Variation of Emission Line Width in Mid- and High-Latitude Corona

Spectroscopic studies of the solar corona, using the high spatial and spectral resolution 25-cm coronagraph at the Norikura Solar Observatory for equatorial off-limb observations, indicated that the variation of radiance and line width with height is different for different temperature lines. The line width of the forbidden red emission line [Fe x] 6374 Å was found to increase with height, and that of the green emission line [Fe xiv] 5303 Å decreased with height. This had been interpreted in terms of the interaction between different temperature plasmas but needed to be confirmed. Further observations were made on several days during 2004, in two emission lines simultaneously covering the mid-latitude and polar regions to investigate the existence of the observed variation in other parts of the solar corona. In this study, we have analysed several raster scans that cover mid- and high-latitude regions of the off-limb corona in all four bright emission lines [Fe x] 6374 Å, [Fe xi] 7892 Å, [Fe xiii] 10747 Å, and [Fe xiv] 5303 Å. We find that the FWHM of the red line increases with height and that of the green line decreases with height, similar to the observations in the equatorial regions. The line widths are higher in the polar regions for all of the observed emission lines except the green line. Higher values of FWHM in polar regions may imply higher non-thermal velocities, which could be further linked to a non-thermal source powering the solar-wind acceleration, but the reason for the behaviour of the green emission line remains to be explored.     

Modified Homogeneous Data Set of Coronal Intensities

The Astronomical Institute of the Slovak Academy of Sciences has published the intensities, recalibrated with respect to a common intensity scale, of the 530.3 nm (Fe xiv) green coronal line observed at ground-based stations up to the year 2008. The name of this publication is Homogeneous Data Set (HDS). We have developed a method that allows one to successfully substitute the ground-based observations by satellite observations and, thus, continue with the publication of the HDS. For this purpose, the observations of the Extreme-ultraviolet Imaging Telescope (EIT), onboard the Solar and Heliospheric Observatory (SOHO) satellite, were exploited. Among other data the EIT instrument provides almost daily 28.4 nm (Fe xv) emission-line snapshots of the corona. The Fe xiv and Fe xv data (4051 observation days) taken in the period 1996?–?2008 have been compared and good agreement was found. The method to obtain the individual data for the HDS follows from the correlation analysis described in this article. The resulting data, now under the name of Modified Homogeneous Data Set (MHDS), are identical up to 1996 to those in the HDS. The MHDS can be used further for studies of the coronal solar activity and its cycle. These data are available at http://www.suh.sk .     

The 3D Geometry of Active Region Upflows Deduced from Their Limb-to-Limb Evolution

We analyze the evolution of coronal plasma upflows from the edges of AR 10978, which has the best limb-to-limb data coverage with Hinode’s EUV Imaging Spectrometer (EIS). We find that the observed evolution is largely due to the solar rotation progressively changing the viewpoint of nearly stationary flows. From the systematic changes in the upflow regions as a function of distance from disc center, we deduce their 3D geometrical properties as inclination and angular spread in three coronal lines (Si vii, Fe xii, and Fe xv). In agreement with magnetic extrapolations, we find that the flows are thin, fan-like structures rooted in quasi separatrix layers (QSLs). The fans are tilted away from the AR center. The highest plasma velocities in these three spectral lines have similar magnitudes and their heights increase with temperature. The spatial location and extent of the upflow regions in the Si vii, Fe xii, and Fe xv lines are different owing to i) temperature stratification and ii) line of sight integration of the spectral profiles with significantly different backgrounds. We conclude that we sample the same flows at different temperatures. Further, we find that the evolution of line widths during the disc passage is compatible with a broad range of velocities in the flows. Everything considered, our results are compatible with the AR upflows originating from reconnections along QSLs between over-pressure AR loops and neighboring under-pressure loops. The flows are driven along magnetic field lines by a pressure gradient in a stratified atmosphere. Our interpretation of the above results is that, at any given time, we observe the superposition of flows created by successive reconnections, leading to a broad velocity distribution.     

