WILMINGTON — As Hurricane Irma — the largest and most powerful so far during the 2017 hurricane season — made its push toward landfall in the United States last week, meteorologists made predictions both laden with caveats, and that changed often, sometimes in just a matter of hours.
The latest: Hurricane Irma on westerly path, North Carolina cautions residents of flooding, mudslides and rip-tides
Irma was initially anticipated to cut across Florida, steaming away across the Gulf of Mexico. But, by Wednesday, its trajectory had changed. By then, the massive storm was expected to make landfall in the Florida Keys in the early hours of Sunday morning, before swinging north to head up the East coast.
Where it would head after that soon became the focus.
For a while, the Cape Fear Region was told to brace for impact. A day later, on Friday afternoon, Governor Roy Cooper, who had already called a state of emergency and activated the National Guard, said the storm was going to head west, hitting western North Carolina. He adjusted the state’s emergency services in response.
But, by Saturday, the storm was expected to only partially have an impact on the Tar Heel state after it eventually followed a path even more toward the west.
So, what changed?
The problem with these predictions, according to National Weather Service Meteorologist Steve Pfaff, is the unpredictable nature of hurricanes.
“Areas of low pressure, like a hurricane, they’re not like magnetic’s, where positive and negative magnets attract each other. Areas of high pressure and low pressure pretty much deflect away from each other,” Pfaff said on Thursday, when the storm was believed coming toward the Carolinas, but without absolute certainty. “So, areas of low pressure like a hurricane, will seek out other areas of low pressure.
“With Irma, what’s happening is you have the western periphery of the Bermuda high, which had been anchored in place to its north kind of steering it westward,” he said. “But, as it gets toward the western periphery of that area of high pressure, it’s going to want to make its turn.”
Pfaff said that a number of other aspects come into play too, when predicting the path of hurricanes. They include troughs in the upper level of the atmosphere, which can “get hold of the storm,” steering it as a cold front moves into the area. The same goes for Hurricane Jose, which is closely following Irma’s path through the Atlantic.
“This has been an extremely difficult forecast, with significant impacts depending on the scenario that does evolve,” he said.
Modeling a storm – from European to Spaghetti
This is where weather models come in. These computer simulations allow forecasters to make predictions on the path of the storm, using current conditions and observations to track systems from three to five days out.
“The National Weather Service uses a variety of weather models when we look at and assess the atmosphere, especially with a storm like Hurricane Irma,” Pfaff said. “There are models, like the European model, and the GFS, which is the U.S. model, and they’re all run approximately four times a day.”
Pfaff said that there are hundreds of models, all based on different physics packages. These packages enable forecasters to simulate various weather conditions, allowing the computer to show variations in weather as time goes on, producing the traditional “cone,” or “spaghetti” models.
When using these products, meteorologists look for consistency between various models. This allows them to predict the most likely path, often blending projections for a more complete picture.
But, according to Pfaff, these models have certain strengths and weaknesses. Forecasters rely heavily on physical observations that include information from hurricane hunters, a buoy network, observation stations, Doppler radar and satellites.
“So, modeling is one thing, but observations are another, they need to work hand in hand. We wouldn’t have the ability to do what we do if it wasn’t for both of those.”
“So, modeling is one thing, but observations are another, they need to work hand in hand,” he said. “We wouldn’t have the ability to do what we do if it wasn’t for both of those.
“I think an important thing that people don’t think about, is the fact that you can have the best model in the world, but, if you don’t give it a spot to start with observations, if you don’t provide observations to feed into the model, then you’re pretty much putting garbage into it, and you’re going to get garbage out of it.”
Pfaff describes the process of building a forecast out of these models as a “neural network” of sorts, with entirely separate branches coming together to form a whole.
“It’s not just sitting here coming up with a forecast, it’s a lot more involved,” he said. “None of this could happen without the observations.”
‘The double-edged sword of social media’
With the popularity of social media, many folks turn to their favorite sites for information on impending storms. While these can be beneficial for rapidly spreading information, as well for as gathering observations from the ground, there is a downside as well.
“The problem is, you have people who have no background on these model biases and don’t know what to look for; consistency, continuity, those sorts of things,” he said. “So, you’ve got people who have access, who are making their own predictions, without any idea what they’re looking at.
“With the hundreds of models that are out there, there’s going to be an outlier that really does something that is an eye grabber, impact causing ‘if it were to come to fruition kind of thing,’ he said. “But, that’s just one, when the other 99 models are showing consistency, which is key.”
Often, this is the information that gets spread, painting a picture of disaster and causing panic, when there should be reason. Pfaff said that with Irma, it had gotten out of hand.
“These people who are doing this are showing potentially the least likely scenario, getting people in an uproar, and possibly causing them to make the wrong decision, based on misinformation, the fake-weather type of information,” he said. “And that’s run rampant with this storm.”
— NWS (@NWS) September 1, 2017
From the talk of “category 6 and 7 storms,” to people illicitly using the logos of legitimate weather organizations, Pfaff said your best chance of getting real information, is through professional outlets like the National Weather Service, Emergency Management, FEMA or the Red Cross.
“People shouldn’t be basing decisions that include the protection of life and property on social media, with something that has no credible source behind it …”
“If it’s coming from, ‘hey, look what I saw,’ and it’s coming from ‘Jimbo’s Weather House’ or something like that, you’ve got to take it with a grain of salt,” he said. “People shouldn’t be basing decisions that include the protection of life and property on social media, with something that has no credible source behind it, it’s troublesome, like ‘War of the Worlds’ all over again.”
Pfaff said that while the National Weather Service and others like them work to create a realistic picture of hurricanes, the spreading of false information across social media can be detrimental to everyone.
“It’s fairly malicious, and it can impact the unified voice we need before and during a storm, and even after the fact with rumors and things,” he said. “This is the double-edged sword of social media. This has probably been one of the most prolific examples of the spreading of false information, I hope we learn from this and don’t make it worse.”