Status: Finalised Material from: Linus
1. Impact
Hurricane Dorian formed on 24 August in the central Atlantic, and caused huge damages on Bahamas on 1-2 September. The cyclone later affected the U.S east-coast and also Canada.
https://en.wikipedia.org/wiki/Hurricane_Dorian
https://www.bbc.co.uk/news/world-latin-america-49602445
2. Description of the event
An animation of infrared satellite images for the full lifetime of the cyclone can be found here (from Kimberly Wood, Mississippi State University): http://arashi.geosci.msstate.edu/images/2019Dorian.mp4
A radar loop from Bahamas can be found here (thanks to Brian McNoldy, University of Miami): http://bmcnoldy.rsmas.miami.edu/tropics/dorian19/Dorian_1-3Sep19_bahamas.gif
3. Predictability
3.1 Data assimilation
The plot below shows the track and intensity of Dorian in BestTrack (hourglass) and analysis (circle). The analysis missed the position at 28 August 12UTC when the cyclone was in the Caribbean Sea.
The assimilation cycle on 28 August 12UTC had problem to determine the position of the cyclone. This was caused by large first-guess errors, as seen in the plot below.
3.2 HRES
3.3 ENS
The plots below show tropical cyclone strike probability on 25-28 August, in forecasts from 25 (first plot) to 22 August (last plot).
The plots below show the tracks (ensemble -grey, HRES - red, ENS control - blue, best track - black), position and intensity on 3 September 00UTC (ensemble - squares, best track - hourglass) in forecasts from 3 September (first plot) to 23 August (last plot). The cyclone was recognised as a tropical storm on 25 August.Three main shifts in the tracks: Majority of members towards Bahamas (26 Aug), easterly path (28 August) and northward turn (31 August)
.
The plot below shows the cyclone intensity in terms of central pressure (top), maximum wind speed (middle) and the propagation speed (bottom) for the forecasts from 3 September 00UTC (first plot) to 29 September (last plot), all 00UTC.
3.4 Extended-range forecasts
Tropical storm strike probability 2-9 September in extended-range forecasts, starting from the latest forecast (2 September). The first extended-range forecast to pick up the signal was 26 August, an expected result from the medium-range plots.
Accumulated cyclone energy 2-9 September in extended-range forecasts. All forecasts from 26 August and earlier indicated less activity than normal.The signal for the activity in the Atlantic in the early forecasts originated from the south-eastern part of the basin, where Gabrielle later formed.
3.5 Comparison with other centres
4. Experience from general performance/other cases
5. Good and bad aspects of the forecasts for the event
- Missed rapid intensification starting on 31 August
- Reasonable capturing of the probability for slow propagation over the Bahamas and the northward turn before Florida
- Short notice about cyclogenesis