Densities and mass motions in transition-zone plasmas in solar flares observed from Skylab

We studied the EUV line spectra of three flare observed with the NRL slit spectrograph on Skylab. The electron densities in the flare transition-zone plasmas are determined from density-sensitive lines of Si iii and O iv. The electron densities in all three flares studied were greatest during the flare maximum with values of the order of 10¹² cm^–3. The density decreases by a factor of 2 to 3 in the decay phase of the flares. The intensities of EUV lines from the flare chromospheric and transition-zone plasmas all are greatly enhanced. In contrast to lines for Oi, Ci, Feii and other chromospheric ions, the lines of Oiv and Nv and other transition-zone lines are not only enhanced but also very much broadened.Fitting of the N v 1242 Å line with a two-gaussian model shows that for two of the flares studied, there is a red-shifted component in addition to an unshifted component. The shifted component in the N v line profiles is interpreted as due to a dynamic and moving plasma with a bulk motion velocity of 12 km s^–1 for one flare and more than 70 km s^–1 for the other. The broadened line profiles indicate that there are large turbulent mass motions with random velocities ranging from 30 to 80 km s^–1.Ball Corporation. Now with NASA/Marshall Space Flight Center.     

Absolute redshifts in the C iv 1548 Å line in the transition region of the quiet Sun

Observations with the UVSP instrument on the SMM spacecraft were made at the polar limb and disk center for the accurate determination of Doppler shifts of the Civ 1548 Å emission line formed at 10⁵ K in the transition region of the quiet Sun. Individual data points representing 3 arc sec square pixels yield both redshifts and blueshifts, but the mean values from four different days of observations are toward the red. The mean redshifts are in the range 4–8 km s^-1 and are produced by nearly vertically directed flows; the uncertainty associated with the mean values correspond to ±0.5 km s^-1. The redshift increases with brightness of the Civ line.     

The far UV spectrum of the Be star AX Monocerotis

In this paper we study the main features of the far UV spectrum of the binary star AX Mon, observed with the IUE satellite at phase 0.568.Ions indicating a large range of ionization stages, going fromCi,Oi,Ni toSiv,Civ,Nv are present.The spectrum is dominated by shell absorption lines of Feii, Feiii, Siiii,Cii, Alii, Mgii and Niii.Two satellite components are clearly indicated in all these lines except for Niii which presents only one. Their mean velocities are +10±5 km s^–1, –75±10 km s^–1, and –260±15 km s^–1.Red emission wings are observed in the Mgii resonant doublet at 2800 Å, which shows a P Cygni profile. Each of the absorption lines of the Mgii doublet is formed by a narrow component, which is blended with the Mgii interstellar line and a broad component, which shows a complex structure.Broad and asymmetrical profiles are observed for the Siiv,Civ, andNv resonance lines with blue edge velocities about –700±30 km s^–1.     

Analysis of the emission line spectra of a solar flare observed from Skylab

The EUV emission spectra in the wavelength range 110–1900 Å of the 5 September 1973 flare observed with the NRL slit spectrograph on Skylab are studied. The results are: (1) The chromospheric and transition-zone lines are greatly enhanced during the flare. In particular, the allowed lines are enhanced more than the intersystem lines. The Ni ii and P ii lines show the greatest enhancement with a factor of 800 increase in intensity. Other lines such as O i, C i, Si iii, S iii, S iv, O iv, O v, and N v show increases in intensity 10–100 times during the flare. (2) The chromospheric lines, although greatly enhanced during the flare, maintain their sharp and gaussian profiles and are not appreciably broadened. The transition zone lines, on the other hand, show a red-shifted component during the initial phase of the flare. The deduced downward velocity in the transition zone is 50 km s^–1. In addition, there are large turbulent mass motions. The downward mass motion is probably caused by the pressure imbalance between the flare hot plasma at 13 × 10⁶ K and the cooler plasma at 10⁵ K. (3) The density of the 10⁵ K flare plasma, as deduced from density-sensitive lines, is greater than 10¹² cm^-3. The depth of the 10⁵ K plasma in the flare transition zone is only of the order of 0.1 km, giving a steep temperature gradient. Consideration of the energy balance between the conductive flux and the radiative energy losses shows that, indeed, the high density in the transition zone requires that its thickness be very small. This is a consequence of the maximum radiative efficiency at the temperature around 10⁵ K in the solar plasma.Ball Brothers Research Corporation.     

