Van Allen radiation belt electron dynamics are governed by a wide range of physical processes that can simultaneously drive acceleration, transport and loss. However, each individual process can be linked to a specific energy-dependent pitch angle distribution (PAD). We employ a new, unsupervised machine learning technique on 7-years of Van Allen Probe Relativistic Electron-Proton Telescope data and discover that six PADs, two each of: pancake, butterfly, and flattop, successfully describe >70% of classified relativistic PADs. We investigate the occurrence and storm-time evolution of each PAD through 45 geomagnetic storms. We find new populations of PADs, including: “shadowing-like” and wave-particle interaction signatures at low-L, and radial diffusion and substorm injections at higher-L, as well as determining that wave-particle interaction dominated PADs are swamped by radial diffusion processes through geomagnetic storms. Our results clearly demonstrate that PAD characterisation is a key component of understanding Van Allen radiation belt electron dynamics

We investigate the effects of intense chorus waves (wave power > 10-4 nT2) on relativistic electrons (E > 0.5 MeV) in the heart of the outer radiation belt (L* = 4 - 6) using superposed epoch approach. Combining electron flux and electromagnetic wave measurements from 70 geomagnetic storms during the Van Allen Probes mission, we show the relationship between integrated chorus wave power (0.1 - 0.8 equatorial electron gyrofrequency) and changes in relativistic electron flux on two hour timescales. During the loss-dominated storm main phase (Superposed Epoch -0.5 to 0 days), intense chorus waves mitigate the net loss of relativistic electrons. Conversely, in the early recovery phase (Superposed epoch 0 to 0.5 days), flux increases across a range of relativistic energies regardless of chorus wave power. The amount of electron flux at keV energies appears to have an influence on the consequences of chorus wave activity during geomagnetic storms.

Earth’s inner magnetosphere contains multiple electron populations influenced by different factors. The cold electrons of the plasmasphere, warm plasma that contributes to the ring current, and the relativistic plasma of the radiation belt often seem to behave independently. Using omni-directional flux and energy measurements from the HOPE and MagEIS instruments aboard the Van Allen Probes, we provide a detailed density and temperature description of the inner magnetosphere, offering a comprehensive statistical analysis of the entire Van Allen Probe era. While number density and temperature data at geosynchronous orbit are available, this study focuses on the inner magnetosphere (2 < L∗ < 6). Values of density and temperature are extracted by fitting energy and phase space density to obtain the distribution function. The fitted distributions are related to the zeroth and second moments to estimate the number density and temperature. Analysis has indicated that a two Maxwellian fit is sufficient over a wide range of L∗ and that there are two independent plasma populations. The more energetic population has a median number density of approximately 1.2 × 104 m−3 and a temperature of around 130 keV, with a temperature peak observed between L* = 4 and L*= 4.5. This population is relatively uniform in MLT. In contrast, the less energetic warm electron population has a median number density of about 2.5 × 104 m−3 and a temperature of 7.4 keV. Strong statistical trends in density and temperature across both L* and MLT are presented, along with potential sources driving these variations.

Waves which couple to energetic electrons are particularly important in space weather, as they drive rapid changes in the topology and intensity of Earth’s outer radiation belt during geomagnetic storms. This includes Ultra Low Frequency (ULF) waves that interact with electrons via radial diffusion which can lead to electron dropouts and rapid acceleration and inward transport of electrons during. In radiation belt simulations, the strength of this interaction is specified by ULF wave radial diffusion coefficients. In this paper we detail the development of new models of electric and magnetic radial diffusion coefficients derived from in-situ observations of the azimuthal electric field and compressional magnetic field. The new models use L* as it accounts for adiabatic changes due to the dynamic magnetic field coupled with an optimized set of four components of solar wind and geomagnetic activity, Bz, V, Pdyn and Sym-H, as independent variables (inputs). These independent variables are known drivers of ULF waves and offer the ability to calculate diffusion coefficients at a higher cadence then existing models based on Kp. We investigate the performance of the new models by characterizing the model residuals as a function of each independent variable and by comparing to existing radial diffusion models during a quiet geomagnetic period and through a geomagnetic storm. We find that the models developed here perform well under varying levels of activity and have a larger slope or steeper gradient as a function of L* as compared to existing models (higher radial diffusion at higher L* values).

Characterising the location of the outer boundary of the outer radiation belt is a key aspect of improving radiation belt models and helps to constrain our understanding of the mechanisms by which the source and seed electron populations are transported into the radiation belts. In this paper, we hypothesise that there are statistical differences in the electron distribution function across the radiation belt outer boundary, and thus analyse electron flux data from the THEMIS (Time History of Events and Macroscale Interactions during Substorms) satellites to identify this location. We validate our hypothesis by using modelled electron L* values to approximately characterise the differences between electron distribution functions inside and outside of the radiation belts. Initially, we perform a simple statistical analysis by studying the radial evolution of the electron distribution functions. This approach does not yield a clear discontinuity, thus highlighting the need for more complex statistical treatment of the data. Subsequently, we employ machine learning (with no dependence on L*) to test a range of candidate outer boundary locations. By analysing the performance of the models at each candidate location, we identify a statistical boundary at ≈8 Earth radii, with results suggesting some variability. This statistical boundary is typically further out than those used in current radiation belt models.