Predicting equatorial plasma bubbles with SWARM



After sunset, variations in air density can lead to the formation of heated pockets of gas known as "plasma bubbles" above the equator of the Earth, which can interfere with satellite-to-earth communication. Scientists are now using new AI algorithms to foresee plasma bubble episodes and predict their occurrence. The project was presented this week at the National Astronomy Meeting (NAM 2022) by University College London PhD candidate Sachin Reddy.

Plasma bubbles, pockets of very hot gas that extend into space shortly after sunset and are located up to 900 kilometers above the Earth's surface, develop in the upper atmosphere. These bubbles begin tiny and expand quickly, expanding in only a few hours from the size of a football field to that of a small nation. As the bubbles get bigger, they can obstruct and twist radio signals from satellites, making it impossible for them to communicate with the Earth.

A group of scientists has compiled 8 years' worth of data from the SWARM satellite mission in order to forecast plasma bubbles. The Ionospheric Bubble Index, an autonomous bubble detector, is a feature of the spacecraft. In order to determine if bubbles are present, this examines variations in electron density and magnetic field intensity; a high connection between the two suggests the presence of a plasma bubble.

The satellite passes in the midst of the majority of plasma bubbles at a height of 460 km, or around 30 times higher than a commercial aircraft. The model makes predictions on the probability of a plasma bubble event occurring at any moment by combining the data gathering from SWARM with a machine learning method.

The findings demonstrate that the frequency of plasma bubble occurrences fluctuates with solar activity and the seasons, precisely as the frequency of weather events. The South Atlantic Anomaly is an area in the Atlantic where most occurrences take place, and the model indicates that location is a far more important factor in forecasting plasma bubbles than the time of year. Across many tests, the AI model predicts occurrences with an accuracy of 91%.

Reddy claims "We need to be able to predict bubbles, just like the weather on Earth, to avoid significant satellite service interruptions. We want to be able to say something like, "There is a 30% possibility of a bubble forming above the Horn of Africa at 8 o'clock tomorrow." For spacecraft operators and others like you and me who depend on satellite data every day, this type of information is incredibly helpful."



Story Source:

Materials provided by Royal Astronomical Society. Note: Content may be edited for style and length.

Comments

Popular posts from this blog

A Nuclear War Could Starve Billions, But One Country May Be Safer Than The Rest

Plants Have Been Keeping a Secret From Us About How Thirsty They Actually Are

It Really Is in Your Head: Thinking Hard and Long Can Cause Brain Drain