From a Clemson University press release, another peer reviewed study refutes the “global warming to hurricane” linkage that supposedly is causing stronger storms.
The increasing frequency of storms in the last 50 years is to be expected, due to better reporting and improved technology like satellites, Hurricane Hunter planes, and Doppler Radar. NOAA agrees on the improved reporting issue in a study here.
This echoes what I reported on April 11th 2008 about Emanuel’s findings as well as what I reported on February 21st 2008 from Roger Pielke Jr. and Chris Landsea at the National Hurricane Center. On May 15th, 2008, Tom Knutson, a meteorologist with the National Oceanic and Atmospheric Administration’s fluid dynamics lab in Princeton, N.J. reversed his position on the issue in an AP story and argues “against the notion that we’ve already seen a really dramatic increase in Atlantic hurricane activity resulting from greenhouse warming.”
Plus, according to Florida State University’s Ryan Maue, Accumulated Cyclone Energy has hit a 30 year low. The Global Warming linkage simply isn’t there.
CLEMSON — In a new study, Clemson University researchers have concluded that the number of hurricanes and tropical storms in the Atlantic Basin is increasing, but there is no evidence that their individual strengths are any greater than storms of the past or that the chances of a U.S. strike are up.
Robert Lund, professor of mathematical sciences at Clemson, along with colleagues Michael Robbins and Colin Gallagher of Clemson and QiQi Lu of Mississippi State University, studied changes in the tropical cycle record in the North Atlantic between 1851 and 2008.
“This is a hot button in the argument for global warming,” said Lund. “Climatologists reporting to the U.S. Senate as recently as this summer testified to the exact opposite of what we find. Many researchers have maintained that warming waters of the Atlantic are increasing the strengths of these storms. We do not see evidence for this at all, however we do find that the number of storms has recently increased.”
The study represents one of the first rigorous statistical assessments of the issue with uncertainty margins calculated in. For example, Lund says “there is less than a one in 100,000 chance of seeing this many storms occur since 1965 if in truth changes are not taking place.”
He adds, “Hopefully such a rigorous assessment will clear up the controversy and the misinformation about what is truly happening with these storms.”
The study, submitted to the Journal of the American Statistical Association, also found changes in storm pattern records starting around 1935. This was expected at the onset of aircraft reconnaissance, which allowed record-keepers to identify and document storms occurring in the open ocean.
While the study did conclude that more storms are being documented, researchers found no evidence of recent increases in U.S. landfall strike probability of the strongest of hurricanes. Lund notes that “because these types of storms are so uncommon, it will take many more years of data to reliably assess this issue.”
Related Links
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
This can’t be correct. Mann and his foggy bottom bog boys determined something else from three sediment deposits. /sarc
http://wattsupwiththat.com/2009/08/13/mann-hockey-sticks-hurricanes-hurricanes-in-the-atlantic-are-more-frequent-than-at-any-time-in-the-last-1000-years/
Its beginning to make me sick ….. the blatant mis-utilisation of fancy scientific words and fancy statistics to make something out of NOTHING. How can anyone possibly make a graph that stretches back 150 years and make it look like the uncertainties at one end are the same as the uncertainities at the other. Of course there were less observed storms earlier … there was less observation. Its pathetic.
The great thing about comparing landfall events of major storms is that there is a genuinely good chance that all major storms that actually hit land were reported in some scientific manner or other over the last 150 years. And a comparison of THAT shows absolutely nothing in terms of any increasing trend. If anything, it shows that previous decades has more storms.
Its pathetic.
It is great that real statisticians are looking at the data. Now, if we could only have somebody who could write/speak simple declarative English, we might know what the hell was going on.
Yesterday I thought I might get sick. After reading this I did.
Uh oh. Jaxa shows Artic sea ice leveling off or on a slight downturn. Levels seem headed toward the 2007/2008 trend lines. Perhaps those sunspots are already having an effect. I guess the cooling period is over and it’s time to book spots on those Northwest passage cruise lines before all the polar bears drown.
Well, I took a look at the current stretch of silence in the Atlantic. It’s approaching a record for the last 30 years during the month of September.
http://wxtalk.wordpress.com/2009/09/25/tropical-atlantic-approaching-record-september-silence/
Can I have that in proper POHM England please?
Prisoner of her majesty???
‘For example, Lund says “there is less than a one in 100,000 chance of seeing this many storms occur since 1965 if in truth changes are not taking place.”
Radar (05:26:42) is partially correct. This statement will be used by Romm and his ilk to promote their view, but Radar is wrong in believing that the statement supports that view.
