Free Access
Issue
A&A
Volume 537, January 2012
Article Number A38
Number of page(s) 17
Section Atomic, molecular, and nuclear data
DOI https://doi.org/10.1051/0004-6361/201117592
Published online 03 January 2012

© ESO, 2012

1. Introduction

This paper is one in a series where atomic data and line identifications are benchmarked against experimental data (Del Zanna et al. 2004, Paper I). A substantial effort was devoted (within the Iron Project (IP) and the UK APAP network) in the past few years to calculate new atomic data for coronal ions and extreme-ultraviolet (EUV) wavelengths, which are important for many solar missions, in particular for the Hinode EUV Imaging Spectrometer (EIS, see Culhane et al. 2007), which covers two wavelength bands (SW: 166–212 Å; LW: 245–291 Å). New atomic data for Fe vii (Witthoeft & Badnell 2008), Fe viii (Del Zanna 2009b), Fe ix (Storey et al. 2002), Fe x (Del Zanna et al. 2004), Fe xi (Del Zanna et al. 2010), Fe xii (Storey et al. 2005), Fe xiii (Storey & Zeippen 2010), Fe xiv (Liang et al. 2010), and Si x (Liang et al. 2009) have been produced. These data represent a significant improvement over previous ones. The data for Fe ix, Fe x, and Fe xii are already available within the CHIANTI database (Dere et al. 1997), and those for Fe viii, Fe xi, Fe xiii, Fe xiv, and Si x have been included into CHIANTI v.7 (Landi et al. 2011).

These ions are emitting most of the lines in the Hinode/EIS wavelengths. The instrument was radiometrically calibrated on the ground (Lang et al. 2006) with an accuracy of about 20%, hence it is well-suited for benchmarking atomic data. The benchmark for Fe vii, Fe ix (Del Zanna 2009a), Fe viii (Del Zanna 2009b), Fe xi (Del Zanna 2010), Fe xiii (Del Zanna 2011) has recently been made also using Hinode/EIS spectra. However, the benchmark of Fe x (Del Zanna et al. 2004) and Fe xii (Del Zanna & Mason 2005) was made with limited experimental data, and many identifications were tentative.

The main aim of this paper is to reassess all previous identifications for the main coronal ions, and suggest which lines are best for diagnostic purposes for use with EIS. The work presented here is a follow-up of the preliminary discussion of the main lines in the EIS spectra given in Young et al. (2007) and is part of the assessment that is routinely made to ensure that the CHIANTI database has the best atomic data.

Brown et al. (2008) provided an extensive list of lines observed with Hinode/EIS, but line identifications were not substantiated with a quantitative analysis. Even at the high spectral resolution of EIS, most lines are blended, and blending strongly depends on the source region. The EUV range is very crowded, and the best way forward is to study clean spectra of different sources. The main hot flare lines have been identified in Del Zanna (2008) and Del Zanna & Ishikawa (2009). An almost pure EIS spectrum containing only transition-region lines was presented by Del Zanna (2009a). It allowed a detailed benchmark of low-temperature transition-region (TR) emission from a range of cool ions, including Fe vii, Fe viii and Fe ix. A similar work was independently carried out by Landi & Young (2009), although the EIS spectrum did contain a mixture of cold and warm (1 MK) emission.

This paper focuses on the coronal lines. To limit contamination from cool lines, off-limb spectra have been carefully selected for this study. This has been done because in on-disk observations of active regions a large number of lines that become visible only at high electron densities appear, and the line identification becomes even more complex. This discussion is left to a separate paper.

The identification and the atomic data for Hinode EIS are of particular relevance for many EUV imagers such as SOHO/EIT and the Solar Dynamics Observatory (SDO) Atmospheric Imaging Assembly (AIA), because many of their bands cover the same spectral region (O’Dwyer et al. 2010; Del Zanna et al. 2011). Indeed, the interpretation of imaging data relies on the accuracy of the underlying atomic data. Section 2 outlines the benchmark method. Sections 3 and 4 present off-limb active region and quiet-Sun observations. Section 5 presents a combined summary of the benchmark, while Sect. 6 discusses the relative abundances. Section 7 draws the conclusions.

2. Benchmark method

All identifications of the strongest lines were checked with laboratory and solar spectra. Line intensities, whenever available, were compared to confirm identifications and assess the possible presence of blending. For the line identification, we made use of the “emissivity ratio” technique, which divides the observed intensity of a line by its emissivity: (1)calculated at a fixed electron temperature Te (or density Ne) and plotted as a function of the density Ne (or temperature Te). Iob is the observed intensity in the line. The scaling constant C is chosen so that the curves are close to unity. The same constant is chosen for each plot. If theoretical and observed intensities agree, all Fji curves should either overlap or cross, if the plasma is nearly isodensity and isothermal. This allows one to assess at once, in one single plot, how good observed vs. theoretical intensities are for a group of lines (see Del Zanna et al. 2004 for details).

In the emissivity ratio plots below, the observed wavelength λob (Å) and the indices of the transition(s) contributing to the line are shown. The observed intensity is also shown. Whenever a line is blended with a known line, this is indicated, and the observed intensity decreased by a factor shown in the plot. For example, 95% of the line at 256.94 Å in the active region observation is caused by the 1–14 transition in Fe xi, and 5% by the 2–15 transition in Fe xii. This is indicated in the plots as λob 256.94 Å (1–14) 0.95 × Iob = 28.6 (Fe xi), and λob 256.94 Å (2–15) 0.05 × Iob = 1.5 (Fe xii), given that the total observed intensity is 30.1. When a line is a known self-blend of transitions form the same ion, this is indicated as “sbl”.

For many lines, the observed intensities are higher than the predicted ones. In these cases, the possible presence of a blend with an unidentified line is listed as “(bl u)”, and sometimes the observed intensity is decreased by a factor displayed in the plots to have its emissivity ratio agree with those of the other lines. This provides an indication for the degree to which the observed line appears to be blended. When other effects such as instrument calibration problems could be present, a question mark is shown. New identifications are shown as “N”, and tentative new ones as “TN”.

Any emission measure modelling is specifically avoided for the benchmark because of the various problems associated with it (Del Zanna et al. 2002). The emissivity ratio method was carried out for all the ions, and the main blending contribution of the various transitions to each observed line was established.

