The focus of this dissertation is on improving the understanding and prediction capability of thermal (1–10 keV) to superthermal (10–100 keV) electron flux in the near-Earth (6–12 RE, 18–00–06 MLT; RE ≈ 6370 km, MLT = magnetic local time) plasma sheet and especially its dependence on solar wind driving.

The solar wind and interplanetary magnetic field flow past and interact with Earth’s magnetic field, forming a magnetic cavity called the magnetosphere. In Earth’s nightside magnetosphere, the dynamics of electrons within the near-Earth magnetotail are highly dependent on variations in the solar wind and overall solar activity. The electrons are the source population for the Earth’s outer radiation belts, contribute to storm-time ring current pressure and energy, and precipitate into the ionosphere forming part of the aurora. Additionally, this population of electrons are most responsible for the spacecraft surface charging environment at geosynchronous (GEO) orbit. Some dependencies of superthermal electrons on solar wind variations have been identified through several previous investigations; however, there is much that is not known, and it has been difficult to predict their behavior.

Utilizing more than twelve years of data from THEMIS spacecraft, we have investigated the long-term and short-term dependencies of electron flux in the near-Earth plasma sheet to solar wind variations. On time scales of a solar sunspot cycle, we explore the response of energetic electron flux in the near-Earth plasma sheet to solar wind and geomagnetic activity. We show that with only slight solar wind driving (solar wind flow speed, VSW = 400–500 km·s-1) and weak geomagnetic activity (Auroral Electrojet index, AE = 100–300 nT), there is a substantial increase in median 12–52 keV electron flux. We also explored the role that solar wind driving has on short-term (1–2 hours) electron flux variations and found that VSW is the most significant contributor to severe spacecraft surface charging environments at GEO. Furthermore, simply the presence of elevated geomagnetic activity (as indicated by the AE index) is a sufficient indicator of risk for an extreme charging environment. In examining 101 observations of the change of electron flux in the near-Earth plasma sheet during substorms, we failed to find a dependence of flux changes on substorm strength, nor spacecraft location, nor location relative to the peak of auroral activity.


Finally, we develop an empirical, machine-learned neural network model of electron flux in the near-Earth plasma sheet, dependent on inputs of solar wind parameters and their time history. Our model overcomes limitations of previous models by including only inputs that are external to the magnetosphere and predicting differential flux at a wide range of energies. We calculate several model–observation metrics—our model predicts electron flux with a Pearson correlation coefficient between 0.55–0.77 and has a median symmetric accuracy of between 41–140% (metric ranges depend on energy); and, we rank which solar wind parameters are most relevant to predictions. We show that including short time (5-minute) resolution inputs to the model does not result in predicting small scale (1-hour) variations of plasma sheet electron flux. Overall, this dissertation advances knowledge of the dependence of superthermal electron flux in the near-Earth plasma sheet to solar wind variations.

For details, please visit: https://rackham.umich.edu/navigating-your-degree/oral-defense-dates/