Two Atmospheric Effects in the Orbital Acceleration of Artificial Satellites

Abstract

As I first pointed out¹, the orbital acceleration of artificial Earth satellites shows fluctuations which cannot be explained on gravitational grounds. In satellite 1957 β 1, these oscillations were rather irregular, although cycles of 19 and 37 days could be discerned^2,3. According to E. C. Cornford⁴, a periodicity of about 25 days was indicated, and this gave rise to speculations about possible tidal effects from the Moon. The same hypothesis was tentatively advanced by D. G. King-Hele⁵ to explain oscillations with a period of 27 days in satellite 1958 δ 1. Earlier³, in collaboration with R. E. Briggs, I had found oscillations with a period of about 30 days in satellite 1958 β 2, but had rejected the tidal hypothesis in view of the many observed irregularities and also because lunar tides should result in oscillations with about half the observed period. It was suggested, rather, that the rotation of the Sun, which has a synodic period of 27 days, could be responsible for changes in upper-atmosphere densities through variable radiation.