He studied the climate maps and picked Dallas as their destination, because historically it had excellent chances for clear skies in early April.
“Sometimes the weather gods like to laugh at you,” said Epstein, 58, whose mood this week has trended upward thanks to modest improvements in the still-iffy Dallas forecast since Monday.
Nationwide, the eclipse forecast is rather cloudy — as in unclear, fuzzy, murky, but also as in literally full of obnoxious clouds that could obscure this grand spectacle.
A total eclipse is both astronomically predictable and meteorologically fickle. The experts know exactly when the moon will completely cover the sun. They can’t predict if human beings on the ground will be able to see it happen.
And while the moon needs nearly three hours to eclipse the sun, the exquisitely weird period of totality — when the sun is completely obscured but for its entrancing atmosphere, and bright stars and planets pop out in the darkened sky — lasts only a few minutes. People in the contiguous United States will not have another chance to see such a thing for 20 years.
With less than a week before the April eclipse, New England looks like it has the best chance for perfect weather. Mexico is also sitting pretty. But these are anxious times for eclipse aficionados in the 2,000 miles in between.
“I’m going to cross my fingers,” said astrophysicist Adam Frank of the University of Rochester, noting that his city in Upstate New York experiences lake-effect weather and is often cloudy in spring. He will stay put in Rochester no matter what, because he’s committed to giving televised eclipse commentary.
“I have high hopes, low expectations,” he said.
The tricky work of predicting clouds
Cloud forecasts are shot through with ambiguities, uncertainties and hard-to-fathom probabilities. It’s fair to ask: What exactly does “cloudy” mean?
Clouds form when air rises and there’s enough moisture in the air. Lower pressure, which allows air to more easily rise, often generates clouds. Higher pressure, which prevents air from rising, tends to promote sunnier skies.
Some weather systems create large areas of rising, moist air, leading to large areas of solid cloud cover. Other systems only generate pockets of rising air here and there, with some pockets moist enough to make clouds and others not. These clouds — both their location and timing — are much harder to predict, especially more than a day or two ahead of time.
What people really want to know is whether it will be cloudy over their exact location during the exact minutes and hours of Monday’s eclipse. Models, however, can’t accurately predict clouds with that kind of precision this far ahead of time. Instead, they forecast the percentage of the sky that may be covered by clouds at three-hour intervals.
With that in mind, eclipse-goers in the path of totality should probably be worried about any forecast for over 60 percent cloud coverage, and cautiously optimistic about any forecast for less than 30 percent. In between, the situation is pretty fuzzy.
The kind of clouds also matters. High clouds are made of ice crystals, while lower clouds are made of water droplets. High, thin clouds won’t completely obscure the eclipse, but low, dense, dark sun-blotting clouds could spoil the show.
Adding to the anxiety, spring is a particularly tough time of the year to predict cloud cover.
For one, the lingering chill from winter can lead to cool, moist air that creates overnight clouds, while daytime sun and warmth aren’t yet strong enough to dissipate the clouds as quickly as forecast models may anticipate. And the jet stream tends to move weather systems along more slowly in the spring than in winter. That can also lead to cloud cover that’s slower to clear out than predicted.
Yet another variable is the direct effect of the eclipse. The air temperature drops dramatically as the sun is obscured and it ceases to heat the ground, causing air to stop rising. One potential effect, noted by many eclipse-goers, is the creation of an “eclipse hole” in the cloud cover.
This doesn’t happen for every kind of cloud, however. Low-level cumulus clouds — those beguiling, puffy cotton balls — are most likely to dissipate during an eclipse, according to a paper published earlier this year in the journal Communications Earth & Environment.
Where the forecast stands for Monday
Models are currently in pretty good agreement for April 8, showing lower pressure and a cold front from Texas into Arkansas, and then higher pressure heading to the northeast. So that’s most encouraging for New York, Vermont and Maine, and least encouraging for Texas and Arkansas.
There are two caveats, though. First, we’re five days away. At that range, things can still change, no matter how confident the forecast may seem now. Thursday or Friday is when people should start taking the cloud forecast more seriously. That said, cloud forecasts can sometimes be a challenge even on the same day.
Second, just because the models may be right on the overall weather pattern, that doesn’t mean they are right on the timing. At this range, the models could still be off by perhaps 12 to 24 hours in either direction. If that’s the case, then it’s not impossible for the cloud forecast to change significantly for the better or worse, depending on location.
Epstein, the real estate developer, said his friends have assured him that they will have a good time even if the skies over Dallas don’t cooperate. Still, when the forecast was particularly bleak eight days before the eclipse, he felt awful.
“I know I’m not responsible for the weather, but nevertheless, a lot of people had put their trust into the concept that this was going to be a great event,” he said. “To think that it was all going for naught was pretty upsetting.”
At the Dallas Arboretum, the eclipse will be celebrated with three days of events, and organizers expect 10,000 people, as well as NASA scientists and national news media, on Monday. But arboretum vice president for marketing Terry Lendecker said Tuesday that she isn’t worried about the weather.
“They’re forecasting 30 percent chance of rain. In Texas, that really doesn’t mean anything. It changes so rapidly all the time,” Lendecker said. “While we watch the weather, mainly due to safety reasons for our guests, the show has to go on when you’re an outdoor venue.”
And, she added, “it’ll be a beautiful day in the garden, regardless.”
