17:15 〜 18:45
[PEM13-P10] An Empirical Plasmapause Model using Arase/PWE Data and Machine Learning
キーワード：プラズマポーズ、プラズマ圏ヒス、機械学習
The plasmapause is a boundary that clearly separates the high-density plasmasphere from the low-density plasma trough region. Determining the variable plasmapause position is important for understanding the dynamics of the inner magnetosphere. The plasmapause position is generally identified as a sharp gradient in electron density. Tracking of the upper hybrid resonance (UHR) frequency is the most popular way to determine the plasmapause location in space. In this study, we propose a machine learning approach to determine the plasmapause location from the electric and magnetic field plasma wave measurement data, instead of using the UHR frequency tracking.
We used electric and magnetic field power spectral density data observed by the Onboard Frequency Analyzer (OFA), a subsystem of the Plasma Wave Experiment (PWE) aboard the Arase satellite. To improve the accuracy of plasmapause determination, we used satellite orbit data (Mcilwain-L value, altitude, magnetic latitude, and magnetic localtime) and geomagnetic activity index. We examined averaged plasmapause locations under several geomagnetic conditions: geomagnetic quiet (Kp<2), moderate disturbance (2<=Kp<4) and large disturbance (4<=Kp) cases. We confirmed that the shrink and erosion of plasmasphere depending on the geomagnetic activity are clearly seen in the result. Furthermore, a distinctive features of the plasma plume are observed around 16h magnetic localtime during geomagnetic disturbance period. In this presentation, we introduce an empirical plasmapause model based on the least squares method applied to the machine learning results. We confirmed that the model accurately reproduced the variations in the plasmapause location corresponding to changes in the Kp index.
We used electric and magnetic field power spectral density data observed by the Onboard Frequency Analyzer (OFA), a subsystem of the Plasma Wave Experiment (PWE) aboard the Arase satellite. To improve the accuracy of plasmapause determination, we used satellite orbit data (Mcilwain-L value, altitude, magnetic latitude, and magnetic localtime) and geomagnetic activity index. We examined averaged plasmapause locations under several geomagnetic conditions: geomagnetic quiet (Kp<2), moderate disturbance (2<=Kp<4) and large disturbance (4<=Kp) cases. We confirmed that the shrink and erosion of plasmasphere depending on the geomagnetic activity are clearly seen in the result. Furthermore, a distinctive features of the plasma plume are observed around 16h magnetic localtime during geomagnetic disturbance period. In this presentation, we introduce an empirical plasmapause model based on the least squares method applied to the machine learning results. We confirmed that the model accurately reproduced the variations in the plasmapause location corresponding to changes in the Kp index.