A key NASA discovery could protect mega-constellations like SpaceX Starlink

NASA's new A.I. could step in.

Satellites could soon get a whole lot more protection, thanks to NASA’s latest research.

The agency has discovered a way to use machine learning techniques to enhance the readings from hardware components. In this instance, a team has created a virtual component for NASA’s Solar Dynamics Observatory that boosts the ability to measure extreme ultraviolet radiation. The research was published Wednesday in the journal Science Advances. The breakthrough enables the agency to better understand space weather, critical to avoiding satellite malfunctions or radio blackouts.

The research could prove vital as more and more satellites take to the skies. While satellites are used today for applications like locating your phone or communicating around the world, planned mega-constellations would use a lot more crafts to provide fast internet access to people around the world. There’s currently around 5,000 satellites in space total, but SpaceX’s Starlink service alone is expected to number around 12,000 crafts when complete. That’s a lot more crafts to worry about, and better radiation monitoring could help ensure they all run smoothly.

Alexandre Szenicer, a PhD student in Computational Geophysics at the University of Oxford and one of the paper’s authors, explains to Inverse that it’s best to think of the resultant data as a “nowcast” rather than a weather forecast, although he does note that work is ongoing to predict future radiation.

“The EUV radiation from the Sun is the dominant driver of the Earth’s thermosphere and ionosphere system,” Szenicer says. “During periods of elevated solar activity, enhanced solar EUV driving causes adverse space weather effects such as radio communication blackouts, increased aerodynamic drag on satellites in low-earth orbit, and scintillation of global navigation satellite systems signals. As such, having a good understanding of the EUV driving is important in order to react and mitigate effects of adverse space weather.”

While its benefits are likely to be felt across all satellites, smoother operation of mega-constellations could improve the reliability of these emergent internet services.

Starlink in action.

Mark Handley 

The researchers created a virtual component for the Multiple EUV Grating Spectrograph A channel, also known as MEGS-A. This channel experienced an electrical malfunction five years ago. It forms part of the extreme ultraviolet variability instrument, which has otherwise continued working since that time. A separate instrument, the Atmospheric Imaging Assembly or AIA, collects data about the other layers of the sun’s atmosphere.

The team collected together the data from the four years where MEGS-A and the AIA worked together correctly. They were then able to produce observations that filled in missing spectral information. The best-performing model from the research has a maximum error of 4.6 percent.

The system can take AIA images on the left, and make MEGS-A predictions on the right.


That means the model could help enhance existing components, ideal for fueling the future of space.

“Starlink’s satellites, given their altitude (550km), are subject to atmospheric drag,” Szenicer says. “Therefore, it is very important for the operators to have a good understanding of the dynamics of the thermosphere/ionosphere, for which information about the EUV spectral irradiance is crucial. Hence, I believe our work will indeed be useful for these missions.”

It’s not just satellites that could benefit from the research. Enhanced radiation tracking could also help with rocket launches, depending on the mission.

“Another interesting application could be the impact of EUV radiation on electronics onboard rockets,” Szenicer says.

Mega-constellations have caused concern, particularly among astronomers worried about sun reflections and from other satellite operators looking to avoid collisions. Thanks to NASA’s new research, radiation is one thing their creators will have less to worry about.

Related Tags