A quiescent prominence observed in the H-alpha line by the COMP-S instrument at the Lomnicky Peak Observatory


Schwartz, P.1, Rybak, J.1, Kucera, A.1, Kozak, M.1, Ambroz, J.1, Gömöry, P.1

1 Astronomical Institute of the Slovak Academy of Sciences, 05960 Tatranská Lomnica, Slovakia


Abstract: 
A prominence above the NEE limb was observed by the COMP-S instrument attached to the ZEISS coronagraph located at the Lomnicky Peak Observatory. Observations were carried out on Nov 2, 2011 between 14:01:13 and 14:11:34 UT. The filter of the instrument was tuned during measurements sequentially in five wavelengths within the profile of the H-alpha line: 0, +-1, +-2 A around 6563 A. FWHM of the transmission function of the filter was ~0.4 A at these wavelengths. Data were not absolutely calibrated, therefore they could be fitted using only a simple cloud model (1D geometry, a complete frequency redistribution, a source function independent of the optical depth) to diagnose the prominence plasma. As five wavelength points in the profile were not enough for an automatic fitting, five positions at the prominence were chosen for the analysis. Observed data from the five positions were simulated using the cloud model and groups of difeerent models were found for each position. It means that the wavelength scale of a step as large as 1 A when used for the H-alpha line is not fine enough for estimation of a correct and unique model. Simulating observations using three difeerent finer wavelength scales it was found that the wavelength scale with a step of 0.3 A and even more finer in the line core (step of 0.1 A) is already suitable for more precise and unambiguous plasma diagnostics. We also show that for correct plasma diagnostics it is crucial to take into account an effect of a finite width of the transmission function of the filter. If observed data were fitted irrespectively of this important effect, an error in estimated model parameters could exceed even 100%, except for the Doppler velocities, for which the error would be much smaller, e.g. for velocities up to 20 km/s the error is below 1%.