Pioneers of celestial measurement

kw: book reviews, nonfiction, science, scientists, astronomy, astrophotography, spectroscopy

Look carefully at the white line across the gray band, where the ink marks are in each section. The ink marks were made by Edward Pickering in 1889, when he noticed the doubling of the Calcium K line (λ=393 nm, near UV) in the upper photo of the spectrum of Mizar. Mizar, also known as Zeta Ursa Majoris, is the brighter of the Mizar-Alcor double star in the "corner" of the handle of the Big Dipper. One needs keen eyes to see that it is double.

Mizar itself was found to be double by comparison of these two spectra photographed a week apart in 1887. It is the first known "spectroscopic binary". Two stars of roughly equal brightness circle each other in about 20½ days. The splitting of the K line (and all the other lines shown if you look closely) is because of the Doppler effect: when one star is moving toward us, and the other is moving away, the wavelength of the light that reaches us is shifted, one toward the blue, the other toward the red end of the spectrum.

This discovery was made possible by the "glass universe" being compiled at Pickering's behest by pioneers of astrophotography and astrospectroscopy whom he had commissioned to photograph the whole sky, over and over, on glass/emulsion plates using telescopes owned by Harvard Observatory.

This immense photographic effort, and the numerous women—and a few men—who made literally hundreds of thousands of discoveries using the plates, are chronicled in The Glass Universe: How the Ladies of the Harvard Observatory Took the Measure of the Stars, by Dava Sobel. I believe I must declare this book the most fascinating I have read so far this year. I have known for many years of "Pickering's Harem", the female "computers" who carried out manual calculations for the Harvard Observatory, and I knew of a few astronomers, names now to conjure with!, such as Annie Cannon, Mina Fleming, Cecelia Payne, and Henrietta Leavitt, whose work in the late 1800's and early 1900's literally opened the heavens by classifying the stars, discerning nebulae, measuring stars' temperatures, and discovering the period-luminosity relationship that became the yardstick for measuring the size of the Universe. This book brings them all to life for us.

Please forgive me a sort of quibble at the outset (not about what the author wrote, however): I read the Large Print edition by Thorndike Press. On the copyright page the publisher put a standard disclaimer for a work of fiction. I contacted Ms Sobel, and she assured me that there is no fiction in this book. I am very glad of that!

There is a common notion that the Harvard computers were given mainly "grunt work" and had little else of value to contribute. Not so! When Edward C. Pickering assumed leadership of the Harvard Observatory in 1877 a number of computers already worked there, most of them female, and he added more and more, eventually hiring more than 80. He always looked for hidden talents and helped the computers develop as far as they could. Williamina ("Mina") Fleming was originally hired as a maid, but he soon found she was a capable computer, and she went on to co-develop the system for classifying stars that we still use, based on the mnemonic "O, Be A Fine Girl, Kiss Me", which sorts the spectral types by temperature from hottest to "coolest" (still hotter than the filament in a light bulb). In this picture, Mrs. Fleming is standing, and the computers of the day, including Annie Cannon just below her, are shown searching photographic plates and calling out their readings to a compatriot seated nearby.

Ms Sobel presents the life stories of a dozen or so of the computers-turned-astronomers and their colleagues, and shows equal interest in the instruments and methods used to make their discoveries. And the entire narrative is wrapped around the philanthropy of two women whose fortunes underwrote much of the work. Firstly, Anna Draper, wife and collaborator of astronomer Henry Draper, came to Pickering after her husband's untimely death in 1883 and offered to support a continuation of the stellar spectroscopy and cataloging work that Henry had begun with her assistance. The support continued until her death in 1914, and her bequest to the Observatory allowed the rest of the Henry Draper Catalog (still in use) to be completed thereafter. Stars with labels such as HD217014 (AKA 51 Pegasi, the star around which the first exoplanet was discovered) are cataloged therein. Secondly, Catherine Bruce funded the production of the Bruce Telescope, a 24-inch-diameter refractor, which was used first at Cambridge, Massachusetts in 1893, then in Peru and finally in South Africa until 1950. This telescope and several of smaller size were used to take the photographs that make up the bulk of the Glass Universe, a four-dimensional archive of the sky from 1885 to 1993!

This is a small segment of one of the glass plates, showing the globular cluster 4 Tucanae. Astrophoto plates are negative images, and were read directly by researchers such as the Harvard computers, one of whom has inked directly on the plate a few arrows and lines to point out certain stars of interest, most likely variable stars.

