search

absey22 / SpectroZ

Spec1d/2d Pipelines
0 stars 0 forks source link

Left-out exposures? #15

Open anjavdl opened 7 years ago

anjavdl commented 7 years ago

In the plan files, some exposures are commented out, e.g.

# Plan file auto-generated by deimos_planfile.pro Tue Jan 31 21:49:50 2017
# Grating: 600ZD       Grangle:      -8.15040
MASK: M0018L1
#RAWDATADIR: /home/aseymour/DEIMOSraw
polyflag   - use polyflag for fitting lambda
FLATNAME: d1030_0027.fits
FLATNAME: d1030_0028.fits
FLATNAME: d1030_0029.fits
ARCNAME: d1030_0026.fits
SCIENCENAME: d1030_0079.fits
# SCIENCENAME: d1030_0080.fits
# SCIENCENAME: d1030_0081.fits
SCIENCENAME: d1030_0082.fits
# SCIENCENAME: d1030_0083.fits
SCIENCENAME: d1030_0084.fits
LINELIST: lamp_NIST_blue.dat

How did you decide which ones to process, and which ones to leave out? Or does the pipeline do so automagically?

absey22 commented 7 years ago

Automagically :grin: That choice came from what the written log sheets. At least for some masks, It was usually marked which were too cloudy, and so I made sure to remove them from the reduction. Looking back at this one, I am confused as to why I commented those out. Maybe a mistake since the logsheet contradicts these choices. However, I did use the header itself as a reference as well. (The routine that produces this plan file will not comment out objects, only find the proper exposures in the data dir and append them to the file.)

anjavdl commented 7 years ago

Is there a record of which ones were marked worse, and a quantification of that assessment? The most useful number would be a relative extinction measured from the spectra of the alignment stars.

absey22 commented 7 years ago

A quantification from the headers? Or from me. I did not do quantitative calculations like that, other than check that the exposures were the same length.

anjavdl commented 7 years ago

How did the pipeline assess which exposures are worse than the others? Does it give some kind of number?

absey22 commented 7 years ago

The pipeline only deal with removing data on the slitlet basis, this info should be found in each verbose_out.log in a mask directory. However, only I had the power to choose which exposures to plug in. (i.e. the plan files)

anjavdl commented 7 years ago

How did you choose which exposures to use? In the example above, why are d1030_0080, 81, 83 commented out?

absey22 commented 7 years ago

A combination of blinking the masks in ds9 alongside the logsheet comments. I may have been too harsh in choosing, I was going by the metric

it takes roughly 6 hours to reduce a single DEIMOS mask with 3 science frames and ~125 slitlets.