A BRAND-SPANKING-NEW IPCC Screw-UP, Hurricane Data Does Not Support Conclusions

IPCC head Rajendra Kumar Pachauri must be scared to get out of his bed in the morning. During the past few weeks, the reputation of his agency, has fallen off a cliff.  There was that humiliating retraction after it emerged  that statements about the melting of Himalayan glaciers were inaccurate. But that was only the start:

• Their claims about disappearing mountain ice was based on anecdotal evidence in a student’s dissertation and an article in a mountaineering magazine.

• The IPCC’s panel had wrongly reported that more than half of the Netherlands was below sea level because it had failed to check information supplied by a Dutch government agency.

• They mis-used a report about the shrinking rain forests that had nothing to do with climate and applied it to their cause.

•  The Times of London reported a bogus claim that global warming could cut rain-fed north African crop production by up to 50% by 2020.

And through it all researchers insist the errors are minor and do not impact on the overall conclusions about climate change. If that's the case, it should not concern them that another "minor" error has sprung up. This time it is their claims about global hurricane activity. An analysis of the IPCC's raw numbers shows that the claims that global hurricane activity has increased cannot be supported.

Les Hatton once fixed weather models at the Met Office. Having studied Maths at Cambridge, he completed his PhD as metereologist: his PhD was the study of tornadoes and waterspouts. He's a fellow of the Royal Meterological Society, currently teaches at the University of Kingston, and is well known in the software engineering community - his studies include critical systems analysis.

Hatton has released what he describes as an 'A-level' statistical analysis, which tests six IPCC statements against raw data from the National Oceanic and Atmospheric (NOAA) Administration. He's published all the raw data and invites criticism, but warns he is neither "a warmist nor a denialist", but a scientist.

Hatton performed a z-test statistical analysis of the period 1999-2009 against 1946-2009 to test the six conclusions. He also ran the data ending with what the IPCC had available in 2007. He found that North Atlantic hurricane activity increased significantly, but the increase was counterbalanced by diminished activity in the East Pacific, where hurricane-strength storms are 50 per cent more prevalent. The West Pacific showed no significant change. Overall, the declines balance the increases.

"When you average the number of storms and their strength, it almost exactly balances." This isn't indicative of an increase in atmospheric energy manifesting itself in storms.

Even the North Atlantic increase should be treated with caution, Hatton concludes, since the period contains one anomalous year of unusually high hurricane activity - 2005 - the year Al Gore used the Katrina tragedy to advance the case for the manmade global warming theory.

The IPCC does indeed conclude that "there is no clear trend in the annual numbers of tropical cyclones." If only the IPCC had stopped there. Yet it goes on to make more claims, and draw conclusions that the data doesn't support.

Thre IPCC's WG1 paper states: "There are also suggestions of increased intense tropical cyclone activity in some other regions where concerns over data quality are greater." Hatton points out the data quality is similar in each area.

The IPCC continues: "It is more likely than not (50%) that there has been some human contribution to the increases in hurricane intensity." But, as Hatton points out, that conclusion comes from computer climate models, not from the observational data, which show no increase.

To flesh that out a bit, the IPCC is saying there is no increase in hurricane activity, so we think that humans have contributed to the increase in hurricane activity. Got That? I don't blame you.

"The IPCC goes on to make statements that would never pass peer review," Hatton told us. A more scientifically useful conclusion would have been to ask why there was a disparity. "This differential behaviour to me is very interesting. If it's due to increased warming in one place, and decreased warming in the other - then that's interesting to me."


Hatton has thirty years of experience of getting scientific papers published, but describes this one, available on his personal website, as "unpublishable".


"It's an open invitation to tell me I'm wrong," he says. He was prompted to look more closely by the Climategate emails, and by his years of experience with computer modelling. All code and data on which policy conclusions are made should be open and freely downloadable, he says - preferably with open tools.


You can download both the paper and the code and tools from here.

This is not the first time that the hurricane data has been called into question.

The IPCC's AR4 chapter lead was Kevin Trenberth, who features prominently in the Climategate emails. In 2005, the National Hurricane Center's chief scientist Chris Landsea resigned his post in protest at the treatment of the subject by Trenberth.


"I personally cannot in good faith continue to contribute to a process that I view as both being motivated by pre-conceived agendas and being scientifically unsound. As the IPCC leadership has seen no wrong in Dr. Trenberth’s actions and have retained him as a Lead Author for the AR4, I have decided to no longer participate in the IPCC AR4."

Indeed !