fix gray time definition code

[sbailey]

There is a bug in the implementation of the definition of gray time. Below is a plot of moon altitude vs. moon illumination fraction from a surveysim run for dark (blue) bright (red) and gray (black) observations. The upper panel is the assignments from obslist_all.fits; bottom panel is what they should be based upon the DESI-0311 site alteratives definition:

• dark = moon down
• gray = moon up and illum<0.6 and illum*alt < 30 deg
• bright = everything else

Note: there are no illum>0.85 points on these plots because of the 1 week engineering/other downtime around full moon.

I haven't tracked down if this definition code is in desisurvey or desisim, but I'm filing the ticket against desisurvey since that is where it should be :)

Guess: a radians vs. degrees bug.

[sbailey]

Update: the current code uses the alternate gray time definition from DESI-0333 of illum<0.2 or illum*alt<12 deg. It also allows gray tiles to be observed during dark time if all dark tiles for that LST have already been completed, which is reasonable. i.e. I'm not sure if there actually is a bug here, or if it is just using a different gray time definition (we should switch to baseline definition anyway).

[mlandriau]

This was discussed in the thread desi-data 2455. I left it as it was as it is more conservative to use stricter criterium. Martin

…

On Mon, Dec 19, 2016 at 4:17 PM, sbailey ***@***.***> wrote: Update: the current code uses the alternate gray time definition from DESI-0333 <https://desi.lbl.gov/DocDB/cgi-bin/private/ShowDocument?docid=333> of illum<0.2 *or* illum*alt<12 deg. It also allows gray tiles to be observed during dark time if all dark tiles for that LST have already been completed, which is reasonable. i.e. I'm not sure if there actually is a bug here, or if it is just using a different gray time definition (we should switch to baseline definition anyway). — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#13 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ASznwi4PfoMm7fAJYwgMFMH01tyfNcj9ks5rJx6egaJpZM4LRJjf> .

-- Martin Landriau Lawrence Berkeley National Laboratory 1 Cyclotron Road Mailstop 50R5008 Berkeley, CA 94720 USA -- Tel: +1 510 486 7934 Fax: +1 510 486 6738