U band filter swap (FBS 1.3) - SAVING IN PROGRESS SO I CAN WRITE SOMETHING ELSE (IN CONSTRUCTION)

run-evaluation
Tags: #<Tag:0x00007f61a42da4d0>

(Lynne Jones) #1

This post follows up on the evaluation of when to swap the u-band filter that was started with FBS 1.2 - please read the FBS 1.2 u-band swap post for more background on the issues.

The currently identified relevant points are:

  • do we use u band under conditions that are appropriate for using u band?
  • do we achieve the desired number of visits and coadded depth in u band (and other bands) across the sky in WFD? The SRD defines this as u =26.1, equivalent to about 56 visits under ideal conditions.
  • do we achieve the required depth in u band in DD sequences? The AGN DD white paper (Brandt et al) request a depth of u =28.3 in each DD field, equivalent to approximately 2700 visits under ideal conditions.
  • by clustering u band near new moon, can we improve timing for sequences of grizy observations (as requested in the DESC WFD and DD papers)?
  • if we cluster u band near new moon, do we lose sensitivity for transient detection and identification? This is also influenced by how the u band visits are obtained - observing u band in pairs of visits with g and/or r band is beneficial, particularly for TDE detection.

The previous FBS 1.2 simulations investigated a limited set of filter-loading times, at 15%, 30%, and 60% lunar illumination, corresponding to +/- 3.5 (thus, 3 or 4), 5, and 8 days around new moon.
The results indicated that we could restrict the u band availability even more tightly around new moon and still achieve the needed number of visits and depth, so in this set of runs we also investigated filter load times of 3%, 4%, 5%, 7% and 10% (as well as 15%, 30%, and 60%). The FBS 1.2 runs also showed that the DD sequences were being too tightly constrained in u band, so these have been reconfigured to allow shorter inter-night gaps and to add more visits per sequence in u band.

FBS 1.3 u-band swap
Moon Illumination (%) 3 4 5 7 10 15 30 60
Days (+/- new moon) 1.5 1.75 2 2.5 3 3.5 5 8

image

Do we use u band under conditions that are appropriate for using u band?

FBS 1.2 results indicated that yes, this was true - even if the u band was available outside of new moon, it was only used when the sky brightness was good for u band observations. This continues to be true with the new simulations. The median u band single-image depth is consistent across a wide range of u band availability windows, with some slight benefits seen if it is available for a longer period, presumably because there is a wider range of options for obtaining u band visits.

image

Do we achieve the desired number of visits and coadded depth in u band (and other bands) across the sky in WFD?

The SRD defines this as u =26.1, equivalent to about 56 visits under ideal conditions. The FBS 1.2 runs indicated that a filter swap at 15% left the sky somewhat patchy in u band, but that there was more time available - so with a slightly higher weight on u band visits, we should be able to reach the desired numbers.
need numbers over WFD footprint