The statement actually supports both views, or none at all. It is practically meaningless. With or without humanity, the Earth is in constant flux. With or without humanity ‘changes’ would have taken place over the last 44 years and the number of storms would have fluctuated. By definition, we would call these changes ‘natural’. It is through the recognition of natural changes that the increase in Atlantic Hurricane activity in 1995 was correctly predicted almost a decade earlier. Nothing in the CO2 theory supports the step change in hurricane activity that we have observed. Natural changes explain it very nicely, along with the general trend since 1965.
Lund’s statement by itself says nothing about AGW, but Romm and Radar will spin it the way they want to. I think Lund said it that way just so they could spin it. Most AGW studies, showing no hard evidence of a human influence, phrase there conclusions that way.
Is the historic record of total hurricanes gathered on the same basis as that of today?
Can a good estimate of historical numbers be made?
Answer: GIGO
Many scientific papers fail this simple sanity check – this is yet another one.
When will real Climate science start replacing this pseudo-science rubbish?
Here’s part of the abstract:
This paper studies changepoint detection in time-ordered sequences of categorical data. When the data are sampled from a multinomial distribution, the proposed test statistic is the maximum of correlated Pearson chi-square statistics. This test statistic is linked to cumulative sum statistics and its null hypothesis asymptotic distribution is derived in terms of the supremum of squared Brownian bridges. The methods are used to identify changes in the tropical cyclone record in the North Atlantic Basin over the period 1851-2008.
Here’s my translation:
Since nobody really has wind speed data of massive storms prior to 1920 and only crappy incomplete data until 1950 we used highly confusing mathematics to translate the well known increase in detection into more severe storms. Send checks to…
It is unbelievable that they want to pass this crap off. The true answer is — WE DON’T KNOW!
Whatever,
Lucy Skywalker has an interesting post at tAV.
http://noconsensus.wordpress.com/2009/09/25/circling-yamal-delinquent-treering-records/
Studies have shown that the hurricane frequency/intensity is on a cyclical basis (approximately 60 years). It is closely related to the AMO.
See: http://www.appinsys.com/GlobalWarming/GW_4CE_Hurricanes.htm
Did anyone see that NOAA is looking to change their hurricane strength rating system? I wonder what this will do to the hurricane frequency vs strength statistic? http://www.nhc.noaa.gov/aboutsshs.shtml
This is junk science as there is no scientific way to make a comparison of hurricane frequency or intensity prior to satellites. It is sad when this type of analysis comes from a university when a bit of common sense would provide limits for selecting the appropriate data.
Well let’s consider the fact for this year Unisys is showing an El Nino wannabe (that means a non-nino) in the Pacific and the fact it’s been reported cold fronts are making it all the way down to Florida. It seems like hurricanes are having a hard time forming this year considering this even without a good El Nino to induce wind shear.
Also regarding Jaxa, one half day does not even make a short term trend and it takes time for the Sun to affect the Earth’s climate.
Also, I do note the AO index is strongly positive right now, looking at NOAA’s graphic of the polar vortex strength it seems like positive readings are often associated with a strong vortex. Intellicast’s forecast maps seem to predict below freezing temps. in more places in North America currently than in Siberia, cold air still possible in parts of South America and cool air masses washing over parts of Australia again. Also in North America they’re forecasting colder air across large sections of Canada with perhaps some of it spilling into the U.S.
For our area we get a warm up to the low 80’s in 2 days followed by a possible streak of temps. in the 60’s to start October with.
“”” “The study represents one of the first rigorous statistical assessments of the issue with uncertainty margins calculated in. For example, Lund says “there is less than a one in 100,000 chance of seeing this many storms occur since 1965 if in truth changes are not taking place.” “””
First of all; who says changes aren’t taking place ? Natural variability would be one example of changes taking place.
But that last sentence is a humdinger. Does that mean that if 100,000 research groups studied the number of storms occurring since 1965, that only one of the studies would show a (significant) difference in storm number from this study.
How do you apply statistics to an event that has only happened once in all of time; namely the sequence of storms that has occurred (or is believed to have occurred) since 1965.
Assuming that these mathematicians know how to count storms, one would presume that they got the number since 1965 correct; are they suggesting there is one chance in 100,000 that they counted them incorrectly.
Now I’m perfectly happy to let mathematicians perform statistical AlGorythms on any arbitrary set of numbers they want to concoct; such as the number of animals bigger than a flea per hectare of the earth’s surface.
I start to cringe when they start assigning some meaning to their results.
It is amusing to read the above comments. Edcon, I suggest you read before speaking. You made by bulletin board.
Point of facts are that as statisticians, we are not invested in the outcome. We don’t care what the end conclusions are….we just analyze the record as it is. And we know the laws of probability very well (so we cannot cheat).
In this case, the counts are clearly going up recently, but we see no changes in the strengths of the storms. Blame it on whatever you want, or spin it, but that is what we find.
If you will read the paper, you will find that the crux is about how to develop methods to rigorously assess the issue, not whether global warming is responsible for any changes that we see. It could very well be that the recent increase in counts is due to increases in very small storms that “barely make it into the record”, i.e., observational biases.
I have asked this question before, and gotten no reply, so I will ask again.
Through what possible mechanism can an el niño event raise the mean global temperature?
Thank you for coming by Dr. Lund.
People, please treat Dr. Lund with respect and courtesy.
In other words, “DON’T MAKE ME TURN THIS BOARD AROUND!”
enduser (11:40:38) :
“Through what possible mechanism can an el niño event raise the mean global temperature?”
At the beginning, the tropical trade winds blowing out of the east near the equator stop, so there is no wind friction to cause upwelling of cold water off the coast of South America and to blow the warm surface water westward. So the surface water in the eastern tropical Pacific stagnates and warms which shows up as an increase in global temperature. Strong trade winds have the opposite effect, a “La Nina”.
“Patrick Davis (04:36:45) :
Homo praesumitur bonus donec probetur malus?
Can I have that in proper POHM England please?”
My taylor is rich.
re: Jim Clarke
I’m no warmist. My point was to question if wuwt (and those of us who frequent wuwt) should be trumpeting one conclusion of this paper, (no increase in intensity) and rejecting another (increased frequency). Seemed sort of like cherry picking to me.
Dr. Lund,
Your quote from the article “we do find that the number of storms has recently increased.”
Is completely different from,
“it could very well be the recent increase in counts is due to increases in very small storms that barely make it into the record”
Having read your reasoned comment above, I now take the original quote to have meant “we do find that the number of [observed] storms has recently increased.”
But the original quote will be used by the loudest doomsayers of AGW.
“”It could very well be that the recent increase in counts is due to increases in very small storms that “barely make it into the record”, i.e., observational biases.””
There you go. Back in the sailing ship days they wouldn’t have noticed these.
Charles aka “board wielding” moderator:
Is WUWT going to hit twenty million hits today?
Reply: According to the sitemeter bug yes, but if you click on it and view stats, then no. ~ charles the board wielder of grumdorr
REPLY: the 20 million counter is from WordPress and is internal, which yields the greatest count accuracy. The sitemeter bug is external, with less accuracy, and I also started it a few months AFTER I started the WordPress blog, so it won’t line up, ever. I use it to see some details I can’t get from WordPress. – Anthony
Accurate statistical treatment done on inaccurate data results in an erroneous representation of reality. Due to historically poor observational ability you can’t use the early part of the record in a storm frequency count and say the increase has any validity in the real world. Satellites have changed the game with respect to all of the older non-landfall frequency counts.
Observations are recorded every six hours. The Atlantic database has 31 storms with 1 observation and 3 others with 4 observations. All those storms were recorded prior to 1871. Two thirds of those were found because they hit land or were sighted within 200 miles of land. 14 of those were full blown hurricanes when discovered. Gimme a break! There were two storms in 1990 each of which lasted 2 weeks that never came within 300 miles of land. Can there be much doubt that many storms in the early record weren’t recorded at all?
It was concluded the frequency of landfall storms hasn’t changed. Not very surprising since that is where humans predominately spend their time and likely have an accurate storm count. The vastness of the ocean has hidden a multitude of older data.
radar (12:16:49) : “My point was to question if wuwt (and those of us who frequent wuwt) should be trumpeting one conclusion of this paper, (no increase in intensity) and rejecting another (increased frequency). Seemed sort of like cherry picking to me.”
Here’s the thing. “Rejecting” has a specific scientific meaning as in “the hypothesis is rejected”. “Trumpeting” does not. However, one has to be ready to go where the evidence takes one self. The evidence that the paper cites for increasing frequency fails tests of “robustness” (like:
http://go2.wordpress.com/?id=725X1342&site=wattsupwiththat.wordpress.com&url=http%3A%2F%2Fwww.noaanews.noaa.gov%2Fstories2009%2F20090811_tropical.html )
And therefore is rightly rejected as a hypothesis. Now, I want to see the paper before buying their conclusion about storm intensity, however since I do not have a reason to yet, I cannot reject there hypothesis. And it sounds like they probably didn’t screw up in some major way.