A technical note: Very early on it was found that the best discernment of star images could be had by exposing a plate until the background skyglow produced a "density 1" gray, which passes 10% of the light when the plate is back lit. Depending on the darkness of the sky at the observing location, and the photographic speed of the plates used, this would usually take an exposure of between 30 and 90 minutes. In one night of observing at a very dark site with fast plates, one might take 20 or more exposures.

A very major part of the work at HAO was to discover variable stars and chart their variation over time. The stars marked in this plate were found by comparing plates taken over several days' or weeks' time. The long- and medium-period variables were typically the most consistent, and this was fortuitous, because Henrietta Leavitt later discovered that the greatest number of these are Cepheid variables whose period of variation is proportional to their brightness at its peak.

Cepheid variables are giant stars with masses 4 to 20 times that of our Sun, and are as much as 100,000 times as bright. This makes them visible over very great distances, up to tens of millions of light-years, using larger telescopes. The Leavitt Law, or Period-Luminosity Relationship, allows measurement of the distance to galaxies within that range of distance. Such measurements led to Edwin Hubble's discovery of the expansion of the Universe.

If you look at a group of bright stars such as the Orion constellation or the Hyades cluster (the horns of Taurus, the Bull), after a while you can notice that a few stars are yellowish or reddish compared to the rest. Orion, in particular, has the star Betelgeuse (and, yes, it is pronounced "beetle juice"), which is visibly rather orange, in one corner. Most of the rest of the stars in Orion are quite bluish.

For stars, blue means very hot, white means "sorta hot", yellow is not as hot, and orange-red is the coolest. To put numbers on it, Rigel, in the corner of Orion opposite Betelgeuse, is a blue-white giant star with a visible-surface temperature of 12,000 K (over 21,000°F), as compared to our Sun, a yellow-white star of temperature about 5,800 K (9,900°F). Betelgeuse, while not the coolest visible star, comes close at a temperature of 3,500 K (5,800°F). The tungsten filament in a (now nearly obsolete) halogen light bulb typically has a temperature of 3,300 K (just below 5,400 °F), so a "piece" of Betelgeuse brought into your living room would look slightly less yellow than a halogen lamp.

The observers and computers at Harvard took advantage of spectroscopy to do much better than just making visual color estimates. The gray and black streaks on this plate image are spectra of stars, photographed with the help of an objective prism. The objective prism for the Bruce Telescope was a thin wedge of glass more than 24 inches in diameter that turned the whole telescope into a multi-stellar spectrograph. It also incorporated a slight curvature in one direction to turn a stellar point of light into a thin streak, so that the spectra would have useful width.

These little streaks may not look like much but they record an amazing amount of information about a star. Much more than the proportion of blue to red light—which are hard to determine from such photos, though it is not impossible—, the spectra include Fraunhofer lines. These are absorption lines caused by elements in the gaseous upper atmosphere of the star. The kinds of lines that are present are a much more sensitive indication of both the composition and the temperature of the star.

If one were to look through a telescope set up in this way, it might look something like this, from a photo taken at the University of Virginia. This shows the Hyades cluster; if you concentrate on the position of the red end of each spectrum you can see the tilted "V" shape of the cluster.

This photo as shown here is too small for Fraunhofer lines to be seen, so I made this clip of just a few of the spectra:

The dark lines are the more prominent Fraunhofer lines. The K line mentioned above is barely visible in one of these spectra at the far right. Its wavelength of 393 nm is just beyond the traditional edge of Ultraviolet (400 nm), but that wavelength is actually visible to most people if the spectrum is bright enough. Each line is characteristic of a particular element. A dark line in the narrow yellow area would indicate Sodium, for example, just as the K line indicates ionized Calcium. The very strong lines for Hydrogen and Helium found in the spectra of the hotter stars of categories O and B led to the discovery that stars are primarily made up of Hydrogen, about ¼ Helium, and all the other elements add up to no more than about 1%.

Annie Cannon and others excelled in looking at the gray streaks, the hundreds to thousands of them that populated each plate, and categorizing each star by temperature and "spectral type" such as G2 (the spectral type of our Sun). Miss Cannon eventually categorized a third of a million stars.

I could go on and on, but this is long enough already. I love a book like this, that tells about the people and the work they did and why it is important. Without the "boring" work of the Harvard computers and astronomers, nearly all female, we would know only a tiny fraction of what we have learned about the Universe.

Follow-up: The Harvard plates are presently being digitized for Digital Access to a Sky Century @ Harvard, or DASCH ("Dash"). Have a look, but beware, there is a large learning curve. If you want to have a turn at stellar classification, check out Stellar Classification Online Public Exploration, or SCOPE., a Citizen Science project. While a few million stars have been classified, the great majority of the billions of stars, just in the Milky Way galaxy, have yet to be classified. Enjoy!