Coronal Diagnostics from Narrowband Images Around 30.4 nm

Images taken in the band centered at 30.4 nm are routinely used to map the radiance of the He?ii Ly?α line on the solar disk. That line is one of the strongest, if not the strongest, line in the EUV observed in the solar spectrum, and one of the few lines in that wavelength range providing information on the upper chromosphere or lower transition region. However, when observing the off-limb corona, the contribution from the nearby Si?xi 30.3 nm line can become significant. In this work we aim at estimating the relative contribution of those two lines in the solar corona around the minimum of solar activity. We combine measurements from CDS taken in August 2008 with temperature and density profiles from semiempirical models of the corona to compute the radiances of the two lines, and of other representative coronal lines (e.g. Mg?x 62.5 nm, Si?xii 52.1 nm). Considering both diagnosed quantities from line ratios (temperatures and densities) and line radiances in absolute units, we obtain a good overall match between observations and models. We find that the Si?xi line dominates the He?ii line from just above the limb up to ≈?2?R _⊙ in streamers, while its contribution to narrowband imaging in the 30.4 nm band is expected to become smaller, even negligible in the corona beyond ≈?2?–?3?R _⊙, the precise value being strongly dependent on the coronal temperature profile.     

Velocities and Linewidths in the Network and Cell Interiors of a Polar Coronal Hole Compared with the Quiet Sun

The relative Doppler velocities and linewidths in a polar coronal hole and the nearby quiet-Sun region have been obtained from the Solar and Heliospheric Observatory (SOHO)/Coronal Diagnostic Spectrometer (CDS) observations using emission lines originating at different heights in the solar atmosphere from the lower transition region (TR) to the low solar corona. The observed region is separated into the network and the cell interior, and the behavior of the above parameters were examined in the different regions. It has been found that the histograms of Doppler velocity and width are generally broader in the cell interior than in the network. The histograms of Doppler velocities of the network and cell interior do not show significant differences in most cases. However, in the case of the quiet Sun, the Doppler velocities of the cell interior are more blueshifted than those of the network for the lowermost line He?ii 304 Å, and an opposite behavior is seen for the uppermost line Mg?ix 368 Å. The linewidth histograms show that the network–cell difference is more prominent in the coronal hole. The network has a significantly larger linewidth than the cell interior for the lowermost TR line He?ii 304 Å for the quiet Sun. For the coronal hole, this is true for the three lower TR lines: He?ii 304 Å, O?iii 599 Å, and O?v 630 Å. We also obtained the correlations between the relative Doppler velocity and the width. A mild positive correlation is found for the lowermost transition-region line He?ii 304 Å, which decreases even more or become insignificant for the intermediate lines. For the low coronal line Mg?ix 368 Å, the correlation becomes strongly negative. This might be caused by standing waves or waves propagating from the lower to the upper solar atmosphere. The results may have implications for the generation of the fast solar wind and coronal heating.     

On-Orbit Sensitivity Evolution of the EUV Imaging Spectrometer on Hinode

Since its launch on 22 September 2006, the EUV Imaging Spectrometer onboard the Hinode satellite has exhibited a gradual decay in sensitivity. Using spectroheliograms taken in the Fe viii 185.21 Å and Si vii 275.35 Å emission lines in quiet regions near Sun center we characterize that decay. For the period from December 2006 to March 2012, the decline in the sensitivity can be characterized as an exponential decay with an average time constant of 7358±1030 days (20.2±2.8 years). Emission lines formed at temperatures ??10^6.1 K in the quiet Sun data exhibit solar-cycle effects.     

Supernova remnants, planetary nebulae and the distance to NGC 4214

We present narrow band, continuum subtracted Hα, [S ii], Hβ, [O iii] and [O ii] data taken with the Wide Field Camera 3 on the Hubble Space Telescope in the nearby dwarf starburst galaxy NGC 4214. From these images, we identify seventeen new planetary nebula candidates, and seven supernova remnant candidates. We use the observed emission line luminosity function of the planetary nebulae to establish a new velocity-independent distance to NGC 4214. We conclude that the PNLF technique gives a reddening independent distance to NGC 4214 of 3.19±0.36 Mpc, and that our current best-estimate of the distance to this galaxy ids 2.98±0.13 Mpc.     

