Dorian is now just past St. Lucia/St. Vincent.  The storm is still hovering around 50mph wind speeds, a middling tropical storm, and from the current (6:30am ET) high resolution satellite image showed some big thunderstorms are passing over Martinique.  The dry air in the mid levels of the atmosphere in the Caribbean is keeping the storm suppressed, and development seems to be stalled.  The present National Hurricane Center forecast has  backed off on forecasting the storm to become a hurricane before it reaches Puerto Rico and the Dominican Republic (the eastern half of the island of Hispaniola), although there are still hurricane watches up for both.  Preparations in PR and the eastern DR for a strong tropical storm or minimal hurricane make sense at this point, especially for the inevitable localized flash flooding that accompanies these storms.  Wednesday will likely be a wet and blustery, but hopefully not catastrophic day for them.

As to what happens next, it depends on some subtle wobbles and not so subtle mountains.  If the storm threads the Mona passage between the islands of Hispaniola and Puerto Rico, it could maintain its circulation enough to rebound in the warm waters of The Bahamas. This is the scenario NHC is showing.  Butf it wobbles either left or right of this track, it could well be shredded, and the islands and Florida just get a messy tropical rain system.  Hard to tell at this point, but by tomorrow we will have a better handle on this since we will know how much the dry air inhibits (or even decreases) the storm, and if is going to run aground on Hispaniola.  The obligatory “spaghetti map” shows the primary dynamic models taking the storm as the NHC has it, mostly towards the Dominican Republic. This map is showing all of the available tracks … for those of you in Georgia/SC/NC, don’t freak out about the lines headed up towards your area.  Those are mostly ensemble track members that are low probability outcomes, and often represent tracks where the storm is little more than a big thunderstorm.  They are designed to help understand the uncertainty in the forecasts.   It is pretty irresponsible to cherry pick from those tracks, as some local weathercasters and amateurs do.  The colored lines are the “primary” model tracks that are the ones to examine – but that red line is the Official NHC Track that is the one to care about.

Here’s this morning’s Haetta/TC impact map, based on the official forecast …

The National Hurricane Center started advisories on the low pressure system off the Southeastern US Coast as Tropical Depression Number 6.  They are expecting it to become a minimal tropical storm before skimming the coast and impacting Nova Scotia and Newfoundland …

Dorian is an interesting storm, very compact, and hard to forecast the intensity since there are evenly balanced factors for and against.  So far, “for” is winning.  Here’s an animation of the last 30 minutes or so (4:00 to 4:30pm EDT) visual band one minute GOES images (click to embiggen/animate):

That central cloud feature is called a “central dense overcast” or CDO.  The National Hurricane Center is sticking with a forecast taking the storm over Western Puerto Rico and Hispaniola as a minimal hurricane. The island is now under a tropical storm watch, and will likely get hurricane watches/warnings.  That’s a smart bet, given how the storm is behaving.  The dynamical intensity models are really mixed: some virtually kill the storm, some spin it up.  The big factor, as noted this morning, is the very dry air in the lower and middle atmosphere.  Some is getting pulled in to the storm, but not as much as if it were bigger.  The track isn’t in much doubt: they all pretty much go up towards the Mona Passage (between Puerto Rico and Hispaniola).  Here is the HWRF model, which clearly oscillates in intensity and never brings it up to hurricane strength.

But NHC has good reasons to be cautious with this one.  Here is the official forecast track and intensity impacts:

On this track, the worst impacts would be over western PR, damage could be upwards of $500 million, but the fragile infrastructure of Puerto Rico really doesn’t need this additional insult.  Rain and flash floods, and electrical system problems are the most likely impacts.  As for The Bahamas and Florida, it’s a touchy (but not panic time) situation.  If the storm is a little weaker than the NHC forecast, and gets shredded by the mountains of Hispaniola, then it might just be a wet tropical low.  But the water in The Bahamas is very warm, and upper air conditions favorable – if Dorian can survive to reach that area somewhat intact.  if so, bad news for them and potentially Florida, with a real Hurricane landfall possible.  As for Georgia/SC/NC, the more likely scenario is for the storm to cross Florida (assuming it survives), based on the longer range models. But is something we won’t have a handle on until Thursday, so if you have a plan plenty of time to wait and see.  As always, for the official word check the NHC Key Messages graphic!

