US Tornado Trends–Updated to 2011
Guest post by Paul Homewood
2011 has obviously been a disastrous year for tornadoes in the USA, but does this reflect an increasing trend in either the number or severity of tornadoes?
The NOAA graph below is often used to show trends of severe F3+ tornadoes and certainly indicates that there has been no upward trend up to 2010.
The provisional March-August figure for 2011 is 79 EF3-EF5 tornadoes, which would be much less than 1974 and on a par with 1965. However, this only gives part of the picture as it only runs from March – August, and excludes weaker tornadoes. Nevertheless comparing total tornado numbers can be misleading.
(Added by Anthony – For the entire 2011 year, the count is slightly higher:)
An NOAA study by McCarthy and Schaefer in 2002 identified a sharp rise in total tornado numbers between 1970 and 2002, but made it absolutely clear that this was changes in methods of detection, e.g.
There was a significant increase in tornado occurrence during two periods in the last 33 years – in the early 1980s when National Weather Service (NWS) warning verification began, and in 1990 when the WSR-88D became operational.
The increase in reported tornado frequency during the early 1990s corresponds to the operational implementation of Doppler weather radars. Other non meteorological factors that must be considered when looking at the increase in reported tornado frequency over the past 33 years are the advent of cellular telephones; the development of spotter networks by NWS offices, local emergency management officials, and local media; and population shifts. The growing “hobby” of tornado chasing has also contributed to the increasing number of reported tornadoes.
(The WSR-88D Doppler system was phased in between 1992 and 1997).
As a result of these changes, the proportion of F0 tornadoes (which Doppler detects) increased from an average of 39% between 1970 and 2002, to 64% for 1998-2002. As the authors stated,
“ Finally, it will be seen that the number of strong and violent tornadoes has not varied much since 1970.”
So bearing this in mind, and using data supplied by NOAA’s Storm Prediction Center, I have analyzed all reported tornadoes by category since 1970.
AVERAGE NUMBER OF US TORNADOES PER YEAR
(2000’s are 2000-2011)
| 1970’S | 1980’S | 1990’S | 2000’S | |
| F0 | 274 | 331 | 739 | 812 |
| F1 | 343 | 334 | 331 | 369 |
| F2 | 188 | 124 | 109 | 108 |
| F3 | 50 | 33 | 38 | 33 |
| F4 | 14 | 9 | 9 | 7 |
| F5 | 3 | 1 | 1 | 1 |
| TOTAL | 872 | 832 | 1227 | 1330 |
| TOTAL F1+ | 598 | 501 | 488 | 518 |
F0-F5 are based on the Fujita tornado intensity scale
The following graph shows the trends well.
(P.A. – per annum)
Conclusions
There has been no significant trend in F1+ tornadoes since 1980, and each decade since then has been less than the 1970’s. The 2000’s show more F0 tornadoes, but this is the first decade Doppler systems have been fully operational, so we should now have a good basis for the future.
==========================================================
Addendum by Anthony –
The graph for 2011 Total Tornadoes (all F0-F5 reported) shows that even though 2011 had some notable and terrible disasters, such as May 22nd Joplin EF5 tornado and the April 27th Mississippi-Alabama outbreak, it did not exceed the year 2008:
The ONLY things that are increasing are: the technologies and data sources used to detect and report tornados, and the hysteria created by CAGW acolytes that wish to push forward an agenda.
Any other interpretation of the scientific tornado data is hokum.
Maybe they could do it like they do it with sunspots – detect all events using available technology, then adjust these numbers to be comparable with historical records.
A year or two of “old method” and “new method” overlap would be good here as well. It’s the same issue as with the surface stations: If the instrumental methods themselves are a function of time, making anything better than a rough “proxy” for something is self-delusional.
Don’t forget that people have spread out since the 70’s. Towns have spread out and become small cities and there are more subdivisions where people like to live close to town but away from the traffic and noise. So they can’t really say there was an increase, just better reporting of them now than in the past.
In the UK, the Met Office failed to do its job once again.
A spokesman said: “The squally nature of the winds meant that gusts were much stronger than we were expecting.
