03 December 2012

Record US Intense Hurricane Drought Continues

The graph above provides an update to data on the remarkable ongoing US "intense hurricane drought." When the Atlantic hurricane season starts next June 1, it will have been 2,777 days since the last time an intense (that is a Category 3, 4 or 5) hurricane made landfall along the US coast (Wilma in 2005). Such a prolonged period without an intense hurricane landfall has not been observed since 1900.

Some thoughts:

Even with hurricane Sandy and its wide impacts, things will indeed get worse. The US coastlines as a whole have actually been very lucky with respect to hurricanes since 2005, with aggregate damage (even including aggressive estimates for Sandy) 2006-2012 falling at or below the historical average. Sandy made landfall as a post-tropical cyclone of hurricane strength -- a phenomena that has only been documented 3 times since 1900 (1904, 1924, 1925 -- later this week I'll have a post on Sandy damage estimates).

The long-term intense hurricane drought means that a mere "regression to the mean" will see more hurricane landfalls and considerably higher damage in the years to come. The fashionable talk these days of a "new normal" is of course utter bullsh*t. Just wait until we return to the "old normal" -- I know that it may be hard to believe, but both hurricane damage and climate hype are set to increase dramatically in the years to come.


  1. Roger I like your nice guy truth to science approach, but let's be honest it is no match for the cimate change PR machine amassed by NGOs and pushing the notion that man has caused more hurricanes.

    The only new normal I see is for climate blame irrationalism.

  2. Meanwhile, reasonable efforts to mitigate known risks and address common consequences are not forthcoming. If I was cynical, I would describe this as a premeditated enterprise to foment drama, but I will refrain from speculating to the preferred outcome. Although, I will suggest the tactics associated with this cause bear a striking resemblance to those employed during the nutritional upheaval of the 80s and 90s, where we were advised by authorities to wander aimlessly from one extreme to another. It was a profitable time for the authorities and their clients. It was a confusing and unproductive time for their intended beneficiaries.

  3. I have a question, Professor. Why do experts, especially when presenting to the general public, insist on using isolated statistics such as "mean" and describe it as "normal"? There is no effort to inform or remind people that a stochastic process is modeled by a distribution, where normal is not a constant value. There is also no effort to inform or remind people why it is modeled as a stochastic and not deterministic process.

  4. "Meanwhile, reasonable efforts to mitigate known risks and address common consequences are not forthcoming."

    Indeed. Still no signficant discussion of designing and deploying a portable hurricane storm surge barrier that can protect any city on any coast on a few days' notice. Such a system could have prevented the flooding of New Orleans, and would have signicantly reduced the flooding damage of Superstorm Sandy.


  5. Roger,

    You say

    "The long-term intense hurricane drought means that a mere "regression to the mean" will see more hurricane landfalls and considerably higher damage in the years to come."

    I agree with your statment in the sense that the lucky streak is guaranteed to end eventually, and we will have one or more Cat. 3+ hurricanes strike the U.S. and cause huge damage.

    On the other hand, in the broader sense I don't agree that we should see more hurricane landfalls (not even intense ones) in the next few years or decades. The AMO, for example, reached its peak around 2005 and is now on its way down. Hurricanes should be less likely and less powerful during that phase, which should last a few more decades. I expect you'll be retired before we see a significant upswing in the overall number or strength.

  6. "...later this week I'll have a post on Sandy damage estimates."

    If you have any breakdown on damage from storm surge versus wind (and inland flooding, if there was any of that) I'd sure like to see it. I haven't found anything in that regard.

  7. What is the significance of showing just landfalls?

  8. That graph sucks. The bottom axis is ridiculously badly labeled. It's not clear wtf the numbers on the bottom actually mean. I assume they represent the average days between landfall, but that's not clear at all. I'm not doubting the content, I'm just noting the presentation is bad.

    I assume the actual marks, however, represent five-count years since 1900? This is what is normally presented in that location, while the numbers you're providing should be overdrawn on the graph itself... or something other than what you're doing.

  9. }}} Such a system could have prevented the flooding of New Orleans, and would have signicantly reduced the flooding damage of Superstorm Sandy.

    New Orleans was screwed, and this fact was know a decade and more beforehand.

    There was an article in the early-mid 90s in American Heritage of Invention and Technology (a quarterly) that talked about the 1900s-era pumping stations in NO.

    They acked -- in the article -- that if a class 4 or above hurricane made landfall near NO that the pumping capacity would be overloaded.

    I'd been watching every class 3 and above hurricane since reading that article for it to happen, and several times in a row, NO dodged an airborne bullet. Finally, in 2005, what they acked a decade or more before would happen, did.

    So it wasn't at all a surprise to anyone who knew jack about the Big Easy.

  10. -8-OBloodyHell

    Thanks ... you are correct, the presentation could be better explained, and I have neglected to bring forward the explanation that accompanied the original posting of the graph.

    The X-axis shows landfalling storm count in chronological order starting in 1900 (with the 78th landfall occuring in 2005, we still await the 79th). The bars show days between successive storm landfalls. The red line shows the trend. Further questions welcomed.

    Thanks for the feedback!


  11. "Better land regulations may be able to decrease flood exposure, but they cannot do so in a significant manner – in the absence of a large-scale buys-out and house destruction program that appears extremely unlikely"

    But many houses have been destroyed by the destroyed by the storm. All that is required is for the state to condemn that land and preclude further development on it. What is nutty is for the state to pay for rebuilding in the same proven high risk location.

  12. It strikes me we're talking about two very different kinds of policy here, and I'm not sure one merits the name 'policy' at all.

    Clearly it is entirely in the power of the US or the states to enact legislation discouraging development on flood-prone areas, funding the erection of flood barriers, etc.. Those things can and should be done.

    On the other hand, I'm not sure what legislation the US could enact that would have more than a minimal effect on sea-level rise. We could certainly tax carbon fairly heavily; that would possibly reduce our own direct carbon dioxide emissions. It would not change the rate of carbon emission elsewhere; in fact, by depressing the market price for fossil fuels, it might well lead to increased consumption elsewhere.

    That isn't policy. That's simply meaningless and rather expensive symbolism.

  13. As someone who has read a lot of online comments, I was very pleased to note the uncommon intellectual vigor of the comments posted here. I, too, was disturbed by the incompleteness of the graph, and I'm glad that Roger Pielke Jr replied to OBloodyHell's criticism of it. But there's another thing bothering me: identical type storms are called hurricanes in the Atlantic, typhoons in the north Pacific, and cyclones in the south Pacific and Indian oceans. I, therefore, would like to know the difference between a cyclone of hurricane strength ("Sandy") and a hurricane.

  14. Further question:

    Can you point us to the source data?

  15. -14-Unknown



  16. Regression to mean, yes, but I disagree that means "hurricane damage and climate hype are set to increase dramatically in the years to come." What we're like to see a relative change only, The average will return but not anything dramatic. Only the media will make it seem otherwise.

    Also, that assumes there's no real long-term climate change going on. If destructive hurricane landfalls are down, it MAY be just the normal swing around a stable mean. But it could also mean the mean is changing, meaning going down, and we may have many decades with fewer big hurricanes.

    Who was it that said, "prophecy is difficult, especially about the future." I think that applies here.

    --Michael W. Perry, My Nights with Leukemia: Caring for Children with Cancer