A comparison of predicted and measured times of propagation over high latitude paths for which direction of arrival measurements are available

Computer-based, long-term HF prediction methods employ numerical maps of the ionosphere which are frequently based on vertical ionospheric soundings. Furthermore, they assume that propagation occurs over the great circle path (GCP) between transmitter and receiver. Recent observations of signals propagated over two paths which lie tangential to, and equatorward of, the auroral oval indicate that propagation is commonly observed at times which differ appreciably from the predicted `path open´ times. Measurements of the direction of arrival of these signals indicate that at times when propagation is predicted to occur the signals arrive close to the great circle direction, whereas when propagation is not predicted the signals travel along off-great circle paths, sometimes with large bearing displacements of up to 100°. At these times the signals are thought to reflect from the electron density gradients associated with the mid-latitude trough, a band of electron density depletion in the night-time ionosphere which forms just equatorward of the auroral oval. The days on which these effects occur depend on the formation and intensity of the trough which is associated with the level of geomagnetic activity