“Given the impact the winds have caused, had we known, we would have put out a warning across the south.”
http://www.telegraph.co.uk/topics/weather/8995705/Britain-battered-by-hurricane-force-winds-after-forecasters-underestimate-strength-of-gusts.html
Doesn’t this bit -> 2000’s are 2000-2011
Mean the 2000’s cover 12 years and the rest only 10 oer decade ?
It is good to see that F3-5 tornadoe numbers are actually falling. These storms would be obvious over the years so these accumulated numbers will tell the real story. F0-2 numbers have risen due to detection technology so historic comparisons are difficult if not impossible.
Eyeballing the first chart :-
I see that from 1950-1974 when the climate was cooling the tornado trend was up. Since then with a warming climate the trend has been down, which is opposite to what the MSM tell us. So now if the climate cools the trend should one again go positive/up.
Makes perfect sense to me.
Let’s see what happens over the next few years.
But wait – they’ve been telling us its the worst in eleventy billion years or something. It couldn’t be as simple (as most rational folks understand) as better communication, better equipment and detection and better and more extensive storm reporting?
!970 shows a significantly higher incidence of strong tornado’s …. hmmm, wasn’t that when the scientists were warning of impending global cooling?
Man Bearpig
Doesn’t this bit -> 2000’s are 2000-2011
Mean the 2000′s cover 12 years and the rest only 10 oer decade ?
That’s right. The SPC data is batched together as 1970-79, etc, so it seemed sensible to continue with 2000 as the start point for the current “decade” of 12 years!
Excellent work. The chart tells the story! ALL the increase is in F0, which generally went unnoticed before the improvements in radar and communication.
In the types that have always been noticed (because they destroyed things and killed people) there’s no change.
You miss MBP’s point. That would mean the 2000s should be 2000-2009. The next decade is 2010-2019.
Study it. It’s not all that hard once you realize that under this convention a decade ends on a ‘9’, not a ‘0’ or a ‘1’. You could choose to start with ‘1’ and end with ‘0’, but never the reverse.
In dong trend spotting most people end up seeing what suits their mindset. The claim “still no upward trend” is so vague as to me impossible to agree or disagree with.
If I squint at the graph I’d say there was a rise from the beginning until about 1975, a decline until 2001 and a small but definite rise since.
Tornadoes are an event that accompanies cooling (according to some of the more credible articles on this site). So what I would see here is a fairly clear indication that we are in a cooling period (or at least the US mainland is).
Invert this graph and you have a rough match to last 60 years global temps.
Someone could look at doing some proper stats on that but I’d say there is a clear correlation in that sense.
RexAlan says:
January 6, 2012 at 2:34 am
Eyeballing the first chart :-
I see that from 1950-1974 when the climate was cooling the tornado trend was up. Since then with a warming climate the trend has been down, which is opposite to what the MSM tell us. So now if the climate cools the trend should one again go positive/up.
This was my thought on an simple eyeball / guesstimate level too. Which to me, if indications play out as I have seen and believe to project forward – then over the next few years, we could expect the numbers to start increasing again as the temperatures drop (similar to the 70’s). But, that’s just my non-climate ‘edumacated’ guess and not looking much deeper than the graph itself.
Brian H
You miss MBP’s point. That would mean the 2000s should be 2000-2009. The next decade is 2010-2019.
I could have run 2000-2009, but then it would not have been up to date and I could stand accused of deliberately ignoring 2011. (And taking 2010/11 as a two year slice would hardly give a long term trend).
Don’t get too tied up with “decades”. The data shows that tornadoes in the last 12 years are at a similar level to 1980-99.
P Solar
If I squint at the graph I’d say there was a rise from the beginning until about 1975, a decline until 2001 and a small but definite rise since.
McCarthy and Schaefer say “Finally, it will be seen that the number of strong and violent tornadoes has not varied much since 1970.” (This of course is up to 2002).
The average of F3+ since 2000 = 41 p.a., compared to 43 in the 80’s and 48 in the 90’s. I am not sure if these differences would be regarded as significant.
F0’s 1980’s 331 + doppler radar = 1990’s 739
A. Scott says:
January 6, 2012 at 2:36 am
“But wait theyve been telling us its the worst in eleventy billion years or something.”
