CRIRES User Manual
Doc. Number: ESO-254264
Doc. Version: P109.4
Issued on:
2021-12-01
Page:
52 of 99
Document Classification: ESO Internal Use [Confidential for Non-ESO Staff]
background is effectively removed by subtracting one frame from the other and vice versa.
This process is sometimes called
double subtraction
.
The total on-source integration time is split in
N
pairs of exposures;; the exposures pair (i.e.,
usually referred to as AB pair) are taken always along the slit but at a given separation. The
telescope nods between the two positions, A and B. The number of nodding cycles (i.e.,
number
N
of AB pair spectra) and the nodding throw (i.e., separation between the A and B
positions) are free observing parameters. Given that in spectroscopy mode the slit length is
10 arcsec, a nodding throw of 9 arcsec will put both beams close to the edges of the slit.
For larger nodding throws, both beams will be outside the slit (this can be useful when
studying extended objects). The maximum allowed nodding throw in spectroscopy is 30
arcsec, and any nodding throw larger than 8 arcsec requires a waiver in service mode.
In spectro-polarimetry mode the nodding throw is fixed at 2.5”. The number of defined AB
pairs affects the total amount of overheads, therefore already at Phase 1 we strongly
encourage the user to carefully plan the observing strategy.
We note in passing that no M2 chopping can be done with CRIRES.
5.4.1.2 Jittering
The purpose of jittering is to correct for bad pixels and decrease systematics originating
from the detector. This is particularly important for CRIRES observations, as the detectors
suffer from a relatively large number of bad pixels. In addition, the spatial extent of a
spectrum is at most a few pixels in nominal conditions with AO. Jittering is obtained by
adding a small, random o
ff
set to the telescope in addition to the nodding offset. The
recommended maximum size of the jitter offset is a free parameter. It must be smaller than
half the nodding offset but larger than the spatial extent of the minimum feature that one
hopes to detect;; for point sources, it should be larger than the width of the spatial profile.
The jitter offset is always along the slit.
5.4.2 Flux calibration and telluric correction
5.4.2.1 General Procedure
Flux calibration and telluric correction are generally carried out in three steps:
1. Removing the telluric absorption features by dividing the wavelength calibrated
science spectrum by the one of a telluric standard star, or a suitable synthetic telluric
model spectrum.
2. Removing the intrinsic spectral features of the telluric standard imprinted in the
science spectra after performing step 1.
3. Setting the absolute flux scale by using a spectro-photometric standard.
The spectro-photometric standard and the telluric standard can be the same star. Note that
CRIRES, as spectrometers in general, is not meant to provide high absolute spectro-
photometric accuracy due to slit losses. A list of spectro-photometric standards supported
by the pipeline is provided in Table 14.
The Observatory does not take observations of telluric or spectro-photometric
standard stars as part of the calibration plan
. Observers who wish to correct for telluric
features or flux calibrate their spectra therefore need to supply a standard star OB.
