Pielke contra mundum

75349072_1304046896458294_3397302797375373312_n Oh noes, not again. Still it's an entertaining distraction from the faithful war reenactors. My headline is from Brideshead of course (I was of that vintage) but apparently it has history. How appropriate.

This is all about RP Jr's No, Hurricanes Are Not Bigger, Stronger and More Dangerous. A statement with which - to revel5 my prejudices - I have a great deal of sympathy6; perhaps also Climate emergency? applies. That's in well-known academic journal Forbes1, and is a response to Normalized US hurricane damage estimates using area of total destruction, 1900−2018 by Aslak Grinsted, Peter Ditlevsen, and Jens Hesselbjerg Christensen in PNAS November 11, 20193. The bastard paper (which I'll call G19 following RP) is paywalled alas, though happily that spares me the trouble of reading it, but I can show you the abstract:
Hurricanes are the most destructive natural disasters in the United States. The record of economic damage from hurricanes shows a steep positive trend dominated by increases in wealth. It is necessary to account for temporal changes in exposed wealth, in a process called normalization, before we can compare the destructiveness of recorded damaging storms from different areas and at different times. Atmospheric models predict major hurricanes to get more intense as Earth warms, and we expect this trend to eventually emerge above the natural variability in the record of normalized damage. However, the evidence for an increasing trend in normalized damage since 1900 has been controversial. In this study, we develop a record of normalized damage since 1900 based on an equivalent area of total destruction. Here, we show that this record has an improved signal-to-noise ratio over earlier normalization schemes based on calculations of present-day economic damage. Our data reveal an emergent positive trend in damage, which we attribute to a detectable change in extreme storms due to global warming. Moreover, we show that this increasing trend in damage can also be exposed in existing normalized damage records by looking at the frequency of the largest damage events. Our record of normalized damage, framed in terms of an equivalent area of total destruction, is a more reliable measure for climate-related changes in extreme weather, and can be used for better risk assessments on hurricane disasters.
So, the problem here is... actually, let's go back a step. The problem here is that people are desperate for something sexy to be caused by CO2 increasing. The obvious one - that it's getting warmer, duh - just isn't sexy enough because it's too slow. But it's quite well observed and statistically tractable: you can clearly see the signal from the noise. It would be so much better if something deeply erotic like hurricanes could clearly be getting worse under GW. Unfortunately, the hurricane observation record isn't brilliant, and the record of economic damage is dominated by enormous increase in "exposed wealth" as more and more is built on shorelines; and as money shifts to Florida and so on. G19 acknowledge this problem and attempt to dance around it.

Notice that the abstract talks of the record of economic damage from hurricanes, but the sexed-up "significance" header says The frequency of the very most damaging hurricanes has increased at a rate of 330% per century. This of course what made its way into every newspaper's headlines; and it is the first point to draw RP's ire: it purports to say something about climatological trends in hurricanes, but it uses no actual climate data on hurricanes. That’s right, it instead uses data on economic losses from hurricanes to arrive at conclusions about climate trends. Since RP has done a lot of analysis on economic damage - indeed, G19 are using his dataset, or some version thereof - this sorta sounds like a quibble. But he continues From 1900 to 1958, the first half of the period under study, NOAA reports that there were 117 total hurricanes that struck the mainland U.S.. But in contrast, G19 has only 92. They are missing 25 hurricanes. In the second half of the dataset, from 1959 to 2017, NOAA has 91 hurricanes that struck the U.S., and G19 has 155, that is 64 extra hurricanes4. And that sounds like it might matter. So I think G19 owe him a reasoned reply on that.

His second and larger point is that G19's dataset isn't homogeneous: The dataset on losses from hurricanes used by G19... was initially created about a decade ago by a former student and collaborator of mine, Joel Gratz, based entirely on our 2008 hurricane loss dataset (which I’ll call P08)... created a new hybrid dataset, from 1900 to 1980 the ICAT dataset is based on P08 and for 1980 to 2018 it is based on NCEI. This is hugely problematic for G19... the result is a data incontinuity that introduces spurious trends to the dataset. This also sounds like a problem that needs to be addressed.

AG responds, but alas he does so on Twatter which is a totally shit venue for writing coherent thoughts in. He does have his own website but it's badly out of date; hopefully he'll write something clearer and at length soon.

