Next Contents Previous

7. ABSORPTION BY MILKY WAY HALO GAS AND THE ABSORPTION LINE SYSTEMS SEEN IN THE SPECTRA OF QSOS

Table 4 lists the approximate absorption line strengths expected for a sight line through the Milky Way disk and halo at the position of the sun. The measurements are appropriate for a sight line which is inclined at 60° from the galactic disk and which passes through halo and disk gas with positive and negative z. The resulting absorption line system would be called a damped Ly-alpha system with mixed ionization. The HI Ly-alpha line would mostly arise in disk gas while the highly ionized lines of NV, CIV and SiIV would mostly arise in halo gas. The lines of lower ionization would be produced in disk and halo gas. The absorption line system would exhibit multiple components and have a velocity spread as recorded in the strongest lines of CII, SiII and MgII of about 140 km s-1. As shown by Savage and Jeske (1981) absorption line systems with the basic characteristics of the system listed in Table 4 are sometimes found in the spectra of QSOs. The frequency of occurrence of the subset of QSO systems having the characteristics of the Milky Way `solar region' system listed in Table 4 is small because of the presence of the strong damped Ly-alpha line from disk gas having a HI column density of about 6 x 1020 cm-2. Even if only the halo part of the absorption is considered, the fact that the strong low ionization lines of SiII and CII are about 2x stronger than the CIV and SiIV absorption appears to place the absorption system listed in Table 4 in a category which is different from the most common QSO mixed ionization systems having absorption redshifts between approximately 1 and 2 (Wolfe 1986; Danly, Blades and Norman 1987).

Table 4. Milky Way halo and disk gas absorption characteristics a
species lambda(Å) b log [flambda] c Wlambda(Å) d species lambda(Å) b log [flambda] c Wlambda(Å) d
HI 1215.67 2.70 18. SII 1259.52 1.30 0.35
CI 1656.93 2.35 0.07 SII 1253.81 1.13 0.25
CI 1560.31 2.10 0.06 SII 1250.59 0.83 0.14
CI 1277.21 2.30 0.08 MnII 2576.11 2.87 0.16
CII 1334.53 2.20 0.70 MnII 2593.73 2.76 0.14
CII* 1335.71 2.20 0.12 MnII 2605.70 2.62 0.12
CIV 1548.20 2.48 0.34 FeII 2599.40 2.77 0.80
CIV 1550.77 2.18 0.18 FeII 2585.88 2.25 0.75
NI 1200.71 1.72 0.25 FeII 2382.03 3.04 0.80
NI 1200.22 2.03 0.25 FeII 2373.73 2.08 0.40
NI 1199.55 2.20 0.30 FeII 2343.49 2.70 0.80
NV 1242.80 1.97 0.06 FeII 1608.46 2.19 0.30
NV 1238.81 2.28 0.12 NiII 1741.56 2.07 0.08
OI 1302.17 1.81 0.60 CrII 2055.59 2.54 0.07
MgI 2852.13 3.75 0.30 ZnII 2062.02 2.62 0.06
MgI 2026.48 2.35 0.06 ZnII 2025.51 2.92 0.12
MgII 2802.70 2.92 1.25
MgII 2795.53 3.22 1.35
AlII 1670.79 3.50 0.50
AlIII 1862.79 2.70 0.13
AlIII 1854.72 3.00 0.25
SiII 1808.01 0.83 0.13
SiII 1526.72 2.55 0.45
SiII 1304.37 2.28 0.47
SiII 1260.42 3.08 0.60
SiII 1193.28 2.78 0.50
SiIII 1206.51 3.30 0.55
SiIV 1402.77 2.57 0.12
SiIV 1393.76 2.87 0.23

a For a galactic sight line at the position of the sun extending through halo and disk gas to each side of the galactic disk at an inclination of approximately 60° from the disk. The list is incomplete for lambda(rest) < 1200 Å, which is near the short lambda limit of the IUE spectrograph.
b Wavelengths from Morton and Smith (1973) are vacuum for lambda < 2000 Å and air for lambda > 2000 Å.
c The f values are from the recent compilation by Pwa and Pottasch (1986).
d The equivalent widths listed are inferred from IUE absorption measurements toward HD100340 (l = 250°, b = 61°, z = 5.2 kpc) and HD18100 (l = 218°, b = -63°, and z = -3.4 kpc). The absorption toward these two stars seems representative of absorption toward the north and south galactic poles, respectively. For highly saturated lines the entry is entirely determined by the measurements for HD100340 since the strongly saturated lines toward this star have FWHM approx 140 km s-1 compared to about 90 km s-1 for HD18100. Samples of the spectra are found in Figure 6. For unsaturated lines the equivalent widths listed are a straight addition of the measurements toward HD100340 and HD18100. In some cases the data have been supplemented with measurements based on other directions. Numerous weaker lines are not listed. For a reasonably complete line list for the 1200 to 3000 Å wavelength region of all the interstellar lines detected toward zeta Oph see Pwa and Pottasch (1988). The absorption toward HD100340 resembles that seen toward the LMC at velocities less than 140 km s-1, while that toward HD18100 resembles the absorption seen toward the SMC at velocities less than 80 km s-1. Data for the LMC and SMC sight lines were compared to QSO absorption line measurements in Savage and Jeske (1981). The absorption line equivalent widths listed above have contributions from disk and halo gas. Those lines which primarily are produced in the disk include all the weak low ion lines and the HI Ly-alpha line. The strong low ionization lines and the lines of intermediate and high ionization are mostly produced in halo gas.

There are many reasons why we might expect the character of the Milky Way halo absorption line system found for gas in the solar neighborhood and those systems most commonly found in QSO spectra to appear different. First, the frequency of occurrence of the most common systems implies that if they are associated with galaxy halos then the absorption likely occurs many Holmberg radii from the center of the absorbing galaxy. Thus we need to consider how the appearance of the Milky Way halo system might change as the sight line passes further and further from the galactic center. Second, the character of the extragalactic background radiation which illuminates and ionizes the galaxy halo gas will change with redshift. Third, the character of the halo gas will change as the vigor of the fountain changes. Thus the halo gas characteristics may be sensitive to galactic evolution and in particular to the massive star production rate.

The appearance of a galaxy disk-halo absorption line system will change with galactocentric distance because of a number of factors. Some of these include:

  1. changes in the vigor of the galactic fountain flow with galactocentric distance because of changes in the production rate of massive stars;
  2. changes in the gravitational acceleration in the z direction, gz, with galactocentric distance. gz determines the degree of heating necessary to have gas reach large z distances. The observed increase of the HI scale height with galactocentric distance (see Section 2) may be associated with a decrease in gz;
  3. galaxy radial abundance gradients which will influence the detectability of halo gas observed via metal absorption lines;
  4. changes in the relative importance of photoionization to collisional ionization with galactocentric distance which will modify the ionization state of the halo and disk gas;
  5. changes in the physical density and extreme ultraviolet optical depth of the halo gas and associated disk gas with galactocentric radius which will modify the ionization state of the gas; and
  6. changes in the kinematical properties of the halo gas with galactocentric distance which will almost certainly occur because of changes in the vigor of the fountain and changes in the gravitational acceleration.

It would be interesting to consider the consequences of the six effects listed above and then also evaluate the consequences of changing the extragalactic background and the stellar content of the underlying galaxy. Unfortunately such a task is full of uncertainties.

The best way to establish what the Milky Way looks like at large galactocentic distances is to look in those directions where galactic rotation permits a velocity separation of distant matter from local matter, i.e. in or near the directions l approx 90° and l approx 270° and at low enough galactic latitudes to permit the viewing of very distant gas at large |z| distances. Some objects which have been observed at optical and/or ultraviolet wavelengths which may meet these criteria are:

  1. SN1980K in NGC6946 at l = 96° and b = +12° (Pettini et al. 1982). The published data for SN1980K reveal strong absorption in MgII lambda2800, MgI lambda2852 and FeII lambda2599 to high negative velocities of about -150 km s-1 which can be attributed to absorption by high |z| gas situated in the outer parts of the Milky Way. Unfortunately, the spectra obtained were too noisy to study absorption by highly ionized gas;
  2. SN1983N in M83 at l = 315° and b = +32° (Jenkins et al. 1984; d'Odorico, Pettini and Ponz 1985). The published optical data for SN1983N reveal multicomponent CaII absorption associated with the Milky Way at velocities of -45, -8 and +43 km s-1. In addition a component of unknown origin is found at +248 km s-1. Although IUE ultraviolet data were obtained for this object, they have not been published;
  3. the nuclear region of the galaxy NGC3783 in the direction l = 287° and b = +23° shows Milky Way absorption in the CaII H and K lines at vLSR = -39, -21, -1, +40, and +60 km s-1. In addition there is a CaII feature of unknown origin at 241 km s-1 (West et al. 1985); and
  4. stars and other sources in the LMC near l = 280° and b = -33° (Savage and deBoer 1981; Dupree et al. 1987; deBoer et al. 1987). The most extensive data currently available which may provide clues about what the Milky Way looks like in absorption at large galactocentric distances is the extensive collection of ultra-violet and optical data for sight lines to the LMC. If Savage and deBoer (1981) are correct in assigning the absorption features seen near 60 and 120 km s-1 to absorption in distant parts of the Milky Way, then those features provide the best existing information about what the outer parts of the Milky Way looks like in the resonance lines of abundant elements (see Figure 3 and 4 and the many spectra shown in Savage and deBoer 1981). Even if some or all of the 60 and 120 km s-1 absorption is instead associated with sheets of matter in the vicinity of the LMC as advocated by Songaila et al. (1986), it is possible that the absorption characteristics of this gas may be representative of the absorption produced in the outlying regions of galaxies at redshifts of zero. It is noteworthy that these absorption features have mixed ionization characteristics with the strong low ionization lines of CII and SiII being two to three times stronger than the higher ionization lines of CIV and SiIV (see Savage and deBoer 1981; Savage and Jeske 1981). It seems reasonable to propose that such a behavior is a general characteristic of gas in the outer regions of spiral galaxies at small redshift.

I am grateful to my colleagues at the University of Wisconsin-Madison for their many helpful discussions relating to the properties of Milky Way halo gas. I'm appreciative of support for this work through NASA grant NAG 5-186.

Next Contents Previous