Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Most distant galaxy

  1. Jul 15, 2008 #1
    I have been reading this interesting article http://www.subarutelescope.org/Pressrelease/2006/09/13/index.html on one of the earliest galaxies that has ever been observed (z=6.96), that existed when the universe was just 760+/-15 million years old or about 6% of its curent age. See also http://www.nature.com/nature/journal/v443/n7108/abs/nature05104.html

    The article is very readable and nicely illustrated, but a couple of things puzzle me and I hope someone here can enlighten me. It describes the period of "recombination" when the universe transitions from an opaque period due to scattering of light by ions to a period when the universe becomes transparent when the ions combined to form hydrogen atoms. It then goes on to say that we do not see many galaxies at the early epoch (@760 million years) because neutral hydrogen absorbs the light from the early stars and they only later become visible when the young stars reionize the neutral hydrogen. That seems to contradict the earlier statement but maybe I am missing some important point??

    Another issue is that they describe the early universe as being different to later epochs because of the relative rarity of galaxies in the early epoch. That would seem natural if all galaxies did not form simultaneously in the early universe and I would have thought that Malquist bias would be an additional factor at the enormous distances we are talking about here. See http://en.wikipedia.org/wiki/Malmquist_bias . However, they do not mention Malmquist bias anywhere in the article. Have they taken Malmquist bias into account and simply not mentioned it or is it a non issue in this context?
    Last edited by a moderator: Apr 23, 2017
  2. jcsd
  3. Jul 15, 2008 #2


    User Avatar
    Science Advisor

    The recombination is the epoch after which the cosmic radiation background mainly ceases to interact with electrons and allows them to get bounded to protons to form neutral hydrogen. Due to expansion of space most of the photons in the cosmic radiation background will lose energy and will never be able again to ionize hydrogen. However, later on photons are created in stars and quasars that have enough energy to ionize the neutral hydrogen. This process is called reionization.

    The point you make with Malquim bias is insightful, but I would expect that they have analyzed it carefully and taken it into account. They are talking about a decrease in the number density for epochs separated 0.4 in redshift. I would expect that similar accuracy in the detection can be expected in that range. However, I do not know their work so I might be wrong.
  4. Jul 16, 2008 #3
    I would think that the Malquim bias is to be applied to large surveys where one has a large number of objects and wants to consider the completeness of their survey. For galaxies at these high redshift there aren't large samples, there are a handfull of isolated observations. The bias is extreme and there is no point to go into detailed analysis. However even a single galaxy discovered at these high redshifts, corresponding roughly to the first billion years after the big bang, is interesting. A very active area of research is when the first (population III) stars formed and when they gathered into objects we would call galaxies (and how long the transition to population II stars took and ...). The James Web Telescope is going to make observations of this epoch when it goes up and large radio array telescopes being built now will tell us about the epoch of reionization (when the first luminous objects reionized the neutral Hydrogen)
  5. Jul 16, 2008 #4


    User Avatar
    Science Advisor
    Gold Member

    An even earlier galaxy: A Lyman Break Galaxy Candidate at z~9
  6. Jul 16, 2008 #5
    That is impressive work, but the dropout technique is in a different category then a spectroscopically confirmed galaxy. Note the mention of degeneracy with a z = 2-3 model in the last sentence.
  7. Jul 16, 2008 #6


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    z = 7.6 ?

    If Garth's z = 9 is not confirmed, there is this z = 7.6 candidate

    the press release is dated 12 February 2008

    light from the galaxy would have been emitted when expansion was an estimated 700 million years old.

    Ned Wright expresses skepticism ("NASA's budget must be tight") because no lines have been observed. I will see if I can find more about this one.

    Here is the article
    it was accepted by Astrophysical Journal for publication and was reported in press as of May 2008
    it is another "Lyman break" determination, unless I'm mistaken it will take a more powerful infrared telescope like the James Webb to see lines and be more certain about the redshift.
    Last edited by a moderator: May 3, 2017
  8. Jul 16, 2008 #7


    User Avatar
    Science Advisor
    Gold Member

    Obviously the red shift of these galaxies will need to be confirmed once telescopes are sensitive enough to observe spectral lines in the IR but 96% probability that z > 7 is fairly (>3[itex]\sigma[/itex]) convincing.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook