|
Thursday, September 02, 2004 |
| I've been staring at the satellite loop since my last postand it's hard to tell what is actually going on with respect to Frances' movement because it appears that she is in the middle of an eyewall replacement cycle. Brendan Loy is not a meteorologist, but he must have slept in a Holiday Inn last night as he does a good job of showing and explaining what an eyewall replacement cycle is. I'm headed out for a bit, so there probably won't be a 8:00 PM update from me. Next update will probably be around 11 PM EDT. Weather comment []6:12:44 PM  trackback []
|
| AT 5 PM EDT, Hurricane Frances was near San Salvador Island in the Bahamas and 375 miles east-southeast of Florida and moving to the northwest at only 10 mph. Winds have fluctuated down to 140 mph and minimum central pressure rose to 946 mb. The forecasters believe the fluctuation in strength to be temporary due to eyewall restructuring (the alteration in course could also be a mirage caused by the eyewall restructuring). The current landfall forecast is for Vero Beach area on Saturday afternoon. After that the prognostication is for it to go northwest and enter the Gulf of Mexico north of Tampa on Sunday. We will have to see a couple of more frames from the satellite loop to be sure, but the 4:15 image seems to show a return to a west-northwest track after a few frames worth of moving to the northwest. As far as wind speed goes, remember that Hurricane Andrew's windspeed fluttered downward when it passed over the Bahamas, only to resurge in the hours before landfall. The Home Depot in Palm Beach appears to be a little bit busy. Also, television stations there seem to be on the verge of doing wall-wall coverage (soul sucking registration required). Hurricane Frances is forcing local television
stations to stockpile food, sandbag their doors, buy mattresses for
staff and even consider canceling the "Jerry Lewis MDA Telethon" and
President Bush's acceptance speech.
Sounds pretty early to start doing that. I would think people
would be totally exhausted of the coverage well before landfall
even happened. People need to be kept well informed of the
situation and what they should be doing, but this is a touch ridiculous.Officials at the three local news stations — WPTV Channel 5, WPEC Channel 12 and WPBF Channel 25 — were swamped Wednesday as they prepared for the storm to slam their viewing area, which includes the Treasure Coast. If the storm continues on its forecast track, at least two stations anticipate switching to wall-to-wall coverage — and dropping most if not all commercials — around midday today.Weather comment []5:03:38 PM trackback []
|
| Hurricane Frances will make landfall in Fort
Pierce about 8 p.m. Saturday, travel across the state and out into the
Gulf, then come ashore on the Panhandle around St. Marks about 8 a.m.
Monday as a tropical storm, according to a Florida State meteorology
professor T.S. Krishnamurti.
The season opener for Florida State football against Miami has been
postponed to Friday September 10. One wonders if his storm is
God's way of letting us know that we should not play college football
on Monday night.FSU meteorologist predicts Tallahassee area effects Weather comment []4:26:33 PM trackback []
|
| I'm kind of skittish about doing this, as this not quite the proper way to do it. The correct way is to look at each 6 hour forecast, not just ones in 24 hour increments, and keep a cumulative average of the forecast error for each period (i.e. 120 hour, 72 hour, etc). Doing that helps to smooth out some of the oddities that can pop up in any given forecast. But as this does give a touch of insight on how the forecasts have been doing, here goes: I took a look at the forecasts that have brought us to this point, going back five days in 24 hour increments. Here's what I found. The time that the forecast is for is 8 AM EDT today. At that time today, Frances was at 23.2N 73.5W. 5 days ago, the forecast was 23.0 N 73.0 W, an error of only 34 miles. 4 days ago, the forecast was 23.5 N 72.6 W, 55 miles error 3 days ago, the forecast was 23.5 N 73.5 W, 46 miles error 2 days ago, the forecast was 23.5 N 74.0 W, 34 miles error 1 day ago, the forecast was 24.0 N 74.0 W, 93 miles error. Note that while the 3 day forecast may seem good, it goes completely wrong after that as it had the storm going north, parallel to the Florida coast. I've looked back at the forecast and discussion from yesterday, and I can't find a reason why the error was suddenly so large. However, it is on the right track, it's just the timing that is off. --- Weatherblog has a nice graphic comparing the size of Frances and Charley. Am headed to my Atmospheric Dynamics now. Next post will be at approximately 5:00 PM EDT. Weather comment []1:49:46 PM trackback []
|
| Joined the congregation poring over the latest super-ensemble output. The models have finally converged very close to each other and project landfall in the vicinity of Fort Pierce and Vero Beach. They start to diverge in the later periods, but not in a totally unreasonable manner. Some more blog stuff:
To all those Florida bloggers out there: if you are reading this you
are procrastinating. Get your asses off those comfy chairs and away
from your keyboards and get them in gear. Frances looks pretty ominous and she is BIG. The whole state is gonna be under her spell. Seriously, get your asses going.
