From familial discussions tonight, it seems to me that many in my circles have come to view the “cone” as the area where a storm is “definitely going.” The constant shifts of the cone, however important & instructive they mean to be, are causing serious messaging problems. 1/ https://twitter.com/1900hurricane/status/1297738862169657344

We deal with uncertainty in our lives every day. We apply probabilities in nearly every aspect of our lives to deal with those uncertainties. When it comes to the weather, our local daily forecasts include probabilities: “there is a 30% chance of rain today.” 2/

Unfortunately, it seems, at least in my circles, that the cone has become a line of certainty, rather than of uncertainty, definitely not a result that anyone intended, I don’t think.

The definition of the NHC Track Forecast Cone can be found here: https://www.nhc.noaa.gov/aboutcone.shtml  3/

The messaging behind the cone is always extremely difficult, but as it pertains to #Laura, it is becoming even more difficult and is causing lots of confusion. For example, the shifting west then east then back west then back east is very confusing & anxiety-inducing to many. 4/

Some look at the cone as the 100% likely direction: “that’s where they say it will hit.”

Some see the shifting east or west, possibly into or even out of the cone, as a sign of certainty, too: “I am no longer in the cone, so I am good,” or “I am now back in the cone, wth?” 5/

Some look at the constant shifting of the cone as a sign of an uncertainty in the forecasts, though, and, therefore, start to lose faith in the system.

I, for one, love using the probabilities of landfall statistics. But, that stat isn’t typically disseminated to the public. 6/

Every storm is different. Every synoptic setup is different. Every meso & micro setup is different & will affect tropical systems differently. And, just like the argument that the Saffir-Simpson scale doesn’t necessarily explain the dangers associated with each particular... 7/

...storm, we ought to strive to improve the messaging of the forecast track. Saying things like “our confidence in the track is low” or “is at 3 out of 10, so not very high” would go a long way towards managing expectations & explaining that aspect of the forecast. 8/

For example, messaging for #Laura might include the following for Galveston at 11pmCT tonight:
• Forecast track confidence: low, ~3/10
• Probability of landfall within 50 miles of downtown Galveston: 30%, up from 17% yesterday

... or something similar. 9/

The track for #Laura is going to change again, & when it does, the official messaging should include some level of confidence in the track. It might be that models are not really tightly clustered & therefore offer low confidence in the forecast track. That would be helpful. 10/

Not trying to bad-mouth anyone or bash any of the current systems in place. I know that public messaging has always been a critical factor & is always carefully reviewed & improved. 11/

I know NHC, Hurricane Forecast Improvement Project (HFIP), & everyone in the weather community is striving to improve the accuracy & reliability of forecasts as well as the messaging. Just offering my thoughts & opinions, hoping to further improve messaging going forward. End/