Extreme ultraviolet spectra of solar active regions and their analysis

Extreme ultraviolet spectra of several active regions are presented and analyzed. Spectral intensities of 3 active regions observed with the NRL Skylab XUV spectroheliograph (170–630 Å) are derived. From this data density sensitive line ratios of Mg viii, Si x, S xii, Fe ix, Fe x, Fe xi, Fe xii, Fe xiii, Fe xiv, and Fe xv are examined and typically yield, to within a factor of 2, electron pressures of 1 dyne cm^–2 (n _e T = 6 × 10¹⁵ cm^–3 K). The differential emission measure of the brightest 35 × 35 portion of an active region is obtained between 1.4 × 10⁴ K and 5 × 10⁶ K from HCO OSO-VI XUV (280–1370 Å) spectra published by Dupree et al. (1973). Stigmatic EUV spectra (1170–1710 Å) obtained by the NRL High Resolution Telescope and Spectrograph (HRTS) are also presented. Doppler velocities as a function of position along the slit are derived in an active region plage and sunspot. The velocities are based on an absolute wavelength scale derived from neutral chromospheric lines and are accurate to ±2 km s^–1. Downflows at 10⁵ K are found throughout the plage with typical velocities of 10 km s^–1. In the sunspot, downflows are typically 5 to 20 km s^–1 over the umbra and zero over the penumbra. In addition localized 90 and 150 km s^–1 downflows are found in the umbra in the same 1 × 1 resolution elements which contain the lower velocity downflows. Spectral intensities and velocities in a typical plage 1 resolution element are derived. The velocities are greatest ( 10 km s^–1) at 10⁵ K with lower velocities at higher and lower temperatures. The differential emission measure between 1.3 × 10⁴ K and 2 × 10⁶ K is derived and is found to be comparable to that derived from the OSO-VI data. An electron pressure of 1.4 dynes cm^–2 (n _e T = 1.0 × 10¹⁶ cm^–3 K) is determined from pressure sensitive line ratios of Si iii, O iv, and N iv. From the data presented it is shown that convection plays a major role in determining the structure and dynamics of the active region transition zone and corona.     

Discrete subresolution structures in the solar transition zone

During operations on the Spacelab-2 Shuttle mission, the NRL High Resolution Telescope and Spectrograph (HRTS) recorded spectra of a variety of solar features in the 1200–1700 Å wavelength region which contains spectral lines and continua well suited for investigating the temperature minimum, the chromosphere and transition zone. These data show that, at the highest spatial resolution, the transition zone spectra are broken up from a continuous intensity distribution along the slit into discrete emission elements. The average dimensions of these discrete transition zone structures is 2400 km along the slit, but an analysis of their emission measures and densities shows that the dimensions of the actual emitting volume is conciderably less. If these structures are modelled as an ensemble of subresolution filaments, we find that these filaments have typical radii of from 3 to 30 km and that the cross-sectional fill factor is in the range from 10^–5 to 10^–2. The transport of mass and energy through these transition zone structures is reduced by this same factor of 10^–5 to 10^–2 which has significant consequences for our understanding of the dynamics of the solar atmosphere. Because the HRTS transition zone line profiles are not broadened by resolved large-spatial-scale solar velocity fields, the line widths of the Civ lines have been analyzed. The average line width is 0.195 Å (FWHM) and requires an average nonthermal velocity of 16 km s^–1 (most-probable) or 19 km s^–1 (root-mean-square) which is lower than previously observed values.     

Transition-zone observations of rapid flare events as observed by OSO-8

The rapid dissipation of flare energy has been observed in the transition-zone line of C iv at 1548.2 Å using the University of Colorado spectrometer aboard OSO-8. Impulsive brightenings have been resolved with characteristic risetimes as low as 3.5 s. One event is analyzed in detail, in which it is inferred that the electron density is greater than 2 × 10¹¹ cm^–3 at T = 60 000 K, and that the flare energy is deposited at a rate of 2 ergs cm^–3 s^–1 or greater. The temporal behavior of the intensity at the center of the C iv line is consistent with a non-equilibrium ionization of C iii through C v. If this event is a result of the multiple tearing mode instability as the primary energy release mechanism, then the observations indicate a pre-flare magnetic field of about 175 G.The National Center for Atmospheric Research is sponsored by the National Science Foundation.     

