MULTIINSTRUMENT OBSERVATIONS OF MESOSPHERIC MOTIONS OVER ARECIBO - COMPARISONS AND INTERPRETATIONS

-The MF/HF partial-reflection technique of observing the mesosphere and lower thermosphere has been employed for more than two decades to measure motions, but there has never been complete agreement as to what motions were being detected. This paper reports on observations made during a major international campaign-AIDA '89-that was initiated with the objective of resolving this question. The partial-reflection system employed was an Imaging Doppler Interferometer operating at 3.175 MHz, but it stands here as a prototype for all MF/HF partial-reflection radar systems: its raw data were analyzed both in its own basic mode, derived on the assumption that it sees wind-borne multiple scattering centers, and in modes adopted by other interferometric and 'spaced antenna' systems. The motions thus revealed are compared here with those found by what we consider to be more certain measurers of winds: an incoherent-scatter radar at heights of 65 95 km, a meteor-wind radar at heights of 80-100 km, and a Fabry-Perot interferometer measuring 0(1S) emissions near a height of 97 km. Comparisons of the different sets of observations oblige us to conclude that (1) MF/HF partial-reflection systems may be expected to give a good representation of ambient winds up to a height of about 80 km; (2) they fail to give a consistently reliable measurement of the ambient winds above a height of about 80 km; (3) they yield, at the greater heights, what appears in our data to be some convolution of the horizontal phase velocities of atmospheric gravity waves, with the wave spectrum having been modified by passage through the underlying wind system and containing, on occasion, locally generated Kelvin-Helmholtz waves; and (4) when the underlying winds change, the local wave spectrum will change in response and, in MF/HF partial-reflection measurements, will give the appearance of a changing local wind: if the underlying winds undergo tidal changes, the wave spectrum will undergo tide-like changes that will masquerade as true tidal winds. These results are, of course, limited to a single site over a limited period of observation. Nevertheless, taken at face value they suggest that current methods of data reduction are inappropriate for partial-reflection velocities at heights above 80 km, and that new methods of data reduction, perhaps extending certain older methods that have been applied successfully in the past to total-reflection measurements, should be employed in their place if the full potential of the MF/HF partial-reflecton technique is to be realized.