Machine Learning for Predicting the Bz Magnetic Field Component From Upstream in Situ Observations of Solar Coronal Mass Ejections. Issue 12 (3rd December 2021)

• Record Type: Journal Article
• Authors: Reiss, M. A.; Möstl, C.; Bailey, R. L.; Rüdisser, H. T.; Amerstorfer, U. V.; Amerstorfer, T.; Weiss, A. J.; Hinterreiter, J.; Windisch, A.
• Is Part Of: Space weather. Volume 19:Issue 12(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-03
• Subjects: Coronal mass ejections; Space environment -- Periodicals; 551.509992
• DOI: 10.1029/2021SW002859
• Languages: English
• ISSNs: 1542-7390

Abstract

Abstract: Predicting the Bz magnetic field embedded within interplanetary coronal mass ejections (ICMEs), also known as the Bz problem, is a key challenge in space weather forecasting. We study the hypothesis that upstream in situ measurements of the sheath region and the first few hours of the magnetic obstacle provide sufficient information for predicting the downstream Bz component. To do so, we develop a predictive tool based on machine learning that is trained and tested on 348 ICMEs from Wind, STEREO‐A, and STEREO‐B measurements. We train the machine learning models to predict the minimum value of the Bz component and the maximum value of the total magnetic field Bt in the magnetic obstacle. To validate the tool, we let the ICMEs sweep over the spacecraft and assess how continually feeding in situ measurements into the tool improves the Bz prediction. We specifically find that the predictive tool can predict the minimum value of the Bz component in the magnetic obstacle with a mean absolute error of 3.12 nT and a Pearson correlation coefficient of 0.71 when the sheath region and the first 4 hr of the magnetic obstacle are observed. While the underlying hypothesis is unlikely to solve the Bz problem, the tool shows promise for ICMEs that have a recognizable magnetic flux rope signature. Transitioning the tool to operations could lead to improved space weather forecasting. Plain Language Summary: At any time, our solar system is populated with interplanetary coronal mass ejections (ICMEs). Solar scientists and space weather forecasters track ICMEs when they are ejected from the Sun and follow their path into the vast reaches of interplanetary space. They do so because if an ICME hits Earth, it could damage our infrastructure such as power‐grids and GPS satellites, which are a mainstay of our modern civilization. The possible damage is primarily determined by the magnetic field embedded within the ICME. The North‐South magnetic field component, Bz, plays a decisive role, especially if it is pointing opposite to Earth's magnetic field. Currently we cannot predict Bz with sufficient accuracy. Scientists often refer to our limited predictive abilities as the Bz problem. Here we shine a new light on the Bz problem by developing a predictive tool based on machine learning that is trained and tested on 348 ICMEs. By feeding measurements of the ICME into machine learning algorithms, we find that our predictive tool can forecast the Bz component reasonably well. While our tool does not solve the Bz problem, it shows promise in forecasting and potentially mitigating the effects of ICMEs on our planet Earth and its inhabitants. Key Points: We hypothesize that upstream in situ measurements are sufficient to predict Bz in solar coronal mass ejections We present a predictive tool that forecasts the minimum of Bz in an interplanetary coronal mass ejections with a MAE of 3.12 nT and a pearson correlation coefficient of 0.71