The Van Allen radiation belts contain ultrarelativistic electrons trapped in Earth's magnetic field. Since their discovery in 1958, a fundamental unanswered question has been how electrons can be accelerated to such high energies. Two classes of processes have been proposed: transport and acceleration of electrons from a source population located outside the radiation belts (radial acceleration) or acceleration of lower-energy electrons to relativistic energies in situ in the heart of the radiation belts (local acceleration). We report measurements from NASA's Van Allen Radiation Belt Storm Probes that clearly distinguish between the two types of acceleration. The observed radial profiles of phase space density are characteristic of local acceleration in the heart of the radiation belts and are inconsistent with a predominantly radial acceleration process.

[1] Thirty years ago Paulikas and Blake (1979) showed a remarkable correlation between geosynchronous relativistic electron fluxes and solar wind speed (Vsw). This seminal result has been a foundation of radiation belt studies, space weather forecasting, and current understanding of solar wind radiation belt coupling. We have repeated their analysis with a considerably longer-running data set from the Los Alamos National Laboratory energetic particle instruments with several surprising results. Rather than the roughly linear correlation between Vsw and log (flux), our results show a triangle-shaped distribution in which fluxes have a distinct velocity-dependent lower limit but a velocity-independent upper limit. The highest-electron fluxes can occur for any value of Vsw with no indication of a Vsw threshold. We also find a distinct solar cycle dependence with the triangle-shaped distribution evident in 2 declining phase years dominated by high-speed streams but essentially no correlation in 2 solar maximum years. For time periods that do show a triangle-shaped distribution we consider whether it can be explained by scatter due to other parameters. We examine the role of time dependence and time lag in producing the observed distribution. We also look at the same statistical relationship but at energies 1 MeV. We conclude that the relationship between radiation belt electron fluxes and solar wind velocity is substantially more complex than suggested by previous statistical studies. We find that there are important ways in which the "conventional wisdom" stating that high-velocity wind drives high-MeV electron fluxes is, in general, either misleading or unsupported.

[1] Electromagnetic ion cyclotron (EMIC) waves have been observed during geomagnetic storms and are thought to contribute to ring current and radiation belt particle loss during the main phase. Ground-based storm time studies alternatively observe the majority of storm time Pc1-2 pulsations during the recovery phase. In this study we look at the occurrences of EMIC waves during 119 storms occurring throughout the CRRES mission. The storms were defined using the Sym-H index. The storm was divided into three phases: pre-onset, main, and recovery (80% of minimum Sym-H value). The majority, 56.25%, of storm time EMIC waves were found to occur during the main phase, while 35.57% were observed in the recovery phase. The recovery phase definition was then extended to 6 days after the minimum in order to compare with past ground studies. Although more EMIC waves were observed during this extended recovery phase, the maximum occurrence rate of EMIC waves remains in the main phase. A slight increase does appear in the 4-6 days after the minimum Sym-H value. The mean occurrence location of EMIC waves was at L = 6.07, MLT = 15.12. This suggests that EMIC waves may have been generated when the ring current and plasmasphere or plasma plume particle populations overlap, often observed during the main phase due to storm dynamics. The rise in the occurrence rate in the extended recovery phase potentially is dominated by the process of plasmaspheric expansion after the end of the storm.

Quantitative assessment of modeling and forecasting of continuous quantities uses a variety of approaches. We review existing literature describing metrics for forecast accuracy and bias, concentrating on those based on relative errors and percentage errors. Of these accuracy metrics, the mean absolute percentage error (MAPE) is one of the most common across many fields and has been widely applied in recent space science literature and we highlight the benefits and drawbacks of MAPE and proposed alternatives. We then introduce the log accuracy ratio and derive from it two metrics: the median symmetric accuracy and the symmetric signed percentage bias. Robust methods for estimating the spread of a multiplicative linear model using the log accuracy ratio are also presented. The developed metrics are shown to be easy to interpret, robust, and to mitigate the key drawbacks of their more widely used counterparts based on relative errors and percentage errors. Their use is illustrated with radiation belt electron flux modeling examples.

