and this is a strong event, so X per cent of the rainfall is attributable to climate change."

I don't think this makes any sense, and it leads people to conclude things like "To be clear, Hurricane Harvey would have formed over the western Gulf of Mexico and wreaked havoc on Texas, regardless of a warming climate" which doesn't actually make any sense. This was written by Paul Douglas, a meteorologist, and I'm afraid it indicates an alarming lack of understanding of fluid dynamics.

If the proverbial butterfly can change the weather, (small perturbations lead to vastly different weather states) a trillion tons of CO2 certainly will. You just can't compare scenarios like that.

No, causality is conceptually an all-or-nothing thing. We don't want to say "Harvey was caused 99% by nature and 1% by humans" or even worse "Harvey was caused 101% by nature and -1% by humans".  Nobody will understand this, and I think that is not because they don't understand the subtleties of scientific statistics, but because it makes no sense.

Does it really make more sense to say "Harvey was not caused by anything in particular but climate change made it rain 5% harder?" I am finding this just as hard to make sense of. Five per cent harder than what? Than that other Harvey that couldn't have formed in that other climate?

To be fair, it does make some sense to talk about storm surge riding on top of sea level rise. I think it's easy to see how a 12 foot surge on top of a foot of recent sea level rise is a lot like (though not exactly like) a 13 foot surge in a more stable climate. So sea level rise makes storm surge worse in a measurable way. I get that. It's worse than if sea level hadn't risen rapidly. And this matters to Rockport and Corpus Christi. But it doesn't matter to Houston, which was the main story.

The thermodynamic state of the atmosphere and ocean are all an intrinsic part of "this" storm. If there were a storm with 5% less rain, it would be a different storm in the first place. So I just don't get "made Harvey worse". If Harvey were less bad, it wouldn't be Harvey. These events are unique and distinctive. That's why we give them names.

It reminds me of the German aphorism my parents were fond of, "If my grandmother had wheels she would have been a bus."

"If Harvey weren't Harvey, it would be 5% weaker" just doesn't add up to a sensible claim to me.


What we really ought to be doing is seeing where the statistics of tropical storms are heading. If Harvey is part of a trend, to increasing storms, or increasing stalling, or whatever, we could say that. At least that's a well-formed claim.

But can we even say that?

The evidence is rather equivocal. NOAA says:
It is premature to conclude that human activities–and particularly greenhouse gas emissions that cause global warming–have already had a detectable impact on Atlantic hurricane or global tropical cyclone activity. That said, human activities may have already caused changes that are not yet detectable due to the small magnitude of the changes or observational limitations, or are not yet confidently modeled
Yes, they also say:
There are better than even odds that anthropogenic warming over the next century will lead to an increase in the occurrence of very intense tropical cyclone in some basins–an increase that would be substantially larger in percentage terms than the 2-11% increase in the average storm intensity. This increase in intense storm occurrence is projected despite a likely decrease (or little change) in the global numbers of all tropical cyclones.
 But that's a long way from "global warming definitely made Harvey worse".

See for yourself. There's an excellent chart that shows what the Atlantic has been doing hurricane-wise, here (excerpted and linked at the top of this article as well). Do you see an unambiguous trend? Or do you see a remarkably noisy and peculiar pattern about which it is hard to generalize?


So why all this talk of hurricanes? It's obvious. It's a news hook.

Climate change is very serious. Hurricanes are very serious. They're not unrelated, of course, but the relationship is complicated and as yet unclear. Claiming that "climate change made Harvey worse" is not a good look for climate science. It doesn't make semantic sense, and even if translated into terms that do, it's not a slam dunk that it's true.

Climate change is a very serious problem even if tropical storms go away altogether.

The linkage has already caused damage to the reputation of climate activism, as it was much touted in 2005, after which Atlantic hurricanes promptly subsided. People remember this. Going for a remake is not a good idea when the original movie wasn't any good.

I've been criticized for "not being a team player" on this matter. That's too bad. I really hate going against my friends and allies, and it has costs for me. But as far as I'm concerned, the way the Harvey and climate story has been told is not something I am buying into.

Our job as communicators of science is to tell people the scary truth, not just to scare people.


(*)  (Fortunately for my peace of mind though a bit unfortunately for my catchy opening paragraph, as I finish writing this Irma is trending a bit west, which would be quite a bit less disastrous overall, though that's little compensation for the Southwest coast of Florida and the Keys, or Cuba.)


UPDATE: Phil Plait does a good job taking up the contrary position. But again I don't think it applies to Harvey very well.


UPDATE 2: Scott Denning makes a point on Twitter that I had also made earlier this week: that the language "climate change causes weather event X" is very confusing to the public regarding the relationship between weather and climate. Climate doesn't cause weather - climate is made up of weather. This point is a bit pedantic, admittedly. There's a conceptual underpinning here that, true or not, seems to need a coherent phrasing. But the problem that this phrasing is much like saying "the base hit was caused by the batter's improved batting average". In baseball you say "he's hitting well these days" or something. Then you can talk about his improved statistics. But you don't say the statistics caused the hit. The trouble is, I am having trouble formulating a clean way to say what is trying to be said.

For instance, I'm fine with making a connection between anthropogenic climate disruption and the forest fires now raging in the northwest US and western Canada. But I'm struggling with how to make the case clearly without mangling the language.


BACKPEDAL: Really, I think the case for a climate disruption footprint on Irma is rather stronger than that for Harvey.

What we know.

Somebody on a Facebook thread said something silly about climate change. Somebody else said "I know a climate scientist, let's see what he says." That scientist was me.

I responded at perhaps greater length than the occasion deserved, so I have decided to preserve it here. Here's my first pass at "what we know".

Here's what I am absolutely (not 97% or 99%) sure of:

We understand how the atmosphere works pretty darn well. See how the hurricanes are going almost exactly where the models predicted? That's an amazing achievement, and there's a lot of science behind it.

To understand the weather at the surface of a solid planet, one of the keys is the composition of the atmosphere, particularly of gases and suspended particles that absorb radiation. That radiation can be either from the sun (mostly visible and ultraviolet) or thermal radiation from the surface (infrared). Over the long term, the energy coming in matches the energy going out. So the composition of the atmosphere sets the surface conditions. If it's harder for the infrared to get out, the planet is warmer than otherwise. This is not just about the Earth, it's universal.

We call gases that absorb infrared "greenhouse gases", and in understanding the complicated climate history of the Earth, carbon dioxide is one of the two most important of these, the other being water vapor.

Carbon flows in closed cycles between the soil, the plants, the atmosphere and the ocean. Natural changes in CO2 concentration, while very important, are relatively slow. Humans recently have been digging up carbon at a prodigious rate, and burning it to create CO2, altering the amount of carbon in the system. This has been accelerating with economic activity, and has recently become large enough that it matters to the climate. About half the extra CO2 stays in the atmosphere, warming and changing the climate at rates that are much faster than natural systems evolved for nor than human systems are prepared for. Despite increasingly loud and consistent (yes, consistent) warnings from scientists going back as far as the 1950s, the problem continues to accelerate.

Extra carbon in the system does not go away for a very long time (unless we undertake very expensive measures to slowly draw it down artificially). This means that with each passing day the burden we leave to our descendants gets worse.

The main sources of damage are expected to include ecological stress, sea level rise, and direct heat stress. Patterns of drought and flood are likely to change, and unexpected events will take specific places by surprise.

Also, the extra carbon alters ocean chemistry (ocean acidification) which directly impacts shellfish and plankton and upsets the ocean food chain. This is just beginning to be noticeable and is going to get much worse.

People say "97% of scientists" agree with this, but in my experience it is an underestimate. I don't personally know of any competent scientist who disagrees with the above. The "3%" seem to be outsiders who don't know their stuff.

Consider this editorial in the Houston Chronicle, which had a similar purpose to this posting. It asserts "to the best of our knowledge, there are no climate scientists in Texas who disagree with the mainstream view of climate science." As far as I know, nobody stepped up to contest that claim. Texas is a big state not known for reticence. The naysayer contingent in Texas is zero per cent.

There's a great deal more that we know, but that's basically the core of what we are sure of about climate change. There's a lot we suspect but aren't sure of, mostly about the consequences of all this and what to do about it. There's much debate about all that, but most people who think about it seriously are very concerned to say the least.

Sunday, September 3, 2017


It's apparent that it's no longer feasible for me to have an unmoderated blog.