FSU Professor Links Tornado Strength, Frequency to Climate Change

[bobbutts]

Unfortunately it appears that the paper is behind a $40 paywall.

Official Link

 

 

 

although tornadoes are forming fewer days per year, they are forming at a greater density and strength than ever before

 

[blizzard1024]

Unfortunately it appears that the paper is behind a $40 paywall.

Official Link

 

These paywalls are a scam. These studies are publically funded. It costs several thousand dollars to publish the paper and then you have to pay again?

 

Without even reading the paper...what a joke! Really? Stronger tornadoes. Anyone with a degree of meteorology knows that you need more vertical wind shear and CAPE for strong tornadoes. Well, if you think about it, it if warms aloft more than below with AGW, then the atmosphere becomes more stable. Also with weakening horizontal temperature gradients hypothesized by AGW, you would have less vertical wind shear.  Plus we are finding more and more tornadoes because of Doppler radar and denser spotter networks and more aggressive verification schemes. Anyway, this kind of stuff really takes credibility away from climate science.

[Jonger]

These paywalls are a scam. These studies are publically funded. It costs several thousand dollars to publish the paper and then you have to pay again?

 

Without even reading the paper...what a joke! Really? Stronger tornadoes. Anyone with a degree of meteorology knows that you need more vertical wind shear and CAPE for strong tornadoes. Well, if you think about it, it if warms aloft more than below with AGW, then the atmosphere becomes more stable. Also with weakening horizontal temperature gradients hypothesized by AGW, you would have less vertical wind shear.  Plus we are finding more and more tornadoes because of Doppler radar and denser spotter networks and more aggressive verification schemes. Anyway, this kind of stuff really takes credibility away from climate science.

 

If true, this deserves congressional attention.

[ben4vols]

If true, this deserves congressional attention.

 

That is what got us in trouble in the 1st place.  

[ben4vols]

Unfortunately it appears that the paper is behind a $40 paywall.

Official Link

 

Anyone willing to pay $40 to read that rubbish should have their head examined.  

[sokolow]

If true, this deserves congressional attention.

Libraries have been pushing this for a literal generation and there's a boycott movement in the universities. There's been slow movement by the Feds pushing their agencies to implement open access policies. The Association of American Publishers, of course, hates all this and has fashioned custom lawsuits to sue universities for supposed copyright violations in classroom use while whining about how publically owned or subsidized open access journals unfairly exclude commercial enterprises from scientific publishing, in what amounts to a socialistic attack on free enterprise. I will let you imagine what the profit margins for some of those operations are.

[blizzard1024]

Libraries have been pushing this for a literal generation and there's a boycott movement in the universities. There's been slow movement by the Feds pushing their agencies to implement open access policies. The Association of American Publishers, of course, hates all this and has fashioned custom lawsuits to sue universities for supposed copyright violations in classroom use while whining about how publically owned or subsidized open access journals unfairly exclude commercial enterprises from scientific publishing, in what amounts to a socialistic attack on free enterprise. I will let you imagine what the profit margins for some of those operations are.

 

wow. we agree on something. cool. 

[bobbutts]

So who is actually paying $45 a pop for this research?

Is there some correlation with the quality of work in free studies vs. paid?

[ORH_wxman]

So who is actually paying $45 a pop for this research?

Is there some correlation with the quality of work in free studies vs. paid?

 

 

I'm not sure...but its unfortunate that this one is behind a paywall...the press release quotation says "although tornadoes are forming fewer days per year, they are forming at a greater density and strength than ever before"....I find that last bit in the quote an interesting claim given that stronger tornadoes (>=EF2) have actually shown a slight decrease in the dataset. But since I can't read the paper, I have no idea whether the quote is just offtopic drivel (this happens sometimes in press release quotes) or if they actually argue this in the paper with some other tornado dataset that perhaps I am unaware of.

[sokolow]

@bob

Not really. What there is is a relationship between traditional outlets and prestige b/c a pretty big chunk of the high impact / big name journals & their archives are pay for play. Not too much most individual researchers can do about that because junior scholars have to gun for big name journals (to get hired), tenure track scholars have to gun for big name journals (to stay hired), and senior scholars have to gun for big name journals (to raise the profile of their department & their teams in order to secure funding from the Uni and the granting agency / as well to make sure your 2nd, 3rd, 4th, etc. authors who might be junior or int'l scholars get their contributions in recognized outlets).