3. Active region off-limb, 2007 August 19

thumbnail Fig. 1

Monochromatic images (negative, integrated counts) of some of the strongest lines observed by EIS on 2007 Aug. 19 and formed over a range of temperatures. The box represents the area chosen to obtain a sample spectrum. The coordinates are solar X and solar Y, in arcseconds from disk centre.

thumbnail Fig. 2

Hinode EIS off-limb active region spectra recorded on 2007 Aug. 19 (abscissa: wavelengths in Å; ordinate: averaged counts per pixel). The locations of the estimated EIS CCD bias are shown as dashed lines. The locations of all the spectral lines fitted in the spectrum are also shown.

Here we consider a long-exposure (90 s) full-spectral observation of an active region at the limb on 2007 August 19. Long exposures are needed to increase the signal-to-noise, but have the drawback of increasing the incidence of cosmic rays. The data had to be processed with custom-written software by visually inspecting each exposure for cosmic ray hits. A complex data processing, which included various geometrical corrections and a wavelength calibration procedure was applied to the data as described in detail in Del Zanna (2009b). More than 200 lines were fitted with Gaussian profiles using the cfit package (Haugan 1997), and their “morphology” (the aspect of the line radiances) was examined in detail, one by one. Figure 1 shows the resulting monochromatic images for a selection of lines, and the off-limb area chosen to obtain an averaged spectrum, displayed in Fig. 2. This off-limb region is in the unresolved “background” corona outside the core of the active region (which is in the lower part of the field-of-view). This diffuse emission was first described using lower-resolution SOHO/CDS observations by Del Zanna & Mason (2003) as being pervasive and multi-thermal.

The spectrum was wavelength-calibrated using a selection of coronal lines, assuming that there are no Doppler-shifts in the lines, a common and reasonable assumption. The rest wavelengths for EUV coronal ions found in the literature and all databases (e.g. NIST) mostly originate from Behring et al. (1976), who provided a whole-Sun spectrum in the 160–770 Å range with excellent resolution (0.06 Å). The maximum difference between reference and observed lines was 5 m Å for both EIS channels, and in many cases the EIS wavelengths are actually more accurate than the Behring et al. (1976) ones.

Table 1 provides the list of the strongest lines present in this spectrum, with their measured wavelengths λo and intensities. Notice that both the intensities in terms of total counts in the lines are given, as well as the calibrated ones. This was done to highlight the fact that many intrinsically weak spectral lines that fall near the peak sensitivity of the channels do indeed have high count rates, so are very well observed. Many wavelengths are within a few mÅ from the literature values, also shown in Table 1. This agreement is remarkable. The table also clearly indicates that a considerable number of rest wavelengths need to be revised.

Figure 3 shows the EM loci curves (see Del Zanna et al. 2002 for an explanation of the method), indicating a multi-thermal plasma, as expected (Del Zanna & Mason 2003). Strong decays to the ground states have been selected for iron (Fe viii 185.2, Fe x 184.5, Fe xii 192.4, Fe xiii 202.0, Fe xiv 264.8, Fe xv 284.2, Fe xvi 263 Å) sulfur (S x 264.2, S xi 281.4, S xii 288.4, S xiii 256.7 Å), argon (Ar xi 188.8 and Ar xiv 194 Å) and calcium (Ca xiv 193.9 and Ca xv 201 Å). The curves are displayed with the Asplund et al. (2009) photospheric abundances, with the exception of sulphur, decreased by a factor of 1.7. Nearby off-limb areas show very similar EM loci curves.

thumbnail Fig. 3

EM loci curves for the 2007 Aug. 19 off-limb active region spectrum and a selection of iron (solid lines), sulfur (dashed), argon (dot dash), and calcium (dash triple dots) ions. The Asplund et al. (2009) photospheric abundances were adopted (with the exception of sulphur, decreased by a factor of 1.7).

Table 1

List of emission lines from the 2007 Aug. 19 off-limb active region spectrum.

4. Quiet Sun off-limb, 2007 March 11

Here we consider a long-exposure (90 s) observation on 2007 March 11 of the quiet Sun at the limb, the same as used for the Fe xi benchmark (Del Zanna 2010), although the data were re-processed here with visual inspection for cosmic ray hits. Figure 4 shows the monochromatic images for a selection of lines to show how they are different for each ion stage. This fact was taken into account for the assessment.

An averaged spectrum was obtained over eight exposures from an off-limb area shown in Fig. 4. This area is farther off-limb compared to the one selected in Del Zanna (2010), hence the contribution of cooler TR lines is more reduced. The averaged EIS spectrum is shown in Fig. A.1. Table A.1 provides the list of the strongest lines present in this spectrum.

thumbnail Fig. 4

Monochromatic images (negative, integrated counts) of some of the strongest lines observed by EIS on 2007 March 11 and formed over a range of temperatures. The box represents the area chosen to obtain a sample spectrum, shown in Fig. A.1. The coordinates are solar X and solar Y, in arcseconds from disk centre.

thumbnail Fig. 5

EM loci curves of the 2007 March 11 off-limb region for Ar xi (dot dash), and a selection of iron (solid lines), and sulfur (dashed) ions, using the Asplund et al. (2009) photospheric abundances, with the exception of argon, increased by a factor of two.

Figure 5 shows the EM loci curves, which indicate a nearly isothermal plasma around log T[K] = 6.15. Strong decays to the ground states were selected for iron (Fe viii 185.2, Fe x 184.5, Fe xii 192.4, Fe xiii 202.0, Fe xiv 264.8, Fe xv 284.2, Fe xvi 263 Å) sulfur (S x 264.2, S xi 281.4, S xii 288.4, S xiii 256.7 Å) and argon (Ar xi 188.8 Å). The curves are displayed with the Asplund et al. (2009) photospheric abundances, with the exception of argon, increased by a factor of two.

A few exposures were averaged to obtain a series of good spectra along the slit, crossing the solar limb. Figure A.2 shows the intensities in a few selected lines as they vary across the solar limb, to illustrate the approximate formation temperature of some of the unidentified lines. This is a rough estimate, considering that even lines from the same ion do have different profiles because of the varying density and temperature across the limb.

5. Discussion

A summary of the main findings based on the analysis of both off-limb observations now follows. The atomic data are either mentioned or are from CHIANTI v.6 (Dere et al. 2009). The indices and spectroscopic notation for each transition can be found in the cited papers or directly within the CHIANTI database.

