Hurricane Dorian, as you are all well aware, with sustained winds of 185 mph, is pulverizing Abaco island today and will slowly grind its way through the Grand Bahama Island over the next 36 hours. There are a few other diaries today covering news about the hurricane. In this diary, we take a look at the future of Dorian as it nears Florida’s East coast Tuesday morning and some of the science behind the forecasts.
The big question that is lingering in everyone’s mind is whether it will make landfall in Florida and even if it doesn’t, how close will it be and will it make landfall in the Carolinas or veer off into the Atlantic ocean.
Here is the 11 p.m. forecast and advisory from the NHC —
Dorian is not expected to pick up any more speed (famous last words) but its wind field will widen as it approaches Florida. Its movement has slowed down to 6 mph and it is likely to stall over the Grand Bahama island before veering North. Hurricane-force winds extend outward up to 45 miles (75 km) from the center and tropical-storm-force winds extend outward up to 140 miles (220 km).
FORECAST POSITIONS AND MAX WINDS
INIT 02/0300Z 26.6N 77.9W 155 KT 180 MPH
12H 02/1200Z 26.8N 78.5W 145 KT 165 MPH
24H 03/0000Z 26.9N 79.0W 135 KT 155 MPH
36H 03/1200Z 27.3N 79.3W 125 KT 145 MPH
48H 04/0000Z 28.2N 79.8W 120 KT 140 MPH
72H 05/0000Z 30.6N 80.2W 105 KT 120 MPH
96H 06/0000Z 33.5N 77.7W 85 KT 100 MPH
120H 07/0000Z 37.5N 71.8W 75 KT 85 MPH
There are mandatory evacuation orders in many coastal areas in Florida, Georgia and South Carolina. Please check your local advisories.
Here is a look into the eye of the monster today, as it intensified into a Cat 5 hurricane with sustained winds at 185 mph -
The predicted path of Dorian has shifted to the right quite a bit over the past 2 days of forecasts. The animation below shows the progression of the forecasts. Keep in mind that the forecasts never specify a single path but a range of possibilities with what’s known as the “cone of uncertainty”. The NHC forecast cone has shifted to the right, so much so that the the center line of the cone now lies offshore.
There are a multitude of weather prediction models that have been hard at work churning out their predictions round the clock. All of them predict a sharp northern turn after 36-48 hours, but there are differences on when exactly that turn occurs. An earlier turn soon after passing the Grand Bahama island will result in Dorian remaining offshore, while a later turn causes landfall in Florida during the northward movement.
Many models have shown a westward creep of the projected path this evening, which is bad news for Florida.
The European model is keeping Dorian offshore but it might make landfall briefly near the outer banks of NC -
The HWRF model (used by NOAA) is also keeping the track offshore but the track has shifted a bit left in the last 12 hours -
GFS is in agreement with ECMWF -
More comparisons — models are still shifting their forecasts ...
More Insights and Forecasts
Here are some additional insights from weather experts from around the world —
Levi Cowan tells us not to dwell too much on the actual wind speeds and categorization — Cat 4 is devastating enough.
A great video summary by Levi Cowan -
This is a wind map for Tuesday 2:00 p.m. EDT. It shows Dorian’s eye located off the coast of Florida but sustained winds of 85 km/h slamming the shore. One can imagine the storm surge this slow moving hurricane is going to cause.
The same chart drawn 8 hours later shows a bigger Dorian closer to the coast and higher wind speeds at the shore — 170 km/h.
Hurricane and Weather Forecasting
This is a quick and simplified description of weather forecasting techniques. I am no expert in this area, so hopefully readers will add corrections and additions to the description.
Most weather models use computer simulations to predict global weather conditions several days into the future. They use equations that model the laws physics and represent how atmospheric quantities such as temperature, wind speed and direction, humidity, heat transfer, solar radiation, relative humidity, cloud formation, phase changes of water and surface hydrology, change given a set of initial values.
The model requires a set of inputs (initial conditions) that represent the actual values of these parameters from around the world and at different altitudes. Ideally, small cells (say 1 km or smaller cube) would be used from the around the world going up several km, but the actual data is less granular and more spotty.
The main inputs come from country-based weather services which use devices (called radiosondes) in weather balloons that measure various atmospheric parameters and transmit them to a fixed receiver, as well as from weather satellites. Additional info is obtained from commercial aircraft, ships and from reconnaissance aircraft.
Once initialized, the models are run on supercomputers to solve these partial differential equations and compute new values for every cell in the 3D grid for the next time interval (typically a few minutes into the future), as a function of its own parameters and those of its neighboring cells. The models repeat computations for interval after interval until we get predictions for the desired time in the future (typically 12 - 72 hours).
Obviously, predictions have errors caused by inaccurate and/or incomplete data (e.g., measurements over oceans are spotty), the low resolution of the cell grid (several km across), limitations of the models in accounting for terrain effects, atmospheric pollutants, solar effects, absorption and reflection variances, etc) and other random events. A more fundamental problem lies in the chaotic nature of the partial differential equations that govern the atmosphere. It is impossible to solve these equations exactly, and small errors grow with time (doubling about every five days). Present understanding is that this chaotic behavior limits accurate forecasts to about 14 days even with perfectly accurate input data and a flawless model. en.wikipedia.org/… Also see the famously named Butterfly Effect.
