# Anomalous O IV line ratios in the solar

- Slides: 10

Anomalous O IV line ratios in the solar spectrum Alessandra Giunta 07 -10 -11 ADAS workshop, Auburn University 1

O IV lines in the solar spectrum Intensity So. HO/SUMER Wavelength/Å Intensity So. HO/CDS Wavelength/Å Intensity Hinode/EIS Wavelength/Å 07 -10 -11 ADAS workshop, Auburn University 2

Problem Muglach et al. (2010) found that the measured O IV I(787. 7 Å)/I(279. 9 Å) ratio is lower than the predicted ratio by a factor 2 -5 and the derived electron temperature is in the range of log. T=5. 58 -6. 07, much higher than the formation temperature in ionisation equilibrium. ▪ They used: - observations taken during SUMER/EIS campaign in April 2007 - atomic data from CHIANTI v. 5. 2. 1. ▪ They investigated various possible causes: - instrumental and observational effects - presence of additional processes (photo-excitation, self-absorption, non-Maxwellian distribution). ▪ They did not solve the discrepancy, with the implication that non equilibrium ionisation state may be present. 07 -10 -11 ADAS workshop, Auburn University 3

Approach ▪ A new observational data set taken during the joint SUMER/CDS/EIS campaign of April 2009 is explored. ▪ The ratio I(787. 7 Å)/I(279. 9 Å) is re-examined and other three O IV ratios are investigated: - I(554. 5Å)/I(279. 9 Å) temperature dependent - I(787. 7 Å)/I(554. 5 Å) temperature dependent - I(787. 7 Å)/I(b 790 Å) temperature independent ▪ The atomic data for these ratios are revised comparing three datasets: - CHIANTI v. 5. 2. 1 - Revised (CHIANTI + Sampson, 1995) - Aggarwal & Keenan (2008) 07 -10 -11 ADAS workshop, Auburn University 4

Observational domain ▪ Observational data taken from SUMER, CDS and EIS have been calibrated using the standard procedures. ▪ The co-alignment has been done using common emission lines or lines emitted from the same ion. ▪ The cross-calibration has been performed through direct line comparison and spectroscopic methods. ▪ The observed R=I(787. 7 Å)/I(279. 9 Å) ratio obtained is: log. R=1. 8 -2 -1 (the intensity units are photons cm s ) 07 -10 -11 ADAS workshop, Auburn University 5

Atomic domain Three atomic datasets are examined: ▪ CHIANTI ▪ Revised ▪ Aggarwal & Keenan (2008) CHIANTI The main source for the effective collision strengths (Y-values) is Zhang et al. (1994) for the first 15 levels and Sampson (1995) for the other 110 levels. However only transitions which involve the first 5 levels are included in the model. A-value for transitions amongst levels from 6 to 15 are also included, but not the corresponding Y-values. Revised This dataset has been built up merging the CHIANTI data with the DW calculations of Sampson to add the whole set of missing Y-values. Aggarwal & Keenan (2008) 3+ These are the most recent R-matrix calculations for O available in literature and include transitions which involve the n=4 levels. 07 -10 -11 ADAS workshop, Auburn University 6

Atomic data comparison The main difference lies in the population of the upper level (2 s 23 s 2 S 1/2), called N 279, from which the line at 279. 9 Å originates. rv = Revised ch = CHIANTI ak = Aggarwal & Keenan 07 -10 -11 ADAS workshop, Auburn University 7

Results Comparison between the observed and theoretical values for the ratio I(787. 7 Å)/I(279. 9 Å). The electron temperature derived in this work is log. T=5. 17 -5. 39, close to the temperature expected under the assumption of ionisation equilibrium. rv = Revised ch = CHIANTI ak = Aggarwal & Keenan log. T=5. 58 -6. 07 formation log. T=5. 17 -5. 39 temperature 07 -10 -11 ADAS workshop, Auburn University 8

Other line ratio comparison Temperature dependent line ratios I(554. 5Å)/I(279. 9 Å) As further check, the three other O IV ratios are illustrated, comparing the measurements with theoretical data. Since the discrepancy is caused by the intensity of the line at 279. 9 Å, the I(554. 5Å)/I(279. 9 Å) ratio shows the same behaviour as I(787. 7Å)/I(279. 9 Å), unlike the other two ratios. Temperature independent line ratio log. T=5. 17 I(787. 7 Å)/I(b 790 Å) I(787. 7 Å)/I(554. 5 Å) rv = Revised ch = CHIANTI ak = Aggarwal & Keenan 07 -10 -11 ADAS workshop, Auburn University 9

Discussion The Muglach et al. (2010) O IV discrepancy has been solved, however some issues arise from this work: ▪ the uncertainties of atomic data and model can affect substantially the interpretation of the observed emission ▪ the completeness (complete system of A-values and Y-values) and top-up does matter in the population model ▪ new R-matrix calculation for O 3+ are desirable to add more confidence in the currently available atomic data. 07 -10 -11 ADAS workshop, Auburn University 10