William Jack Welch Experiments with the CMBR

Jack Welch retired from teaching Astronomy and Electrical Engineering at UC Berkeley in 2005 but continues as the Alberts Professor in the Search for Extraterrestrial Intelligence. He was Director of the Radio Astronomy Laboratory at Berkeley from 1972 to 1996 during which time the BIMA Millimeter Telescope Array was built and operated. He continues his research in the interstellar medium and star formation and is currently working on completion of the Allen Telescope Array.

My introduction to the question of the absolute radio brightness of the sky came from a talk that I heard at a meeting of the IEEE Antennas and Propagation group held in Palo Alto in 1961 or 1962. The Speaker was R. W. De Grasse, one of the team of engineers at the Bell Telephone Laboratories that had developed a communication system for the Echo project. He described the horn-reflector antenna and maser receiver amplifier that had been built at Crawford Hill in New Jersey. As a young radio engineer just beginning work in radio astronomy at Berkeley, I was enormously impressed with the quality of the instrumental work and the care taken with the system noise measurements. I remember him saying that they assumed the sky background temperature to be zero but were uncertain about an excess of a couple of degrees or so in their summary of system noise contributions. The excess was thought to be pick-up in the antenna side lobes (Ohm 1961). At the time, I had no idea what to expect for the background.

A few years later, I read the letter in The Astrophysical Journal by Penzias and Wilson (1965a) describing their beautiful background measurements with that same antenna. Using a new receiver at 4.08 GHz with a new reference load (Penzias 1965), they were able to report with certainty an excess of about 3.5 K that had to be ascribed to the cosmic background. The companion paper by the Princeton group (Dicke et al. 1965) with the plausible interpretation that the radiation was the blackbody radiation remnant of an earlier stage of an expanding universe was very exciting. George Field, who had recently joined the Berkeley Astronomy Department, was very taken with the new finding and realized that earlier observations of the excitation of interstellar CN (Herzberg 1950) might be consistent with the new radio observations. The excitation of the first rotational level of the CN line corresponded to background radiation at a wavelength of 2.6 mm, suggesting that the excess radiation was that of a blackbody in agreement with the Princeton group interpretation. At the time, our group was developing receiving equipment at wavelengths near 1.0 cm for radio astronomy and studies of atmospheric emission with a small antenna. George urged us to attempt a measurement of the CMBR to help determine its spectrum at the shorter wavelengths.

We decided to take a detour from our other program to study the background at a wavelength of 1.5 cm. An important piece of information about the universe had been found. We might be able to add to that, and it would be an interesting instrumental challenge. At 1.5-cm wavelength, the background emission from the atmosphere is rather high at sea-level sites, and we planned an observation from the High Altitude Barcroft Laboratory of the University of California White Mountain Research Station. Sam Silver, the Director of the Space Sciences Lab at the University of California, had outfitted a trailer for remote observations, and we were able to take it to the Barcroft Laboratory for our observations. The atmospheric emission brightness is typically only 3-4K at the 12,400-ft altitude of the Barcroft Laboratory. Our technique was conventional. We used a Dicke radiometer that compared the brightness of the sky as detected by a standard gain horn and associated receiver with that of blackbody loads at known temperatures, and we made tipping measurements to extrapolate the brightness to zero air mass. One difference in our system was that we used a load at the temperature of liquid nitrogen as our low-temperature reference rather than a liquid helium load. We felt that we could characterize it well and it would be easier to manage at the remote site than a liquid helium load such as those used by the other groups. As a check, we measured a liquid helium load in the lab at Berkeley with our system and found the correct temperature. We spent the summer of 1966 making background observations at the high-altitude site.

Our reported result, 2.0 ± 0.8 K, was disappointing (Welch et al. 1967). The final uncertainty was large. The reproducibility of individual measurements was limited by the scatter in the measurements of the liquid nitrogen load brightness. Because of the greater temperature difference between the sky brightness and that of liquid nitrogen, the extrapolated results were subject to greater random errors. In addition, our mean value was low in comparison with the results of the other measurements available at the time of our publication. The average, particularly including the first radio detections (Penzias and Wilson 1965a; Roll and Wilkinson 1966) and the temperatures derived from CN measurements (Field and Hitchcock 1966; Thaddeus and Clauser 1966) were pointing to a blackbody temperature of 3.0K or even higher, outside our error limit. As the more accurate measurements, shortly thereafter from the Wilkinson group (Stokes, Partridge and Wilkinson 1967) and others, and finally from the COBE satellite (Mather et al. 1990) came in, we were somewhat relieved that the limit of our error just included the final blackbody temperature, 2.725 K.

A year or so after our publication, I was reexamining the characterization of the pyramidal horn for some other calibrations that we were planning and discovered that I had made a mistake in the model tipping curve that we had used for the background measurements. Correcting for that properly, we would have had 2.3 ± 0.8 K for our result, a little closer to the final accurate temperature. Since that miscalculation was small compared to our random errors, we did not think it appropriate to publish it. In retrospect, I realize that was a mistake. The systematic error is, of course, different from the random errors, and it should have been reported.

We subsequently returned to our original program of getting a short wavelength telescope running for other astronomical observations, particularly for studies of Solar System objects and the interstellar medium. There we had some nice results with the first discoveries of polyatomic molecules in the interstellar medium revealing the molecular clouds where stars are born (Cheung et al. 1968, 1969). Then we proceeded to develop interferometry at short wavelengths for interstellar medium and star formation studies as well as for other fields.

Our most recent encounter with CMBR studies occurred when we discovered that we were making some accurate ground-based flux measurements of Jupiter at the same time that they were being made by the WMAP satellite in the course of its calibration (Page et al. 2003a). We had just completed our study when the WMAP results were announced. Our measurement was made at a wavelength of 1.05cm (Gibson, Welch and de Pater 2005), in between the two longest WMAP receiver bands and close to the center of the Jovian ammonia inversion absorption band. Our accuracy for the Jovian flux was about 1.5% and it fell nicely between the Jovian fluxes of the two adjacent WMAP observations that had comparable accuracies. I think everyone was pleased with the good agreement between these independent calibrations of Jupiter. Our result enabled us to get a fairly accurate measure of the upper Jovian atmospheric ammonia abundance. Absolute calibration to 1-2% accuracy was essential for getting a good Jovian atmospheric model, and the WMAP results helped with that as well.

Some of the best memories from the earlier period were of discussions with Dave Wilkinson, an experimentalist of extraordinary capability.

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment