For discussion – the tornado "hockey stick"

A Doppler on Wheels (DOW) unit observing a tor...

A Doppler on Wheels (DOW) unit observing a tornado near Attica, Kansas. (Photo credit: Wikipedia)

I’m caught between sessions but wanted to post this. From a presentation at AGU that I couldn’t attend. This from SciAm:

James Elsner at Florida State University has a killer curve, and lots of caveats. The curve indicates that tornadoes in the U.S. may be getting stronger. The caveats indicate they may not be.

“If I were a betting man I’d say tornadoes are getting stronger,” he noted on Tuesday during a lecture at the annual American Geophysical Union fall meeting in San Francisco.

But when asked directly at a press conference whether that is the case, he would not commit. “I’m not doing this [work] to establish the future intensity of tornadoes,” he explained, but to establish a method that someday could indeed determine if the storms are becoming more powerful.

Because the lecture was titled “Are tornadoes getting stronger?” the audience expected an answer. And their consternation rose when Elsner showed his final graph, adding up the kinetic energy of tornadoes each year from 1994 to 2012.

More here: http://blogs.scientificamerican.com/observations/2013/12/11/tornadoes-may-be-getting-stronger-or-not/

This reminds me of Dr. Ryan Maue’s ACE (accumulated cyclone energy) for hurricanes.

tornadoes_kinetic_energy_trend

Since measuring tornado wind speed is a hit/miss proposition, even with doppler radar I have many reasons to suspect the data in this graph.

Elsner has 18[years of data]. His data begin in 1994 because that’s when Doppler radar, the best at tracking tornadoes, began covering the entire U.S.

The point of the curve, however, is to show that measuring the length and width of a tornado’s damage path gives an accurate indication of its strength, which is driven by the storm’s peak wind speed. It is difficult if not impossible to measure that speed directly, as is done for hurricanes by ground instruments and planes that fly into the storms.

So, like Mann’s hockey stick, it is a proxy, not the actual measurement.

Advertisements

  Subscribe  
newest oldest most voted
Notify of

What happened to the 1970’s?

TomL

Looks like a one-year spike.

From NOAA’s On LIne Tornado FAQ page:
What was the biggest known tornado? On 31 May 2013, a deadly, multiple-vortex tornado near El Reno, OK carved a maximum path width of 2.6 miles (from preliminary data). That width barely exceeded that of the Hallam, Nebraska F4 tornado of 22 May 2004. Two and a half miles probably is close to the maximum size for tornadoes; but it is quite possible that others this size or somewhat larger have occurred that weren’t sampled by high-resolution radars or surveyed so carefully in the field.
http://www.spc.noaa.gov/faq/tornado/index.html#Climatology
Rather says it all.
More junk science from Elsner.

Meanwhile, EF3+ tornadoes have clearly declined since the 1950-70’s.
http://www1.ncdc.noaa.gov/pub/data/cmb/images/tornado/clim/EF3-EF5.png

albertalad

These people are more and more displaying characteristics with the insane guy on the street holding up his sign – The end of the world is nigh – repent!

Berényi Péter

Bullshit. Tornado intensity scale is logarithmic, therefore annual integrated kinetic energy is dominated by the strongest ones. Now, there were almost twice as many F5/EF5 tornadoes in the first half of the epoch since 1950 than in the second one.

Doug Proctor

If it is not definitive, then the data must be bogus.
Strikes me he is trying to have it both ways: make a claim that isn’t a claim, so he can’t have to produce data to back it up, but have a useful graphic added to the toolbox of CAGW supporters.
It looks like another model, one of his imagination or “in principle”, something he would claim we should see with CAGW …. but haven’t yet, but will be “observations” to the unsuspecting.
More gaming the system.