Note that while Dorian is in a mixed environment, seems to be slowly intensifying and organizing.  NHC has increased their forecast intensity, and upgraded the tropical storm watch for St. Lucia to a Hurricane Watch.  That said, bottom line for the Windard Islands hasn’t really changed much, the main risk is flash flooding, with perhaps a bit more risk of isolated wind damage.  For the Puerto Rico/Hispaniola, NHC is hinting at hurricane watches later.  That is a reasonable precaution; as always, use NHC’s “Key Messages” graphics for a good summary of actions you should be taking.

—

Dorian (AL052019) continues to wander towards Barbados and the Windward Islands as a very small tropical storm.  Warnings are up for Barbados, St. Lucia, and St. Vincent.  The main threat to these islands is localized flash flooding on the steep slopes of these islands; unless something breaks that shouldn’t, winds and waves should not cause extensive damage.

The official NHC forecast brings the storm to near hurricane strength as it approaches the Mona Passage (the gap between Puerto Rico and Hispaniola (Dominican Republic).  That is at the extreme upper end of the intensity forecasts.  I understand why NHC is doing that – they want people in those areas to pay attention and to be prepared just in case, and if you live in those areas you should certainly check your plans and be ready.  But it is pretty unlikely the storm will be that strong when it gets to the Puerto Rico area.  There is a lot of dry air over the region (click to embiggen)…

So what does that mean?  Again, if you are in the Windward Islands, you should be wrapping up preparations for a tropical storm.  If you are in Puerto Rico or on Hispaniola, and you have a plan, watch and be ready, but no reason to get too worried at this point.  The dynamical models like the new GFS/FV3 and the sainted ECMWF decay Dorian fairly quickly.  Normally I avoid showing individual model outputs, but will make an exception here given the situation.  Here are the forecasted impacts based on the GFS Ensemble blend, using my Haetta/TC model …

For those in Florida, definitely don’t freak out that the storm seems to hit you in 6 days: just because the line comes your way doesn’t mean it will have significant impacts, in this case, look at the color key: “Small Trees Sway” and “Large Trees Sway” (30mph winds or so).  That and rain: no big deal.  Kite flying weather.  Well, except the lightning.  So don’t do that.

But for watch and warning purposes, and taking action, the official NHC forecast is what you should be using.  It’s almost always the conservative (safe) choice, and if you follow it you will rarely be caught off guard.  Here is the same impact map using the official forecast:

Note the official forecast only goes out five days.  There is a very good reason for that: beyond that time frame the forecast accuracy reaches the point the forecast is virtually unusable, and five days is more than enough time to prepare (and three days is really enough, assuming you’ve made your pre-season plans appropriately).

Tropical Storm Dorian (AL052019) is a small, weak tropical storm about 900km (that’s 560 miles for you backwards ‘Muricans!) east of Barbados. A tropical storm watch has been issued for that island since most of the objective track models take the storm near or over the island by late Monday, and the official forecast track shows a direct hit.  As usual, there are two questions: where is it going to go, and how bad will it be when it gets there?

Two key tools are used for the “where”: objective track models, and forecaster experience to create the “official” track.  In this day with dozens of tracks (if you include individual ensemble members, which you shouldn’t in and of themselves) to choose from, amateurs often overlook that second bit and spend a lot of time arguing over this or that individual track model. The Official NHC forecast incorporates a third component: watch and warning considerations.  In this case, Dorian is a small storm, in a mixed environment.  The present track has the storm crossing Barbados and the Windward Islands as a small but mature tropical storm, briefly hit hurricane strength in the mid-eastern Caribbean sea, then weaken as it approaches the island of Hispaniola (Dominican Republic/Haiti).  The tracks are fairly well grouped for a weaker storm …

For Puerto Rico: the forecast track has shifted away, and will probably shift away more later today. It’s a good ways away – unless it speeds up it looks like a Wednesday event for PR, and while the infrastructure is still (resisting the urge to rant) weak, and the rain potentially a problem, it doesn’t look like it cause significant problems.  But it might be a close call.  Also, even if the storm gets a bit stronger at first, it will be entering a pretty dry mass of air that should inhibit strengthening (or even cause it to decay).  Here’s the latest water vapor band image; all that orange is dry air that can choke off a storm:

As for “how bad”, the main threat to Barbados and the Windward Islands is rain and flash flooding.  But it’s a small storm in dry air, so that may be pretty localized, so if folks button up they should be ok.  The big question is more if it can intensify later and threaten the Greater Antilles (PR/Hispaniola or, if the track shifts further west, Jamaica).  As for The Bahamas and Florida, that’s at least a week away and too far out to forecast, but if it survives passage over the Caribbean (and isn’t shredded by Hispaniola) it could reform/rebuild, but that’s absolutely not worth worrying about at this point.  Here’s the forecast intensity …

Update at 1pm ET: we’re now at “Meh” for sure: NHC has started official advisories on Tropical Depression number 5 (AL052019).  I’m surprised at how aggressive they are in the initial advisory, taking the storm up to hurricane status in 5 days.  But this means folks in the Caribbean should pay attention now.

