Colorado State’s inaugural two week hurricane forecast busting (spectacularly)
August 18, 2009
Back on August 5, Philip Klotzbach and William Gray issued their very first two week forecast of hurricane activity.which replaced their monthly forecasts from previous years. So far, the forecast has not gone well, and may end up being a spectacular bust. Rather than trying to predict the number of storms, they chose to predict the Accumulated Cyclone Energy for the forecast period August 6-20. Also, rather than predicting a particular number, they went categorical, stating that ACE would be at below average (less than 70% of climatological average), i.e. less than six. While the exact figure may vary slightly with how Ana and Claudette are ultimately recorded in the books, ACE for August 6-present is now higher than seven. Wouldn’t be that bad of an outcome for the forecast, except that there is still two days left in the forecast period and Bill is still out there just below major hurricane status. Just by staying at 95 knots for the next two days, Bill would double the ACE for the period causing it to be around 15. That puts it out of the ‘average’ level and into ‘above average’! It’s not clear if they are going to be doing verification (after-action reports) on their forecasts, but if they did, they may have a bone to pick with the forecast of the Madden-Julian Oscillation. While the forecast was for a phase of the MJO that would be unfavorable for tropical cyclone formation in the Atlantic, last week’s MJO discussion from the Climate Prediction Center featured the following statements:
Enhanced convection over the past week rapidly shifted eastward to the eastern Pacific and Atlantic Oceans, which is indicative of subseasonal variations…. Tropical cyclogenesis is favored across the eastern Pacific and east-central Atlantic.
And with that, bust went the CSU forecast.