The Hurricane Center’s forecast features the “Cone of Uncertainty.” The name describes it fittingly, but it’s still a dumb name because everyone immediately thinks of Get Smart.

Cred buster!

The cone starts narrow and widens over time. The idea is the farther out a forecast goes the more fudge factor is necessary. With Hurricane Earl as close as it is we are now on the outside of the narrow end of the cone. There is nearly no chance of Earl hitting Connecticut directly.

My friend Ryan who forecasts for Channel 30 tweeted:

Tomorrow evening will look like any other “rainy night”. I don’t expect any damage or anything more than isolated power problems.

He very well could be right though he is a shade more optimistic than I am. A nasty rainy night sounds more reasonable to me, but we are separated by shades. I do expect there will be some gusty winds especially in Southeastern Connecticut. Isolated power problems? Absolutely.

In any event Ryan and I and every other meteorologist around has done their best to allay fears. There may be people at the TV stations who’d like the storm’s impact perceived larger than it is. None of them work in weather.

I got an email this evening from Bob, a writer friend in Guilford.

Your blog pics of that storm, and your awestruck description of it, seriously scare the hell out of me.

Is that the impression I left? Though most casual viewers only remember the big storms I look at them all–even those six hour wonders that get named then deteriorate in under a day! Hurricane Earl was (it isn’t anymore) an exceptional work of nature. It had everything you should fear from a hurricane except a threatening path.

Now the waiting begins. The forecast will hardly change unless Earl totally flies by without saying hello.

There is little as satisfying as nailing the forecast. You probably don’t remember the storms I downplay as much as the ones forecast to strike. You would if I was wrong!

Over the past few minutes I’ve looked at loops from the WHRF, GFS and NAM models. They’re all slightly different though very similar. The worst of the three lives up to my earlier forecasts. The others bring less.

It’s possible the storm will pass in such a way half the state doesn’t even realize it was there! That would make me happy.

My website traffic’s up. I would guess you’re here wondering what I think of Earl? I’m in awe of this storm.

When the satellite map appeared on my screen tonight I marveled at the natural beauty of Hurricane Earl. Not all hurricanes are alike. Earl is classic.

Tonight Earl is exquisitely curved. The eye, 30 nautical miles in diameter, is nearly round. Earl is undisturbed, gorging on energy transferred from the warm waters below.

Few storms look like this. Most have faults or flaws. There’s a reason not every storm is as strong as Earl. A lot of things have to fall into place. It seldom happens.

At some point Earl will interact with land or colder water or the strong westerlies still to come. He will weaken.

I don’t know everything, but I’ve watched a lot of these storms. I am very intellectually curious in matters of science and technology.

Often during storms I chat with my friend Bob down in Tallahassee. He is one of a handful of the brightest minds in this field. Our conversations often center around interesting and esoteric observations. It’s stuff almost no one looks at. We talk about buoy readings a lot. Sometime we rate the hurricane forecasters at NHC as if they were eligible to be drafted onto some “fantasy meteorology” team.

“2 min,” he’ll type and two minutes later a link arrives. At the other end a beautifully rendered map or chart created on-the-spot to illustrate a point. Few people think this way. Fewer have this skill. It’s sort of amazing.

It’s funny how some viewers interpret what I’m doing. This was blogged this evening:

Our local meteorologist Geoff Fox says Earl should not be that much of a threat to the Connecticut coast, but you can hear the excitement in his voice. You just know he’s waiting for the big one.

Really? I just want to grit my teeth and let out a small scream. Everything I’ve done has been to try and balance what we’ll see with what at the moment is a freak of nature! I don’t want to see the big one or even the medium one. I have too much respect… too much fear.

I often get emails and phone calls trying to sell me on a more exciting forecast. Their logic always has multiple ifs. It can come true but it’s unlikely and deserves to be treated that way. It’s easy to make outlandish forecasts when you’ve got no skin in the game.

Ann Nyberg came to the Weather Center tonight. She interviewed me for her website.

