These data are electron flux measurements in the Earth's radiation belt obtained with the MagEIS instrument on the NASA Van Allen Probe B satellite. They are an extension of a previously published data set (https://doi.org/10.5068/D1RQ2W). In the original dataset, the fluxes were averaged between 80 and 100 degrees local pitch angle (i.e., equatorial pitch angles between ~70 and 110 degrees) and used to study electron decay time scales or "lifetimes" (see https://doi.org/10.1029/2019GL086053 and https://doi.org/10.1029/2019GL086056). In this new data set, the data are obtained and processed in the same way, but here multiple equatorial pitch angle bins are provided. The contents of the data files are described in greater detail in a "readme" file.

These data are daily (UTC) flux averages. The L parameter is McIlwain L and was obtained from the Olson-Pfitzer quiet 1977 magenetic field model. The data are provided in L bins of 0.1L width from L = 1.0 to 7.6 and in equatorial pitch angle bins from 0 to 180 degrees with 12 degree bin width.

Three flux variables are provided:

1. FEDU: Uncorrected differential electron flux (cm² s sr keV)^^-1
2. FEDU_CORR: Background corrected (standard algorithm) differential electron flux (cm² s sr keV)^^-1
3. FEDU_CORR_ALT: Background corrected (alternative algorithm)  differential electron flux (cm² s sr keV)^^-1.

All flux arrays are 4D and have dimensions ~ [n_time x n_energy x n_L x n_alpha].

Notes on FEDU_CORR_ALT:

• For energies < 700 keV, it is identical to FEDU_CORR  (i.e., the "standard" background correction algorithm was used, doi:10.1002/2015JA021171).
• For L > 3.0, it is identical to FEDU_CORR (standard algorithm).
• For L < 3.0 and E > 700 keV, the "alternative" background correction algorithm was used (doi:10.1029/2018JA026349)
• Data from the "MED" unit, pixel 8 is not included, as it has considerable Poisson noise and there's a "HIGH" unit channel with nearly identical energy (~1 MeV, HIGH-P0).

Notes (general)

• The lowest L bin (L=1.0-1.1) is set to fill in all of the flux variables. 
• Data from LOW/MED pixel 0 (P0) and pixel 1 (P1) are not included, as they are noisy (see https://doi.org/10.1007/s11214-021-00855-2).
• Days with HIGH unit operational testing at end of May 2014 (all pixels) and July 2013 (pixel 3 only) are set to fill (see https://doi.org/10.1007/s11214-021-00855-2).
• Data on 2017/12/07 is set to fill due to missing magnetic ephemeris