1. Introduction
Many competing processes contribute to the dynamic formation and depletion of the Earth’s radiation belts, including radial transport, local wave acceleration, particle loss to the magnetopause, particle precipitation into the atmosphere, and others [see review byThorne et al. , 2010]. These competing mechanisms typically occur simultaneously and are energy dependent; understanding the importance of each is fundamental to radiation belt physics.
Electron microbursts are impulsive (<1s) injections of energetic (few keV to >MeV) electrons into the atmosphere that may represent a major loss source from the radiation belt during storm main phase and recovery [Millan and Thorne , 2007]. Microburst electron precipitation was first observed in the 1960s by balloon measurements of bremsstrahlung X-rays produced by precipitating electrons as they enter Earth’s atmosphere [Winckler et al. , 1962; Anderson and Milton , 1964]. Microbursts have been observed in situ on satellites [Imhof and Nightingale , 1992,Blake et al ., 1996], and via X-rays on balloons [Millan et al., 2002]. Despite a long history of observations, we still do not understand the detailed physics of the underlying scattering mechanism nor the importance of microburst precipitation as a loss mechanism for outer belt energetic electrons [Blum et al. , 2015; Breneman et al ., 2017, Blum and Breneman, 2020 review]. This is in part due to a lack of data coverage during events when microbursts are observed over a large region and extended time. Although microbursts are short-lived, they can have fluxes more than an order of magnitude higher than the background precipitation [Lorentzen et al ., 2001a]. As such, microburst precipitation can potentially be a significant loss mechanism for outer radiation belt electrons.
The inherently bursty nature of whistler mode chorus waves, and their ability to resonate with a wide range of electron energies, make them the primary candidate for the generation of electron microbursts. Surveys have shown similarities in occurrence of microbursts and whistler mode chorus waves, including distributions in L and MLT and variation with magnetic activity level [Douma et al. , 2019]. Both share similar sub-second durations and occur at a similar cadence. The evidence to date thus strongly suggests that the dominant cause of microbursts is resonant scattering into the loss cone by chorus [Nakamura et al. , 2000; Lorentzen et al. , 2001a, 2001b;Thorne et al. , 2005, Kersten et al. , 2011; Breneman et al. , 2017]. However, details of the scattering, including scattering magnetic latitude (equatorial or off-equatorial), and the exact nature of resonance remain unverified or unknown. Recently,Miyoshi et al . [2020] has proposed a model that relativistic electron microbursts are the high-energy tail of the pulsating aurora. The latitudinal propagating chorus waves cause wide energy electron precipitations from keV energy (pulsating aurora) and MeV energy (relativistic electron microbursts) because the resonance energy depends on the magnetic latitudes. This model is consistent with recent studies that use conjugate observations between ground-based observations and SAMPEX and FIREBIRD [Kawamura et al ., 2021; Shumuko et al ., 2021; Zhang et al ., 2022].
One way to address these questions is with intercomparison of high-altitude chorus observations made on near-equatorial satellites and microburst observations on low Earth orbit satellites during magnetic conjunctions. This dataset has been realized in recent years with a collaboration from early 2016 to late 2019 between the Electric Fields and Waves (EFW) team on the near-equatorial Van Allen Probes (RBSP) and the low-altitude Focused Investigations of Relativistic Electron Burst: Intensity, Range, and Dynamics (FIREBIRD) CubeSats. More than 5900 minutes of EFW burst waveform data and >3300 minutes of EMFISIS burst waveform data were taken within +/- 1hr of ~813 magnetic conjunctions with FIREBIRD.
This significant conjunction dataset has uncovered details of the connection between chorus and microbursts. For example, Breneman et al. [2017] analyzed a close bounce-loss cone conjunction and found that the chorus-induced microburst precipitation from 220 – 985 keV was scattered at off-equatorial latitudes of +/-20-30 degrees. The latitudinal location of the scattering has been found to have a strong effect of the overall scattering rates and energy range [e.g.Thorne et al. , 2005; Shprits et al ., 2006]. Another study [Colpitts et al. , 2020] showed the first direct evidence of discrete chorus elements propagating from an equatorial satellite (Van Allen Probe A) to a satellite at higher latitudes (Arase). This study confirmed that these waves can propagate unducted, without selective amplification, to the latitudes necessary to scatter electrons producing microburst precipitation.
The conjunction dataset also allows us to estimate the scale size of the scattering region. Breneman et al. [2017] used FIREBIRD data to estimate the differential flux loss rate to the atmosphere due to 220 – 985 keV microbursts. The time to deplete the entire outer radiation belt through this mechanism can be determined by integrating the estimated average loss rate over the entire chorus source region. The chorus source size has been estimated from transient auroral flashes [Ozaki et al. , 2019] and pulsating auroral patches related to isolated chorus elements [Ozaki et al ., 2018a] and the source size can depend on the chorus wave propagation [Ozaki et al ., 2021]. However, this critical source region size is not well constrained.
An earlier study by Anderson et al. [2015] also estimated the size of a microburst region using observations from the Balloon Array for Radiation belt Relativistic Electron Losses (BARREL) mission, FIREBIRD, and AC6 and found the region to extend over 5 \(R_{E}\) in L shell and 4 hours in MLT. However, the MLT extent was underestimated due to lack of early MLT coverage from any of the datasets, and they did not include observations of chorus waves. Lorentzen et al.[2001b] also assumed microbursts occurred over 6 hours in MLT and 2\(R_{E}\) in L shell for their global loss calculation. The overall contribution that microburst precipitation has on outer belt loss is uncertain.
To resolve this open question, we use this new database, augmented with additional satellite-borne and ground-based data, to determine the typical size and duration of this region of wave-particle interaction following a magnetic storm on 5 December 2017.