Observations of winds and gravity waves (GW) by MF radars from the Arctic t
o the Equator are used to provide frequency spectra and spectral variances
of horizontal motions, and information on the predominant azimuthal directi
ons of propagation for the waves. The years used are mainly 1993/4; the hei
ght layer 76-88 km; and the GW bands 10 100 min. and 1-6 hrs. The high/mid-
latitude locations of Troms empty set, Saskatoon, London/Urbana, Yamagawa,
generally demonstrate similar behaviour: the monthly spectra have slopes ne
ar -5/3 in winter months, but smaller (absolute) slopes at higher frequenci
es (<2 hrs.) in summer. Corresponding to this, the spectral densities (10-1
00 min.) are larger for conditions of higher mean background windspeed-this
is related by means of a new correlation-vector technique to GW propagatin
g anti-parallel to the mean zonal winds, and the closure of the solstitial
mesospheric jets. Also consistent with this, the sizes and orientations of
perturbation ovals (fitted to the wind variations), demonstrate strong semi
-annual-oscillations (SAO), and generally similar monthly and latitudinal d
irections. This suggests strong control, especially of the high-frequency G
W band, by the dominant zonal wind-structures of the mesosphere. In contras
t the low-latitude locations of Hawaii and Christmas Island demonstrate uni
quely different behaviours, with indications of significant inter-annual va
riability The frequency spectra for all months tend to have smaller slopes
at higher frequencies. Also the dependence of spectral density in both GW b
ands, upon background wind speed, is negative rather than positive, and is
shown to be generally consistent with GW propagating parallel to the mean-g
lobal winds. This is consistent with weaker vertical shears in the zonal wi
nds (76-88 km), and lower GW momentum depositions. The perturbation ovals r
eveal much weaker SAG, and more variable orientations, consistent with more
dependency upon GW sources, and less control by the mean winds of the meso
sphere.