Why is sporadic-E propagation so weird?

Mid-latitude sporadic-E ("Es") clouds are transient, thin layers of dense but patchy ionization which appear in the E region of the ionosphere. The process of formation of Es is different from that of the background ionosphere and can, by comparison, produce much higher ionization densities. Sporadic-E propagation is well known to radio amateurs because it allows communication at higher frequencies and/or over shorter skip distances than is possible via the background ionosphere, but Es is also currently the subject of much academic research. This is partly because of the disruptive impact of Es on satellite communications and satellite radars, but there is also growing scientific interest in the Mesosphere / Lower Thermosphere region of the upper atmosphere, which is where Es mainly occurs. The thin, intense, and variable nature of Es means that reflected signals can have quite extreme temporal, spatial, and polarization characteristics. My PhD research showed that the reflection process at 50 MHz is primarily magnetoionic in nature, but many detailed features remain to be explained. To explore the observed polarization behaviour in more detail and to link that behaviour to the physical properties of the Es layers, a PHaRLAP-based raytrace simulation has been developed which predicts polarization parameters a signal passes through an Es cloud. As a case study, the observed significant and systematic differences in the polarization of the signals received in the UK over very similar paths from beacons in Hungary and Slovenia have been investigated in detail. Each of the two beacons shows strongly defined elliptical polarization, but the sense of rotation and predominant tilt angle are consistently opposite from each other, over multiple Es reflection events and on multiple days. This presentation will summarize the earlier work and then describe the building and testing of the PHaRLAP simulation model and the case study results obtained so far. Finally, outstanding questions about the weird nature of Es propagation will be discussed and opportunities for further work described.