5.1. Cooler lines

The atomic data and identifications for Fe vii have been presented in Witthoeft & Badnell (2008) and Del Zanna (2009a), although some contradicting line identifications were presented at the same time by Young & Landi (2009). Given this uncertainty, the Witthoeft & Badnell (2008) atomic data are not included in CHIANTI v.7. The atomic data and identifications for Fe viii have been discussed in Del Zanna (2009b) and have been included in CHIANTI v.7. There is hardly any Fe vii residual emission in the off-limb spectra and the Fe viii lines are weak.

The CHIANTI atomic data for Fe ix are from Storey et al. (2002). Young (2009) identified four new lines. The three main decays from the 3s2 3p4 3d2. were confirmed by Del Zanna (2009a). The fourth line was identified by Young (2009) with the main decay from the 3s2 3p5 4p configuration (13-140), observed at 197.862 Å. Del Zanna (2009a) provided a different tentative identification for the 13–140 transition, as the 194.784 Å line. This identification appears to be incorrect, given that in the off-limb spectra the observed line is too weak. There are no other observed levels from the 3s2 3p5 4p configuration, so it is not easy to identify any lines originating from this configuration.

There are consistent discrepancies between the strongest 171.0 Å resonance line and the other strong Fe ix lines in the sense that the 171.0 Å has an observed intensity too high compared to the predicted one. This could be caused by a problem in the EIS calibration at the very edge of the EIS responsivity in the SW channel. More analysis is in progress.

Finally, the cooler emission that is residually present in the spectra and caused by Si vi, Si vii, Si viii, Mg v, Mg vi, Mg vii, O vi, and O v is discussed in Del Zanna (2009a).

5.2. Fe x

thumbnail Fig. 6

Emissivity ratio curves relative to the main Fe x EUV transitions observed by Hinode EIS on 2007 Aug. 19.

thumbnail Fig. 7

Emissivity ratio curves relative to the main Fe xi EUV transitions observed by Hinode EIS on 2007 Aug. 19.

The electron scattering calculations adopted within CHIANTI for this ion are from Del Zanna et al. (2004). The identifications are also from the benchmark work described in that paper, where most lines were identified. Many weaker transitions were only tentatively identified, however, given the paucity of high-resolution solar spectra available at the time. All identifications of lines in the Hinode/EIS wavelengths proposed by Del Zanna et al. (2004) are now confirmed.

The emissivity ratio curves relative to the Fe x EUV transitions observed by Hinode EIS are shown in Fig. 6. The curves indicate very good agreement between theory and experiment, with the exception of the 257.26 Å self-blend (identified in Del Zanna et al. 2004), which has a temperature sensitivity and also a slight density dependence above log Ne [cm-3] = 9.

thumbnail Fig. 8

Emissivity ratio curves relative to the main Fe xii EUV transitions observed by Hinode EIS on 2007 Aug. 19.

5.3. Fe xi

This ion has been the most complex of all coronal ions. Final agreement between observation and theory has been achieved after a few years of work along the S-like sequence. The R-matrix scattering calculation performed as part of the Iron Project is described in Del Zanna et al. (2010). The new calculation, obtained with an optimal target, finally allowed the identifications of most transitions. The details are given in Del Zanna (2010), and the data have been made available within CHIANTI v.7. Fe xi lines represent a good fraction of all the coronal lines in the Hinode/EIS spectra, so solving the problems for this ion was necessary for this work. Figure 7 gives one example of emissivity ratio curves for this ion, confirming the results presented in Del Zanna (2010). There are some interesting temperature diagnostics for this ion, discussed in Del Zanna (2010). There is an ample range of line ratios that can be used to measure the electron density, as Fig. 7 shows.

5.4. Fe xii

CHIANTI adopted for Fe xii (n = 3 configurations) the Storey et al. (2005) R-matrix electron collisional data and the new identifications of Del Zanna & Mason (2005). For the first time, good agreement between theory and observations for the strongest lines was found, and most lines were identified. As in the Fe x case, many weaker transitions were only tentatively identified.

thumbnail Fig. 9

Emissivity ratio curves relative to the main Fe xiii EUV transitions observed by Hinode EIS on 2007 Aug. 19.

Almost all of the proposed identifications for the Hinode/EIS wavelengths are now confirmed (see Fig. 8). A few new tentative identifications are proposed here. Del Zanna & Mason (2005) gave a tentative suggestion for the identification of the 6–84 line, as a self-blend with the 5–40, observed at 191.05 Å. The 6–84 is the strongest decay from the 3s 3p3 3d configuration, from the 4D7/2. In the present spectra the 191.05 Å line is well observed, and it is true that the predicted intensity of the 5–40 transition is not sufficient to explain the observed intensity, but this line cannot be a self-blend with the 6–84 line because it is too weak. The energy difference between observed and theoretical values for the other configurations suggests a wavelength around 188 Å. There are a few possibilities, in particular that the line is blended with the strong Fe xi 188.2, 188.3 Å lines, but a few other decays from the 4D should be observable with the present deep exposures.

A search over many possibilities has resulted in one reasonable option, namely that the 6–84 is blended with the Fe x 190.04 Å line. The residual intensity is about right. The difference between observed and theoretical energies is about –17 500 cm-1. The decay from the 4D5/2 level (87) to level 7 is the second strongest line from this configuration, and there is a line at the predicted wavelength (192.17 Å, from the 4D theoretical splittings) and intensity, although it would be blended with a cooler line on-disk. The 6–87 (weaker) line would be blended at 188.68 Å. The weak decays from the 4D3/2 (level 88) to levels 8, 7 should be observable, and would be at 193.41 and 191.82 Å respectively. The former would be blended, while the latter has the proper wavelength and about the correct intensity.

The best density diagnostic is the ratio of the two self-blends at 186.88 and 195.12 Å, both of which were identified as such for the first time by Del Zanna & Mason (2005). At high densities, the 196.65 Å is the best choice, although this line is blended in on-disk observations.

5.5. Fe xiii

The most accurate collisional data for this ion have recently been published by Storey & Zeippen (2010), as part of the Iron Project. These data have been benchmarked by Del Zanna (2011), where a number of new identifications have been suggested. These data have been made available within CHIANTI v 7. Figure 9 shows the emissivity ratio curves relative to the main Fe xiii EUV transitions observed by Hinode EIS on 2007 Aug. 19. The Fe xiii lines discussed here are not significantly temperature-dependent, i.e. similar emissivity ratios are obtained for different temperatures. The observed intensities of the 246.2 and 251.95 Å are too low by a factor of 1.5 because of an EIS calibration problem (see Del Zanna 2011 and below).

Del Zanna (2011) suggested two possible identifications for a new strong self-blend (7–60, 8–60): either this self-blend is blending the Fe xi 202.42 Å or the important self-blend at 203.82 Å. Evidence that supports both choices has been found, with the quiet-Sun off-limb measurements favouring the first option, and the active region ones favoring the latter, which is adopted here. In either case, the 4–26 196.52 Å line appears to be blended in both off-limb spectra. The latter choice lets the emissivity ratio of the 203.82 Å blend agree very well with those of the 202.04, 209.62, 209.92, 261.74 Å lines, providing densities of log Ne = 8.5 and 8.35 for the active region and quiet Sun, both measurements being very close to those obtained from Si x, but slightly lower than Fe xii. The slightly higher emissivity ratio curves for the Fe xiii 200.02, 204.26, 203.16 Å lines could mean that these (weaker) lines are blended. Laboratory measurements for this ion are necessary to clarify the various blends.

5.6. Fe xiv

Significant discrepancies between observed and predicted intensities of the very strong Fe xiv EUV coronal lines existed until the calculation of Storey et al. (2000). Liang et al. (2010), as part of the UK APAP network, have performed a much more extensive R-matrix calculation that has been made available within CHIANTI v.7 and is adopted here.

Figure 10 shows the emissivity ratio curves for the off-limb active region spectrum, where the Fe xiv lines are very strong. There are obvious discrepancies between observed and predicted intensities, in particular when branching ratios are considered, as also noticed in Liang et al. (2010). The 211.3 Å is at the edge of the SW channel, and therefore it makes sense that the radiometric calibration at the edge is not correct. Notice that the very nearby Fe xiii lines do agree well, so the problem seems to be just at the edge. There appears to be another calibration problem for lines around 250 Å, as also found looking at the Fe xiii lines. The comparison improves if the intensities of the lines around 250 Å is increased by 20%.

Malinovsky & Heroux (1973) published an excellent rocket EUV spectrum, radiometrically calibrated with an accuracy of 10–20% for lines close in wavelength. Figure 11 shows the emissivity ratio curves from their data, with the Si vii contribution removed from the 274.20 Å line. The agreement is impressive, within a relative 10% for all strong lines, and confirms the Hinode/EIS calibration problem around 250 Å.

Several very weak transitions are tentatively identified as Fe xiv. The benchmark work on Fe xi (Del Zanna 2010) has suggested the presence of a new Fe xi line blending the strong Fe xiv 264.79 Å line. Indeed, in the quiet Sun off-limb spectrum the Fe xiv lines are very weak, but the Fe xi are not, and excellent agreement between observed and predicted intensities for both Fe xiv and Fe xi is obtained considering that Fe xi contributes a significant 20% to the blend. The width of the 264.79 Å line is indeed broader than the widths of the other Fe xiv lines. The exact wavelength of the Fe xi line is difficult to assess. Laboratory measurements are needed. The presence of the Fe xi line could affect Doppler measurements of the 264.79 Å line. The best density diagnostic for this ion is the ratio of the 264.79 and 274.20 Å lines, although both lines are blended.

thumbnail Fig. 10

Emissivity ratio curves relative to the Fe xiv EUV transitions observed by Hinode EIS on 2007 Aug. 19.

thumbnail Fig. 11

Emissivity ratio curves relative to the Fe xiv EUV transitions observed by Malinovsky & Heroux (1973).

5.7. Fe xvi and Fe xvii

For Fe xvi the atomic data are from Sampson et al. (1990). There are only three lines at the EIS wavelengths, the strongest of which is the 262.98 Å line. Observed and predicted intensities agree well, but again the branching ratio between the 251.06 and the 265 Å lines indicates that the 251.06 Å is too weak by 20%, again confirming the Hinode/EIS calibration problem.

It is interesting to notice that the EM loci plots (see Fig. 3) indicate that the Fe xvi emission measure is much lower than what it should be, unless the S/Fe abundance changes with temperature. This is difficult to establish, given that Fe xvi is one of the “anomalous ions” discussed in a series of papers (see the review in Del Zanna et al. 2002). See the discussion on the calcium lines below.

For Fe xvii the scattering data calculated as part of the UK APAP network by Liang & Badnell (2010), together with the identifications of the EUV lines described in Del Zanna & Ishikawa (2009), have been made available within the CHIANTI database v.7 and were used for the present study.

There is no clear evidence of any measurable Fe xvii emission in the active region spectrum. This is a significant result because it puts constraints on active-region heating models. Various weak coronal lines are present at the same wavelengths as the Fe xvii lines, as already described in Del Zanna & Ishikawa (2009).

5.8. Ca ions

The Ca xiv lines are difficult to assess because two out of the three strongest lines are significantly blended. The 183.46 Å line is a known blend with at least Ar xiv. The 186.61 Å line is a known blend with at least a strong Fe viii transition. The only line possibly not blended is the strongest, at 193.87 Å. However, the estimate of the Fe viii contribution from the 186.61 Å line suggests that the 193.87 Å is blended as well.

Several Ca xv transitions should be observable with Hinode/EIS, at 200.97, 208.72, 181.90, 176.93, 208.32 Å. However, it appears that all lines, with perhaps the exception of the strongest one at 200.97 Å, contribute very little (less than 20%) to the observed intensities.

5.9. Si ions

thumbnail Fig. 12

Emissivity ratio curves relative to the Si x EUV transitions observed by Hinode EIS on 2007 Aug. 19.

There are a few weak Si ix transitions in the EIS wavelengths. The strongest, predicted at 290.69 Å and observed at 290.70 Å is possibly the only un-blended line from this ion. The others, at 258.08, 259.76, 261.43 Å are all blended according to the present atomic data.

Liang et al. (2009), as part of the UK APAP network, has produced an extensive R-matrix calculation for Si x, finding significant differences with previous calculations. This ion is potentially very important for Hinode/EIS because it provides some of the best density diagnostics, which turn out to be slightly modified when the new atomic data, now made available through CHIANTI v.7, are used. The new atomic data for Si x provide an impressive agreement between predicted and observed intensities, as Fig. 12 shows. The best density diagnostics are the 258.37 and 261.06 lines, because they are strong, unblended and nearby.

5.10. Al ix

thumbnail Fig. 13

Emissivity ratio curves relative to the Al ix EUV transitions observed by Hinode EIS on 2007 Mar. 11.

The strongest of the Al ix lines at 284.02 Å is particularly important because it is in the blue-wing of the Fe xv 284.16 Å resonance line. The Al ix is a strong line, and work on the Fe xv line profile should take it into account. In active region conditions, the Al ix line is blended in the blue-wing of the Fe xv and is difficult to measure. The quiet Sun off-limb spectrum is better suited to study the Al ix lines. The emissivity ratio curves for that observation are shown in Fig. 13. Excellent agreement for the 282.42 and 284.02 Å is found, providing a density in line with what was found from the other ions. The very weak 280.14, which would blend an Fe xvii self-blend first identified in Del Zanna & Ishikawa (2009), appears to be even more blended in these quiet Sun conditions. The strong Al ix 284.02 Å line is also blended in solar flares with an Fe xvii line identified for the first time in Del Zanna & Ishikawa (2009).

5.11. Mg x

Brown et al. (2008) suggested the presence of a few Mg x lines in the EIS spectra. In the off-limb observations analysed here, there is no clear evidence of any significant Mg x emission. If the very weak 190.56 Å line is entirely caused by Mg x, then the other 187.07 and 187.18 Å lines are significantly blended.

5.12. Ni ions

It is probable that a few Ni x are present in the EIS spectra. Regarding Ni xi, Brown et al. (2008) lists a few possible Ni xi lines in the EIS spectra, and Landi & Young (2009) identified a few transitions, despite noting some unexplained inconsistencies, which could be caused by problems with the atomic data for this ion. The present observations indicate that most Ni xi lines are blended, but the benchmark is complicated by the presence of some unidentified transitions. At first sight, the strongest Ni xi is the 1–4 transition, with a NIST wavelength of 207.93 Å. There is indeed an observed line in the EIS spectra at 207.94 Å. Assuming that this line is caused entirely by Ni xi would predict that other Ni xi transitions (1–10: 186.98 Å; 1 − 7: 198.38 Å; 1–3: 211.43 Å) are blending some observed lines. However, there are two still unidentified lines (4–16 and 5–16) that should be well observable in the EIS LW channel, around 270 and 280 Å. There are no observed lines with the required strength, so the only option appears to be that they would blend the Si x 271.99 Å and the S xi 281.41 Å. The Si x is indeed a wide line and could be blended, but the current atomic data for S xi do not indicate that the 281.41 Å is blended. Therefore, with the current atomic data for Ni xi and S xi, the only explanation is that all the Ni xi are blended. A more in-depth benchmark of Ni x and Ni xi lines is left to a future paper.

In the case of Ni xii, Ni xiii and Ni xiv it appears that most lines, if present, are blended. For Ni xv the only line that might be observed is the strongest one at 176.74 Å. For Ni xvi the only line which might not be blended is the very weak 198.31 Å transition. Three other lines (194.02, 195.27, 185.23 Å) provide significant contributions to blended lines, the main one being the 185.23 Å one (80% of the observed line caused by Ni xvi, the rest by Fe viii). The 195.27 Å forms a branching ratio with the 185.23 Å one, and could be used in principle to remove the Ni xvi contamination from the Fe viii in on-disk active region observations, but the presence of a further unidentified coronal line complicates the matter.

5.13. S ions

thumbnail Fig. 14

Emissivity ratio curves relative to the S x EUV transitions observed by Hinode EIS on 2007 Aug. 19.

thumbnail Fig. 15

Emissivity ratio curves relative to the S xi EUV transitions observed by Hinode EIS on 2007 Aug. 19.

The comparison between predicted and observed intensities for S x is satisfactory once all the blends are taken into account, as Fig. 14 shows. The strongest and unblended line is the 1–6 264.23 Å line. The 3–12 180.73 Å line can be used to measure the electron density. The values obtained agree well with those from ions formed at similar temperatures.

The strongest and unblended line for S xi is the 1–8 281.40 Å line. It can be used to obtain the electron density in combination with the 2–7 285.82 Å or with the 3–12 246.89 Å line, as Fig. 15 shows. The densities obtained are close to those obtained from ions formed at similar temperatures. Many lines for this ion are blended. The most notable one is the 1–14 because it blends the most important density diagnostic for Hinode/EIS, the Fe xii 186.88 Å. There have been reports that densities obtained from this line are too high, which could be explained by the presence of blends. For both off-limb cases, the S xi contribution to the Fe xii 186.88 Å line is very low (3% or less), so any additional significant blending could only come from some lines that become bright at high-densities.

There is only one line from S xii, observed at 288.41 Å, so it is difficult to assess if any blending is present. The same is true for the strong S xiii 256.7 Å line, although the emission measures for both these lines (cf. Fig. 3) suggest that there is no blending.

5.14. Ar ions

thumbnail Fig. 16

Emissivity ratio curves relative to the Ar xi EUV transitions observed by Hinode EIS on 2007 Aug. 19.

Brown et al. (2008) suggested the presence of a few weak Ar xi lines in the EIS spectra. Figure 16 shows that in the off-limb active region spectra the only lines that are possibly not blended are the strongest one, at 188.806 Å, and the weaker 194.1 Å line (within 20%). These lines are intrinsically weak, but are well observed by Hinode/EIS, given the fact that they are located not far from the peak sensitivity of the multi-layer coatings for the SW channel.

Ar xiv has a typical formation temperature of log T[K] = 6.55, and a few lines are observable with EIS in active region spectra. Unfortunately, the analysis has shown that all lines seem to be significantly blended, with the exception of the 194.40 Å line. These Ar xi and Ar xiv lines are potentially very important for studying the relative argon abundance in active regions.

6. Relative abundances

The EM loci curves offer the possibility to estimate the abundances of a few elements relative to iron. The active region EM loci curves (Fig. 3) suggest that either the strongest Ca xiv and Ca xv lines are blended or that the calcium abundance is too low. Alternatively, the strongest lines are not blended, the calcium abundance is correct, the emission measure for Fe xvi is also correct, but then both the sulphur and argon abundances are much lower in higher-temperature regions. This second option is favoured here.

If this result is confirmed, it could explain some of the large discrepancies found in the literature by assuming that abundances vary at different heights. For example, observations in the X-rays, associated with hot (log T[K] = 6.5–7.0) plasmas often suggest lower “coronal” abundances when compared to observations at lower heights (transition-region temperatures) in the EUV/UV (see, e.g. Veck & Parkinson 1981), which often produce measurements close to photospheric values. A detailed discussion of these questions is deferred to a future paper.

