Observing Checklist

There are so many things to remember while observing that it is easy to forget them. This is a list to help. Not all items may be relevant, but I hope it is reasonably comprehensive. The idea is that you be able to go down the list crossing off every item in a fairly mindless manner and at the end of it, be reasonably sure that all is OK.

Pre-observing run:

• Have you booked lodge/residencia/hotels?
• Passport + visa
• Tickets
• Money/credit card etc
• PATT time award letter or similar documentation.
• Finding charts and positions.

At the start of the run:

• What grating(s)?
• What central wavelength(s)?
• Do you need order blocking filters? [> 7000A, probably yes]
• What dichroic? [Multi-beam spectrographs only]
• Are the gratings in the right way round? Might have to ask the support astronomer to jog their memory as you probably won't be able to tell yourself unless you go out and watch them put them in; if the blaze is pointing wrong way you can lose 50% or more of the light.
• What binning and spatial coverage?
• Carry out tests of readout time.
• Decide on final CCD formats and readout speed. Record them; they may be monkeyed with during the day.
• Do you need a dekker? [e.g. on ISIS to cut out scattered light]
• Which arc lamps are best? How long exposures will be needed?
• Will you need to take flats for e.g. fringing? How long an exposure is needed?
• Check the alignment of the arc lines with CCD. Get someone to rotate if need be, but only if fairly far out as it can take a while to do a full setup. It is more important to align the arc lines with rows (or columns) than the object spectrum which gets tilted by dispersion anyway.
• Focus spectrographs [I normally do this with a moderately wide slit, say 2" to ensure that the profile is well-sampled during Hartmann tests]
• Check that bad columns don't wreck crucial regions of the spectrum. (Might need to tweak the central wavelength to avoid this.)
• Do you need to observe standards for flux, spectral type, radial velocity or telluric absorption line calibration?
• Record the collimator positions for the focus.

Before the start of each night:

• Check that the collimator positions, central wavelengths, CCD window formats, readout speeds and binning factors are all as you had them (you did record them didn't you?).
• Re-check the focus; it can drift or might have been moved by someone during the day. [Experts only.]
• Check any order blockers etc are in place.
• Ensure both Hartman shutters are open.
• Check CCD format and readout speed again.

During the night:

• Have you bracketed objects with arc exposures if need be?
• Are you keeping up with the parallactic angle?
• Have you taken flats for removal of fringing?
• If using large spatial pixels, have you adjusted object position to fall as far as possible into one pixel?
• If changing slit width at all, (e.g. for a flux standard), have you taken appropriate arcs when the slit is narrow?
• Are the counts/pixel/gain reasonably above the (readout noise)**2?
• If observing with two stars on the slit:
  • Did you check their alignment with a filter in place to reduce atmospheric dispersion?
  • Did you try to choose a comparison star East or West of target?
  • Have you taken a wide slit spectrum of them at some point?
• Beware pointing too close to the zenith on alt/az telescopes. Also check that the rotator will not hit any limits during the run on a given target.

Other calibrations:

• Sky frames e.g. at twilight, to calibrate response along slit.
• Bias frames. Should always do these. Take a load (~50) every night; no point adding any noise because you wanted to save a bit of disk space.
• Noise calibration [Experts only.]. Tungsten flats at a series of count levels to calibrate noise characteristics. Usually I do a series from very low levels, well below readout to as high as possible, stepping by factors of 1.5 or 2. If your tunsten frames have a wide range of counts you may be able to step by a larger factor.
• Dark frames [Experts only.]. Useful for examination of cosmic ray stats, and if you are running the CCD warm. Often difficult to be sure that light is not getting in somewhere so need all dome lights off.