Include stellar parameter info in Paper 1

[adrn]

Do we include stellar parameter info inferred from photometry in Paper 1 about the candidate sample? My gut says "yes" since it seems like it is fairly easy for @timothydmorton to run on a sample with the numbers we expect, but it could also move to the population inference paper. Any strong opinions?

[timothydmorton]

I'd like it to be part of Paper 1. Given the strong likelihood that we're not the only group doing this, I think this adds some additional good value to the candidate catalog paper.

[adrn]

I'm changing this from a question to a command and assigning you!

[timothydmorton]

@adrn do you have a list of pairs for me to fit (even if not final, it can get me going)?

[adrn]

@timothydmorton Will have a list for you by end of the day!

[timothydmorton]

Just had another thought here-- My default [Fe/H] prior for isochrones fitting that I've been using is based on the local [Fe/H] distribution from SDSS, which has very little power at low [Fe/H]. I suppose I should improve this by adding another underlying component that could allow for a low-level halo component? @davidwhogg, @adrn, do you have any particular suggestions for a better prior to use here? Or should I just add a low-level uniform component that extends to low-metallicity just so as not to totally reject the possibility of low [Fe/H]?

[adrn]

Well, I'm always partial to mixtures of Gaussians

[timothydmorton]

What would you suggest for the halo feh distribution, and the mixing ratio relative to the disk?

[adrn]

Gaussian mu,sigma = (-1.5,0.4) https://arxiv.org/abs/1211.7073

[timothydmorton]

Thanks! Approximately what fraction of local stars would you expect to belong to the halo population? According to this, the fraction is something less than 1 in 500... what would you suggest?

[timothydmorton]

Though come to think of it, perhaps this is something that the space motions from TGAS tell us better than anything else?

[adrn]

OK the new plan: we're going to do 3 tests sets for Paper 1, and @timothydmorton will then write a Paper 2 on the stellar parameters of all pairs. The 3 pairs are:

1. All pairs within some radius of the Pleiades
2. 1024 pairs outside of all known clusters / associations (or @smoh 's "mutually exclusive" samples)
3. 1024 random pairs of TGAS stars

[timothydmorton]

@adrn @smoh do we have lists of IDs for these test sets yet?

[timothydmorton]

And if we figured out the cross-matching issues, presumably we also have a standard matched photometry table to use?

[smoh]

@timothydmorton I did figure out how to retrieve cross-matches, so I'll put an update on that here today. About candidate list, sorry, I was focused on getting handle on the right false positive rates. I can make and send the list today -- but it maybe well the case we have to do it again.

[smoh]

So for Gaia team's cross matches, they have local copies of each external catalog, and you can do something like below

select top 100 tgas.source_id, tgas.parallax, tmass.j_m, wise.w1mpro from gaiadr1.tgas_source as tgas
join gaiadr1.allwise_best_neighbour
on tgas.source_id = gaiadr1.allwise_best_neighbour.source_id
join gaiadr1.allwise_original_valid as wise
on gaiadr1.allwise_best_neighbour.allwise_oid = wise.allwise_oid
join gaiadr1.tmass_best_neighbour
on tgas.source_id = gaiadr1.tmass_best_neighbour.source_id
join gaiadr1.tmass_original_valid as tmass
on gaiadr1.tmass_best_neighbour.tmass_oid = tmass.tmass_oid

in Gaia archive adql form.

[timothydmorton]

@smoh could you give me a hand with this query? I've uploaded a table of source_ids corresponding to the stars you sent me (user_tmorton.gwb_isochrone_unique_source_ids), and tried to drop this into a query to get the photometry I want to use all in a single table. But this returns me zero rows (no error though). What am I doing wrong?

SELECT tgas.source_id, tgas.parallax, tgas.parallax_error, 
    tgas.phot_g_mean_mag, tgas.phot_g_mean_flux, tgas.phot_g_mean_flux_error,
tmass.designation, tmass.j_m, tmass.j_msigcom, tmass.ks_m, tmass.ks_msigcom, 
wise.designation, wise.w1mpro, wise.w1mpro_error, 
                  wise.w2mpro, wise.w2mpro_error, 
                  wise.w3mpro, wise.w3mpro_error
FROM gaiadr1.tgas_source AS tgas
JOIN user_tmorton.gwb_isochrone_unique_source_ids
        ON tgas.source_id = user_tmorton.gwb_isochrone_unique_source_ids.col1
JOIN gaiadr1.allwise_best_neighbour
    ON tgas.source_id = gaiadr1.allwise_best_neighbour.source_id
JOIN gaiadr1.allwise_original_valid AS wise
    ON gaiadr1.allwise_best_neighbour.allwise_oid = wise.allwise_oid
JOIN gaiadr1.tmass_best_neighbour
    ON tgas.source_id = gaiadr1.tmass_best_neighbour.source_id
JOIN gaiadr1.tmass_original_valid AS tmass
    ON gaiadr1.tmass_best_neighbour.tmass_oid = tmass.tmass_oid

[smoh]

Hmm.. I just tried your snippet on 171 unique stars related to pleiades and it executed just fine, returning 160 rows, so I'm not sure what's wrong.

By the way:

• You can do [full table name] [alias]

JOIN user_tmorton.gwb_isochrone_unique_source_ids mytable
        ON tgas.source_id = mytable.col1

• It seems that for some of the stars not all photometry is available. If you want row-by-row matches with your table, I think you want to set main table as your table, and join other stuff by LEFT JOIN.

[smoh]

Since there was no error raised, I think you want to make sure your source id's are correct. (Note the thing I send you is indices in stacked_tgas, not gaia source ids).

[timothydmorton]

OK great thanks-- and yes, I just looked at my table (for the first time) and apparently I had accidentally written the TGAS indices rather than source_ids... I guess I have such low confidence with SQL that I'd assumed there must be a problem with the query! And thanks for the LEFT JOIN tip; that was going to be my next question.