Unlike conventional telescopes, that have metal structures supporting the primary or secondary optics, the GBT's aperture is unblocked. The GBT is a 100 x 110 meter section of a conventional, symmetric 208 meter paraboloid. An off-axis feed-arm supports the primary and secondary optics (see Jewell & Prestage, 2004, SPIE, 5489, 312 for additional details). The GBT's asymmetric structure increased the cost of the telescope in both it's design and construction. What are the benefits?
The clear aperture of the GBT improves the sensitivity in two ways: (1) reduced stray radiation that effectively lowers the system temperature; and (2) an increase in the aperture efficiency. Measurements of the system temperature for the same receiver on the 140 Foot compared to the GBT indicates a 25% improvement. At the longer wavelengths the aperture efficiency is 0.55 for the MPIfR 100m telescope and 0.70 for the GBT. Combined this corresponds to an increase in the signal-to-noise ratio by a factor of 1.7.
The figure below shows two spectra of the 22 GHz water maser emission observed toward the galaxy IC 2560 from the GBT (top) and the 140 Foot (bottom). The GBT spectrum consist of a single polarization, 1 hr integration taken during commissioning in April 2002. Only the inner portion of the 800 MHz spectrum is shown. The 140 Foot spectrum resulted from a 2 hr integration in dual polarization taken in December 1997. Although it is difficult to make a direct comparison with these data, the GBT spectrum is clearly more sensitive. (See Braatz et al., 2003, ApJS, 146, 249 for details.)
22 GHz water maser emission observed toward the galaxy IC 2560 from the GBT (top) and the 140 Foot (bottom).
Parabolic antennas have a response that consists of a main beam in the direction the antenna is pointing and side-lobes in other directions (see Condon & Ransom for a discussion of antenna fundamentals). The clear aperture of the GBT produces side-lobes that are down by about an order of magnitude from traditionally designed telescopes. For example, the MPIfR 100 m telescope has side-lobes from the support feed legs that are 17 db lower than the main beam (Sun et al. 2006, A&A, 447, 937). The GBT first side-lobes are about 28 db down from the main beam - a factor of 12 improvement.
Below are two images of the Omega nebulae (M17) taken with the GBT and the MPIfR 100 m telescope at 9 GHz. The half-power beam-width is about 80 arcsec in each image. The MPIfR image was formed by scanning the telescope in both right ascension and declination. A zero level baseline was removed from the edge of the image. The data were calibrated and combined using the program WADDSC (see Balser et al 1995, ApJS, 100, 371 for details). The GBT image was formed by scanning the telescope only in one of the cardinal directions. A zero level baseline was removed from the edge of the image. The data were calibrated and the image smoothed using aips++ (Priv. Comm. Frank Ghigo.)
The rms noise level is 0.03 and 0.20 Jy/beam for the GBT and MPIfR images, respectively. The dynamic range is 1900 and 230 for the GBT and MPIfR images, respectively. The GBT image has a factor of 8 better dynamic range.
The Omega Nebulae (M17) taken with the GBT at 9 GHz.
The Omega Nebulae (M17) taken with the MPIfR 100m at 9 GHz.
Improved Spectral Baselines
Traditionally designed single-dish telescopes that have a blocked aperture are not only less sensitive but produce spectral baseline structure. Reflections from the superstructure cause spectral baseline ripples that are pseudo-sinusoidal and vary with sky frequency and therefore with the time of year. For the study of wide, weak spectral lines this structure can dominate the random, thermal radiometer noise.
Below are two spectra of the Galactic HII region S209 taken from the 140 Foot telescope and the GBT. The observing procedures were very similar for the two data sets. Total power position switching was employed with an ON and OFF integration time of 6 minutes each. The total time is about 10 hours for each spectrum. The GBT data have been smoothed to the velocity resolution of the 140 Foot data, 2.7 km/s (see Bania et al. 2002, Nature, 415, 54). The GBT receivers, consisting of cooled HFET amplifiers, were used on both telescopes. (The GBT receivers were installed on the 140 Foot before the GBT structure was complete.) S209 is an extended object that fills both the 140 Foot telescope beam (3.5 arcmin) and the GBT beam (1.4 arcmin). Therefore we expect similar antenna temperatures from each telescope. The vertical lines mark various recombination lines and the hyperfine line of 3He+ (at 8665.65 MHz).
The GBT spectra are relatively flat and reveal two spectral lines: the H171η recombination line and the 3He+ hyperfine line detected with a signal-to-noise ratio of about 3. There is no obvious detection in the 140 Foot spectrum. It required an additional 20 hours or a total integration time of 30 hours with the 140 Foot telescope to detect these lines to the same level of confidence. The better sensitivity of the GBT is due primarily to the lower system temperatures from less spillover and reduced spectral baseline structure from the unblocked aperture.
The Galactic HII region S209 taken from the 140 Foot telescope.
The Galactic HII region S209 taken from the GBT.