The computers we use at work depend on specialized software which reads weather data and formats it so it can be plotted or contoured on-the-air. It’s helpful and powerful, but because there are so many different types of data and different ways to show things there are dozens… maybe hundreds of individual settings.

I spent dinner at my desk working on a few graphics I want to show tonight.

The models remain close to where they were. There are outliers which are more pessimistic. I will anguish over that through the night. I haven’t changed much in the past two days. Rotten–yes. Direct hit–no

You’ll be glad to know most of the dependable guidance continues to show Hurricane Earl south of Long Island then out-to-sea. Some of the models are more westerly than others, but for most we get a rotten day with some limbs/power down and enough rain to gum things up.

This time before a storm I get emails and other messages asking if the author can do something on a given day. I don’t answer those. I don’t want someone to get into an accident and then, whether my fault or not, say “I was unsure, so I asked Geoff if I could go.”

The answer is always “No!” You cannot.

He’s not requesting advice. He’s asking permission!Not granted.

Sorry.

I have worked hard to allay fears, to reassure the viewers. Now that scenario better come true. Being wrong here would be pretty awful for all concerned!

I’ll be sweating all the models and anything else I can get my hands on. Now that Earl is closer there will be shorter term models available. More confusion!

So far the track has stayed reasonably in line. This would seem a logical path based on past storms.

Most likely Friday on TV you’ll have a good view as Earl slides by. He should be within radar range. We’ll have hi-res imagery to show. If you enjoy this sort of thing you”ll have fun, but from afar.

I have been forecasting in Connecticut over 26 years. The pressure to find the correct forecast answer has never seemed greater.

Nothing’s changed!

I guess that’s a good thing. The more consistent the forecast over time the more likely it is to come true.

The HWRFTM is a high resolution coupled air-sea-land prediction model with a movable nested grid and advanced physics for high resolution

Like you didn’t already know!

This is a fairly recent model. It’s made especially for hurricanes. Hurricanes need an extremely fine mesh or high resolution because so much is going on in a small space!

Most computer guidance is optimized for larger scale events like the type of fall-through-spring storms we get in Connecticut.

Forecasting has really improved in my 28 years in weather. I wish I was better at it. Mistakes are never fun.

It’s not that the forecasting isn’t good–it needs to be great. Mistakes have a huge downside. Both over and underforecasting have costly consequences.

It’s in.

A quick peek shows it still west of the official forecast. I lean west too.

Interesting. Bothersome. Inconveniencing. Not tragic.

There is an unfortunate inconsistency to this data. In nearly every case Earl tracked west of the forecast!

If this remains the case Earl snuggles a little closer to Connecticut than has been said. Not good!

The Hurricane Center is populated with some of the smartest minds in tropical weather. Hopefully they’ll catch on or there’s a method to their madness I just don’t see.

Radar — that was my first step this morning. I needed to check the radar from Saint Maarten. As I type this the southern eyewall of Hurricane Earl is touching the coast of Anguilla in the Caribbean. The eye is well defined on radar.

The Hurricane Center says the top winds are 125 mph. If that’s true they’re in an extremely small area. I usually feel NHC’s estimated maximum wind is higher than warranted. It’s all academic. You don’t need 125 mph to rip a Caribbean island to shreds!

There have been no official observations from the Anguilla airport since yesterday afternoon. In Saint Maarten just to the south winds are sustained at 33 mph with gusts to 53 mph. Anguilla is getting it worse.

At St. Thomas the wind is gusting to 49 mph. Earl isn’t there yet. The Virgin Islands and Puerto Rico get their run-in later this afternoon.

These storms tend to wobble. They are not fully symmetrical. There is uneven friction from nearby landmasses and the interaction with other weather systems. This current path is north of where I thought it would be as recently as last night, but reasonably close.

Hurricane Earl is definitely a threat to the US East Coast. Will it hit? Too early to say, but it’s certainly enough of a possibility that I’m watching its every move.

