- Title: NEW CONSTRAINTS ON COSMIC REIONIZATION FROM THE 2012 HUBBLE ULTRA DEEP FIELD CAMPAIGN
- Authors: Brant E. Robertson et al.
- First Author’s Institution: University of Arizona
Often, the most profound questions in physics are those you wouldn’t think to ask. A perfect example is that leading to the discovery of the Higgs boson: why do particles have mass? Today’s paper is on a question in this category in astronomy: why is the Universe ionized and not neutral? After all, the background temperature today, 2.73 K, is far lower than the ~10,000 K needed to ionize hydrogen. Well, what keeps us, and the Universe, warm? Stars. In particular, the Universe became ionized (for the 2nd time) because of starlight heating up the neutral hydrogen and popping electrons off their nuclei.
When this occurred, and how quickly, is a tremendous open question in astrophysics. If you know the answer, please write us and we will be glad to co-author the paper and share your Nobel prize. Nonetheless, we do have some constraints, which Robertson et al, assemble. They particularly focus on what new things we learned from the latest Ultra-Deep-Field Hubble Space Telescope observations, which go to z~8 and limiting magnitude -17. Redshift z is a proxy for distance, while magnitude is a proxy for how faint the minimally detecting objects are. Ultra-Deep-Field just means it goes to high-z (very deep out into space).
The constraints are:
- Number of stars and luminosity. Stars produce the light to ionize hydrogen—so the more stars, the more ionization, and the hotter and brighter these stars, the more ionization. So measuring how numerous and how luminous and hot stellar populations were at different times in the Universe’s history (this can be derived, with some additional assumptions, from stellar mass density over time) gives you a budget of photons to spend reionizing.
- Thomson optical depth. Electrons scatter photons in a process called Thomson scattering. The more ionized the Universe, the more electrons are out there scattering photons, and the less far back into the past/ far out from us you can see, as photons from farther away have higher chance of being scattered.
- Lyman-alpha emission from galaxies passing through the inter-galactic medium (IGM). Lyman alpha emission is just when a photon leaves a hydrogen atom be causes its electron drops from n>1 to n=1. Galaxies emit Lyma-alpha photons and these are then absorbed by neutral hydrogen in the IGM. Thus, the more galaxies we can see, the less neutral hydrogen in the IGM we can infer. This constrains the history of neutral hydrogen in the IGM.
- The kinetic Sunyaev-Zeldovich (kSZ) effect. Whew, what a mouthful. Cool facts: I have met Sunyaev (and got his autograph!), and Zeldovich basically did everything in cosmology after retiring from heading the Soviet nuclear weapons program. The kSZ effect has to do with the motions of electrons during reionization. In general, photons from the cosmic microwave background (CMB) scatter off of electrons in between it and us. Normally, the electrons’ motion is completely random so the scattering averages out to zero. But reionization can make the electron’s motion coherent in different patches, and so net on net the scatterings no longer cancel out. We can detect this and use it to constrain the total length of time reionization could have gone on.
Robertson et al. also include a few other constraints omitted for brevity here. Overall they conclude that the galaxy population measured in the Hubble UDF program is not enough to reionize the Universe by the time we know it is (z~6) and simultaneously get the right Thomson optical depth (see 2) above). So they do the simplest thing they know how to: extrapolate from the known properties of galaxies in the sample and assume that galaxies continue to exist with the same properties out to z~12-15 and limiting magnitude -13. In other words, that there are more galaxies out there, basically the same as those we see, but too far away and too faint to show up in Hubble’s UDF.
Reionization is an incredibly complex area of research that is only just being attacked with both observations and numerical simulations, so we should expect many more exciting constraints in the near future. After all, how many papers do you read a day that have a plot showing the history of the element hydrogen over the last 10 billion years?