ING Banner
Home > Astronomy > LIRIS > LIRIS Multi-Object Spectroscopy (MOS)

LIRIS Multi-Object Spectroscopy (MOS)

Planning MOS Observations
Science Target Acquisition and Observations
MOS Spectroscopy Scripts
Checking Mask Acquisition
Standard Star Observations

See also:
Calibrations: biases, darks, flats, arcs
Slits and Grisms
LIRIS Long-Slit Spectroscopy
Detector: Bright sources and reflected ghost images

Planning MOS Observations


If you have been awarded time for a MOS proposal, please contact the LIRIS instrumental team ( and the ING LIRIS MOS Specialist (Cecilia Farina: as soon as possible after telescope time allocation for advice. More details of slit-mask design and timelines are available in Considerations about LIRIS MOS observations.

LIRIS can accomodate about 6-8 MOS masks at a time. To change the masks, the LIRIS cryostat has to be warmed up, which is preferably done only once per semester since there is some risk involved for the sensitive IR detector array. If the number of applications for MOS mode is very high, two warmups may be able to be scheduled per semester or, preferably, either one of the long-slits or the coronographic mask will be traded in to accommodate more MOS masks. It is however possible that we will not be able to accommodate all the requested masks if demand is high.

MOS mask preparation

The coordinate list must be sent to as soon as possible after the telescope time allocation, as the external company that produces the masks needs one month for the production (but see above: they can only be inserted once per semester usually). It is your responsibility to provide the coordinate lists in due time. If you fail, you run the risk that by the time of your observing run your masks will not be available in LIRIS. Any questions concerning the scheduling and production of MOS masks should be directed to the IAC.

The MOS masks are produced using an electromagnetic discharge process. The slitlets are 0.8" wide and by default 10" long. About 20 such slitlets can be fitted per mask. The length of the slitlets is user-definable. They can be made shorter, but should ideally still allow for a 3-point dither pattern for optimal sky background subtraction. The slitlets of one mask do not all need to have the same length.

The MOS masks are inserted into a part of the focal plane of the telescope which is free from distortions. Therefore, it is sufficient that the observer provides a target list with accurate (0.1" or better) RA and DEC coordinates. The target list must include the coordinates of at least three reference stars which are used to align the mask on the sky accurately. Ideally, these three reference stars should form a triangle that extends over a large part of the detector area. The reference stars must be highlighted in the list since they will not get slitlets assigned in the mask but instead circular holes. Please note that in order to be able to properly trace the spectra, if every science target is fainter than the 19th mag, assigning a brighter star to a science slitlet is now mandatory.

The figure shows a 600s exposure through a MOS mask. The spectra are dominated by sky lines, hence it is important that the slitlets are chosen long enough (default: 10") to accomodate three dither positions. The spectra of three reference stars used for mask alignment can be seen here too. 

Science Target Acquisition and Observations

The acquisition of a MOS mask is similar to that for long-slit LIRIS Spectroscopy. The only major difference is that you have to specify three reference sources instead of one. The procedure is:
  1. Ask the telescope operator (OSA) to point the telescope to your field and rotate LIRIS to the required sky position angle (the information about the PA for a given mask is included in the slit mask's files: <mask_name> and <mask_name>
  2. Take an image of the field. If the sky background is high or the objects very faint you will need to take two images with an offset of about 5" in order to subtract the sky emission. This can be done using the RTD (Real-Time Display) or the task imarith of IRAF.
  3. Display the image and overplot the corresponding mask from the DS9 display, either using the DS9 ‘LIRIS MOS’ button or load the regions directly from the DS9 top bar menu: Region -> Load Regions -> /wht/var/liris_<mask_name>.reg
  4. If at this stage the offsets are large (over about 15"), calculate the offsets very roughly, to the nearest 5 pixels is fine, and ask the telescope operator to perform this movement in units of arcseconds (1 pix = 0.25"), since doing a large offset will make it hard to reacquire the guide star. 
  5. After the telescope has moved, take a new image. Measure and note the <x,y> coordinates of the reference stars using the Star profile task in the RTD control or imexamine IRAF task.
  6. To acquire the mask:

