Starlink loss highlights current space weather prediction capabilities, coming advancements

by Haygen Warren & Chris Gebhardt

On February 8, 2022, SpaceX announced the loss of around 40 of the 49 Starlink satellites launched on the Starlink Group 4-7 mission from LC-39A on February 3. For spacecraft initially orbiting at low altitudes, as the Starlinks did at the time upon deployment, many factors can lead to a craft not reaching its final intended orbit.

However, the loss of the Group 4-7 satellites ultimately traces back to the Sun.

The loss was due to a geomagnetic storm in Earth’s magnetic field and atmosphere. In the case of the 4-7 mission, the geomagnetic storm caused Earth’s atmosphere to heat, leading to a 50% increase in atmospheric density at the orbital altitude the newly-deployed Starlinks were in.

This increase in atmospheric density increased the drag on the satellites, and despite SpaceX turning them edge-on to minimize drag as much as possible — a capability that is part of their larger safe mode application — the geomagnetic storm’s impact proved too intense.

With too much drag and rapidly decaying orbits, many of the Starlinks could not recover and reentered Earth’s atmosphere. Approximately nine of the 49 Starlinks launched on the 4-7 mission survived per a SpaceX statement.

Although the reentry of the Starlinks was unplanned, all v1.5 Starlinks are designed to completely disintegrate upon reentry — posing no risk to populated areas.

So what happened here: how was an expensive mission lost without warning? What exactly was this storm? And how can we predict these types of storms in the future?

“It’s the geomagnetic storm that caught the Starlink spacecraft,” said Dr. James Spann, space weather lead in the heliophysics division of NASA’s Science Mission Directorate. “There was some [EUV (extreme ultraviolet)] going on also, but it was really the surprise — well, not the surprise, but the sudden arrival of the Coronal Mass Ejections that caused the geomagnetic storm.”

A geomagnetic storm is a disturbance in Earth’s magnetic field, typically caused by high amounts of solar wind or solar radiation interacting with Earth’s magnetosphere. In particular, geomagnetic storms are most commonly caused by massive eruptions from the Sun’s surface called Coronal Mass Ejections, or CMEs.

Geomagnetic storms are notorious for causing issues in GPS and radio communications, power usage, satellite operations, and interference with aviation. The largest and most intense geomagnetic storm ever recorded occurred in September 1859, called the Carrington Event.

During the peak of the Carrington Event, telegraph lines in the U.S. were severely damaged, leading to the start of fires and the shocking of telegraph operators. Additionally, auroras were observed worldwide — even near the equator — as the most intense solar radiation interacted with Earth’s magnetic field.

In all, there are three categories of solar storm events that impact Earth: flares, energetic protons, and CMEs.

Flares — packed with EUV and x-rays — arrive quickest; traveling at the speed of light, they arrive within 500 seconds or eight minutes. This particular class of storm will heat the ionosphere, the portion of Earth’s atmosphere beginning around 80 km in altitude, altering the density characteristics of the lower atmosphere.

Energetic protons arrive within a few hours and carry the increased radiation risks that would pose a problem for airline passengers as well as astronauts in space.

The slowest of the storm elements to arrive would be those stemming from Coronal Mass Ejections, portions of the Sun that are ejected into space as balls of energetic plasma. Depending on the interplanetary density between the Sun and Earth, these storms can take anywhere from 18 to 36 hours to arrive at Earth’s orbital distance, and if they strike Earth, they will pump the radiation belts full of energy and create dazzling auroral displays.

In the case of the Starlinks and their initial low deployment orbit, there was no warning to give SpaceX on the day of launch as the CME event that ultimately doomed the satellites had not yet occurred.

Yet this loss has — in part — drawn attention to our lack of ability to accurately predict this kind of space weather event and its effects on Earth’s atmosphere and our technology, both things that are actively being worked on.

“So the prediction of these things is really, really complicated. We can predict them with observations and with our models,” said Dr. Spann. “Activity on the Sun generally occurs in areas that are energized. Sunspots are typically the way we would describe that.”

Richard Carrington’s drawings of the sunspots of 1 Sept. 1859, including notations (“A” and “B”) where the solar flare erupted from (“A”) and where it disappeared (“B”). (Credit: American Scientist, Vol. 95)

“But it’s complicated because we can kind of tell when these areas become active because they change their character as we look at them on the face of the Sun. But we can’t see when they’re going to suddenly erupt and whether it’s a flare or a Coronal Mass Ejection. We have not understood it enough to understand when that’s going to happen.”

