Get PJ Media on your Apple

Weather Nerd

For about a week-and-a-half now, meteorologists and weatherbloggers, myself included, have been watching a potential Cape Verde hurricane that some computer models were forecasting way back when it was just a vigorous batch of thunderstorms over the African continent, many days away from the warm ocean waters that fuel tropical cyclones. Countless grains of salt were consumed, especially given that the previous hypothetical “proto-Dorian,” forecasted for mid-July by some models — you might recall me mentioning it in a Chantal update — had proved to be a total dud. But when the Canadian model (CMC) joined the premier American model (GFS) on July 16 in predicting a late-month wave with significant potential for development, things started to get interesting.

Dorian as a wave

By July 20, as the vigorous African wave (represented by the green line above) neared the Atlantic Ocean, the model runs were becoming more and more consistent in predicting storm formation (or “cyclogenesis,” to use the fancy #science term). But then, even as the wave emerged over water late on July 21, and was officially designated as “Invest 98L” on Monday (July 22), the models backed off and stopped predicting development.

Invest 98L now looked like a marginal threat — an unlikely “proto-Dorian,” even after all that computer-model hype the previous week. 98L, however, was apparently committed to living up to the hype. Once it got over the unseasonably warm waters just off Africa, 98L did unexpectedly well. Its odds of development steadily increased, and by late Tuesday, it was a “WINO” (Wave In Name Only) on the verge of Tropical Depression status. Early yesterday (Wednesday) morning, 98L officially became Tropical Depression 4, and six hours later, it was Tropical Storm Dorian. (Not DeLorean!)

At the risk of overly anthropomorphizing tropical systems, it sometimes feels like certain storms are “survivors,” tending to consistently outperform the expectations of computer models and forecasters, while other storms are consistent underperformers, never quite “getting their acts together” even when the meteorological conditions suggest that they “should.” Thus far, Dorian seems to fall into the “survivor” category. Even after it defied the odds by becoming a depression and then a storm yesterday, the naysayers were everywhere, noting Dorian’s seemingly “dim future” thanks to cool sea-surface temperatures ahead. The conventional wisdom was that “significant strengthening [was] unlikely,” and that total dissipation was entirely possible. I was one of the naysayers:

I intended to post a “Weather Nerd” update yesterday, and I had the lede pre-written in my head. Referencing my first Chantal update, I was going to say: “Stop me if you’ve heard this one before. An unseasonably early ‘Cape Verde’ tropical storm has formed in the eastern Atlantic, and could eventually pose a threat to the southeastern U.S. — if it lasts that long.”

Alas, politics took precedence over weather in my blogging endeavors yesterday, and because of the time I spent explaining why I support Justin Amash and oppose lawless NSA/FISA spying, I never got around to publishing that post. Then again, perhaps that’s just as well, because 24 hours later, the story looks quite different.

As Levi Cowan (@TropicalTidbits) explained in an excellent video update last night, Dorian held together remarkably well yesterday, and it has continued to do so overnight and into today, even as it traverses the coolest waters it will face in its life as a tropical cyclone. The storm shows no sign of significant weakening; indeed, its maximum sustained winds ramped up to 60 mph yesterday, making it a relatively strong tropical storm, and they’ve remained steady ever since. It’s possible the National Hurricane Center — which is relying primarily on satellite estimates right now, as Dorian is too far away for Hurricane Hunter reconnaisance flights, and will remain so until Sunday — is being a little bit generous with those winds, but what’s clear is that Dorian is by no means falling apart, as might have been expected by now. Here’s how it looked this morning:

And here’s how it looks this afternoon:

Like I said, Dorian feels like a “survivor.” I know that’s unscientific, but I can’t tell you how many times I’ve seen storms persistently “live up” — or “live down” — to their seeming “character” as either overachievers or underachievers. Of course, I suppose there’s some confirmation bias at work: for every Irene (a bit of an underachiever) and Epsilon (epic overachiever and “survivor”), there are probably countless storms that one doesn’t remember as having a “character,” because no particular pattern emerges. But whatever. It may be silly, but I’ll confess I’m skeptical of those computer-model projections — even though one of them is the almighty Euro — which show Dorian significantly weakening or dissipating due to upcoming encounters with wind shear, dry air and the like. For instance:

I don’t buy it. I know there’s wind shear ahead, several days down the line. But Dorian is a geographically small storm with a well-developed, tightly-wound circulation, which may be a plus, as Joe Bastardi says: “This was supposed to fire up and start falling apart…I think its being compact is helping it beat models.” As for the dry air that’s supposed to impede it? Well, that seems to be shrinking. And again, my gut just tells me he won’t go quietly into the night. That’s why I lean, unscientifically, toward the more bullish model runs, like this one from last night:

Please don’t focus on the details of that particular track, though. Not only is it an old model run — and using an old supercomputer to run its projections, no less; the new one, this afternoon, spit out a more southerly track — but there’s just a huge amount of uncertainty about where this storm will ultimately go, a week to 10 days hence:

That’s why I urge people NOT TO #PANIC about individual computer model graphics like this one (today’s 06z GFS):

Again, DO NOT #PANIC!! It’s waaaaaay too early to get exercised about specific landfall scenarios. At this distance, you can expect the tracks to keep shifting around from model to model and from run to run. We really don’t have a remotely clear idea yet of where Dorian will go, and we won’t for a while yet. Everybody on the East Coast should keep a loose eye on it.

At this point, rather than specific landfall scenarios, the focus should be on big-picture trends, like the issue articulated this morning by Charles Fenwick:

Track forecast continues to appear to be simple for the forecast period [i.e., the next five days]; same course [as previously predicted], albeit slowing, puts Dorian N of PR by Monday night. It’s in the days over the horizon of the forecast period that the track forecast for Dorian becomes complicated. Still a split in how models depict Dorian several days out. Those that keep it a weak entity continue to move it westward. The GFS, on the other hand, keeps Dorian as an organized storm and turns it northwest *slowly* & moves it to the Carolinas. Speculative stuff.

Indeed. Dr. Jeff Masters has more.

Anyway, stay tuned. I’ll keep updating Dorian periodically here on Weather Nerd. For even more frequent updates, follow me on Twitter (@brendanloy). Another good resource for the latest information, in tweet form, is Amy Sweezey’s “Wx Tweeps” Twitter list.