We are causing more hurricanes: by stopping polluting

This is interesting don\’t you think?

Researchers from the Met Office established a direct link between levels of industrial pollution and the frequency of hurricanes in the North Atlantic.

For much of the 20th century, sooty pollution in the atmosphere has made conditions unfavourable for the storms, causing their numbers to drop.

But since the 1980s, cleaner air over the Atlantic has created better conditions for hurricane formation, with more tropical storms developing and battering American and Caribbean coasts as a result.

This rather over turns the current mantra: that hurricanes are Gaia\’s rage at the carbon in the atmosphere. They\’re actually Gaia\’s coughing fit after we stop polluting said atmosphere.

So, all those who said that we must stop polluting because of the risks of more hurricanes should now be saying we should pollute more because of the risks of hurricanes, no?

12 thoughts on “We are causing more hurricanes: by stopping polluting”

  1. So Much For Subtlety

    The hottest years on record were in ther 1930s (using the partial record of the US alone) and in the 1990s. That is to say, after the collapse of the Soviet Union and around the Great Depression.

    What they both have in common is a collapse of coal burning.

    I think it is entirely possible that “global warming” is simply a return to the long-term average caused by a decline in the production of and distribution of sulphates.

    But we don’t know and as long as the warmists are controling journals to restrict publication we probably never will.

  2. It’ll be a day for ice-skating in Hades before the warmistas admit they might be wrong!

    Besides, ‘climate change’ isn’t the main focus of their aim – it’s just the hook they hang it on.

    Their aim is control.

  3. “established a direct link “: unfortunately the link to the article on that link won’t work from my computer at the mo’. Your problem or mine, Tim?

  4. (Link now works.) “Met Office scientist … model results” are a combination to which I would not myself attribute the verb “establish”.

  5. According to the telly yesterday the sun spot record shows we could be heading for another Little ice Age ,like when ,for instance, the Thames froze over regularly. Man-made global warming activities might then not be out of place. Because of the intense cold we would be burning things we should n’t anyway.

  6. “According to the telly yesterday the sun spot record shows we could be heading for another Little ice Age ,like when ,for instance, the Thames froze over regularly”
    Oh gawd! This is where i came in & we didn’t even get the Pathe news or intermission for an ice cream. Next global warning panic’ll be coming around in 2035 then?

  7. Here’s a fun thing for any with a statistical bent: grab a table of hurricane frequencies year by year. I found a series of files for the North Atlantic from 1948 – 2007 and processed them into a time-series table thus {6, 7, 11, 8, 6, 6, 8, 9, 4, 3, 7, 7, 4, 8, 3, 7, 6, 4, 7, 6, 4, 12, 5, 6, 3, 4, 4, 6, 6, 5, 5, 5, 9, 7, 2, 3, 5, 7, 4, 3, 5, 7, 8, 4, 4, 4, 3, 11, 9, 3, 10, 8, 8, 9, 4, 7, 9, 15, 5, 6}

    Fitting a linear trend to this gives f(x) = 5.82994 + 0.0115866 x, which is not much of a trend. Mean of this series is 371/50 ~ 6.18 with a standard deviation of 2.57 – much, much bigger than any purported ‘trend’. We can bin these frequencies into a histogram.

    Now form a Poisson process P(k,λ) = λ^k exp(-lambda;)/k! and overlay it on the histogram. Hey presto! The fit is striking. What this tells me is that a) there is no statistically significant trend towards more frequent hurricanes b) frequency is well characterised by a Poisson (i.e. random) process and the hurricane predictors are talking out of their arses.

  8. Whoops, now that the content encoding bug has been fixed, next thing is a preview button. Baby steps. Anyway, λ = 371/60, P(k,λ) = λ^k exp(-λ)/k! (scaled by a factor of 60 to normalise it to the histogram, natch.) P gives the likelihood of observing k events in a given time period (here, one year) given an expected number of events in that period (the ‘intensity’, λ) assuming that events occur independently of one another i.e. that the process is memoryless. If your data fits a Poisson process then it’s a pretty good indication that it’s random.

  9. Ooh, that’s new, thank you Mr R. I think I had a JavaScript thingy that was nobbling it before. Let’s see if can do HTML entities. E = mc². Jag gillar svenska köttbullar. N?G ? ?n ? N, ?g ? G, gng?¹ ? N

    If I have time (unlikely), I’ll run a χ² test on that hurricane data to see if a Poisson process can be rejected as the null hypothesis.

Leave a Reply

Your email address will not be published. Required fields are marked *