Like we were explicitly instructed by our mentors to sit on our first papers rather than take the CV hit comes from publishing record made up entirely of "mediocre" journals.

Also think about how much of the argumentative structure every paper rests on a lit review & field history which resides in journal archives held behind paywalls. Not only can persons without comprehensive access not read work their taxes paid for, you can't even work through the citation chain for papers that are free. Changing institutions or going private changes access; I was abroad for a while and my overseas affiliation had a different access set which meant I was nagging colleagues to send me articles. Long enough out of the academy and you end up having to pay to access the published versions of your own research.

It's terrible.

@ORH / everyone if noone else has access I'll do my best to summarize it in enough detail for everyone else to critique it

[Msalgado]

FYI if you don't have access to an academic library with journal access, your local public library likely will.  They can get most of these articles for free via inter library loan.  Open access is preferable, but the access is attainable with a minimum amount of legwork for almost everyone as it stands.

 

You have to understand that this system is a relic from a time before the internet.  Its easy to publish an entirely electric publication without much overhead, but that wasn't the case when journals were actual paper journals with no other means of distribution.  The research is paid for by grants, but the publication/review costs have to be covered as well.  

[bobbutts]

Thanks for the answer, sololow.  

One interesting thing about climate change in general is I think there will be lots of surprises as far as effects go especially localized ones like tornadoes in the USA or say snowfall in New England.  

 

People start with a simple assumption, like warmer = less snow, and then as the weather plays out, often times we observe the opposite.  

Another example is Atlantic Hurricanes.  More OHC = More Hurricanes.  But in reality there are so many other factors that there may be no or even a negative correlation.

 

I think the bottom line point I'm trying to make is that predictions are sketchy when compared to observations.  I'm not saying this automatically makes the Tornado Paper I mentioned in the OP valid, but it would fit in to this kind of pattern.  It's certainly easier and I think more fruitful to debate conclusions based on observations rather than model predictions.

[ORH_wxman]

Thanks for the answer, sololow.  

One interesting thing about climate change in general is I think there will be lots of surprises as far as effects go especially localized ones like tornadoes in the USA or say snowfall in New England.  

 

People start with a simple assumption, like warmer = less snow, and then as the weather plays out, often times we observe the opposite.  

Another example is Atlantic Hurricanes.  More OHC = More Hurricanes.  But in reality there are so many other factors that there may be no or even a negative correlation.

 

I think the bottom line point I'm trying to make is that predictions are sketchy when compared to observations.  I'm not saying this automatically makes the Tornado Paper I mentioned in the OP valid, but it would fit in to this kind of pattern.  It's certainly easier and I think more fruitful to debate conclusions based on observations rather than model predictions.

 

 

That's one reason I'm interested in their claim of stronger tornadoes. The actual observations do not support that claim by itself. Though perhaps they are saying something like "the stronger tornadoes are being clustered more in less frequent events but higher numbers of them on a per event basis" or something to that effect.

 

Rather than getting X number of stronger tornadoes spaced out normally throughout the season, perhaps they are being clustered in spurts more recently. But hard to say since the press release didn't specify that and I haven't/cannot read the paper without paying for it.

 

 

Or maybe they used some other dataset that I am unaware exists. I only know of the NOAA one.

[sokolow]

That's one reason I'm interested in their claim of stronger tornadoes. The actual observations do not support that claim by itself. Though perhaps they are saying something like "the stronger tornadoes are being clustered more in less frequent events but higher numbers of them on a per event basis" or something to that effect.

 

Rather than getting X number of stronger tornadoes spaced out normally throughout the season, perhaps they are being clustered in spurts more recently. But hard to say since the press release didn't specify that and I haven't/cannot read the paper without paying for it.

 

Or maybe they used some other dataset that I am unaware exists. I only know of the NOAA one.