Kip Hansen

Lots of talk about this topic based on Richard Muller’s “The Truth About Tornadoes,” Op-Ed article asserting there was a measurable decline in strong tornadoes over at Dot Earth.
http://dotearth.blogs.nytimes.com/2013/12/09/a-closer-look-at-tornadoes-and-global-warming/
A raft of experts rebutted Muller here: http://www.livescience.com/41632-the-truth-about-tornadoes.html
and have filled in some details for Revkin as a follow up.
Paul Markowski and Harold Brooks have been participating in the comments section.
[Spoiler Alert]
Bottom line — there is not enough data to prove no decrease…but much stronger case can be made for there not being an increase.

aaron

My goodness, does Elsner think that we are all idiots? If the proxy is the size of the “swath path” then this metric can only be used when the twister rips through a community where the damage path can be accurately measured and the media interest is high.
Most tornados hop across open farmland where the so called “swath path” is very difficult to measure and often is not. No one cares much about a long lived F5 that ruffles up a recently harvested open farm field. The measurements he is using are skewed by the probability of how many tornados actually hit populated areas. Recently there has a rash of these striking populated areas mainly because of urban sprawl.
When and if accurate direct wind speeds are made on a large sample of tornados in all areas over a long period of time, a more meaningful metric in the meantime is how many tornados form every year. So far this simple metric shows a down turn in the quantity in spite of the fact that more eyes are on the skies now than in the past.
This silly paper is just one baby step above using divining rods, basically warmist pseudo-science masquerading as science.

Janice Moore

(fyi — link to scientific American article didn’t work as of a few minutes ago)
Okay. Just for the fun of it, Mr. Eisner, let’s (just for grins) say that tornadoes are increasing in intensity. So what?
You have proven exactly: nothing.

KNR

Facts be damned, Elsner knows the take away ‘impact ‘ is in the image and he also knows that which comes at the end is best remembered. Grant seeking science by PR in action.
How long this enters the dogma of ‘the cause ; and we it held up as ‘proof’ of AGW ?

Have a look at NOAA United Stated Annual Trend of LSR Tornadoes, A plot of the annual running total of U.S. tornadoes.
http://www.spc.noaa.gov/wcm/torngraph-big.png

Janice Moore

Sorry — ELsner (I know some Eisners, never heard of “Elsner” — likely very few others have either, bwah, ha, ha, ha, haaa)

clipe

In all this REMEMBER that RADAR does not measure the speed of ‘air’ movement directly (although frontal activity can often be seen at the juncture of two differing airmasses), but rather any debris, rain, hail or other aqueous meteors (Hydrometeors) which may be present.
Also, be aware that there are various algorithms running on the various data processing platforms involved in the processing of Doppler imagery as detected by the WSR-88D network … some of these algos have gotten ‘revisited’ over the years as the various digital processing systems have been upgraded in the WSR-88D ‘hardware chain’ to ‘COTS’ (Commercial Off the Shelf) processors rather than the specialized hardware originally fielded back in the late in the 90’s.
The first installation of a WSR-88D for operational use in everyday forecasts was in Sterling, Virginia on June 12, 1992. The last system of this installation campaign was installed in North Webster, Indiana on August 30, 1997
.

Larry Ledwick (hotrod)

Berényi Péter says:
December 11, 2013 at 3:22 pm
Bullshit. Tornado intensity scale is logarithmic, therefore annual integrated kinetic energy is dominated by the strongest ones. Now, there were almost twice as many F5/EF5 tornadoes in the first half of the epoch since 1950 than in the second one.

One minor little detail there, prior to doppler radar, and the development of the EF ratings, tornadoes were only rated based on ,b>observed damage to structures. The strongest tornadoes went unrated if they did not manage to hit a building to allow them to be rated, and then you only got a brief snapshot of the damage potential at that moment when the storm impacted the structure. Ripping up fences, trashing cars and slam dunking cattle did not provide any rating information even though the storm winds might have been well into the F5 range.
A friend of mine (storm chaser) was investigating a very powerful tornado that struck in the extreme northeast part of Colorado and South Western Nebraska years ago. It was officially unrated, but its winds sucked the pavement up off of roadways and shredded cars, breaking them up so badly, that durable components like engine blocks were stripped of their accessories like intake manifolds, and transmissions broken off the engine. While trying to reconstruct the damage track, he talked to a farmer who was trying to find a big enough piece of his farm equipment to prove to the insurance company it had been destroyed by the tornado. It was probably a high 200 mph wind speed tornado due to incidental damage reports he discovered during his investigation, but it missed all structures and was never rated by the National Weather Service.
As a result the reports earlier than the wide spread use of doppler and the EF rating system will be significantly biased low on reports of the strongest tornadoes. This is a strong built in hockey stick bias to the analysis.

