Winds from lensed star-forming galaxies

Oh, you thought Jane was off the mountain so she was done blogging? Incorrect, my friend! Jane strikes again below.

by Jane Rigby

This trip to Las Campanas is the last run in a six-year project to obtain high-quality rest-frame ultraviolet spectra for fifteen gravitationally lensed galaxies.  I’m using the MagE instrument on Magellan, which is a simple, powerful echellette spectrograph.

This graphic shows a reconstruction (at lower left) of the brightest galaxy whose image has been distorted by the gravity of a distant galaxy cluster. The small rectangle in the center shows the location of the background galaxy on the sky if the intervening galaxy cluster were not there. The rounded outlines show distinct, distorted images of the background galaxy resulting from lensing by the mass in the cluster. The image at lower left is a reconstruction of what the lensed galaxy would look like in the absence of the cluster, based on a model of the cluster's mass distribution derived from studying the distorted galaxy images. From NASA, ESA, and Z. Levay (STScI). Science credit: Sharon et al. (2012).

These spectra are ridiculously rich in spectral diagnostics that probe different aspects of the galaxies.

First, these spectra contain diagnostics of the massive stars in these galaxies.  Such massive stars that will burn hot, die young, explode as supernovae, and drive winds of gas that may escape for good, or may rain back down and trigger future star formation.  Since we can almost never obtain good spectra of the combined stellar output of galaxies as they appeared billions  of years ago (z~>1), we don’t actually know how important are Wolf-Rayet stars, or how much the stars are enriched in heavy elements such as carbon, oxygen, silicon, sulfur, and iron.

Second, these spectra also contain diagnostics of the gas in these galaxies.  In distant galaxies, that gas is usually blueshifted toward us, which means the gas is flowing out of each galaxy.  That’s interpreted as winds, driven by the supernovae of the aforementioned massive stars.  This dataset connects the wind of a galaxy to its population of massive stars.  The thing I’m currently most excited about with this dataset is that we've gotten spectra of multiple physical regions within a few galaxies.  What we’ve seen so far in one galaxy (Bordoloi et al., submitted to MNRAS) is that the properties of the wind correlates with the properties of the closest star-forming region.  The wind appears to be “locally sourced”, arising close to the star-forming regions, rather than some uniform wind out at many kpc.  We now have spectra of multiple regions in additional galaxies to test this picture.

Third, these spectra contain nebular emission lines of magnesium and carbon (Mg II and [C III]+C III]).  These lines are quite bright in distant galaxies, and there’s been a flurry of recent papers trying to understand why.  I’ve written two myself.  These nebular lines must be powered by the hot stars, though the location of this emission within the galaxies is not well constrained.

Figure 4 from Bayliss, Rigby, & Sharon et al. (2014). Top: GMOS spectrum covering the rest-frame wavelength range Δλ = 1200–1600 Å. Spectral lines are indicated by type: black dashed lines are nebular emission lines, short solid red lines are stellar photospheric absorption features, medium length blue lines are ISM absorption lines, and long purple lines indicate transitions that could be either stellar photospheric or ISM (or more likely, a blend of the two). The error array is over plotted as the black dotted line, and the fit to the continuum level across the spectrum is plotted as a thin green line. The apparent emission feature that we observe at ~6290 Å is the result of a pernicious sky subtraction residual, and lines resulting from intervening absorption systems are indicated with downward facing arrows. Bottom: GMOS spectrum covering the rest-frame wavelength range Δλ = 1600–1950 Å. Lines are indicated according to the same scheme as the top panel. The N iii] 1750 emission line is only detected at ~2σ, but we indicate its location here because it is used later to constrain the relative nitrogen abundance.

None of this is science generally possible for typical galaxies with current telescopes.  Galaxies are just too faint, and our telescopes are just too small.  What makes this project possible is that our targets are among the brightest known galaxies that have been gravitationally lensed.  We’re using galaxy clusters as natural telescopes, to bend extra light from these galaxies toward us.

I’ve been working on this project with my pals & collaborators Keren Sharon, Matt Bayliss, Mike Gladders, and Rongmon Bordoloi, all of whom rock.