Roger Pielke Jr.’s 2020 Hurricane Season Recap

2020 Atlantic hurricane season ends today
Here are updated CONUS normalized losses through 2020
2020 ranks 15th of 121 years
Based on @JessicaWeinkle et al

And CONUS landfalling hurricanes 1900-2020
Updated from @philklotzbach et al

And CONUS landfalling major hurricanes 1900-2020
Updated from @philklotzbach et al

And for readers preferring a different trend
CONUS landfalling hurricanes 1970-2020
Updated from @philklotzbach et al

Originally tweeted by Roger Pielke Jr. (@RogerPielkeJr) on November 30, 2020.

0 0 vote
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
December 1, 2020 6:27 pm


How is anyone going to make screaming headlines with those plots??? Can’t you at least colour them so that they go from a nice mild blue on the right-hand side to a screaming red at 2020 on the left-hand side?

Reply to  JaneHM
December 2, 2020 4:48 am

It can always be, “worse than we thought.”

Carbon Bigfoot
Reply to  Scissor
December 2, 2020 5:32 am

As crack meteorologist Joe Bastardi of Weatherbell Analytics has enumerated many times is that the goalposts for named storms has been changed to fit the CAGW Narrative.
Where is the chart that shows how the frequency would have been measured in the good old days?
Easy read even for Lukewarmers and Warmistas is Joe’s ” The Weaponization of Weather in the Phony Climate War”.

Caligula Jones
Reply to  JaneHM
December 2, 2020 7:35 am

Well, they did yesterday:

I mean, when Mann, the Master Modeler speaks, don’t YOU tremble?

Reply to  Caligula Jones
December 2, 2020 10:37 am

With laughter.

Chris Morris
December 1, 2020 6:30 pm

Can we agree that the MSM will only use the last graph and include none of the above commentary

Reply to  Chris Morris
December 2, 2020 3:34 am

Yes, as always with fitting “trends”, it depends when you start and finish.

I know Pieker jr’s expertise is the wealth adjusted damage figures but it would be interesting to see ACE ( energy) plotted too.

Reply to  Greg
December 2, 2020 6:44 am

Wikipedia reported ACE in 2020 is 178.8. This ranks it just out of the top 10. By comparison, 2017 was more intense with ACE of 224.9 (ranked #7 worst). Unusually, the last 5 years the ACE has been above 125 each year, the first time that has happened, as far as I can tell.

Total hurricane season $damage ranking for 2020 (not just CONUS) is ranked #7 worst. Of course damage depends on where they landfall so we got a bit lucky this year. The 1926 ACE was around 228, 4th worse. The worst was 1933 at around 239.

Matthew Schilling
Reply to  menace
December 2, 2020 8:25 am

The “A” in ACE is for accumulated. The ACE for the most recent five years that you cite is 811. For comparison, the ACE for the five consecutive years ending in 2005 was 832 – the highest I could find. The five year ACE ending in 2007 comes in third, with 806. Fourth place goes to 2002-2006, with 800, and Fifth Place ended in 1999 with 792.

The 21st Century has been quite active: Seven of the Ten biggest 5 Year ACE’s (I know, my own made-up category) are in the 2000’s. Two come from the 20th Century and one from the 19th Century: 1999 sits in Fifth place, 1936 in Eighth place, and 1896 in Tenth.

Hurricane activity this century suggests Willis’ escape valve has been busy lately – like a global air conditioner in a planetary summer.

Reply to  Matthew Schilling
December 5, 2020 6:10 am

Gee, I wonder why the 21st century has been unusually active. Perhaps the answer lies in catching more of those 3 hour tropical storms, or those 6 hour hurricanes. The detection of these small changes makes a difference, especially because they used to be missed but now are modeled and recorded as data.

Caligula Jones
Reply to  Chris Morris
December 2, 2020 7:37 am

From good old I learned years ago I learned of “start and end date bias”…

Danley Wolfe
Reply to  Caligula Jones
December 2, 2020 10:00 am

Very true. On the last plot start at 1985 and you have a very decidedly decline in hurricane incidence.