RP asserts that the IPCC backs up his position, which from memory I think is correct. Browsing, the IPCC AR5 SPM says nowt about hurricanes; in particular the Detection and Attribution of Climate Change section very sensibly concentrates on the unsexy but observable stuff. Chapter 2 says that AR4 concluded that it was likely that an increasing trend had occurred in intense tropical cyclone activity since 1970 in some regions but that there was no clear trend in the annual numbers of tropical cyclones. Subsequent assessments, including SREX and more recent literature indicate that it is difficult to draw firm conclusions with respect to the confidence levels associated with observed trends prior to the satellite era and in ocean basins outside of the North Atlantic... In summary, this assessment does not revise the SREX conclusion of low confidence that any reported long-term (centennial) increases in tropical cyclone activity are robust, after accounting for past changes in observing capabilities. More recent assessments indicate that it is unlikely that annual numbers of tropical storms, hurricanes and major hurricanes counts have increased over the past 100 years in the North Atlantic basin. Evidence, however, is for a virtually certain increase in the frequency and intensity of the strongest tropical cyclones since the 1970s in that region. And so on. I got bored. If you can find the IPCC saying something dead exciting on the subject do let me know.

I have a feeling that some of the sound and fury of the argument comes from not clearly distinguishing between all hurricanes and the most powerful ones. Implicitly, G19 only finds trends in the most destructive? That sounds like a familiar theory. But of course, they are rarer, and so your stats become even noisier.


1. I snark by sheer reflex; of course, a rapid-response is entirely reasonable; this is this week's news; asking for RP to wait until next week, let alone for the sluggish and uncertain journal response times2, would be unreasonable.

2. There's a fun anecdote in the bio of Schroedinger I'm currently reading: the Prussian academy was noted for the speed of publication of its Sitzungberichte: two days after a manuscript was received, the proofs were brought to the author by a messenger boy who would wait while the author corrected them.

3. Notice how carefully I link this all up. I do it at work too. So many people don't, and you'd be amazed at how hard it is to get from a to b if you haven't left breadcrumbs.

4. Why the miscounts? According to RP Part of this difference can be explained by the fact that G19 focus on economic damage, not hurricanes. If a hurricane from early in the 20th century resulted in no reported damage, then according to G19 it did not exist. That’s one reason why we don’t use economic data to make conclusions about climate. A second reason for the mismatched counts is that G19 counts many non-hurricanes as hurricanes, and disproportionately so in the second half of the dataset.

5. I meant reveal of course, but now I notice the typo it's also rather appropriate, I do revel in this stuff and my prejudices in particular.

6. Pointing out that everything is so statistically woolly that what RP really means is not that we have strong evidence for no trend, but that we have no strong evidence for a trend, would be tedious.


Here's Pielke Jr in 2014 feeling good about ICAT data - AG still dumping stuff on Twatter instead of writing a coherent response.
* Gavin mentions the war, at RC; and at Gizmodo


...and Then There's Physics said...

Your footnote 6 seems a little generous. As far as I can tell when Roger says there's no trend, he really means "there's no trend". As for as Tropical Cyclones (TCs) specifically, it's my understanding that there are indeed trends associated with increased sea surface temperatures (Elsner et al. 2008). This Realclimate post also seems to suggest that there has indeed been an increase in the frequency of the strongest TCs.

Tom said...

Waitaminnit, dammit. Hurricanes are not " the most destructive natural disasters in the United States." Floods are.

William M. Connolley said...

> there are indeed trends

That isn't obviously consistent with the IPCC I quoted and looked at; possibly the intense stuff?

> Floods are

Wiki disagrees with you (https://en.wikipedia.org/wiki/List_of_disasters_by_cost) in terms of individual events; I don't know about in sum.

...and Then There's Physics said...

That isn't obviously consistent with the IPCC I quoted and looked at; possibly the intense stuff?

Possibly, but I'm not sure it has to be. Also, what I was highlighting was that there are indications of a trend associated with increasing sea surface temperatures. IIRC, what Elsner et al. (2008) showed was that the maximum wind speed of the strongest TCs increased with increasing sea surface temperature. AFAIA, this is what would be expected from our physical understanding of TCs. Given that AGW has almost certainly increased SSTs, this would seem to indicate that AGW may well already have influenced the strongest TCs. This is probably somewhat Bayesian thinking, though, and a full frequentist detection and attribution analysis may still not be feasible.

As far as I can tell, it's also possible that the above could be true and it could also be true that there is no detectable trend in TC frequency. Given that strong TCs are rare, it seems quite possible that the strongest TCs could have got slightly stronger without this necessary producing a detectable trend in the frequency of strong TCs. However, the RC post seems to suggest that there has indeed been an increase in the frequency of the strongest TCs, but I haven't managed to find a paper with the same kind of results in it.

