Why is there such a big disconnect between the landfall model (chance of no landfall is going up) and the ACE model (chance of hyperactive season going up)?
The two models were derived independently from one another by choice-- really, for just this type of situation. A couple of points:
1. The ACE season is only about 5% over, while CONUS ACE season is 12% over. So there's been more time attrition (relatively) in the landfall model.
2. The model training set for U.S. ACE goes back to 1900, and the basin ACE set starts in 1950. So the landfall model perspective is a little longer, though the predictand is also noisier.
3. I think what the divergence is *probably* signaling is that the extreme warmth in the eastern Atlantic means rising odds of a busy season, though perhaps one in which TC activity preferentially forms in regions of the Atlantic which pose less of a U.S. landfall threat. If upper-level winds in the Gulf and Caribbean remain hostile, as they likely will, that shuts down an avenue of CONUS risks. More activity doesn't always mean more landfalls, as in 2010.
I'll be looking at this topic in more depth in the upcoming seasonal outlook update in a couple of weeks.
Why is there such a big disconnect between the landfall model (chance of no landfall is going up) and the ACE model (chance of hyperactive season going up)?
Good question.
The two models were derived independently from one another by choice-- really, for just this type of situation. A couple of points:
1. The ACE season is only about 5% over, while CONUS ACE season is 12% over. So there's been more time attrition (relatively) in the landfall model.
2. The model training set for U.S. ACE goes back to 1900, and the basin ACE set starts in 1950. So the landfall model perspective is a little longer, though the predictand is also noisier.
3. I think what the divergence is *probably* signaling is that the extreme warmth in the eastern Atlantic means rising odds of a busy season, though perhaps one in which TC activity preferentially forms in regions of the Atlantic which pose less of a U.S. landfall threat. If upper-level winds in the Gulf and Caribbean remain hostile, as they likely will, that shuts down an avenue of CONUS risks. More activity doesn't always mean more landfalls, as in 2010.
I'll be looking at this topic in more depth in the upcoming seasonal outlook update in a couple of weeks.