One last hurrah on my end for the evening before I head to bed...
Introduction -- Frances continues to have an excellent satellite interpretation, with a well-defined CDO and excellent outflow on the west side of the storm. It is somewhat losing the outflow channel with the ULL around 23 N/45 W, but outflow remains pretty good on the east side of the storm. We're currently in the satellite eclipse for the night, though I don't expect much to have changed by the time we are out of the eclipse period. The upper low to the west of the storm near 24 N/79 W is weakening to some degree, but remains a distinct entity. The steering ridge remains in place both near and ahead of Frances. Recon is enroute and we should have a vortex message/center fix in the next hour or so.
Track Reasoning -- Frances remains on a just slightly north of due west track, a track that should continue for the forseeable future. I expect the storm to make landfall in the 84-96hr time frame (i.e. around Saturday) in between West Palm Beach and Miami.
As mentioned above, the ridge is firmly in place north of Frances, guiding the storm along at a pretty fast clip. A weaker ridge center is located over southern Georgia -- think of it as an extension of the ridge. The upper low to Frances' west is retrograding, but slowly. I think it's almost done retrograding, to tell the truth. The trough in the east central U.S. is becoming elongated east-west (partially as Gaston races out to sea) and slowing it's forward progression as a low forms in the Mississippi region. Another low is currently over western Texas and moving towards the east, but I don't think it is going to be a big factor in this case.
Frances should continue moving west or west-northwest ward in the short-term. There may be some slight jogs to the noth, particularly in the 24-48hr time frame, but the general west-northwestward motion should resume thereafter. I'm not sure I'm going to buy into the slowing at 96hr or so, but I do see why the models are doing what they are doing (and for what they forecast to evolve, it's a very valid response. I just don't agree with that evolution). The ridge is very strong and I don't see anything out there to really kick it. If anything, Gaston exiting is going to make it slightly stronger, while any energy from Canada is currently getting shunted eastward as well, not allowing the trough to dig any further. I don't see much of a reason for this to change, not until the storm makes landfall.
Using the Univ. of Wisconsin deep-layer flow analyses, one can see this. With such a strong storm, a deeper layer flow regime is going to steer the storm; with a weaker storm, I could see a turn, but that'd be a cat 2/3 storm or lower. The current deep-layer regime supports a more westerly - and faster - track through Florida. I favor the 850-250 and 700-200 mb analyses, both of which support this argument, and I don't see much of a reason for this to change over the next few days.
With that established, what does the storm do after landfall? The ridge may begin to erode with time, but should remain around in some manner. As the storm begins to weaken over land, it may get pulled slightly further to the north as well by the mean flow. (Note: the NHC considers the 1000-100mb mean flow to be a fairly good indicator of storm motion, according to a lecture by Richard Pasch. I don't think the entire atmosphere is a good indicator, though, just a subsection.) The storm should re-enter the Gulf at some point, give or take a few (+/- 15-25mi) miles of Tampa Bay. Everywhere from Biloxi to Apalachicola needs to watch for a recurving storm in the 5-6 day time frame; I'm trending towards Panama City, but that's more of a hunch right now.
Models -- Yes, this track is to the left of most of the models. If I've learned anything in classes and research, it's that the models are usually wrong to the left. I just don't think they are initializing these features well, and I'd like to see what they were able to do if all eastern and central U.S. NWS officers were launching 6hrly baloons for upper-air observations. Really, I can see why they are doing what they are doing, save for the extreme outliers, but I'm not in agreement at this time with those cases. If I'm off on the track forecast above, however, it will almost assuredly be to the right. It will be a really close call on all of the features, however. Note that through Frances' lifetime, the GFS has had the smallest track errors out to 120hr (the FSU Superensemble has a major advantage through 96hr over *all* models, though it falls slightly behind the GFS at 108 and 120hr). As the synoptic patterns and track come more under the influence of the midlatitudes, the global models should still do fairly well -- but 5 day track errors are still 150+ miles. Bottom line -- the models aren't out to lunch at all, but as with any guidance, use it as just that and add in your own analysis of the situation.
Intensity -- Frances remains a healthy cat 4 hurricane. Wind shear is low -- not negligible, but around 10m/s, more than sufficient for maintenance of the storm and some strengthening. (Aside -- useful for diagnosing shear are the UWisc/CIMSS wind shear analyses. You can get it by doing a google search or by going to the PSU/FSU Tropical Cyclone Model page.) The storm has about a day to make it to strengthen further, then a combination of an eyewall replacement cycle and slightly increased shear should level the intensity off for awhile. There will be another window for intensity increases closer to landfall, around 60-72hr, as waters in the Bahamas and off of Florida are very warm and shear is forecast to be very low. It's not inconcieveable to see Frances strengthen to about 150mph in the short-term, level off around 140-150 for a day or so, and then strengthen again to borderline 4/5 status as it nears landfall.
Further strengthening will likely be slightly inhibited by interactions with the Bahamas, albeit not to any large degree. There's also the chance for solely maintaining the storm until landfall...and the further north the storm makes landfall, the weaker it is likely to be. If the storm nears Cape Canaveral, it's likely to be a low end 4; Jacksonville, a 3; and further north, borderline 3. Bottom line -- a very dangerous hurricane is going to make landfall somewhere in the next few days. It should weaken some overland and only strengthen slightly if it makes it out over the Gulf before a second landfall. The northern Gulf coast should expect a cat 1/2 storm if this scenario unfolds.
Closing -- I'll add in this disclaimer: remember, the storm is a big storm, not a point. The entire Florida peninsula and panhandle, except for extreme SW Florida and the Keys, are likely to feel some impact of this storm. Regions further north in Georgia and the Carolinas are by no means out of the woods yet. This storm does not have very good guidance consensus yet, though I expect this to change some tomorrow. Everyone up and down the SE US and northern Gulf coast needs to follow this storm over the coming days -- no one is out of the woods yet. Tomorrow will be a pivotal day. Get some rest everyone, we'll see where this thing is going, how fast, and how strong in the morning.
Current Tropical Model Output Plots
(or view them on the main page for any active Atlantic storms!)