4 ms·
131 and 304 Å Emission Variability Increases Hours Prior to Solar Flare Onset
- ggm 2y ago> We find significantly increased emission variability in the 2–3 hr before flare onset Is this within the band of existing "advance notice of flare" signalling? Without being a competent astronomer (soler or otherwise!) I can't tell if this in itself is significant, or if the analytics of why these spectral elements change behaviour is the (sole) key point.