“There are a lot of models, and there are some signatures. But the best we can do is kind of statistically indicate that … it has an X% chance to become unstably active and erupt over the next day or so. We can’t say ‘Oh, in three hours, this is going to go.’ We don’t have that fidelity.”

But once an event happens, can we then predict the effects? Even if only on short notice?

For flare events that move at the speed of light, their intensity can only be ascertained once they arrive at an observation satellite or Earth. “Obviously, we only see it after the light has come here. So by the time that’s happened, the x-rays are already here.”

“If it’s a particle event on the other hand, we’ve got maybe a couple of hours to look at it. And we can only predict what the particle energy is and how much. In a Coronal Mass Ejection, we have a much better idea of it because we can kind of see the explosion,” said Dr. Spann.

For CMEs, the timing of when that event will impact the planet can usually be determined within a six to seven-hour window, and while the rough effect can be calculated, scientists still are not able to approximate how strong a geomagnetic storm will be until the CME passes the L1 (Lagrange Point 1) point directly between Earth and the Sun — just a few hours ahead of the planet in terms of CME travel time.

The Sun-Earth Lagrange Point locations — with L1 directly between the Sun and Earth. (Credit: NASA/WMAP Science Team)

That L1 point is where some, but not all, of our ever-watchful Sun satellites — early warning systems, so to speak — reside.

But it is data from all Heliophysics missions combined that helps form the larger picture. Spacecraft such as the SOHO (Solar and Heliospheric Observatory), SDO (Solar Dynamics Observatory), Solar Orbiter, and the Parker Solar Probe constantly watch the Sun and take measurements to try to accurately predict what our sometimes unpredictable — from our knowledge standpoint — star will do.

But even with those missions, there are limitations in predicting how extreme these events will be.

“That’s pretty hard,” related Dr. Spann. “How much [atmospheric] heating is going to take place depends on how intense [the event] is. And so there are just a lot of factors. We are still surprised all the time when something becomes very active when we didn’t anticipate it. Or we thought something really big’s coming, and it was kind of a big nothing burger. So we’re still learning this. We’re getting better. But it’s difficult. It’s difficult.”

So how can accuracy be improved? The answer, in part, lies not just with our Sun sentinels but with Cubesats as well.

“We’ve recently launched a lot of Cubesats, very small satellites, that are in low Earth orbit and are making measurements of the ionosphere. And we’re getting ready to launch several others. Each one kind of has a different focus,” said Dr. Spann.

The world’s Heliophysics fleet of spacecraft that keep constant watch on the Sun. (Credit: NASA)

“But very importantly, now in development, is a mission called Geospace Dynamics Constellation, GEC. And that is anticipated to be a six spacecraft mission that’s going to make some very precise measurements as a constellation to really address a lot of this energy transfer between the magnetosphere, the ionosphere, and thermosphere.”

“And all of that is coupled with a lot of ground-based observations that are going on all the time anyway — with radar and ionosondes and magnetometers and all that sort of stuff — to really get a good handle on how this whole process works. Once we understand how it works, then we can do better predictions. And once we can do better predictions, then we can set up a forecast, or enable NOAA to do forecasting, which will help us be better prepared for all of these things.”

But more immediately, this most-recent example with Starlink shows how much we still don’t know about how our parent star operates. NASA’s Parker Solar Probe is trying to help scientists unlock the mysteries of the corona by literally diving through it — a feat of engineering that took 60 years to realize.

Meanwhile, ESA’s Solar Orbiter is — for the first time — allowing observations of the Sun’s polar regions.

“It is a complicated system,” offered Dr. Spann. “But that’s part of what NASA’s job is. That’s part of what this particular division’s, the Heliophysics Division at NASA, goal is: to understand the solar and space physics of this connected system.”

But Dr. Spann also pointed out that what often lacks in conversation about space weather events like the geomagnetic storm and the Starlinks is the fact that we’ve been in a solar minimum period for some years and are now starting to enter a more active period.

“We’ve been very very quiet. And a lot of new technologies and a lot of new things have happened since we’ve had a very strong solar maximum where the Sun gets very dynamic and space weather becomes much more active and stronger.”

“So now that we’re starting to go up to solar max, we’re beginning to see systems that have been launched during the solar minimum which had zero problems now experiencing problems… even though we really haven’t had much space weather. But what little we have had, we’ve already seen some impacts.”

“And so it’s a little bit of a learning curve for everyone. And it’s just going to become more and more impactful over these next several years, as we get up to solar max.”

(Lead image: Rendering of Starlink satellites deploying from the second stage. Credit: Mack Crawford for NSF)

Related Articles