December 1, 2020 6:35 pm

The two graphs that show Continental Landfalling Hurricanes have different trends. The graph that shows the period 1900-2020 has a negative trend. The one that shows 1900-2020 has a positive trend.

If your purpose is to persuade people of your case, you can usually tailor the evidence to support your case without actually telling a lie.

Reply to  commieBob
December 1, 2020 7:12 pm

What does it matter if it is a negative or a positive trend? Both undeniably proves that AGW is real.

David Guy-Johnson
Reply to  Dmtr
December 1, 2020 11:52 pm

Ha ha ha, I like your sense of humour

Reply to  commieBob
December 1, 2020 7:12 pm

One shows major hurricanes, the other shows all hurricanes.

Reply to  commieBob
December 1, 2020 7:13 pm


The graph that shows the period 1900-2020 has a negative trend. The one that shows 19001970-2020 has a positive trend.

Reply to  commieBob
December 2, 2020 12:40 am

None of those show any trend. A “trend” of 0.5 of a hurricane over 50 years is nothing more than noise.

Reply to  ggm
December 2, 2020 7:38 am

Good point. One way to express the degree of unusualness of the now-concluded 2020 season, or of the recent 5 years’ ACE, would be to do a Bayesian Analysis. Given that ACE in 1900, 1901, 1902, 1903, and 1904 was [x1, x2, x3, x4, x5], how likely was it to have an ACE of [x6] in 1906? Carry forward to 2019: given subsequent ACE values for 1900 through 2019, how likely or unlikely was it to have the ACE we did have in 2020?

Reply to  commieBob
December 2, 2020 3:33 am

Pielke’s graphs show that landfalling U.S. hurricanes (both total and major) have a slight downward trend since 1900, but since 1970 it looks like a slight upward trend. What conclusions can be drawn from that?

Could it be that there was a hiatus of ‘canes between 1960 and 2015? That is what my eyeball sees in the 1900-2020 charts. If so, the 1970-2020 chart would see that as increasing. No?

Also, could it be that the criteria for classifying hurricanes has changed? It used to be based on sustained surface winds. Now it seems that surface gusts or satellite observations are now being used. Would that not change the substance of these reports? Satellites can spot events that would have been completely missed a century ago.

Steve Keohane
Reply to  Johanus
December 2, 2020 5:11 am

I am under the impression that they have gone from surface airspeed measurements to aircraft/satellite measurements at 10K feet or so. At least that what it looks like on Nullschool Earth. The claims of windspeed only correlate with the speed at 10K feet(700 hPa).

Mumbles McGuirck
Reply to  Steve Keohane
December 2, 2020 5:54 am

The ‘standard’ is always maximum 1-minute surface (10 meter) sustained wind. That value is never directly measured. So it is estimated by a hurricane specialist by whatever data is available. Preferably a landbased or buoy anemometer at standard height, but often it is aircraft using remote sensing of sea surface or a reduction of flight level winds. If no aircraft are in the storm the it’s estimated from satellite. Prior to 1970s there were no satellite estimates, prior to 1940s no aircraft estimates. The further back in time you go in hurricane database the larger the error bars.

Reply to  Johanus
December 2, 2020 5:54 am

“The claims of windspeed only correlate with the speed at 10K feet(700 hPa).”

If so, then they are looking at winds above the planetary boundary layer (PBL), where winds tend to be stronger and geostrophic (i.e. flowing parallel with isobars). Below the PBL, winds can actually flow towards low pressure areas (as if ignoring Coriolis effect).

AFAIK, the U.S. NHC still ‘officially’ uses the Saffir-Simpson scale, where ‘sustained’ means ‘one minute or longer’. (WMO and others are even more strict, requiring 3 to 10 minutes to ‘sustain’). But my own ‘observations’ (using on line tools/weather sites) suggest that the media were too eager to accept dire forecasts based on wind gusts alone.