Key words here:
Elsner has 18[years of data]. His data begin in 1994 because that’s when Doppler radar, the best at tracking tornadoes, began covering the entire U.S.
. . . “began”
It took a good part of the 1990’s to install the entire network. The system designers and the Install crews had their work cut out for them, from site selection, site acquisition, preparation (grading, electric power, the ‘laying’ telco T1 “T-span lines” for data-backhaul to the various NWS field/forecast offices), tower and building construction, transmitter and processing equipment installation and commissioning at both the RADAR site and local field offices. Then came the training of the NWS personnel for operational use …
This was a big project.
.

Bruce Cobb

They just really really need for them to be getting stronger, because it helps sell the CAGW sky-is-falling-we’re all-gonna-die cause. Especially since the warming has stopped for some 17 years, and people are starting to notice.

Larry Ledwick (hotrod) says December 11, 2013 at 3:45 pm

One minor little detail there, prior to doppler radar,

We aren’t going to totally ignore the WSR-57 and WSR-74 RADARs that covered the eastern (basically, east of the Rockies) US, are we?
RADAR ops were ‘trained’ to look for specific features on ‘returns’ (hook echo anyone?) in these days too. I recall in the mid 1970’s onward Storm spotters here in the Dallas/Ft. Worth area depended on the RADAR operators in Stephenville (SW of Ft. Worth) as well as the several TV stations (such as CH 5 and Harold Taft) for ‘heads up’ from the RADAR perspective …
.

Chris @NJSnowFan

Chart looks like the increase in doplar radar sites.

Jarrett Jones

I suspect the data point for 2013 will be about the same or less than 2012. This presentation is one of the finer cherries I have seen picked. The historic F3/5 graph show this clearly. Not including this graph in the presentation for context speaks loudly regarding its level of credibility.

Jimbo

Has the process of peer review changed?

“If I were a betting man I’d say tornadoes are getting stronger,”

If it has then I want to say:

“If I were a betting man I’d say tornadoes are getting weaker,”

Is James Elsner funded by an insurance company or does he have shares in any insurance company?
PS I don’t have any fossil fuel interests whatsoever – except I use them.

highflight56433

Looks like another hockey stick. More breaking news on break wind.

u.k.(us)

_Jim says:
December 11, 2013 at 3:57 pm
===========
Even worse, the tail-end “charlies”.
It is not so much the line of storms sometimes, it’s the independent super-cells that need watching.
Thus the radar.

Bill_W

Aside from all of the standard issues: comparing apples and oranges, only starting at a time when more sensitive instruments are coming online, inconsistent error bars, etc. – the trend is defined by a few years. Weak data, weak conclusions.

u.k.(us) says December 11, 2013 at 4:27 pm
Even worse, the tail-end “charlies”.
It is not so much the line of storms sometimes, it’s the independent super-cells that need watching.

Friend, these independent super-cells have MORE than just an identifiable RADAR signature that allows them to be identified … have you ever been to a NWS-conducted Skywarn class? The first one for me back in the 70’s was an eye-opener … conducted by (now-retired) FT NWS meteorologist (Dr.) Alan Moller. Alan storm chased with Dr. Chuck (Charles) Doswell who wrote many, many papers on super-cell thunderstorms …
.

Chris @NJSnowFan says December 11, 2013 at 3:58 pm
Chart looks like the increase in doplar radar sites.

Good point; later in the series it could be coincidence with the various enhanced or improved ‘data’ products output by the (now) Open Source RPG (Radar Products Generator) via ‘software upgrades’ as described here in a paper from about 2003:
https://ams.confex.com/ams/pdfpapers/57682.pdf
Fair-use excerpt:

2. FUNCTIONALITY
Functional enhancements have been deployed in Open RPG Builds 1, 1.2, 2 and development is progressing for future builds scheduled for release every six months. Enhancements developed by the NWS for RPG Builds 1.2, 2, and 3 are described below.

2.1 Full Resolution Base Products
Full resolution base reflectivity and velocity products were provided with RPG Build 1.2 and are displayable with AWIPS Build 5.2.1. These products are at the same resolution and coverage area as the RDA base data.
Consequently, they are very large (160-330 Kbytes) and so the high speed RPG-AWIPS TCP/IP product distribution interface is needed to acquire them in a timely manner. In RPG Build 2 these products became replayable, meaning that a product will be generated and distributed immediately upon one-time request. In AWIPS OB1, the capability to remove a specified storm motion from the full resolution base velocity product will be available.
Full resolution base products will improve the visual identification of mesocyclone and tornado vortex signatures.