They use the SPC archive for all reported tornadoes over the period 1950–2013 with the download link for the files they used as www.spc.noaa.gov/gis/svrgis/zipped/tornado.zip , saying they restrict their analysis to tornadoes rated EF1 and higher per SPC guidance, their description of the data and guidance taken from Verbout & crew (2006) "Evolution of the U.S. tornado database: 1954–2003."

[ineedsnow]

Maybe there is the same amount just more structures to be hit so its easier to determine intensity then before

[ORH_wxman]

Maybe there is the same amount just more structures to be hit so its easier to determine intensity then before

 

 

The tornado climatology is mucked up enough as it is. EF1 and higher is definitely better to use than all tornadoes due to the EF0 observational coverage bias when compraing past decades versus recent decades...but even EF1 isn't perfect. Many use EF2 or greater.

 

However, in the SPC/NOAA dataset, the EF1 and higher trend is essentially flat. There is a very slight statistially insignificant increase actually. EF2 and greater shows a slight decrease and EF3 or greater shows a more solid decrease. However, then we run into the trouble of rating past tornadoes...did they overrate them?

 

All these issues make tornado climatology very difficult.

[bobbutts]

Maybe there is the same amount just more structures to be hit so its easier to determine intensity then before

Being an academic paper one would assume they took such factors into account though.  Maybe it's a trash study based on trash data, thanks to the paywall we can only guess.

[sokolow]

First they look at tornadoes per year, and the authors note that mean annual rate is 505 tornadoes per year and the median rate is 474 tornadoes per year. They then observe that even given the inter-annual variation in the annual number of tornadoes, their statistical methods identify no long term trend -- its flat.

They go on to plot the number of days with least one tornado, calling those days 'one-tornado days, which varies a lot around a mean of 128/year -- max in 1979, min in 2013.

They then say that in contrast to tornadoes per year,

the number of tornado days has

been declining since the 1970’s. The total number of torna-

does is not trending so the atmosphere must be producing

more tornadoes on tornado days (Fig. 2). Here we define

four thresholds for tornado days starting with

N ± 4 tornadoes and continuing in powers of two. The starting thresh-

old is based on the fact that a four-tornado day is just above

the long term daily mean as indicated by dividing the mean

annual tornado count (505) by the mean number of tornado

days (128).

There is a downward trend in the number of tornado

days when a tornado day is defined by at least four torna-

does but upward trends in the number of tornado days when

a tornado day is defined by at least eight, 16, and 32 torna-

does. In particular before 1980 the number of days per year

with at least 16 tornadoes averaged between three and four.

Since 2000 the average has doubled to seven. The change

in the number of days per year with at least 32 tornadoes is

even more dramatic. Most years prior to 1990 had no days

with 32 tornadoes. However, since 2001 there has been no

year without a tornado day with at least 32 tornadoes and in

2011 there were six such days (see Fig. 2d).

Their next move is to say that the ratio of blockbuster days to a one-tornado day stands as the conditional probability of a big day

Said another way, given a day with at least one tornado

the ratio is the probability that the

day’s tornado count will exceed N tornadoes (Fig. 3). A

big [day] is defined for at least four, eight, 16, and 32 tornadoes.

Upward trends are noted in all cases (blue lines) with the

largest trend occurring when a big day is defined as hav-

ing at least 32 tornadoes. The trends are a constant percent-

age increase per year and result in curved lines on the plots

with the curvature most pronounced for the biggest tornado

days.

The next subsection of the paper deals with spatial clustering.

[sokolow]

The upshot of the section on clustering is they look at touchdown locations and conclude,

There is a significant upward trend in the annual num-

ber of tornadoes per cluster area (cluster number density)

(Fig. 8). The percent changes over the 60-year period

range from a low of 123 % for tornado days defined by at

least four tornadoes to 200 % for tornado days defined by

at least 32 tornadoes. Thus, the upward trend in the pro-

portion of tornadoes occurring on big days appears to be

related to an increasing concentration of tornadoes rather

than to larger areas favorable for deep convection. These

results are similar to increases in heavy precipitation days

during the 20th century over the United States (Groisman

et al. 2004).

