Letters to the Editor

Accelerating Universe:  Who Knew What When?

Ed. Note: We featured the accelerating universe in our “This Month in Physics History” column in January. This was followed by a series of letters in February and April. With the two letters below, we close our discussion of the history of how this discovery was made. 

Michael Riordan’s letter to APS News (April 2009) contained a vivid account of the scene as Saul Perlmutter presented the results of the Supernova Cosmology Project at a Santa Cruz colloquium in December 1997. His narrative of Saul reporting evidence for low cosmic mass density and unlimited expansion is fully consistent with what both the High-Z Team and the SCP said in January 1998. This confirms the point I was trying to make in my February 2009 letter to APS News: that the persuasive evidence for something qualitatively different–cosmic acceleration–came just a little later. February 1998 if you like Alex Filippenko’s conference talk at the Dark Matter meeting, March 1998 if you like the High-Z Team’s Astronomical Journal submission. Ruth Daly’s point in her letter to APS News is that ideas about acceleration were in the air in January, mine is that data of adequate precision to make the case in a refereed journal were just ahead.

Riordan makes three additional points. He says that Saul Perlmutter pioneered the technique of taking successive photographs four weeks apart in the dark of the moon to discover supernovae, that experimental particle physicists “familiar with manipulating vast quantities of data” felt up to the task of finding supernovae in digital images from CCDs while astronomers did not, and that the SCP result had “significantly better statistics” than the High-Z Team result. Each deserves a brief factual response.

Saul Perlmutter led the SCP with determination, but he did not invent the rhythm of the moon, which approximately matches the 21-day rise time of supernovae and leads to a four week cycle for efficient searches. This pattern of observing in the dark of the moon was pioneered by Caltech’s Fritz Zwicky, starting in the 1930’s. Monthly spacing of the search was used extensively in the Calan/Tololo search in Chile to find objects for scheduled follow-up. This set of data on nearby supernovae was used by both groups to establish the reality of cosmic acceleration. Monthly searches were employed, but not invented, by the SCP.

Replacing eyes with computers to find supernovae in digital data was an important development. In 1988, Danish astronomers used their 1.5m telescope at the European Southern Observatory to search for supernovae. They took digital images each month of galaxy clusters, registered the new images with the old, scaled the sky, convolved the images to match atmospheric blurring, and subtracted to find new stars in distant galaxies. All this manipulation of data was done in real time at the observatory. By astronomers.Their discovery of SN 1988U, a Type Ia supernova at the cosmologically interesting redshift, z =0.31, was reported in Nature. I wrote the “News & Views” explaining the importance of this work. Experimental particle physicists invented their own methods, but pixel-by-pixel subtraction to find supernovae for cosmology was carried out first by Danish astronomers (who didn’t think it was very difficult.) If their detector had been a little bigger, perhaps they would have found cosmic acceleration.

Finally, comparing the error ellipses shows that the statistical uncertainty in the values of Omega_lambda and Omega_matter from the High-Z Team in our September 1998 AJ paper was every bit as good as that from the SCP in their June 1999 Ap J paper, despite our having a smaller sample of high redshift objects. That’s because we had a larger sample of low-redshift objects from both Calan/Tololo and the Center for Astrophysics, an object-by-object way of determining the reddening to each supernova, and a larger fraction of excellent measurements from the Hubble Space Telescope. A bigger sample does not always yield a more precise answer.

All of us have had a great scientific adventure in learning that the universe is accelerating. That thrill of discovery is part of the fun of doing science, and it is based on a long sequence of contributions by many minds and hands. That long chain of cooperation is what I'd call success.

Robert P. Kirshner
Cambridge, MA