Prominence Cavity Regions Observed Using SWAP 174 Å Filtergrams and Simultaneous Eclipse Flash Spectra

SWAP images from PROBA2 taken at 174 Å in the Fe ix/x lines are compared with simultaneous slitless flash spectra obtained during the solar total eclipse of 11 July 2010. Myriad faint low-excitation emission lines together with the He i and He ii Paschen α chromospheric lines are recorded on eclipse spectra where regions of limb prominences are obtained with space-borne imagers. We analyzed a deep flash spectrum obtained by summing 80 individual spectra to evaluate the intensity modulations of the continuum. Intensity deficits are observed and measured at the prominences boundaries in both eclipse and SWAP images. The prominence cavities interpreted as a relative depression of plasma density, produced inside the corona surrounding the prominences, and some intense heating occurring in these regions, are discussed. Photometric measurements are shown at different scales and different, spectrally narrow, intervals for both the prominences and the coronal background.     

Observations of Unresolved Photospheric Magnetic Fields in Solar Flares Using Fe i and Cr i Lines

The structure of the photospheric magnetic field during solar flares is examined using echelle spectropolarimetric observations. The study is based on several Fe i and Cr i lines observed at locations corresponding to brightest Hα emission during thermal phase of flares. The analysis is performed by comparing magnetic-field values deduced from lines with different magnetic sensitivities, as well as by examining the fine structure of I±V Stokes-profiles’ splitting. It is shown that the field has at least two components, with stronger unresolved flux tubes embedded in weaker ambient field. Based on a two-component magnetic-field model, we compare observed and synthetic line profiles and show that the field strength in small-scale flux tubes is about 2?–?3 kG. Furthermore, we find that the small-scale flux tubes are associated with flare emission, which may have implications for flare phenomenology.     

Optical spectral decomposition of NGC 4051

Considering the host galaxy contribution, a spectral decomposition method is used to reanalyzed the archive data of optical spectra for a narrow line Seyfert 1 galaxy, NGC 4051. The light curves of the continuum f _λ (5100 Å), and Hβ, He ii, Fe ii emission lines are given. We find strong flux correlations between line emissions of Hβ, He ii, Fe ii and the continuum f _λ (5100 Å). These low-ionization lines (Hβ, Fe ii, He ii) have “inverse” intrinsic Baldwin effects. Using the methods of the cross-correlation function and the Monte Carlo simulation, we find the time delays, with respect to the continuum, are $3.45^{+12.0}_{-0.5}~\mbox{days}$ with the probability of 34 % for the intermediate component of Hβ, $6.45^{+13.0}_{-1.0}~\mbox{days}$ with the probability of 65 % for the intermediate component of He ii. From these intermediate components of Hβ and He ii, the calculated central black hole masses are $0.86^{+4.35}_{-0.33}\times 10^{6}$ and $0.82^{+3.12}_{-0.45}\times 10^{6}~M_{\odot }$ . We also find that the time delays for Fe ii are $9.7^{+3.0}_{-5.0}~\mbox{days}$ with the probability of 36 %, $8.45^{+1.0}_{-2.0}~\mbox{days}$ with the probability of 18 % for the total epochs and “subset 1” data, respectively. It seems that the Fe ii emission region is outside of the Hβ emission region.     

Automatic Detection and Classification of Coronal Holes and Filaments Based on EUV and Magnetogram Observations of the Solar Disk

A new method for the automated detection of coronal holes and filaments on the solar disk is presented. The starting point is coronal images taken by the Extreme Ultraviolet Telescope on the Solar and Heliospheric Observatory (SOHO/EIT) in the Fe ix/x 171 Å, Fe xii 195 Å, and He ii 304 Å extreme ultraviolet (EUV) lines and the corresponding full-disk magnetograms from the Michelson Doppler Imager (SOHO/MDI) from different phases of the solar cycle. The images are processed to enhance their contrast and to enable the automatic detection of the two candidate features, which are visually indistinguishable in these images. Comparisons are made with existing databases, such as the He i 10830 Å NSO/Kitt Peak coronal-hole maps and the Solar Feature Catalog (SFC) from the European Grid of Solar Observations (EGSO), to discriminate between the two features. By mapping the features onto the corresponding magnetograms, distinct magnetic signatures are then derived. Coronal holes are found to have a skewed distribution of magnetic-field intensities, with values often reaching 100?–?200 gauss, and a relative magnetic-flux imbalance. Filaments, in contrast, have a symmetric distribution of field intensity values around zero, have smaller magnetic-field intensity than coronal holes, and lie along a magnetic-field reversal line. The identification of candidate features from the processed images and the determination of their distinct magnetic signatures are then combined to achieve the automated detection of coronal holes and filaments from EUV images of the solar disk. Application of this technique to all three wavelengths does not yield identical results. Furthermore, the best agreement among all three wavelengths and NSO/Kitt Peak coronal-hole maps occurs during the declining phase of solar activity. The He ii data mostly fail to yield the location of filaments at solar minimum and provide only a subset at the declining phase or peak of the solar cycle. However, the Fe ix/x 171 Å and Fe xii 195 Å data yield a larger number of filaments than the Hα data of the SFC.     

