As unpredictable solar storms ravage Earth, AI comes to the rescue

tech.hindustantimes.com:

Solar storms are hitting Earth almost constantly as the Sun reaches the peak of its current solar cycle. And the worst part is that we cannot predict whether the dangerous sunspots currently present on the Sun will explode to create more storms. That is because we do not have a method to predict solar storms. We only know about them after the Sun has erupted, which gives us a warning period between 2-48 hours, depending on the type of solar disturbance. But now, some scientists believe that artificial intelligence (AI) could be a crucial tool to predict solar storms weeks or even months ahead of time.

Dr. Andy Smith from Northumbria University in UK has been awarded more than 5,00,000 pounds to develop an AI program that can reliably predict solar disturbances ahead of time, according to a report by SpaceRef. Notably, the UK had added severe space weather to the UK National Risk Assessment in 2011. Severe space weather refers to solar storms, coronal mass ejections (CME)-based storms, solar flare eruptions and solar wind attacks that can have significant impact on satellites, mobile phone communications, Internet as well as power grids.

Dr. Smith recently received a Research Fellowship from the Natural Environment Research Council (NERC) to find out whether artificial intelligence and data-set based learning models can be created to develop systems capable of forecasting space weather and issue warnings for severe solar storms much ahead in time.

For this, Dr. Smith and his team will now analyze huge amounts of data around the Sun collected over a period of 20 years. This data comes from various satellites that have been monitoring the Sun as well as Earth-based observatories which has been tasked with the same.

“This