< link rel="DCTERMS.isreplacedby" href="http://sirhumphreys.com" > Sir Humphrey's: Hurricane frequency in USA has declined over last 50 years


SITE MOVED:Sir Humphrey's has moved

Please join us at our new site: www.sirhumphreys.com.

The RSS feed for sirhumphreys.com is now here.

Thursday, September 01, 2005

Hurricane frequency in USA has declined over last 50 years

Must be more of that "Global Warming" eh? And consider this: its possible some hurricanes were missed back then, given lower population densities and inferior technology.

All credit to EU Rota.

Posted by Antarctic Lemur | 9/01/2005 03:14:00 pm


Blogger Murray said...

No fool!

Democrats were in office!!!

9/01/2005 03:34:00 pm  
Blogger Antarctic Lemur said...

It's all JFK's fault ?! Just like Cuba...

9/01/2005 03:39:00 pm  
Blogger Murray said...

You said it brother!

9/01/2005 03:57:00 pm  
Blogger Bernard Woolley said...

Ahem, I hate to pull you up on this AL, but the last bin is only 2001-2004. To be truely representative we have to get through to 2010 and then compare. Now, the US National Hurricane Center predictions are for a bad season in 2005. They are almost certain (95-100%) of an above normal season. They are predicting 9-11 tropical storms becoming hurricanes, and 3-5 becoming major hurricanes. In fact, this graph presents a contrary picture and suggests that we are entering a period that will spike higher that the 1951-1960 decade. I'm not saying it is caused by global warming, but one would have to call BS to anyone suggesting that the Atlantic Tropical Cyclone Basin are entering a quiet period. I would suggest that there is a much stronger linkage to the El Nino Southern Oscillation than global warming. Now, if the question qould then become what impact does global warming (if it is occuring) have on ENSO.

9/01/2005 04:04:00 pm  
Blogger Bernard Woolley said...

It would be interesting to see the bins broken up into 5 year intervals. I would posit that the last two 5 year periods would in fact show that the rate of activity is growing.

9/01/2005 04:06:00 pm  
Blogger Bernard Woolley said...

Actually, a 5y or 10y moving average would probably show any trends more clearly.

9/01/2005 04:22:00 pm  
Blogger ZenTiger said...

But what's the 500 year moving average? I seem to recall Noah complained about the rain a fair while ago. Unseasonal flooding caused by new fangled irrigation methods?

9/01/2005 04:39:00 pm  
Blogger Rob O'Neill said...

Factoid: 12 per cent of Americans believe Joan of Arc was Noah's wife.


Which proves not a lot really, just thought I'd mention it.

9/01/2005 05:01:00 pm  
Blogger Antarctic Lemur said...

Erm, Bernard you're joking right?

9/01/2005 05:07:00 pm  
Blogger reid said...

I think the relevant indicator is the ferocity and not the frequency? Weather extremes are occuring all over the place. Heat waves, storms, freezes, ...

Surely that's a bit of a warning in flaming letters 10 feet high, I would have thought.

Combine that with the threat of the gulf stream which moves the weather round the globe slowing or even stopping and we're in deep shit.

And even if scientists can't agree on a correlation between global warming and the weather, does that mean we should do nothing and just watch as the US, European, Japanese, and in particular the growing Chinese and Indian vehicle fleets pour millions more tonnes of poison p.a. into the atmosphere? WTF is wrong with our species that we (a) tolerate this without a murmur, and (b) that we accept the word of the vested interests who, of course, say there's nothing to worry about.

9/01/2005 06:49:00 pm  
Blogger Bernard Woolley said...

I've found the source data used for the graph above, and I'm going to have a look at it tonight sometime. And AL, yes I'm serious on this one. I'm not creating a linkage between the hurricanes and global warming, but I believe we are seeing a short term uptick in hurricane activity - which the NHC reported this season and has been warning about for the last few months. I'm not however picking the cause of this uptick.

9/01/2005 07:00:00 pm  
Blogger Chefen said...

"weather extremes" now are no more extreme than in recent history. The weather has been much warmer and colder in recent history. People take more notice of "extremes" now, which makes them more noticable, but "ferocity" and "frequency" are certainly not on an upward trend when known cycles are accounted for.

If scientists don't agree on it then it would be wiser to do nothing. Back in the 70s the fear was a new iceage, what if they had acted hastily then to counteract that given that now the globe is supposed to be warming? You must also account for the opportunity cost of whatever you propose doing, you can't ignore the foregone benefits of whatever could have been developed due to crippling the economy by arbitrarily reducing CO2.

