Propagate L1 solar wind measurements to Earth with the help of machine learning

Abstract

The propagation of solar wind measurements from L1 to the bow shock nose of Earth is the basis of the frequently used OMNI dataset. Depending on the solar wind speed, the propagation time delay from L1 to Earth lies between 20 and 90 minutes. In this study we present a machine learning algorithm that is suitable to predict the solar wind propagation delay between Lagrangian point L1 and the Earth. This work introduces the proposed algorithm and investigates its applicability to propagate ACE data to Earth. The propagation delay is measured from interplanetary shocks passing the Advanced Composition Explorer (ACE) first and their sudden commencements within the magnetosphere later, as recorded by ground-based magnetometers. Overall 380 interplanetary shocks with data ranging from 1998 to 2018 builds up the database that is used to train the machine learning model. We investigate two different feature sets. The training of one machine learning model will use all three components of solar wind speed , the other only bulk solar wind speed. Both feature sets also contain the position of the spacecrafts. The major advantage of the machine learning approach is its simplicity when it comes to its application. After training, values for the different features have to be fed into the algorithms only and the evaluation of the propagation delay can be continuous. Both machine learning models will be used to propagate ACE data to Earth. The propagated ACE measurements are compared to OMNI data during different solar wind conditions. Future assessments will include a comparisons of propagated solar wind data to satellite measurements of the IMF just outside of Earth magnetosphere.