Solar X-ray spectra observed from the ‘Intercosmos-4’ satellite and the ‘Vertical-2’ rocket

Results are given of the detailed analysis of fourteen Fe xxv-xxiii lines ( = 1.850–1.870 Å) in the spectra of a solar flare on 16 Nov. 1970. The spectra were obtained with a resolution of about 4 × 10^–4 Å, which revealed lines not previously observed and allowed the measurement of line profiles. The measured values of the wavelengths and emission fluxes are presented and compared with theoretical calculations. The analysis of the contour of the Fe xxv line ( = 1.850 Å) leads to the conclusion that there is unidirectional macroscopic gas motion in the flare region with the velocity (projection on the line of sight) ± 90 km s^–1.Measurements of the 8.42 Å Mg xii and 9.16 Å Mg xi lines in the absence of solar flares indicate prolonged existence of active regions on the solar disk with T _e = 4–6 × 10⁶K and emission measure ME 10⁴⁸ cm^–3. The profile of the Mg xii line indicates a macroscopic ion motion with a velocity up to 100 km s^–1.     

Mass flow and the validity of ionization equilibrium on the Sun

Ionization equilibrium is a useful assumption which allows temperatures and other plasma properties to be deduced from spectral observations. Inherent to this assumption is the premise that the ion stage densities are determined solely by atomic processes which are local functions of the plasma temperature and electron density. However, if the time scale of plasma flow through a temperature gradient is less than the characteristic time scale for an important atomic process, deviations from the ionization stage densities expected for equilibrium will occur which could introduce serious errors into subsequent analyses. In the past few years, significant flow velocities in the upper solar atmosphere have been inferred from observations of emission lines originaing in the transition region (about 10⁴–10⁶ K) and corona. In this paper, three models of the solar atmosphere (quiet Sun, coronal hole, and a network model) are examined to determine if the emission expected from these model atmospheres could be produced from equilibrium ion populations when steady flows of several kilometers per second are assumed. If the flows are quasi-periodic instead of steady, spatial and temporal averaging inherent in the observations may allow for the construction of satisfactory models based on the assumption of ionization equilibrium. Representative emission lines are analysed for the following ions: C iii, iv, O iv, v, vi, Ne vii, viii, Mg ix, x, Si xii, and Fe ix–xiv. Two principle conclusions are drawn. First, only the iron ions are generally in equilibrium for steady flows of 20 km s^–1. For carbon and oxygen, ionization equilibrium is not a valid assumption for steady flows as small as 1 km s^–1. Second, the three models representing different solar conditions behave in a qualitatively similar manner, implying that these results are not particularly model dependent over the range of temperature gradients and electron densities thus far inferred for the Sun. In view of the flow velocities which have been reported for the Sun, our results strongly suggest caution in using the assumption of ionization equilibrium for interpreting spectral lines produced in the transition region.The National Center for Atmospheric Research is sponsored by the National Science Foundation.     

Dynamic evolution of recurrent mass ejections observed in Hα and Civ lines

During a coordinated SMY program, the consecutive formation of two new active centers merging together within AR 2646 was observed from 28 August, to 5 September, 1980. The two preceding spots compressed an inverse polarity spot on 1 September 1980, causing recurrent ejecta of matter with time intervals around 10 min. The observations of the MSDP spectrograph operating in H at the Meudon Solar tower and of the UVSP spectrometer on SMM in the Civ 1548 Å line show that cold and hot material had the same projection, although the upward Civ velocity structure was more extended than the H one. We present evidence that observed contrasts of the H absorbing structure can be interpreted in terms of a dynamic cloud model overlying the chromosphere. H matter follows a magnetic channel with upward velocity around 20–30 km s^–1 in the first phase of the event and with downward velocity ( - 40 km s^–1) in the second phase. The stored energy is not sufficient to trigger a flare, nor even to propulse matter along the full length of an arch, because of the periodic reorganisation of the magnetic field.     

Photospheric and chromospheric oscillations in the base of coronal holes

In this paper we carry out an analysis of the spatial–temporal line-of-sight velocity variations measured in the chromospheric (H, H) and photospheric (Fei 6569 Å, Fei 4864 Å, Nii 4857 Å) lines at the base of 17 coronal holes. Time series of a duration from 43 to 120 min were recorded with the CCD line-array and the CCD matrix. Rather frequently we observed quasi-stationary upward flows with a measured velocity of up to 1 km s^–1 in the photosphere and up to 4–5 km s^–1 in the chromosphere (equivalent radial velocity of up to 3 km s^–1 and up to 12–15 km s^–1 accordingly) near dark points on the chromospheric network boundary inside polar CH. Line-of-sight velocity fluctuation spectra contain meaningful maxima in the low-frequency region clustering around the values 0.4, 0.75, and 1 mHz. Usually, the spatial localization of these maxima mutually coincides and, in our opinion, coincides with the chromospheric network boundary. Acoustic 3- and 5-min oscillations are enhanced in the coronal hole region and reach 1 km s^–1 in the photosphere and 3–4 km s^–1 in the chromosphere. These oscillations are not localized spatially and are distinguished throughout the entire region observed.     