You give yourself away with your (b). Vested interests lie just as much on the "save the planet with our plan" as they do on the other side

BTW, CO2 is not a poison.

9/01/2005 07:13:00 pm  
Blogger Bernard Woolley said...

Chefen, the main problem is just that there are more people, with more to lose. Thats why things are appearing worse. It we did not have population growth, had not clustered people into high-density urban areas, natural hazards wouldn't be claiming the attention that they do. Claiming it is global warming is nothing more than a cheap shot. It may be the fault of people, but only because of how we choose to live.

9/01/2005 07:29:00 pm  
Blogger Chefen said...

Yeah, I should have mentioned that too. Sometimes there is too much to say at once.

9/01/2005 07:36:00 pm  
Blogger Bernard Woolley said...

Right, I found the source data that was used for the graph above, and also found some stats and threw in 2005 up to and including Katrina.

(National Hurricane Center)
U.S. Hurricane Strikes by Decade (National Hurricane Center)
2005 Stats (Unisys Weather)

My resulting graph can be found here.

The reason I objected to the original chart was that it creates an arbitrary decade bin (xx01-xx10 etc) when that has absolutely no relevance to natural systems, and hence doesn't accurately display the data in a meaningful manner. Nature doesn't work to our silly abstractions of decades - she has her own cycles that we haven't even figured out yet.

So, I decided that the better way to display and analyse the data was to create some moving averages. In the end I decided on 3,5 and 11 year moving averages. I'm sure if you went out to a 21 or 31 year MA, you'd probably still manage to get it on a downtrend.

It shows that there does appear to be a short term uptick in activity since 2003.

My point? It's a bit like the share market, at this stage it could go further up or further down. We don't know. The current short term trend however appears to be up. The long term trend is open for debate - it could be trending down, or we could just be coming out of a trough and heading to higher activity. We won't truely know for another 20 or more years.

Put another way, if this was a chart of a shares moving averages, I'd consider (amongst other factors) investing in this business, because it appears to have some upwards potential.

9/01/2005 11:26:00 pm  
Blogger Chefen said...

The graph seems to show some cycles, a Fourier spectrum would be interesting. Of course the weather is driven by physical phenomena so solar cycles and the like tend to show up readily, unlike the sharemarket which is subject to much more stochastic effects. I think the point is that hurricanes have been in a low patch for a few decades and look to be on the upswing again, although it is sketchy and there seems to be a way to go before the highs of the 1880s are reached. Alternatively the industrial age seems to have reduced hurricanes.

9/01/2005 11:37:00 pm  
Blogger Antarctic Lemur said...

Yes the binning criticism is correct, but then applying any regression curve to such data is scientifically incorrect, as you'd be implying links between discrete and probably unrelated events.

i.e to believe such a graph, you'd have to already think the isolated events displayed were controlled by some greater property or event.


I'm going to email someone who studied such weather phenomenon relative to cimate cycles and see what she thinks.

One of our commenters knows advanced stats. Pity he doesn't appear to be reading today.

9/02/2005 12:04:00 am  
Blogger Bernard Woolley said...

It would be interesting to put the figures up against ENSO and solar activity - those are the two most significant factors that I could think of that would be influencing hurricane generation. You'd also have to compare with other basins - asia and pacifc to see if similar results were seen there. Arrggh. Too much for my head this time of night. Off to bed.

9/02/2005 12:25:00 am  
Blogger Chefen said...

I can guarantee that those sort of studies exist. Where to find them though? John Ray might have something among all his info...

9/02/2005 12:31:00 am  
Blogger Antarctic Lemur said...

The person who does ENSO work is a lecturer at SGES (Auck Uni).

9/02/2005 01:10:00 am  
Blogger reid said...

Fox News story: New research suggests global warming may be making hurricanes more ferocious.

Cause it's short, I've copied the story. The study itself is in the Nature journal, but it looks like you need to pay for it, I haven't read it. Luckily for me however, I note that the Fox story backs up exactly the point I was making above. Tee hee.

"A study by an MIT climatologist finds that major storms in both the Atlantic and the Pacific since the 1970s have increased in duration and intensity by about 50 percent.

These trends are closely linked to increases in the average temperatures of the ocean surface, and they correspond to increases in global average atmospheric temperatures during the same period.

Scientists call the findings both surprising and "alarming," because they suggest global warming is influencing storms now — rather than in the distant future.

However, the research doesn't suggest global warming is generating more hurricanes and typhoons.

9/02/2005 01:56:00 pm  

Post a Comment

<< Home