The Challenge of Explaining Joaquin’s Complexity

The hurricane intensified overnight Wednesday, and that has continued Thursday


BREAKING UPDATE: Hurricane Joaquin has strengthened into a Category 4 storm. Click here for more on the National Hurricane Center's update.

Earlier story is below:

Those who pay attention know the routine well - a storm with a name enters the Western Atlantic. Invariably, at least one or more of hundreds of various available weather forecasting solutions will indicate a potential strike to the United States, and from that point forward, it's off to the races.

Broadcast media lights up with tracks and cones from the National Hurricane Center, "spaghetti" is thrown all over our screens, and a multitude of varying and vastly different available numerical weather predictions suddenly and magically are boiled down in to just two - "the European" and "the American" model. Sometimes, "the European" and "the American GFS" if the presenter is feeling especially bold. Social media tweets at us, feeding our faces with books worth of information, images and unfortunately, sometimes, ignorance, depending on the source.

This circus of weather malady plays out repeatedly, tweaked and updated with each computer run and each issued update from the National Hurricane Center, adding to the hope that perhaps from four days in advance, the key piece of information will come forward that makes the forecast suddenly crystal clear.


The uncomfortable reality in this presentation game is, behind all of the showmanship, there is often a desperate desire to marry a complex science to a simplified presentation. The challenge is this: the modern media environment encourages the informer to either fit this complex science into 2-3 minutes of video, 140 typed characters, or a status update that won't bore a casually scrolling reader on a smartphone bathroom break.  

The result in a low predictability situation, understandably, is desperation. If one cannot successfully explain the complexities of the atmosphere and earth system interactions in the space allotted, perhaps one can simply display two mismatched scenarios in a floundering attempt to illustrate either just how complicated the situation is, or, in more self-serving exhibits, just how difficult the forecast is for the meteorologist.

Though the above description fits impending blizzards, nor'easters, and sometimes severe weather outbreaks, the storm at hand is Joaquin - a formidable, Category Three hurricane churning over the Bahamas.  

Within 48 hours, forecasts for Joaquin have been reliable with regard to track - beyond that two day window, a wide array of potential scenarios is posited. Let's be honest - this is nothing outside the ordinary in the world of meteorology, even Bill Belichick could tell you that... when the New England Patriots' head coach proclaimed the meteorologists hadn't gotten one forecast right last season, I agreed.  

That's not because the technology, or the wherewithal, or the dedication isn't there - it's because of over-extension: trying to deliver more than what the science and the technology can reasonably provide. Timing of rain for Sunday's game from the Monday prior? Tuesday? Wednesday? Time and again, our best verification for precipitation onset will come within 48 hours, and becomes more precise within 30 hours. That means until Friday night or Saturday morning, Bill shouldn't interpret the forecast as anything more than "it could rain."  

Daring talk from a meteorologist who offers New England's only 10-day forecast, but here's why: to indicate temperature patterns or precipitation patterns is a very different undertaking than nailing the hour raindrops move into Boston. Even temperature and precipitation patterns, though predictable enough to secure NECN a 5.5 degree error at Day 10, can go all wrong several days out - my 10-day last week, one week prior to this week's flooding rains, had sunshine. Sunny turned into a flood.  

That said, thankfully such an about-face for a large-scale pattern is the rare exception, rather than the rule. The same could be said about knowledge of storm paths and tracks - the key came from Kenny Rogers, decades ago. Know when to hold 'em, know when to fold 'em. Know when to walk away, know when to run.

So let's dispense with the circus and cut to the chase: Will Joaquin hit the United States?

Simple question. So, here's the quick, bathroom break answer: Nobody knows for sure. Check back later.

Here's the more scientific answer...

Hurricane Joaquin intensified overnight Wednesday, and that intensification process has continued Thursday. Thriving over warm ocean waters in the Bahamas and enjoying an environment of low wind shear - a Tropical Atlantic condition hard to find in an El Nino year - the storm organized around its core and cranked up to Category Three, likely to max out at Category Four on the Saffir-Simpson scale that maxes out at five.  