Interestingly, the sequence of sulphur ions (S x, S xi, S xii, S xiii) presents the opportunity to measure the sulphur abundance relative to iron quite accurately. The Asplund et al. (2009) recommended photospheric abundance S/Fe ratio is 0.42. The present quiet Sun off-limb spectrum confirms this value. Gloeckler & Geiss (2007) used Ulysses measurements of solar-wind particles emanating from coronal holes and found a similar S/Fe ratio (0.52).

A range of values are found in solar flares. One of the most recent and direct line/continuum analyses of RESIK X-ray spectra (Chifor et al. 2007) produced an S/Fe ratio of 0.25, much lower than photospheric. Monsignori Fossi et al. (1994) used a SERTS 1989 rocket spectrum over an active region (on-disk) to obtain the same value, a S/Fe ratio of 0.25. The present is possibly the first measurement of the sulphur abundance in the diffuse AR emission, suggesting the same S/Fe ratio of 0.25, i.e. a first ionization potential (FIP) bias of 1.7.

A similar result was found by Feldman et al. (1998) from off-limb SOHO / SUMER observations of a quiescent streamer. Mohan et al. (2000) used off-limb SOHO / SUMER observations above an active region to find a FIP bias of 2–3 in the S/Si abundance. Brooks & Warren (2011) used Hinode/EIS observations to find a FIP bias of about 3 in the S/Si abundance in the “coronal outflow” regions, where blueshifts in coronal lines are observed.

The active region spectrum indicates an Ar/Fe relative abundance of 0.08, while the quiet Sun one indicates a value of 0.16. Other nearby regions within the two off-limb observations have been analysed, and the results are the same.

These are the first measurements obtained using Ar xi. There is no measurement of the photospheric abundance of argon, and there is ample literature on the large discrepancies between all previous solar measurements and those obtained from stars or nebulae in the solar neighbourhood. Most solar measurements have produced an abundance about a factor of two lower. For some details see, e.g. the recent Lanz et al. (2008) study. Asplund et al. (2009) used some solar measurements to recommend an Ar/Fe relative abundance of 0.08. Lanz et al. (2008) recommended instead a higher photospheric value, between 0.11 and 0.17. There are plenty of solar flare measurements based on the he-like X-ray argon lines, but they all provide low argon abundances (e.g. an Ar/Fe abundance of 0.05, Chifor et al. 2007). There are indications, however, that the argon abundance increases when low-activity X-ray spectra are examined. For example, Sylwester et al. (2010) found Ar/Fe values of 0.09. Monsignori Fossi et al. (1994) measured an Ar/Fe abundance of 0.04, from the SERTS 1989 active region rocket spectrum. Doschek & Bhatia (1990) used an Ar xi forbidden line to measure an Ar/Fe abundance of 0.031 from Skylab S082-B slit spectrograph exposures taken off-limb above an active region.

Argon is a high-FIP element, hence it is expected to show a variation in its abundance depending on the region of the solar atmosphere considered. It is expected that its abundance in active regions should be lower than in the quiet Sun. Indeed, there are measurements far off-limb of very weak argon lines observed by SOHO/UVCS that suggest sub-photospheric argon abundances above active regions (Raymond et al. 1997), although most other elements were also found to have sub-photospheric abundances.

The present EIS measurements do indicate that the argon abundance decreases, as does the sulphur one, by a factor of about two from the quiet Sun to active region diffuse emission. The present quiet-Sun measurement remarkably agrees with the value recommended by Lanz et al. (2008), hence if confirmed would end the 40-year long puzzle of the solar photospheric argon abundance.

It is obviously possible that all argon lines are blended with unidentified coronal lines in the Hinode/EIS spectra. If that was the case, the quiet-Sun argon abundance would decrease. However, any blending would still be present in the off-limb active region spectrum, where the argon abundance is already a factor of two lower and is close to the RESIK measurement of Sylwester et al. (2010). A detailed discussion of these questions is deferred to a future paper.

7. Conclusions

Spatially-resolved high-resolution EUV solar spectroscopy is very valuable for line identification purposes. After very careful data analysis, Hinode EIS spectra can provide very accurate wavelengths (down to a few m Å) and line intensities. The overall benchmark of the most recent atomic data using off-limb coronal spectra is very satisfactory, at last. All the strongest lines in the spectra have now finally been identified, and agree well (within a relative 20% or better) in terms of expected and observed line intensities when blending is taken into account. Indeed, the majority of spectral lines are blended. Only blending with coronal lines was taken into account here. Some wavelengths have been revised.

A large number of transitions still await firm identifications, however, the results presented here clearly indicate the approximate formation temperature of the unidentified lines, which will aid the identification process. Work is in progress to improve some of the atomic models for a range of ions, but new laboratory measurements are essential to confirm the many tentative identifications provided here.

A few important temperature diagnostics for Hinode/EIS were discussed in Del Zanna (2009a,b and 2010) and have not been repeated here. The various calibration problems for the Hinode/EIS instrument, highlighted here, will need to be properly assessed before these new temperature diagnostics can be used reliably. The main density diagnostics for Hinode/EIS have now been assessed. Very good agreement from different ions is found, which supports the accuracy of the new atomic data.

The results of the emission measure analysis are new and potentially important, if confirmed. A quiet-Sun argon abundance in line with Galactic measurements is found. Argon and sulphur show the well-know FIP bias, i.e. their abundances (relative to iron) are much reduced (by a factor of about two) in active region diffuse emission when compared to the quiet Sun. The emission measures at high temperatures (log T[K] = 6.4–6.7) suggest that the argon and sulphur FIP bias might increase even more with temperature. These questions will be discussed in a forthcoming paper.

Acknowledgments

Support from STFC (Advanced Fellowship and APAP network) is acknowledged. The work of the UK APAP Network was funded by the UK STFC under grant No. PP/E001254/1 with the University of Strathclyde. The anonymous referee is thanked for useful suggestions. I warmly thank B. C. Fawcett for his encouragement and help since 2003, when the present long-term benchmark programme started. CHIANTI is a collaborative project involving researchers at the Universities of: Cambridge (UK), George Mason, Michigan (USA). The excellent Hinode Science Data Centre Europe was used to search the EIS database. Hinode is a Japanese mission developed and launched by ISAS/JAXA, with NAOJ as domestic partner and NASA and STFC (UK) as international partners. It is operated by these agencies in co-operation with ESA and NSC (Norway).

