search

desihub / redrock

Redshift fitting for spectroperfectionism
BSD 3-Clause "New" or "Revised" License
21 stars 13 forks source link

set max_velo_diff to 100 km/s for stars #300

Closed moustakas closed 2 months ago

moustakas commented 2 months ago

Simple fix for #277.

Use 100 km/s as the "exclusion" zone for identifying "good minima" in the redshift scans for stars, versus 1000 km/s (which remains the same for galaxy and quasar scans).

With this branch, using the example object discussed in #276 and #277, we now get the following minima with the PCA/Jura templates:

     targetid               z                     zerr          zwarn        chi2        ... spectype subtype ncoeff znum     deltachi2
----------------- ---------------------- ---------------------- ----- ------------------ ... -------- ------- ------ ---- ------------------
39627688712865504    0.16364439776838893 1.0766243480836462e-05     0  19214.88787150383 ...   GALAXY             10    0  27842.03577605635
39627688712865504    0.17690234647531164  2.692106748341094e-05     0  47056.92364756018 ...   GALAXY             10    1  6043.541360843927
39627688712865504    0.18493315451662629  3.556348128910181e-05     0 53100.465008404106 ...   GALAXY             10    2  9627.262869153048
39627688712865504  0.0015657665423507877 1.1923063512031652e-05     0 62727.727877557154 ...     STAR       K      5    3  21.07149459215725
39627688712865504  0.0007735323930400607 1.0333191156667991e-05     0  62748.79937214931 ...     STAR       K      5    4 49417.100658225216
39627688712865504      1.039699278826281 2.6425853489245436e-05     0 112165.90003037453 ...      QSO     LOZ      4    5  2591.760555922985
39627688712865504     0.5302398868198309  3.203875319557626e-05     0 114757.66058629751 ...      QSO     LOZ      4    6  468.9983800649643
39627688712865504     0.5193319912431936  2.671646591573515e-05     0 115226.65896636248 ...      QSO     LOZ      4    7 15535.551420591131
39627688712865504 -0.0014325498503955333  6.594297818415482e-06     0 130762.21038695361 ...     STAR       M      5    8                0.0

And the following minima with an early version of the NMF templates:

     targetid               z                     zerr          zwarn        chi2        ... spectype subtype ncoeff znum     deltachi2
----------------- ---------------------- ---------------------- ----- ------------------ ... -------- ------- ------ ---- ------------------
39627688712865504  0.0015657665423507877 1.1923063512031652e-05     0 62727.727877557154 ...     STAR       K      5    0  21.07149459215725
39627688712865504  0.0007735323930400607 1.0333191156667991e-05     0  62748.79937214931 ...     STAR       K      5    1  3646.557508873142
39627688712865504     0.2592844076863871 1.8832468985475416e-05     0  66395.35688102245 ...   GALAXY             10    2 1659.1592923998833
39627688712865504    0.24543019059875015 2.5085768138572075e-05     0  68054.51617342234 ...   GALAXY             10    3 1106.9018910750747
39627688712865504    0.16378398397398541 2.0080150217357837e-05     0  69161.41806449741 ...   GALAXY             10    4 43004.481965877116
39627688712865504      1.039699278826281 2.6425853489245436e-05     0 112165.90003037453 ...      QSO     LOZ      4    5  2279.647886157036
39627688712865504     1.7249275860747917 0.00011775404198232079     0 114445.54791653156 ...      QSO     HIZ      4    6 312.11266976594925
39627688712865504     0.5302398868198309  3.203875319557626e-05     0 114757.66058629751 ...      QSO     LOZ      4    7 16004.549800656096
39627688712865504 -0.0014325498503955333  6.594297818415482e-06     0 130762.21038695361 ...     STAR       M      5    8                0.0

The main point here is that the best-fitting STAR redshift (with either set of templates) is now z=0.0015657 vs z=-0.001475 previously (i.e., with main). (Also note the much smaller delta-chi2 value between the STAR minima.)

In @sbailey's beautifully labeled graphic, the new redshift corresponds to the better / best "first" (i.e., non-bad) redshift minimum: image

So I think this PR is non-controversial and should be good-to-go.

coveralls commented 2 months ago

Coverage Status

coverage: 39.505% (-0.02%) from 39.528% when pulling 19c7e84c44877e204daaa556307c1578e1c92a5d on max-velo-diff-star into 04653df87a123c920446bbf4380f68bfdf190457 on main.