William M. Connolley said...

Hmm. I guess what I'm trying to say, rather badly, is that even if your "optimistic" assumptions are correct then the effect, if there, is sufficiently weak to stand out rather poorly from the noise (and unlike the wealth-at-risk signal, which is unmissable). Unlike increases in temperature. So it seems a bit of a shame to be too hung up about it.

Tom said...

Hurricanes, tornadoes, and earthquakes might seem like the most dangerous natural hazards you could ever face, but floods and droughts actually kill more Americans over time.

Better predictions for hurricanes and other tropical cyclones, as well as tornadoes, have reduced the death tolls from such events in recent decades. But flooding deaths are on the rise.

On average, U.S. flooding kills more than 100 people a year — more than any other single weather hazard, including tornadoes and hurricanes, according to the University Corporation for Atmospheric Research (UCAR). Most flood deaths are from flash floods, however, and about half of those are because people try to cross swollen streams or flooded roads. Victims often underestimate the power of water when driving into flooded areas, UCAR scientists note, adding that it takes only 18 inches of water to float a typical vehicle.


...and Then There's Physics said...

What I'm trying to suggest is that there are analysis that do suggest trends in some circumstances (for example, trends with respect to SST for the strongest TCs). So, the suggestion that there is no evidence for any kind of change seems wrong. However, I agree that trying to detect some kind of trend with respect to the frequency of strong TCs seems very challenging, given the noise. It doesn't surprise me that we haven't yet seen something unambiguous. I guess what I'm hung up about (if I am) is with suggestion that there are no indications of any kind of trend, when that seems to not be true (i.e., there are indications, even if not everyone yet agrees).

William M. Connolley said...

> U.S. flooding kills more than 100 people a year

Oh, but that's trivial. No-one would bother do anything about that. https://www.cdc.gov/nchs/fastats/deaths.htm says:

Number of deaths for leading causes of death:
Heart disease: 647,457
Cancer: 599,108
Accidents (unintentional injuries): 169,936
Chronic lower respiratory diseases: 160,201
Stroke (cerebrovascular diseases): 146,383
Alzheimer’s disease: 121,404
Diabetes: 83,564
Influenza and Pneumonia: 55,672
Nephritis, nephrotic syndrome and nephrosis: 50,633
Intentional self-harm (suicide): 47,173

Even if there were 10 or even 100 times as many deaths from flooding it would still be unimportant.

> suggestion that there are no indications of any kind of trend

Yes, I think that's wrong.

Tom said...

Yes, WMC, in the U.S. climate isn't really a major cause of death. I am pretty sure that flooding is more dangerous worldwide--I think Indur Goklany published on that at some point, but I'm too lazy to look for it.

Anonymous said...

Tom opines that "in the U.S. climate isn't really a major cause of death..."--I think Indur Goklany published on that at some point, but I'm too lazy to look for it."

The right word for steering clear of Indur's Heartland deliverables is "asute" , not lazy:


Thanks to WMC for fixing this link if it balks

Tom said...

Mr./Ms. Anonymous, I am most probably to the left of you and am pretty much rabid on the subject of Republican destruction of American domestic politics. I despise Donald Trump and the phenomenon of Trumpism and yearn for its passing.

Do you or the author of vvattsupwiththat take any issue with the research Indur Goklany has published? If so, you would be the first that I have seen.

Russell Seitz said...

Your politics wont save you if you can't tell hack from a handsaw.

Indur is a coal-fired electrical engineer who works for Heartland and writes for the GWPF


Tom said...

Is his analysis of mortality due to natural disasters and cllmate/weather related deaths accurate?

William M. Connolley said...

> but I'm too lazy to look for it... accurate?

Aren't you obliged to at least say what analysis you mean before asking insistently if it's accurate?

Tom said...

I think the 'insistency' you cite is inferred and not intended.

His claims about the decline of mortality due to extreme weather can be found here:


I personally hate it when Trump does something right, like criminal justice reform. Thankfully it's quite rare. But hey, you know? Sometimes the skeptic brigade is right.

Russell said...

The Bot is strong in this one.


The wet elephant in the room is the disparity between flood death trends arising from climate change, and the jumbo actuariall anomaly arising from the 2004 Indonesian tsunami.

The wave killed more people in one day than ordinary floods have in the last two decades.

Andy Mitchell said...

Didn't GWPF tell us that global warming stopped in 1998?

