Observing Constraints and Classification Rules

General Observing Constraints

Every requested observation has multiple observing constraints. Typical observing constraints are:

  • the allowable brightest lunar phase
  • the allowable smallest moon-to-object angular separation
  • the allowable maximum airmass
  • the allowable maximum image size (i.e. FWHM at observed wavelength, 'image quality' )
  • the allowable sky transparency
  • for CRIRES, NACO and SINFONI, the Strehl ratio on the reference star (as applicable).
  • for instruments observing in the mid-IR (CRIRES and VISIR), the allowable maximum Precipitable Water Vapour (PWV)
  • the allowable twilight constraint that defines the earliest time in minutes with respect to the end of the astronomical twilight when the execution of the OB can be started (see the note below).
  • the allowable absolute time window for the start of the observation (i.e. for time critical events, multi-epoch monitoring)
  • the allowable local sidereal time range for the entire observation (e.g. for ADI observation)
  • for VLTI instruments, the availability of the desired baseline

The Observing Constraints are specified by the user at Phase 2 for each Observation Block. Since the execution conditions required by each programme are an important ingredient in the process of building up the Long Term Schedule of an observing semester, and thus determines which programmes can or cannot be scheduled, users are not allowed to specify at Phase 2 constraints that are more strict than those specified in the original proposal. Users can however relax the constraints during the submission of their Phase 2 material. The values in the OB constraint sets that are selected (and approved) during Phase 2 preparation (and review) cannot be changed later during the observing period.

Note about the twilight constraint: this observing constraint has been introduced to allow specifying start of observation with respect to the start of the night: e.g. to delay start of observations for faint targets until the sky gets darker, or allow starting observations for very bright targets during the twilight. The original motivation for this constraint is related to sky brightness in near-IR that is affected by OH lines excitation, and is not affected by other constraints (e.g. moon distance/phase). It does not apply to astronomical twilight at the end of the night (i.e. sunrise). 

General Classification Rules

Quality Control of OBs executed in Service Mode will be based on the specified constraints in the OB for airmass, atmospheric transparency, image quality/seeing, moon constraints, twilight constraint, as well as Strehl ratio for Adaptive Optics mode observations. If all constraints are fullfilled the OB will get assigned Quality Control grade "A", while the "B" quality control is assigned if some constraint is up to 10% violated. The observations with quality control grades A or B are completed, while those with quality control grade "C" (out of constraints) will be re-scheduled and may be repeated. In exceptional cases an OB may get status completed with quality grade "D", meaning that it is taken out of constraints but will not be repeated.

Note: for most instruments the image quality constraint as defined in the OB is judged against the full width at half maximum (FWHM) of a point source in the resulting image (or spectral image). For the instruments where the image quality cannot be directly measured (AO, VLTI, fibre instrument), it is either not used for classification or is obtained from the wavefront sensor of the active optics of the telescope.

Special Note for UT4 OB Classification Rules

Ellipticity was detected in some HAWK-I, MUSE and SINFONI observations from 07 May 2017 onwards when pointing away from the wind. The problem is under investigation and not yet understood.  In the interrim there is an additional criterion imposed during OB classification, related to elongation, defined as 100*(1-B/A)%, where A and B are the FWHM on the major and minor axes, respectively.

  • For HAWK-I:
    • A. If elongation < 10% for most stars
    • B. If 10% < elongation < 20% for most stars
    • C. If 20% < elongation for most stars 
  • For MUSE:
    • If there are stellar objects in the reconstructed cube FoV, adopt HAWK criteria.
    • If there are no stellar objects in the reconstructed cube FoV, use the SGS (slow guidance sensor) with criteria as above, but relaxed to 15% and 25% to account for the SGS distortions
    • If there are no stellar objects in the FoV or SGS the classification is based only on the average FWHM on the auto-guider.
  • For SINFONI:
    • For LGS/NGS no special ellipticity criteria are applied.
    • For NoAO the HAWK-I criteria are adopted only if
      • the target is a point source
      • the FWHM can be reliably measured (>100 ADU peak counts)
      • the PSF is resolved (FWHM > 4 pixels)