Yes – THEY (aka the mainstream climate scientists) can now tell us UNEQUIVOCALLY that storms are worse now than ever by reading…you know… tree rings…and other proxies. Proxies can tell us SO MUCH using the right mathematical processing techniques…
Here in Blighty we have had some bad weather recently and, sadly, someone died when a fallen tree crushed him in his car (gale force winds, gusting to 100+ on high ground in the North and Scotland). My apologies to those of you in the USA, Canada, Russia, Australia, India, et al. You are probably all thinking: “with storms like that, what the hell are they worrying about?” Please understand that over here, ably led by our politicians and the BBC, all weather is disastrous and the worst ever known. The article below puts present reality into perspective:
http://www.dailymail.co.uk/news/article-2082939/Think-weeks-gales-bad-Theyre-storm-drowned-Britain.html#comments
From your table, even including F2’s would still show a decline. Only F0 and F1’s have “increased”.
I think it’s confusing to use TOTAL to mean “total average”. That’s why people are confused about you using 2000-2011 as a group, even though the graph title states that it’s the average per year.
“No upward Trend in Tornadoes” is clearly caused by Climate change. Try to keep up 😉
Yet more Windmills destroyed…..by the wind.
http://www.dailymail.co.uk/news/article-2083149/Wind-turbines-cope-UK-weather-3-blown-pieces.html
Whirled off pain. (It’s a pun).
http://www.telegraph.co.uk/news/picturegalleries/picturesoftheday/8997144/Pictures-of-the-day-6-January-2012.html?image=9
Show a trend line for the stronger ones from 1950 on. It’s easy to see for the commenters here, but the general population – if not the media – might benefit a bit if it’s spelled out for them.
Wow! Imagine if they had the same current methods of tornado detection during the 1960 to 1975 time frame! It must have been hell on earth!! How did we manage to survive with all that ignorance??!!
I recall a previous argument here at WUWT that claimed that mild global warming might reduce the incidence of tropical storms, as it would reduce the conditions of temperature and pressure gradients that produce the storms, at least over the oceans.
Are there any meteorologists out there willing to make the same argument for tornados? The trend line in the data above seems to raise the possibility.
I’m sure someone will bemoan the statistic, based on the top graph, that 2011 saw the second highest number of strong tornadoes in the last 61 years of record-keeping, and that increasingly severe extreme weather events are consistent with a warming climate.
Great post – interesting read.
One potential correction: based on the updated 2011 US Tornado Count (inserted by Anthony) the F0-F5 chart by decade should also be noted to reflect the 6 F5s from 2011.
Researchers at the SPC say prelim results from current research show (extremely simplified here):
Warming = stability = fewer tornados;
Cooling = instability = more tornados.
“Yet more Windmills destroyed…..by the wind.”
“Whirled off pain. (It’s a pun).”
==========================================
If the same people that are pushing eco/green/windmills….
…are telling us to expect more damaging winds
Thanks Jon C and sorry for any confusion, but all the figures and graphs are Annual Averages, not Decadal Totals. I perhaps should have made this clearer.
So, for instance , in the “2000’s”, i.e. 2000-2011, there have been 8 F5 tornadoes including 6 in 2011. Therefore the average over the 12 years is 0.66 p.a (that I have rounded up to 1). The 1990’s saw 10 in total and the 1980’s saw 5, so I guess there is not much of a trend here either.
Other noted researchers have found similar results for the claimed increased in tornado activity.
This from: Verbout, S. M., H. E. Brooks, L. M. Leslie, D. M. Schultz, 2006: Evolution of the U.S. tornado database: 1954–2003. Wea. Forecasting, 21, 86–93.
http://journals.ametsoc.org/doi/full/10.1175/WAF910.1?prevSearch=%5BContrib%3A+verbout%5D&searchHistoryKey=
“Over the past half century, the number of tornadoes reported in the United States has doubled from roughly 600 per year in the 1950s to around 1200 in the 2000s (Fig. 1). The 1374 reported tornadoes in 2003 were second only to the 1426 reported in 1998. The changes are not likely due to meteorological causes alone. Report discrepancies, public awareness, Doppler radar, and National Weather Service vigilance all have contributed to the increasing trend. In addition to the general increase in the annual number of tornado reports, the tornado record possesses interannual variability, too. In the late 1980s, a relative deficit of tornadoes occurred (Fig. 1).
