Hello, I wanted to bring the high profile case of Category 5
Hurricane Michael in 2018 to the forefront for examination. I
went back and plotted an hourly max sustained wind swath and
hourly max gust swath for the RTMA around when Michael made
landfall. Unfortunately, I don't have URMA files, so I can't see
how much URMA improved on the RTMA. However, the RTMA needed a
lot of improvement in this case based on the plots I
created. I have attached these plots as well as the official NHC
tropical storm and hurricane force sustained wind swath. The PSH
product from NWS Tallahassee which lists several wind speeds can
be found here.
The max sustained wind swath is not even coherent. The gust swath
is at least coherent and not too bad for some peripheral areas,
while other areas near the core remain way too low. There was a
100 knot gust measured as far inland as extreme southwest
Georgia. I would imagine the lack of observations due to
communication problems from power losses were interfering with the
analysis in this case. I'm wondering though that if tropical
cyclones are this bad in general, then perhaps the URMA would
benefit from pasting in some kind of gridded TCM wind analysis
into the URMA, sort of like what offices do now as part of their
forecast winds. Regardless, as a forecaster I don't want to be
graded against an analysis like this when it comes to
verification. Hopefully we'll see continued improvements.
--
Don Van Dyke
Lead Forecaster
WSR-88D, Storm Data, and Verification Focal Points
NWS Tallahassee, FL
(850) 942-8833
“Until somebody shows me that a machine can out-forecast
me, or out-forecast other people I consider good forecasters,
I’m not willing to concede that a machine should be issuing a
forecast.” –Al Moller
