search

LSSTDESC / DC2_visitList

Code to create the list of visits to simulate for DC2.
1 stars 0 forks source link

WFD visits to field 1427? #5

Open humnaawan opened 6 years ago

humnaawan commented 6 years ago

There's some uncertainty in terms of what happens to the visits that are "duplicates", i.e. are classified as WFD and uDD, i.e., WFD visits considered here for field 1427. @rbiswas4 mentioned that they should be classified as WFD.

humnaawan commented 6 years ago

Okay, since the all-band lists are generated using MAF, the list-generation code does not impose any checks but rely on MAF handling the duplicated visits. I am thinking that this problem might only be relevant for Run 1.1 since the per-band lists did not use MAF, but opsimsummary (linked in the Run 1.1 notebook).

Based on the notebook I just pushed, the WFD visits to field 1427 are not in the uDD list and there are no duplicates across WFD and DD lists.

@rbiswas4 what do you think?

rbiswas4 commented 6 years ago

Both MAF and opsimsummary know enough to remove duplicates. The question is if a visit is both in DDF and WFD do we do this as part of our Main Survey or our modified DDF. OpSimSummary treats this as part of our main survey (based on decisions on design) and thus keeps the WFD row. I don't know what choice was made in MAF if at all there is a specific choice made.

humnaawan commented 6 years ago

Hm I think MAF treats the duplicates as WFD ones since there are WFD visits to field ID 1427. If it was treating them as DD visits, we would not see any WFD visits to field 1427, no?