Since Hermine has passed we have been looking back at the forecasts since about a week beforehand.
One of the things I always think about before, during and after a tropical system, is the cone.
- The cone used to illustrate Hermine's potential path is not a forecast
- It's a graphical representation of historical error
- No matter the severity of the storm, that graphic never changes
Some call it the “Cone of Uncertainty”, the “Cone of Error”, the “Forecast Cone,” etc. The last example is the worst because the cone is NOT a forecast cone.
Let me say that again…
THE CONE IS NOT A FORECAST.
It is simply a graphical representation of historical error at each time period. In fact, when a forecast is made for a tropical depression, tropical storm, or hurricane, a forecast point is determined at various time periods -- 12, 24, 36, 48, 72, 96, and 120 hours are used as those forecast time periods.
So, each storm will have a single point forecast for each time period. Then, a circle is drawn around each point. After that, the circles are connected, forming the cone.
But, here is where it is interesting, the circles at certain time periods, in a given hurricane season are the same size from one storm to the next. In other words, the cone’s size doesn’t change from one storm to the next.
It has nothing to do with the size of the storm, the strength of the storm, the shape of the storm, or the forecast confidence of the storm. It is always the same.
At 12 hours the “historical error” surrounding the point has a radius of 30 miles. That means, that in this season EVERY 12 hour forecast cone will be 60 miles wide. Compare that to at 120 hours, the radius is 237 miles. Every 5 day cone is 474 miles across! So, looking at Hermine, compare some early forecasts with their cones to some recent ones (a few hours before landfall) and you will notice that there really wasn’t much fluctuation in the forecast track (the line connecting the points).
But, you will notice that in one of the earlier forecasts, when Tampa Bay was “inside the cone”, that that cone stretched from Ft. Myers to near Tallahassee, yet as time passed, the Tampa Bay Area eventually was no longer in the cone, yet the forecast was essentially unchanged. It isn’t that confidence in the track forecast changed, it was simply that the historical error at sooner time periods shrinks.
I already mentioned the 12 hour and the 120 hour. But, in the chart below you will see the error used for the cone width at the various forecast time points.
- 12 hour 30 mile radius
- 24 hour 49 mile radius
- 36 hour 66 mile radius
- 48 hour 84 mile radius
- 72 hour 115 mile radius
- 96 hour 165 mile radius
- 120 hour 237 mile radius
Here is how you should use this information: Know that each storm is unique, that each forecast presents its own set of challenges and quirks in the models.
Being in the cone isn’t necessarily bad and being outside the cone doesn’t necessarily mean you are in the clear. It will all depend on the particular storm. It is our job to communicate the variables to you. How big is the storm, how far out does the wind field extend, how confident are we in the forecast track, intensity, etc.
In the case of Hermine, and Colin in June, we said many times that we were on the “bad” side of the storm or the “wet” side because most of the significant weather was to the right of the center. Again, the next storm might be different. Use the cone as one of the tools you look at when a tropical system is forecast, not the only source of information.
By mid-afternoon on Thursday the cone is even smaller as Hermine is already a hurricane and about to pound out coast with two rainbands bringing tropical storm force winds by evening.
By the 5 a.m. advisory on Thursday, the Tampa Bay region had experienced very wet and stormy weather for 24 hours. Hermine was now expected to be a Category 1 Hurricane just before landfall in the Big Bend as it turned out to be.
But Tampa Bay and all of our counties are no longer “in the cone” which gives some a false sense of security. The impacts from future Hurricane Hermine would extend 120 miles east of the center by Thursday afternoon bringing strong winds and storm surge to our coast.
On Wednesday morning, the cone is smaller as future Hermine is expected to make it all the way off the New England Coast by the weekend.
Notice how accurate the track forecast ended up being with the landfall near Tallahassee late Thursday night.
By Tuesday, the cone shrinks as the forecast is “stretched” out in time. But the impacts and over weather forecast for us doesn’t change.
One of the first cones issued on future Hermine when it was still a Tropical Depression Sunday night.
