The Intersection

Hurricanes can vary dramatically in size, and it’s my understanding that there is not any meaningful correlation between storm size and storm strength. For instance, last week’s Category 4 Cyclone Favio, which caused serious damage to Mozambique, was a relatively small storm, as can be seen in this image (with Favio located in the southern Mozambique Channel):


But now look at the latest Meteo France satellite image of the Southwest Indian Ocean, showing the same area as before, but with a huge Cyclone Gamede (945 millibar central pressure, a strong Category 2) in the middle of it. Technically, this storm is still at sea; but practically speaking, because it’s so big, its various rainbands are already affecting Mauritius, La Reunion, and mainland Madagascar:

i-ebb34e0c132f6ae70b0a4dc46afac581-S Indian Feb24.jpg

As you can see by comparing the two images above, Gamede is massively larger than Favio was. Just eyeballing it suggests to me that several Favios could fit comfortably within Gamede’s circulation.

In an earlier post, I discussed how our Saffir-Simpson scale for rating hurricanes focuses on peak intensity rather than a storm’s duration at a strong intensity, and how this can be misleading. On a similar note, it goes without saying that a large storm can cause damage over a much broader area than a small one. But once again, traditional Saffir-Simpson categories do not take storm size into account.

In this way, too, our current categorization system misses something really crucial about hurricanes. In my opinion, simply calling the gigantic Gamede a “Category 2” fails to capture its awe inspiring nature….


  1. #1 Jonathan Vause
    February 25, 2007

    Maybe worth adding that most fatalities during tropical storms are a result of the storm surge, and that large storms tend to cause greater surges because they move water over a larger area.

  2. #2 derek
    February 25, 2007

    Category is not and should not be a classification of a storm. It’s the classification of the damaging effect of the wind at a particular place and time. The sensational tendency to identify the highest category wind to be found anywhere in a storm (usually near the center) and to call the storm “a Category X storm” on the basis of that, is why storms fluctuate in category so fast, and why, as you say, it doesn’t really capture the size of it.

    Modern computer graphics should now be up to the task of classifying whole storms by the square kilometers currently being battered by winds of each category, and at the end of the storm’s life, the land area affected by a peak wind of each category over the entire track. The result in both cases would be a histogram of area against category, which sadly is not simple enough for news media. Perhaps a new system could be devised that takes that histogram and boils it down to a single number, or two numbers.

    Then again, there’s the sea surge, which isn’t accounted for at all in this system of wind damage.

  3. #3 Chris Mooney
    February 25, 2007

    Jonathan, Derek,
    Good points. As far as the surge goes, Katrina is the poster child for the relationship with storm size:

    “The massive storm surge produced by Katrina, even though it had weakened from Category 5 intensity the previous day to Category 3 at landfall in Louisiana, can be generally explained by the huge size of the storm.”

  4. #4 llewelly
    February 25, 2007

    The ATCF track file for a storm will typically contain estimates of the maximum extents of 64+ kt, 50+ kt, and 34+ kt winds, in each quadrant. Hit the link, count to the 11th column, which will be 0, 34, 50, or 64 . 0 means there is no estimate. 34 means columns 13-16 contain estimates for the max extent of 34+ kt winds in each quadrant. 50 means columns 13-16 contain estimates for the max extent of 50+ kt winds in each quadrant. 64 means columns 13-16 contain estimates for the max extent of 64+ kt winds in each quadrant.

    Unfortunately, I don’t have a link to graphic. However, if you look at the graphical forecasts the JTWC provides for Gamede and Humba you can see the circles for the forecasts (the ATCF file contains real time estimates) of 34+ kt, 50+ kt, and 64+ kt winds.

    This gives some notion of the size and the distribution of a storm’s winds, but I get the impression there is also much variation in the distribution winds above 64 kts, which is not well monitored for storms for which there is no aerial recon.

  5. #5 Chris Mooney
    February 25, 2007

    Gamede has claimed its first victims…a person, and a 520 meter bridge (!)…the latter may have something to do with a large surge associated with the storm size.

  6. #6 llewelly
    February 25, 2007

    Oh, and since we’re talking about Katrina and storm surge anyway, here is a public advisory for Katrina which contains a forecast of storm surge:

    Coastal storm surge flooding of 18 to 22 feet above normal tide
    levels…locally as high as 28 feet along with large and dangerous
    battering waves…can be expected near and to the east of where the
    center makes landfall.

    (De-capitalizatiion mine)
    Just another item for those of us who can’t get over claims akin to ‘nobody could have predicted’ the Katrina disaster.

New comments have been disabled.