I am surprised by the Hurricane Center’s forecast. The ‘out days’ center of the track is well offshore though all of the East Coast from the Maryland shore to the Canadian Maritimes are within the cone of uncertainty. How helpful is a forecast when it has to alert Washington, Baltimore, Philadelphia, New York and Boston? There’s lots of fudge factor still at work.

Based on the GFS past few runs I’d shift the track even farther west (left) than they have. I’m sure there’s a little “better safe than sorry” in my thoughts as well.

From the latest Hurricane Center discussion:

EARL IS FORECAST TO TURN NORTHWARD...THEN NORTHEASTWARD AHEAD OF A MID-LATITUDE TROUGH THAT MOVES ACROSS THE GREAT LAKES AND INTO THE NORTHEAST UNITED STATES IN 4-5 DAYS. THE TRACK GUIDANCE HAS SHIFTED WESTWARD AGAIN AND THE OFFICIAL FORECAST HAS BEEN ADJUSTED IN THAT DIRECTION. THE UPDATED TRACK FORECAST IS NEAR THE MIDDLE OF THE TIGHTLY CLUSTERED GUIDANCE.

THIS IS A GOOD TIME TO REMIND EVERYONE THAT NHC AVERAGE TRACK FORECAST ERRORS ARE 200 TO 300 MILES AT DAYS 4 AND 5. GIVEN THIS UNCERTAINTY...IT IS TOO SOON TO DETERMINE WHAT PORTION OF THE U.S. EAST COAST MIGHT SEE DIRECT IMPACTS FROM EARL.

There’s a full workweek of this still to come! I will become enveloped in the storm. Actually I already have.

I captured the radar image above a little before midnight. You’re looking at Hurricane Earl on the heretofore unknown St Maarten radar. St Maarten is in the middle. On the left is Puerto Rico. The Virgin Islands sit between the two.

Hurricanes are the Gisele Bundchen of meteorology! They are stared at and studied. They are incessantly photographed. They are penetrated by flying meteo labs. All the stops are pulled for hurricanes.

Hurricanes obey the laws of physics. All weather does. Still, hurricanes are distinct.

At its simplest a hurricane moves heat from the equator where there’s an overabundance to the poles where there’s a paucity. The Earth appreciates it. This would be a very different world without hurricanes to transfer heat.

Wind rushes inward toward the hurricane’s centerr. Yes–hurricane’s suck!

As the wind moves friction turns it right. That rightward turn translates into the storm’s counterclockwise spin. As the wind gets closer to the center it gets faster until it finally reaches the eye where it is carried aloft!

The base of a hurricane sucks. The top of a hurricane blows.

A BROAD UPPER-LEVEL TROUGH IS CURRENTLY SITUATED TO THE NORTHWEST OF EARL. GLOBAL MODELS ARE IN GOOD AGREEMENT THAT THIS TROUGH WILL SHIFT WESTWARD AND WEAKEN. THIS SHOULD CREATE A ENVIRONMENT CONDUCIVE FOR INTENSIFICATION AND THE OFFICIAL FORECAST...LIKE THE PREVIOUS ONE...SHOWS EARL BECOMING A MAJOR HURRICANE OVER THE NEXT DAY OR SO. THE NHC WIND SPEED FORECAST IS ABOVE THE INTENSITY MODEL CONSENSUS BUT BELOW THE LATEST GFDL HURRICANE MODEL GUIDANCE.

Pretty standard stuff from Dr. Richard Pasch at the Hurricane Center. The storm is his responsibility tonight.

I keep looking at the long range guidance because this storm has been modeled to pass very close to Connecticut. The 00z (8:00 PM EDT) run of the GFS isn’t in yet, but yesterday’s 18z shows Earl brushing past Cape Hatteras then hooking northeasterly south of Long Island and possibly striking Cape Cod.

Right now on St Maarten the wind is out of the northnortheast at 14 mph. There’s a thunderstorm in progress–not unusual on a tropical island. Tomorrow will be different.

When Hurricane Earl hits (or goes a little south of the island) every single person on St. Maarten will feel threatened! No one goes through a hurricane in any kind of shelter without fear. It’s well founded.