“Base Products” refers to (basically) raw-type ‘imagery’ acquired by the RDA (the RADAR Data Acquisition system), which is the ‘dish’ and accompanying receiver and transmitter and supporting site that actually ‘acquires’ the RADAR picture as seen by transmitting and receiving the narrow RADAR RF (radio wave energy) pulses. ‘Base’ products as opposed to ‘derived’ products such as re-constructing (in software) synthetic 3-D images from multiple ‘sweeps’ (at different elevation angles or scans) of the RADAR dish antenna.
Also see the paper for more info and accompanying graphics.
.

timetochooseagain

18 years? Odd, then, that he takes his hockey stick record back further. But an important question here is, what is the actual relationship of this variable to temperature? That last point there, which I am assuming is 2012, is a large drop from the highest value (2011?) Which is interesting, because the temperature in the US underwent a big jump-*up* when it looks like the tornado energy went *down*. I’d need the full dataset to do a proper correlation analysis but it looks to me like there is *no* clear relationship to temperature here *at all*.

Bill Illis

Tornado activity is going to be the lowest on record in 2013 (going back to 1954).
This is an adjusted for under-over-counts chart from the NOAA. The Pink line was the previous record low and 2013 is obviously going to come in under that.
http://www.spc.noaa.gov/wcm/torgraph-big.png

ossqss

Ok, so who has a coverage map showing the fully operational doppler sites as they progressed over the time period referenced? Don’t tell me they smooth or homoginize those too /$arc

tokyoboy

The tornado Hockey Stick (TM) period perfectly overlaps with the temp Pause (TM) period.
WUWT??

Chuck L

It seems to me that an increase in tornado intensity would be more likely the result of cooling – the temperature contrast between air masses increases the instability, and tightens and intensifies the jet stream. Global warming, if it were to occur, ultimately would reduce tornado intensity since the contrast between warm and cold air masses would be less and the jet stream slower.

mojo

Money where yer yap is, pal…

Ban Ki-moon
Secretary-General of the United Nations says:
December 10, 2013
“Ice caps are melting, sea levels are rising and the oceans are becoming more acidic. Greenhouse gas emissions continue to rise – we are the first humans ever to breathe air with 400 parts per million of carbon dioxide. Extreme weather events – heat waves, floods, droughts and tropical cyclones – are more frequent and severe.
We need look no further than the recent catastrophe in the Philippines. All around the world, people now face and fear the wrath of a warming planet.”

u.k.(us)

_Jim says:
December 11, 2013 at 4:40 pm
… have you ever been to a NWS-conducted Skywarn class?
============
No, but I worked outside everyday from 1985 – 2005.
I’ve seen every kind of weather She could throw at the northwest suburbs of Chicago.
Never a tornado though.
Used to watch the sky,… it was before real-time radar displayed on the “phone”.
Our electronics didn’t like the rain, or wet snow, other than that we needed to meet deadlines and try to make a profit.
I’ve lived by radar returns and forecasts.
It’s still fun and exciting.

John Morpuss

One thing I see is We are in such a hurrey to understand how these systems work that if the conditions even look like producing a tornado man with all their electromagnetic senscing equipment attacking the potential tornado producing storm with electromagnetic radiation from all sides, meanwhile the national weather service is doing their thing tracking it with their radar systems only add energy to the system and help it to intencify. To see were I’m comming from you will have to look at a tornado as a eletromagnetic process .
http://www.sott.net/article/229191-Tornado-A-Natural-Charged-Sheath-Vortex
http://en.wikipedia.org/wiki/Field_(physics)

Janice Moore

@ John M (6:21pm today). “To see were I’m comming from you will have to look {through a powerful telescope}… .”
Sorry, John M., no one here (except, perhaps, Ja! M!tchell) is likely to EVER get it. I hope you can find a little peace of mind… somehow.