Then comes the wrap-up. The gist of their summary & conclusions can be drawn from the latter half of the abstract, but the following portions bear on their interpretation of results and caveats on those findings, which I gather are gonna be of interest to you all,

The increasing density of tornado occurrences within

clusters suggests that the explanation for an increasing pro-

portion of tornadoes occurring on big days might involve

local-scale thermodynamics. In particular we hypoth-

esize that increases in both CAPE driven by increases

in low-level moisture and CIN driven by warming aloft

could lead to fewer days with tornadoes and to smaller,

but more active, areas of severe convection on days with

tornadoes. These findings and speculations are broadly

consistent with numerical modeling studies (Genio et al.

2007; Trapp et al. 2007; Diffenbaugh et al. 2013) of future

tornado environments especially those indicating that when

deep convection occurs it may more likely become severe

(Klooster and Roebber 2009).

Interpretation of the results rely on a consistent set of

tornado data. As noted by an anonymous reviewer of an

earlier draft, recent reporting practices with greater skill

at interpreting damage might have changed some events to

wind reports that would have been reported as tornadoes in

the past [see Speheger et al. (2002) for a discussion]. This

potential report bias might have some impact on the decline

in the number of tornado days.

[andyhb]

The tornado climatology is mucked up enough as it is. EF1 and higher is definitely better to use than all tornadoes due to the EF0 observational coverage bias when compraing past decades versus recent decades...but even EF1 isn't perfect. Many use EF2 or greater.

 

However, in the SPC/NOAA dataset, the EF1 and higher trend is essentially flat. There is a very slight statistially insignificant increase actually. EF2 and greater shows a slight decrease and EF3 or greater shows a more solid decrease. However, then we run into the trouble of rating past tornadoes...did they overrate them?

 

All these issues make tornado climatology very difficult.

 

This.

 

If 4/3/74 happened today, I'm fairly certain there wouldn't be 30 violent tornadoes (or at the very least, there wouldn't be 7 EF5 tornadoes).

[sokolow]

FYI if you don't have access to an academic library with journal access, your local public library likely will. They can get most of these articles for free via inter library loan. Open access is preferable, but the access is attainable with a minimum amount of legwork for almost everyone as it stands.

You have to understand that this system is a relic from a time before the internet. Its easy to publish an entirely electric publication without much overhead, but that wasn't the case when journals were actual paper journals with no other means of distribution. The research is paid for by grants, but the publication/review costs have to be covered as well.

All true & good points -- replied in banter thread

http://www.americanwx.com/bb/index.php/topic/38652-climate-change-banter/?p=3033670

[ORH_wxman]

Thanks for the posts sokolow...so I guess they never addressed the strength of tornadoes?

 

That would seem a bit of a discrepency with the press release quote in the original post.

[sokolow]

Thanks for the posts sokolow...so I guess they never addressed the strength of tornadoes?

That would seem a bit of a discrepency with the press release quote in the original post.

n/p ... and yeah i didn't catch tornado strength addressed in the paper or in his quotes from the press release. My guess is the author of the science daily splasher took Elsner's spoken / the paper's claim that when deep convection kicks off, its more likely to go severe and took that to mean "tornadoes are more severe"

Also I went to go reread the splasher and turns out he's active on Twitter as @hurricanejim, and he's posted a personal copy of both the paper as well as his code for anyone who wants to screw with it. Looks like he's amenable to questions

@hurricanejim

Interested in our new tornado study? Find it here ... Code for replication is here

paper (pdf link top of the pubs list)

http://t.co/Y2OAA5DM5L

code

http://t.co/iPI82a7afP

[LocoAko]

That's one reason I'm interested in their claim of stronger tornadoes. The actual observations do not support that claim by itself. Though perhaps they are saying something like "the stronger tornadoes are being clustered more in less frequent events but higher numbers of them on a per event basis" or something to that effect.

 

Rather than getting X number of stronger tornadoes spaced out normally throughout the season, perhaps they are being clustered in spurts more recently. But hard to say since the press release didn't specify that and I haven't/cannot read the paper without paying for it.

 

Or maybe they used some other dataset that I am unaware exists. I only know of the NOAA one.

