An Algorithm to Separate Ionospheric Turbulence Radar Echoes From Those of Meteor Trails in Large Data Sets

Coherent scatter echoes from disintegrating meteors and from the unstable ionospheric E-region can overlap considerably between 90 and 110 km altitudes. As the physical origin of plasma irregularities produced by meteor trails differs starkly from that of E-region auroral irregularities, this has consequences for winds as well as electrodynamic studies, thereby introducing a need to distinguish between the two types of echoes. To that goal, we have developed a novel separation algorithm to automatically sort through arbitrarily large data sets in the region of overlap. This proves very useful when the 3D location of echoes is available. The algorithm uses a definition of crowding, or clustering, in both time and space and has been developed and tested with a comprehensive data set obtained from the recently built Canadian icebear 3D radar. We discuss the characteristics belonging to the two classes of echoes, and present statistical results about the location of each type of echo as a function of conditions. Our proposed algorithm can be applied to any coherent scatter echo data with high resolution 3D location information.