New England

Will New England see Northern Lights this week? Maybe not, after all

Some scientists believe that an auroral forecast for Maine, New Hampshire, Vermont and more than a dozen other U.S. states to see the light show Thursday is based on a miscalculation

NBC Universal, Inc.

There's been widespread reporting that the Northern Lights will be visible across much of the northern United States this week, including several states in New England, as well as further south in Maryland.

The Associated Press reported that the aurora borealis would be visible further south than usual because of an 11-year solar cycle that's expected to peak in 2024, citing an auroral activity forecast from the Geophysical Institute at the University of Alaska at Fairbanks.

But there's been some chatter among the scientific community that there may have been a miscalculation in the algorithm that led to an incorrect date being reported.

Error in Carrington Rotation calculation?

From what I’ve gathered, there’s a heartbeat to solar flares triggering geomagnetic storms that flare up into what we know as auroras. And it seems like the cycle, or rhythm, known as the Carrington Rotation, was miscalculated.

Here's how the Geophysical Institute explains it:

Just as Earth rotates on its axis making a complete rotation every 24 hours, the sun spins on an axis making a complete rotation in 27 days. Solar phenomena on the surface of the sun distribute high speed plasmas that often result in increased auroral activity on earth. These phenomena may be sunspots, coronal mass ejection, filaments, or a prominence. As the sun rotates, the solar phenomena and resulting areas of high-speed plasma are likely to reoccur every 27 days until the phenomena dissipate. Therefore, if there is high auroral activity today it’s possible that there will be high auroral activity again in 27 days. This 27 day rotation is called a Carrington Rotation.

It appears to me and other scientists, like Tamitha Skov, that this window for a rotation was this past Thursday, not the one coming up. She explains a little here — and also, this past Thursday’s viewing was a dud.

Forecast confidence

There are two points that I want to make here.

First, the AP and other articles that sourced the AP cited a University of Alaska at Fairbanks calculation that Kp values would near 6 on Thursday. The geomagnetic index known as Kp ranks auroral activity on a scale from 0 to 9, with 0 being not very active and 9 being bright and active.

Again, I think there were miscalculations on the University’s behalf. This was the forecast from their website Tuesday morning, with maximum Kp values of 4, meaning auroral viewing is much farther northward than initially stated, and where it’s typically visible.

A map showing an auroral activity forecast from July
University of Alaska

According to the NOAA’s Space Weather Center — yes, that entity exists — if Kp values were near 6, viewing would drop as far south as Idaho and New York, not Maryland, like the AP article suggests.

A second point that I think should be examined closer is the forecasting scope of auroras. More specifically, the confidence interval and lead time on when forecasts are issued — it’s fairly minimal.

The northern lights are a beautiful natural phenomenon, but do you know what causes them?

The Space Weather Center's forecast only goes out three days in advance. From what I can see, the AP article from late last week cited the University of Alaska at Fairbanks’ forecast that was nearly a week out, which is out of range and scope of solid forecasting.

The University even states:

Remote sensing spacecraft also monitor the sun for indications of impulsive eruptions that eject material toward Earth, or regions where continuous high-speed streams of material are escaping and heading Earthward. In either case, the travel time to Earth for such material leaving the sun is 1 to 3 days. This allows good forecasts to be made on these look-ahead time scales, based on events and conditions observed on the sun.

What would tip the scale is a substantial eruption and solar flare. But that isn’t the case.

We've reached out to the university to ask about the discussions over the forecast.

One dead giveaway

From what I’ve been able to gather, Alaska was one of the states listed that could see the aurora. 

There’s 24 hours of daylight in Alaska during the summer, so that’s the first clue. It has to be pitch black to see the aurora. Even light pollution from city lights is enough to offset and disrupt viewing.

In the same vein, because the aurora borealis is based on latitude (from the North Pole and southward), there are several states that are between New Hampshire and Maryland that were omitted from the list.

If the aurora were actually visible in Maryland as the article suggestions, Connecticut, Massachusetts, New Jersey and Pennsylvania should also be considered for viewing. They were not mentioned.

The full list was Alaska, Oregon, Washington, Idaho, Montana, Wyoming, North Dakota, South Dakota, Minnesota, Wisconsin, Michigan, New York, New Hampshire, Vermont, Indiana, Maine and Maryland.

The Geophysical Institute at the University of Alaska at Fairbanks responded to the conflicting chatter Tuesday evening, stressing that space weather forecasts, like more earthy ones, do change and that they do not produce aurora forecasts.

Don Hampton, a space physicist at the institute, offered further thoughts:

“Recent predictions of auroral activity visible much further south than Alaska have been making the rounds. This prediction, several days ahead of time, is based on models run by the Space Weather Prediction Center, a part of NOAA, and located in Boulder, Colorado. The Geophysical Institute does not make long-term auroral predictions. Our aurora forecast is from SWPC.”

“The energy and therefore level of activity comes directly from the stream of charged particles that come from the sun — the solar wind. The solar wind is always present, but when the sun is near solar max, when there are more sunspots indicating a complicated solar magnetic field structure, we often see more erratic and often very energetic bursts of solar wind. These are what cause the active auroral periods when the auroral oval expands and allows viewing south of the US/Canada border. This sort of activity is what was predicted by NOAA for the coming week.”

“The accuracy of the models to predict the auroral activity depend strongly on the accuracy and number of input measurements of the activity on the sun and the intervening space where the solar wind is flowing and evolving after it leaves the sun. There are only a few satellites and instruments dedicated to collecting these data, so the models typically have a wide range of predictions since the observations are relatively sparse. While large solar storms can be seen leaving the vicinity of the sun, and their direction and speed can be estimated, once they leave the local solar vicinity they cannot be tracked. During this time the solar storms can be slightly diverted or even reduced, and the final impact on Earth’s magnetic field may be different than predicted.”

“In terms of viewing aurora, the current solar cycle is showing many more sunspots than earlier predicted, so this promises to be an active period for aurora and space weather. This is good for aurora observers, but maybe not as good for operators of satellites and those wanting to communicate with satellites or by radio. Large solar storms can also affect power grids and pipelines by inducing large currents and potentially disrupting power. This is why NOAA produces the predictions, so that operators can take steps to mitigate potential issues.”

Contact Us