Just released.

http://www.sciencemag.org/content/346/6207/349

 

Increased variability of tornado occurrence in the United States

Harold E. Brooks 1*, Gregory W. Carbin 2, Patrick T. Marsh 2

1 National Oceanic and Atmospheric Administration (NOAA)/National Severe Storms Laboratory, Norman, OK 73072, USA.

2 NOAA/National Weather Service Storm Prediction Center, Norman, OK 73072, USA.

Whether or not climate change has had an impact on the occurrence of tornadoes in the United States has become a question of high public and scientific interest, but changes in how tornadoes are reported have made it difficult to answer it convincingly. We show that, excluding the weakest tornadoes, the mean annual number of tornadoes has remained relatively constant, but their variability of occurrence has increased since the 1970s. This is due to a decrease in the number of days per year with tornadoes combined with an increase in days with many tornadoes, leading to greater variability on annual and monthly time scales and changes in the timing of the start of the tornado season.

Not sure if it is behind a wall since I'm on a federal computer right now, but here's the entire journal article: http://www.sciencemag.org/content/346/6207/349.full. Some really interesting stuff in there, especially the SD trendline.

Here's an article from Reuters about the study:

http://in.mobile.reuters.com/article/idINL2N0S92UP20141016?irpc=932

Severe weather alert: U.S. study finds tornadoes coming in swarms

Thu, Oct 16 23:30 PM IST

By Will Dunham

WASHINGTON, Oct 16 (Reuters) - Tornadoes in the United States are increasingly coming in swarms rather than as isolated twisters, according to a study by U.S. government meteorologists published on Thursday that illustrates another trend toward extreme weather emerging in recent years.

Looking at tornado activity over the past six decades, the study in the journal Science found the total number of tornadoes annually remaining rather steady, averaging 495. Since the 1970s, there have been fewer days with tornadoes but plenty more days with many of them, sometimes dozens or more.

On the list of the 10 single days with the most tornadoes since 1954, eight have occurred since 1999, including five since 2011. That year alone had days with 115, 73, 53 and 52 twisters.

The meteorologist who led the study, Harold Brooks of the U.S. National Oceanic and Atmospheric Administration's National Severe Storms Laboratory in Norman, Oklahoma, said emergency management agencies and insurers should be prepared to deal more often with days with lots of tornado damage.

The study analyzed the official U.S. tornado database for the six-decade period ending last year, excluding twisters below Category F1, with wind speeds of 73-112 mph (117-180 kph), on the Enhanced Fujita Tornado Intensity Scale.

Some experts have blamed weather intensity seen in recent years on global climate change they attribute to human activities. This study did not, however, offer a conclusion as to a cause.

"Knowing that the climate now has changed from that of the 1970s makes for a circumstantial argument in favor of a changing climate playing at least some role in the tornado changes," said meteorologist Patrick Marsh of NOAA's Storm Prediction Center.

"There are indications that heavy rainfall events are occurring with greater frequency globally, and given a warmer climate, this makes sense," added Storm Prediction Center meteorologist Greg Carbin.

But "any trend in tornado events is much more difficult to discern," Carbin added.

The average number of days annually with at least 20 tornadoes has more than doubled since the 1970s to upwards of five days per year in the past decade. For days with at least 30 tornadoes, there has been an average of three per year in the past decade, compared to 0.6 days per year in the 1970s.

Records for both the most and fewest tornadoes over a 12-month period have come in the past five years, with 1,050 from June 2010 to May 2011 and 236 tornadoes from May 2012 to April 2013. May is the month with the most tornado activity, followed by June and April.

Tornadoes, rapidly spinning columns of air usually spawned by rotating thunderstorms, can be among the most violent weather events. They have been reported on every continent except Antarctica but most often hit a U.S. region covering the Great Plains and parts of the Midwest and South.

Tornadoes can cause extensive loss of life and property damage like the May 2011 twister in Joplin, Missouri, that killed about 160 people and wrecked thousands of homes. (Reporting by Will Dunham, editing by G Crosse)

[CoastalWx]

Still looks like there is a wall Loco. Blah.

 

I see some references to post 1999 data. Was there any conclusion to things like decadal cycles? -PDOs love to produce tornadoes with an increased frequency of troughs ejecting out of the SW US. That doesn't explain there findings since the 70s, but perhaps something to think about recently.