The main tropical story is in the West Pacific, Tropical Storm (Tropical Cyclone) Bailu is making landfall on southern Taiwan.  Main risk is flash flooding and landslides, estimated impacts in the low hundreds of millions of USD.

The Atlantic might look interesting at first glance, but if you have your hurricane plans in place, it’s more of a yawn at this point.  There are two “invest” areas in the Atlantic –  Chantal (AL042019) is no more, fading out in the North Atlantic.  The first area, AL982019 (recalling INVEST areas get temporary ID’s so you may see that code again!) has moved inland over Flordia.  It is causing rain in that state as well as over The Bahamas.  This is the expected scenario, that it will drift out to sea and head northeast away from land (either Florida or the northern Islands).  At that point it has a good chance of becoming a depression or storm, but by then it won’t be a problem except maybe for Bermuda.  Here’s the full track map for this morning (Saturday 24 August) … there is no official forecast yet, but I think the most likely track is like the red line (the GFS Ensemble).

Further south, over the mid-Atlantic, the system tagged AL992019 in the Automated Tropical Cyclone Forecast System (ATCF) is lurching towards the Caribbean and almost rises to the “meh” level.  NHC gives it a 70% chance of becoming a minimal tropical storm, but conditions are unfavorable once you get close to the Islands, so it’s not likely to get strong enough to cause much damage.  Still, worth paying some attention to, given the risk of flash flooding to cause localized damage on many Islands, and there is a small chance it could spin up more than expected..  Still to far to anticipate impacts to Puerto Rico, should know more by Monday.  Again, still just an invest, so no official tracks yet. (OK, now there is at 11am, as AL052019).  Highlighted red line is the GFS Ensemble, which at this point is as good a guess as any.

There is a tropical system passing over South and Central Florida this weekend (23-25 August 2019).  Not very unusual – Florida gets systems like this just about weekly during the summer!  But this one does show some signs of organization, and once it moves back out to sea might cross the threshold to become a system officially tracked by the US National Hurricane Center.  Before they are “officially” tracked (in other words, with formal 6 hourly public advisories), these systems are called “Investigation Areas” or INVEST areas. It’s not too likely this thing will be strong enough over land to cause problems, and any strengthening shouldn’t happen until after it moves away from the US or Islands.

Right now, this system is called “AL982019”.  What does that mean? All tropical systems are given tracking codes in the form BBNNYYYY where BB is the basin, NN is a number code, YYYY is the year.  The basin codes are pretty straightforward: AL is the Atlantic, EP is the East Pacific (off of Mexico), CP is the Central Pacific (Hawai’i), WP is the West Pacific (Asia), IO the Indian Ocean, and SH the Southern Hemisphere.  The number codes are a little tricky.  Numbers 90 and above are reserved for INVEST areas and are reused during the season.  This bears repeating:

Be careful searching on teh intertubs for invest areas by the codes:
they are re-used during the season!

If the system meets the criteria to become a system with advisories, it is given a “permanent” code, the next in line starting at 01 for each year (January 1st in the Northern Hemisphere.  In the Southern Hemisphere, the year starts September 1st of the previous year, so a tropical storm forming off of Australia in November this year might have the code SH032020! ).  Note this is different from the name of the storm.  The criteria for a storm being “tropical” and being “named” are different!  So AL03 may not be the “C” named stormed.  In fact, the presently active Chantal is AL04 because there was a previous system that was organized enough to start advisories, but did not meet the criteria to become a “named” storm.

Here’s the track map as of 7am this morning.  Some models kill it off, some spin it up after it moves off shore into the Atlantic.  NHC is saying 30% chance of it meeting the criteria to be a minimal tropical system in the next two days, and 60% in the next 5.  Again, even if it does, this system is not likely to have any significant impacts.