Forecasting hurricanes

Please consider donating to Behind the Black, by giving either a one-time contribution or a regular subscription, as outlined in the tip jar to the right or below. Your support will allow me to continue covering science and culture as I have for the past twenty years, independent and free from any outside influence.

NOAA today announced its prediction for the upcoming Atlantic Ocean hurricane season, calling for between 9 and 15 tropical storms in 2012, with 4 to 8 becoming full blown hurricanes. The NOAA release can be seen here.

To me, the range of the prediction is so wide it really doesn’t mean anything. Moreover, I wonder about the reliability of these predictions.

For the last few years NOAA has consistently expected more hurricanes than have actually shown up. For example, last year NOAA predicted between 6 to 10 storms, well above the 1944-2005 average. The final number was 6, exactly what the average has been.

More worrisome to me is that, though they overestimated the number of actual hurricanes, last year they somehow got the number of named storms right. Named storms are tropical storms that become strong enough to merit a name. Some strengthen into hurricanes. NOAA had predicted there would be 12 to 18 named storms. In the end, they named 18, right at the top end of their prediction.

I wonder if their naming process was fudged to get them the numbers they wanted. While it might be possible to do that with the naming process of tropical storms, it is far more difficult to fudge the number of actual hurricanes.

My skeptical nature and the recent willingness in the climate field to fiddle with data probably makes me more suspicious than I should be. We shall see what transpires. Stay tuned. We shall return to this sometime early in 2013, after the hurricane season has ended.



  • Patrick Ritchie

    Wow that is quite suspicious. Assuming one were sufficiently motivated, how would you go about verifying this? My understanding is that a tropical storm must have sustained winds of at least 34 knots. Does anyone publish the raw data?

  • Chet Twarog

    Best to review their webpage info at We’ve already had two Atlantic tropical storms before the official start of the hurricane season. Tropical Storm Beryl impacted GA/FL/SC. It’s good to be skeptical but they have certainly improved their products over years–lots of variables to consider in such dynamic systems.

Leave a Reply

Your email address will not be published.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