Galaxy Zoo Starburst Talk

Recommendations & Advice Coming Soon

  • trouille by trouille scientist, moderator, admin

    I'll be posting soon with suggestions, recommendations, and advice for carrying out data analysis, using tools.zooniverse.org and discussions within Talk.

    Posted

  • JeanTate by JeanTate

    That will be most welcome! 😃

    Will you be waiting until the classification phase is almost (say ~90+%) done, before posting those?

    Posted

  • trouille by trouille scientist, moderator, admin

    In order to hit the ground running, it'll be great for people to start playing with the data that's available now (i.e., magnitudes, redshift, mass, emission line strengths, etc.) to get a feel for the sample of post-quenched galaxies and their properties. One we're at 100% complete with the classifications, we'll upload the results into Quench Tools.

    I've just created a post with a framework for the data analysis phase. Check it out and add your suggestions!

    Posted

  • mlpeck by mlpeck

    Will you have a downloadable catalog of program objects and (preferably also) controls once classifications are complete? An ASCII file in csv or some other straightforward format would be ideal.

    I'm proficient in R ([http://r-project.org]) and have some tools of my own that might be of some use. I realize the tools at tools.zooniverse.org are still in a preliminary state, but I didn't immediately find them as usable as I'd like.

    Thanks!

    Posted

  • trouille by trouille scientist, moderator, admin

    Yes! You can download the data now and we'll update the tables with the classification results once those are complete. Check out this post.

    Great to hear that you'll be analyzing the data with R. Definitely keep the group updated on your progress. I'm sure each of our different approaches will inspire others to look at the data in new ways. And if anyone else is working with R, you might consider starting a discussion space to share tips and tricks with each other.

    And yes, tools.zooniverse.org is a work in progress. We'll be learning a lot from you all through this Quench project of what functionality would be helpful. As you work with the data, if you could touch base now and again with what you think would be helpful to add to Quench Tools, that would be extremely useful to us.

    Posted

  • mlpeck by mlpeck

    Thanks, that was exactly what I was hoping you'd publish.

    A couple things I've noticed so far (this has nothing to do with R):

    -- The emission line flux errors are identical to the flux values for all objects. Same for NaD.

    -- The spectroscopic object with plateid/mjd/fiberid = 2812/54633/303 apparently doesn't exist. Plate 2812 was exposed on mjd 54639, and in DR9 the spectrum with fiberid 303 is classified as a QSO with z = 0.96 +- 0.90.

    Back to classifying...

    Posted

  • trouille by trouille scientist, moderator, admin

    This is why collaborations are so useful! I didn't realize that I'd inputted the same piece of the pie into 2 lines of my code.

    This is the relevant piece of my code, pulling out the information from a much larger catalog. You can see that I've given both OII_flux_all and OII_flux_all_err (for example) the same piece of the catalog.

    I've updated my code, rerun it, and sent the new table to Ed, our most wonderful Quench Tools software genius.

    ;LINES

    OII_flux_all = cat1.OII_3726_flux

    OIII_flux_all = cat1.OIII_5007_flux

    Hbeta_flux_all = cat1.H_beta_flux

    Halpha_flux_all = cat1.H_alpha_flux

    NII_flux_all = cat1.NII_6584_flux

    NaD_abs_flux_all = cat1.NA_D_ABS

    OII_flux_err_all = cat1.OII_3726_flux

    OIII_flux_err_all = cat1.OIII_5007_flux

    Hbeta_flux_err_all = cat1.H_beta_flux

    Halpha_flux_err_all = cat1.H_alpha_flux

    NII_flux_err_all = cat1.NII_6584_flux

    NaD_abs_flux_err_all = cat1.NA_D_ABS

    Mlpeck -- Thanks for posting this!

    Posted

  • klmasters by klmasters scientist

    So one thing I'd like to see now the morphologies are in the sample, are plots of the u-r colour versus dn4000 for the quench and control sample split by morphology. I think that might reveal some interesting things. If I were directly a summer student to do this project that's what I'd be asking them to do right now. 😃

    Posted