move westward along the southern periphery of the Bermuda-Azores
ridge for the next couple of days, followed by a west-northwestward
motion on days 3-5 as the cyclone approaches a weakness in the ridge
located between Bermuda and the Bahamas. The global and regional
models are in general agreement on this track scenario through 72
hours, After that time, the models diverge significantly with the
GFS, Navy COAMPS, and GFDL models taking Karl farther west as a
weaker and more shallow cyclone, whereas the ECMWF, UKMET, and HWRF
models move Karl more northwestward as a stronger and deeper system.
For now, the official forecast track remains a compromise between
these extremes, close to the consensus model solutions, due to the
lack of a stable, consistent run-to-run forecast by the models.