Energetic electron pitch angle distribution parameters at 6.6 Re, as deduced from GOES X-ray observations

X-ray sensors that measure the Sun's radiant output in two soft X-ray channels, 1-8 and 0.5-4 Å, are carried on all GOES geostationary equatorial weather satellites. A comparison of X-ray measurements from two co-operational GOES reveals a systematic difference signal that shows periodic diurnal and seasonal variations. These effects are seen during geomagnetically quiet times as well as disturbed times and are most noticeable when solar activity is low to moderate. The GOES orbit lies just above the main outer electron belt of the van Allen radiation belts but it falls inside the region containing >2MeV trapped electrons; thus the local particle environment includes electrons of sufficient energy to cause significant Bremsstrahlung on the walls of the ion chamber as well as direct deposition of energy through the entrance aperture. These background effects occur despite passive shielding of the ion chambers and in-orbit electronic suppression of the spurious particle contribution. However, because of the regularity of the difference signal it is possible to exploit this X-ray contaminant to infer certain properties of the energetic electron pitch angle distribution in anisotropy and in local time, on the assumption that these energetic electrons are responsible for the spurious X-ray detector response. The basic attributes of the observed diurnal and seasonal effects can be re-created in a model that incorporates a tilted dipole magnetosphere and local-time-dependent, generic pitch angle distributions. It is possible to infer the anisotropy index, n, for dayside sin ⁿ(α) distributions and the anisotropy index, m, for nightside sin ^m(2α) butterfly distributions as well as the local times where these distributions convert from normal loss-cone to butterfly in the afternoon and return to normal loss-cone in the morning. Examples of the diurnal and seasonal variations in the observed X-ray difference signal are shown, and these waveforms are re-created by a model simulation.