Multi-long-slit Spectroscopy for Kinematic Studies. I.

1
Multi-long-slit Spectroscopy for Kinematic
Studies. I. Implementation and Demonstration Rene
A. M. Walterbos, Jiehae Choi, Sophia Cisneros,
Maria T. Patterson, Cat Wu New Mexico State
University, Las Cruces, NM 88003
Comparison to Other Methods
Introduction
Compared to most multi-fiber and lenslet
integral field instruments, this setup allows for
larger field of view. For our system, the 16
slits are distributed over a 4x5 arcmin sq. area.
Compared to a Fabry-Perot, our method can be
faster for obtaining 2-D velocity fields for disk
galaxies since we do not require scanning an
etalon through all velocity channels. Complete
2-D coverage can be obtained by stepping the
telescope in position between exposures. Galaxies
at different redshifts or other emission lines
can be observed by changing the narrow band
filter. The implementation is very inexpensive in
terms of hardware cost and involves only
production of a set of multi-long-slit masks and
suitable filters. The spectral resolution we can
obtain is typical of that of a medium resolution
spectrograph, about 1.5-4 Å depending on slit
width and grating. We demonstrate the
implementation through observations of the Owl
nebula and the nearby spiral M33.
Several methods exist for spectroscopy of
ionized gas in galaxies to measure the kinematics
of galaxies. These include single long-slit
spectroscopy, integral field spectroscopy where a
fiber bundle or lenslet-array feeds a
spectrograph, or an imaging Fabry-Perot. We
demonstrate here another method for 2-dimensional
spectroscopy, using a mask with several parallel
long slits in combination with a narrowband
filter. This enables observations at Ha for up to
16 slits simultaneously with the Apache Point
Observatory's ARC-3.5m telescope, using a
conventional optical spectrograph (the Double
Imaging Spectrograph). The method is not new
(e.g. Wilson et al. 1959) but is inexpensive to
implement on our spectrograph and quite efficient
for certain applications.
Fig. 1
Blue/Green/Red composite of the Owl Nebula.
The picture measures about 8.5' on a side. The
Owl nebula is a bright source of diffuse Ha
emission which makes it suitable as test object
to verify our velocity consistency across the 16
slits. Image credit Gary White and
Verlenne Monroe/Adam Block/NOAO/AURA/NSF
Fig. 4
ARC 3.5m image of the central region of the
nearby spiral M33 through a B filter. The image
is 4.6x4.6 arcmin. The major axis position
angle of M33 is 22 degrees.
Fig. 2
Four exposures of the Owl Nebula with our
16-slit setup. One exposure was centered on the
nebula, while the other three were offset by 2'
in different directions, as shown. The offset
exposures allowed us to measure how consistent we
can measure velocities across the slits. Our goal
is not to resolve the expansion of the shell seen
in high resolution long-slit echelle spectra
(Guerrero et al. 2003) but to verify we measure
the same velocities when different slits are
positioned at the same location. On occasion,
NII lines from the neighboring slits are
visible.
Fig. 5
Exposure of M33's central region with our
16-mask setup. The PA for the slitmask was 22
degrees, parallel to the major axis. A sky
spectrum was obtained from a separate sky
exposure, offset from the galaxy, and subtracted.
We obtain signal across much of the field. The
nucleus of M33 is the continuum source near the
center it demonstrates how the filter cuts off
light outside its narrow passband, which allows
the use of 16 slits. The Ha line is the most
prominent line, but in many cases one of the 6548
Angstrom NII line is visible as well.
Table 1. Muli-slit setup for the testdata

Telescope ARC 3.5m
Instrument Double Imaging Spectrograph, red side
Slitmask 16 slits, each 4.5 x 2, spaced 15 apart
Area sampled 3.8 x 4.5 (limited by filter size)
Filters 2 x 2 inch, 6560 and 6570 Ha, 25 Å FWHM
Dispersion 0.82 Å/pix, 3.2 Å FWHM
Throughput 20 (including telescope and spectrograph)
Exposure times Owl nebula 2x300 sec M33 2x900 sec
Wavelength H-arc lamp
Positioning Through offsetting from slitviewer on single long slit
Conclusions
Fig. 3
Comparison of the velocities measured across
the Owl Nebula for the various slits in the
central and offset frames. Our systematic
velocity offsets between slits are typically less
than 5 km/s, and at worst about 10 km/s. These
systematic errors likely stem from small changes
in the wavelength calibration between arc
exposure and data exposure, related to rotation
angle of the spectrograph at the Nasmyth focus.
We can minimize these with more frequent arc
exposures.
We demonstrate we can obtain velocities
across 16 slits simultaneously in one emission
line to an accuracy of about 5-10 km/s over a
4'x5'field of view. This multi-plexing method
enhances the efficiency in obtaining 2-D velocity
maps compared to single longslit spectroscopy. In
comparison with other integral field spectroscopy
methods, our setup is inexpensive, has a
relatively large field of view, can work faster
than Fabry-Perot for certain applications (at
expensive of incomplete spatial coverage). It
generally has higher throughput than the typical
fiber spectrographs. The multi-slit setup can be
used at redshifted wavelengths using other arc
lines. Using different slit mask designs, we can
maximize spectral resolution (e.g. 12 slits of 1"
spaced 20" apart with a higher dispersion
grating) or detection of faint emission (e.g. 45
slits spaced 5" apart when used with our medium
resolution grating).
Fig. 6
Upper Panel Major axis stellar and gas
velocities from a single-long slit spectrum
obtained by Corbelli Walterbos (2007). Right
panels Velocities measured along the 16 slits
shown in Figure 5 for the central region of M33.
Slit 8 is the major axis spectrum. In Corbelli
Walterbos we argue for the presence of a bar in
M33 based on the gas kinematics. Notice the sharp
jump in velocities across the nucleus. In
addition, the maximum velocities are not observed
on the major axis but in slits 10 and 11. We have
obtained complete mapping for this region by a
series of 8 exposures with 2" spacing, for better
constraining the parameters of the bar. These
data are currently being analyzed.
References
Acknowledgements
This research was supported by an award from
Research Corporation. We acknowledge early
discussions about this mode of spectroscopy with
Robert Braun and the late Michael Ledlow.
Corbelli, E., Walterbos, R.A.M., 2007, ApJ 669,
315 Guerrero, M.A., et al., 2003 AJ, 125,
3213 Wilson, O.C., 1959, ApJS 4, 199