It really was a matter of Joaquin finding the right place at the right time, and that's evident in this map of wind shear - higher values indicate more wind shear, or changing wind speed and direction with height, which is damaging to a tropical system. Notice the high wind shear values all around the storm... but not over it:


The slow movement of the storm has come in the absence of strong steering winds, but soon enough, there'll be plenty of wind approaching that can transport Joaquin. This weekend, a strong disturbance at the jet stream level will dig into the Southeastern United States - keep in mind the jet stream is the fast river of air, high in the sky, that steers storm systems. So, a big dip in the jet stream over the Southeast leaves the wind blowing from south to north near the coastline, which ensures Joaquin will move north with time:


There are two other important features involved in the determining the future of Joaquin, however - a building high pressure "ridge" over the Western Atlantic and a small disturbance moving around the edge of that high pressure dome - the old remnant of what was once Tropical Storm Ida.  

High pressure, as you hear in my daily weathercasts, is a dome of fair weather... so to have one strengthening over the Atlantic really discourages an eastward track, and this is what many guidance products that have been taking Joaquin into the Mid-Atlantic are keying in on - for a storm to penetrate a building area of high pressure would take some magic. That magic, however, may be available as Ida's remnant.  

In the image below - a plot of mid-level atmospheric energy, Joaquin is the purple/pink bundle of energy, and Ida's remnant is the smaller, weaker energy to the northeast of the storm as of Saturday morning.


The timing of this disturbance is uncanny - it opens the possibility of weakening the bottom of the building high pressure ridge just enough, and for just long enough, to allow a possible northeast escape of Joaquin. So now you're talking about trying to accurately time and gauge the strength of:  

  1. A building Western Atlantic high pressure dome that hasn't built yet;
  2. Joaquin;
  3. The leftover energy moving west across the Atlantic.

Any of the above three important parameters can change, and depending upon how heavily any or all of them are weighed by various guidance, you're either going to get a storm that slides northwest into the coastline, or is able to escape northeast.  

That would be why you see guidance broken into those two, distinct camps:


One thing you don't see in these solutions is a strike to New England or New York City. So, why, then, does the National Hurricane Center track look like this?


The answer comes back to a high-stakes game - one different from the media meteorologist's desperation of trying to convey complexity through simplicity, but rather, a game of verification. Any scientific entity is held accountable by how well they perform, and how well you perform in science and particularly forecasting, is determined by the numbers.

What was your average error? In the case of the National Hurricane Center, what was the average error in distance for track forecasts, or speed for intensity forecasts? If they bet the farm on a left track into the coast, the public rises into preparatory mode. On top of that, if the storm ends up going right and out to sea, error is huge. Similarly, a bank on a track out to sea would lead to huge error, and a public with less time to prepare if the storm goes left.  

So the best bet for both verification score and public readiness is to keep the official forecast track in the middle of the potential solutions - technically, in a position where very few guidance products actually are tracking the storm and along a path Joaquin seems unlikely to actually track, but a track that will limit potential error, and buy time before either raising public alarm or taking residents off-guard.  

It doesn't hurt, too, that this track doesn't actually predict landfall for anyone yet - it gives time to change the track and include landfall, or keep it out to sea.

It's for this reason that it is critically important to always pay attention to the cone of probability associated with the forecasts, not the actual track. Not because the cone shows the National Hurricane Center meteorologists have no idea what may happen (they're excellent forecasters) and not because the cone illustrates who may be impacted on the edges of the storm (it doesn't), but rather because the actual track may very well be in the middle of two possible solutions, and technically in an area of reduced likelihood to occur.

Until we start conveying probability in more effective - and for media, more efficient and concise - ways, however, this is the most effective, mass-distributed way to display the possibility.

Having said all of this, surely one would imagine that a track in the middle of available guidance certainly is a "possibility."  That said, in its current position and motion, the most similar storms in history have either gone into North Carolina before meandering up the coast, or have missed out to sea.  

There is, of course, a first time for everything, but historically, of all similar storms, 28 percent have made landfall in North Carolina, 48 percent have missed the United States entirely, and 16 percent have made landfall in New England.  

Add to that the two camps of solutions explained above, and the weather setup described, and you have more indications of Joaquin having to take one path or the other, left or right. That said, for now, the middle of the potential solutions is the track opted for by the National Hurricane Center, and the probability of it verifying exactly may be low, but admittedly is existent.

I could have made this much simpler and shown you the American and European model. But I wouldn't have been able to sleep tonight.

Contact Us