    SYS> lobject_inslit <mask_name> <x1> <y1> <x2> <y2> <x3> <y3 >

    The reference stars must be given in the right order. The order can be checked with the overlaid mask in DS9 or by looking at the file: /wht/var/liris_slitdb_pos.dat  Therein, you will find an entry for your MOS mask in the form: maskname 3 x1 y1 x2 y2 x3 y3  with xi and yi being the pixel coordinates of the reference sources. The routine lobject_inslit calculates shifts and rotations, and moves the telescope and rotator accordingly if the movements are accepted. The rotations are expected to be small (but can still lead to large movements of the guide star, which is why autoguider is not used at this stage). The mask is now roughly acquired. 
  7. Start autoguiding (the telescope operator does this, but please check with them that the "I" filter is used, especially if the time on target is more than about an hour).
  8. For fine-tuning the mask position, take another image, display it and note again the x,y coordinates of the target star, then use the lobject_inslit command again to calculate the offsets required but answer “no” when the move confirmation is queried. At this point it is quicker and more accurate to center the mask manually, asking the telescope operator to make these small offsets.
  9. Check the final position on at least 2 images, as sometimes the autoguider needs a little time to stabilize, especially when seeing is not very stable.
  10. When the reference stars are well centered, insert the mask only, not the grism:

    SYS> lslit <mask_name>

  11. Take an image through the mask and check that the center of the reference stars are in the centres of the holes. It may be necessary to adjust the contrast in the RTD display (switch off the autoscale option and select Zmin and Zmax values to have enough dynamic range in the display). If a further adjustment is needed, it is best done manually by the telescope operator, or following the procedure described in the next section, Checking Mask Acquisiton, since at this point the Star Profile task used previously may not work due to the background level.
  12. If the reference stars are in position, then you can switch to spectroscopy mode:

    SYS> lspec <grism_name> <mask_name>

  13. Use the appropriate script, ag_spec_nod or ag_spec_nod3, according to the mask's design (if the slitlets are large enough, you can fit 3 nodding points). Normally, the number of nod positions and the offset value are calculated during the mask design process and this information can be found in the mask's files: <mask_name> and <mask_name> Using this offset will ensure that the targets are not offset too far and go out of the slits.

Spectroscopy Scripts for MOS:

[ag_]spec_nod <int time> ["title"] [-ncyc=int] [-nruns=int] [-offset=float] [-mndr=int] [-jitter=float] [-clean=int] [-start_center] [-coave=int]

This nods the telescope in an AB-BA-AB-BA pattern up and down by the offset amount, with A and B being the two nodding positions. The nodding step size is given by offset in arcsec, defaulted to 12". The -start_center option should be specified, so that the telescope will consider the starting position as the middle point between A and B and will therefore offset half the offset value down before starting the script. If the autoguider is not desired (i.e. omit "ag_" at the start), use the spec_nod script (not recommended, unless you are observing the target for only about 3 minutes, e.g. for standards).

At each nodding position nruns images are taken with the specified integration time. Every subsequent coave (default 1) exposures are coaveraged (note if you enter a value for coave in the script that this overrides the default or any value previously used).The AB pattern is repeated ncyc times. When the script has finished, the telescope returns to the starting point. You can set the jitter parameter up to a maximum value of 30% of the offset parameter. Each nodding point will then be offset from its nominal position by a random vector of the specified length. For the guided version of the script only, starting with ag_ , the number of multiple non-destructive reads is automatically set to  mndr=4, if the mndr parameter is not set. Like that the exposures get read out 4 times and averaged, in order to suppress the read-noise. You can also set the clean option which is the number (default 3) of clearing reads before the first exposure in each position. 

