Probabilistic Short-wave Fadeout Detection in SuperDARN Time Series Observations

Short-wave fadeout (SWF) is one of the first space weather effects to occur in the ionosphere following a solar flare and leads to severe disruption of ionospheric HF systems. The disruption is produced by flare-enhanced energetic radiations that penetrate to the D-layer where they enhance ionization that leads to heavy absorption of high-frequency (HF, 3–30 MHz) radio signal over much of the dayside for an hour or more. In this paper, we describe two probabilistic anomaly detection schemes that have been used to detect SWF events produced by M and X class flares in Super Dual Auroral Radar Network (SuperDARN) observations. The two schemes are based on statistical Z-score and nonlinear energy operators. Performance of the detection schemes varies with flare intensity and parameters of the detection schemes. We find a correlation coefficient ~0.73 between flare counts per month and SWF counts per month detected using the Z-score scheme.