On the interpretation of Fraunhofer line Doppler shifts at supergranule boundaries

We have used the SPO tower telescope and echelle spectrograph to study differences in the profiles of three Fei lines, between magnetic network and cells. Ca K slit-jaw pictures were used to identify the network and cell areas, and mean network and cell profiles were computed from digitized spectra for the g = 0 lines 4065, 5434, and the g = 1.5 line 5233. The profile bisectors show that the wings of all three lines are red-shifted in the network by between 75–200 m s^–1 relative to the cell profiles. But the redshift decreases in the line core and becomes less than the standard error of 20 m s^–1 near the line core minimum. This disappearance of the redshift at the cores of all 3 lines formed over the height range 250–500 km above _0.5 = 1, argues against a steady downflow at supergranule boundaries. We show that such red-shifted wings and a relatively unshifted core can result if granular convection is suppressed near the network flux tubes, without implying any downflow in the vicinity of these flux tubes. Our results also indicate that searches for large-scale convective velocity patterns should measure shifts of the line core, rather than the line wings which appear to be very sensitive to inhomogeneities in granule structure.Visiting Astronomers, Sacramento Peak Observatory.     

Evolution of the high-temperature plasma in the 15 June 1973 flare

The analysis of the high temperature plasma in Fe xxiii–xxiv in the 15 June 1973 flare is presented. The observations were obtained with the NRLXUV spectroheliograph on Skylab. The results are: (1) There was preheating of the active region in which the flare occurred. In particular, a large loop in the vicinity of the flaring region showed enhanced brightness for many hours before the flare. The loop disappeared when the flare occurred, and returned in the postflare phase, as if the energy flux which had been heating the large loop was blocked during the flare and restored after the flare was gone. The large magnetic fields did not change significantly. (2) The flare occurred in low-lying loop or loops. The spatial distribution of flare emission shows that there was a temperature gradient along the loop. (3) The high temperature plasma emitting Fe xxiii and xxiv had an initial upward motion with a velocity of about 80 km s^–1. (4) There was large turbulent mass motion in the high temperature plasma with a random velocity of 100 to 160 km s^–1. (5) The peak temperature of the hot plasma, determined from the Fe xxiii and xxiv intensity ratio, was 14 × 10⁶ K. It decreased slightly and then, for a period of 4 min, remained at 12.6 × 10⁶ K before dropping sharply to below 10 × 10⁶ K. The density of the central core of the hot plasma, determined from absolute intensity of Fe xxiv 255 Å line, was of the order of 10¹¹ cm^–3.The persistence of the high level of turbulence and of the high temperature plateau in the decaying phase of the flare indicates the presence of secondary energy release. From the energy balance equation the required energy source is calculated to be about 3 to 7 ergs cm^–3 s^–1.Ball Brothers Research Corporation.     

The Si IV λ 1393 line in a coronal hole compared to the quiet Sun from OSO-8 observations

We report on studies of the 1393 line of Si iv, formed in the transition region at about 80 000 K, made using the Colorado experiment on OSO-8. Results indicate that the line width is somewhat greater in coronal holes compared to the quiet Sun, implying a difference in the broadening mechanism. There is no evidence that the line is Doppler shifted in coronal holes relative to the quiet Sun implying there is no mass flow in holes, at the 80 000 K level, greater than 4.3 km s^–1. Within the uncertainty of our experiment the integrated line intensities are the same in a coronal hole as in the quiet Sun.     

Identification of forbidden coronal lines of Fe x and Ni xii

The transitions 3s ²3p ⁴3d ⁴ D, ⁴ F–3s3p ⁵3d ⁴ F have been studied in the Cl i isoelectronic sequence from Ca iv to Fe x. The determination of the 3s ²3p ⁴3d quartet intervals in Fe x has led to the identification of eleven coronal lines in the region from 1463 Å to 5539 Å as forbidden transitions within this configuration. By extrapolation, an additional coronal line is identified with a similar transition in Ni xii.     

O VI (λ = 1032 Å) profiles in and above an active region prominence,compared to quiet Sun center and limb profiles

O vi ( = 1032 Å) profiles have been measured in and above a filament at the limb, previously analyzed in H i, Mg ii, Ca ii resonance lines (Vial et al., 1979). They are compared to profiles measured at the quiet Sun center and at the quiet Sun limb.Absolute intensities are found to be about 1.55 times larger than above the quiet limb at the same height (3); at the top of the prominence (15 above the limb) one finds a maximum blue shift and a minimum line width. The inferred non-thermal velocity (29 km s^–1) is about the same as in cooler lines while the approaching line-of-sight velocity (8 km s^–1) is lower than in Ca ii lines.The O vi profile recorded 30 above the limb outside the filament is wider (FWHM = 0.33 Å). It can be interpreted as a coronal emission of O vi ions with a temperature of about 10⁶ K, and a non-thermal velocity (NTV) of 49 km s^–1. This NTV is twice the NTV of quiet Sun center O vi profiles. Lower NTV require higher temperatures and densities (as suggested by K-coronameter measurements). Computed emission measures for this high temperature regime agree with determinations from disk intensities of euv lines.     

Motions in solar magnetic tubes

The line-centre-magnetogram technique has been used to measure the average velocity in magnetic elements in plages and isolated magnetic elements (including dipoles) in Ca ii 8542, Mg i 5183, Fe i 8688 and C i 9111. The velocities vary from 0.6 km s^–1 downflow in the line of deepest origin to zero in the highest. The smooth curve obtained by combining these with the results of other investigators is in conformity with Giovanelli's (1977) theory of inflow in the neighborhood of the temperature minimum.This material is based upon research supported by the National Science Foundation under its contract No. AST 74-04129 with the Association of Universities for Research in Astronomy, Inc., for management, operation and maintenance of the Kitt Peak National Observatory.Visiting Astronomer, Kitt Peak National Observatory.     

Solar flare X-ray spectra

Results are presented of an investigation of solar flare X-ray spectra in the region 1.70–1.95 Å, obtained aboard the Intercosmos-4 satellite during the maximum of solar activity (October–November, 1970). With the use of 6 high resolution spectra in the region 1.85–1.87 Å the identification of lines due to 18 transitions of 2p 1s type, consisting of the resonance, intercombination and forbidden Fe xxv ion lines and the satellite Fe xxiv lines has been performed. With the use of the recent laboratory data the averaged wavelengths of the lines were obtained confirming the theoretically calculated ones with an accuracy about ± 0.0004 Å. A variable Doppler shift of the Fe xxv resonance lines was observed for the flare of November 16, 1970, which points to hot plasma motions with velocities up to 400 km s^-1.     

Multiple loop activations and continuous energy release in the solar flare of June 15, 1973

The spatial and temporal evolution of the high temperature plasma in the flare of 1973 June 15 has been studied using the flare images photographed by the NRL XUV spectroheliograph on Skylab.The overall event involves the successive activations of a number of different loops and arches bridging the magnetic neutral line. The spatial shifts and brightenings observed in the Fe xxiii–xxiv lines are interpreted as the activation of new structures. These continued for four or five minutes after the end of the microwave burst phase, implying additional energy-release unrelated to the nonthermal phase of the flare. A shear component observed in the coronal magnetic field may be a factor in the storage and release of the flare energy.The observed Fe xxiii–xxiv intensities define a post-burst heating phase during which the temperature remained approximately constant at 13 × 10⁶ K while the Fe xxiv intensity and 0–3 Å flux rose to peak values. This phase coincided with the activation of the densest structure (N _e = 2 × 10¹¹ cm^–3). Heating of higher loops continued into the decay phase, even as the overall temperature and flux declined with the fading of the lower Fe xxiv arches.The observed morphology of individual flaring arches is consistent with the idea of energy release at altitude in the arch (coincident with a bright, energetic core in the Fe xxiv image) and energy flow downward into the ribbons. The Doppler velocity of the Fe xxi 1354 Å line is less than 5 km s^–1, indicating that the hot plasma region is stationary.The relation of this flare to the larger class of flares associated with filament eruptions and emerging magnetic flux is discussed.     

The supergranule velocity field

A study of supergranule motions confirms horizontal velocities with peak values of typically 0.36 km s^–1 as observed in Fe i 8688 Å. These show no significant variation with height over the range of formation of C i 9111, Fe i 8688, and Mg i 8806, but there is a substantial reduction to about one-half of this at the level of Ca ii 8542.Near disk center, supergranule vertical velocities in Fe i 8688 have rms values ±0.01 km ^–1, after allowance for the residual effects of the line-of-sight component of the horizontal supergranule motions, the five-minute oscillations, granule motions, and detector drift. There is a marginally-significant association of magnetic elements, and hence of cell boundaries, with downward motions; but this requires further testing.Measurements of downward velocities 0.1 km^–1 in regions of strong magnetic field when using unpolarized light are attributed to the much higher downflow inside the elements themselves and have nothing to do with supergranule motions.Visiting Astronomer, Kitt Peak National Observatory.Operated by the Association of Universities for Research in Astronomy, Inc. under contract with the National Science Foundation.