The Atlantic Basin recently roared to life after a seemingly lackluster summer devoid of significant tropical threats. Weeks would pass before a disturbance of interest appeared in the Atlantic. Tropical Storm Ernersto acquired full Tropical characteristics on August 16, and is the most recent named storm of the 2018 Atlantic Hurricane season. No tropical systems have developed in the more than two weeks since during what is typically the start of the most active period of Hurricane Season.

Like the flip of a switch,  three systems of concern suddenly formed in the Atlantic Basin only within the last few days.

Tropical Storm Florence formed along a train of moisture flowing off the coast of western Africa Saturday. It is trailed by a cluster of thunderstorms emerging from the coast of Senegal designated Tropical Disturbance 2 . As of September 2, both are too far to the east to ascertain a threat to land,  but recent trends in numerical guidance have suggested the possibility that Florence could take a swipe at parts of the East Coast. Meanwhile, a broad cluster of thunderstorms is developing circulation over the western Caribbean and is likely to impact the United States throughout the upcoming week.

Seemingly lackluster is the keyword in understanding this hurricane season. In an average year, six named systems, three of which become hurricanes, form by September 8 according to climatology statistics from the National Hurricane Center. Tropical Storm Florence marks the sixth named storm of the 2018 Atlantic Hurricane Season. If no additional tropical cyclones develop within the next six days, this hurricane season would fall exactly on the nose of the climatological average. But this is no average year.

This hurricane season started much earlier than average. The first named storm was given to Subtropical Storm Alberto on May 25, one week before the official start of hurricane season. The first named storm typically does not form until July 9. No tropical systems formed in June, but the head-start to hurricane season resumed in early July with Hurricanes Beryl and Chris. One Atlantic Hurricane is rare in early July, let alone two. The second hurricane of the season does not form in an average year until August 28.

Later in July, the Saharan Air Layer (SAL) resurged into the Atlantic. Its withdrawal is part of the reason why tropical activity is coming back to life. The hot dust emerging from the Saharan Desert rides over the cool Atlantic, forming an inversion that inhibits convection and suppresses tropical cyclone growth. Over a month passed until Tropical Storms Debby and Ernesto formed within breaks in the suffocating dust in the first half of August.

Alberto is thus far the only storm to have directly impacted the United States this hurricane season.  All other storms have stayed out at sea. With no direct threats for land after Memorial Day Weekend and with long breaks between Tropical activity, it is difficult to shake the feeling that this is a below-normal season. As peak hurricane season nears, several of the environmental factors that resulted in this dissonant hurricane season will have been vanquished so that even if the three ongoing tropical systems pose little harm to land, future systems may not be so evasive.

The primary factor inhibiting tropical development for several week periods throughout the summer was the continuous flow of the SAL into the western Atlantic. The majority of Atlantic tropical cyclones sprout from clusters of thunderstorms that develop south of the African Easterly Jet (AEJ), a mid-level jet stream that develops in response to the stark contrast in temperature between the Saharan Desert and Sub-Saharan Africa. The thermal circulation that prompts thunderstorm development south of the AEJ works to reduce the difference in temperature across Africa. When an intensifying temperature gradient or other synoptic conditions push the AEJ southward, the SAL is able to spread further south, greatly suppressing tropical cyclogenesis.

This summer, dust from the SAL traveled across the Atlantic all the way to Texas, suppressing organized convection across a corridor spanning the Atlantic Ocean and Gulf of Mexico. The most recent ejection of Saharan dust is visible in water vapor infrared imagery from GOES 16.×4320.jpg

The last injection of the SAL into the Atlantic is denoted by the swath of warm colors in the center of the image. Most of the water vapor contained in a column of dry air is close to the surface, where the atmosphere is the warmest. The AEJ recently weakened and retreated north, permitting tropical waves to breach the open Atlantic. Immediately trailing the SAL is Tropical Storm Florence. A secondary dry layer associated with subsidence from the Azores High is present just to the north of Florence, and is perhaps the most critical feature in Florence’s evolution.

As of 5:00 pm EDT September 2, Florence was a 1000 mb Tropical Storm located roughly 700 mi west-northwest of the Cabo Verde Islands with maximum sustained winds estimated to be 51 mph. Since it still has a weak upper-level vortex, Florence will be steered by mid-level winds over the next several days. A ridge over the Saharan Coast will steer Florence northwest into the aforementioned swath of dry air and over a region of the ocean with marginal sea surface temperatures for tropical cyclone intensification. Sea Surface Temperatures (SSTs) of at least 26°C (79°F) are required to maintain tropical cyclone strength. Florence is poised to spend the next several days over sea surface temperatures of 25-26°C, per the most recent Reynolds SST analysis depicted below.

There are a wide variety of solutions to what happens when Florence reaches the cooler SSTs and nearby swath of dry air.  All models depict minor weakening due to cooler sea-surface temperatures and at least some interaction with dry air. As Florence approaches the dry layer, its circulation will wrap the dry air around the west flank of the cyclone. Some model solutions, like recent GFS runs, only slightly entangle Florence with dry air. It essentially cuts off the flow of dry air around the storm and causes dry air to pool to the west of Florence. If this were to happen, Florence would only modestly weaken on Wednesday, then quickly intensify Thursday upon emerging into an environment of warmer SSTs and low wind shear, achieving Major Hurricane status with relative ease.  It would develop a strong  upper-level vortex which would be steered rightward by the Bermuda High to the north before it approaches land early next week. The GEFS ensemble mean aligns with its parent GFS in a track that makes an early right turn to the north. Only 3 of the 12z September 2 GEFS  run forecast a track that could impact the East Coast of the US.

The alternative solutions are not so benign. The ECMWF and CMC, for example, each draw more dry air into the system. They weaken the storm upon entering the cooler SSTs before approaching the dry layer to the north. The weaker systems are incapable of drawing enough convection to combat intrusion of dry-air.  Conversely, the GFS continues to intensify the storm until it marginally interacts with dry air.  The ECMWF, CMC and other models consequentially keep Florence  a weak tropical storm or tropical depression until the end of this week.  Since these solutions do not forecast a robust upper-level vortex, the mid-level winds will dominate and steer the storm further south and west before an eventual northward turn near the coast of the Southeast US as a major hurricane.  Each passing run of the ECMWF has introduced more ensemble members that depict such a track, causing the mean to shift closer to the coast. The HWRF only forecasts to 126 hours, but it depicts a similar track further to the south.

Currently the slower intensification and more southerly track is favored over the faster intensification, more northerly track. The GFS is mostly by itself, and oddly intensifies Florence before interacting with the dry air despite trekking over marginal SSTs. Additionally, it has historically proven to have an eastern track bias since its convective parameterization frequently results in an intensifying ridge to the right of cyclones.

The following images courtesy of Tropical Tidbits are images of the GFS and ECMWF, respectively, from 12z September 2 depicting these radically different scenarios for Florence. Both are one week forecasts of 500mb Geoptential Height and MSLP valid 12z Sunday. The GFS has one strong Category 3 or Category 4 Hurricane and two tropical depressions or tropical storms. This is in stark contrast to the three tropical storms painted by the ECMWF. The furthest most left storm is Florence, followed by what is now Tropical Disturbance 2 and what could be Tropical Disturbance 3 intensified into tropical storms. It is obvious that the upper-level ridge is stronger in the GFS solution (denoted by red colors) than the ECMWF, which could be a result of the aforementioned bias. Also notice that the GFS solution is further north by 1-2 degrees of laitude and further east by 4-5 degrees of longitude. This is because the upper-level low is strong enough to be picked up by and change the direction of the winds at the upper-levels of the atmosphere. Advection of winds around Florence by the storm’s forward momentum will be sufficient to punch a breach in the circulation around the upper level ridge, as depicted by Florence surrounded by larger values of geopotential height on the GFS.

Each model also has a vastly different suite of ensemble solutions, though their means reflect the discussion of their parent model. An ensemble of spaghetti plots  from global deterministic models, their ensemble means, and hurricane models courtesy of South Florida Water Management District is presented below to demonstrate some of the vastly different possible tracks Florence may take.

Fortunately, potential impacts from Florence are more than one week away. Any speculation about potential landfall is mostly guesswork. Model solutions will continue to evolve overtime since any small change in the forecast can produce much larger changes later in the model forecast.  Such changes will partially be the result of weather events that have yet to occur.  Perhaps the most significant event to impact the forecast is Category 5 Super Typhoon Jebi in the western Pacific. This deadly storm is expected to weaken as it continues northward but will indubitably ravish Japan as a Category 2 Typhoon.  Recurving typhoons like Jebi tend to prompt a hemisphere-wide pattern change that results in the formation of troughs over the Eastern U.S. within 6-9 days. If such a trough invades the eastern US early enough, Florence could be deflected before around the trough before it can do more than generate rough surf along the Eastern Seaboard.

More pressing in the short-term but less threatening overall is Tropical Storm Gordon. The thunderstorms circled on the satellite image below have organized over the last several days. The system has developed a closed surface circulation, and models are in strong agreement that ridging over the Southeastern US. will steer the cyclone between Florida and Cuba into the Gulf of Mexico by Monday afternoon. Low wind shear and warm SSTs will provide for ideal conditions for steady intensification.

The only limiting factor is time, as upon formation the storm will have only 48- 60 hours to intensify before making landfall Tuesday night as Tropical Storm Gordon, most likely between New Orleans and Pensacola, per unusually strong model agreement. The global models like the ECMWF and GFS have a tropical depression make landfall near New Orleans. Roughly half of the ECMWF’s 51 members depict a weak tropical storm at landfall whereas none of the GFS members depict a tropical storm. Hi-RES guidance does depict a Tropical Storm, which would make landfall further east. This solution is favored given that convective parameterization of global models tends to slightly enhance ridging, a consequence that already resulted in them under-representing Sunday’s organization.

A brief period of gusty winds and heavy rain along the Gulf Coast is the most likely outcome of what is now Tropical Storm Gordon.  The potential exists for 3-6″ of rain with locally higher amounts to fall over a short period of time. Impacts are only likely to last for 24-36 hours so flooding and damage will be minimal.

Hurricane season is intensifying rapidly. New threats will continue to emerge over the Atlantic as tropical waves continue to enter the ocean from Africa. We will continue to update you as Florence and Gordon continue to evolve and as new threats become apparent.


As Head Meteorologist, Josh bridges together weather forecasting with product quality and innovation. He vigilantly monitors weather threats across the country and directly engages with clients to outline hazards posed by expected inclement weather. He also offers insights into meteorology and numerical weather prediction to aid the development team in improving and expanding the diverse set of products. Feldman graduated from Stony Brook University in 2018 with Bachelor of Science degrees in Atmospheric and Oceanic Sciences and Physics.

Comments are closed.