First Commissioning of the IR Spectrograph
LIRIS
J. A. Acosta Pulido, E. Ballesteros, M. Barreto, R. Barreto, E. Cadavid,
J. Carrillo, S. Correa, J. M. Delgado, C. Domínguez-Tagle, E. Hernández,
R. López, A. Manescau, A. Manchado, H. Moreno, J. Olives, L. Peraza,
F. Prada, P. Redondo, V. Sánchez, N. Sosa, F. Tenegi (IAC)
The
near-infrared spectrograph LIRIS was first commissioned on the WHT on the
nights of 15–19 Feb 2003. During all that time the instrument functioned with
no major failures. At the time of commissioning LIRIS included imaging and
spectroscopic observing modes. Spectroscopy could be done at low resolution
(about 700 using the narrowest slit of 0.65") using two grisms: One simultaneously
covers the bands
Z and
J (9230 – 15,425 Å) with a dispersion
of 6.1Å/pix; and the other covers the bands
H and
K (14,600
– 24,946 Å) with a dispersion of 10 Å/pix.
|
|
Figure 1 (top left). LIRIS mounted
on the WHT Cassegrain focus. LIRIS cryostat can be seen at the bottom of the
telescope focus, with the two electronics racks at both sides [ JPEG | TIFF ]. Figure 2
(top right). The globular cluster M5 in the J band. The FWHM of the PSF was
2 pixels corresponding to 0.5". The image quality is very good over the whole
field of view (4.2 arcmin2) [ JPEG
| TIFF ]. Figure 3 (right). Arrival of LIRIS at
the WHT in January [ JPEG | TIFF ].
|
|
The weather conditions during the commissioning were good; 3 out of 4 nights
were photometric and only half a night was lost due to high humidity. On two
of the nights the seeing was about 0.5", as measured by LIRIS. The image quality
over the whole field appeared to be very good (see
Figure 2). The PSF over
the whole field remained very uniform, with variations of width smaller than
half a pixel (0.13") across the whole field of view. The image quality was
very good in the different bands, and the best telescope focus results constant,
independent of the filter used.
|
|
Figure 4 (left). The planetary nebula
NGC2346 in the emission of H2 ν =1 - 0 S(1) at 2.122μm. The field of view
covered in this picture is approximately 3.2x2.8 arcmin2. North
is to the right and East is at the top. Note the clumpy structure in the
lobes and the bright central star, only visible in the infrared [ JPEG | TIFF ]. Figure 5
(right). The Seyfert 2 galaxy NGC4388 observed in the J filter. The field
of view covered in this picture is 2.5x2 arcmin2. North is at
the top and east to the left. Note the very bright active nucleus and the
patchy structure of the spiral arms, revealing the presence of obscuring
dust lanes [ JPEG | TIFF ].
|
LIRIS was funded and built by the IAC, the optical and the conceptual mechanical
designs were provided under contract by the UKATC. The Spanish contractor
INGOVI manufactured the LIRIS vacuum vessel and optical bench. For more detailed
information about LIRIS’ design, manufacturing and capabilities see Acosta-Pulido
et al. (
2002, ING Newsletter, 6,
22). For updated information, including instrument simulator, please
consult our web site at:
http://www.iac.es/proyect/LIRIS/.
LIRIS is equipped with a 1024×1024 Hawaii detector, using a SDSU controller.
The engineering detector was used during the commissioning. The detector temperature
was kept stable at 61 K. The readout noise was 4.8 ADU or 24e
–
in double correlated mode. This value can be effectively reduced using multiple
non-destructive readouts (for instance it reduces to 12e
– when
4 readouts are made). The minimum integration time allowed by the controller
is 1s. LIRIS is always limited by background noise for imaging mode in
H
and
Ks bands, and in
J band for exposures longer
than 4.5s. In spectroscopic mode the same condition is reached for exposure
times longer than 380s and 42s in the ranges
Z –
J and
H
–
K, respectively.
The photometric zero point and the system efficiency (optics & detector)
were measured in the different bands (see Table 1). We also report the average
sky brightness. Remarkably, the sky background in Ks measured with LIRIS is
among the lowest reported with similar instrumentation at different telescopes.
We would like to point out the fact that the WHT is not an IR optimised telescope.
The limiting magnitude was computed for detection at 3s in an hour of on-source
integration with a seeing of 0.7".
Filter |
J |
H |
Ks |
Zero Point |
24.83 |
25.17 |
24.55 |
Efficiency |
0.34 |
0.53 |
0.52 |
mlim |
23.4 |
22.4 |
21.8 |
<msky/arcsec2> |
15.4 |
14.2 |
13.0 |
Table 1. LIRIS photometric
characteristics.
|
The rigidity of the instrument, in particular the flexures of the slit wheel
with respect to the rest of optics is a critical point for spectroscopic observations.
A displacement along the spectral axis during a LIRIS exposure will introduce
several unwanted effects, such as smearing of the spectral features, and
flux losses due to light coming from the object not passing through the slit.
Moreover the position of the slit on the detector needs to be known at any
time in order to accurately centre the target object. The LIRIS rigidity
was checked with good results. It was found that the maximum shift (with
respect to zenith position) along the spectral direction does not exceed
0.5 pixel, or 0.12", at 45º zenith distance (ZD), although it reaches
0.8 pixel at 60º ZD. The flexures along the spatial direction were slightly
worse, reaching about 1pixel (0.25") at 45º ZD and certain rotation
angles of the Cassegrain turnplate.
A key issue in near-IR astronomy is the sky background subtraction. This
is generally performed by following dithering patterns on the sky, which involves
a good deal of interaction of the instrument data acquisition with the telescope.
The WHT was already prepared for this based on the INGRID experience, although
the instrument LIRIS introduces new demands in the spectroscopic mode. For
IR spectroscopy the target is often offset along a narrow slit and should
be always maintained well centred on it to avoid flux losses. For this purpose
the telescope and the auto-guider should work in synchronisation with the
instrument data acquisition. It was found that the repeatability of the offsets
could not be guaranteed beyond 1pixel or 0.25", which involved target re-centring
after a couple of movements. However the WHT auto-guider is going to be changed
and the situation should improve.
The next commissioning period is foreseen for February 2004. During that
period the multi-object spectroscopy mode will be the main focus. We also
expect that LIRIS can provide polarimetric and coronographic capabilities.
The instrument has so far been used to observe several astrophysical targets
of interest. Some of the initial results are presented in the accompanying
figures (see Figures
4,
5 and
6). One of the most remarkable results was the
observation of the most distant quasar known at the time (SDSS J1148+5251,
z = 6.41). A spectrum in the bands
Z and
J was obtained,
in which several broad features were detected.
|
|
Figure 6. Two-dimensional spectrum
of the most distant QSO at z=6.41 (top bright row) [ JPEG | TIFF ]. The extracted
spectrum is shown in the left panel. A fit to the spectrum is also shown,
where several broad emission lines are identified [ JPEG | TIFF ]. The most
intense feature is the CIV line, detected with a S/N ratio of 10. The spectrum
is the co-addition of 5 frames of 850s exposure time each, giving an approximate
total time of 70 minutes.
|
We would like to thank all ING staff and the IAC Instrumentation Area for
their excellent support during the preparation and commissioning periods.
¤
Email contact: José
Acosta (
jap@ll.iac.es)