Often, the model is run multiple times, each produces a slightly different output. The outputs are “averaged” to produce a more reliable “ensemble” forecast. Often, the output of different models are plotted on the same chart, resulting in the so-called “spaghetti” plots.
You may have seen these acronyms for weather models in news, tweets and spaghetti graphs. These are the most commonly used weather models, used for global weather as well as hurricane and cyclone modeling.
||The European Center for Medium-Range Weather Forecasting model
||The Global Forecast System model run by the NWS
||The Hurricanes in a Multi-scale Ocean-coupled Non-hydrostatic model, used by the NWS
||The NWS/Hurricane Weather Research model
The Model for Prediction Across Scales, used by the National Center for Atmospheric Research (NCAR)
||The United Kingdom Met Office model
For those of your friends who are pooh-poohing weather forecasting (hey they can’t even get this right a few days out), ask how many of them would look at this picture and predict that Dorian is not going to slam into Florida and instead follow the coastline.
NOAA and the 53rd Weather Reconnaissance Squadron (53rd WRS) send aircraft with special instrumentation several times a day into the hurricane to make more precise and frequent measurements of parameters within hurricanes. The aircraft are equipped with radar and other instruments. Typically, they fly straight across the hurricane at altitudes between 1,000 feet to 10,000 feet. It sounds dangerous to fly into through the eye-wall of a hurricane swirling at 185 mph, but they have been doing so for decades with a near perfect record.
A key part of hurricane monitoring involves the dropping of “dropsondes”, a small tube with instruments and a parachute, at various places in the storm such as in the eyewall and at the center of a hurricane. As it falls, and right up until it hits the water, it sends temperature, humidity, barometric pressure, wind speed and wind direction back to the aircraft twice each second. The data is sent via satellite to data centers in the U.S., from where it is made available to researchers and meteorologists within minutes after collection.
A look inside the cockpit of the NOAA Lockheed WP-3D Orion four-engine turboprop aircraft -
The terror inside the cockpit, but all in a day’s work for these talented and dedicated men and women.
And the glorious view from inside the eye of Dorian -
www.tropicaltidbits.com/… is a terrific site that plots the data in near real-time in 10-minute increments. Here is an example of the flight path across Dorian today of the AF recon plane. Each number in the chart represents the dropping of a dropsonde.
Here is data displayed at the website for an AF flight today between 8 and 10 p.m. EDT, which included two passes through the eye. The most interesting graph is the one on the top right, which shows wind speed (in knots, 1 knot = 1.15078 mph).
Hurricane Tracks and Distribution
The following map from NASA shows the tracks of all tropical cyclones during the 1985-2005 time period. The Pacific Ocean west of the International Date Line sees more tropical cyclones than any other basin, while there is almost no activity in the Atlantic Ocean south of the Equator. Tropical cyclones do not form around the equator.
A large number of Atlantic hurricanes originate near the Cape Verde islands, off the west coast of Africa and are know as the “Cape Verde hurricanes.” These hurricanes form from a tropical wave that has passed over or near the Cape Verde islands after exiting the coast of West Africa. Hurricanes Dean (2007), Ivan (2004), Floyd (1999) Hugo (1989), Fran (1996), Isabelle (2003) are examples of Cape Verde hurricanes.
There is another storm currently brewing off the Cape Verde islands, which NHC is keeping an eye on.
Yes, hurricanes are not uncommon, but there is a link between climate change and increased hurricane intensities.
As is usually the case, forecasts will firm up as we get closer to Tuesday. The forecast for the U.S. mainland at this point contains a narrow range of possibilities, but a wide range of possible damage and destruction. So, the best advise is to listen to forecasts and instructions from the NHC and local authorities, prepare for the worst and hope for the best.
References and further reading
- Hurricane Forecast Computer Models — www.wunderground.com/…
- Numerical weather prediction — en.wikipedia.org/…
- NHC Track and Intensity Models — www.nhc.noaa.gov/…
- National Hurricane Center wiki — en.wikipedia.org/…
- Hurricane Hunters wiki — en.wikipedia.org/…
- NOAA Hurricane Hunters — www.omao.noaa.gov/…
- Hurricane Hunters of the Air Force Reserve — www.hurricanehunters.com
- Hurricane Hunters FAQ — www.hurricanehunters.com/…
- Hurricane Dorian Forecasts, Updates and Science — www.dailykos.com/…
How To Tame a Hurricane — www.dailykos.com/
Monday 5:00 a.m. —
- Dorian has weakened a bit, 165 mph still Cat 5
- The eye has expanded to a diameter of 23 miles, this is consistent with development of a concentric eyewall
- No change in projected trajectory
- Dorian is almost stalled, now moving at only 1 mph
- Grand Bahama island will experience a prolonged period of the wrath of Dorian
Model HWRF is trending away from Florida, GFS and ECMWF are trending closer. Best to assume the worst and make preparations.