Despite the increase in the number of reported tornadoes, the numbers of F1 and greater tornadoes has remained fairly consistent over the 50 yr at around 500 reports per year (Fig. 1). Brooks and Doswell (2001) suggested that stronger tornadoes have been reported more consistently over time. Therefore, nearly all the doubling of tornado reports over the last 50 yr is most likely due to the increased reporting of F0 tornadoes. “
The situation in Canada is similar.
This from: Hage, K. D., 2003: On destructive Canadian prairie windstorms and severe winters. Natural Hazards, 29, 207-228.
http://www.springerlink.com/content/rw766847h733145r/
“All storm frequencies exhibited an apparent maximum between about 1910 and
1940. Tornado counts rose appreciably after about 1970 in contrast to rather steady
frequencies of other destructive windstorms. The recent rise in tornado counts
was discovered in the United States in the late 1970s (Fujita, 1978; Forbes and
Bluestein, 2001) and high counts have continued there in recent years (Doswell,
Moller and Brooks, 1999). The combined provincial data are shown in Figures 3
and 4 which also illustrate the frequencies of intense storms. The recent increase in
observed tornado counts occurred only in the frequency of weak tornadoes as will
be seen by a comparison of Figures 1 and 2 with Figures 3 and 4. Several factors
contributed to improved identification of weak tornadoes after about 1960. In the
1960s and 1970s hail report cards mailed in by farmers to the Alberta Hail Research
Project were scanned for windstorm information that was then added to data from
other sources. Letters to the editors of scores of weekly newspapers were published
in the early 1970s resulting in reports of weak tornadoes that were not reported
elsewhere. In 1980 networks of hundreds of voluntary severe weather observers
were established by Environment Canada in Alberta and Saskatchewan. After the
devastating Edmonton tornado of July 31, 1987, public awareness and reporting
of tornadoes was increased in response to Environment Canada press releases and
other media announcements at the start of the tornado season each year. Because of
these changes in sampling procedures after about 1960 the long-term time trends
in the frequencies of weak tornadoes and all tornadoes are not reliable.”
kwinterkorn says:
January 6, 2012 at 7:36 am
“I recall a previous argument here at WUWT that claimed that mild global warming might reduce the incidence of tropical storms, as it would reduce the conditions of temperature and pressure gradients that produce the storms, at least over the oceans.
Are there any meteorologists out there willing to make the same argument for tornados? The trend line in the data above seems to raise the possibility.”
I will jump in on that. What you say is indeed a possibility. Tornadoes owe their existence to strong wind shear. The shear is in turn a product of the strength of the temperature gradient, or baroclinic zone as meteorologists call it. One of the theories of global warming is to increase the temperatures more at high latitudes as compared to the lower latitudes, ie. the poles warm more that the tropics. This would have the effect of decreasing the temperature contrast between the higher and lower latitudes. The decreased strength of the baroclinic zone would in turn decrease the wind shear necessary for strong tornado development. This would imply that, on average, one could expect fewer or less intense (or both) tornado activity.
Alpha Tango says:
January 6, 2012 at 7:07 am
Actually, your attempt to make light of a serious situation underscores how bad things really are: we now have hundreds of serious tornadoes “in the pipeline”*, which can suddenly burst forth, like a tidal wave of gophers destroying a golf course maintained by Bill Murray, killing 120-150% of the population of the United States (though mostly low income single mothers and minorities) overnight.
*I believe this pipeline may be laid in the deep oceans and could be very hot, but is likely undetectable with modern methods.
It’s tough to top ’74.
I think this year’s tornadoes have been taken out of context.
Nick Shaw says:
January 6, 2012 at 7:32 am
. . . survive . . .
Many didn’t Nick. You are the only “ignorance” here. And you owe an apology to those readers that have lost family and friends to these storms.
{bad puns alert}
“More tornadoes” alarmism is just spin. All Hail AGW!!