Forecasting the Maxima of Solar Cycle 24 with Coronal Fe xiv Emission

The onset of the “Rush to the Poles” of polar-crown prominences and their associated coronal emission is a harbinger of solar maximum. Altrock (Solar Phys. 216, 343, 2003) showed that the “Rush” was well observed at 1.15 R _o in the Fe xiv corona at the Sacramento Peak site of the National Solar Observatory prior to the maxima of Cycles 21 to 23. The data show that solar maximum in those cycles occurred when the center line of the Rush reached a critical latitude of 76^°±2^°. Furthermore, in the previous three cycles solar maximum occurred when the highest number of Fe xiv emission features per day (averaged over 365 days and both hemispheres) first reached latitudes 20^°±1.7^°. Applying the above conclusions to Cycle 24 is difficult due to the unusual nature of this cycle. Cycle 24 displays an intermittent Rush that is only well-defined in the northern hemisphere. In 2009 an initial slope of 4.6^°?year^?1 was found in the north, compared to an average of 9.4±1.7^°?year^?1 in the previous cycles. An early fit to the Rush would have reached 76^° at 2014.6. However, in 2010 the slope increased to 7.5^°?year^?1 (an increase did not occur in the previous three cycles). Extending that rate to 76^°±2^° indicates that the solar maximum in the northern hemisphere already occurred at 2011.6±0.3. In the southern hemisphere the Rush to the Poles, if it exists, is very poorly defined. A linear fit to several maxima would reach 76^° in the south at 2014.2. In 1999, persistent Fe xiv coronal emission known as the “extended solar cycle” appeared near 70^° in the North and began migrating towards the equator at a rate 40 % slower than the previous two solar cycles. However, in 2009 and 2010 an acceleration occurred. Currently the greatest number of emission features is at 21^° in the North and 24^° in the South. This indicates that solar maximum is occurring now in the North but not yet in the South.     

Variability of Fe ii in Two NLS1s, PG 1700+815 and NGC 4051

We analyze the spectral variability for two narrow line Seyfert 1 galaxies, PG 1700+518 and NGC 4051 using the spectral decomposition method. We focus on their optical Fe ii variability to investigate the origin of Fe ii in AGNs. For PG 1700+518, we find that the Fe ii size is about 200 light-days, which is consistent with the Hβ size derived from the empirical R–L relation. For NGC 4051, the [O iii] 5007 Å flux is strongly correlated with continuum flux, suggesting that we should recalibrate the spectral flux on a scale defined by [O iii] flux. The corrected light curves of Fe ii, Hβ, He ii, f _λ (5100 Å) are given here. A detailed analysis will be given in the near future.     

Doppler Shifts of the Hα Line and the Ca ii 854.2 nm Line in a Quiet Region of the Sun Observed with the FISS/NST

The characteristics of Doppler shifts in a quiet region of the Sun are compared between the Hα line and the Ca?ii infrared line at 854.2 nm. A small area of 16″×40″ was observed for about half an hour with the Fast Imaging Solar Spectrograph (FISS) of the 1.6 meter New Solar Telescope (NST) at Big Bear Solar Observatory. The observed area contains a network region and an internetwork region, and identified in the network region are fibrils and bright points. We infer Doppler velocity v _m from each line profile at each individual point with the lambdameter method as a function of half wavelength separation Δλ. It is confirmed that the bisector of the spatially averaged Ca?ii line profile has an inverse C-shape with a significant peak redshift of +?1.8 km?s^?1. In contrast, the bisector of the spatially averaged Hα line profile has a C-shape with a small peak blueshift of ??0.5 km?s^?1. In both lines, the bisectors of bright network points are significantly redshifted not only at the line centers, but also at the wings. The Ca?ii Doppler shifts are found to be correlated with the Hα ones with the strongest correlation occurring in the internetwork region. Moreover, we find that here the Doppler shifts in the two lines are essentially in phase. We discuss the physical implications of our results in view of the formation of the Hα line and Ca?ii 854.2 nm line in the quiet region chromosphere.     

