October 2019 Update - FBS 1.3 Runs

Tags: #<Tag:0x00007f61a3c4bd58>

(Lynne Jones) #1

An OCTOBER 2019 set of extended and updated runs, the “FBS 1.3” set of simulations. This set of runs has a bug fix regarding the limiting magnitude for most runs (calculating u band limits for 1x30s visits instead of 2x15s visits where relevant), changes regarding DD sequence execution, a general rotational dithering plan, some updates regarding the filter load time (restricting the u band filter availability slightly closer to new moon), and many new runs - including simulations adding cadences for the galactic bulge, an NEO survey during twilight, and a WFD footprint explicitly limited by dust extinction.

There are a large number of simulations in this release, even after dropping some of the previous simulations which seemed the least promising. However, note that the most useful previous runs have been recreated with a consistent set of software and settings (and bug fixes), even though names of the runs may not have changed - thus please be aware of which version of these runs you are comparing (and generally try to compare runs in the same release). If a run that was particularly interesting to you was dropped, please let us know - we can recreate it again (and you can tell us what we were missing).

Version tracking
This set of runs are labelled as “FBS 1.3”, while the previous set (July 2019 and prior) will have version numbers of 1.2 or 1.1. The version information is preserved inside the output database, in the “info” table, and is clearly labelled in the online MAF tables.

The simulation output databases are available at https://lsst-web.ncsa.illinois.edu/sim-data/sims_featureScheduler_runs_1.3/: follow the directories in that location (mirrored in the descriptions below) to find specific runs. If you are familiar with the general python configuration scripts used to set up these surveys, you may find the configurations at https://github.com/lsst-sims/sims_featureScheduler_runs_1.3 useful – the READMEs in the github repo can be useful for understanding the output contents as well.
A limited variety of MAF outputs (and additional links to the databases) for these runs are also available at http://astro-lsst-01.astro.washington.edu:8081 – more outputs will come online here over the next few weeks. The MAF outputs are identified by the same run names as the DB links.

FBS 1.3 runs
We have several sets of runs to investigate specific topics.

  • baseline Variations on the baseline intra-night visits, using the standard (old) footprint and minisurveys. These compare 1x30s vs. 2x15s visits, as well as pairs in the same filter vs mixed filters. For other sets of simulations, we have taken the 1x30s + mixed pairs of filters configuration as the standard intranight cadence … however please take this as a working premise, but NOT an official constraint or choice. The final decision to choose 1x30s vs. 2x15s visits will not be made until on-sky observations with the survey camera are acquired, and 2x15s visits must be carried as the official baseline until that point.

  • presto These are a further variation on the intra-night cadence, but the pairs of visits are in non-adjacent filters.

    • presto Pairs of visits in each night: g-i, r-z, i-i, z-r, with u and y visits unpaired.
    • presto_third Add a third visit into each night, in the same filter as the first of the pair.
  • Filter_load The SAC recommended confining u band observations to within +/- 2 days of new moon, and understanding when to change the u band filter (for either z or y band) is useful. These sets of runs complete an (updated and extended) evaluation of the impact of changing the u band filter loading time. The time of filter swap is varied from 3% lunar illumination - 60% lunar illumination. 10% is the default carried to other simulations.
    Here is a thread evaluating the effects for FBS 1.2: this will be updated for FBS 1.3.

  • footprints Variations on the WFD footprint and overall survey footprint. For now, we’ve continued to take the ‘traditional WFD and minisurvey’ footprint to be the basis for most of the other experiments, but this may change in a future set of simulations.

    • baseline … equivalent to the baseline_v_1.3_10yrs above, and following the traditional baseline footprint for WFD
    • gp heavy continue traditional WFD coverage throughout the galactic plane (low galactic latitude regions + bulge)
    • bluer footprint This is a standard survey footprint, but with more observations in bluer bands (shifted from z and y bands).
    • big sky a footprint roughly matching that suggested by the “Big sky” whitepaper - extends WFD north (+12) and south (-72), limited by |b|<15 from low galactic latitude regions. This drops any concept of the North Ecliptic Spur (NES) or other minisurveys as separate designs.
    • big sky no uiy just like above, but no observations in u, i or y bands outside the WFD footprint.
    • big sky with dust cut The big sky concept, but defining the galactic plane cutout by a dust extinction cut, rather than a galactic latitude.
    • newA This is based on a similar concept to the big sky simulations, but takes more observations from WFD to extend coverage in the NES and galactic plane regions. This simulation runs WFD from -72 to +12 degrees declination.
    • newB Similar to newA, but moves observations from the galactic anti-center to the galactic center to increase coverage over the galactic plane. This simulation runs WFD from -72 to +12 degrees declination (a bugfix updated this from FBS 1.2).
    • no gp north This is a standard footprint, but without the small spur of the galactic plane that has traditionally stuck up north of the usual WFD footprint.
    • add magellanic clouds This is a standard footprint that adds extended (WFD style) coverage over the Magellanic clouds.
    • ‘stuck’ rolling This is a test simulation, an extreme case of what could possibly happen if there was a simple rolling cadence that always had bad weather in the even (or odd) years. Basically, this is an example simulation that should show that your metrics break when sky coverage is bad.
    • Here is a visualization of these footprints, including their filter coverage. (If github fails to load the notebook repeatedly, try it on nbviewer).
  • alt_sched These simulations attempt to come closer to recreating the alt scheduler outputs, without some of the drawbacks to the alt scheduler. The alt scheduler schedules without consideration of lunar phase or weather conditions, which we would like to account for. The benefits to the alt scheduler cadence include the more rapid inter-night revisit rate in given filters. In general, we’ve attempted to come closer to the alt scheduler by strongly weighting a region of sky in a particular night, defined by declination (to come closer to matching the per-night regions observed by the alt scheduler). In FBS 1.2, we had a series of weightings and filter swap times for this; here we’ve dropped to a single weighting scheme (closest to very_alt3), and used the standard u band filter swap time, but experimented with a ‘large’ and ‘small’ overall footprint. A footprint that looks more like the ‘standard’ footprint is included in the ‘roll_alt’ set below. These simulations illustrate what a basic North-South interleaving of observations gives you, over some simple footprints.

    • altLike A swath across the sky from Dec = -68 to Dec +7, with an alt scheduler like cadence.
    • altLarge A wider dec band footprint version of the alt scheduler cadence - Dec = -90 to + 20.
  • roll_alt These two simulations start to take the short-timescale internight cadence imposed by the alt scheduler (alternating N-S means fields tend to get observed every other night, or short multiples of this), but bring them back to the standard footprint.

    • alt_WFD Alternating N/S bands per night, on a standard footprint.
    • alt_roll A hybrid of the alt scheduler N/S nightly emphasis and a delayed rolling cadence, where bands of declination are emphasized in alternating years. Here, the footprint is essentially sliced in 4 parts, two in northern declination bands and two in southern declination bands; like in the 2-dec band delayed rolling cadences below, half of the sky is chosen to roll in each of the alternating years (with a season that corresponds to their RA range), but the ‘bands’ are actually one each of a N and S section, and these are further alternated from night to night.
  • alt_roll_dust Same as above, but with a dust-map based footprint instead of the older standard WFD footprint.

  • Rolling_cadence A range of variations on the rolling cadence. There is more to do here, but we need better metrics to distinguish between the effects of these difference cadences. The rolling cadences tested so far are all declination band-based. Then, in general, there are two categories: a ‘simple’ version where the cutoff for changing between dec bands is a single time for all points on the sky (so seasons at some points in the sky can be split), and a ‘delayed’ (previously, ‘modified’) version where the time for changing between dec bands depends on RA (so seasons are always contiguous). The delayed version requires waiting a little longer before starting and after ending the rolling period. For each, there are variations on the number of dec bands: 2, 3, 5, and 10 for the ‘simple’ version, and 2, 3 and 6 bands for the ‘delayed’ version. For the ‘delayed’ runs, there are also variations on the weight for the ‘background’ (non-rolling / non-emphasized) portions of the WFD, from 10-20%. For the ‘simple’ runs, the background WFD is carried at 20% of normal weight. In each run, the intranight visits are in mixed filter pairs. For now, we’ve taken a non-rolling cadence as the basis for other experiments, but this may change in a future set of simulations.

  • DDF For general simulations, we are using the ‘traditional’ non-dithered DD sequence, unless otherwise noted. In most simulations, the u band filter is swapped at 10% illumination - in the experiments below, the u band filter is swapped at varying illumination levels (identified by “illum_X” in the database filename).

