Center for Astrophysical Research in Antarctica
Click image for a larger, readable version.The intense cold of the Antarctic winter night dramatically darkens the sky at infrared wavelengths. Telescopes located there can produce images which would be difficult or impossible to obtain at other earth-based sites. In the near-infrared spectrum from 3-5 µm, where the gains are largest, there are many unique tracers of gas, stars, and dust. Three of these were used to construct this picture of the star-forming region NGC 6334. The data were taken at the South Pole during the winter of 1998 with the 60-cm SPIREX telescope and the million-pixel Abu infrared camera.
In this composite photograph, red traces the emission from ionized hydrogen gas in a spectral line at a wavelength of 4.05 µm, green traces the "L-band" continuum light emitted from stars in the spectral region between 3.20 and 3.82 µm, and blue traces the fluorescent emission from very small "PAH" dust particles in a narrow spectral band centered at 3.28 µm.
Most of the stars in the picture are very young. They are still surrounded by remnants of the massive interstellar cloud of gas and dust from which they formed. Ultraviolet radiation from the hottest and most luminous of these stars illuminates the cloud material, ionizing nearby gas and causing both gas and dust to radiate profusely at infrared wavelengths. Some stars, the reddest in this picture, are so heavily enshrouded that they can only be seen at infrared wavelengths. Some have yet to be born. Observations in the infrared are crucial for understanding the cycle of star formation in the interstellar medium. The intensity and spectral distribution of the infrared light is intimately related to the heating and cooling of the interstellar material, and it penetrates obscuring dust clouds much more readily than visible light. Without the clues provided by infrared measurements, solving the mystery of how stars are born would be impossible.
Click image for a larger, readable version.Dwarf galaxies like the Large Magellanic Cloud provide nearby laboratories which simulate conditions in galaxies which formed soon after the Big Bang. The matter produced in the Big Bang was composed almost exclusively of hydrogen and helium. Except for lithium, the amounts of heavier elements or "metals" were negligible. The elemental abundances seen in our solar system today were built up through successive generations of star formation in lower-metallicity interstellar clouds similar to the Large Magellanic Cloud's 30 Doradus nebula.
The L-band (3.5 µm) infrared image used for the red component of this composite photograph is the most sensitive ever obtained of the 30 Doradus region. It was made at the South Pole during the winter of 1998 with the 60-cm SPIREX telescope and the million-pixel Abu infrared camera. K-band (2.2 µm) and H-band (1.6 µm) images taken at the National Optical Astronomy Observatories' Cerro Tololo Inter-American Observatory were used for the green and blue components, respectively. Analyzing such multiband, widefield, infrared images is the most efficient way to discover the birthplaces of the types of massive stars which are the principal agents of the elemental enrichment of the universe. They stand out clearly as the reddest stars in the picture because of their high L-band fluxes.

In these plots, we present a comparison of the measured sky brightness
at zenith between the South Pole and Siding Spring (Australia) from
2-4.3 µm. This data was collected using the IRPS using its CVF. In the first figure
(2-3 µm), the South Pole
spectra was collected on 31 May 1994 when the ambient temperature
was -62 °C, while in the second figure (3-5 µm) the
South Pole spectra was collected on 2 June
1994 when the ambient temperature was -66 C. In both plots, the
Siding Spring data was collected on 9 December 1993 at +10 C.
The units on the vertical axis of the left-hand plot are micro-Jy
arcsec
.
Note the large dip in the South
Pole sky brightness between 2.25-2.5 µm compared to the temperate
latitude site. The South Pole sky is 20-100 times darker at these
wavelengths than any other sight on earth which is accessible for
large telescopes. CARA specifically designed the so-called K_dark
(2.29-2.43 µm) filter to exploit this atmospheric window.
The units on the vertical axis of the right-hand plot are milli-Jy
arcsec
.
Note that between 2.9-4 µm, the South Pole sky is
more than 10 times darker than at a temperate latitude site.
CARA is currently exploring the utility of this window with the
Abu camera mounted on SPIREX.
For more information on the IRPS
observations, please see Ashley et al.
1996.
For more information on site characterization, please see the site characterization section.
Click image for a larger, readable version.To make the point that the South Pole sky is dark at the upper end of the K photometric band, we present 2 images of Baade's window taken with the GRIM camera mounted on SPIREX during the 1996 season. Both images are mosaics consisting of 14 separate images, each of which has a field-of-view of 2 arcminutes on a side. The integration time for the individual images was 10 minutes, and each image has been sky-subtracted before inclusion in the mosaic. The only difference between the 2 mosaic images is that the one on the left was taken using a standard K-band (2.00-2.41 µm) filter, while the one on the right was taken through the K_dark (2.29-2.43 µm) filter. Clearly, one can see more stars in the K_dark image because one can detect fainter stars in this band. The K_dark window offers a tremendous advantage in the search for faint objects that emit starlight, including low-mass stars and low surface brightness galaxies. (Images courtesy of Dr. John Bally, University of Colorado.)
Click image for a larger, readable version.
We present a sky dip (also known as a zenith scan) obtained with the GRIM
camera, which we used to measure the sky brightness at the South Pole in the
K_dark (2.29-2.43 µm) band. During the 1994 observing season, we made 22
such measurements from April through August. For each sky dip, we varied the
zenith angle of the telescope from nearly overhead to nearly horizon-pointing
(black dots). Each scan consisted of 4 sky dips taken at different azimuths,
and the plot shows one 4-dip data set. The background brightens in proportion
to the airmass to about 50 degrees from the zenith, then rolls over within about
20 degrees of the horizon. We attribute the roll off to atmospheric absorption.
We have fit a portion of the zenith scan with a model (solid line) consisting of
the secant of zenith angle plus a constant term. This model allows us to
separate the fixed telescope contribution to the background from the
contribution made by the sky which varies with the airmass. From our
measurements, we conclude that at the zenith the winter South Pole
background is 162 +- 67 micro-Jy
arcsec
in K_dark. This can be compared to a K' sky
brightness at zenith of about 4000 micro-Jy
arcsec
at Mauna Kea, Hawaii, a
very good observatory site at temperate latitudes. Thus, during the winter the
South Pole appears to be the darkest earth-based location from which to make
K-band observations. We also conclude that the SPIREX/GRIM instrument
emissivity is no more that 5%, which was our design goal. For more information
on this work, please see Nguyen et al. 1996.
For more information on site characterization, please see the site characterization section.

