THE SOUND OF THE BIG BANG
For the January 2001 issue of Analog, I wrote a column entitled "BOOMERanG and the Sound of the Big Bang." It described a then-recent Antarctic balloon flight that had mapped the small-angle temperature variations of the cosmic background radiation (CBR) and discussed the implications of these measurements for the flatness and structure of the universe. Following the lead of the scientists involved in the project, I described the temperature variations that they observed as, in effect, a recording of the "sound of the Big Bang" when the universe was 376,000 years old.
That column was published two years ago, and for some time it has been available on the web at http://www.npl. washington.edu/AV/altvw104.html . Two months ago I received an e-mail from a mother, who said that her 11-year-old son Daniel was doing a school project on the Big Bang. They had found my column on the web, and she was wondering if the sound of the Big Bang mentioned in my column was actually recorded anywhere, so that Daniel could play it for his class.
My answer was, of course, "No." However, the question caused me to consider the problem of how one might go about producing such a recording. With the available data from BOOMERanG and more recently from WMAP, it would not be very difficult to simulate the sound, using the symbolic algebra program Mathematica (Wolfram Research). Im a fairly skilled user of Mathematica, which I use for data analysis and calculations, and it includes the feature of rendering mathematical functions as sound that can be captured as .wav files on a computer.
I was fascinated by the idea of synthesizing the Big Bang sound, and it ran around in my head for a day or so. I decided that I wanted to hear what the Big Bang sounded like. So one Saturday morning, when I should have been doing something else, I downloaded the frequency spectrum measured by WMAP, the satellite probe that has done a definitive job of mapping the cosmic background radiation and was featured in one of my recent columns. Then I quickly wrote a 16-line Mathematica program that read in the WMAP data, produced the sound as a mathematical function, and saved it to my hard disk as a .wav file.
My PC has a good sound card and a substantial sub-woofer, so it reproduced the .wav file well. When I ran the program for the first time and the sound started echoing from my office, our two male Shetland Sheepdogs, Alex and Lance, came running into the room, barking with agitation. After they had looked around and determined that nothing terrible was happening, they lay down on the floor and listened attentively to the sound of the Big Bang, giving their characteristic Sheltie Stare to my sub-woofer.
My Mathematica program (or "notebook" in Mathematica-speak) combines the frequencies measured by WMAP and adds up the cosine waves corresponding to all of the data points, with each wave given an intensity that corresponds to the area it would represent on a bar graph. (This is done so that closely-spaced data points are not over-weighted as compared to widely spaced points.) The actual Big Bang frequencies, which had wavelengths that were some fraction of the size of the universe at that time, were far too low in frequency to be audible to humans (even had any been around). Therefore the WMAP frequencies had to be boosted upward by a huge factor (about 1026). The waves of all frequencies were set to start at their maximum values at t=0, which in the simulation corresponds to the start of the Big Bang.
The simulation includes three important effects: (1) The multiply peaked frequency spectrum measured by WMAP is made into a single sound wave (monaural, not stereo) by the process described above; (2) According to the WMAP analysis, the emission profile of the cosmic background radiation peaked at 379,000 years and dropped to 60% intensity at 110,000 years before and after the peak emission time. The simulation lasts 100 seconds and represents the first 760,000 years of evolution of the universe, as the emitted CBR rises and falls in intensity following the WMAP profile; (3) The universe was expanding and becoming more of a "bass instrument" while the cosmic background radiation was being emitted. To put it another way, the expanding universe "stretches" the sound wavelengths and thereby lowers their frequencies. To account for this effect, the program shifts the waves downward in frequency to follow the expansion in the first 760 thousand years of the universe. How fast the universe initially expanded depends on what cosmological model is used. I decided to follow the predictions of the flat-space Robertson-Walker metric with zero cosmological constant. That model predicts that the radius of the universe grows as time to the 2/3 power (R ~ t2/3). Therefore, instead of the component cosine waves varying as frequency x time, they vary as frequency x time1/3 to implement the cosmological Doppler shift. The resulting .wav file, along with several variants of it with other running times, can be downloaded from my web site at http://faculty.washington.edu/ jcramer/BBSound.html.
After I had produced the .wav file, I sent a copy as an E-mail attachment to Daniel. His mother reports that his science project was a great success. I also mentioned the simulation to Marcus Chown, a science writer with whom I frequently interact and who writes for New Scientist. Marcus wrote a short article about it, and it was publicized in a New Scientist press release about their upcoming issue. The result was an amazing media explosion. The story was picked up by newspapers and news services around the world, including Aljazeera, the Telegraph (UK), Ananova (UK), news.com.au (Australia), The National Post (Canada), EurekaAlert!, The Mirror (UK), The Australian, The Brisbane Courier Mail (Australia), and The Frankfurter Allgemeine (Germany). It was also the feature front page story in the Turkish newspaper Sabah, crowding out reports of the October California fires. I was interviewed by three different parts of the BBC in the same 24-hour period, and I did several other radio interviews including a "Living on Earth" segment for National Public Radio. Many thousands of file requests for the .wav file hit our local web server, which was making the audio file available, and filled its hard disk, causing an ugly system crash. We had to move the file to a more robust system to deal with the traffic. Things have calmed down by now, but our server still gets many download requests.
Apparently the sound file is also of interest to musicians. I was interviewed by a news service that focuses on music. The sound file in several running lengths was placed on a server that provides sound files to radio stations and the music industry.
Frankly, Im amazed by the vibrant chord of public response that was struck by this little exercise in data presentation. My hour of work on a quiet Saturday morning has generated more interest in my work than all the rest of my scientific output in a long and varied career of scientific research and popular science writing. Im now wondering what I might do for an encore. The Sound of the Galaxy? The Sound of the Sun? The Sound of a RHIC (Relativistic Heavy Ion Collider) Collision? Suggestions (in moderation) would be appreciated.
Finally, here are the answers to two recurring questions that have been asked about the sound file:
Q: How can you represent it as a sound? Sound is supposed to be a wave that travels through air, and there was no air in the early stages of the Big Bang?
A: The Big Bang Sound in the simulation is derived from the sound propagating as compression waves through the plasma/hydrogen medium of the early universe some 100 to 700 thousand years after the initial Big Bang. The density of this medium was changing as the universe expanded, but should have been considerably more dense than air on our little planet. One does NOT need air to have sound, only some medium in which compression/rarefaction waves can propagate. The sound waves were very low in frequency and had wavelengths comparable to some fraction of the size of the universe. For the convenience of humans, who could not hear such low frequencies, I have increased them to the audio range of the human ear.
Q: How could the universe have a radius of 18 million light years only 0.76 million years after the Big Bang. It would have to be expanding faster than the speed of light.
A: The VISIBLE radius of the universe at age 760,000 years is, of course, 760,000 light-years. However, even at this early age the scale radius of the universe is already larger than its visible radius, with regions already out of causal contact. The SCALE radius of 18 million light years mentioned in the New Scientist article is an estimate based on assuming that the universe expands as time to the 2/3 power (as it would for a flat Robertson-Walker metric with zero cosmological constant) and propagating backwards 3/2 of the present visible radius of the universe to its scale radius at age 760,000 years.
AV Columns Online: Electronic reprints of over 120 "The Alternate View" columns by John G. Cramer, previously published in Analog, are available on-line at: http://www.npl. washington.edu/av. Electronic preprints of papers listed below are available at: http://arxiv.org.
"First Year Wilkinson Anisotropy Probe (WMAP) Observations: Preliminary Maps and Basic Results", C. L. Bennett, et al, submitted to Astrophysics Journal, preprint astro-ph/0302207 available at http://arxiv.org.