    • DESC_DDF The DD’s in these simulations are based on the requested DESC DD sequence, but with varying u band filter swap times.
    • AGN_DDF The DD’s in these simulations are based on the requested AGN DD sequence, but with varying u band filter swap times.
    • Euclid DDF Adding a fifth field, in the location desired for Euclid ground-based observations, using traditional DD sequences.
  • bulge These experiments vary the number and strategy of observations within the galactic bulge region. The “background” survey footprint is the “big sky” footprint - i.e. an extended WFD region, but minimal coverage north or south of this region, or beyond a galactic latitude of +/-15 degrees unless otherwise modified in the simulation. The ‘low galactic latitude’ region in question as specified by the SAC is then |b| < 10 degrees, with a further specification of galactic ‘bulge’ |l| < 20 degrees and |b| < 10 degrees.

    • bulge big sky This is approximately the baseline ‘big sky’ footprint, with an enhancement throughout the low galactic latitude region to increase the number of visits to approximately 250.
    • bulge wfd This further increases the enhancement in the bulge region within the galactic low-latitude region.
    • bulge i heavy Similar to the bulge enhancement above, but shifts visits into i band in the bulge.
    • bulge cadence big sky This is the same overall footprint and number of visit target goal as the ‘bulge big sky’ survey above, but with additional constraints on the cadence of observations within the bulge region, tightening the season to 2.5 months and intra-night revisit time of < 2.5 days.
    • bulge cadence wfd This combines the WFD coverage in the bulge region with the cadence constraints described above (tighten season to 2.5 months and a revisit time < 2.5 days).
    • bulge cadence i-heavy Adds cadence constraints onto the i-heavy bulge footprint.
  • twilight_neo The previous twilight survey indicated that twilight time was not necessary to meet the SRD requirements; with the degraded weather added in FBS 1.2, this is no longer true. These experiments tested adding an NEO-specific survey during twilight.
    The NEO survey included fields with |ecliptic latitude| < 40 degrees and Dec < 30 degrees, then attempted to take 3 observations per pointing with about 3 minutes spacing, in one of riz filters, up to airmass of 2 and with a preference for pointing as near as possible toward the Sun.

    • mod_1 Repeat the neo twilight survey every night.
    • mod_2 Repeat the neo twilight survey every other night.
    • mod_3 Repeat the neo twilight survey every third night.
    • mod_4 Repeat the neo twilight survey every fourth night.
  • DCR Add high airmass observations in u and g to enhance estimates of colors/spectra via DCR. The number of high airmass observations per filter per season was varied in the different simulations.

    • nham 1 Add 1 high airmass visit in u and 1 in g each season.
    • nham 2 Add 2 high airmass visits per filter per season.
    • nham 3 Add 3 high airmass visits per filter per season.
  • templates Difference imaging will require a usable template image. This set of simulations experimented with adding weighting to parts of the sky which had not received at least 3 visits within the last ~300 days. The run parameters are similar, just with different weighting for this feature, the weight ranging from 0 to 10. The default, 6, has been carried out as a baseline for the other simulations. Note that we need better information from Data Management regarding what is required for a template!

    • weight_6 The new default, template weight = 6.
    • weight_0 Zero weighting on this template feature.
    • weight_10 Heavy weighting on the template generation.
    • There are more runs spanning the weight range of 0 to 10 in the directory.
  • WFD depth This set of runs experiments with weighting the WFD portion of the footprint to different depths. The survey footprint is the standard traditional footprint. The WFD % in each case represents an estimate of the requested number of visits in WFD; the fraction of WFD visits in the final result may vary slightly.

  • WFD only A series of runs containing only WFD visits. The WFD footprint here is the traditional WFD footprint. The runs mirror the baseline runs above - mixed filter pairs, non-mixed filter pairs, snaps and no snaps. The intent is to evaluate the overall efficiency of observing if we only covered the WFD footprint.