References

  1. Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009, ARA&A, 47, 481 [NASA ADS] [CrossRef] [Google Scholar]
  2. Behring, W. E., Cohen, L., Doschek, G. A., & Feldman, U. 1976, ApJ, 203, 521 [NASA ADS] [CrossRef] [Google Scholar]
  3. Brooks, D. H., & Warren, H. P. 2011, ApJ, 727, L13 [Google Scholar]
  4. Brown, C. M., Feldman, U., Seely, J. F., Korendyke, C. M., & Hara, H. 2008, ApJS, 176, 511 [NASA ADS] [CrossRef] [Google Scholar]
  5. Chifor, C., Del Zanna, G., Mason, H. E., et al. 2007, A&A, 462, 323 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Culhane, J. L., Harra, L. K., James, A. M., et al. 2007, Sol. Phys., 60 [Google Scholar]
  7. Del Zanna, G. 2008, A&A, 481, L69 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  8. Del Zanna, G. 2009a, A&A, 508, 501 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  9. Del Zanna, G. 2009b, A&A, 508, 513 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  10. Del Zanna, G. 2010, A&A, 514, A41 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  11. Del Zanna, G. 2011, A&A, 533, A12 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  12. Del Zanna, G., & Ishikawa, Y. 2009, A&A, 508, 1517 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  13. Del Zanna, G., & Mason, H. E. 2003, A&A, 406, 1089 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  14. Del Zanna, G., & Mason, H. E. 2005, A&A, 433, 731 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  15. Del Zanna, G., Landini, M., & Mason, H. E. 2002, A&A, 385, 968 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  16. Del Zanna, G., Berrington, K. A., & Mason, H. E. 2004, A&A, 422, 731 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  17. Del Zanna, G., Storey, P. J., & Mason, H. E. 2010, A&A, 514, A40 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  18. Del Zanna, G., O’Dwyer, B., & Mason, H. E. 2011, A&A, 535, A46 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  19. Dere, K. P., Landi, E., Mason, H. E., Monsignori Fossi, B. C., & Young, P. R. 1997, A&AS, 125, 149 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Dere, K. P., Landi, E., Young, P. R., et al. 2009, A&A, 498, 915 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  21. Doschek, G. A., & Bhatia, A. K. 1990, ApJ, 358, 338 [NASA ADS] [CrossRef] [Google Scholar]
  22. Feldman, U., Schühle, U., Widing, K. G., & Laming, J. M. 1998, ApJ, 505, 999 [NASA ADS] [CrossRef] [Google Scholar]
  23. Gloeckler, G., & Geiss, J. 2007, Space Sci. Rev., 130, 139 [NASA ADS] [CrossRef] [Google Scholar]
  24. Haugan, S. V. H. 1997, CDS software note, 47 [Google Scholar]
  25. Landi, E., & Young, P. R. 2009, ApJ, 706, 1 [NASA ADS] [CrossRef] [Google Scholar]
  26. Landi, E., Del Zanna, G., Young, P. R., Dere, K. P., & Mason, H. E. 2011, ApJ, accepted [Google Scholar]
  27. Lang, J., Kent, B. J., Paustian, W., et al. 2006, Appl. Opt., 45, 8689 [NASA ADS] [CrossRef] [PubMed] [Google Scholar]
  28. Lanz, T., Cunha, K., Holtzman, J., & Hubeny, I. 2008, ApJ, 678, 1342 [NASA ADS] [CrossRef] [Google Scholar]
  29. Liang, G. Y., & Badnell, N. R. 2010, A&A, 518, A64 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  30. Liang, G. Y., Whiteford, A. D., & Badnell, N. R. 2009, A&A, 499, 943 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  31. Liang, G. Y., Badnell, N. R., Crespo López-Urrutia, J. R., et al. 2010, ApJS, 190, 322 [NASA ADS] [CrossRef] [Google Scholar]
  32. Malinovsky, L., & Heroux, M. 1973, ApJ, 181, 1009 [NASA ADS] [CrossRef] [Google Scholar]
  33. Mohan, A., Landi, E., & Dwivedi, B. N. 2000, A&A, 364, 835 [NASA ADS] [Google Scholar]
  34. Monsignori Fossi, B. C., Landini, M., Thomas, R. J., & Neupert, W. M. 1994, Adv. Space Res., 14, 163 [NASA ADS] [CrossRef] [Google Scholar]
  35. O’Dwyer, B., Del Zanna, G., Mason, H. E., Weber, M. A., & Tripathi, D. 2010, A&A, 521, A21 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  36. Raymond, J. C., Kohl, J. L., Noci, G., et al. 1997, Sol. Phys., 175, 645 [NASA ADS] [CrossRef] [Google Scholar]
  37. Sampson, D. H., Zhang, H. L., & Fontes, C. J. 1990, Atomic Data and Nuclear Data Tables, 44, 209 [Google Scholar]
  38. Storey, P. J., & Zeippen, C. J. 2010, A&A, 511, A78 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  39. Storey, P. J., Mason, H. E., & Young, P. R. 2000, A&AS, 141, 285 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  40. Storey, P. J., Zeippen, C. J., & Le Dourneuf, M. 2002, A&A, 394, 753 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  41. Storey, P. J., Del Zanna, G., Mason, H. E., & Zeippen, C. 2005, A&A, 433, 717 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  42. Sylwester, B., Sylwester, J., & Phillips, K. J. H. 2010, A&A, 514, A82 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  43. Veck, N. J., & Parkinson, J. H. 1981, MNRAS, 197, 41 [NASA ADS] [Google Scholar]
  44. Witthoeft, M. C., & Badnell, N. R. 2008, A&A, 481, 543 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  45. Young, P. R. 2009, ApJ, 691, L77 [NASA ADS] [CrossRef] [Google Scholar]
  46. Young, P. R., Del Zanna, G., Mason, H. E., et al. 2007, PASJ, 59, 857 [Google Scholar]

Appendix A: The quiet-Sun off-limb spectrum

thumbnail Fig. A.1

Hinode EIS spectrum (units are averaged counts per pixel) of the quiet-Sun off-limb region observed on 2007 March 11 (see Fig. 4).

thumbnail Fig. A.2

Intensities (ordinate) in a selection of Hinode/EIS lines as they vary across the solar limb (abscissa: in arcseconds along the N-S direction from disk centre, see Fig. 4). The top row shows strong well-known lines for reference, the rest are unidentified.