Spectral Line Selection for HMI: A Comparison of Fe I 6173 Å and Ni I 6768 Å

We present a study of two spectral lines, Fe I 6173 Å and Ni I 6768 Å, that were candidates to be used in the Helioseismic and Magnetic Imager (HMI) for observing Doppler velocity and the vector magnetic field. The line profiles were studied using the Mt. Wilson Observatory, the Advanced Stokes Polarimeter and the Kitt Peak-McMath Pierce telescope and one-meter Fourier transform spectrometer atlas. Both Fe I and Ni I profiles have clean continua and no blends that threaten instrument performance. The Fe I line is 2% deeper, 15% narrower, and has a 6% smaller equivalent width than the Ni I line. The potential of each spectral line to recover pre-assigned solar conditions is tested using a least-squares minimization technique to fit Milne-Eddington models to tens of thousands of line profiles that have been sampled at five spectral positions across the line. Overall, the Fe I line has a better performance than the Ni I line for vector-magnetic-field retrieval. Specifically, the Fe I line is able to determine field strength, longitudinal and transverse flux four times more accurately than the Ni I line in active regions. Inclination and azimuthal angles can be recovered to ≈2^° above 600 Mx cm^?2 for Fe I and above 1000 Mx cm^?2 for Ni I. Therefore, the Fe I line better determines the magnetic-field orientation in plage, whereas both lines provide good orientation determination in penumbrae and umbrae. We selected the Fe I spectral line for use in HMI due to its better performance for magnetic diagnostics while not sacrificing velocity information. The one exception to the better performance of the Fe I line arises when high field strengths combine with high velocities to move the spectral line beyond the effective sampling range. The higher g _eff of Fe I means that its useful range of velocity values in regions of strong magnetic field is smaller than Ni I.     

Properties of the ionised gas of circumnuclear star-forming regions in early-type spirals

A study of circumnuclear star-forming regions (CNSFRs) in several early-type spirals has been carried out in order to investigate their main properties: stellar and gas kinematics, dynamical masses, ionising stellar masses, chemical abundances and other properties of the ionised gas. Both high resolution (R～20,000) and moderate resolution (R～5000) have been used. In some cases, these regions (about 100–150 pc in size) are composed of several individual star clusters with sizes between 1.5 and 4.9 pc, estimated from Hubble Space Telescope images. Stellar and gas velocity dispersions are found to differ by about 20 to 30 km?s^?1, with the Hβ emission lines being narrower than both the stellar lines and the [Oiii]λ5007 Å lines. The twice ionised oxygen, on the other hand, shows velocity dispersions comparable to those of stars. We have applied the virial theorem to estimate dynamical masses of the clusters, assuming that the systems are gravitationally bounded and spherically symmetric, and using previously measured sizes. The measured values of the stellar velocity dispersions yield dynamical masses of the order of 10⁷ to 10⁸ M_⊙ for the full CNSFRs. We obtain oxygen abundances which are comparable to those found in high-metallicity disc Hii regions from direct measurements of electron temperatures and consistent with solar values within the errors. The region with the highest oxygen abundance is R3+R4 in NGC3504, 12+log(O/H)=8.85, about 1.5 times solar. The derived N/O ratios are, on average, larger than those found in high-metallicity disc Hii regions, and they do not seem to follow the trend of N/O vs. O/H which marks the secondary behaviour of nitrogen. On the other hand, the S/O ratios span a very narrow range—between 0.6 and 0.8 times solar. Compared to high-metallicity disc Hii regions, CNSFRs show values of the O₂₃ and the N2 parameters whose distributions are shifted to lower and higher values, respectively. Hence, even though their derived oxygen and sulphur abundances are similar, higher values would in principle be obtained for the CNSFRs if pure empirical methods were used to estimate abundances. CNSFRs also exhibit lower ionisation parameters than their disc counterparts, as derived from [Sii]/[Siii]. Their ionisation structure also seems to be different, with CNSFRs showing radiation-field properties more similar to Hii galaxies than to disc high-metallicity Hii regions.     