We present K_dark (2.29-2.43 µm) images of the southern hemisphere spiral
galaxy ESO 240-G11. For the left-hand image, north is up and east is to the
left, and the field-of-view is 8.7 arcminutes on a side. The total on-source
integration for this image was 3.3 hours. We have highlighted the surface
brightness contour at 23.7 mag
arcsec
,
which is where the signal-to-noise
ratio is 3. This is unprecedented depth for a ground-based near-infrared
image.
ESO 240-G11 is interesting because it is morphologically and dynamically similar
to the northern hemisphere galaxy NGC 5907 around which observers have detected
a stellar corona that appears to trace a dark-matter halo. From our deep
near-infrared image, we conclude that ESO 240-G11 does not have such a corona,
and we therefore hypothesize that formation of such a corona around spiral
galaxies is influenced by the local environment of the galaxy. This image was
taken with the GRIM camera mounted on the SPIREX
60-cm telescope during
the 1996 season. For more information, please see
Rauscher et al. 1998.
The image on the right is a similar image of ESO 240-G11 acquired during the
1997 season. The total integration time on-source was 6.6 hours. We are
currently analyzing this image in order to verify the 1996 results. For
more information, please contact David
Barnaby. (Images
courtesy of David Barnaby, Yerkes Observatory.)
Click image for a larger, readable version.SPIREX/GRIM had a nearly uninterrupted view of Jupiter throughout the period when Comet Shoemaker-Levy was colliding with Jupiter during the 1994 observing season. Of the 20 fragments for which timing was available, SPIREX/GRIM captured 16 and saw obvious evidence of an impact in 10 of them. The sequence of cometary fragment impacts and their aftermath forms the basis of Scott Severson's dissertation. A journal paper is in preparation. For more information, please contact Scott Severson.
The NSSDC has a nice summary of the SPIREX/GRIM SL9 images.

We present a K_dark image of ESO-LV 1450250, a low surface brightness galaxy in
the southern hemisphere. Because the sky brightness at the South Pole is
low in the in near-infrared,
low surface brightness galaxies are relatively easy to detect and observe, and
we conducted a survey of 6 such galaxies during the 1997 observing season.
These galaxies are interesting because they contain relatively few stars, yet
sometimes possess morphological and dynamical properties of much larger
galaxies. They may also represent a large percentage of the galaxies which
populate the universe. We are investigating what limits star formation in these
galaxies and what is supplying the gravtity that holds them together. This
image is the culmination of 6.5 hours of on-source integration and is about 4
arcminutes on a side. North is up and east is to the left. At galaxy center,
the B band surface brightness is 23.1 mag
arcsec
, while the K-band central
surface brightness is 20.4 mag
arcsec
.
(Total brightness: B = 14.2, K =
12.2). We conclude from our observations of ESO-LV 1450250 that
Click image for a larger, readable version.From May to September 1997, we measured the near-infrared extinction (magnitudes per airmass) in the K_dark band at the South Pole on 43 different days using the GRIM camera. On most days, we made at least two measurements, and on 12 days we made three or more measurements, for a total of 91 separate measurements. The histogram above shows the distribution of extinction for all 91 measurements. The median extinction is 0.11, the 3rd quartile is 0.31, and mean is 0.18. The negative values are not physically meaningful but indicate times of high variability. With a statistically sampled parameter like exinction, negative numbers sometimes arise, an artifact that other researchers have also noted.
Click image for a larger, readable version.
This histogram shows the distribution for the average value of K_dark extinction for the 43 days. The median extinction (magnitudes per airmass) is 0.14, the 3rd quartile is 0.2, the mode is 0.083, and the mean is 0.19.
Click image for a larger, readable version.
During a 7 day period in July 1997, we measured photometric scatter in the K_dark band by monitoring solitary calibration stars over a 30-minute period, taking an image each minute. In this way, we collected a total of 45 30-minute series. The stars ranged in airmass from 1.1 to 1.7, and during the week the extinction ranged from 0.022 to 0.212 mag/airmass. We then measured the mean and the standard deviation of the instrumental magnitude during the 30-minute series. This histogram shows the distribution of the standard deviation for the 45 measurements. The median scatter is 0.024 mag, the 3rd quartile is 0.04 mag, the mode is 0.016 mag, and the mean is 0.035 mag.
We are preparing a paper summarizing these and other results from 1997. For more information on related work, please see Marks et al. 1999 or Barnaby et al. 1997.
For more information on site characterization, please see the site characterization section.
Questions? Comments? email us at caraweb@astro.uchicago.edu Last modified Friday, 07-May-1999 13:46:20 CDT