Sunday, September 02, 2007

Book Review - Parallax: The Race to Measure the Cosmos

Back at the MARAC conference last April, one of the keynote speakers was by Alan Hirshfeld who was selling copies of his newest book The Electric Life of Michael Faraday. I'm not really into electricity so I didn't grab a copy, but my friend Luis told me about his last book on Parallax (which I outlined the idea of in this post).

I'd been wanting to get a copy, but never could find a bookstore that carried it. Finally, when I got back to school this semester, I checked out a copy from the library.

Overall, it was a pretty good book. One of the main points is raises is the problem of acceptance of the heliocentric model. By the time of Galileo, it was pretty well established that the "crystalline spheres" of the geocentric Ptolemaic models didn't really exist and that, while it was a nice idea, it wasn't really much more than a useful mathematical model. As long as models matched with observation, they're considered to be good. The problem for the heliocentric model is that it didn't do any better of a job than the old one. Thus, there was no reason to rock the boat and try to supplant it.

Unless, of course, observations could be made that were in agreement with the heliocentric model that the geocentric one could not explain. Since an Earth orbiting the Sun gives a larger baseline than the simple width of the Earth as it rotates on its axis would, this would mean that the heliocentric model would allow for measurements of stellar parallax while the geocentric one would not.

Astronomy classes often tell of Galileo's observations of Jupiter and its four largest moons and how, seeing that smaller bodies could orbit larger ones, this convinced the astronomical community of the reality of the heliocentric model. However, this is a bit of a stretch. While it was compelling, what was really needed to put the nail in the coffin was the determination of parallax.

But this is no easy feat. Even the closest star only has a parallax angle of .75 arc seconds (recall 60 arcminutes = 1 degree & 60 arc second = 1 arcminute). In other words, .0002ยบ. That's a tiny angle! Given the additional problem of the distortion of the atmosphere, the wobble of Earth's orbit, and a host of other detrimental factors, it's quite a challenge to take on. Of course, that's just for the nearest star. If a star is further, it will make an even smaller angle. So aside from just measuring the angle, there's also the trick of choosing the right star.

But that didn't stop a long line of astronomers from trying.

The infamous Tycho Brahe was one of the first to attempt this despite not believing in the heliocentric model and having his own Tychonic model. However, his instrument was only a large quadrant and far too coarse to measure such small angles.

To improve accuracy, the invention of the telescope was needed. Although Galileo was the pioneer of the basic telescope design, the quality of the optics was still far too poor to allow for precise angles. In the late 1600's, Robert Hooke attempted to measure parallax, but still with no success.

By 1725, James Bradley and Samuel Molyneux picked up the challenge, but although they claimed that their telescope could measure to better than one arcsecond (approaching the necessary limit), they chose the wrong star (Gamma Draconis). However, while they failed to measure any movement due to parallax, the two did discover another oddity: There is another movement that had not been expected. What they had discovered was the annual aberration of starlight. While this wasn't what they had been looking for, it was ultimately an observation that fit only with the heliocentric model and could not be explained by the geocentric one. Although the heliocentric model was now proven beyond any reasonable doubt, a measurement of stellar parallax had still never been made. Astronomers continued to look for it.

In the 1780's renowned astronomer William Herschel picked up the task. Working under the assumption that all stars were really about the same overall brightness (a completely unfounded and incorrect assumption), his idea was to observe systems of visual double stars (stars which are close together but are not bound gravitationally). Under his assumption, the fainter one would be further away and provide a measuring stick by which to measure the parallax of the brighter and presumably closer star. He cataloged thousands of double star systems. In 1802, he decided to go back to several to check their separations again, but discovered that although they had changed with respect to one another, the manner in which they had done so was inconsistent with parallax. Instead, they were orbiting around one another. His method was doomed to failure.

The next to join the race was Thomas Henderson, who assumed that stars that drifted the most quickly (what's known as proper motion) were the ones that were nearest to us for the same reason that a car nearby looks to move much faster while a plane which is actually moving far faster, seems to glide slowly across the sky. Working in the southern hemisphere, he chose the star with the highest proper motion he could find: Alpha Centauri. Since southern observatories weren't as well equipped, he was using an inferior telescope, but by 1833, with 19 measurements, he had appeared to have determined a parallax. But by then he had moved back to England. He decided to wait for his assistant still at the southern observatory to make additional observations before announcing anything definite.

in 1834, William Bessel turned his attention to trying to determine a measurable parallax. He turned his attention to 61 Cygni which also had a high proper motion. But right behind him was Wilhelm Struve who had at his disposal a telescope made by the legendary Fraunhaufer, with the ability to measure stellar positions to a few hundreds of a second of arc. Struve was looking at Vega. By the end of 1837, Struve had determined a parallax angle of 1/8 of an arcsecond, but since he only had 17 measurements, he decided to wait for more before making his announcement.

Bessel then redoubled his efforts on 61 Cygni. By October of 1838, he had compiled hundreds of measurements and announced his findings: a definitive parallax angle of .314 arcseconds. Henderson published his results of Alpha Centauri only two months later and Struve, his results for Vega during the following year.

Overall, this book wasn't a bad read. It explores the lives of the people involved more than necessary, but this adds a nice bit of flavor. For people more familiar with astronomy, parts tend to get boring when basic concepts are explained in great detail, but this makes it easily accessible for even someone who's never read anything on astronomy in their lives. My biggest dislike was that Hirshfeld tends to weave in personal anecdotes that add absolutely nothing to the story he's telling. While the majority of the book is a good historical recounting, these bits, although intended to bridge topics, break the simple chronological flow and seem highly out of place.

Still, it's a book I'd recommend reading. Assuming you can find a copy...