Narrow-Band Imaging System for the Multi-application Solar Telescope at Udaipur Solar Observatory: Characterization of Lithium Niobate Etalons

The Multi-application Solar Telescope is a 50 cm off-axis Gregorian telescope that has been installed at the lake site of Udaipur Solar Observatory. For quasi-simultaneous photospheric and chromospheric observations, a narrow-band imager has been developed as one of the back-end instruments for this telescope. Narrow-band imaging is achieved using two lithium niobate Fabry–Perot etalons working in tandem as a filter. This filter can be tuned to different wavelengths by changing either voltage, tilt, or temperature of the etalons. To characterize the etalons, a Littrow spectrograph was set up in conjunction with a 15 cm Carl Zeiss Coudé solar telescope. The etalons were calibrated for the solar spectral lines Fe i 6173 Å, and Ca ii 8542 Å. In this work, we discuss the characterization of the Fabry–Perot etalons, specifically, the temperature and voltage tuning of the system for the spectral lines proposed for observations. We present the details of the calibration set-up and various tuning parameters. We also present solar images obtained using the system.     

Spectroscopic Studies of Solar Corona VIII. Temperature and Non-Thermal Variations in Steady Coronal Structures

With a view to investigate variations in parameters of coronal emission lines over a large range of radial distance from the limb, raster scans were made with sufficiently long exposure times on several days during September – October 2003. An analysis of the data shows that (i) in most of the coronal structures, the FWHM of the Fe xiv 5303 Å line decreases up to 300″±50″, (ii) the FWHM of the Fe x 6374 Å line increases up to about 200″ and then remains unchanged up to about 500″, and (iii) the FWHMs of the Fe xi 7892 Å and Fe xiii 10747 Å lines show an intermediate behaviour with height. The analysis of the data also shows that the ratio of FWHM of 6374 Å to that of 5303 Å increases from 0.93 at the limb to 1.18 at 200″ above the limb. From this and the ratio of intensities of the two lines we infer that the plasma in steady coronal structures at a height of about 200″ has a temperature of about 1.5 MK and a non-thermal velocity around 17 km s^?1. The observations also show that non-homogeneous temperatures and non-thermal velocities largely exist in the lower corona up to about 300″±100″ above the limb. Amplitudes of variations in FWHM of different emission lines with height in the coronal loops are similar to those in the diffuse plasma around the coronal loops.     

The High-Resolution Coronal Imager (Hi-C)

The High-Resolution Coronal Imager (Hi-C) was flown on a NASA sounding rocket on 11 July 2012. The goal of the Hi-C mission was to obtain high-resolution (≈?0.3?–?0.4′′), high-cadence (≈?5 seconds) images of a solar active region to investigate the dynamics of solar coronal structures at small spatial scales. The instrument consists of a normal-incidence telescope with the optics coated with multilayers to reflect a narrow wavelength range around 19.3 nm (including the Fe xii 19.5-nm spectral line) and a 4096×4096 camera with a plate scale of 0.1′′?pixel^?1. The target of the Hi-C rocket flight was Active Region 11520. Hi-C obtained 37 full-frame images and 86 partial-frame images during the rocket flight. Analysis of the Hi-C data indicates the corona is structured on scales smaller than currently resolved by existing satellite missions.     

Silicon Abundance from RESIK Solar Flare Observations

The RESIK instrument on the CORONAS-F spacecraft obtained solar flare and active-region X-ray spectra in four channels covering the wavelength range 3.8?–?6.1 Å in its operational period between 2001 and 2003. Several highly ionized silicon lines were observed within the range of the long-wavelength channel (5.00?–?6.05 Å). The fluxes of the Si?xiv Ly-β line (5.217 Å) and the Si?xiii 1s ²?–?1s3p line (5.688 Å) during 21 flares with optimized pulse-height analyzer settings on RESIK have been analyzed to obtain the silicon abundance relative to hydrogen in flare plasmas. As in previous work, the emitting plasma for each spectrum is assumed to be characterized by a single temperature and emission measure given by the ratio of emission in the two channels of GOES. The silicon abundance is determined to be A(Si)=7.93±.21 (Si?xiv) and 7.89±.13 (Si?xiii) on a logarithmic scale with H=12. These values, which vary by only very small amounts from flare to flare and times within flares, are 2.6±1.3 and 2.4±0.7 times the photospheric abundance, and are about a factor of three higher than RESIK measurements during a period of very low activity. There is a suggestion that the Si/S abundance ratio increases from active regions to flares.