Supernovae in external galaxies provide rare, fleeting opportunities to probe the interstellar media of the host galaxies via absorption-line spectroscopy. In principle, such observations can reveal the behavior of various tracers of the ISM under somewhat different environmental conditions from those typically sampled in the local Galactic ISM. Toward the recent Type Ia SN 2014J in M82, complex absorption is seen from Na I, K I, Ca I, Ca II, CH, CH+, and CN; many of the diffuse interstellar bands are also detected. Comparisons of the column densities of the atomic and molecular species and the equivalent widths of the DIBs reveal both similarities and differences in relative abundances, compared to trends seen in the ISM of our Galaxy and the Magellanic Clouds. Shifts in the centroids of some of the DIBs, relative to the envelope of the K I absorption, are likely due to component-to-component variations in W(DIB)/N(K I) that may reflect differences in molecular content. The column densities of Na I and K I and the equivalent widths of the DIBs can be used to estimate the host galaxy reddening and extinction -- which can be compared with values derived from optical/near-IR photometry of the SN.
Return to top of this page
For the small currently available sample of quasar absorption-line systems (QSOALS; mostly at redshifts z near 2) having relatively well determined abundances for heavy elements such as Zn, Cr, Si, and Fe, the overall metal abundances seem to be roughly 0.1 X Solar. The distinctly non-Solar relative abundances are likely due to differences in nucleosynthetic history, further modified by some degree of depletion (dust is apparently present, though not at the levels commonly seen in our Galaxy). If accurate abundances for enough key species can be determined, for QSOALS at various redshifts (0.5 < z < 4.0), then one can separately follow the build-up of the heavy elements and of dust over an appreciable fraction of the history of the Universe - with substantial implications for understanding the evolution of galaxies.
The current generation of large optical telescopes, when equipped with spectrographs capable of moderately high resolution (5-6 km/s), allow some of the detailed analyses used in studies of the Galactic ISM to be applied to the "individual" components seen in QSOALS at 0.5 < z < 4.0. At that resolution, components due to predominantly neutral gas can be better separated from those due to ionized gas, metal lines in the Lyman alpha forest can be separated from the Lyman alpha lines, and individual Lyman alpha forest lines can be resolved. With achievable rest frame detection limits of several mA, accurate abundances can be determined for many species in damped Lyman alpha and Lyman limit systems. The abundances of Zn, S, and O (typically little depleted) yield initial estimates for the overall metallicity and nucleosynthetic mix; comparisons of Mg, Si, Al, and Ti with S, and of Ni, Cr, Mn, and Fe with Zn (grouping elements with similar nucleosynthetic origins) should yield the level and pattern of depletion - and thus more precise information on the underlying total abundances (see figure below). The level and spread of the abundances of various elements, as functions of z, will provide important constraints for ideas on the evolution of galaxies: What were the rates of Type I and Type II supernovae? Do the damped Lyman alpha systems arise in early galactic disks or in gas-rich dwarfs with active star formation? Renewed searches for trace species (e.g., C I) and excited state lines (e.g., C I*, C II*, Si II*), will provide information on the radiation fields (including the evolving cosmic microwave background) and local densities in the QSOALS. Comparisons of abundances and physical conditions for associated (z_abs = z_em) systems with those for systems with z_abs < z_em will yield information on the QSO's and their environment. Observations of similar absorption-line systems toward gamma-ray bursts (GRBs) are providing information on the (typically somewhat higher column density) ISM in the galaxies hosting the GRBs.
While the QSO spectra obtained for the Sloan Digital Sky Survey (SDSS) are of lower resolution, the sheer number of spectra now available allows interesting statistical tests for the properties of QSOALS and estimation of average properties via careful averaging of the individual spectra. Such averages can yield detections of weak absorption features and indications of the reddening/extinction of the QSOs due to the intervening absorbers (e.g., York et al. 2006). The typical extinction curve characterizing the QSOALS appears to be similar to that found in most of the SMC (i.e., steep far-UV rise with no 2175 A bump).
Direct detection of silicate dust in some of the higher column density QSOALS is now possible via observations of (redshifted) 9.7 micron absorption with the Spitzer Space Telescope (Kulkarni et al. 2007). In combination with observations of gas-phase absorption from Si II, the dust-phase abundances derived from Spitzer will thus yield both total silicon abundances and direct measures of the depletion.
(Top) Here we compare the relative abundances found for the SMC and LMC ISM with the average Galactic warm, low density cloud (W), cold, dense cloud (C), and halo cloud (H) gas-phase abundance patterns (for Al, Si, Ti, Cr, Mn, Fe, and Ni), expressed as ratios with respect to Zn, relative to Solar ratios (i.e., [X/Zn]). These patterns reflect both the nucleosynthetic history of the gas and any selective depletion. The [Zn/H] ratio is shown at the far right; the points for the LMC and SMC reflect primarily the lower than Solar metallicity of those systems. The pattern seen for the main LMC (L) component (blend) is quite similar to the Galactic warm cloud pattern; the SMC (S) components resemble more the (less depleted) halo cloud pattern. Since the stellar abundance ratios are not much different from Solar for these elements in the LMC and SMC, the depletion patterns may also be similar to those found in the ISM of our Galaxy -- despite the lower metallicities and dust-to-gas ratios in the LMC and SMC.
(Bottom) The pattern seen for a small number of QSOALS (0.7 < z < 3.4; from current literature) is somewhat similar to the halo and SMC patterns (though a large range is evident for most [X/Zn]). Note also the wide range in [Zn/H] -- from -0.3 to <-1.7. Both nucleosynthetic and depletion effects (comparable in magnitude) may be present. The similarities between the depletion patterns in the LMC and SMC and those seen in low-density environments in the Galactic disk and halo suggest that we may infer similar patterns in the QSOALS -- which should allow the underlying total abundances in the QSOALS to be determined more accurately.
Relative gas phase abundances for four individual QSOALS, with respect to Zn, relative to Solar photospheric values, compared to relative abundances found for Galactic warm (W) and halo (H) clouds and for Galactic metal-poor stars (*) with [Fe/H] < -1. Limits with a cross-bar denote cases where a limit (4-sigma) is available for one of the two species (generally Zn). Possible 2-4 sigma detections are indicated by circles superposed on the arrows denoting the 4-sigma limits. Error bars are 1-sigma. A range of relative abundance patterns, presumably reflecting different nucleosynthetic history and depletion effects, is apparent. For example, the two systems on the left appear to have relative abundances more consistent with the metal-poor halo star pattern, with little (if any) depletion ([Si/Zn] high, [Mn/Zn] low). The relative abundances found for the two systems on the right, however, seem more consistent with the halo cloud (and SMC ISM) pattern, suggesting the presence of some dust/depletion ([Mg/Zn], [Si/Zn], [Ti/Zn] all low).
You may also view a similar plot of the relative abundances for 18 QSOALS . (The data have been taken largely from work by Lu, Sargent, Wolfe, Pettini, and their collaborators.)
I have compiled a catalogue of QSO absorption-line systems for which elemental abundances have been determined from moderate-to-high-resolution optical spectra (in the literature to 2002 Feb -- in need of updating).
See also Jason Prochaska's DLA abundance compilation .
See also discussions in:
Return to top of this page
In determining the statistical properties (column densities, line widths, clustering) of the Lyman alpha forest (H I) lines, it is important both to resolve the individual Lyman alpha components and to remove any lines due to heavier elements from the samples.
Return to top of this page
Return to my home page
Go to UC Astro home pageDan Welty / email@example.com
Last modified 22 Aug 2014