Dr. Masters co-founded wunderground in 1995. He flew with the NOAA Hurricane Hunters from 1986-1990. Co-blogging with him: Bob Henson, @bhensonweather
By: Dr. Jeff Masters , 2:46 PM GMT on September 23, 2008
Tropical disturbance 93L continues to hang out just inland along the northeast coast of the Dominican Republic, and is not a hurry to go anywhere. Dominican Republic radar shows a broad circulation with a large area rain, but the rain is not organized into low-level rain bands, and is mostly to the southeast. Visible satellite loops show very little heavy thunderstorm activity near the center of 93L, near the northeast coast of Hispaniola. The heavy thunderstorm activity is mostly to the south and east, but there are no signs that the center of 93L will relocate itself under the heavy thunderstorm activity, and I do not expect this to happen. The thunderstorm activity is displaced to the southeast because strong upper-level winds from the west are pushing the thunderstorms away from the west side of the storm. The wind shear remains about 15 knots, which is marginal for tropical storm development.
The record rains of 93L--as much as 24 inches in 24 hours--have left a gigantic mess in Puerto Rico. Several highways are blocked by flood waters and landslides, and one bridge was washed out in Guayanilla by a flooded river. All but two rivers have fallen below flood stage this morning, and the worst of the rains and flooding are done. Damage to agriculture in Puerto Rico from the storm has been estimated at $14 million.
In the Dominican Republic, heavy rain has been limited to the extreme eastern end, near Punta Cana, where satellite estimates indicate up to six inches of rain has fallen. Rainfall in the capital, Santo Domingo, has been less than an inch, according to three personal weather stations there. Additional heavy rains of 4-8 inches are likely today and Wednesday in many portions of the Dominican Republic from 93L.
Haiti has thus far escaped heavy rains from 93L. It will be a close call, but I believe Haiti will only receive 2-4 inches of rain from 93L, except for the extreme northeastern part of the country near the Dominican Republic border, where 3-6 inches may fall. Highest rains in the Turk and Caicos islands and southeastern Bahamas should be in the 3-6 inch range.
Figure 1. Rainfall rate from 93L as estimated by microwave satellite imagery at 6:30 am EDT 9/23/08. The heaviest rains of 1.4 inches per hour were just south of the eastern tip of the Dominican Republic. Image credit: Navy Research Lab, Monterey.
Well, 93L has not done what we expected, and continues to resist the models' efforts to pull it to the north. The northward turn has been delayed by about a day behind what the models originally forecast, likely delaying the eventual impact of 93L on New England until Saturday. I have no choice but to continue to forecast that 93L will turn to the north in the next 24 hours. A developing coastal storm off the coast of North Carolina should impart enough of a northward motion in the steering currents to make this happen. Wind shear over 93L has also refused to obey model forecasts, and remains at a higher-than-expected 15 knots. The current wind shear forecast from the SHIPS model keeps the shear at 15-20 knots for the remainder of the week. Other models show lower shear, but intensification of 93L into a Category 1 or 2 hurricane later this week, as has been consistently forecast by the GFDL and HWRF models, is probably overdone. There is also a large amount of dry air to the northwest of 93L that will probably interfere with development.
The models are now in fairly good agreement that a strong coastal storm--which could be extratropical or subtropical--will develop off the coastal of North Carolina today. This low will probably bring winds near 40 mph to the coastal waters of North Carolina on Wednesday. As 93L is drawn northwards, the two storms will probably rotate cyclonically around a common center (the Fujiwhara effect), sending 93L hurtling into the coast somewhere between North Carolina and Nova Scotia. Considering that we are trying to forecast a complicated interaction between a storm that hasn't formed yet and another storm that refuses to obey the forecast models, confidence in the forecasts for both of these storms is low. Residents along the entire U.S. and Canadian coast from Georgia to Nova Scotia should anticipate the possibility of tropical storm conditions this week. In Georgia, South Carolina, and North Carolina, this would likely be due to the extratropical low pressure system. New England may get a 1-2 punch on Saturday and Sunday: a tropical storm, followed by heavy rains drawn up from the south from the remnants of the extratropical storm.
Links to follow
Dominican Republic radar
Puerto Rico radar
Video of a boat that weathered Hurricane Ike in Galveston Bay
The Hurricane Ike "NEXT TRUCK CHALLENGE" continues
Two wunderground members, presslord and violet312s, have announced that they will match two dollars for every dollar in contributions made to portlight.org. This charity has really made a difference in some of the hard-hit areas of Texas and Louisiana affected by Hurricane Ike neglected by the traditional relief efforts. A quote from Paul Timmons (AKA Presslord), who has helped coordinate this effort:
My wife just called...they will NOT be staying in Bridge City. They unloaded half the truck there, and the folks from Winnie TX (on the Bolivar Peninsula) called by radio and begged them to bring the rest of the supplies there. So, they are on the road to Winnie, where they will stay at the fire house in the dark. She and SJ are in tears...my wife quoted one of the Bridge City officials: "Thank God for y'all.
Figure 2. A Florida fireman sits behind supplies that Portlight Strategies delivered to the Winnie-Stowell fire station. Image credit: Storm Junkie.
Your contributions do make a difference, and you can read more about the effort at at stormjunkie's blog.
I'll have an update later today, after the Hurricane Hunters have had a chance to check out the storm.
Comments will take a few seconds to appear.