I've a humble request for South Florida bloggers, or anyone else who happens to be familiar with the Vero Beach area. It would be nice for someone to post a description of the area, i.e. what's the population density, what sort of homes are in the area and greater region, and anything else that the poster deems of interest. I'm very much a North Florida person, having ventured south on only a couple of occasions enroute to Key West, so I've no familiarity whatsoever with the area from the space coast to the treasure coast. Weather comment []12:59:12 PM trackback []
|
| At 11 AM EDT, dangerous Hurricane Frances was 55 miles southeast of San Salvador Island in the Bahamas and 450 miles east-southeast of Florida. Forward motion is to the west-northwest (295 degrees, specifically). Winds remain at 145 mph and mimimum central pressure is unchanged at 939 millibars (27.73") A hurricane warning is now in effect from Florida City to Flagler Beach. This means that hurricane conditions are expected in that area in the next 24-36 hours. The forecast models are finally coming a bit closer together. The current forecast is a little bit slower as it has landfall ocurring on Saturday morning a little bit north of Palm Beach. (following for meteorologists and other weather junkies) I walked by Dr Krishnamurti's office a little while ago. The 12Z super-ensemble projection is not up yet. Looking at the projections for the history of the storm, it is amazing how disjointed the models got on Tuesday and yesterday. As I've mentioned, the GFDL and GFS have been the worst offenders. For the history of the storm though, NOGAPS has had the worst long term track error, as it took Frances way too far north in the early days of the storm. However, it has had none of the issues that have plagued the GFDL and GFS in the past two day, and is now actually looking kind of good (although it may now be bringing the storm too far south and west). Anyhow, now is the time for people in south and central Florida to be getting ready. Don't procrastinate!!! Weather comment []11:09:28 AM trackback []
|
| At 8:00 AM EDT, Hurricane Frances was 80 miles southeast of San Salvador island in the Bahamas and 470 miles east southeast of Palm Beach. Forward motion remains to the west-northwest at 14 mph, winds remain at 145 mph, and minimum central pressure has blipped up to 939 mb (27.73") Hurricane Frances intermediate advisory number 34A (National Hurricane Center) Weather comment []8:01:52 AM trackback []
|
| At 5:00 AM EDT, Hurricane Frances was 35 miles north of Mayaguana Island in the Bahamas and 520 miles east southeast of Palm Beach. She is moving to the west-northwest at 13 mph . Winds remain 145 mph and minimum central pressure has fallen to 936 mb. A hurricane watch has been extended southward to Craig Key and covers the coast northward to Flagler Beach The forecast track is a little further south and takes the storm further west, but also as Frances hasn't shown much sign of slowing down, projected landfall has been moved up to Friday night/Saturday morning. The forecaster notes that feeding the models observation data from the area surrounding the storm didn't help the ones that have been incorrectly handling the ridge that is pushing the storm more west than north. The discussion notes that the storm could intensify a tiny bit more. Remember my fellow Floridians, that a hurricane is not a point on a map. Hurricane force winds extend outward 80 miles from the center of Frances and tropical storm force winds extend out 185 miles. People in south Florida are now looking at a two day timetable to prepare for the storm. If I were preparing (assuming that I were not under evacutation orders), today would be shopping day (for food, water, batteries, etc.) and initial preparation for the house (start bringing in or tying down anything that could get blown away by the wind), while tomorrow would be the final preps that I would want done by sunset. Weather comment []7:25:57 AM trackback []
|
| At 2:00 AM EDT, Frances was 35 miles northeast of Mayaguana Island in the Bahamas and 555 miles east-southeast of Palm Beach. Forward motion remains 14 mph to the west northwest. Winds have increased to 145 mph. and the minimum central pressure has dropped to 937 millibars (27.67"). Models continue to conflict with regards to track because of how they handle the high pressure ridge. The NFS model (WARNING: Insanely huge file, takes a few minutes to load on broadband) has an insanely rapidly retreating ridge and has Frances making landfall in southern North Carolina on Tuesday morning. The GFDL is a bit more reasonable, but still seems to be handling the situation wrong. Recall that these two models have been suspect because of their handling of the ridge for the past couple of days NOGAPS, (which has kind of been looking kind of good lately) is one of the strong ridge proponents and puts Frances on a scary track where she makes landfall in south Florida and proceeds up the west coast a little before hooking northwest to make a second landfall in Louisiana. The early part of that scenario seems reasonable, but what follows does not. Weather Underground has a graphic displaying these models (pay attention though to the times the models were run as sometimes they are slow at updating the image). It's naptime now... Weather comment []2:40:48 AM trackback []
|
| Before getting into this there are a couple of things to note. 1. Our skill at position forecasting has gotten better in recent years. Our three day forecasts have gotten pretty good and our five day forecasts have gotten as accurate as our three day forecasts of five to ten years ago. 2. Our intensity forecasting leaves something to be desired. Right now we are at a bit of a brick wall in improving. We tend to go with persistence (i.e. a strengthening storm tends to strengthen, a steady storm stays steady, etc.) This has bitten us badly on multiple occasions. I looked at the five day forecast for this evening from Friday night (11 PM EDT) and each of the 72 hour forecasts from that day on. Five day forecast issued Friday night: Off by 105 nautical miles (121 statute). Too far north, not west enough. Intensity was 5 knots low. Three day forecast issued Friday night: Off by 98 nautical miles (90 statute). Almost all of the error is from placing it too far north. Intensity forecast was 5 knots high. It's interesting that the five day forecast was in a sense better than the three day (fewer miles of error per day) Three day forecast issued Saturday night: Off by 46 nautical miles (53 s). Too far north, not west enough. Intensity forecast five knots high. Three day forecast issued Monday night: Position-wise the best of the bunch. Only 25 nautical miles error (29 s). Just a touch too far north and not quite west enough. However, worst intensity forecast of my sample (10 knots too low) as the forecaster got duped by the SHIPS and GFDL models. Not too shabby. It's encouraging that the three day forecasts have been improving each day (reduction by about half each time). (Of course, it is understandable that readers in the Fort Pierce area are hoping for that trend to reverse itself). Note though, that Frances' landfall is expected inside the three day period, so theoretically our error for landfall should be reasonably low at this point. However, the situation remains tricky, so stay tuned... Weather comment []1:38:10 AM trackback []
|
| More blogging goodness
Weather comment []12:28:18 AM trackback []
|
© Copyright
2004
Charles Fenwick
.
Last update: 10/2/2004; 11:03:30 PM .
Last update: 10/2/2004; 11:03:30 PM .
