NBP2 -Table 5.3

Table 5.3: Summary of frequencies (mHz) and correlations for NBP2. See Section 5.3.2 for details.

	From:	1.00	1.33	1.67	2.00	2.33	2.67	3.00	3.33	3.67	4.00	4.33
λ	To:	1.22	1.56	1.89	2.22	2.56	2.89	3.22	3.56	3.89	4.22	4.55
Ca II K3							99,95	99,95	99,95	99	95	95
Hα		99	99		99					95		95
Mg I b2			99,99	99,95					99,99			95
Mg I b1-0.4 Å		95,99	95,95	95,95	95

Wavelengths		Correlation Coefficient	Frequency (mHz)	Timelag (Units of 45 s)	Speed (km s-1
λ1	λ2
Ca II K3	Hα	0.72 - 0.76	2.4 - 2.6	0 - 4	0-17+
Ca II K3	Mg I b2	0.66 - 0.78	1.25 - 1.45	2 - 4	6-14
		0.68 - 0.68	1.75 - 1.95	3 - 8	3-10
Ca II K3	Mg I b1-0.4 Å	0.67 - 0.78	1.2 - 1.4	1 - 3	12-40
		0.70 - 0.73	1.7 - 1.9	7 - 11	3-6
Hα	Mg I b2	0.65 - 0.67	2.97 - 3.17	1 - 2	5-22
		0.63 - 0.65	3.85 - 4.05	3	3-7
Hα	Mg I b1-0.4 Å	0.61 - 0.62	3.2 - 3.4	5	4-7
		0.62 - 0.63	4.7 - 4.9	-2	11-16
Mg I b2	Mg I b1-0.4 Å	0.83 - 0.93	1.3 - 1.5	0 - 1	11+
		0.95 - 0.97	1.6 - 1.8	4 - 6	2-3
		0.94 - 0.94	4.2 - 4.4	-2	5-6

Correlations of Ca II K₃ to Mg I b₂ and Mg I b₁-0.4 Å, and Mg I b₂ to Mg I b₁-0.4 Å, reveal a possible travelling wave at ~1.3 mHz. The FFTs also show power at this frequency, up to the Hα level. There is some power at the Ca II K₃ level, but it appears inside the COI and hence is not included here. However, Ca II K₃ does contain power around 2.8 mHz, and there is some correlation around 2.6 mHz between Ca II K₃ and Hα. This is all consistent with Tests 1 to 4, excepting the fact that there is no correlation from Hα to Mg I b₂. However, this lack of correlation shows the benefit of multi-wavelength analysis. Although the Hα to Mg I b₂ correlation is missing, we can safely rely on correlations across the other wavelengths. It is worth noting that a transverse frequency of 1.4 mHz, and longitudinal frequency of 2.8 mHz, are low compared to the suggestions of Kalkofen (1997).

The Ca II K₃ to Mg I b₂ and Mg I b₁-0.4 Å correlations, along with that of Mg I b₂ to Mg I b₁-0.4 Å, also show a possible ~1.8 mHz wave. There is some power at this frequency in the lower lines and a weak peak in Hα at ~3.8 mHz, with a possible travelling wave at ~3.9mHz (Hα to Mg I b₂). However, there is no peak in the higher lying lines at the original frequency. Again, there is some power below 1.3 mHz, with no apparent correlation, and several frequencies seem to be localised in height (e.g., ~3.4 mHz in Mg I b₂).

There are several other correlations evident between the lower lines. However in many cases, when the frequencies of these correlation peaks are compared to the FFT tables, there is no associated power in the FFT, or wave-packets in the wavelet power transforms. This demonstrates the advantage of using wavelet analysis to obtain temporal information and allows us to dismiss the 3.07 mHz (Hα to Mg I b₂), 3.3 mHz and 4.8 mHz (Hα to Mg I b₁-0.4 Å) correlations. The 4.3 mHz correlation (Mg I b₂ to Mg I b₁-0.4 Å) does correspond to a possible downward- moving wave-packet. Waves at around this frequency have been noted before (Bocchialini & Baudin 1995).