The sky's changing phenomena have always been a source of wonder. Nearly all the changes seen with the unaided eye result from the movements of solar-system objects. But far in the background, a number of bright variable stars can also be followed through their brightness cycles without optical aid. Compared to most of their telescopic counterparts, the naked-eye variables are easy to find. Once you know a few of them, you can check on how they are doing whenever you look up at the sky.
Thirty-four variable stars have a range of at least 0.4 magnitude and become brighter than visual magnitude 4.0, according to the authoritative General Catalogue of Variable Stars (GCVS) and its supplements the Name-Lists of Variable Stars. (This doesn't include novae or supernovae, which occasionally reach naked-eye brightness.) Among these stars are many eclipsing binaries, Cepheid variables, and semiregular red variables, as well as a few long-period stars of the Mira type and the recurrent nova T Coronae Borealis. As many as 24 do not fade below magnitude 5.1 and so remain visible to the unaided eye all the time. It's interesting that only seven of these are south of the celestial equator, compared to 17 north of it. Could there be several undiscovered naked-eye variables in the southern sky waiting to be noticed?
Here is a personal list of my dozen favorite northern naked-eye variables, in no particular order. Many types of variable stars go unrepresented in this sample, since many are low-luminosity objects too faint to be visible without a telescope or vary too slightly for their changes to be visually noticeable. Small binoculars will help in observing the fainter phases of some of these stars, especially if you don't have really dark skies.
All the light curves in this article resulted from hundreds of naked-eye observations by the author. Each plotted point is the average of between seven and 11 brightness estimates made on different dates. For clarity, data are repeated to cover more than one cycle.
Algol and Lambda Tauri
Algol (Beta Persei), the prototype eclipsing binary, dips from magnitude 2.1 to 3.4 every 2.87 days. Each eclipse, including the partial phases, takes nearly 10 hours. The dimmings are most obvious for the two hours that the star stays close to minimum light. Algol is a star of fall and winter evenings. A comparison-star chart for estimating Algol's magnitude appeared in the November 1996 issue of Sky & Telescope, page 66. (For tips on estimating a variable's brightness, see "The Lure of Variable Stars.")
Nearby Gamma Andromedae, magnitude 2.1, makes a handy comparison for first-glance checks. To monitor Algol more closely, make a visual brightness estimate every half hour for as many hours as possible spanning a predicted eclipse. You can derive the time of mideclipse from a graph of your magnitudes. Such timings provide a useful check on the accuracy of the predictions.
Alternatively, you can estimate the star's magnitude once or twice a night and start making more frequent observations if it is noticeably fainter than normal. In this way, I obtained the star's complete light curve. It showed that the eclipses were coming appreciably later than predicted at the time. My light curve even contains a hint of the secondary minimum halfway between the primary eclipses. This is only 0.05 magnitude deep, though, and I was completely unaware of it at the time of the observations.
Lambda Tauri in the back of the Bull is another Algol-type eclipsing binary, less well known due to its smaller magnitude range of 3.4 to 3.9. The eclipses last 14 hours, too long to cover in a single night. But enough random observations will define the light curve well. The period (3.953 days) is just an hour short of four days, so once the eclipses start coming in the evening, they will repeat every four days for about a month.
In addition to the primary minimum, my light curve shows the 0.2-magnitude secondary minimum. As in the case of Algol, there was a significant difference between the observed and predicted times of minimum light. Accurate photoelectric magnitudes for suitable comparison stars are readily available in modern sky atlases.
Beta Lyrae and Delta Cephei
Beta Lyrae is an eclipsing binary of a different type. Algol-type binaries appear nearly constant between eclipses, because the brighter star of the pair is approximately spherical. But Beta Lyrae's components are so close together that they are distorted into ellipsoids by each other's gravity. As the system revolves in its 12.94-day orbital period, we see continuous change at all phases of its light curve.
The magnitude range is cataloged as 3.3 to 4.4. The mean light curve from my own observations shows a smaller range, perhaps partly because of observational bias. See Sky & Telescope:, June 1993, page 72, and June 1994, page 72, for more about watching this rapidly evolving binary star.
Delta Cephei is the prototype Cepheid variable, a class of giant stars that pulsate with periods proportional to their luminosity. This relationship is exploited by astronomers to determine the brightness and thus distance of Cepheids in other galaxies.
