Chinese researchers have developed a new algorithm to automatically derive kinematic parameters of coronal mass ejections (CMEs) based on machine learning, according to a recent research article published in the Astrophysical Journal Supplement Series, highlighting the great significance of this algorithm in predicting catastrophic space weather.

CMEs are large scale masses of plasma thrown from the sun into interplanetary space and are considered the largest form of energy release in the solar system. They constitute the major source of severe space weather events, with the potential to cause enormous damage to humans and spacecraft in space.

It is becoming increasingly important to detect and track CMEs, since there are now more space activities and facilities, the study noted.

The study of the revolution of CMEs in solar corona and interplanetary space is a major topic in the field of space weather, and so too the positional relations between CMEs and Earth's orbit, according to Shen Fang, a researcher with the National Space Science Center of the Chinese Academy of Sciences.

Their method consisted of three steps -- recognition, tracking, and determination of parameters.

First, the researchers trained a neural network to judge whether there were CMEs observed in images. Next, they acquired binary-labeled CME regions. Finally, they tracked a CME's motion in time-series images and determined the CME's kinematic parameters such as velocity, angular width, and central position angle.

The algorithm can identify relatively weak CME signals and generate accurate morphology information concerning CMEs, said Shen. It is expected to assist with real-time CME warnings and predictions.