Alternatively, you can use this script:
[ag_]spec_nod3 <int time> ["title"] [-ncyc=int] [-nruns=int] [-offset=float] [-mndr=int] [-jitter=float] [-clean=int] [-coave=int]

One cycle consists of an ABC pattern. The first exposure will be taken at position A=(0, 0), the second one at B=(0, +offset), and the third one at C=(0, -offset). The offset should not be larger than about 1/3 of the length of your slitlets, otherwise the star will be driven outside the slitlet.

Checking Mask Acquisition

Check the centreing of the mask about once every hour, or more frequently if you notice a drop in flux not due to cloud. The relative position between the autoguider star and the science target may change due to differential atmospheric refraction, especially if the telescope elevation is low.

To check the centreing, switch to imaging mode without moving the mask (to keep the mask in the light path):

SYS> limage <filter> <mask_name>

and take an image through the mask. If the reference stars are not in the right position, recentering is needed. The required offset is best estimated maually, rather than using a script, and passed to the telescope operator to make the offset. Then take another image to check it is now centred.

Finally switch back to spectroscopy mode:

SYS> lspec <grism_name> <mask_name>

and continue your observations.

You can also use the routine recenter_mos (which runs only on the whtdrpc1 computer) to check your acquisition:
1. Create a working directory:  whtobs@whtdrpc1$ cd /scratch/whta/ - whtobs@whtdrpc1$ mkdir
2. Copy there the file liris_.mask which can be found in the directorywhtobs@taurus:/wht/var/: - whtobs@whtdrpc1$ scp whtobs@taurus:/wht/var/liris_.mask - whtobs@whtdrpc1$ cd
3. Load the LIRIS utilities package by doing: - whtobs@whtdrpc1$ source /home/whtobs/liris/setup_lirisutils
4. Run the task recenter_mos: - whtobs@whtdrpc1$ recenter_mos  The output of the recenter_mos routine will give the offset needed to correct the acquisition.

Standard Star Observations

You might want to observe a standard star in a few or all the slitlets. Observing the standard star in all the slitlets is useful to determine the trace (since spectra at the bottom and top of the detector suffer geometrical distortion by about 10 pixels) along the spectral direction for targets located within each slitlet. The recommendation is to do it only if it is really necessary, since it is very time-consuming (over 30 minutes). In most cases it will be enough to observe the standard in three different positions: in the extreme left and right slitlets, and in one slitlet close to the middle of the mask. With this procedure full wavelength coverage is guaranteed.

The method to acquire the standard star in the selected slitlet is similar to that used in long-slit spectroscopy:

Point the telescope to the standard star, take an image and then roughly position the star in the desired slitlet N by:

SYS> lacq_mask <mask_name>_slitN

(to find the slitlet number look at the mask overlays in DS9 with the <mask_name> button).

Fine-tune the target acquisition by measuring the position of the centroid of the star (x and y coordinates in pixels) and running the command:

SYS> lobject_inslit <mask_name>_slitN <x> <y>

Once the centreing is finished (overlay your mask to check it), switch to spectroscopy mode:

SYS> lspec <grism> <mask_name>

and take your exposures using the same scripts as for the science observations.

Once you have observed the standard star in the chosen slitlets ask the telescope operator to reset to 0 0 the telescope aperture introduce by the task lacq_mask.

Notes about this method:
* It provides a very controlled way to obtain the standard spectra.
* It is possible to use the autoguider.
* It requires several switches between imaging and spectroscopy mode, which increases the overheads.


The telescope operator will take care of all aspects of the autoguiding, if required. All the observer needs to do is communicate with him/her about the details of the observations, in particular the size of any dither pattern being used. In general, for MOS observations, guiding will always be needed in order to keep the targets in the slits, except if the total time on the target is very small (< 3 mins), which may be the case for standards. Guiding overheads are in any case small: less than about a minute per target at the start and then a few seconds per exposure.

Top | Back

Contact:  (LIRIS MOS Specialist)
Last modified: 05 February 2015