Model now predicts satellite-killing radiation storms TWO days before they strike

Space scientists have successfully predicted satellite-killing radiation storms two days before they strike – beating out the previous model that alerted experts only one day in advance.

The new model, called PreMevE 2.0, uses machine-learning to improve forecasts by incorporating upstream solar wind speeds from the Van Allen belts.

The technology compiles existing data sets to ‘learn’ patterns and predict future storms so satellite operators can take protective measures, including temporarily shutting down part of or even the whole satellite to avoid damage.

The model’s creators have also noted that it can be used to capture earthquake patterns on earth in order to predict when these natural disasters will strike.

Scroll down for video 

PreMevE 2.0 was developed by space scientists at Los Alamos National Laboratory, who are working in a NASA and National Oceanic and atmospheric Administration (NOAA).

Yue Chen, a space scientist at Los Alamos National Laboratory and principal investigator on the project, said: ‘Radiation storms from the Van Allen belts can damage or even knock out satellites orbiting in medium and high altitudes above the Earth, but predicting these storms has always been a challenge.’

‘Given that the Van Allen Probes, which provided important data about space weather, recently de-orbited, we no longer have direct measurements about what’s happening in the outer electron radiation belt.’

‘Our new model uses existing data sets to ‘learn’ patterns and predict future storms so satellite operators can take protective measures, including temporarily shutting down part of or even the whole satellite to avoid damage.’

The new model does build on its predecessor that was able to predict these radiation storm one day in advance.

However, scientists have learned that incorporating upstream solar wind speeds can add another day to their prediction.

Youzuo Lin, a computational scientist at Los Alamos who developed the machine-learning algorithms for the model, said: ‘With the expectation that similar patterns may reveal themselves in the future, our model is capable of making predictions by capturing some critical signatures as a precursor to those future events.’

By testing the model with multiple machine-learning algorithms, this work confirms the predictability of MeV electrons, as well as the robustness of using low-Earth-orbit electron observations to drive predictions,’ added Chen. 

‘In addition, the framework set up in this work allows us to easily include more input parameters to predict more energetic electrons in the next step.

The machine learning framework developed for PreMevE 2.0 can also be applied to many broad applications that use time-related measurements, such as capturing earthquake patterns among large volumes of seismic time-series data, enabling detection of small earthquakes out of the noisy environments.