Figure A.1 shows the Hinode EIS spectrum of the quiet-Sun off-limb region observed on 2007 March 11. Table A.1 lists a selection of the main lines in this spectrum. Figure A.2 shows how line intensities vary across the limb, to aid the identification process. Figures A.3A.7 show the emissivity ratio curves relative to the Fe x,Fe xi,Fe xii, Fe xiii, Fe xiv and Si x for the quiet-Sun off-limb region observed on 2007 March 11.

Table A.1

List of emission lines from the 2007 March 11 off-limb quiet-Sun spectrum.

thumbnail Fig. A.3

Emissivity ratio curves relative to the main Fe x EUV transitions observed by Hinode EIS on 2007 Mar. 11. Iob: observed intensity; bl: blend; sbl: self-blend; N: new identification proposed here.

thumbnail Fig. A.4

Emissivity ratio curves relative to the main Fe xi EUV transitions observed by Hinode EIS on 2007 Mar. 11.

thumbnail Fig. A.5

Emissivity ratio curves relative to the main Fe xii EUV transitions observed by Hinode EIS on 2007 Mar. 11.

thumbnail Fig. A.6

Emissivity ratio curves relative to the main Fe xiii EUV transitions observed by Hinode EIS on 2007 Mar. 11.

thumbnail Fig. A.7

Emissivity ratio curves relative to the Fe xiv and Si x EUV transitions observed by Hinode EIS on 2007 Mar. 11.

All Tables

Table 1

List of emission lines from the 2007 Aug. 19 off-limb active region spectrum.

Table A.1

List of emission lines from the 2007 March 11 off-limb quiet-Sun spectrum.

All Figures

thumbnail Fig. 1

Monochromatic images (negative, integrated counts) of some of the strongest lines observed by EIS on 2007 Aug. 19 and formed over a range of temperatures. The box represents the area chosen to obtain a sample spectrum. The coordinates are solar X and solar Y, in arcseconds from disk centre.

In the text
thumbnail Fig. 2

Hinode EIS off-limb active region spectra recorded on 2007 Aug. 19 (abscissa: wavelengths in Å; ordinate: averaged counts per pixel). The locations of the estimated EIS CCD bias are shown as dashed lines. The locations of all the spectral lines fitted in the spectrum are also shown.

In the text
thumbnail Fig. 3

EM loci curves for the 2007 Aug. 19 off-limb active region spectrum and a selection of iron (solid lines), sulfur (dashed), argon (dot dash), and calcium (dash triple dots) ions. The Asplund et al. (2009) photospheric abundances were adopted (with the exception of sulphur, decreased by a factor of 1.7).

In the text
thumbnail Fig. 4

Monochromatic images (negative, integrated counts) of some of the strongest lines observed by EIS on 2007 March 11 and formed over a range of temperatures. The box represents the area chosen to obtain a sample spectrum, shown in Fig. A.1. The coordinates are solar X and solar Y, in arcseconds from disk centre.

In the text
thumbnail Fig. 5

EM loci curves of the 2007 March 11 off-limb region for Ar xi (dot dash), and a selection of iron (solid lines), and sulfur (dashed) ions, using the Asplund et al. (2009) photospheric abundances, with the exception of argon, increased by a factor of two.

In the text
thumbnail Fig. 6

Emissivity ratio curves relative to the main Fe x EUV transitions observed by Hinode EIS on 2007 Aug. 19.

In the text
thumbnail Fig. 7

Emissivity ratio curves relative to the main Fe xi EUV transitions observed by Hinode EIS on 2007 Aug. 19.

In the text
thumbnail Fig. 8

Emissivity ratio curves relative to the main Fe xii EUV transitions observed by Hinode EIS on 2007 Aug. 19.

In the text
thumbnail Fig. 9

Emissivity ratio curves relative to the main Fe xiii EUV transitions observed by Hinode EIS on 2007 Aug. 19.

In the text
thumbnail Fig. 10

Emissivity ratio curves relative to the Fe xiv EUV transitions observed by Hinode EIS on 2007 Aug. 19.

In the text
thumbnail Fig. 11

Emissivity ratio curves relative to the Fe xiv EUV transitions observed by Malinovsky & Heroux (1973).

In the text
thumbnail Fig. 12

Emissivity ratio curves relative to the Si x EUV transitions observed by Hinode EIS on 2007 Aug. 19.

In the text
thumbnail Fig. 13

Emissivity ratio curves relative to the Al ix EUV transitions observed by Hinode EIS on 2007 Mar. 11.

In the text
thumbnail Fig. 14

Emissivity ratio curves relative to the S x EUV transitions observed by Hinode EIS on 2007 Aug. 19.

In the text
thumbnail Fig. 15

Emissivity ratio curves relative to the S xi EUV transitions observed by Hinode EIS on 2007 Aug. 19.

In the text
thumbnail Fig. 16

Emissivity ratio curves relative to the Ar xi EUV transitions observed by Hinode EIS on 2007 Aug. 19.

In the text
thumbnail Fig. A.1

Hinode EIS spectrum (units are averaged counts per pixel) of the quiet-Sun off-limb region observed on 2007 March 11 (see Fig. 4).

In the text
thumbnail Fig. A.2

Intensities (ordinate) in a selection of Hinode/EIS lines as they vary across the solar limb (abscissa: in arcseconds along the N-S direction from disk centre, see Fig. 4). The top row shows strong well-known lines for reference, the rest are unidentified.

In the text
thumbnail Fig. A.3

Emissivity ratio curves relative to the main Fe x EUV transitions observed by Hinode EIS on 2007 Mar. 11. Iob: observed intensity; bl: blend; sbl: self-blend; N: new identification proposed here.

In the text
thumbnail Fig. A.4

Emissivity ratio curves relative to the main Fe xi EUV transitions observed by Hinode EIS on 2007 Mar. 11.

In the text
thumbnail Fig. A.5

Emissivity ratio curves relative to the main Fe xii EUV transitions observed by Hinode EIS on 2007 Mar. 11.

In the text
thumbnail Fig. A.6

Emissivity ratio curves relative to the main Fe xiii EUV transitions observed by Hinode EIS on 2007 Mar. 11.

In the text
thumbnail Fig. A.7

Emissivity ratio curves relative to the Fe xiv and Si x EUV transitions observed by Hinode EIS on 2007 Mar. 11.

In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.