The idea behind hurricane classification is not merely a contest to record the highest winds, but to indicate the scale of destruction to be expected from each scale. A single puff of wind will not reliably predict or measure the overall destruction to be expected. Indeed, wind alone will not adequately predict destruction due storm surge and rainfall. But, if one depends on wind alone, then a sustained wind, over some suitable period of time, must be applied. In other words, the predict relies on ergodic measurements (the idea that sampling over an sufficient interval will provide a reliable basis for prediction).

Robert of Texas
December 1, 2020 6:45 pm

The year 1900 gives us some historical perspective but seems rather random.

Not as random as 1970 though – what the heck is up with starting on that year? Why not 1964? Or 1985? I hate random starting points…it’s almost like someone trying to trick you with a graph.

Ron Long
Reply to  Robert of Texas
December 2, 2020 2:00 am

Robert of Texas, although NASA launched a weather satellite, TIROS, in 1960, it was not very useful, and the first actual useful weather satellite, NIMBUS, launched in 1969, so 1970 is a common start year for some “weather” data, and I’m guessing that was his reason to start that one chart at 1970. Stay safe.

HD Hoese
December 1, 2020 7:08 pm
“Page not found
The requested content could not be found.”
Nature article no problem. Been asked about this recently

Rud Istvan
December 1, 2020 7:16 pm

Just personal note comparing the 2004-5 graphic to 2020.
I moved to Fort Lauderdale, on the ocean, yearend 2000. In 2004? we took three three hits, two direct. Katrina as a then east to west Cat 1, then Wilma as a west to east marginal Cat 3.
Wilma was real bad. Took two years to repair the grounds damage. And took months to repair on of my significant other’s rental properties: back yard community shed destroyed, roof on a two unit complex blown off, windows blown in. I was hoisting wallboard to repair ceilings for weeks.

Then two years ago we did not evacuate, but rode out Irma. We were 120 miles from the eye over Naples, yet sustained near Cat 3 damage. To grounds and adjacent state park.

Point is, is hit or miss. And those Pielke bars do not say what was hit, only that something was.

Reply to  Rud Istvan
December 2, 2020 1:55 am

And those Pielke bars do not say what was hit, only that something was.

The first graph, normalised losses, does give some idea of what was hit in terms of economic value.

Anyway, the data provided by the numerous authors* who produced those graphs runs counter to the alarmist narrative. example

The alarmists can get their evidence by torturing the data. (If you torture the data, it will confess. Coase) At least the data presented here has the advantage of relative simplicity.

*numerous authors: Pielke is seventh author on the first cited paper. When I clicked on the second link, it returned a ‘not found’ error. At least we know Pielke wasn’t the first author.

John Andrews
December 1, 2020 7:46 pm

Those graphs look like log-normal data, not normal random values. If true, the slopes shown are probably incorrect because the slope equation is for a randomly distributed variable.

D. J. Hawkins
Reply to  John Andrews
December 2, 2020 6:16 am

Yes, I noticed there was a significant difference between the mean and median. That points to a non-normal distribution, whether log normal, bi-modal, or other.

John F Hultquist
December 1, 2020 8:06 pm

How about putting the “normalized losses” on a chart with highest “1926” on the left and the other years strung out to the right – seems there is a long tail. Then mark the 2020 bar.

I understand why this first chart is as drawn, but much information is left to the viewer to extract.

Wm. Briggs has critiqued bar charts with several posts.

Steve Reddish
December 1, 2020 8:35 pm

These graphs make it hard for anyone wanting to confirm the prediction of global warming causing more major hurricanes.

Reply to  Steve Reddish
December 2, 2020 7:42 am

Reddish: just wait until we get ot that Tipping Point!

John Pickens
December 1, 2020 10:58 pm

The graphs would look significantly different if two of those 2020 “hurricanes” were correctly recorded as the tropical storms they actually were at landfall.

Rob Thomson
December 2, 2020 12:01 am

This was reported in our local paper today only a few hours before our PM declared a climate emergency.

Reply to  Rob Thomson
December 2, 2020 2:44 am

Thanks for the link.