 

There is no doubt tornadoes have been documented much more with the help have increased spotter network, smartphone recording, social media etc. Did this get accounted for? I think limiting the study to EF1 or greater helps, because those usually aren't missed as much as EF-0.  

 

In the days of dual-pol radar and spotter network/social media etc, it's getting hard to miss tornadoes unlike even 20 yrs ago.

[Hoosier]

Granted I'm not able to read the article that Jake posted but what they did seems pretty straight-forward, so I ran the numbers just on a regional basis for the Lakes/OV region and a similar trend is revealed.  The average annual number of days with (E)F1 or greater tornadoes has fallen significantly since the 1970s.  The 1950s through 1970s saw around 30-35 days per year on average with at least one (E)F1 or greater tornado, falling off by several days per year on average in the 1980s with the 1990s through 2013 holding relatively steady in the low 20s days per year on average.  So it's roughly 10 fewer days per year with (E)F1 or greater tornadoes nowadays compared to the 1950s-1970s.  Didn't get into the various thresholds of tornadoes per day to see if there's been an increase in big days, but I'm guessing there would be, much like the national trend.     

[Rainman]

That makes sense.  The decrease in baroclinicity will result in fewer favorable environments for tornado formation.  But I think you can also consider the increase in saturation vapor pressure through the column due to warming, and you have a scenario where it may be more difficult for intense updrafts to form, but the ones that do form will be more intense due to greater latent heat release.  The argument is similar to why total hurricanes might decrease while intensity increases.  A warmed environment could therefore be favorable for decrease in total number of tornadoes, but an increase in the ceiling on intensity.

 

I am not familiar with your idea that greenhouse warming disproportionately warms the mid/upper atmosphere.  My understanding is that it uniformly affects the depth of the troposphere, resulting in no net change in lapse rates.

 

These paywalls are a scam. These studies are publically funded. It costs several thousand dollars to publish the paper and then you have to pay again?

 

Without even reading the paper...what a joke! Really? Stronger tornadoes. Anyone with a degree of meteorology knows that you need more vertical wind shear and CAPE for strong tornadoes. Well, if you think about it, it if warms aloft more than below with AGW, then the atmosphere becomes more stable. Also with weakening horizontal temperature gradients hypothesized by AGW, you would have less vertical wind shear.  Plus we are finding more and more tornadoes because of Doppler radar and denser spotter networks and more aggressive verification schemes. Anyway, this kind of stuff really takes credibility away from climate science.

[donsutherland1]

Unfortunately it appears that the paper is behind a $40 paywall.

Official Link

I clicked on the blue "Download PDF" button and the whole article opened. I'm on my home computer and don't subscribe to Climate Dynamics. Perhaps a full version has been made available due to the media coverage?

[donsutherland1]

The news stories describing the paper might be going beyond what the authors themselves concluded. The authors cautiously suggested a possible link to climate change.

 

In their summary and discussion, the authors wrote:

 

We present empirical evidence of changes in tornado climatology possibly related to a changing climate…

 

The increasing density of tornado occurrences within clusters suggests that the explanation for an increasing proportion of tornadoes occurring on big days might involve local-scale thermodynamics. In particular we hypothesize that increases in both CAPE driven by increases in low-level moisture and CIN driven by warming aloft could lead to fewer days with tornadoes and to smaller, but more active, areas of severe convection on days with tornadoes. These findings and speculations are broadly consistent with numerical modeling studies…

 

As noted by an anonymous reviewer of an earlier draft, recent reporting practices with greater skill at interpreting damage might have changed some events to wind reports that would have been reported as tornadoes in the past (see Speheger et al. (2002) for a discussion). This potential report bias might have some impact on the decline in the number of tornado days.

[skierinvermont]

Don, that fits with what ORH was speculating earlier. 

 

"I'm not sure...but its unfortunate that this one is behind a paywall...the press release quotation says "although tornadoes are forming fewer days per year, they are forming at a greater density and strength than ever before"....I find that last bit in the quote an interesting claim given that stronger tornadoes (>=EF2) have actually shown a slight decrease in the dataset. But since I can't read the paper, I have no idea whether the quote is just offtopic drivel (this happens sometimes in press release quotes) or if they actually argue this in the paper with some other tornado dataset that perhaps I am unaware of."