NASA Frontier Development Lab Uses Deep Learning to Virtually Monitor the Sun’s Ultraviolet Emission

Why it matters: Powerful solar storms, bursts of solar plasma and charged particles can harm satellites in orbit and even cause major problems for power grids on Earth. NASA’s Solar Dynamics Observatory (SDO), which can spot solar storms in near real-time, is a key part of that. One of the SDO instrument sensor MEGS-A was designed to keep an eye on extreme ultraviolet (EUV) radiation levels, which correlate with a ballooning of the Earth's outer atmosphere that control the longevity of satellites in near-Earth orbit. This instrument stopped working in 2014.

What the team did: A deep-learning network created during 2018 NASA FDL sprint can now be used to replace the data from the non-working sensor by inferring what ultraviolet radiation levels that sensor would have detected based on what the other instruments on SDO are observing at any given time. Using deep learning, this work demonstrates away of virtually monitoring the Sun's extreme ultraviolet emission, which is an important driver of space weather. 

What is the benefit: Deep learning can help us get more value out of our current ability to monitor the Sun by providing virtual instruments that can supplement our physical instruments. AI models like this could also be used for other future missions. Instead of putting say three instruments on a satellite to measure different aspects of the space environment, you could potentially launch two and use the data collected to infer the information that would have been measured by a third.