Delta's magnitude ranges from 3.5 to 4.4 in a cycle of 5.37 days. The fade from maximum to minimum is slower than the rise back to maximum, which takes less than two days.
Zeta Geminorum, Eta Aquilae, and Others
Zeta Geminorum is another Cepheid variable. But its light curve is more symmetrical than the preceding two, with the declining and rising phases each taking half of the 10.15-day cycle. The magnitude range is 3.6 to 4.2. According to my light curves, all three Cepheids reached maximum light at the times predicted by the GCVS. Many Cepheids do have slightly variable periods, but the deviations from prediction are usually smaller than those often seen in eclipsing binaries.
Among the naked-eye variables are several red giant or supergiant stars that change irregularly or semiregularly, with transient or multiple periodicities. Mu Cephei (magnitude 3.4-5.1, periods 2 and 12 years) and Alpha Herculis (2.7-4.0 with rough cycles of about 100 days and 6 years) show noticeable variations if you observe them patiently for long enough.
Eta Geminorum is a binary star whose bright component is a semiregular red giant. Most of the time it varies only slightly from magnitude 3.2. But every 8.2 years the bright component is eclipsed by its companion star, causing the system's total light to drop to about 4.0. The next of these eclipses is expected in October 2012.
Eta Aquilae, a Cepheid of summer and fall, closely resembles Delta Cephei in its magnitude range, 3.5 to 4.4, and the general shape of its light curve. But its period is 7.18 days. About halfway down the decline there is a temporary reversal, visible as a small hump in my light curve.
Mira, Gamma Cassiopeiae, and Betelgeuse
When it comes to pulsating giants we can hardly pass up Mira, Omicron Ceti. With a declination of -3° it doesn't quite meet our criterion of being in the north celestial hemisphere, but this seems like splitting hairs. Mira had an unusually brilliant maximum in the winter of 1996-97. At a typical peak of its cycle Mira reaches magnitude 3.4, but in early February 1997 it topped out at about 2.5 and stayed there for the rest of the month.
Mira is the brightest of the red long-period variables, and (again barring novae and supernovae) it presents the most radical changes that can be seen with the naked eye beyond the solar system. Mira's period of 332 days means its maxima come one month earlier each succeeding year. In 2009, maximum brightness occurred in late November; in 2010 maximum is predicted for late October. But the date, like the peak brightness, is never exactly predictable.
The unstable hot star Gamma Cassiopeiae is generally low in the north. Previously magnitude 2.25, it rose to 1.6 for many months in 1937 when it ejected a shell of gas. The familiar W pattern of Cassiopeia looked noticeably different. After fading to 3rd magnitude in 1940, Gamma slowly brightened to 2.2 by 1966. Since then it has shown little visible change, but there is no telling when it might again act up.
AAVSO: The Variable-Star Experts
The American Association of Variable Star Observers (AAVSO) keeps track of all these stars, but isolated estimates of Algol and Lambda Tauri made at random times are of little value. For these eclipsing stars, the AAVSO prefers to receive series of estimates made during the course of an eclipse, from which the time of minimum can be determined. In the case of Lambda Tauri with its long eclipses, this means observing the fade into an eclipse on one occasion and the rise from another eclipse on a different night. Several nights' observations can be combined into a single graph if the star's period is well known. For more information, charts, and observing techniques, contact the AAVSO at 49 Bay State Road St., Cambridge, MA 02138 USA
Most bright naked-eye variables have small magnitude ranges, so care is needed in making estimates. Above all, avoid using comparison stars whose heights above the horizon differ greatly from that of the variable, as they will then be seen through different thicknesses of atmosphere. If circumstances force you to use such stars, apply atmospheric extinction corrections to each one.
These corrections are especially important with Betelgeuse, Alpha Orionis, the brightest noticeably variable star of all. It fades and brightens slowly and semiregularly with a suggested underlying period of about 6 years.
How many skywatchers know that Betelgeuse can grow nearly as bright as Rigel or as faint as Aldebaran? Its visual magnitude has ranged from about magnitude +0.3 (in late winter 1988 and early winter 1990-1991) to as faint as +0.9 (in late winter of 1989 and 1993 and early spring of 1995).