The DREAM3D diffusion model is applied to Van Allen Probes observations of the fast dropout and strong enhancement of MeV electrons during the October 2012 "double-dip" storm. We show that in order to explain the very different behavior in the two "dips," diffusion in all three dimensions (energy, pitch angle, and L * ) coupled with data-driven, event-specific inputs, and boundary conditions is required.Specifically, we find that outward radial diffusion to the solar wind-driven magnetopause, an event-specific chorus wave model, and a dynamic lower-energy seed population are critical for modeling the dynamics.In contrast, models that include only a subset of processes, use statistical wave amplitudes, or rely on inward radial diffusion of a seed population, perform poorly. The results illustrate the utility of the high resolution, comprehensive set of Van Allen Probes' measurements in studying the balance between source and loss in the radiation belt, a principal goal of the mission.

The radiation belts and plasma in the Earth's magnetosphere pose hazards to satellite systems which restrict design and orbit options with a resultant impact on mission performance and cost. For decades the standard space environment specification used for spacecraft design has been provided by the NASA AE8 and AP8 trapped radiation belt models. There are well-known limitations on their performance, however, and the need for a new trapped radiation and plasma model has been recognized by the engineering community for some time. To address this challenge a new set of models, denoted AE9/AP9/SPM, for energetic electrons, energetic protons and space plasma has been developed. The new models offer significant improvements including more detailed spatial resolution and the quantification of uncertainty due to both space weather and instrument errors. Fundamental to the model design, construction and operation are a number of new data sets and a novel statistical approach which captures first order temporal and spatial correlations allowing for the Monte-Carlo estimation of flux thresholds for user-specified percentile levels (e.g., 50th and 95th) over the course of the mission. An overview of the model architecture, data reduction methods, statistics algorithms, user application and initial validation is presented in this paper.

We present a statistical study of relativistic electron counts in the electron radiation belt across a range of drift shells (L * > 4) combining data from nine combined X-ray dosimeters (CXD) on the global positioning system (GPS) constellation. The response of the electron counts as functions of time, energy and drift shell are examined statistically for 67 solar wind stream interfaces (SIs); two-dimensional superposed epoch analysis is performed with the CXD data. For these epochs we study the radiation belt dropouts and concurrent variations in key geophysical parameters.At higher L * we observe a tendency for a gradual drop in the electron counts over the day preceding the SI, consistent with outward diffusion and magnetopause shadowing. At all L * , dropouts occur with a median time scale of 7 h and median counts fall by 0.4-1.8 orders of magnitude. The central tendencies of radiation belt dropout and recovery depend on both L * and energy. For 70 per cent of epochs Sym-H more than −30 nT, yet only three of 67 SIs did not have an associated dropout in the electron data. Statistical maps of electron precipitation suggest that chorus-driven relativistic electron microbursts might be major contributors to radiation belt losses under high-speed stream driving.

On 30 September 2012, a flux "dropout" occurred throughout Earth's outer electron radiation belt during the main phase of a strong geomagnetic storm. Using eight spacecraft from NASA's Time History of Events and Macroscale Interactions during Substorms (THEMIS) and Van Allen Probes missions and NOAA's Geostationary Operational Environmental Satellites constellation, we examined the full extent and timescales of the dropout based on particle energy, equatorial pitch angle, radial distance, and species. We calculated phase space densities of relativistic electrons, in adiabatic invariant coordinates, which revealed that loss processes during the dropout were > 90% effective throughout the majority of the outer belt and the plasmapause played a key role in limiting the spatial extent of the dropout. THEMIS and the Van Allen Probes observed telltale signatures of loss due to magnetopause shadowing and subsequent outward radial transport, including similar loss of energetic ring current ions. However, Van Allen Probes observations suggest that another loss process played a role for multi-MeV electrons at lower L shells (L* <~4).