NICS Spectroscopic Modes


NICS slits
Name Width Length
0.5 0.5" = 2 pix 4'
0.75 0.75" = 3 pix 4'
1.0 1.0" = 4 pix 4'
1.5 1.5" = 6 pix 4'
2.0 2.0" = 8 pix 4'

Long slit spectroscopic observations are performed by inserting a slit at the entrance focal plane and a disperser (grism or prism) in the collimated beam. The table here lists the slits available in NICS which, thanks to the refurbishment works of Feb-Mar 2003, have now a very stable and repeatable positioning. All spectroscopic modes make use of the LF camera with a scale of 0.25"/pixel.


NICS dispersers
Name wl-range (µm) dispersion (Å/pix) Res.power with 1" slit
 Low resolution
Amici 0.8-2.5 30-100 50
IJ 0.9-1.45 5.5 500
JH 1.15-1.75 6.6 500
JK' 1.15-2.23 11.6 350
HK 1.40-2.50 11.2 500
 Medium resolution
1mic 0.96-1.09 2.0 1250
Js 1.17-1.33 2.5 1200
J 1.12-1.40 2.5 1200
H 1.48-1.78 3.5 1150
KB 1.95-2.34 4.3 1250

The instrument is equipped with one prism and a number of grism dispersers whose main characteristics are displayed in this figure and listed in the table on the left. Note that the grisms have a fairly constant dispersion (Å/pix) throughout the spectrum and, therefore, their resolving power increases going towards the red. The Amici prism, on the contrary, delivers a spectrum with a quasi-constant resolving power and, therefore, its dispersion varies by more than a factor of 3 over its spectral range.
A few technical details can be obtained by clicking on the name of the disperser in the table.
All the low resolution dispersers can be used in combination with the grey filters to take spectra of very bright objects which would otherwise saturate the array.

Wavelength calibration and sky spectra

Lamps and sky spectra
Disperser Ar lamp Xe lamp Sky

GRISMS An accurate wavelength calibration and a proper rectification of the (curved) slit images can be achieved using a two dimensional polynomial of 3rd or 4rd degree in the dispersion (X) direction and 2nd or 3rd degree in the spatial (Y) direction. All the spectra are correctly oriented (i.e. blue is to the left), click on the rainbows in the table to get a view of the bi-dimensional spectra. The labelled wavelengths are in air and refer to lines which are bright and isolated. Please note that while the Argon lamp provides good calibration frames for basically all the grisms, the Xenon spectrum is less rich of lines in the K band. Examples of wavelength calibrated spectra in the form of ascii files can be also found by clicking on the grism name in the table.
Thanks to the intrinsic stability of grism dispersers, wavelength calibration lamps are usually taken once every several weeks. More frequent measurements are generally unnecessary because the sky is also rich of emission lines which can be conveniently used to check and correct for instrumental effects (if any). To avoid excessive overheads, users are discouraged from taking Ar/Xe calibration frames during the night.

AMICI PRISM The spectrum is flipped (i.e. blue is to the right) and occupies only the central part of the array. Due to the very low resolution, virtually all the Ar/Xe lines are blended and cannot be easily used for standard reduction procedures. For this reason, wavelength calibration is normally performed using a look-up table which is based on the theoretical dispersion predicted by ray-tracing and adjusted to best fit the observed spectra of calibration sources. Slit curvature is very modest and evident only in the reddest part of the spectrum, it can be usually neglected.


Example of an halogen exposure
with the IJ grism

The main reason why one needs a flat is to correct the "granulation" (not-uniform pixel-to-pixel response) which is instrinsic to the array and can be corrected for by using deep halogen lamps exposures taken within several days from the observations.
Please note that the calibration unit available in NICS does not produce an uniform illumination along the slit. To avoid that the flatting procedure may introduce spurious effects into the data, it is convenient to use only the high spatial frequency component of the flats, i.e. normalizing the halogen frames raw by raw to a 2nd-3rd degree 1D-fit of the background.


The observational performances for spectroscopy can be estimated using our Exposure Calculator which is based on the measured zero points, backgrounds, array noise and yields the average s/n ratio achievable for a source with a given magnitude or, alternatively, the time necessary to achieve a chosen s/n for a given source property.
Please note that the program assumes a maximum on-chip integration time of 900 sec and provides just a representative figure for the central wavelength of each photometric band for which a spectral-averaged background level is also adopted. The actual s/n varies significantly with wavelength following the instrumental efficiency curve and the spectral distribution of the sky emission.

The limiting fluxes for line detection can be roughly estimated by multiplying the flux per wavelength unit corresponding to a given object magnitude, times the line width which, for unresolved lines, is set by the slit width (in pixels) times the dispersion (Å/pix).

For any comments please contact Francesca Ghinassi.