Galaxy Zoo Starburst Talk

Does our Quench data contradict Hubble's Cosmos survey conclusion?

  • twins by twins

    When we create a scatter diagram relating redshift (x axis) to Log Mass of quenched galaxies (y axis), we get a roughly horizontal scatter around 10 or 11 Log Mass (although data points do lessen as redshift increases). However, if we accept the conclusions of the Hubble COSMOS survey as described in the recent Astronomy article, we would expect the datapoints to go from the upper left to the lower right. In other words, more massive quenched galaxies would have a lower redshift and less massive galaxies would have a higher redshift. Can anyone explain this seeming contradiction? Is this perhaps because the sample was chosen so that they all have about a 10 or 11 Log Mass?

    Posted

  • JeanTate by JeanTate in response to twins's comment.

    Are you referring to Hubble’s COSMOS survey solves “quenched” galaxy mystery, "By Hubble ESA, Garching, Germany — Published: August 1, 2013"?

    If so, then:

    • it's always important - critical in our case - to read the actual, peer-reviewed published papers; all too often, in my experience, what appears in a popsci article can be misleading, because important details are omitted, or 'word pictures' inadequately reflect the 'number pictures' painted by graphs, plots, charts and equations. And so on. Does anyone have a copy of the paper (or papers) that this Astronomy article is based on?
    • the 'redshift depth' of the current Quench project is shallow - there are only ~10 with z > ~0.3, and < 300 with z > 0.2 - while the COSMOS one is quite a bit greater (the article doesn't say, but "throughout the last 8 billion years of cosmic history" implies it goes to at least z ~1)
    • we have to address the Malquist bias in our sample (presumably the astronomers who did the COSMOS research carefully addressed this bias in their work)
    • we don't - yet - know how the QS and QC were selected (other than that the QC is 'matched' to the QS, by having comparable stellar masses and redshifts within 0.02). This question has come up at least twice already (here and here).

    Posted

  • jtmendel by jtmendel scientist, moderator

    It is also important to keep in mind that there are two relevant (but different!) timescales to consider here: one is the age of the Universe, which can be estimated from the redshift, and one is the age of stars in the galaxies themselves, which is related to when they formed.

    In the COSMOS article (which can be found here, for anyone interested) the authors are focused mainly on the evolution of galaxy sizes as a function redshift (~age of the Universe) at fixed stellar mass. Deep imaging surveys have found that high-redshift quenched galaxies are much more compact at fixed stellar mass than their counterparts in the local Universe. This has led to the idea that galaxy mergers play an important part in "puffing up" passive galaxies over time, and there has been a lot of work trying to confirm that this is the case. In this article they argue that evolution in passive galaxy sizes is not from increasing the size of already passive objects in situ (as in the merger scenario described above), but instead that new, larger galaxies are being added to the passive population.

    So, in this framework, it might be interesting to look at the sizes of quenched objects relative to their controls. If the picture discussed by the COSMOS paper is correct, then my naive expectation would be for the quenched galaxies to be larger on average than their controls at fixed mass.

    Posted

  • JeanTate by JeanTate in response to jtmendel's comment.

    It's still early days in the data analysis, but in Quench: Sample vs Control, what's the same, what's different, pretty much the opposite trend is what you can see.

    For example, QS and QC are very closely matched in Log_mass (12 equal-sized log_mass bins):

    enter image description here

    Yet the QS galaxies are smaller, with one (maybe two) marginal exception(s):

    enter image description here

    Posted

  • JeanTate by JeanTate

    And here's that same plot, with Petro_R50 expressed in kpc (i.e. physical size, not apparent on-the-sky size):

    enter image description here

    QS galaxies are - with only one (marginal) exception - smaller than their QC counterparts.

    "Planck cosmology": H0 = 67.3 km/s/Mpc, ΩM = 0.315, and ΩΛ = 0.685

    Posted