Tom said...

I really don't get it, Russell. Is Goklany accurate in his claims or not? His claims are not new, as shown by the dates on his charts. And yet I haven't seen any counter claims or refutations. Noticing that does not make me a 'bot.'

William M. Connolley said...

If IG only publishes at GWPF, it's not stange that people ignore him, since their stuff is generally junk. That deaths-from-disasters have declined is not controversial I think, e.g. https://ourworldindata.org/natural-disasters. Why would you use an unreliable source like GWPF rather than a reliable one?

Tom said...

"Indur M. Goklany is a science and technology policy analyst for the United States Department of the Interior, where he holds the position of Assistant Director of Programs, Science and Technology Policy." (Wikipedia)

Goklany, Indur M. (2006). The Improving State of the World: Why We're Living Longer, Healthier, More Comfortable Lives on a Cleaner Planet. The Cato Institute. ISBN 978-1-930865-98-3.
Goklany, Indur M. (November 2002). The Precautionary Principle: A Critical Appraisal of Environmental Risk. The Cato Institute. ISBN 978-1-930865-16-7.
Goklany, Indur M. (November 26, 1999). Clearing the Air: The Real Story of the War on Air Pollution. The Cato Institute. ISBN 978-1-882577-83-5.
Goklany, Indur M. (September 2015). "Carbon Dioxide The Good News" (PDF). Global Warming Policy Foundation Report. 18: 1–47. Retrieved October 19, 2015.
Goklany, Indur M. (September 19, 2009). "Climate change is not the biggest global health threat". The Lancet. 374 (9694): 973–974. doi:10.1016/S0140-6736(09)61655-X. PMID 19766873.
Goklany, Indur M. (July 29, 2009). "Is Climate Change the "Defining Challenge of Our Age"?". Energy & Environment. 20 (3): 279–302. CiteSeerX doi:10.1260/095830509788066439.
Goklany, Indur M. (February 5, 2008). "What to Do about Climate Change" (PDF). The Cato Institute. p. 1. Retrieved May 23, 2010.
Goklany, Indur M. (November 2002). "From precautionary principle to risk–risk analysis". Nature. 20 (1075): 1075. doi:10.1038/nbt1102-1075. PMID 12410243.
Goklany, Indur M. (August 22, 2002). "The Globalization of Human Well-Being". The Cato Institute. p. 1. Retrieved May 23, 2010.

He seems to be a conservative analyst who finds it easier to get published in friendly environments, although Lancet has some cachet.

As for your question, WMC, the only other reports I see are the ridiculous (and abandoned) ones like 300,000 deaths a year due to climate change.

I am a progressive Democrat. I despise Trump. But is Goklany correct? If not, why haven't others published a refutation? The sources he cites are legitimate--I did a little publishing on this when I was doing State of the Climate summaries, so I know that much.

Sometimes people who advocate positions I don't like are correct on specific things.

William M. Connolley said...

Sorry, I meant has he only published *that stuff* at GWPF.

> But is Goklany correct?

Err, didn't you read what I wrote? As far as I can tell he is trivially and uninterestingly correct, depending on what you think he's saying, of course.

Marco said...

Tom, the Lancet publishes lots of "correspondences" - these are in general not peer reviewed. They're a bit like comments on papers.

Regarding refutations, Jeff Masters had a not-too-old piece on the problems with the stats:

And then there's this:

Tom said...

I believe Goklany's major point is that weather and climate related deaths have dropped by two orders of magnitude over the past century. He repeatedly makes the point that improvements in observation and communications technology are the primary causes of this really good news. It's not like he's saying the weather/climate got better.

As for cold deaths/heat deaths, there are numerous papers on both sides of that issue. I have a personal interest in it as an elderly friend died of a cold winter in Hove because she didn't have enough coins to feed her coin operated meter. But I understand there are arguments in favor of each position.


Has Tom forgotten Goklany's political history ?

He quit the Interior department to become a Cato Institute writer over a decade ago- here's the WashPost's 2018 take on his brilliant career:

Climate and Environment
How a climate skeptic marginalized for years at the Interior Dept. rose to prominence under Trump

"10 days after President Trump took office last year, an Interior Department official suggested a swift change to its website in preparation for Trump’s choice to lead the department, Ryan Zinke.
While Zinke wouldn’t be sworn in for weeks, Office of Policy Analysis senior adviser Indur Goklany emailed Doug Domenech — a Trump appointee who would become the Interior Department’s assistant secretary of insular areas — telling him they would “be doing the new Secretary a favor if … the current ‘Our Priorities’ page visible on the DOI home page were removed before he is confirmed.”

