A bit of a rant re: NHC overcautiousness.

[Cheeznado]

This will be a controversial opinion but so what, I am retired now so bombs away. 

I have been in the forecasting game for over 35 years, I was at TWC for over 30 and have been involved with hurricane forecasting for all of that time period. First off let me say that the mets at NHC work incredibly hard and are dedicated and professional public servants to which the nation owes a great debt. 

However, one tendency that I have noted since way back in the early 80s is their philosophy of "even if the odds are incredibly against a particular storm suddenly deviating from it's forecasted path, we must always err on the side of caution" and deliberately overdo watches/warnings even we know that chances of the storm affecting those areas are very small. Case in point is the current situation with Dorian. The model clustering is about as tight as it ever gets, and the official forecast is following that closely. The winds on this and most hurricanes are weaker on the west side of the storm, the Cman station on the west tip of Grand Bahama Island never had a sustained wind of hurricane force, in fact the highest was only 54 kts, even though this location is much closer to Dorian than Florida is ever going to be. Right now the highest wind gust anywhere in the FL coast right now is about 30 mph, and even if the storm does go NNW as forecast, why would any location in FL or for that matter GA have winds remotely close to hurricane force in gusts, much less sustained? This type of extreme caution is certain to lead to even more public skepticism over future hurricane forecasts and evacuation orders which would result in people staying put when they really should evacuate. My totally worthless and biased opinion....

[Windspeed]

First off, I want to say thank you for your service. Second, you worked as a professional in that environment and I very much respect your opinion. However, though I understand and feel where you are coming from, I do disagree.

 

I would say the NHC errs on the side of caution because they have major players to answer to all the way up to the highest levels of government. Their forecasts, even ones based on sound modeling ensemble consensus and synoptic reasoning could still fail. You can never be 100% certain especially dealing with TCs, more significantly, ones still at a lower latitude versus shoreline that is in close proximity. Sure, TCs are steered by large scale synoptic features and the NHC has become amazingly better at forecasting to a higher degree of accuracy as the seasons come and go. However, there still remains significant chance of error with stalled tropical cyclones in weak steering currents / flow.

 

Perhaps in the time period of weak flow, one large scale dynamic feature does not evolve as modeled. Even multi-day repeated modeling consensus. Perhaps an initial trough misses. Perhaps a second trailing trough lifts further NW not in the modeling consensus. At the chance of sounding like devil's advocate here, you have a hurricane still far enough south that easterlies could become the dominate flow again due to ridging reexpanding NW. Is that a modeled solution right now with Dorian, or even being something mildly resolved at all? No. Absolutely not. But could that be the case in operational runs 12 hours from now? What about 24 hrs? Even slight uncertainty with a system so close to the coast, you keep the public aware. People have mentioned Ophelia. But I will take this a step further.

 

How many hurricanes have we seen in the central Atlantic above 25°N be unanimously modeled to turn NE into the westerlies and yet end up lingering for over a week? There is typically one every three or four seasons that pulls off this Houdini act despite overwhelming modeling to the contrary. One trough misses it because one synoptic feature fails to evolve as modeled. Then a second trough misses. Suddenly the models begin rebuilding ridging, and here we go! That cyclone ends up tracking west again. Sometimes just a little longitude, sometimes even 5 to 10° before finally banking a weakness. Nothing much ever comes of it because it effects nothing but shipping interests and flight patterns. And really only some tropical meteorological enthusiast take note. But let's play that unlikely scenario near the Southeast CONUS. That would end up a catastrophic failure for a government agency where so many other government agencies and the public in general rely. It is their responsibility to see the unexpected and weigh it in doubt against consensus, just in case. Sure, overwhelming unlikely as that may seem, I am glad they account for it. We all realize atmospheric science in 2019 still has percentages of error and it is still as much dynamic and emperical as it is sound fundamental methodology in prognostication. You can never be 100% accurate, and even within great forecasting certainty and modeling support, surprises occur.

 

There is simply too much at stake to not account for the possibility of a stalled system not quite performing as you would like, much less accounting for the unexpected. Having said all of this, I am not suggesting Dorian won't do as modeled from here on out, though this prolonged stall seems to have been undermodeled somewhat. Hopefully it misses and everyone on the SE coast can get back on with their lives.

[Great Snow 1717]

Do you view Dorian as being a Catch 22 situation? I think the strength of Dorian and the proximity of Dorian to the FL/GA/SC led to the caution. At some point decisions have to be made. I'm sure the decision(s) to evacuate some areas where difficult to make.  How long do you think that the elected officials should have waited?

[Cheeznado]

Yesterday you can make an argument. But right now the warnings and watches in FL and GA are not even their own forecast cone!

[Windspeed]

Yesterday you can make an argument. But right now the warnings and watches in FL and GA are not even their own forecast cone!

You are correct that they are outside the cone. And when Dorian is moving on a steady NNW or W of N track verifying greater forecast certainty, I guarantee they will advise for those watches and warnings to be lifted. Until then, we're waiting on Dorian and the large scale features at play to do what they are modeled and forecasted to do.

[Cheeznado]

At 11, they STILL have the warnings even due W and SW of where the center is now. Their own definition of what a hurricane warning means: 

A Hurricane Warning means that hurricane conditions are expected
somewhere within the warning area.  Preparations to protect life and
property should be rushed to completion.

 

However, again their own product- the hurricane wind probabilities- show no more than a 20-30% chance of 74mph winds on the FL coast, is a 25% chance "expected"? Of course not.

Let's use an an analogy from another type of forecast. Would it be acceptable for a NWS office, say Boston, to issue a winter storm warning, but the next set of models have no snow at all so they remove the chance of snow from the forecast but keep the warning in place?

[bigtenfan]

I am in Boca Raton about a half mile form the ocean and we are still under a tropical storm warning and a hurricane watch. I certainly have no problem with either being issued when they were but at this point not having had any wind gust of even minimal TS strength much less hurricane strength I think that at least the hurricane watch could have been taken down and the TS warning downgraded to a TS watch by now.

 

[OSUmetstud]

2 hours ago, Cheeznado said:

At 11, they STILL have the warnings even due W and SW of where the center is now. Their own definition of what a hurricane warning means: 

A Hurricane Warning means that hurricane conditions are expected
somewhere within the warning area.  Preparations to protect life and
property should be rushed to completion.

 

However, again their own product- the hurricane wind probabilities- show no more than a 20-30% chance of 74mph winds on the FL coast, is a 25% chance "expected"? Of course not.

Let's use an an analogy from another type of forecast. Would it be acceptable for a NWS office, say Boston, to issue a winter storm warning, but the next set of models have no snow at all so they remove the chance of snow from the forecast but keep the warning in place?

Hurricane Warnings only verify a small percentage of the time. They warn larger stretches of the coast line to have sufficient lead time for preparations and account for track deviations. I think theyre verification stats are similar to tornado warnings compared to other warning products. 

Im guessing there is some sort of peristence bias with warnings in the sense that there not lifted until conditions markedly improve even though they would know they arent verifying. 

[OSUmetstud]

Also, they use the word "Hurricane Conditions" which is more encompassing than the strict wind definition. It includes thing like storm surge and large battering waves. 

[ohleary]

This is why they're cautious:

i.e. because people die.