Hurricane Matthew

If you live on the east coast of the US or the Bahamas, do not take your eyes off this! 

I've been following this hurricane closely and studying the maps at sites I referred you to in Advanced Hurricane Links last month. (In fact, I've let my fascination with it cut into my writing time, which must stop!)

Friday morning I saw it had intensified overnight and was not going to stop. I tweeted that it was scary and predicted to a friend IRL it'd be Cat 5 before bedtime. It was. I began worrying about my Florida fans and all the Floridians I didn't know, and everyone in Kingston Jamaica, who looked like they were in for terrible times.

I admit I was befuddled about the models, which had it turning north, and I continued to be befuddled Saturday morning, when I pulled this 700hPa wind map off earth.nullschool.net.

I could see the cyclonic and anticyclonic areas over the Ohio Valley and Atlantic, respectively, and the likely path north between them is obvious, a channel right up the coast as of that moment.  But my question--and I couldn't find an answer for it anywhere--was what made forecasters think the storm was going to turn north within minutes of when I took this screen cap? I couldn't see any path north for it to get to the northward channel. It was obviously being steered west.

Leaving the mystery alone, I forced myself away from researching the question to write, did my daily words, and by nightfall my time on Saturday, it was moving NW. One strange thing about this storm (and it's a weird one!) is that on satellite, you can see a big area of convection, bigger than the core of the hurricane, to its east. This happened because that side of the storm was, for a long time, over high mountains in South America, which changed the storm's dynamics. When it moves entirely away from SA, that huge batch of thunderstorms may be wrapped into the main circulation, or it might stay just like that (looks like eyes, doesn't it?)...until it hits Haiti.

Infrared view of Matthew 1140 UT, 2 October

Deforested, poor, and still not recovered from the big earthquake, Haiti will not fare well under that powerful bunch of storms. Mudslides and flooding will kill, and if the death toll is in the hundreds, I won't be surprised.

As to where it's going after it's done with Haiti and Cuba? Too much about the future is uncertain. How far north will a low pressure system now in Ohio move? How strong will the high ridge now building in the Atlantic become? My feeling all day yesterday was that the official track was a little too far east, that Florida and the Carolinas are still in danger. One model (and a good model, not one of the weird ones that is never right) has a major hurricane slamming into Maine in a week! So everyone on the East Coast of the continental US needs to watch this thing.

And remember, if you see a forecast for four days down the road and it isn't aimed at you, that's no guarantee you are safe. On average, that far in advance, they're wrong by 200 miles, which the NHC itself mentions in some of its updates. Weather is complicated, which makes long-range forecasting difficult. Get your supplies today, just in case. If the hurricane curves out to sea, you can be thankful, and the supplies will serve you well for the next storm, power outage, or earthquake, too.

ETA, Tuesday 3 Oct. I answered my own questions about the steering.

1) The turn north did not happen for 30 hours beyond when I looked at steering maps. By then, it made sense to me from looking at the maps. So some forecasts promised the turn too quickly.
2) I've read--but haven't thoroughly investigated--that which steering level to look at depends on the storm itself, barometric pressure, cloud heights, etc. I'll have to dig into that one and see....
3) The blog moderator at WU pointed out that all Northern Hemisphere storms want to "visit the North Pole." I thought about it for five seconds and realized why. I did a thought experiment of the sort that we did physically in junior high physics, imagining a drop of colored fluid on a spinning top, just "north" of the top's equator, and I could see it. This is, of course, another part of the Coriolis effect. If nothing else is steering a hurricane--or if all the forces that are trying to do so balance out, it'll move north. If it's caught in trade winds (as from Africa to the Caribbean), those will steer it. If there are high or low pressure areas, those can steer it. Otherwise, it does indeed want to "visit the North Pole."

..and the models and forecast tracks are inching west today, so that part of my guess did come true. If you're in Florida or the Carolinas, it's time to do early prep: take in the lawn furniture, clean out the freezer, put a bunch of bottles of water in there to freeze solid by Wednesday and keep it colder longer, fill up the spare gas cans on your way to/from work tomorrow, buy or charge batteries. As the forecasts narrow it down, you may need to do more. If they tell  you to evacuate, please, evacuate.