2.3 Particle flux measurements
Electron flux measurements from the Medium Energy Proton and Electron Detector (MEPED) instruments on the POES/MetOp satellites (NOAA15, NOAA18, NOAA19, METOP01, and METOP02) provide a measure of strong electron flux in the bounce loss cone (BLC). This flux is expected to increase in regions where strong microburst precipitation is occurring. While the 2-second cadence of these measurements cannot resolve microbursts, their excellent spatial coverage is invaluable for identifying the likely size of the precipitation region over time. It should be noted that POES/MetOp has limited spatial coverage between 10 and 15 MLT (see Figure S1) in years after the MEPED instrument aboard NOAA16 was no longer in operation (2005). The POES/MetOp Space Environment Monitor (SEM/2) MEPED data used in this study are calibrated as described by Pettit et al. [2021] and is referred to as the MEPED Precipitating Electron (MPE) data set. The POES MEPED data uses three overlapping integral energy channels of with energies of 30 -2500 keV, 100-2500 keV and 300-2500 keV respectively. An additional, virtual high-energy channel (> 700 keV) can be utilized by using one of the proton channels aboard MEPED that is contaminated with relativistic electrons [Yando et al., 2001]. From there, a differential electron flux spectrum using the 4 electron channels spanning an energy range of 27 keV – 8.9 MeV with 27 differential energy levels is created. A flux vs. energy threshold profile for the satellites was developed, which was then compared to the estimated BLC fluxes for energies between 84.7 keV - 320.8 keV (energies commonly scattered by chorus). The threshold profile was generated by filtering all the MPE spectral data from 2014 - 2020 using a threshold daily Kp-index of 5 or greater. From the filtered data a median was calculated for each energy level available in the MPE spectral data set. We then applied the threshold on the days of interest by removing all BLC flux that did not meet or exceed the threshold median. A median was chosen in lieu of a mean since BLC flux does not follow a Gaussian distribution. Since the MPE data set measures all precipitating flux occurring on a given day, the purpose of this analysis was to remove lower levels of BLC flux in an attempt to isolate precipitation caused my chorus waves. However, precipitation caused by other means (EMIC waves, etc.) may still show up in the results.