© 1998. The American Astronomical Society. All rights reserved. Printed in U.S.A.
U. FELDMAN, 1 W. CURDT, 2 G. A. DOSCHEK, 1 U. SCHÜHLE, 2 K. WILHELM, 2 AND P. LEMAIRE 3
Received 1997 October 16; accepted 1998 March 12
Subject headings: line: identificationsSun: coronaSun: UV radiation
FOOTNOTES1 E. O. Hulburt Center for Space Research, Mail Code 7608, Naval Research Laboratory, 4555 Overlook Avenue SW, Washington DC, 20375-5352.
2 Max-Planck-Institut für Aeronomie, D-37191 Katlenburg-Lindau, Germany.
3 Institut d'Astrophysique Spatiale, Unite Mixte Centre Nationale de la Recherche Scientifique, Université Paris XI, Bat 121, F-91405 Orsay, France.
Feldman et al. (1997, hereafter Paper I) published an extensive list of spectral lines in the 5001610 Å range compiled from spectra recorded 1996 June 25 by the Solar Ultraviolet Measurements of Emitted Radiation Spectrometer (SUMER) flown on the Solar and Heliospheric Observatory (SOHO). The spectra were recorded at a height of about 20,000 km above the west equatorial solar limb. Curdt et al. (1997) also published a line list from a SUMER solar disk spectrum that covers the 6601175 Å range. However, from the time that SOHO was launched until recently conditions on the Sun have been typical of solar minimum. Active regions have been scarce and small, and therefore only a few of the brightest lines characteristic of plasmas with temperatures Te > 1.5 × 106 K are visible in SUMER spectra of these regions, and consequently only a few such lines are tabulated in the published line lists. Recently, however, as the Sun has become increasingly more active, more lines emitted by higher temperature plasmas have appeared in SUMER active region spectra.
The main allowed lines (electric dipole transitions) emitted by highly ionized ions that are present in high-temperature (Te > 2 × 106 K) solar coronal plasmas appear at short wavelengths ( < 500 Å) not accessible to the SUMER instrument. However, a number of forbidden transitions arising from levels that do not decay via electric dipole transitions are calculated to be bright in high-temperature solar plasmas and are expected to be present in the SUMER spectral range. These lines arise mainly from transitions within the 2s22pk and 3s23pk (where k = 1, 5) ground configurations. Since bright forbidden lines emitted by lower temperature plasmas (Te < 2 × 106 K) are prominent in quiet-Sun SUMER spectra (Paper I), we expect to see similar high-temperature forbidden lines from the solar abundant elements Ar, Ca, Fe, and Ni in active region spectra.
In this paper we report the identification of spectral lines emitted by plasmas with temperatures between 2 × 106 K that are useful for determining emission measure distributions, electron densities, and mass motions above bright active regions. Except for the 1133.76 Å Ca XIII line, which is present in the Curdt et al. (1997) list, none of the lines is normally seen in disk spectra. Some of the newly identified lines at wavelengths short of 680 Å and those in the 9371361 Å range are also expected to be observed under certain coronal conditions at heights greater than 1.5 R by the SOHO Ultraviolet Coronagraph Spectrometer (UVCS; Kohl et al. 1995). Several of the lines should also be visible in spectra obtained by the SOHO Coronal Diagnostic Spectrometer (CDS; Harrison et al. 1995). Flare lines that are expected to be seen at temperatures greater than 8 × 106 K have not yet been seen by SUMER. However, lines of Fe XVIII and Fe XIX have been observed and are discussed in this paper.
Wilhelm et al. (1995), and first results are described by Wilhelm et al. (1997) and Lemaire et al. (1997).
The spectra used to identify the spectral lines are a set of 300 s exposures recorded 1997 September 6 between about 01:04 and 06:42 UT. The spectra are part of the "reference spectrum" program, i.e., full wavelength coverage spectra are periodically recorded for many different types of solar regions and at different locations. On this day the solar radius was 962. The pointing for these reference spectra was located above an active region on the north-west limb. The slit was oriented in the north-south direction with the center of the slit located 960 west and 370 north of Sun center. The limb distance varied between 227 (bottom/south of the slit) and 1296 (top/north of the slit). This was the site of active region NOAA 8076 that flared on September 5. An image of the region obtained by the Extreme Ultraviolet Imaging Telescope (EIT) on SOHO is shown in Figure 1.
Fig. 1 Table 1. The wavelengths are believed to be accurate to ±0.02 Å. The wavelengths were derived from the spectra after the spectra were flat-field corrected and the geometric distortions were removed ("destretched"). The intensities were obtained from a single active region reference spectrum. Because this spectrum took several hours to record, and active regions exhibit considerable transient behavior, intensity ratios of lines at widely separated wavelengths may not be accurate.
Six of the forbidden lines are shown in Figures 2 and 3. In Figure 2, an active region spectrum is compared with a quiet-Sun spectrum. Note the presence of Ca XIII, Ca XIV, and Ca XV lines in the active region spectrum and their absence in the quiet-Sun spectrum. In Figure 3, an Fe XIX flare line is shown. Note that the flare line shows substantial Doppler broadening and that it is confined to the small spatial regions of the flare.
Fig. 2 Fig. 3 We discuss the methods of identification in this section. Forbidden transitions in ions of neon to sulfur and in Ar XI having maximum fractional abundance temperatures less than 2 × 106 K are present in the SUMER wavelength range and were discussed in an earlier publication (Paper I). Two forbidden transitions from Ar XII, although quite faint, were also seen in the quiet-Sun spectra and were discussed in Paper I. We report new identifications of Ar XIII, Ar XII, Ca XV, Ca XIV, Ca XIII, Ni XV, and Ni XIV forbidden lines, and the transitions are given in Table 1. Using a semi-empirical method, Edlén predicted the energy levels within the ground configurations of the ions (Edlén 1983, 1984, 1985).
Arnaud & Rothenflug 1985). The 2s22p2 ground configuration includes the following levels: 3P0,1,2, 1D2, and 1S0. Two of the forbidden transitions within the ground configuration, i.e., 3P11D2 (1330.54 Å) and 3P11S0 (656.69 Å), are predicted to be in the SUMER detector B wavelength range. The other bright forbidden lines resulting from transitions within the ground configuration are at longer wavelengths and are not accessible with SUMER. A previous identification of the 3P11D2 transition (Sandlin, Brueckner, & Tousey 1977) is in error. Both the 1330.54 and 656.69 Å lines are observed for the first time in the SUMER active region spectra. Details concerning the Ar XIII energy levels are given in Table 2. The energies of the 3P1 and 3P2 levels are taken from Edlén (1985).
Arnaud & Rothenflug 1985). The 2s22p3 ground configuration includes the following levels: 4S3/2, 2D3/2,5/2, and 2P1/2,3/2 (see Fig. 4). Edlén (1984) predicted the energies of the levels within the ground configuration of Ar+11. Four of the forbidden transitions within this configuration, i.e., 4S3/22D3/2(1054.62 Å), 4S3/22D5/2(1018.87 Å), 4S3/22P1/2(670.34 Å), and 4S3/22P3/2(648.93 Å), are predicted to be in the SUMER detector B wavelength range. The other transitions are at longer wavelengths and are not accessible with SUMER. The 1054.57 and 1018.87 Å lines were seen before in the SUMER quiet-Sun spectra (Paper I) where they were quite weak. The 670.34 Å transition is observed for the first time in the active region spectra. The 648.93 Å transition is most likely present in the spectra but is blended with the bright 649.20 Å Si X line. Details concerning the Ar XII energy levels are given in Table 3.
Fig. 4 Arnaud & Rothenflug 1985). The 2s22p2 ground configuration includes the following levels: 3P0,1,2, 1D2, and 1S0. Edlén (1985) predicted the energies of the levels within this configuration. Only two of the forbidden transitions within the ground configuration, i.e., 3P21D2 (1375.96 Å) and 3P11D2(1098.44 Å), are visible in the SUMER detector B wavelength range. The 3P11S0 transition expected to appear at 555.21 Å is not visible, most likely due to the low reflectivity of SUMER at that wavelength. The 1122.7 Å 1D21S0 transition is not visible, and all other transitions are at longer wavelengths. The 1375.96 Å line was previously observed in Skylab spectra by Feldman & Doschek (1977) and Sandlin et al. (1977). Details concerning the Ca XV energy levels are given in Table 2. The energies of the 3P1 and 3P2 levels are taken from Edlén (1985).
Arnaud & Rothenflug 1985). The 2s22p3 ground configuration includes the following levels: 4S3/2, 2D3/2,5/2, and 2P1/2,3/2 (see Fig. 4). Edlén (1984) predicted the energies of the levels within the configuration. Six of the forbidden transitions within the ground configuration, i.e., 4S3/22D3/2(943.61 Å), 4S3/22D5/2(880.43 Å), 4S3/22P1/2 (579.85 Å), 4S3/22P3/2(545.26 Å), 2D3/22P3/2(1291.61 Å), and 2D5/22P3/2(1432.12 Å), are predicted to be in the SUMER detector B wavelength range. The 2D3/22P1/2 transition is predicted to be at 1503.1 Å. This wavelength is outside the SUMER B detector range, but the line is expected to be visible in the wavelength range of detector A. All but the 1432.12 Å are seen in the SUMER spectra. Details concerning the Ca XIV energy levels are given in Table 3.
Arnaud & Rothenflug 1985). The 2s22p4 ground configuration includes the following levels: 3P2,1,0, 1D2, and 1S0. Edlén (1983) predicted the energies of the levels within the configuration. Only two of the forbidden transitions within the ground configuration, i.e., 3P21D2 (1133.79 Å) and 3P21S0(648.68 Å), are predicted to be in the SUMER detector B wavelength range. All other transitions are at longer wavelengths. Both transitions are present in the active region data (Table 1). The 1133.79 Å line was seen before as a very faint feature in Skylab flare spectra (Feldman & Doschek 1991). Details concerning the Ca XIII energy levels are given in Table 4. The energies of the 3P0 and 3P1 are taken from Edlén (1983).
XVII, Fe XVIII, and Fe XIX are also present in active region spectra. The maximum fractional abundance of Fe+16 is at 4.0 × 106 K, Fe+17 is at 6.3 × 106 K, and Fe+18 is at 7.9 × 106 K (Arnaud & Raymond 1992). The 1153.16 Å Fe XVII (2s22p53s 3P12s22p53s 3P0) and the 974.86 Å Fe XVIII (2s22p5 2P3/22s22p5 2P1/2) lines are bright in the active region spectra, whereas the 592.2 Å Fe XIX (2s22p4 3P22s22p4 1D2) and 1118.08 Å Fe XIX (2s22p4 3P22s22p4 3P1) lines are fairly faint. All four transitions were previously observed in Skylab solar flare spectra (Doschek et al. 1975; Sandlin et al. 1977; Dere 1978; Feldman, Doschek, & Seely 1985; and Feldman & Doschek 1991). The Fe XVIII line and the long-wavelength Fe XIX line are also observed in Tokamak spectra (Peacock, Stamp, & Silver 1984). The SUMER spectral range contains additional forbidden lines from Fe XIX, Fe XX, Fe XXI, and Fe XXII which are expected to be present in spectra dominated by flare plasmas. The Fe XXI line at 1354.1 Å is well-known in flare spectra. It was seen in Skylab spectra by Doschek et al. (1975) and was discussed by many subsequent investigators using spectra from the Solar Maximum Mission and rocket experiments. It is also observed in stellar spectra (Maran et al. 1994). The Fe XXII line at 845.1 Å was reported in an Orbiting Solar Observatory flare spectrum by Noyes (1973).
XIII line previously reported in solar spectra by Sandlin et al. (1977). Three highly ionized nickel transitions belonging to Ni XIV and Ni XV are present in the active region spectra. The maximum fractional abundances of Ni+13 and Ni+14 are at 2.1 × 106 and 2.3 × 106 K, respectively (Arnaud & Rothenflug 1985). The 1034.48 Å Ni XIV (3s23p3 4S3/23s23p3 2P3/2) line, the 1174.65 Å Ni XIV (3s23p3 4S3/23s23p3 2P1/2) line, and the 1033.04 Å Ni XV (3s22p2 3P13s23p2 1S0) line are very prominent in the spectrum. The 1174.65 Å line was first seen in Skylab spectra and was reported to be at 1174.72 Å (Sandlin et al. 1977). Predicted wavelengths for the nickel forbidden lines were taken from Kaufman & Sugar (1986).
I isoelectronic sequence are suitable density indicators for solar plasmas. Feldman et al. (1978) showed that intensity ratios between the 2s22p3 4S3/22s22p3 2D3/2 and 2s22p3 4S3/22s22p3 2D5/2 transitions, and between the 2s22p3 4S3/22s22p3 2P3/2 and 2s22p3 4S3/22s22p3 2D3/2 transitions in Mg VI, Si VIII, and S X, and to a certain extent in Ar XII, can provide a good measure of the electron density in various regions of the corona. They were able to use the S X lines in Skylab spectra to derive densities. Subsequently, the first mentioned line ratio has been used by Doschek et al. (1997) and Laming et al. (1997) to derive densities in polar coronal holes and quiet-Sun regions using SUMER spectra.
The above transitions in Ar XII and in particular in Ca XIV are most suitable for probing the electron density of dense and hot active regions. As shown in Figure 5, the various Ar XII forbidden line ratios are sensitive to densities between 109 and 1012 cm-3. The identical transitions in Ca XIV are sensitive to densities between 1010 and 1013 cm-3. Intensity ratios between the 2s22p2 3P12s22p2 1D2 and the 2s22p2 3P12s22p2 1S0 transitions in Ar XIII and between the 2s22p4 3P22s22p4 1D2 and the 2s22p4 3P12s22p4 1S0 transitions in Ca XIII are sensitive to the electron densities that vary between 1010.5 and 1013 cm-3. The density sensitivity of Ar XIII and Ca XIII is shown in Figure 6.
Fig. 5 Fig. 6 The above discussion is included to demonstrate to readers the diagnostic importance of some of the forbidden lines. We have attempted to obtain densities for the active region from which the intensities in Table 1 were derived, but we obtain inconsistent results, i.e., we derive densities that differ by more than seems reasonable based on inaccuracies in the atomic physics of line excitation. This is likely due to time variability of the line intensities. The SUMER observations were not optimized for obtaining densities using the forbidden lines, but a good observational sequence could easily be constructed.
- Arnaud, M., & Raymond, J. C. 1992, ApJ, 398, 394 First citation in article | CrossRef | ADS
- Arnaud, M., & Rothenflug, R. 1985, A&AS, 60, 425 First citation in article | ADS
- Curdt, W., Feldman, U., Laming, J. M., Wilhelm, K., Schühle, U., & Lemaire, P. 1997, A&AS, 126, 281 First citation in article | CrossRef | ADS
- Dere, K. P., 1978, ApJ, 221, 1062 First citation in article | CrossRef | ADS
- Doschek, G. A., Feldman, U., Dere, K. P., Sandlin, G. D., Van Hoosier, M. E., Brueckner, G. E., Purcell, J. D., & Tousey, R. 1975, ApJ, 196, L83 First citation in article | CrossRef | ADS
- Doschek, G. A., Warren, H. P., Laming, J. M., Mariska, J. T., Wilhelm, K., Lemaire, P., Schühle, U., & Moran, T. G. 1997, ApJ, 482, L109 First citation in article | IOPscience | ADS
- Edlén, B. 1983, Phys. Scr., 28, 51 First citation in article | IOPscience | ADS
- Edlén, B. 1984, Phys. Scr., 30, 135 First citation in article | IOPscience | ADS
- . 1985, Phys. Scr., 31, 345 First citation in article | IOPscience | ADS
- Feldman, U., Behring, W. E., Curdt, W., Schühle, U., Wilhelm, K., Lemaire, P., & Moran, T. M. 1997, ApJS, 113, 195 (Paper I) First citation in article | IOPscience | ADS
- Feldman, U., & Doschek, G. A. 1977, J. Opt. Soc. Am., 67, 726 First citation in article | CrossRef | ADS
- . 1991, ApJS, 75, 925 First citation in article | CrossRef | ADS
- Feldman, U., Doschek, G. A., Mariska, J. T., & Mason, H. E. 1978, ApJ, 226, 674 First citation in article | CrossRef | ADS
- Feldman, U., Doschek, G. A., & Seely, J. F. 1985, MNRAS, 212, 41p First citation in article | CrossRef | ADS
- Harrison, R. A., et al. 1995, Sol. Phys., 162, 233 First citation in article | CrossRef | ADS
- Kaufman, V., & Sugar, J. 1986, J. Phys. Chem. Ref. Data, 15, 321 First citation in article | CrossRef | ADS
- Kohl, J. L., et al. 1995, Sol. Phys., 162, 313 First citation in article | CrossRef | ADS
- Laming, J. M., Feldman, U., Schühle, U., Lemaire, P., Curdt, W., & Wilhelm, K. 1997, ApJ, 485, 911 First citation in article | IOPscience | ADS
- Lemaire, P., et al. 1997, Sol. Phys., 170, 105 First citation in article | CrossRef | ADS
- Maran, S. P., et al. 1994, ApJ, 421, 800 First citation in article | CrossRef | ADS
- Noyes, R. W. 1973, in High Energy Phenomena on the Sun, ed. R. Ramaty & R. G. Stone (NASA SP-342), 231 First citation in article | ADS
- Peacock, N. J., Stamp, M. F., & Silver, J. D. 1984, Phys. Scr. T8, 10 First citation in article | IOPscience | ADS
- Sandlin, G. D., Brueckner, G. E., & Tousey, R. 1977, ApJ, 214, 898 First citation in article | CrossRef | ADS
- Wilhelm, K., et al. 1995, Sol. Phys., 162, 189 First citation in article | CrossRef | ADS
- . 1997, Sol. Phys., 170, 75 First citation in article | CrossRef | ADS
Full image (388kb) | Discussion in text
FIG. 1.EIT image of the 1997 September 6 active region
Full image (244kb) | Discussion in text
FIG. 2.Forbidden lines of Ca in an active region spectrum. Note that they are absent in the quiet-Sun spectrum included for comparison.
Full image (391kb) | Discussion in text
FIG. 3.Spectrum showing the Fe XIX flare line near 1118 Å. Top: a tracing through the spectrum; bottom: distribution of emission along the slit length. Note the small regions in which the Fe XIX emission is confined and the large Doppler motions shown by the spectral width of the lines.
Full image (61kb) | Discussion in text
FIG. 4.Energy levels (cm-1) and transitions (Å) for Ar XII (in parentheses) and Ca XIV detected by SUMER.
Full image (69kb) | Discussion in text
FIG. 5.Density-sensitive line ratios for Ar XII and Ca XIV
Full image (53kb) | Discussion in text
FIG. 6.Density-sensitive line ratios for Ar XIII and Ca XIII
|545.26...||545.21||Ca XIV||2s22p3 4S3/22s22p3 2P3/2||0.43|
|579.85...||579.83||Ca XIV||2s22p3 4S3/22s22p3 2P1/2||0.37|
|592.23 cP...||Fe XIX||2s22p4 3P22s22p4 1D2|
|648.68...||648.92||Ca XIII||2s22p4 3P12s22p4 1S0||0.29|
|656.69...||656.60||Ar XIII||2s22p23P12s22p2 1S0||0.29|
|670.34...||670.35||Ar XII||2s22p3 4S3/22s22p3 2P1/2||0.24|
|880.43...||880.35||Ca XIV||2s22p3 4S3/22s22p3 2D5/2||1.67|
|943.61...||943.70||Ca XIV||2s22p3 4S3/22s22p3 2D3/2||2.43|
|974.86P...||974.82||Fe XVIII||2s22p 2P3/22s22p 2P1/2||3.62|
|1018.87P...||1018.72||Ar XII||2s22p3 4S3/22s22p3 2D5/2||1.13|
|1033.04...||1033.2||Ni XV||3s23p2 3P13s23p2 1S0||1.19|
|1034.48...||1034.9||Ni XIV||3s23p3 4S3/23s23p3 2P3/2||1.91|
|1054.62P...||1054.69||Ar XII||2s22p3 4S3/22s22p3 2D3/2||0.31|
|1098.44...||1098.42||Ca XV||2s22p2 3P12s22p2 1D2||0.26|
|1118.08P...||1118.08||Fe XIX||2s22p4 3P22s22p4 3P1||0.17|
|1133.79...||1133.76||Ca XIII||2s22p4 3P22s22p4 1D2||4.95|
|1153.16P...||1153.14||Fe XVII||2s22p53s 3P12s22p53s 3P0||1.24|
|1174.65P...||1174.72||Ni XIV||3s23p3 4S3/23s23p3 2P1/2||1.53|
|1277.22P...||1277.23||Ni XIII||3s23p4 3P13s23p4 1S0||0.78|
|1291.61...||1291.2||Ca XIV||2s22p3 2D3/22s22p3 2P3/2||0.26|
|1330.54...||1330.37||Ar XIII||2s22p2 3P12s22p2 1D2||0.22|
|1375.96P...||1375.93||Ca XV||2s22p2 3P22s22p2 1D2||0.39|
NOTE. P denotes "previously identified." See text.
a Predicted wavelengths for argon, calcium, and iron transitions are from Edlén (1983, 1984, 1985); predicted wavelengths for Ni lines are from Kaufman & Sugar (1986).
b Intensities are given in units of mW st-1 m-2 and are obtained from a single active region reference spectrum.
c This line was observed in a different reference spectrum from the other lines, and therefore no intensity is given.
Image of typeset table | Discussion in text
|LEVEL||Ar XIII||Ca XV|
|3P1...||9859 a||17555 a|
|3P2...||21859 a||35917 a|
a Energy levels are from Edlén (1985). Numbers in parentheses are calculated values.
Image of typeset table | Discussion in text
|Ar XII||Ca XIV|
a Energy levels are from Edlén (1984). Numbers in parentheses are calculated values.
Image of typeset table | Discussion in text
a Energy levels are from Edlén (1983).
Image of typeset table | Discussion in text