Breakout Session on Deep-Drilling Fields and Other Ancillary Time Projects

At the 2016 Project & Community Workshop, it would likely be good to have a breakout session on the Deep-Drilling Fields (DDFs) and other ancillary time projects (i.e., all projects that are outside the main Wide-Fast-Deep survey). All together, these projects are expected to utilize about 10-15% of the LSST time. The breakout presentations/discussions would aim to, e.g., help the Project prioritize task and resource allocation. Some possible topics for presentation/discussion could include the following:

  • Current status and future plans for implementing the DDF white-paper ideas in OpSim. Related plans to document the results and make them public for community feedback.

  • Needed metrics to assess observational success for the DDFs.

  • Cadence and other observational planning for the four currently selected DDFs (ELAIS-S1, XMM-LSS, E-CDF-S, COSMOS); see

  • Additional ideas for new DDFs arising since the original 2011 white papers, and the process for incorporating these ideas.

  • Ongoing efforts to obtain sensitive multiwavelength coverage of the DDFs, especially with limited-lifetime facilities (e.g., Spitzer and X-ray missions).

  • Other ancillary projects including the Northern ecliptic spur, low Galactic latitudes, southern Celestial pole.

  • The mechanism for making good final decisions about the DDFs and other ancillary projects.

Please share your thoughts about possibilities for making such a breakout session successful and useful!

I very much welcome this session. In addition to the topics suggested, it would be very useful if an initial exploration could be made of what Data Management processing would be expected and/or desirable for deep drilling fields (or other “ancillary time projects”).

It is important to understand that no particular requirements in this area are currently “on file”. Please see Section 7 of the Data Products Definition Document, LSE-163, for a discussion of what constraints are understood to apply. Note that it is easy to envision that certain desirable processing schemes might produce additional, presently unapproved, costs to the project.

In particular, it would be good to consider the Level 1 (quasi-real-time) processing to be performed: whether transient alert detection would be run, and, if so, how. A simple choice that would not require significant additional effort from DM would be to treat deep drilling sequences, for purposes of Level 1, as sequences of standard visits that happen to be to the same field - i.e., to generate alerts on every pair of (15-second) exposures in the sequence. Variations on this, such as co-adding all the exposures in a filter in a sequence and performing image differencing against a special deep template, would require additional effort from DM.

Similar questions arise for the Level 2 (annual Data Release) processing to be performed. If the field in question is also part of the main survey, it would be good to understand whether the design would be to produce two sets of catalogs for the field: one at the “normal” depth and a separate one at full depth. This would require planning ahead to ensure that the system can distinguish the “normal” subset of images for the field. The baseline Level 2 design foresees performing forced photometry on every visit for every object detected in the full-survey coadds; for deep drilling, would this also be the case? Do the full-survey coadds for deep fields require special (e.g., crowded-field) processing? Would forced photometry be performed on all fields for both the main-survey-depth object catalog and the full-deep-drilling-depth object catalog? Would it apply only to exposure pairs from the deep stacks, or would one also, say, perform forced photometry on a coadd of each night’s worth of data for a field?

Finally, it would be good to know if deep drilling fields and their processing would yield any novel requirements on the Science User Interface and Tools (SUIT).

On a separate note, within the topics already suggested in the original post, I would suggest a particular focus on the question of what the filter-change requirements for deep drilling operation might be. Understanding the per-night and lifetime filter change counts is of important to the camera design, construction, and maintenance plan.


A breakout on the Deep Drilling Fields might be a good opportunity to discuss coordinating deep fields with representatives from WFIRST. The WFIRST Deep Fields working group and Formulation Science Working Group (I serve on both) have certainly raised the possibility of leveraging LSST Deep Drilling Fields, but exactly how the observations might be coordinated or organized remains uncertain. The WFIRST deep fields have certain constraints that could prevent their placement in any of the previously selected LSST DDFs, and they too will being to organize other ancillary observations.

If this kind of coordination is of interest, please contact me (the FSWG meets in person Jun 13-15, so now is a good time to start a discussion).

Thanks for these good points and suggestions, Gregory and Brant. All these points would indeed seem good for discussion at a breakout session. Hopefully we will be able to get sufficient time to cover the many needed topics well.

Brant, please do bring this up at the WFIRST FSWG and gather relevant information for LSST.

@brant thanks for posting on this. We discussed DDFs at our last Project Science Team meeting, with considerations of the timing of finalizing the DDFs as one topic of discussion. We floated the idea of having a call to the community for DDF input circulated around August or early Fall, with an eye on early 2017 as the timescale for getting input. I would like to have a strawman set of selection criteria to discuss at LSST2016.

One thing on my HW list is to learn more about the timing and selection process for the WFIRST and Euclid DDFs. @brant - What is the timing of the WFIRST selection process? Now-Ish? I need to get more plugged in there.

@nielbrandt, may I assume that you’d be willing to lead a DDF + ancillary time projects breakout and take an abstract from among your bullet points listed about?

Yes, happy to do what I can to help here. Note that I will not be there on Friday, so need to avoid having the breakout on that day.

Hi @bethwillman and @nielbrandt,

I will raise the Deep Fields issue at the WFIRST FSWG (we’re meeting currently). There is a lot of interest in coordinating at least some of the LSST DDFs (if there will be as many as originally planned) with the WFIRST Deep Field(s) – the number of these fields has not been determined, but will be of order 1-5 and will likely be well-matched to the LSST single-pointing FOV (meaning smaller or not much larger). These details are all being discussed by the relevant WFIRST working groups, but now is definitely the right time to begin coordinating. I’m happy to assist the communication between the relevant groups if that’s helpful.

I will be at the meeting Wednesday and Thursday, and would like to attend this breakout if possible.


1 Like

Thanks, @nielbrandt!

Hi All,

About 5 of us have just joined the LSST Project from the International Centre for Radio Astronomy Research (ICRAR). I’m very much looking forward to coming to the meeting to get up to speed. What I hope I can add is awareness of various contiguous spectroscopic campaigns I’m involved in, as well as radio campaigns from ASKAP and SKA. An awful lot of science can be facilitated by combining deep optical, deep near-IR and deep spectroscopic surveys — in particular group science which means directly tracing the dark matter to low halo mass limits and on a halo by halo basis.

I’m currently leading the GAMA survey, PI of the upcoming WAVES survey on 4MOST, and drafted the high-z science case for the Maunukea Spectroscopic Explorer.

I think it might be very exciting to discuss the prospect of an LSST/WFIRST medium-deep drilling field over ~50sq deg with extensive high density redshift coverage (i.e., 10,000/sq deg to r<22mag), and I would be keen to contribute to any break-out discussions on the extragalactic roadmap, the deep drilling fields, the multi-wavelength interface from X-ray to radio, as well as automated bulge-disc decomposition, optimal multi-wavelength photometry, supporting Numerical Simulations, and of course spectroscopic follow-up/pre-survey.

I hope to be at the meeting from Tuesday evening until Saturday morning.