John Morpuss says December 11, 2013 at 6:21 pm
One thing I see is We are in such a hurrey to understand how these systems work that if the conditions even look like producing a tornado man with all their electromagnetic senscing equipment attacking the potential tornado producing storm with electromagnetic radiation from all sides, meanwhile the national weather service is doing their thing tracking it with their radar systems only add energy to the system and help it to intencify.

WT*? Are you simply mad or uneducated in sciences and physics?
Do you have ANY idea what the AVERAGE RF power output is from a WSR-88D RADAR?
ANY IDEA AT ALL?
Can you Google it even?
Do you know how that figure compares to the heat output from your average vehicle or a house air conditioner?
.

David S

What was left unsaid is that the spike in intensity must be due to global warming. But the spike occurs in the last 13 years, and there has been no warming for the last 17 years. Conclusion; any increase in intensity is not due to global warming.

John Morpuss says December 11, 2013 at 6:21 pm
..
To see were I’m comming from you will have to look at a tornado as a eletromagnetic process .

How does this enter into process with the release of latent heat from from condensing water vapor in the parent thunderstorm?
Do you have any idea what I’m asking?
.

John Morpuss

@ Janice Moore Two things that have helped man progress , And the way we see things has been a major impact on knowing While the radio telescope has improved deep space research the radio microscope has accelerated understanding how things work here on Earth at the nano level . No use CERN smashing mater together if they can’t see whats going on. So here on Earth I’d say the advancement in the microscope has a greater effect on our everyday lives then the telescope.

John Morpuss

@ Jim, Heat IS electric potential at WORK ( ELECTRon, ELECTRicity and ELECTRONics ) What do these all have in common . Radio Is short for Radiated electromagnetic wave.

John Morpuss

Here is a couple of other things that may help people see a clearer picture Their about how the fair weather electric field…
[…Snip. Weather control, HAARP, and related comments are not allowed here. — mod.]

John Morpuss says December 11, 2013 at 7:47 pm
@ Jim, Heat IS electric potential at WORK ( ELECTRon, ELECTRicity and ELECTRONics ) What do these all have in common . Radio Is short for Radiated electromagnetic wave.

You didn’t answer the question; please stick to the subject and answer the question, if able. No more hand-waving or gobbledygook.
Let me be plain and straight away. I think you’re an idiot and a prime example of the Dunning–Kruger effect effect, which states:
The Dunning–Kruger effect is a cognitive bias in which unskilled individuals suffer from illusory superiority, mistakenly rating their ability much higher than is accurate. This bias is attributed to a metacognitive inability of the unskilled to recognize their ineptitude.
[Insults, unless they advance the argument being presented, should be avoided. Mod]
.

Janice Moore

Yes, yes, John M., but, I stand by my statement that we would need a powerful telescope to see where you are coming from.

Janice Moore

@ _Jim, I may be mistaken, but I think this is a fellow who needs our sympathy, not our scorn…. far more than he will ever know, I’m afraid. The human mind is such a fragile thing… .
btw: GREAT answers demonstrating your super-extensive knowledge (not to mention sharp analytical thinking) above, blank Jim. Glad you are here on WUWT to so often provide us all with first class science.

Janice Moore

Dear John M.,
Please forgive my joking about needing a telescope above. I hope that you can find a sympathetic ear somewhere. Hang in there.
Take care,
Janice

John Morpuss

@ Jim Do I have to provied you with the difference between a forcing and a feed back come on get over yourself and you might learn something RADAR uses a small forcing to create a large feedbak. Turn what you know about how the sun workes and turn it upside down . IF there have been people following this pathway I’m trying to create they must start seeing how weather modification IS a part of our everyday lives

John Morpuss

@ Janice Moore BAZINGA LOL Cheers

Owen in GA

How does the author get off calling it 18 years of data? The system wasn’t fully fielded until 1997, and was continuously tinkered with on the processing side up to present. The data in my lab has to be collected on the same setup to assure apples to apples comparisons and data consistency. How can you compare the data processed on the 1997 equipment with that collected from the upgrades fielded in the years since? That is just insanity. It is a little like assuming that the ARGOS floats MUST have a COLD bias because the data doesn’t match the MODELED output then adjusting all the data up to compensate! You have to scoop the sensor up and do a full lab calibration to make such a claim and implement it. Undergrads fail lab courses for doing this, is it better to do because these yahoos have PhD.s?