Additional Observing Constraints for VIRCAM

  • It is recommended to specify the weakest possible Constraint Set values.
  • The minimum image quality for VIRCAM is 0.6 arcsec, due to instrumental PSF of 0.51 arcsec.
  • The recommended airmass is up to ~1.7 (if applicable) due to implemented short term scheduling optimization algorithm.
  • The TWILIGHT constraint can be used to define the earliest time with respect to the end of the astronomical twilight when the execution of the OB can be started. While the relation between the time difference from the evening twilight end and sun elevation varies during the year, for Paranal due to its low latitude this difference is small.Therefore the constraint is given in minutes as a difference in time with respect to the end of astronomical twilight (i.e. the time when the solar elevation is -18 degrees). The default value of twilight constraint is 0. A negative number means that it is allowed to start the observation before the end of the astronomical twilight, a positive number means start the observation at least that many minutes after the end of the astronomical twilight.
    The recommended values for the twilight constraint for deep exposures are given in the VISTA/VIRCAM User's Manual.

Additional Classification Rules for VIRCAM

The following QC rules are implemented for all programmes. They are based on the experience gathered in the first years of VISTA operations, aiming to avoid repeating several times the same OB due to variable seeing during the observation, that affects only a small fraction of images in a tile:

Image quality condition classification rules:

  • C QC grade and status Executed (must be repeated):
    • Average image quality (IQ) over one tile is >10% out of constraint
    • At least one pawprint IQ is >20% out of constraint
  • B QC grade and status Completed:
    • At least one pawprint IQ is between 0% and 20% out of constraints _and_ average IQ is within 10% of the constraints
    • The average image quality is between 0% and 10% out of constraints
  • A QC grade and status Completed:
    • The average IQ _and_ every single pawprint IQ is within constraint

Image ellipticity classification rule:

  • C QC grade and status Executed (must be repeated):
    • there are >10 chips over a full tile with ellipticity >0.2
  • B QC grade and status Completed:
    • there are >4 and <10 chips over a full tile with ellipticity >0.2
    • or ellipticity up to 0.25 if FWHM*(1+ell) is ≤ 0.7
  • A QC grade and status Completed:
    • - there are <=4 chips over a full tile with ellipticity >0.2

OB classification is done on Paranal by the observers and is based on on-the-fly pipeline that reduces all the data as they are observed. It should be noticed that this data reduction uses master calibration files that are updated regularly, but may not be the best ones for a particular observation.

 

Special Classification Rules for VISTA Public Survey Programmes

Following the VISTA Public Survey Panel recommendation to introduce measures that minimize the number of repeated observations for the Public Survey programmes, the following specific changes in the quality control rules and OB classification were implemented starting from April 1, 2013:

  1. Surveys with multi-filter OBs (VIKING and VHS)
    If within the same OB, at least one filter has all 6 pawprints completed within constraints (QC grade A or B), and the rest is observed out of constraints, then the OB will get QC0 grade “D” and status (C)ompleted.

  2. Surveys with several concatenated OBs (VVVX) (as well as it was for VVV; VVV was completed in Oct 2015)
    If at least two OBs in a concatenation have all 6 pawprints classified A or B, and the rest of the concatenated OBs fail (out of constraints), the whole concatenation will be classified as (C)ompleted. The OBs that failed will get QC grade “D” instead of “C”, and the OBs that were done within constraints will be classified “A” or “B”.

  3. Surveys with only one filter per OB with no concatenations (Ultra-VISTA, VIDEO, VMC)
    No change with respect to the above QC0 rules that are applicable for all VIRCAM observations.

The responsibility to verify which fraction of the OBs classified "D" are good enough for the science goals of the survey, and which need to be repeated is the responsibility of the survey teams. Following the inspection of the observations, the survey team will determine the fraction of the time corresponding to OBs classified “D” that must be repeated. Within the report for the OPC, delivered every 6 months, the survey PI should request the corresponding "compensation time", that will be then added to the total survey allocation in a future period(s).

 

Instrument selector

On this page: