Analysis of High-Resolution Wind Fields of the Upper Atmosphere Observed with a Multistatic Meteor Radar Network

Observing the winds in the mesosphere and lower thermosphere (MLT) is crucial to understanding the energetics and coupling processes of the upper atmosphere, but they are difficult to measure with current techniques. The altitude range of interest limits in-situ measurements to rockets, which are infrequent and localized. Optical remote sensing instruments are sparsely-located, give only a local view, and are mostly limited to nighttime observations. Recent advances in meteor radars, which observe winds from about 80 to 100 km altitude through the Doppler shift of drifting meteor trails, hold the promise of temporally and spatially continuous coverage. New coded continuous wave (CW) meteor radars operating in a MIMO configuration with multiple inputs (transmitters) multiple outputs (receivers) collect enough statistics to estimate the complete MLT wind field over a wide area. A demonstration campaign with a prototype coded-CW specular meteor radar network called SIMONe was conducted in early November 2018 in northern Germany in a joint effort between the Leibniz Institute of Atmospheric Physics (IAP) and MIT Haystack Observatory. We analyzed 24 hours of data from this campaign to assess the wind observation capabilities of the system and test wind field estimation techniques. We computed altitude-resolved mean horizontal winds averaged over the radar field of view using the standard linear least squares method and observed effective resolutions of approximately 30 minutes and 1 km altitude. By applying a new Gaussian process regression technique, we estimated the complete MLT wind field and found it to be consistent with the mean winds. The wind field estimation technique was also tested with simulated data and shown to reliably reproduce mesoscale features. We conclude that coded-CW MIMO meteor radar networks, along with the new wind field estimation technique, show great promise for resolving outstanding questions about MLT dynamics.