Faraday rotation data from a pair of polarimeters located in Tucson, A
rizona during the first half of 1990 have been examined for evidence o
f traveling ionospheric disturbances. With the use of digital filterin
g techniques, oscillations in total electron content (TEC) with period
s ranging from 20 to 90 minutes and amplitudes of 1-5 x 10(16) m(-2) h
ave been detected in the data. When present, the oscillations persist
for varying time intervals (usually several hours) and show considerab
le variation from day to day. Although the viewing geometry (nearly al
ong the local magnetic field line) is not ideal for observation of TID
s, these oscillations do appear to be real ionospheric oscillations. W
e have examined the possibility that the oscillations are not caused b
y actual changes in electron content but rather are due to changes in
the Faraday rotation resulting from the motion of the ionospheric plas
ma along a magnetic field line and the resulting change in effective m
agnetic field strength. Our calculations indicate waves with displacem
ent amplitudes of a few tens of kilometers would be sufficient to expl
ain the observations. We also examined the possibility that the oscill
ations could be due to actual changes in TEC by using a one-dimensiona
l F region model and imposing a plausible wind oscillation (simulating
a gravity wave). We found that a wave with a velocity amplitude of 15
m s(-1) was also sufficient to explain the observations. We expect th
at the observations are best explained by a combination of the two eff
ects, that is, an actual oscillation in electron content and an oscill
ation in Faraday rotation, both caused by gravity waves in the thermos
phere. We estimate that for the 90-min oscillations, 25% of the amplit
ude is due to the Faraday rotation effect and 75% is due to an actual
change in TEC.