Long-term variability of electron radiation dose in geosynchronous orbit

A neural network that was developed to model the temporal variations of relativistic (>3 MeV) electrons at geosynchronous orbit has been used to estimate the long-term variability of the radiation dose to geosynchronous spacecraft. The input to the neural network consists of 10 consecutive days of the daily sum of the planetary magnetic index SIGMAKp. The output is an estimate of the daily-averaged electron flux for the tenth day. The model was used to compute the daily-averaged electron flux for energies >3 MeV for each day from January 1933 through December 1988. Assuming that the average energy spectrum is independent of time, the annual radiation dose of electrons >300 keV is obtained by scaling the model output to 300 keV and summing the daily-averaged flux for each year. The annual dose is found to vary from 0.25 to 3.4 times its average value. The dose is found to correlate only weakly with the annual mean sunspot number. The results are also applicable to studies of the affect on the middle atmosphere of the long-term variability of the precipitation of these electrons into the atmosphere.