At first I couldn’t reconcile the above graphs with the NOAA article, that is 12 hitting land versus 6 landfall. Then I reread the NOAA piece where it said 12 storms versus the Pielke graph of 6 hurricanes.

To the average reader, just glancing at the article without paying attention to the precise wording, will think 12 hurricanes. Which is probably what NOAA would like.

December 2, 2020 1:41 am

An old paper from the Mississippi state university concluded that the solar activity has an influence on a hurricane season and should be considered for improving pre-seasonal hurricane forecasts and understanding inter-annual hurricane variability. “ …. the Solar Minimum years has the strongest correlation. Decreased sunspots actually increased tropical cyclone frequency”
Currently, solar activity is picking up. SC25 is underway in the earnest.
The early 21st century’s minimum fell short of either one at the early 19th or the early 20th century.
Is the landfall hurricane duration increasing?
I had a look at landfall hurricane acceleration over 6, 12 and 18 hours
for data period 1967-2018
(note: hurricane’s slowing down calculates to a negative acceleration)

Reply to  Vuk
December 2, 2020 2:10 am

Here is correct link
Solar Cycle Extremes as a Seasonal Predictor of Atlantic-Basin Tropical Cyclones

December 2, 2020 2:59 am

The first chart shows the incredible spike caused by The Great Miami Hurricane ( and several others) which devastated the Miami area,parts of Florida and much of the Gulf Coast in September 1926.
For those with an interest in history it features in Historian Frederic Lewis Allen’s work, “ Only Yesterday”, a revue of the 1920s.
As Allen recorded, it showed what a gentle Caribbean zephyr could do when it got a running start from the Bahamas!

December 2, 2020 7:08 am

How does one accurately “normalize losses”? Whatever hits today is obviously going to run into more structures and people than there were earlier.

Caligula Jones
Reply to  Jl
December 2, 2020 7:48 am

A very tough thing to guesstimate, that’s for sure. We simply have more stuff, and more expensive stuff.

If my dad’s basement flooded right now, he’d lost a few new tools, quite a few older tools (most of which he was either given or found in the dump and fixed up)…and furniture that went from someone’s home, to their basement, to his basement…40 years ago.

Even his furnace is almost 50 years old.

If my basement flooded today: newish computer, widescreen monitor, etc. Widescreen TV, etc. Probably a table or two. A few hundred books, many rare and collectable, and furniture my Executive Design Consultant insisted we spend some money on.

Thomas Gasloli
December 2, 2020 7:17 am

Maybe I’m miss-remembering, but didn’t they change the method of measuring wind speed in the hurricane?
Doesn’t the new method result in an increase in the measured speed of the hurricane?
Didn’t this happen around 2016, the place on the graph where there is a jump in the number of hurricanes making landfall?
There should be a flag on the graph making the point where this change in methodology was made.

Tom Abbott
December 2, 2020 9:22 am

Yes, there were a lot of named storms back in 2005, too, such as Katrina. I think 2005 had the old record for number of named storms before this year of 2020.

What happened after 2005? Well, there followed a 12-year hiatus of major hurricanes (Cat 3, 4, or 5) hitting the U.S. So don’t get overly excited over hearing 2020 had the most named storms. Lots of named storms didn’t mean much after 2005, and it may not mean much after 2020.

The alarmists are desperate to connect CO2 to the weather. Desperation, however, doesn’t change the facts: There’s no evidence CO2 is measureably affecting the Earth’s atmosphere or weather.

December 2, 2020 9:57 am

We now know that most of the MSM consists of transition team wannabes.

December 2, 2020 12:08 pm

Where is the graph for total Global Tropical Cyclone Accumulated Cyclone Energy?


Martha K.
December 3, 2020 9:07 pm

What happened to hurricane strength being measured by barometric pressure in the eye at sea level? Lower barometer = stronger storm? I clearly remember that as being more accurate in determining storm strength than one minute sustained wind speed at any height above sea level.

%d bloggers like this: