Stacking Quasars

I heard an interesting talk today by J. Xavier Prochaska, from Santa Cruz. The abstract is below. The basic idea was that we can co-add quasar spectra to get a cleaner signal of opacity (due to lyman absorption) and thus the mean free path as a function of redshift. We should totally do this with BOSS.

~~~
"Direct Measurement of the IGM Opacity to H I Ionizing Photons"
Paper: http://arxiv.org/abs/0910.0009
We present a new method to directly measure the opacity from HI Lyman limit (LL) absorption k_LL along quasar sightlines by the intergalactic medium (IGM). The approach analyzes the average (``stacked'') spectrum of an ensemble of quasars at a common redshift to infer the mean free path (MFP) to ionizing radiation. We apply this technique to 1800 quasars at z=3.50-4.34 drawn from the Sloan Digital Sky Survey (SDSS), giving the most precise measurements on k_LL at any redshift. From z=3.6 to 4.3, the opacity increases steadily as expected and is well parameterized by MFP = (48.4 +/- 2.1) - (38.0 +/- 5.3)*(z-3.6) h^-1 Mpc (proper distance). The relatively high MFP values indicate that the incidence of systems which dominate k_LL evolves less strongly at z>3 than that of the Lya forest. We infer a mean free path three times higher than some previous estimates, a result which has important implications for the photo-ionization rate derived from the emissivity of star forming galaxies and quasars. Finally, our analysis reveals a previously unreported, systematic bias in the SDSS quasar sample related to the survey's color targeting criteria. This bias potentially affects all z~3 IGM studies using the SDSS database.