| Forecast for August 21 - September 4 "We expect that the next fifteen days will be characterized by heightened amounts of activity relative to climatology. These new 15-day forecasts have replaced the monthly forecasts that we have been issuing in recent years. (as of 20 August 2009)" I respect that fact that they gave themselves a poor evaluation for their forecast for the last 15 days: "VERIFICATION OF AUGUST 6 - AUGUST 20, 2009 FORECAST The 15-day forecast of tropical cyclone activity between August 6 - August 20 did not verify well. Activity at below-average levels was predicted, while observed activity was at above-average levels. Table A displays predicted and observed activity over the 15-day period. The primary reason why we believe activity was heightened during the 15-day period was due to the fact that the Madden-Julian Oscillation (MJO) increased in amplitude and intensified over the Indian Ocean towards the end of the period (Figure A). This MJO state is typically favorable for tropical cyclone activity in the Atlantic. The intensification of the MJO and development in the Indian Ocean was hinted at by a few GFS ensemble members. We based our 15-day forecast more on extrapolation of the MJO (e.g., coherent eastward propagation from Phase 6 to Phase 7) than on the GFS ensemble, and therefore we under-forecast activity considerably." |