FBS 1.3 release notes

  • The rotator control that was tested in the FBS 1.2 simulation (rotator) has been incorporated into the defaults for the main survey. Rotator angles should be more uniformly distributed on the sky.
  • The DD field translational dithering that was tested in the FBS 1.2 DD runs is incorporated into the DD sequences by default, imprinting a 0.7 degree dithering pattern.
  • The template generation requirements tested in the ‘templates’ runs above were used as the default for the standard simulations (weight = 6) - driving approximately 3 visits per year per filter.
  • As in FBS 1.2, the mixed filter pairs and 1x30s visits are used as the defaults for the standard simulations (although 2x15s visits must remain the official baseline and will be tested for all final runs).
  • A bugfix in the per-visit point source five-sigma limiting depths was fixed in FBS 1.3 - a notebook documents the as-used scaling relationships; the result is that u band depths in FBS 1.3 runs should be more accurate and about 0.2 magnitudes deeper than in FBS 1.2 (where the depths were all estimated as 2x15s depths), and g band depths almost 0.1 mag deeper.
  • An additional bugfix to the DD sequences should let them execute more frequently during u band time, resulting in more desirable u band depths and intra-night revisit rates.
  • Planet avoidance is explicitly coded into the new simulations, instead of being provided implicitly via the sky brightness and masked regions.

Experiments not repeated in FBS 1.3
These are experiments we did not repeat in FBS 1.3 - the FBS 1.2 links are preserved here for work that may need to investigate these simulations. Where necessary (or as we get further information), we will rework these experiments in a later release.

  • ToO We will rework these again in a later release, but the initial results were that the impact of the ToOs depended extensively on where the ToOs were located and how they were triggered – further information on these aspects would be useful before diving into ToO experiments further. In the meantime, FBS 1.2 showed they could essentially be treated as a simple cost of time.
    (from FBS 1.2) – Experiments to add ToO sample observations, at a varying rate of triggers. In each, the ToO requested observations themselves are similar; a circular region of sky with a 15deg radius (so about 700 sq deg per alert) is covered in each of g, r and i bands 3 times (9 observations per location in the alert area) within 3 days of an alert being issued. The rate of alerts is varied in each simulation, from 1 per year to 100 per year. The takeaway here would include looking at the impact on other science, and how these impacts compare to other downtime.

  • Short exposures There were white paper requests for very short (1s or 5s) exposures over the entire sky in each filter. We ran simulations adding these mini-surveys, taking either 1s or 5s exposures 2 or 5 times per year. This will be repeated in a later release, but the impact was minimal.

  • Exposure Time Based on early feedback from the science collaborations, we had an early look at varying the exposure time to create visits with closer-to-uniform depth. We simply scaled the exposure time between 20s to 100s to maintain more uniform depth. This is not likely to be reworked into future releases - this was not an avenue we were directed to investigate by the SAC and may not be supported by Data Management.

SMWLV observing strategy task force
(Lynne Jones) #2

Note that there is an additional directory of platform runs - these are experiments with making the simulator code platform-independent.
They are not necessary for FBS 1.3 evaluations, although this will be incorporated into future FBS 1.4 runs.

(Lynne Jones) #3

And one further FYI:

@ehneilsen (Eric Neilsen) has been working on updating the seeing distribution used for the simulations. The initial indications are that this new distribution will include more ‘bad’ seeing than the current distribution, as well as stronger seasonal and yearly variations. This will most likely be incorporated in FBS 1.4, but has not been used here.

(Nathan Golovich) #4

Hi @ljones, great stuff.

I’m curious about throughputs for the atmosphere. I’m looking at the twilight NEO cadences, and the airmass rises above 2.5, which is the highest throughput file in the lsst/throughputs/atmos repository as far as I can tell.

Were higher airmass throughput files used for the these runs? If so, where are they?



(Lynne Jones) #5

Hi Nathan,
The five-sigma depths for each visit are not calculated ‘from scratch’ using the throughputs curves, but rather they’re calculated using a scaling law. There’s a short description of this scaling law in SMTN-002 (although now I notice that I need to update those Cm values and indeed most of this section of the smtn-002 document … it was based on the older OpSim code, but the calculation has remained basically the same - see the updated code) .

This means that we can also quickly calculate m5 values across the sky when choosing where to point, using the FWHMeff (also, separately, calculated at each point in the sky) and sky brightness (again separately calculated at each point in the sky). The relevant remaining part of the scaling law for changes in m5 related to the airmass of the observation is the
-k * (X - 1)
portion of the equation, where k is calculated per band, using the X=1 throughput curves.