, Goklany wasn’t a new arrival. He had been working at the Interior Department since Ronald Reagan was president and had spent years questioning whether climate change would damage the planet as much as some of his colleagues predicted.

Goklany, who often goes by the nickname “Goks,” has written papers for several conservative think tanks as well as participated in their events and films while working at the Office of Policy Analysis for more than 30 years. Weeks after Trump’s inauguration, he found himself within the inner circle of the Interior Department’s leadership. He weighed in on climate change discussions, attended meetings with some of Zinke’s senior aides and began working in the office of the deputy secretary.

Tom said...

No, TCW, I haven't forgotten Goklany's history, as noted above. He's conservative. I am not. He's a skeptic. I am not.

I don't really care. If he's right, he's right. If he's wrong, show it.

William M. Connolley said...

> major point is that weather and climate related deaths have dropped by two orders of magnitude over the past century.

Then I think that, as I said before, this is dull; and there's no reason to go to such a dodgy source for such info.

> He repeatedly makes the point that improvements in observation and communications technology are the primary causes

That however seems dubious. A quick scan shows him quoting himself for those points, and although I can't be bothered to check I'm fairly sure the circular references will continue if pursued.

Tom said...

WMC, who else is publishing on the subject?

I google declining mortality due to weather and the first result is an article in Reason. But it references... Goklany. https://reason.org/policy-study/decline-deaths-extreme-weather/

Well, there is this: https://www.nber.org/papers/w18692. But it's US only. Still, "First, we find that the mortality effect of an extremely hot day declined by about 80% between 1900-1959 and 1960-2004. As a consequence, days with temperatures exceeding 90°F were responsible for about 600 premature fatalities annually in the 1960-2004 period, compared to the approximately 3,600 premature fatalities that would have occurred if the temperature-mortality relationship from before 1960 still prevailed. Second, the adoption of residential air conditioning (AC) explains essentially the entire decline in the temperature-mortality relationship."

Unknown said...

Storm-surge damages might be an easier category to attribute, partially, to climate-change. Ironically, the areas with the smallest storm-surge damage might be easiest to attribute because absent sea level rise, they wouldn't have that type of damage at all.

I'm thinking of this from a legal perspective, FWIW.


"I don't really care. If he's right, he's right. If he's wrong, show it."

Tom, he's a NIPCC apparatchk. Read his handiwork and weep tears of laughter.

Tom said...

TCW, I've read his work. I did not weep tears of laughter. In fact it seemed quite reasonable to me.

Brian, I think storm surge might be difficult for a variety of reasons, including unmeasured subsidence. But perhaps we're getting better at that.

Andy Mitchell said...

"...In fact it seemed quite reasonable to me."

As you have already revealed you consider the GWPF to be a reliable source, I don't see how that represents any kind of recommendation.

Tom said...

What on earth would make you think I'm recommending anything? Or that I consider the GWPF to be anything at all, good or bad?

Andy Mitchell said...

You linked to the GWPF in one of your first posts on this very thread. So what are you doing citing its productions as evidence for anything?

Tom said...

I linked to a specific piece that happened to appear in the GWPF. Are you somehow suggesting that that implies endorsement of everything that appears in their various publications?

Are you somehow suggesting that the mere fact of its appearance in that venue automatically guarantees that it is false to fact?

I'm really curious about this. I'm having a spirited discussion with conservative defenders of Donald Trump over at the Blackboard. They refuse to consider anything that was printed in the Washington Post or the NY Times.

Can you explain in what way you are different from them in this regard?

Andy Mitchell said...

"I linked to a specific piece that happened to appear in the GWPF. Are you somehow suggesting that that implies endorsement of everything that appears in their various publications?"

So is the GWPF a reliable source or not?

Tom said...

Is the Lancet a reliable source? Yes. Did they publish Wakefield? Yes. Bad stuff appears in good publications. Good stuff appears in bad publications.

This should be obvious. I fail to understand why you... overlook... the obvious. Do you have any evidence or have you even seen any claims that Goklany is wrong?

Andy Mitchell said...

So is the GWPF a reliable source or not?

William M. Connolley said...

This appears to be one of those situations where no-one has anything new to say but no-one will accept not having the last word. So, I will: any more comments re Golk / Lancet that add nothing new will be deleted.

Andy Mitchell said...
This comment has been removed by a blog administrator.